jr^'W^') 
 
 
 
WBSMUmrmWIf^^jrBm 
 
 
 ^ -V^r.. /:?:. 
 
 
 ^l;/^v -•^OAr';..?^''^ 
 
 
 ^A^wwiuj: 
 
 
 ?^\ r-'^- 
 
Digitized by the Internet Archive 
 
 in 2010 with funding from 
 
 NCSU Libraries 
 
 http://www.archive.org/details/contributionstocOOeric 
 
^.o3. 6 
 
 ^ 
 
 &:, 
 
 '~^ ^' /Z^ ^- 
 
 ■A-'^^-'C--- — ? 
 

 ■y<-<^ 
 
 ^'rf^' 
 
CONTRIBUTIONS 
 
 CESTEffllAl EXHIBITION. 
 
 JOHIf ERIOSSOX, LL.D. 
 
 HONORARY DOCTOR OF PHILOSOPHY OF THE ROYAL UNIVERSITY OF LUND; MEMBER OF THE 
 
 ROYAL ACADEMY OF SCIENCES, STOCKHOLM; MEMBER OF THE ROYAL ACADEMY OF 
 
 MILITARY SCIENCES OF SWEDEN; HONORARY MEMBER OF THE ROYAL SCIENTIFIC 
 
 SOCIETY OF UPSALA; AND MEMBER OF VARIOUS OTHER SCIENTIFIC 
 
 INSTITUTIONS IN EUROPE AND AMERICA; 
 
 KNIGHT COMMANDER, WITH THE GRAND CROSS. OF THE ORDER OF NORDSTJERNAN ; KNIGHT 
 
 COMMANDER OF DANNEBROC, FIRST CLASS; KNIGHT COMMANDER OF ISABEL LA 
 
 CATOLICA; KNIGHT COMMANDER OF SANCT OLAF ; AND 
 
 KKIGHT OF THE ORDER OF VASA. 
 
 NEW YORK: 
 PRmXED FOR THE AUTHOR AT "THE NATION" PRESS. 
 
 1876. 
 
INTRODUCTION. 
 
 The Commissioners of the Ceuteunial Exliil^itiou baviug 
 omitted to invite me to exhibit the results of my labors 
 connected with mechanics and physics, a gap in their re- 
 cord of material progress exceeding one-third of a century 
 has been occasioned. I have therefore deemed it proper 
 to publish a statement of my principal labors during the 
 last third of the century, tlie achievements of Avhich the 
 promoters of the Centennial Exhibition have called upon 
 the civilized world to recognize. 
 
 The nature of the labors referred to will be seen by 
 the following account of philosophical instruments, engines, 
 and other structures described and illustrated in this work 
 — viz. : Apparatus for measuring the intensity of radiant 
 heat at given distances. Instrument for measuring radiant 
 heat emitted by concave spherical radiators within exhausted 
 
 ^ 
 
iv INTRODUCTION. 
 
 euclosiires. Instriimeut showing the rate of cooling of a 
 heated body within an exhausted cold enclosure. Instru- 
 ment showing the rate of heating of a cold body ^vithi^ 
 an exhausted heated enclosure. Instrument showing the rate 
 of cooling of an incandescent sphere within an exhausted 
 cold enclosure. Instrument for measuring the dynamic 
 energy developed by radiant heat at different intensities. 
 Actinometer, for measuring the temperature developed by 
 solar radiation. Solar Calorimeter, for measuring the dyna 
 mic energy developed by solar radiation. Portable Solar 
 Calorimeter. Parallactic mechanism, for measuring the in 
 tensity of radiation from different parts of the solar disc 
 Instrument for measuring the radiant power of the solar 
 envelope. Instrument for measuring the actual intensity of 
 the sun's rays. Solar Pyrometer, for measuring the tempe 
 rature of the solar sui'face. Apparatus for measuring the 
 radiant intensity of flames. Instrument fo]- measuring radi- 
 ation from incandescent planes at different angles. Instru- 
 ment for measuring the radiation from different zones of 
 incandescent spheres. Calorimeter, for measuring the dyna- 
 mic enei'gy developed by radiation from fused iron. Appa- 
 ratus for measuring radiant heat by means of the thermo- 
 electric pile. Barometric Actinometer, for measuring the 
 temperature developed by solar radiation. Apj^aratus for 
 ascertaining the conductivity of mercury. Concave spherical 
 
INTRODUCTION. t 
 
 radiator, for testiug the accuracy of the sohar pyrometer, 
 lustrumeut for measuring the reflective power of silver aud 
 other metals. Rapid-indication Actinometer, for measuring 
 the temperature developed by solar radiation. Apparatus 
 for ascertaining the diathermancy of flames. Dynamic Re- 
 gistei', for measuring the relative power of currents of water 
 and vapor. Distance-instrument, for measuring distances 
 at sea. Steam fire-engine, designed 1841. Engines of the 
 United States steamship Princeton, built at Philadelphia, 
 1842. Twelve-inch wrought-iron gun aud carriage mounted 
 on board the Princeton, 1843. Iron-clad cupola vessel, de- 
 signed 1854. Surface-condenser for marine engines, patented 
 1849, built at New York. Experimental caloric engine, 
 i)uilt at New York, 1851. Caloric engine for domestic 
 purposes, extensively introduced in Europe and America. 
 The iron-clad turret-vessel Monitor, built at New York, 
 1861. Turret-vessels of the Passaic class, built at New 
 York and other places, 18G2. The Monitor engine, applied 
 to the entire iron-clad fleet of the United States during- the 
 war. The turret-vessel Dictator, built at New York, 18G2. 
 Carriages for heavy ordnance, designed 1861, built at 
 numerous mechanical establishments in the United States. 
 Pivot-carriages of the Si"»anish gunboats, Imilt at New York, 
 1869. Rotary gun-carriage and transit platform, built at 
 New York, 1873. Gun-cai'riage for coast defence, designed 
 
Ti INTRODUCTION. 
 
 1861, built at New York. ludepeudent twin sorew-engiues 
 t)f the thirty Spanish gunboats, built at New York, 1869. 
 New system of naval attack, published 1870. Movable tor- 
 pedo, biailt at New York, 1873. Air-compressor for the 
 transmission of mechanical poAver, built at New York, 1873. 
 Solar engine, actuated by the intervention of steam, built 
 at New York, 1870. Solar engine, actuated by the inter- 
 vention of atmospheric air, built at New York, 1872. 
 
 The foregoing, it should be observed, relates to work 
 carried out by me on American soil. It has no reference 
 to my labors in England from 1826 to 1839 connected with 
 locomotion, steam navigation, motive engines, and other 
 branches of mechanical and civil engineering. Nor does it 
 contain a comj^lete enumeration of the original mechanical 
 inventions carried into practice by me in the United States 
 — models of which would have been presented at the Cen- 
 tennial Exhibition had its promoters desired me to furnish 
 a record of my share in the progress of mechanical engi- 
 neering during the last thirty-seven years of the first cen- 
 tury of the Kepublic. 
 
 As our space only admits of a brief reference to the 
 mechanical inventions adverted to and not described or illus- 
 trated in this work, the following statement is appended, 
 furnishing an outline of the principal structures omitted — 
 viz. : Engines of the twin-screw steamship Clarion, built at 
 
INTIWDUVTION. Tii 
 
 New York, 1S40, cousistiug of two vertical cylinders, placed 
 fore and uft iu the vessel, actuating tLe cranks of the 
 screw-shafts by inclined conuectinof-rods. Vertical single 
 engines, actuating twin screws, built at New York, 1842, 
 applied to sev^eral freight vessels on the Delawai-e and 
 Knritan Canal. Single horizontal back-action engine, l)uilt 
 
 1843, applied to the United States screw-steamer Legare. 
 Inclined screw-engines, built 1843, applied to the steam- 
 ship Massachusetts, the steam-cylinders of which weie 
 placed near the deck at the ship's sides, secured to dia- 
 gonal timbers bolted to the planking. Centrifugal suction- 
 fan, built 1843, operated by an independent engine, for 
 producing draught in marine boilers by drawing the air 
 through the furnaces and flues, and forcing the products 
 of combustion into the chimney. Inclined engines, built 
 
 1844, applied to the bark Edith, the connecting-rods ope- 
 rating at right angles to each other and coupled to a 
 common crank-pin on the piopeller-shaft. Vertical engines, 
 built 1844, applied to the twin-screw vessel Midas (the 
 fii-st screw-vessel to round the Cape of Good Hope), the 
 power being transmitted to the propeller-shafts by vertical 
 connecting-rods actuated by horizontal beams placed trans- 
 versely under the deck. Vertical engines applied to nume- 
 rous screw-vessels employed on the coast and inland waters 
 of the United States, the cylinders being placed perpen- 
 
viii INTR OB UCTION. 
 
 dicularly above the propeller-shaft, the connecting-rods act- 
 ing downwards — a form of engine now employed in nearly 
 all sea-going steamers, but at that time (about 1844) 
 severely criticised by marine engineers. Engines of the 
 twin-screw ship Marmora, built 1843, consisting of vertical 
 steam-cylinders which, by means of beams working under 
 the deck and vertical connecting-rods, imparted indepen- 
 dent motion to the pi'opeller-shafts. Horizontal high-pres- 
 sure and condensing engine of the twin-screw steam-tug R. 
 B. Foi'hes, built 1844, provided with detached condenser 
 and air-pump actuated by an independent engine — a vessel 
 which, during a series of years, rendered valuable service 
 on the coast of Massachusetts by towing and relieving 
 ships in distress. Compound stationary engine, actuated by 
 very high pressure, in which the steam was expanded to 
 the utmost extent, elaborately described by Dr. Lardner, 
 who devoted much time to its theoretical consideration. 
 Horizontal engine applied to the screw-vessel Primero, actu- 
 ated by a mixture of steam and atmospheric air. Stationary 
 engines actuated by highly superheated steam, the pistons of 
 which were single-acting and thoroughly protected against 
 the injurious effect of high temperature. Experimental 
 street-car, propelled by a doul^le caloric engine. Hoisting 
 machines, actuated by cold compressed air, applied to several 
 warehouses in New York. Small motors, actuated by cold 
 
INTRODUCTION. ix 
 
 compressed air, successfully applied to the sewing-machines 
 of a large establishment in New York, intended to estab- 
 lish the fact that the present injurious physical exertion of 
 sewing-women may economically be dispensed with. 
 
 Regarding the descriptions and illustrations of the caloi'ic 
 engines contained in this work, it is proper to observe that 
 they relate only to some of the engines which I have built, 
 at least ten different types, unlike those described, having 
 been constructed and practically tested. Nor have I yet 
 wholly suspended the labors connected with this safe and 
 economical engine. The fact that it requires no water, and 
 that its principle is not incompatible with the desirable 
 emplo}Tnent of very high temperature — apart from the im- 
 portant circumstance that the use of atmospheric air admits 
 of returning at each stroke, by the process of regeneration, 
 the heat not converted into mechanical work during the 
 previous movement of the working piston — justify continued 
 endeavors to perfect this remarkable motor. 
 
 J. Ericsson. 
 
 New Yokk, September, 1876. 
 
CONTENTS. 
 
 CHAPTER PAGE 
 
 I. — Transmission of Radiant Heat, .... 1 
 
 II. — Radiation at Different Temperatures, . . 17 
 
 III. — Intensity of Solar Radiation, .... 52 
 IV. — Periodic Variation of the Intensity of Solar 
 
 Radiation, 75 
 
 v.— Mechanical Energy of Solar Radiation, . 91 
 VI. — Thermal Energy transmitted to the Earth by 
 Radiation from Different Parts of the 
 
 Solar Surface, 107 
 
 VII.— The Source of Solar Energy, .... 137 
 VIII.— Radiating Power and Depth of the Solar 
 
 Atmosphere, 153 
 
 IX.— The Feebleness of Solar Radiation demon- 
 strated, 1^^ 
 
CONTENTS. 
 
 CHAPTEE 
 
 X. — Temperature of the Solar Surface, 
 XL — Radiation from Incandescent Planes, . 
 XII. — Eadiation from Incandescent Spheres, . 
 XIII. — Radiation from Fused Iron, 
 XIV. ^ — Radiant Heat measured by the Thermo-Elec 
 
 TRic Method, 
 
 XV. — The Thermoheliometer, .... 
 XVI. — Barometric Actinometer, .... 
 XVII. — Conductivity of Mercury, .... 
 XVIII. — Incandescent Concave Spherical Radiator, 
 XIX. — Reflective Power of Silver and other 
 
 Metals, 
 
 XX. — Rapid-indication Actinometer 
 
 XXI. — Solar Radiation and Diathermancy of Flames, 
 XXII. — Constancy of Rotation of the Earth incom- 
 patible with Solar Influence, . 
 XXIII. — Distance Instrument, for measuring Dis- 
 tances at Sea, 
 
 XXIV. — The Steam Fire-Engine, 
 
 XXV. — The Steamship Princeton, 
 
 XXVI. — Twelve-inch Wrought-iron Gun and Car- 
 riage, 
 
 FAQE 
 181 
 
 209 
 220 
 229 
 
 239 
 254 
 266 
 275 
 
 285 
 
 294 
 310 
 317 
 
 327 
 
 380 
 386 
 391 
 
 400 
 
OONTENTa. xiii 
 
 CEAPTES PAGE 
 
 XXVII.— Application of the Submerged Propeller 
 
 FOR Commercial Purposes, . . . 404 
 XXVIII. — Iron-clad Steam Battery, with Revolving 
 Cupola, submitted to Emperor Napo- 
 leon III., 410 
 
 XXIX. — Surface-Condenser, operated by Indepen- 
 dent Steam Power, 417 
 
 XXX.— The Caloric Engine — Application of 
 
 Heated Air as a Motor, . . . 425 
 XXXI.— Caloric Engine for Domestic Purposes, . 439 
 XXXII.— The Monitor System of Iron-clads, . . 460 
 XXXin. — The Monitor Turret and the Centennial 
 
 Exhibition, 471 
 
 XXXIV.— The Monitor Engine, 478 
 
 XXXV.— The Monitor Dictator, .... 492 
 
 XXXVI.— The Monitor Turret and the Casemate, . 498 
 
 XXXVII.— Carriages for Heavy Ordnance, . . 505 
 XXXVIII.— Pivot Carriages of the Thirty Spanish 
 
 Gunboats, 510 
 
 XXXIX.— Rotary Gun-Carriage and Transit Plat- 
 form, 517 
 
 XL.— Gun-Carriage for Coast Defence, . . 521 
 
xiv CONTENTS. 
 
 CHAJTEB PAOB 
 
 XLI.— The Thirty Spanish Gunboats and their 
 
 Engines, 625 
 
 XLII. — A New System op Naval Attack, . . . 632 
 XLIII. — Submarine Warfare — The Movable Torpedo, 540 
 XLTTV". — Transmission of Mechanical Power by Com- 
 pressed Air, 549 
 
 XLV.— Sun Power— The Solar Engine, . . . 568 
 
LIST OF PLATES. 
 
 PLATX 
 
 1. Apparatus for measttring Radiant Heat. 
 
 2. Instrument for measuring the Intensity ok Radiation 
 
 FROM Enclosed Concave Radiators. 
 
 3. Diagrams showing the Propagation of Radiant Heat 
 
 THROUGH Space. 
 
 4. Instrument showing the Rate of Cooling of a Heated 
 
 Body within an Exhausted Cold Enclosure. 
 6. Instrument showing the Rate of Heating of a Cold 
 Body' within an Exhausted Heated Enclosure. 
 
 6. Instrument showing the Rate of Cooling of an Incan- 
 
 descent Sphere within an Exhausted Cold Enclo- 
 sure. 
 
 7. Instrument for measuring the Dynaiuo Energy deve- 
 
 loped BY Radiant Heat. 
 
XVI LIST OF PLATES. 
 
 PLATE 
 
 8. actinometer for measuring the intensity oe solak 
 
 Radiation. 
 
 9. Diagrams showing the Intensity of Solar Radiation 
 
 AT Different Zenith Distances. 
 
 10. Solar Calorimeter, for measuring the Mechanical 
 
 Energy of Solar Radiation. 
 
 11. Portable Solar Calorimeter, for measuring the Me- 
 
 chanical Energy of Solar Radiation. 
 
 12. Diagrams showing the Radiation from Different Parts 
 
 of the Solar Disc. 
 
 13. Parallactic Mechanism for measuring the Intensity of 
 
 Radiation from Different Parts of the Solar Disc. 
 
 14. Diagram showing the Attraction within the Solar 
 
 Mass at Different Distances from its Centre. 
 
 15. Instrument for measuring the Radiant Power of the 
 
 Solar Atmosphere. 
 
 16. Diagrams showing the Radiant Power of the Solar 
 
 Atmosphere. 
 
 17. Instrument for measuring the Actual Intensity of 
 
 the Sun's Rays. 
 
 18. Instrument for showing the Feebleness of Solar Ra- 
 
 diation. 
 
LIST OF PLATES. xvii 
 
 PLATE 
 
 19. Solar Pyrometer, for ascertainiis^g the Temperature 
 
 OF the Solar Surface. 
 
 20. Apparatus for measuring the Radiant Intensity of 
 
 Flames. 
 
 21. Instrument for measuring the Radiation from Incan- 
 
 descent Planes. 
 
 22. Diagrams showing the Radiation at Different Incli- 
 
 nations OF Incandescent Planes. 
 
 23. Instrument for measuring the Radiation from Diffe- 
 
 rent Zones of Incandescent Spheres. 
 
 24. Diagrams showing the Radiation from Different Zones 
 
 OF Incandescent Spheres. 
 
 25. Calorimeter, for measuring the Energy developed by' 
 
 Radiation of Fused Iron. 
 2G. Appar.vius for measuring Radiant Heat by Means of 
 the Thermo-Electric Pile. 
 
 27. Barometric Actinomf.ter, for sieasuring the Intensity 
 
 OF Solar Radiation. 
 
 28. Apparatus for ascertaining the Conductivity of Mer- 
 
 cury. 
 
 29. Concave Spherical Radiator, for testing the Accuracy 
 
 of the Solar Pyrometer. 
 
xviii LIST OF PLATES. 
 
 PLATE 
 
 30. Instrument for measuring the Reflective Power of 
 
 Silver and other Metals. 
 
 31. Rapid-indication Actinometer, for measurinc+ the In- 
 
 tensity OF Solar Radiation. 
 
 32. Apparatus for ascertaining the Diathermancy of 
 
 Flames. 
 
 33. Diagram representing a Section of the Earth and cer- 
 
 tain River Basins. 
 
 34. Dynamic Register for measuring the Relatia'e Power 
 
 OF Currents of Water and Vapor. 
 
 35. Diagram showing the Result of Experiments with the 
 
 Dynamic Register. 
 
 36. Distance Instrument, for measuring Distances at Sea. 
 
 37. Steam Fire-Enchne. 
 
 38. Engines of the " Princeton"— Transverse Section of 
 
 Semi-Cylinders and Piston. 
 
 39. Engines of the " Princeton "—Front Elevation. 
 
 40. Engines of the " Princeton"— Elevation vieaved from 
 
 the Stern. 
 
 41. Twelve-inch Wrought-iron Gun and Carriage. 
 
 42. Iron-clad Cupola Vessel— Side Elevation and Trans- 
 
 verse Section. 
 
LIST OF PLATES. xix 
 
 PLATE 
 
 43. Sl'RFACE COXDF.XSER. 
 
 44. Experimental Calokic Exgine— Traxsverse Section. 
 4,"). Experimental Caloric Exgixe— Loxcitudixal Sectiox. 
 4G. Caloric Exgixe for Domestic Purposes— Loxgitudixal 
 
 Section. 
 
 47. The " Moxitor"— Side Elevation, Deck Plan, and 
 
 Transverse Sectiox of Hull and Turret. 
 
 48. MoxiTOR " Weeiiawken " AT Sea. 
 
 49. MoxiTOR OF THE " Passaic " Class — Side Elevattox 
 
 AXD Tkaxsverse Sectiox of Turret and Pilot- 
 house. 
 !)0. The Moxitor Exgixe — Top View. 
 
 51. The Monitor Exgixe — Front Elevation. 
 
 52. Monitor "Dictator"— Side Elevation axd Deck Plax. 
 
 53. Monitor " Dictator"— Traxsverse Section of Exgixes 
 
 AXD Ship. 
 
 54. Moxitor "Dictator" — Top View of Engines. 
 
 5.'"). Monitor •• Dictator" on the Stocks, prepared for 
 
 Launching. 
 56. The Monitor Turret and the Casemate — Deck Plan of 
 
 A ^^OXITOR WITH Two TURRETS — Df.CK PlAX OF THE 
 
 Turkish Irox-clad "Moyixi Zaffir." 
 
XX LIST OF PLATES. 
 
 PLATE 
 
 57. CaeeiaCtES foe Heavy Oednanoe — Section showing the 
 Feiction Geae applied to the Gun-Caeeiages of the 
 United States Ieon-Clad Fleet — Section showing 
 Captain Scott's Plagiaeism — Section showing Sie 
 William Aemsteong's Plagiaeism. 
 
 SS. Pivot Caeeiage of the Spanish Gunboats. 
 
 59. Rotaey Gun-Caeeiage and Teansit Platfoem applifj) 
 
 to the Spanish Gunboat " Toenado." 
 
 60. Gun-Caeeiage foe Coast Defence. 
 
 61. The Spanish Gunboat Engines. 
 
 62. A New System of Naval Attack. 
 
 63. Movable Toepedo. 
 
 64. Aie-Compressoe, foe the Transmission of Mechanical 
 
 Power — Perspective View. 
 
 65. AlE-COMPRESSOR, FOE THE TeANSMISSION OF MECHANICAL 
 
 Power — Teansverse Section. 
 
 66. Solar Engine, operated by the Inteevention of Steam. 
 
 67. SoLAE Engine, opeeated by the Inteevention of Atmo- 
 
 spheric Air. 
 
CHAPTER I. 
 
 TRANSMISSION OF RADIANT HEAT. 
 
 No phenomenon connected with radiant heat supposed to 
 have been thoroughly investigated is so imperfectly under- 
 stood as its propagation through space. The recognized doc- 
 trine, which asserts that the temperature imparted to substances 
 exposed to radiant heat diminishes in the inverse ratio of the 
 square of the distance from the radiating body, is true only 
 of a sphere of perfectly uniform temperature at the surface, if 
 the distance be computed from the centre of the sjihere. The 
 temperatui-e produced by radiation of spheres which ai'e not 
 uniformly heated at all points of their surface, and of other 
 bodies of whatever form, we have no exact means of ascertain- 
 ing, although the distance and the temperature of the radiating 
 body be accurately known. Nor will it avail if, in addition 
 to the assumed known uniform tempei-ature and accurate 
 knowledge of distance, we also know the dimensions of the 
 radiator. In fine, notwithstanding our knowledge of these 
 elements, an attempt to solve the problem will be fruitless, 
 
2 L'ADTAXT HEAT. chap. I. 
 
 unless, as before stated, the radiation i^'oeeeds from a }<phere 
 of known diameter having a imiform temperature at its suii'ace. 
 According to Melloui's theory, which Pi'ofe.ssor Tyndall and 
 sOme eminent French scientists assure us has been unanswer- 
 ably demonstrated l)y the celebrated Italian, the matter is 
 very simple, namely: the temperature at intermediate points 
 between a heated body and a thermometer placed at a o-iven 
 distance, exposed to, and indicating the intensity of the radiant 
 heat, may be determined by squaring the respective distances 
 from the radiator. The products, it is asserted, will show the 
 inverse ratio of the intensities, and consequently the tempera- 
 ture at the intermediate points, due to their distance from 
 the radiating sni-face. The question being of great practical 
 importance, I have examined the merits of Melloni's mode 
 of establishing the infalliljility of the propounded doctrine 
 that the temjiei'ature inq)arted to bodies exposed to radiant 
 heat are in the inverse ratio of the distances between the 
 radiating surface and those bodies. The illustration shown 
 on the first plate represents an apparatus constructed for the 
 purpose of testing practically whethei' the exjieriment referred 
 to by Professor Tyndall in " Heat as a Mode of Motion " really 
 fui-nishes positive evidence of the correctness of Melloni's 
 theory accepted by the distinguished experimentalist and ])y 
 certain Fi'ench authors, a represents the section of a rect- 
 angular vessel filled with hot water, the face g h being coated 
 with lamp-black. A hollow cone c, lined with black paper 
 on the inside, is secured to a thermo-electric pile c', supported 
 by an appropriate stand sliding in a groove formed at tlie 
 
CHAP. 1. TL'A.XS.yiSt^IOX Of n.llHAXT IfKAT. 3 
 
 to[) of, and iiarallel with, the ceiitie Hue of the table on Avliicli 
 the apparatus is placed. It will be evident that this airaiit^e- 
 ment prevents any change of direction of the axis of the cone 
 which should be right angular to the heated vessel, during 
 the movement of the pile. Melloni asserts that if the cone 
 c be moved towards the heated vessel, say to the position b, 
 the needle of the galvanometer connected with the i)ile will 
 remain stationary. His accepted reasoning defining the law 
 which governs the transmission of radiant heat being based 
 solely on the correctness of this assumption, I have carried 
 out an elaborate series of experiments to ascertain if the 
 needle of the galvanometer remains absolutely stationary as 
 shown by Prof. Tyndall dui'ing the exhibitions at the Royal 
 Institution, mentioned in his work on heat before alluded 
 to. The result establishes the fact most positively that the 
 needle does not remain stationary as assumed, and that the 
 deflection \\hich takes place increases considerably as the pile 
 is advanced towards the heated vessel. That such should be 
 the case will be evident on due reflection. A careful inspec- 
 tion of the diagram shows that, because the points ;/ and d 
 of the i-adiating surface are not equidistant from c', the i-adiant 
 heat which they emit cannot affect the face of the pile alike. 
 The extent of the irregularity may be approximately deter- 
 mined by squaring the distances g c' and d c'. Now, the pro- 
 portion of these distances, as shown by the diagi-am, is 20 
 to 19, hence the radiant heat transmitted to the pile from 
 the points (/ and d will vary in the invei-se ratio of the pro- 
 ducts of the stated numbers, viz. : as 3(51 to 400. In view of 
 
4 FADIANT HEAT. CHAP. I. 
 
 this great diflference of intensity of the heat transmitted to 
 the face of the pile from different points of the radiating 
 surface, the readers of " Heat as a Mode of Motion " may with 
 propriety ask if the intensity indicated by the galvanometer 
 connected with the thermo-electric pile during the reported 
 experiments at the Eoyal Institution was that due to the 
 radiant heat transmitted from g in the line of g c', or that 
 transmitted from d along d c'. 
 
 On theoretical grounds, therefore, Melloni's demonstration 
 is quite unsatisfactory. Several imperfections of a practical 
 nature, insej^arable from the method adopted by the Italian 
 physicist, demand consideration, all tending to aggravate the 
 theoretical defects. The black paper which lines the inside 
 of the hollow cone h, being very light and incapable of reflect- 
 ing the radiant energy transmitted by the heat-rays 1c r and m 
 0, will at once become heated, consequently the air contained 
 within the cone will speedily have its temperature raised by 
 convection. The heat thus received will radiate towards the 
 face of the pile and thereby contribute to disturb the indi- 
 cation, already seriously affected by the difference of tempe- 
 rature consequent on the varying intensities of the radiant 
 heat transmitted. The relative intensity imparted to the black 
 paper within the cone h, and that transmitted to the pile b', 
 may be ascertained by squaring the length of the heat-rays 
 d and d h'. The diagram shows that these lengths are as 
 3 to 5 ; hence, by squaring and inversion, we find that the 
 temperatui'e imparted by the radiant heat at o will be greater 
 than that at b' in the proportion of 25 to 9. Black pajjer 
 
(UAp. I. TI!ANS^^ssI()^^ of l-adtant heat. 5 
 
 being a powei-ful radiator, wbile air has ver\ small specific 
 heat, it will be perceived that the volume within the cone 
 will rapidly become heated by convection, as already stated, 
 and therefore the temperatui'e of the face of the pile will be 
 elevated to a degree far beyond that Avhich it would attain 
 by the direct radiation of the vessel a. Before presenting 
 a demonstration 2>roving the fallacy of Melloni's assertion, 
 that the temperatures imparted to bodies exposed to radiant 
 heat are " iuverseli/ rt6' (he square of the distances hi-txveett, tlie 
 radiating Sfiirface and those bodies,'^ I propose to give a de- 
 tailed statement of the elaborate experiments before adverted 
 to, instituted for the purpose ctf showing practicaUij that the 
 assertion which forms the basis of the accepted doctrine is 
 false, and that the hollow cone c cannot be moved to the 
 position h without increasing the deflection of the needle of 
 the galvanometer connected with the theniio-pile. It has 
 already been stated that, in order to insure pei-fect parallel- 
 ism during the movement, the stand which supports the cone 
 and pile slides in a groove parallel with the table on which 
 the radiating vessel a is placed, and at right angles to the 
 face of the latter. A scale h, divided into inches, is attached 
 to the side of the table for the purpose of showing the distance 
 between the pile and the vessel a, the face of the latter coin- 
 ciding Avith the zero of the scale. The mode of conducting 
 the experiment will be seen by the following brief explanation. 
 The hollow cone c was placed so that the vertical line of the 
 face d corresponded with the last division of the scale, a screen 
 being introduced for the i>urpose of shutting off the radiation 
 
6 BADIANT HEAT. CHAP. I. 
 
 from the blackened face of tlie vessel a. Maintaining the water 
 in the latter at a constant tenaperatnre of 130° above that of 
 the surrounding atmosphere, it Avas found on removing the 
 screen that the needle of the galvanometer moved 4.7 deg. 
 from zero during an intel•^'al of 30 seconds from the moment 
 of exposing the face of the pile to the radiator. The vessel 
 a was then I'emoved and carried into an adjoining room to 
 allow the pile to cool while thus thoroiighly protected from 
 the radiation of the heated vessel. In the meantime, a spirit- 
 lamp was applied under the vessel in order to make good 
 the heat lost during the preceding operation. Sufficient time 
 having elapsed to allow the pile to cool, the vessel was again 
 put in position on the table. The pile being then advanced 
 to the 10th division, and the screen withdrawn, the needle 
 moved 5.05 deg. from zero in the same time as before, viz., 
 30 seconds. The vessel was removed a third time into the 
 adjoining room, the tempei'ature raised to the fixed point, 
 the pile allowed to cool, and the vessel placed in position as 
 before. The pile being now advanced to the 5th division 
 on the scale, and the screen raised, the needle moved 5.55 
 deg. from zero in the stipulated time of 30 seconds. It will 
 be seen then that, so far from remaining stationary, the deflec- 
 tion of the needle of the galvanometer increases very con- 
 siderably as the pile is advanced towards the radiator. The 
 important fact remains to be noticed, that the needle conti- 
 nues to move rapidly after the expiration of 30 seconds from 
 the time of exposing the pile to the radiator ; a circumstance 
 furnishing additional evidence of the unsatisfactory nature of 
 
TL'AM>MJSt^I<>.\ (>/• A'.( /»/.!. VV HEAT. 
 
 Melloui's methoil. In tuder to asceitaiii the exact extent uf 
 deflection of the needle, referred to, two distinct sets of expe- 
 riments were made, the mean result of which is exhibited in 
 the accompanying table. The position of the needle of the 
 
 
 Distance axd Deflection. 
 
 Time. 
 
 
 1 
 
 20 ins. 
 
 10 ins. 
 
 5 ins. 
 
 Stcond). 
 
 I>tg. 
 
 Dv. 
 
 D^. 
 
 80 
 
 4.70 
 
 5.05 
 
 5.55 
 
 60 
 
 4.80 
 
 5.00 
 
 6.75 
 
 90 
 
 4.r)0 
 
 5.70 
 
 8.25 
 
 120 
 
 4.7;-) 
 
 6.25 
 
 10.15 
 
 galvanometer, it will he seen, was recoi'ded at the expiration 
 of 80, 60, i>0, and 120 seconds. A glance at the table proves 
 the correctness of the objections previously raised against the 
 detail of Melloni's arrangement, especially the distui'bing influ- 
 ence of the rays projected against the interior of the cone 
 from points fir-i/on// the ai'ea whose I'adiatioii is supposed alone 
 to affect the pile. Having thus practically shown the fallacy 
 of the assumption on j^hich the Italian physicist bases his 
 doctrine, I will now prove its unsoundness by the i)rocess f)f 
 demonstration. For this purpose let us suppose that an in- 
 candescent cylindrical block I (see illustration Plate 1) com- 
 posed of cast iron, (1 inches in diameter, be suspended at such 
 a height that its axis passes through the centre of the bulb 
 of a thermometei- //i, jield by a bracket in' sliding,' in a groove 
 
8 BADIANT MEAT. chap. i. 
 
 similar to that formed at the opposite end of the table, before 
 described. A scale 7i, divided into equal parts, is attached to 
 the side of the table, the zero of this scale coinciding with 
 the vertical line drawn from the face of the incandescent cylin- 
 drical block I ; while the last division corresponds with the 
 vertical line passing through the centre of the bulb of the 
 thermometer m. Actual trial shows that, if the incandescent 
 iron block be suspended while its temperature is 1,200°, the 
 thermometer exposed to its radiant heat wall indicate 22° above 
 that of the surrounding atmosphere when the block has cooled 
 so far that its differential temperature has become reduced to 
 1,000.° Now, agreeably to Melloni's doctrine, the thermometer 
 m, if advanced to the 10th division of the scale, will indicate 
 
 20' X 22 
 a differential temperature of ; — = 88° ; if advanced to 
 
 ... 20' X 22 
 
 the 5th division, = 352° ; if further advanced to the 
 
 5' 
 
 20' X 22 
 2d division, ^ = 2,200° ; and, lastly, if advanced to the 
 
 1st division on the scale, the thermometer will indicate an 
 
 20' X 22 
 intensity of = 8,800°. The temperature of the incan- 
 descent radiating block being only 1,000°, Ave have thus demon, 
 strated the utter fallacy of Melloni's theory, based on the result 
 of imperfect experiments with hollow cones and thermopiles 
 exposed to radiant heat. In view of the foregoing facts writers 
 on radiant heat will do well to expunge this eiToneous theory 
 from their works. Practical men, not suspecting that the law 
 of inverse squares is inapplicable to short distances, frequently 
 
CHAP. I. TL'Ays.VISSIoS OF I,'AniA.\T HEAT. 9 
 
 coiiiinit serious mustakes in ralculatiug the effect of phuiiii;' 
 iiu-andescent bodies at eeitain distances from structures in- 
 tended to be heated. But Melloni's device, alt]iout,di inca- 
 pable of elucidatin<f the principles wliicli govern the trans- 
 mission of radiant heat, possesses great value, as it proves 
 the correctness of the ai)parently absurd proposition, that an 
 increase of the surface of a radiator is capa1)le of elevating 
 the temperature of a substance exposed to radiant heat with- 
 out increasing the temperature of the radiator. The import- 
 ance of this fact cannot well be overstated. It contradicts 
 ceitain modem sijeculations regarding radiant heat, and it 
 afl'ords a landmark which, tliough it does not point out what 
 we seek, guards against taking the WToug coui"se. 
 
 The defects of Melloni's method, and its inadequacy to 
 solve the problem under consideration, suggest the question : 
 Is it possible to determine, by calculation, what temperature 
 sul)stances will acquire by being placed at various distances 
 from the surface of a heated body of known temperature and 
 dimensions, the sides of which are stiaight and parallel ( "NVe 
 are compelled to admit that the question thus presented cannot 
 be satisfactorily answered, because each point of the radiating 
 surface, in consequence of varying distance, transmits a dif- 
 ferent degree of intensity to the exposed substances. The 
 difficulty of solution is increased on account of the stipulated 
 form ; the inferior intensity transmitted from the cornei-s of 
 the radiator rendering the question exceedingly complicated. 
 A radiating plane surface of circitlar fonn somewhat simplifies 
 the (juestion; yet the difficulty remains of dealing with the 
 
10 RADIANT HEAT. chap. i. 
 
 inferior power of the heat-rays euiauating from the eirciuii- 
 ference. To overcome this difficulty, I have coustructed au 
 instrument — shown Ijy the illustrations on Phite 2, represent- 
 ing a longitudinal section and a perspective view (cojned 
 from a photograph) — in which the radiating surface is con- 
 cave, forming part of a sphere whose centre is situated near 
 the centre of the bulb of the thermometer employed to ascer- 
 tain the intensity of the radiant heat. But tliis expedient of 
 employing a concave spherical surface, every point of which 
 is equidistant from the bulb of the thermometei', prevents 
 any change of distance between the same and the radiator 
 during experiments. I have accordingly introduced four dis- 
 tinct discs of varying spherical concavity, with a thermometer 
 placed in the centre of curvature of eacli. These four con- 
 cave discs forni the sides of an open vessel filled with oil, 
 heated by a gas flame. In order to facilitate comparison, a 
 regular gradation of curvature has been adopted ; the radius 
 of the deepest concavity being 3 in., the next being 6 in., 
 then 9 in., and, lastly, 12 in. for the least concavity. It will 
 be evident that, owing to the difference of curvature, each 
 disc mil present a different extent of radiating surface, if a 
 uniform size be employed, thereby rendering comparisons be- 
 tween the indicated intensities quite laborious. To obviate 
 this, the diameters of the several discs have been so propor- 
 tioned that the superficial measurement of each is precisely 
 alike. The disc of the least curvature is 4 in. in diameter; 
 the remaining three being respectively 3.982 in., 3.939 in., and 
 3.777 in. diameter. Much pains has been bestowed on the 
 
CHAP. I. TRAXSMISSloy OF EADrAXT HEAT. 1] 
 
 workmanship in order to obtain concavities of precisely equal 
 ai-eas. The ulijection.s raised iu subsequent chapters against 
 conducting experiments with the solar calorimeter and actiuo- 
 iiictcr in the presence of the disturltiii^' inlluence of the sur- 
 rounding air, apply witli equal force to the apparatus now 
 being considered. Accordingly, the heater, with its four con- 
 cave radiating discs and thermometers, are enclosed in an 
 exhausted cliamber, as will be seen by reference to the longi- 
 tudinal section. Evidently, the heat reflected by the sides of 
 this chamber towards the bulbs of the enclosed therniome- 
 teis, will determine their zero precisely as iu the instiuments 
 referred to; hence, it is indispensable that this chamber should 
 be maintained at a constant temperature. This is effected ])v 
 surrounding the same with an external ojjeu vessel filled with 
 water kept at a tempei'ature of (50 deg. Fahr. during experi- 
 ments. The object of adopting 60 deg. is that of facilitating 
 comparisons with solar intensity, the zero of my actinometer 
 being also (iO deg. above Fahrenheit's zero. Let us now con- 
 sider briefly the leading properties of the illustrated device. 
 In the first place, the radiating concave surfaces, as well as 
 the liulbs of the thermometeis, are surrounded by the ether 
 alone, and thei'efore cannot suffer any loss of heat by con- 
 vection. SecDudly, the intensity of the radiant heat received 
 from all i)oints of each concave surface will be precisely 
 alike, because those points are equidistant from their respec- 
 tive thermometers. Thirdly, the concave surfaces presenting 
 e(|ual areas, being composed of the same materials of uiiilonu 
 thickness, and beiuLf heated bv the same medium, will emit 
 
12 JiADTAXT HFAT. CHAP. I. 
 
 an eqiinl amount of licat. Fourtlily, as the several thei'nio- 
 meters are exjioscd to tlie radiation of a surrounding vessel 
 kept at 60 deg., they ^vill acquire that temperature before 
 fire is a})plied to the heater. In consequence of this, the 
 increase of temperature, after heating, will furnish a ti-ue com- 
 parative measure of the energy of the radiant heat transmitted 
 from the concave radiators to tlie bulbs of the thennometers. 
 
 The prevailing opinion that there is a concentration of heat 
 in the focus of spherical radiators, will be urged as an objection 
 against the described method of measuring radiant heat. A 
 careful examination of this question -svill therefore be neces- 
 sary. Fig. 4, Plate ?>, represents a concave spherical radiator, 
 being the focus, o c the axis, and a h a section of the I'adiator. 
 According to the accepted theory of radiation, rays of heat 
 are projected in all directions from every point of the I'adiat- 
 ing surface. In order, therefore, to demonstrate that there is 
 no concentration of heat in the focus, we have merely to draw 
 radial lines representing the heat-rays from points d and /' 
 on the concave surface, as shown on the diagram. It will 
 be seen on close examination that there is only a single ray 
 d c emitted from the point d, which is directed towards the 
 focus c ; the ray /' c being the only one directed from the 
 point/'. The other heat-rays from the points d and/' diverge 
 in all directions, and intersect every part of the field I' 1 ; 
 thus dispersing the radiant heat nearly uniformly over a very 
 large surface. The curve c >i, struck from the point _/', clearly 
 shows that all the heat-i'ays below c, projected from f, are 
 shoi'ter than those dii'ected to the forns from the same point, 
 
cnAP. I. TPA^'s^fTf!fi^o^' of i^adtaxt heat. 13 
 
 and tlu'ivfore impart a hiijJnr tcinpciatiin- to the plane h I 
 than that transmitted to the focus. It will be needless to 
 enter on a detailed computation of the temperature at the 
 intersections of the vai'ious I'ays with the plane h 1, as a 
 mere inspection of the diagram distinctly shows that the 
 focus c receives no increase of heat on account of being the 
 centre of the spherical radiator. Indeed, if we snl>stitute a 
 plane circular disc extending from <i to /', a greater amount 
 of heat will be transmitted to a thermometer placed at <\ 
 than w ith the spherical surface represented ; the only impoi-- 
 tant difference between a plane and a concave i-adiator being 
 that a thei'monieter jilaced in the focus of the latter receives 
 an equal degree of heat from each point of the concave surface. 
 Investigations conducted by means of tlie illustrated instni- 
 nient prove that the intensity of heat transmitted by the radi- 
 ation of concave spliciical suii'aces of eipial temperature, pre- 
 senting equal areas anil having ditt'ei'ent curvature, is in the 
 inverse ratio of the square of their ladii, pi-ovided the sub- 
 stance which receives the radiant heat is placed in the focus 
 of the radiating sui'face. This impoitant fact in coiiiiectiou 
 with the jtrevious demonstrations showing the fallacy of the 
 propounded doctrine that the intensity of ladiation diminishes 
 in the invei-se ratio of the square of the distance, between 
 jdane radiatoi-s and the recijiients of their radiant heat, enables 
 us now to consider the proposition contained in our inti-oduc- 
 tory remarks, viz., that the assumed law is true only of a 
 .sphere of jierfectly uniform temperature at the surface, if the 
 distance to the recipient of the radiating heat be couq)Utcd 
 
14 RADIANT HEAT. CHAP. I. 
 
 from the centre of the hejited sphere. Suppose that .s, Fig. 5, 
 represents a solid s])here of metal maintained at a constant 
 temperature and suspended in the centre of an exhausted 
 metallic s])herical vessel v, the size of which is much greater 
 tlian that of the solid sphere; c a and c h being radial lines 
 dra\vn from the mutual centre c to the circuinference of tlie 
 exhaiisted vessel. The heated sphere being maintained at a 
 unifoi'm temperature -sA'hile the spherical vessel is exhausted, 
 and thereby freed from any disturbing influence of internal 
 currents of air, it will be evident that the intensity of radiant 
 heat, thus transmitted from the sphere through the ether alone, 
 will be alike at every point of the surface of the sui-rounding 
 sjiherical vessel. It needs no demonstration to prove that the 
 temperature resulting from radiation against the latter will 
 be less than that of the central sphere in proportion to the 
 area pi'esented by each. In other words, tie temperature will 
 he inversely as the areas of the tioo spheres. On the soundness 
 of this proposition depends the correctness of the accepted doc- 
 trine I'elating to the propagation of heat and light through 
 space. The remarkable fact niTist not be overlooked, that 
 this projiosition takes no cognizance of distance ; and that it 
 enables us to determine the temperature of the surrounding 
 sphere if we know the temperature of the central one, merely 
 l)y comparing their relative areas, basing our computation on 
 the proportion they bear to each other; the intervening space, 
 whether it be 1 mile, or 1,000 miles, being an element excluded 
 from our calculation. 
 
 That rays of light and heat meet with no appreciable 
 
CHAP. I. TL'A.\SM1SSI<)X OF J:AI)IA.\T HEAT. lo 
 
 resistance in their passage through the etlier is an irresistible 
 inference to be drawn from the fact that the intensity bears 
 a direct pi-oportion to the area over which the rays are di.s- 
 ]K-rt<ed. The conuuoii expression that "tlie intensity of light 
 and heat varies inversely as the square of the distances " leads 
 to the supposition that distance is an element to be taken into 
 account in estimating the intensity of radiant heat. That this 
 is an error will be readily })erceive(l if we reHect on the fact, 
 just referred to, that the intensity of the rays diminishes in 
 the exact pioportion to the areas over which they are dis- 
 jiersed. It is self-evident that if distance, per se, in any way 
 att'ected the question, the intensities could not thus depend 
 solely on the extent of the areas over which the rays are 
 diffused. Much misapprehension would be prevented if the 
 law relating to radiant heat were thus expi-essed : l^ie inttii- 
 sities are inversely as the areas over which the rays are dis- 
 persed. Sir Isaac Newton, referring to the intensity of the 
 sun's radiant heat at dift'erent distances, thus clearly defines 
 the question : " The heat of the sun (at various distances) 
 is as the density of its rays." He also states that the den- 
 sities of the diverging rays are " reciprocally as the square 
 of the distance from the centre of the sun," a fact which 
 obviously has nothing to do with the main proposition, as 
 it simply results from certain geometrical relations, viz., that 
 the areas of transverse sections of a cone are as the square 
 of the distances from the apex. 
 
 I will now briefly show why the theory of Newton, sup- 
 posed to have been proved by Melloni, is true, under certain 
 
16 BADIANT HEAT. chap. i. 
 
 conditions, as regards a uniformly lieated sphere, although 
 fallacious, under all circumstances, as regards plane surfaces. 
 Referring again to Fig. 5, let us assume that the sphere ,s 
 represents the sun, the semi-diameter of ^vhicll is 426,2U2 
 miles ; and that a b v rejjresents the earth's orbit reduced to 
 a circle whose radius corresjiouds Avith the earth's mean dis- 
 tance from the sun's centre, viz., 91,430,000 miles. Now, the 
 arc e f is to the arc a h as the radial line <: e, rejjresenting the 
 sun's radius, is to c a, re2:)resenting the earth's distance from 
 the sun's centre. According to the stated dimensions, there- 
 fore, c e is to c a as 1 to 214.44:; hence, regarding a c h as a 
 cone of very small base, the area of e f is to that of a b as 1 to 
 the square of 214.44 = 45,984. Accordingly, the temperature 
 of the heated central sphere « will be 45,984 times greater 
 than the temi^erature which it imparts by radiation to the 
 sui'rounding sphere a b v. 
 
 We have now fully established the fact that, although the 
 extent of the radiating and recipient surfaces, in connection 
 with the temj^erature of the former, determines the tempera- 
 ture transmitted to the latter, wholly independent of distance, 
 yet, as regards the sun, it is true that the temperature pro- 
 duced by the radiant heat of his rays is inversely as the 
 square of the distance from his centre, the same law applying 
 to all spheres having a uniform temperature at the surface. 
 
CHAPTER 11. 
 
 EADIATIOX AT DIFFERENT TEMPERATURES. 
 
 SiK Isaac Newtox supposed that a lieated body Avould 
 lose by radiation at each instant a quantity of heat propor- 
 tionate to the excess of its temperature above that of the sur- 
 rounding medium. Modern physicists, basing their reasoning 
 on the result of MM. Dulong and Petit's experiments, contend 
 that Newton's supposition is erroneous. Prof. Balfour Ste\vart 
 in his " Elementary Treatise on Heat," published at Oxford 
 1871, in support of his assertion that Newton's doctrine is 
 false, presents the following extract from the work of Duloug 
 and Petit : 
 
 Excess of Temperature of 
 
 the Thermometer. Velocity of Cooling. 
 
 °c. ° c. 
 
 240 10.69 
 
 220 8.81 
 
 200 7.40 
 
18 
 
 BADIANT HEAT. 
 
 
 CHAP. II. 
 
 Excess of Temperature of 
 
 the Thermometer. Veloc 
 
 ity of Cooling. 
 
 
 °c. 
 
 °c. 
 
 
 180 
 
 6.10 
 
 
 160 
 
 4.89 
 
 
 140 
 
 3.88 
 
 
 120 
 
 3.02 
 
 
 100 
 
 2.30 
 
 ' 
 
 80 
 
 1.74 
 
 
 "We see at once from this table," says the author of the 
 " Elementary Treatise on Heat," " that the law of Newton 
 does not hold, for according to it the velocity of cooling for 
 an excess of 200 deg. should be precisely double of that for 
 an excess of 100 deg. ; now we find that it is more than three 
 times as much." The rate of cooling exhibited in this table 
 being at variance with the laws which govern the emission 
 and absorption of radiant heat, I have instituted a series of 
 experiments in order to test the correctness of the tabulated 
 rates accepted by Prof. Stewart. The initiary proceeding of 
 the investigation was that of ascertaining the rate of cooling 
 at temperatures beloAV 200° F., for which purpose I constructed 
 the apparatus shown on PI. 4, Fig. 1, representing three concen- 
 tric spherical vessels placed within each other. As the draw- 
 ing shows the detail of the entire naechanisra, a brief descrip- 
 tion will suffice, a « is a spherical vessel 7 inches in diameter, 
 applied within an exterior casing h. A spherical radiator r, 
 4 inches in diameter, composed of very thin copper, and coated 
 with lamp-black, is sustained in the centre of the spherical 
 vessel a by means of two vertical tubes I and w, composed 
 
CHAP. 11. liAJjJATioy AT diffjjuent tempekatures. ly 
 
 of very thin metal, coramuuicatiug with the interior of the 
 radiator. The upper tube is large enough to admit the bulb 
 of a thermometer, the lower one being only sufficiently large 
 to accommodate a small axle, to \vhich is attached a paddle- 
 wheel provided with curved paddles arranged in such a manner 
 that the bulb of the thermometer may be inserted consider- 
 ably beyond the centre of the sphere, as shown in the illus- 
 ti'ation. The external casing b is provided with nozzles g 
 and d, to which tubes are attached for circulating cold water 
 through the intervening space during experiments. The air is 
 exhausted from the spherical enclosure a a through the tube 
 I; which passes across the said intervening space. The radi- 
 ating sphere c being filled with water, it will be perceived 
 that the centrifugal action of the paddles of the wheel applied 
 within ^vill produce a continuous current from the centre to- 
 wards the circumference, the fluid successively passing over 
 and coming in contact with the thin spherical shell, then 
 returning to the centre to be again thrown off by the centri- 
 fugal action. The rotary motion of the water thus kept up 
 without inteiTnission round the cylindrical bulb of the ther- 
 mometer, will evidently render its indication prompt and reli- 
 able. It is hardly necessary to observe that the rajiid pre- 
 sentation of fresh particles of A\'ater, promoted by the action 
 of the paddles, will eflfectually prevent the reduction of tem- 
 perature proceeding faster at the circumference than at the 
 centre ; hence the radiation at the surface Avill, in virtue of 
 the continuous interchange of particles, affect almost simul- 
 taneously eveiy molecule within the sphere. It will be seen, 
 
^0 RADIANT HEAT. CHAP. ir. 
 
 therefore, that the total energy of radiation will be rendered 
 availaljle in i-edueing the temperature of the contents of the 
 radiator, and tliat the central thermometer Avill indicate, at 
 every instant, the precise degree of temperature of the entire 
 mass. It might be sujiposed that the motion of the watei- 
 within the sphere, consequent on the action of the paddle- 
 wheel, Avould produce an elevation of temperature sufficient to 
 render the indication of the thermometer inaccurate. Before 
 commencing the experiments, several trials were accordingly 
 instituted to ascertain if at the requisite speed, 20 turns per 
 minute, heat could be produced. The result has proved posi- 
 tively that no appreciable elevation of temperatui-e talces j^lace. 
 The diameter of the wheel being 3.4 ins., the maximum speed 
 of the particles of water produced by the rotation scarcely 
 reaches 3.5 ins. per second, a velocity evidently too small to 
 generate appreciable heat. 
 
 The mode of conducting the experiment Avill be seen by 
 the following statement : A capacious wooden cistern, charged 
 with water and crushed ice, is connected by small flexible 
 tubes to the nozzles g and d on opposite sides of the external 
 vessel h, a pump being applied between the said nozzles and 
 the cistern, by means of which the cold water is forced through 
 the apparatus and ultimately returned to the source of supply. 
 
 In view of the great importance of the question at issue 
 the investigation has been conducted with the utmost care, 
 four operators having been employed to cany out the experi- 
 ments, th.e labor being thus divided : 1st operator regulates 
 the temperature of the water in the cistern by continual agita- 
 
CHAP. II. RADIATION AT DIFFEREyr TEMPERATURES. 21 
 
 tiou and supply of crushed ice from time to time ; 2d operator 
 works the pump at a uniform rate ; 3d operator, after having 
 charged the central .sphere with boiling water, turns the paddle- 
 wheel, reads the thermometei", and announces the temperature 
 for each degree, at the instant when the toj) of the mercui-ial 
 column is covered by half, the thickness of the line on the 
 scale ; the 4th operator, provided with a Casella chronograph, 
 records the time and temperature. It should be ol)sen'ed 
 that, notwithstanding this procedure, there is a slight ii-re- 
 gularity in the ratio of temperature and time, the increment 
 for each degree not being quite unifoi-m. Obviously the most 
 practised eye, though assisted by the raagnifying-glass, can- 
 not detennine exactly at what moment the top of the falling 
 column is half covered by the line on the thermometric scale. 
 Again, a pei-fectly graduated thermometer cannot be obtained. 
 The discrepancy referred to is, howevei', so small that it can- 
 not readily be detected in the diagram. It will lie proper 
 to state that, in constnicting the several tables contained in 
 this chapter, a correction has been introduced agreeably to 
 Regnault's method of correcting the irregularities inseparable 
 from thermometric observations. Referring to the anne.xed 
 tables, marked A, it will be noticed that the rate at which 
 the spherical radiator cools has been recorded for each degree 
 of differential temperature from 150' to 15\ The tem])era- 
 ture of the surrounding cold vessel, it will be .«een, is entered 
 in the first column ; the temperature of the radiating sphere 
 in the second, and its e.xcess of temperatuie in the third 
 column. The number of seconds occupied in rftliiciiig the 
 
32 
 
 RADIANT BEAT. 
 
 A 
 
 Table showing the Eate 
 
 OF Cooling 
 
 OF A 
 
 Heated Body 
 
 
 SUSPENDED WITHIN A C'of.D EkCLOSURE. 
 
 
 Is 
 
 g o 
 
 a"? 
 
 Is 
 
 EH !a 
 o 
 
 
 111 
 i 1 § 
 
 0-" 
 
 B 
 II 
 
 1-1 
 
 11 
 
 o 3 
 
 "3 >;> 
 P . 
 
 "go 
 
 o 
 
 ^■^ 
 
 3 ^ 
 £ c 
 
 bs,Q 
 03 > 
 
 II ■ 
 
 •JaA. 
 
 °Fah. 
 
 °Fah. 
 
 Sees. 
 
 &(». 
 
 •Fah. 
 
 °Fah. 
 
 °Fah. 
 
 Sece- 
 
 Sees. 
 
 33 
 
 183 
 
 150 
 
 37.7 
 
 42.01 
 
 33 
 
 160 
 
 127 
 
 47.3 
 
 49.64 
 
 H3 
 
 182 
 
 149 
 
 38.1 
 
 42.29 
 
 33 
 
 159 
 
 120 
 
 47.8 
 
 50.04 
 
 33 
 
 181 
 
 148 
 
 38.5 
 
 42.57 
 
 33 
 
 158 
 
 ]25 
 
 48.3 
 
 50.44 
 
 33 
 
 180 
 
 147 
 
 38.9 
 
 42.86 
 
 33 
 
 157 
 
 124 
 
 48.8 
 
 50.85 
 
 33 
 
 179 
 
 146 
 
 39.3 
 
 43.16 
 
 33 
 
 156 
 
 123 
 
 49.3 
 
 51.26 
 
 33 
 
 178 
 
 145 
 
 39.7 
 
 43.46 
 
 33 
 
 155 
 
 122 
 
 49.8 
 
 51.68 
 
 33 
 
 177 
 
 144 
 
 40.1 
 
 43.76 
 
 33 
 
 154 
 
 121 
 
 50.3 
 
 52.11 
 
 33 
 
 - 176 
 
 143 
 
 40.5 
 
 44.07 
 
 33 
 
 153 
 
 120 
 
 50.8 
 
 52.55 
 
 33 
 
 175 
 
 142 
 
 40.9 
 
 44.38 
 
 33 
 
 152 
 
 119 
 
 51.3 
 
 53.00 
 
 33 
 
 174 
 
 141 
 
 41.3 
 
 44.69 
 
 33 
 
 151 
 
 118 
 
 51.8 
 
 53.44 
 
 33 
 
 173 
 
 140 
 
 41.7 
 
 45.02 
 
 33 
 
 150 
 
 117 
 
 52.3 
 
 53.90 
 
 33 
 
 172 
 
 139 
 
 42.1 
 
 45.34 
 
 33 
 
 149 
 
 116 
 
 52.8 
 
 54.37 
 
 33 
 
 171 
 
 138 
 
 42.5 
 
 45.67 
 
 33 
 
 148 
 
 115 
 
 53.3 
 
 54.84 
 
 33 
 
 170 
 
 137 
 
 42.9 
 
 46.01 
 
 33 
 
 147 
 
 114 
 
 53.9 
 
 55.33 
 
 33 
 
 169 
 
 136 
 
 43.3 
 
 46.34 
 
 33 
 
 146 
 
 113 
 
 54.5 
 
 55.82 
 
 33 
 
 168 
 
 135 
 
 43.7 
 
 46.69 
 
 33 
 
 145 
 
 112 
 
 55.1 
 
 50.32 
 
 33 
 
 167 
 
 134 
 
 44.1 
 
 47.04 
 
 33 
 
 144 
 
 ni 
 
 55.7 
 
 56.83 
 
 33 
 
 166 
 
 133 
 
 44.5 
 
 47.39 
 
 33 
 
 143 
 
 110 
 
 56.3 
 
 57.35 
 
 33 
 
 165 
 
 132 
 
 44.9 
 
 47.75 
 
 33 
 
 142 
 
 109 
 
 56.9 
 
 57.87 
 
 33 
 
 164 
 
 131 
 
 45.3 
 
 48.12 
 
 33 
 
 141 
 
 108 
 
 57.5 
 
 58.42 
 
 33 
 
 163 
 
 130 
 
 45.8 
 
 48.49 
 
 33 
 
 140 
 
 107 
 
 58.2 
 
 58.97 
 
 33 
 
 102 
 
 129 
 
 46.3 
 
 48.87 
 
 33 
 
 139 
 
 106 
 
 58.9 
 
 59.52 
 
 33 
 
 161 
 
 128 
 
 46.8 
 
 49.25 
 
 ; 33 
 
 1 
 
 138 
 
 105 
 
 59.6 
 
 60.09 
 
Cll.vi'. II. liADlATlON AT DlFFEliEM TEMrEliATURES. 
 
 A 
 
 Table .miowino thk Kate 
 
 or Coo LI. NO 
 
 OF A 
 
 Heated Body 
 
 
 SUSPENDED WITHIN A COLU ExCLOSVlit:. 
 
 
 o 
 
 So 
 
 P 
 
 11 
 
 "3 
 
 O 3 
 M ■" 
 'A 
 
 111 
 
 Hi 
 Is 
 
 o- 
 
 S 
 
 So - 
 
 ■s 
 
 11 
 
 2.'% 
 
 as 
 
 i1 
 
 S.1 
 
 Observed time cool- 
 ing enclosed body 
 one deg. 
 
 2 
 
 ll 
 11 
 
 'Fah. 
 
 'Fak 
 
 'Fah. 
 
 Ste». 
 
 Sea. 
 
 •Fah. 
 
 'Fah. 
 
 'Fah. 
 
 Sta. 
 
 Ste». 
 
 33 
 
 137 
 
 104 
 
 60.3 
 
 60.67 
 
 33 
 
 114 
 
 81 
 
 80.8 
 
 78.01 
 
 33 
 
 136 
 
 103 
 
 61.0 
 
 61.27 
 
 33 
 
 113 
 
 80 
 
 81.9 
 
 78.99 
 
 33 
 
 135 
 
 102 
 
 61.7 
 
 61.87 
 
 33 
 
 112 
 
 79 
 
 83.1 
 
 80.00 
 
 33 
 
 134 
 
 101 
 
 62.4 
 
 62.48 
 
 33 
 
 111 
 
 78 
 
 84.3 
 
 81.03 
 
 33 
 
 133 
 
 100 
 
 63.2 
 
 63.11 
 
 33 
 
 110 
 
 77 
 
 85.5 
 
 82.09 
 
 33 
 
 132 
 
 99 
 
 64.0 
 
 63.76 
 
 33 
 
 109 
 
 76 
 
 86.7 
 
 83.18 
 
 33 
 
 131 
 
 98 
 
 64.8 
 
 64.41 
 
 33 
 
 108 
 
 75 
 
 88.0 
 
 84.29 
 
 33 
 
 130 
 
 97 
 
 65.6 
 
 65.08 
 
 33 
 
 107 
 
 74 
 
 89.3 
 
 85.44 
 
 33 
 
 129 
 
 96 
 
 66.4 
 
 65.76 
 
 33 
 
 106 
 
 73 
 
 90.6 
 
 86.62 
 
 33 
 
 128 
 
 95 
 
 67.2 
 
 66.46 
 
 33 
 
 105 
 
 72 
 
 91.9 
 
 87.83 
 
 33 
 
 127 
 
 94 
 
 68.0 
 
 67.16 
 
 33 
 
 104 
 
 71 
 
 93.3 
 
 89.08 
 
 33 
 
 126 
 
 93 
 
 68.9 
 
 67.89 
 
 33 
 
 103 
 
 70 
 
 94.7 
 
 90.36 
 
 33 
 
 125 
 
 92 
 
 69.8 
 
 68.63 
 
 33 
 
 102 
 
 69 
 
 96.1 
 
 91.68 
 
 33 
 
 124 
 
 91 
 
 70.7 
 
 69.39 
 
 33 
 
 101 
 
 68 
 
 97.5 
 
 93.04 
 
 33 
 
 123 
 
 90 
 
 71.6 
 
 70.17 
 
 33 
 
 100 
 
 67 
 
 99.0 
 
 94.44 
 
 33 
 
 122 
 
 89 
 
 72.5 
 
 70.96 
 
 33 
 
 99 
 
 66 
 
 100.5 
 
 95.88 
 
 33 
 
 121 
 
 88 
 
 73.5 
 
 71.77 
 
 33 
 
 98 
 
 65 
 
 102.0 
 
 97.36 
 
 33 
 
 120 
 
 87 
 
 74.5 
 
 72.60 
 
 33 
 
 97 
 
 64 
 
 103.6 
 
 98.89 
 
 33 
 
 119 
 
 86 
 
 75.5 
 
 73.45 
 
 33 
 
 96 
 
 63 
 
 10i).2 
 
 100.48 
 
 33 
 
 118 
 
 85 
 
 76.5 
 
 74.32 
 
 33 
 
 95 
 
 62 
 
 106.9 
 
 102.11 
 
 33 
 
 117 
 
 84 
 
 77.5 
 
 75.21 
 
 33 
 
 94 
 
 61 
 
 108.6 
 
 103.80 
 
 33 
 
 116 
 
 83 
 
 78.6 
 
 76.12 
 
 33 
 
 93 
 
 60 
 
 110.4 
 
 105.55 
 
 33 
 
 115 
 
 82 
 
 79.7 
 
 77.11.) 
 
 33 
 
 92 
 
 59 
 
 112.:! 
 
 107.35 
 
24 
 
 BADIAM HEAT. 
 
 A 
 
 Tahle showing tue Rate 
 
 OF Cooling 
 
 of a 
 
 Heated Body 
 
 
 SUSPENDED WITHIN A COLD ENCLOSURE. 
 
 
 "8 
 
 is 
 P 
 
 li 
 
 c 
 
 k 
 
 1- 
 
 It 
 
 -S 
 
 ii 
 
 MO 
 
 ail 
 
 H 
 
 O 
 
 a a> 
 
 £'g 
 
 l1 
 
 CO 
 
 £13 
 £ 
 H I- 
 
 11 
 
 ii 
 
 CO 
 
 o- 
 
 -S 
 
 .be 
 ^^ 
 
 l| 
 
 ' Fnh. 
 
 'Fall. 
 
 'Fah. 
 
 Sees. 
 
 Sees. 
 
 'Fah. 
 
 'Fah. 
 
 'Fah. 
 
 Secs. 
 
 Sees. 
 
 33 
 
 91 
 
 58 
 
 114.3 
 
 109.22 
 
 33 
 
 69 
 
 36 
 
 189.9 
 
 176.90 
 
 33 
 
 90 
 
 57 
 
 116.4 
 
 111.15 
 
 33 
 
 68 
 
 35 
 
 195.4 
 
 182.03 
 
 33 
 
 89 
 
 56 
 
 118.6 
 
 113.15 
 
 33 
 
 67 
 
 34 
 
 201.2 
 
 187.46 
 
 33 
 
 88 
 
 55 
 
 120.9 
 
 115.23 
 
 33 
 
 66 
 
 33 
 
 207.2 
 
 193.23 
 
 33 
 
 87 
 
 54 
 
 123.3 
 
 117.38 
 
 33 
 
 65 
 
 32 
 
 213.5 
 
 199.36 
 
 33 
 
 86 
 
 53 
 
 125.8 
 
 119.62 
 
 33 
 
 64 
 
 31 
 
 220.2 
 
 205.90 
 
 33 
 
 85 
 
 52 
 
 128.4 
 
 121.94 
 
 33 
 
 63 
 
 30 
 
 227.4 
 
 212.88 
 
 33 
 
 84 
 
 51 
 
 131.1 
 
 124.35 
 
 33 
 
 62 
 
 29 
 
 235.1 
 
 220.35 
 
 33 
 
 S3 
 
 50 
 
 133.9 
 
 126.87 
 
 3:3 
 
 61 
 
 28 
 
 243.4 
 
 228.36 
 
 33 
 
 82 
 
 49 
 
 136.8 
 
 129.48 
 
 33 
 
 60 
 
 27 
 
 252.3 
 
 236.98 
 
 33 
 
 81 
 
 48 
 
 139.9 
 
 132.21 
 
 33 
 
 59 
 
 26 
 
 261.9 
 
 246.27 
 
 33 
 
 80 
 
 47 
 
 143.1 
 
 135.05 
 
 33 
 
 58 
 
 25 
 
 272.2 
 
 256.32 
 
 33 
 
 79 
 
 46 
 
 146.5 
 
 138.02 
 
 33 
 
 57 
 
 24 
 
 283.3 
 
 267.23 
 
 33 
 
 78 
 
 45 
 
 150.0 
 
 141.12 
 
 33 
 
 56 
 
 23 
 
 295.4 
 
 279.11 
 
 33 
 
 77 
 
 44 
 
 153.7 
 
 144.37 
 
 33 
 
 55 
 
 22 
 
 308.7 
 
 292.09 
 
 33 
 
 76 
 
 43 
 
 157.5 
 
 147.76 
 
 33 
 
 54 
 
 21 
 
 323.4 
 
 306.34 
 
 33 
 
 75 
 
 42 
 
 161.5 
 
 151.32 
 
 33 
 
 63 
 
 20 
 
 339.7 
 
 322.05 
 
 33 
 
 74 
 
 41 
 
 165.7 
 
 155.06 
 
 33 
 
 52 
 
 19 
 
 357.9 
 
 339.46 
 
 33 
 
 73 
 
 40 
 
 170.1 
 
 158.99 
 
 33 
 
 51 
 
 18 
 
 378.3 
 
 358.86 
 
 33 
 
 72 
 
 39 
 
 174.7 
 
 163.12 
 
 33 
 
 50 
 
 17 
 
 401.2 
 
 380.60 
 
 33 
 
 71 
 
 38 
 
 179.5 
 
 167.47 
 
 33 
 
 49 
 
 16 
 
 426.9 
 
 405.16 
 
 33 
 
 70 
 
 37 
 
 184.6 
 
 172.06 
 
 33 
 
 48 
 
 15 
 
 455.7 
 
 433.10 
 
CHAP. II. liADlATIOX AT l>lFl'EliEM TEMl'imATVliES. '^'^ 
 
 temperature of the radiating spliere one degree is rec(jrdeil 
 iu the fourth column. The fifth column contains the number 
 of second.s which A\x)uld be requisite to reduce the tempera- 
 ture one degree if the cooling proceeded at the rate shown 
 by the Newtonian law. 
 
 Let ns now examine the diagram Fig. 2, attached to the 
 illustration, in whieli the length of the ordinates of the curve 
 b c represent the observed time for each degree of differential 
 temperature, while the ordinates of the cun^e a d represent 
 the time that would elapse if the rate of cooling were in exact 
 accordance with Newton's doctrine — namely, if the times were 
 inversely as the differential temperatures. The vertical line 
 in Fig. 2, on which the ordinates representing the time of 
 cooling liave been projected, is divided into degrees of Fah- 
 renheit, showing the differential temi^erature, viz., the excess 
 of temperature of the radiating sphere above that of the sur- 
 rounding cold vessel. A careful inspection of this diagram and 
 of the Tables A, rendei-s argument unnecessarj^ to show that 
 our experimental investigation has established the correctness 
 of Newton's assumption that a radiating body loses at each 
 instant a quantity of heat proportionate to the excess of its 
 temperature above that of the surrounding medium. It will 
 be sho-wn hereafter that the slight discrepancy indicated by 
 the different length of the ordinates of the curves a d and 
 i c is owing to the variation of emissive power of the radiator, 
 caused by the difference of molecular motion resulting fi-om 
 change of tempeiature, and the consequent change of dimen- 
 sions, of the radiator. Dulong and Petit's formula being based 
 
26 
 
 BABIANT HEAT. 
 
 on tlie table presented at the commencement of this chapter, 
 copied from Prof. Stewart's "Elementary Treatise on Heat" 
 (compared also with the French original), I have deemed it 
 important to examine carefully whether the rates of cooling 
 presented in the said table arQ consistent. The result of this 
 examination is shown in the accompanying table, in which 
 the 1st column contains the differential temperature of the 
 radiator, the 2d column contains the corresponding rate of 
 cooling for each minute, established by Duloug and Petit ; 
 while the 3d and 4th columns show the ratio of difference. 
 
 240° 
 
 10°. 69 
 
 \ 1.88 
 
 
 220 
 
 8.81 
 
 \ 1.41 
 
 \ 0.47 
 
 200 
 
 7.40 
 
 1 1.30 
 
 j 0.11 
 
 180 
 
 6.10 
 
 ( 1.21 
 
 \ 0.09 
 
 160 
 
 4.89 
 
 ( 1.01 
 
 ( 0.20 
 
 140 
 
 3.88 
 
 ( 0.86 
 
 I 0.15 
 
 120 
 
 3.02 
 
 ( 0.72 
 
 I 0.14 
 
 100 
 
 2.30 
 
 j 0.56 
 
 ; 0.16 
 
 80 
 
 1.74 
 
 
 
 1 
 
 2 
 
 3 
 
 4 
 
 The inconsistency and irregularity of the rates of cooling exhi- 
 bited by the figures in the two last columns prove the um-e- 
 liable character of the temperatures inserted in the two first 
 columns. We are warranted in concluding that a doctrine 
 
CHAP. II. liADIATIoy AT DlFb'EHEST TKMVKUATUREa. 
 
 2: 
 
 B 
 
 Table siiowixc; tiik Rate of Coolint. of a Heatkd IJody 
 
 SUSPENDED WITUIN A VOLD EnCI.OSUKK. 
 
 .S-3 
 
 r 
 
 ^•5 
 
 3 =8 
 ll 
 
 — B 
 B O O 
 
 > g 
 
 S to 
 
 .a c 
 0-" 
 
 i>? 
 
 ."2 
 
 E S 
 
 £1 
 
 c g 
 
 S'3 Sf 
 
 
 &e«. 
 
 Cent. 
 
 Ste». 
 
 See*. 
 
 Suit. 
 
 83 
 
 82 
 
 81 
 
 80 
 
 79 
 
 78 
 
 77 
 
 76 
 
 75 
 
 74 
 
 73 
 
 72 
 
 71 
 
 70 
 
 GO 
 
 68 
 
 67 
 
 66 
 
 65 
 
 64 
 
 63 
 
 62 
 
 61 
 
 60 
 
 59 
 
 58 
 
 38.1 
 
 38.8 
 
 39.5 
 
 40.3 
 
 41.0 
 
 41.7 
 
 42.4 
 
 43.1 
 
 43.8 
 
 44.5 
 
 45.3 
 
 46.1 
 
 47.0 
 
 47.9 
 
 48.8 
 
 49.8 
 
 50.7 
 
 51.6 
 
 52.5 
 
 53.5 
 
 54.5 
 
 55.6 
 
 56.7 
 
 57.8 
 
 59.0 
 
 60.3 
 
 42.29 
 
 42.81 
 
 43.34 
 
 43.89 
 
 44.45 
 
 45.02 
 
 45.60 
 
 46.20 
 
 46.82 
 
 47.46 
 
 48.12 
 
 48.80 
 
 49.49 
 
 50.20 
 
 50.93 
 
 51.68 
 
 52.45 
 
 53.25 
 
 54.09 
 
 54.94 
 
 55.82 
 
 56.73 
 
 57.67 
 
 58.64 
 
 59.64 
 
 60.47 
 
 57 
 
 56 
 
 55 
 
 54 
 
 53 
 
 52 
 
 51 
 
 50 
 
 49 
 
 48 
 
 47 
 
 46 
 
 45 
 
 44 
 
 43 
 
 42 
 
 41 
 
 40 
 
 39 
 
 38 
 
 37 
 
 36 
 
 35 
 
 34 
 
 33 
 
 32 
 
 61.6 
 62.9 
 64.3 
 65.7 
 67.2 
 68.7 
 70.3 
 72.0 
 73.7 
 75.5 
 77.3 
 79.2 
 81.2 
 83.3 
 85.5 
 87.7 
 90.0 
 92.4 
 94.9 
 97.5 
 100.2 
 103.0 
 106.0 
 109.1 
 112.3 
 115.9 
 
 61.75 
 62.86 
 64.01 
 65.21 
 66.46 
 67.74 
 69.09 
 70.48 
 71.94 
 73.45 
 75.02 
 76.68 
 78.40 
 80.20 
 82.09 
 84.07 
 86.14 
 88.32 
 90.62 
 93.04 
 95.58 
 98.28 
 101.12 
 104.15 
 107.35 
 110.76 
 
 31 
 30 
 29 
 28 
 27 
 26 
 25 
 24 
 23 
 22 
 21 
 20 
 19 
 18 
 17 
 16 
 15 
 14 
 13 
 12 
 11 
 10 
 9 
 8 
 7 
 6 
 
 119.9 
 
 124.3 
 
 128.9 
 
 133.9 
 
 139.3 
 
 145.2 
 
 151.5 
 
 158.3 
 
 165.7 
 
 173.8 
 
 182.6 
 
 192.1 
 
 202.4 
 
 213.5 
 
 225.7 
 
 239.4 
 
 255.2 
 
 273.6 
 
 295.4 
 
 321.1 
 
 351.6 
 
 388.1 
 
 432.5 
 
 487.9 
 
 558.0 
 
 656.3 
 
 114.39 
 
 118.27 
 
 122.41 
 
 120.87 
 
 131.05 
 
 136.81 
 
 142.40 
 
 148.46 
 
 155.06 
 
 162.27 
 
 170.18 
 
 178.91 
 
 188.59 
 
 190.36 
 
 211.44 
 
 22r).08 
 
 240.01 
 
 258.44 
 
 279.11 
 
 303.38 
 
 332.27 
 
 307.25 
 
 410.45 
 
 405.11 
 
 536.71 
 
 634.33 
 
28 EADIAKT HEAT. CHAP. ii. 
 
 based on sucli unsatisfactory premises cannot be sound. How 
 far this inference is correct will be seen presently, by a com- 
 parison between tlie ratio of cooling at liigli temperatures, 
 deduced from MM. Duloug and Petit's formula, and the actual 
 rate shown by an incandescent cast-iron sphere enclosed within 
 a vacuum ; also by the amount of mechanical energy developed 
 by the radiation of fused cast iron. The temperatures inserted 
 in the table mai'ked B, relating to the experiments under con- 
 sideration, have been reduced to the Centigrade scale for the 
 purpose of facilitating direct comparison with the result of 
 Dulong and Petit's researches. 
 
 The question has frequently been asked, whether Newton's 
 law holds for mcrease of temperature when a cold body is 
 exposed to the radiation of a surrounding hot medium. In 
 order to decide the question, experimentally, whether the times 
 occupied in heating a certain body are equal, for corresponding 
 differential temperatures, to the times occupied in cooling the 
 same body, a modified apparatus has been constructed (see 
 illustration shown on Plate 5, Fig. 3). The means adopted 
 for measuring the temperature and producing circulation Avithin 
 the central sphere being identical with those of the apparatus 
 already described, it will only be necessary to point out that 
 the exhausted vessel a a which suiTounds the central sphere 
 c is immersed in water contained in a cylindrical vertical 
 boiler b, open at the top and heated by means of a spirit- 
 lamp applied under its bottom. The experiments with this 
 modified apparatus have been conducted in the following 
 manner: The water in the boiler havina: been brought to 
 
CHAP. II. UADIATION AT DIFFEUEST TEMl'EliATUREa. 
 
 29 
 
 c 
 
 Table showing 
 
 THE Eate of Heating of 
 
 A Cold Body 
 
 
 SUSPENDED WITHIN 
 
 A Heated 
 
 Enclosure. 
 
 
 'o 
 So 
 
 53 
 1 = 
 
 
 is 
 
 ss 
 
 M 3 
 
 •° 9, 
 
 IS 
 
 Time agreeably to 
 Newton's law. 
 
 gi 
 H 
 
 s-g 
 
 B 
 
 o 
 
 u 
 
 ~ a 
 o ^ 
 H. ° 
 
 X 3 
 
 
 o 
 
 II 
 
 ' FaK. 
 
 'Fah. 
 
 'FaK 
 
 Sect. 
 
 Stct. 
 
 'Fah. 
 
 'Fah. 
 
 'Fah. 
 
 Sea. 
 
 StM. 
 
 212 
 
 37 
 
 175 
 
 29.1 
 
 29.08 
 
 212 
 
 64 
 
 148 
 
 35.6 
 
 34.41 
 
 212 
 
 38 
 
 174 
 
 29.3 
 
 29.25 
 
 212 
 
 65 
 
 147 
 
 35.9 
 
 34.64 
 
 212 
 
 39 
 
 173 
 
 29.5 
 
 29.42 
 
 212 
 
 66 
 
 146 
 
 36.3 
 
 34.88 
 
 212 
 
 40 
 
 172 
 
 29.7 
 
 29.59 
 
 212 
 
 67 
 
 145 
 
 36.5 
 
 35.12 
 
 212 
 
 41 
 
 171 
 
 29.9 
 
 29.76 
 
 212 
 
 68 
 
 144 
 
 36.8 
 
 35.36 
 
 212 
 
 42 
 
 170 
 
 30.1 
 
 29.94 
 
 212 
 
 69 
 
 143 
 
 37.1 
 
 35.01 
 
 212 
 
 43 
 
 169 
 
 30.3 
 
 30.12 
 
 212 
 
 70 
 
 142 
 
 37.4 
 
 35.86 
 
 212 
 
 44 
 
 168 
 
 30.5 
 
 30.29 
 
 212 
 
 71 
 
 141 
 
 37.7 
 
 36.12 
 
 212 
 
 45 
 
 167 
 
 30.7 
 
 30.48 
 
 212 
 
 72 
 
 140 
 
 38.0 
 
 36.38 
 
 212 
 
 46 
 
 166 
 
 30.9 
 
 30.66 
 
 212 
 
 73 
 
 139 
 
 38.3 
 
 36.64 
 
 212 
 
 47 
 
 165 
 
 31.1 
 
 30.85 
 
 212 
 
 74 
 
 138 
 
 38.6 
 
 36.91 
 
 212 
 
 48 
 
 164 
 
 31.3 
 
 31.04 
 
 212 
 
 75 
 
 137 
 
 38.9 
 
 37.18 
 
 212 
 
 49 
 
 163 
 
 31.5 
 
 31.23 
 
 212 
 
 76 
 
 136 
 
 39.2 
 
 37.46 
 
 212 
 
 50 
 
 162 
 
 31.7 
 
 31.42 
 
 212 
 
 77 
 
 135 
 
 39.5 
 
 37.74 
 
 212 
 
 51 
 
 161 
 
 31.9 
 
 31.62 
 
 212 
 
 78 
 
 134 
 
 39.8 
 
 38.02 
 
 212 
 
 52 
 
 160 
 
 32.1 
 
 31.82 
 
 212 
 
 79 
 
 133 
 
 40.1 
 
 38.30 
 
 212 
 
 53 
 
 159 
 
 32.3 
 
 32.02 
 
 212 
 
 80 
 
 132 
 
 40.4 
 
 38.59 
 
 212 
 
 54 
 
 158 
 
 32.6 
 
 32.22 
 
 212 
 
 81 
 
 131 
 
 40.7 
 
 38.88 
 
 212 
 
 55 
 
 157 
 
 32.9 
 
 32.43 
 
 212 
 
 82 
 
 130 
 
 41.0 
 
 39.18 
 
 212 
 
 56 
 
 156 
 
 33.2 
 
 32.64 
 
 212 
 
 S3 
 
 129 
 
 41.3 
 
 39.49 
 
 212 
 
 57 
 
 155 
 
 33.5 
 
 32.85 
 
 212 
 
 84 
 
 128 
 
 41.6 
 
 39.80 
 
 212 
 
 58 
 
 154 
 
 33.8 
 
 33.06 
 
 212 
 
 85 
 
 127 
 
 41.9 
 
 40.12 
 
 212 
 
 59 
 
 153 
 
 34.1 
 
 33.28 
 
 212 
 
 86 
 
 126 
 
 42.2 
 
 40.44 
 
 212 
 
 60 
 
 152 
 
 34.4 
 
 33.50 
 
 212 
 
 87 
 
 125 
 
 42.5 
 
 40.76 
 
 212 
 
 61 
 
 151 
 
 34.7 
 
 33.72 
 
 212 
 
 88 
 
 124 
 
 42.8 
 
 41.09 
 
 212 
 
 62 
 
 150 
 
 35.0 
 
 33.95 
 
 212 
 
 89 
 
 123 
 
 43.1 
 
 41.43 
 
 212 
 
 63 
 
 149 
 
 35.3 
 
 34.18 
 
 212 
 
 90 
 
 122 
 
 43.5 
 
 41.77 
 
30 
 
 BADIAXT UEAT. 
 
 c 
 
 Tablk snowixG 
 
 THE IiATE OF HeaTIXG OF 
 
 A Cold 
 
 Body 
 
 
 SUSPKXDED WITIIIX 
 
 A Heated 
 
 ENCLOSURE. 
 
 
 o 
 
 -S 3 
 
 as 
 P 
 
 If 
 
 f= o 
 II 
 
 •s, ° 
 
 is of 
 
 -a" g 
 > g 5 
 
 i ti 
 
 J a 
 
 0-" 
 
 o 
 
 || 
 
 1 ^ 
 
 o 
 
 si 
 
 II 
 si 
 
 
 So 
 
 II 
 
 it 
 
 > S o 
 •°.5 
 
 
 ' Fah. 
 
 'Fah. 
 
 'I-ah. 
 
 5eos. 
 
 Sees. 
 
 'Fali. 
 
 ° Fah. 
 
 'Fah. 
 
 Sec». 
 
 S(C3. 
 
 212 
 
 91 
 
 121 
 
 43.9 
 
 42.12 
 
 212 
 
 118 
 
 94 
 
 55.4 
 
 54.28 
 
 212 
 
 92 
 
 120 
 
 44.3 
 
 42.47 
 
 212 
 
 119 
 
 93 
 
 55.9 
 
 54.86 
 
 212 
 
 93 
 
 119 
 
 44.7 
 
 42.83 
 
 212 
 
 120 
 
 92 
 
 56.4 
 
 55.46 
 
 212 
 
 94 
 
 118 
 
 45.1 
 
 43.19 
 
 212 
 
 121 
 
 91 
 
 56.9 
 
 56.08 
 
 212 
 
 95 
 
 117 
 
 45.5 
 
 43.56 
 
 212 
 
 122 
 
 90 
 
 57.4 
 
 56.70 
 
 212 
 
 96 
 
 116 
 
 45.9 
 
 43.94 
 
 212 
 
 123 
 
 89 
 
 57.9 
 
 57.33 
 
 212 
 
 97 
 
 115 
 
 46.3 
 
 44.32 
 
 212 
 
 124 
 
 88 
 
 58.5 
 
 68.00 
 
 212 
 
 98 
 
 114 
 
 46.7 
 
 44.71 
 
 212 
 
 125 
 
 87 
 
 59.1 
 
 58.68 
 
 212 
 
 99 
 
 113 
 
 47.1 
 
 45.11 
 
 212 
 
 126 
 
 86 
 
 59.7 
 
 59.36 
 
 212 
 
 100 
 
 112 
 
 47.5 
 
 45.51 
 
 212 
 
 127 
 
 85 
 
 60.3 
 
 60.06 
 
 212 
 
 101 
 
 111 
 
 47.9 
 
 45.93 
 
 212 
 
 128 
 
 84 
 
 60.9 
 
 60.78 
 
 212 
 
 102 
 
 110 
 
 48.3 
 
 46.35 
 
 212 
 
 129 
 
 83 
 
 61.5 
 
 61.51 
 
 212 
 
 103 
 
 109 
 
 48.7 
 
 46.78 
 
 212 
 
 130 
 
 82 
 
 62.1 
 
 62.27 
 
 212 
 
 104 
 
 108 
 
 49.1 
 
 47.21 
 
 212 
 
 131 
 
 81 
 
 62.7 
 
 63.04 
 
 212 
 
 105 
 
 107 
 
 49.5 
 
 47.65 
 
 212 
 
 132 
 
 80 
 
 63.4 
 
 63.84 
 
 212 
 
 106 
 
 106 
 
 49.9 
 
 48.10 
 
 212 
 
 133 
 
 79 
 
 64.1 
 
 64.65 
 
 212 
 
 107 
 
 105 
 
 50.3 
 
 48.56 
 
 212 
 
 134 
 
 78 
 
 64.8 
 
 65.48 
 
 212 
 
 108 
 
 104 
 
 50.7 
 
 49.03 
 
 212 
 
 135 
 
 77 
 
 65.5 
 
 66.34 
 
 212 
 
 109 
 
 103 
 
 51.1 
 
 49.51 
 
 212 
 
 136 
 
 76 
 
 66.2 
 
 67.22 
 
 212 
 
 110 
 
 102 
 
 51.5 
 
 50.00 
 
 212 
 
 137 
 
 75 
 
 66.9 
 
 68.12 
 
 212 
 
 111 
 
 101 
 
 51.9 
 
 50.50 
 
 212 
 
 138 
 
 74 
 
 67.7 
 
 69.05 
 
 212 
 
 112 
 
 100 
 
 52.4 
 
 51.01 
 
 212 
 
 139 
 
 73 
 
 68.5 
 
 70.00 
 
 212 
 
 113 
 
 99 
 
 52.9 
 
 51.52 
 
 212 
 
 140 
 
 72 
 
 69.3 
 
 70.98 
 
 212 
 
 114 
 
 98 
 
 53.4 
 
 52.05 
 
 212 
 
 141 
 
 71 
 
 70.1 
 
 71.98 
 
 212 
 
 115 
 
 97 
 
 53.9 
 
 52.59 
 
 212 
 
 142 
 
 70 
 
 70.9 
 
 73.02 
 
 212 
 
 116 
 
 96 
 
 54.4 
 
 53.14 
 
 212 
 
 143 
 
 69 
 
 71.8 
 
 74.09 
 
 212 
 
 117 
 
 95 
 
 54.9 
 
 53.70 
 
 212 
 
 144 
 
 08 
 
 72.7 
 
 75.19 
 
CHAP. II. RADIATION AT DIFFEREXT TEMPEBATCRES. 
 
 31 
 
 c 
 
 Table showinm; 
 
 Tin: 1!at 
 
 : 01' IfKATLVG OF 
 
 V Cold 
 
 Bom- 
 
 
 SLSPENDEU WnillX 
 
 A Ilk 
 
 Vl TKIJ 
 
 EMLOHLh-t:. 
 
 
 II 
 
 ll 
 
 CO 
 
 Co 
 
 |S 
 
 .£.■3 
 
 ^ c 
 S 
 
 t s 
 
 ■=1 
 
 lii 
 0- 
 
 Time ng^reeably to 
 Newton's law. 
 
 
 
 11 
 
 = 1 
 
 
 
 is 
 
 0^ 
 
 i p 
 
 Observed time lie«t- 
 
 ing eiiclo.'«<l body 
 
 one dcg. 
 
 
 
 1 
 
 r 
 
 'r<lh. 
 
 'FaA. 
 
 'Fa?>. 
 
 Sto. 
 
 Sax. 
 
 •Fa/i. 
 
 'Fa/i. 
 
 'fah. 
 
 Stet. 
 
 Stei. 
 
 212 
 
 145 
 
 67 
 
 73.6 
 
 70.32 
 
 212 
 
 172 
 
 40 
 
 118.7 
 
 128.48 
 
 212 
 
 146 
 
 06 
 
 74.5 
 
 77.48 
 
 212 
 
 173 
 
 39 
 
 121.6 
 
 131.82 
 
 212 
 
 147 
 
 65 
 
 75.5 
 
 78.68 
 
 212 
 
 174 
 
 38 
 
 124.6 
 
 135.33 
 
 212 
 
 148 
 
 64 
 
 76.5 
 
 79.92 
 
 212 
 
 175 
 
 37 
 
 127.8 
 
 139.04 
 
 212 
 
 149 
 
 63 
 
 77.5 
 
 81.20 
 
 212 
 
 176 
 
 36 
 
 131.2 
 
 142.96 
 
 212 
 
 150 
 
 62 
 
 78.6 
 
 82.52 
 
 212 
 
 177 
 
 3."") 
 
 134.8 
 
 147.10 
 
 212 
 
 151 
 
 61 
 
 79.7 
 
 83.88 
 
 212 
 
 178 
 
 34 
 
 VSS.G 
 
 151.49 
 
 212 
 
 152 
 
 60 
 
 80.8 
 
 85.29 
 
 212 
 
 179 
 
 33 
 
 142.6 
 
 150.15 
 
 212 
 
 153 
 
 59 
 
 82.0 
 
 86.75 
 
 212 
 
 180 
 
 32 
 
 146.9 
 
 161.11 
 
 212 
 
 154 
 
 58 
 
 83.2 
 
 88.26 
 
 212 
 
 181 
 
 31 
 
 151.5 
 
 166.39 
 
 212 
 
 155 
 
 57 
 
 84.5 
 
 89.82 
 
 212 
 
 182 
 
 30 
 
 156.4 
 
 172.03 
 
 212 
 
 156 
 
 56 
 
 85.8 
 
 91.44 
 
 212 
 
 183 
 
 29 
 
 161.6 
 
 178.07 
 
 212 
 
 157 
 
 55 
 
 87.2 
 
 93.12 
 
 212 
 
 184 
 
 28 
 
 167.1 
 
 184.54 
 
 212 
 
 158 
 
 54 
 
 88.6 
 
 94.86 
 
 212 
 
 185 
 
 27 
 
 172.9 
 
 191.51 
 
 212 
 
 159 
 
 53 
 
 90.1 
 
 96.66 
 
 212 
 
 186 
 
 26 
 
 179.1 
 
 199.01 
 
 212 
 
 160 
 
 52 
 
 91.7 
 
 98.54 
 
 212 
 
 187 
 
 25 
 
 185.7 
 
 207.14 
 
 212 
 
 161 
 
 61 
 
 93.4 
 
 100.50 
 
 212 
 
 188 
 
 24 
 
 192.7 
 
 215.95 
 
 212 
 
 162 
 
 50 
 
 95.2 
 
 102.53 
 
 212 
 
 189 
 
 23 
 
 200.3 
 
 225.55 
 
 212 
 
 163 
 
 49 
 
 97.1 
 
 104.64 
 
 212 
 
 190 
 
 22 
 
 208.7 
 
 236.05 
 
 212 
 
 164 
 
 48 
 
 99.1 
 
 106.84 
 
 212 
 
 191 
 
 21 
 
 218.1 
 
 247.56 
 
 212 
 
 165 
 
 47 
 
 101.2 
 
 109.14 
 
 212 
 
 192 
 
 20 
 
 228.7 
 
 260.26 
 
 212 
 
 166 
 
 46 
 
 103.4 
 
 111.54 
 
 212 
 
 193 
 
 19 
 
 240.7 
 
 274.32 
 
 212 
 
 167 
 
 45 
 
 105.7 
 
 114.05 
 
 212 
 
 194 
 
 18 
 
 254.7 
 
 289.99 
 
 212 
 
 168 
 
 44 
 
 108.1 
 
 116.66 
 
 212 
 
 195 
 
 17 
 
 269.6 
 
 307.57 
 
 212 
 
 169 
 
 43 
 
 110.6 
 
 119.41 
 
 212 
 
 196 
 
 16 
 
 287.0 
 
 327.42 
 
 212 
 
 170 
 
 42 
 
 113.2 
 
 122.29 
 
 212 
 
 197 
 
 15 
 
 306.8 
 
 350.00 
 
 212 
 
 171 
 
 41 
 
 115.9 
 
 125.31 
 
 
 
 
 
 
32 
 
 RADIANT HEAT. 
 
 c 
 
 Table siiowtxo the Kate 
 
 OF Heating of 
 
 A Cold Body 
 
 
 SUSPENJMiD 
 
 WITHIN A Hot 
 
 Enci.osuhe. 
 
 
 -3 
 
 11 
 
 "a 
 
 
 too 
 
 Differential tem- 
 
 pei'atui-e of enclosed 
 
 body. 
 
 J|l 
 11 i 
 
 II 
 
 H 
 
 Differential tem- 
 perature of enclosed 
 body. 
 
 Obsen'ed time heat- 
 ing enclosed body 
 one deg. 
 
 E-i 
 
 'Cent. 
 
 Sees. 
 
 Sec«. 
 
 'Cent. 
 
 Sees. 
 
 Sees. 
 
 ° Cent. 
 
 Sees. 
 
 Sees. 
 
 97 
 
 29.3 
 
 29.22 
 
 67 
 
 44.2 
 
 42.40 
 
 37 
 
 74.4 
 
 11. ^A: 
 
 96 
 
 29.6 
 
 29.52 
 
 66 
 
 44.9 
 
 43.04 
 
 36 
 
 76.1 
 
 79.42 
 
 95 
 
 30.0 
 
 29.83 
 
 65 
 
 45.6 
 
 43.71 
 
 35 
 
 78.0 
 
 81.72 
 
 94 
 
 30.3 
 
 30.15 
 
 64 
 
 46.4 
 
 44.40 
 
 34 
 
 79.9 
 
 84.16 
 
 93 
 
 30.7 
 
 30.48 
 
 63 
 
 47.1 
 
 45.11 
 
 33 
 
 82.0 
 
 86.75 
 
 92 
 
 31.1 
 
 30.81 
 
 62 
 
 '47.8 
 
 45.84 
 
 32 
 
 84.2 
 
 89.50 
 
 91 
 
 31.4 
 
 31.15 
 
 61 
 
 48.5 
 
 46.60 
 
 31 
 
 86.6 
 
 92.44 
 
 90 
 
 31.8 
 
 31.50 
 
 60 
 
 49.3 
 
 47.38 
 
 30 
 
 89.2 
 
 95.57 
 
 89 
 
 32.2 
 
 31.86 
 
 59 
 
 60.0 
 
 48.19 
 
 29 
 
 92.0 
 
 98.92 
 
 88 
 
 32.6 
 
 32.22 
 
 68 
 
 50.7 
 
 49.03 
 
 28 
 
 95.2 
 
 102.53 
 
 87 
 
 33.1 
 
 32.60 
 
 57 
 
 51.4 
 
 49.90 
 
 27 
 
 98.7 
 
 106.39 
 
 86 
 
 33.6 
 
 32.99 
 
 66 
 
 52.2 
 
 60.80 
 
 26 
 
 102.5 
 
 110.56 
 
 85 
 
 34.2 
 
 33.38 
 
 55 
 
 53.1 
 
 51.73 
 
 25 
 
 106.6 
 
 115.04 
 
 84 
 
 34.8 
 
 33.78 
 
 54 
 
 54.0 
 
 52.69 
 
 24 
 
 111.1 
 
 119.97 
 
 83 
 
 35.3 
 
 34.18 
 
 53 
 
 54.9 
 
 53.70 
 
 23 
 
 115.9 
 
 125.31 
 
 82 
 
 35.8 
 
 34.59 
 
 52 
 
 55.7 
 
 54.74 
 
 22 
 
 121.0 
 
 131.13 
 
 81 
 
 36.4 
 
 35.02 
 
 51 
 
 56.6 
 
 55.83 
 
 21 
 
 126.6 
 
 137.63 
 
 80 
 
 36.9 
 
 35.46 
 
 50 
 
 57.6 
 
 56.96 
 
 20 
 
 132.7 
 
 144.58 
 
 79 
 
 37.4 
 
 35.91 
 
 49 
 
 58.6 
 
 58.13 
 
 19 
 
 139.4 
 
 152.40 
 
 78 
 
 38.0 
 
 36.38 
 
 48 
 
 59.7 
 
 69.36 
 
 18 
 
 146.9 
 
 161.11 
 
 77 
 
 38.5 
 
 36.86 
 
 47 
 
 60.8 
 
 60.63 
 
 17 
 
 156.4 
 
 170.87 
 
 76 
 
 39.0 
 
 37.34 
 
 46 
 
 61.9 
 
 61.96 
 
 16 
 
 164.9 
 
 181.89 
 
 75 
 
 39.6 
 
 87.84 
 
 45 
 
 63.0 
 
 63.35 
 
 15 
 
 175.4 
 
 194.44 
 
 74 
 
 40.1 
 
 38.35 
 
 44 
 
 64.2 
 
 64.81 
 
 14 
 
 187.1 
 
 208.84 
 
 73 
 
 40.7 
 
 38.88 
 
 43 
 
 65.5 
 
 66.34 
 
 13 
 
 200.3 
 
 225.56 
 
 72 
 
 41.2 
 
 39.42 
 
 42 
 
 66.8 
 
 67.91 
 
 12 
 
 216.3 
 
 245.16 
 
 71 
 
 41.8 
 
 39.98 
 
 41 
 
 68.2 
 
 69.61 
 
 11 
 
 235.9 
 
 268.52 
 
 70 
 
 42.3 
 
 40.56 
 
 40 
 
 69.6 
 
 71.37 
 
 10 
 
 260.7 
 
 296.77 
 
 69 
 
 42.9 
 
 41.16 
 
 39 
 
 71.1 
 
 73.23 
 
 9 
 
 291.0 
 
 331.69 
 
 68 
 
 43.5 
 
 41.77 
 
 38 
 
 72.7 
 
 75.19 
 
 8 
 
 327.3 
 
 375.90 
 
CHAP. II. liADlAllOS AT iniFKUEyr TEMVEUATLRES. -i'd 
 
 the boiling point, cold water, as near the freezing point as 
 possible, is pumped into the central sphere, through suitable 
 2)ipes (the thermometer having been previously removed). 
 The operation of charging with cold water should be per- 
 formed quickly, and the thermometer inserted as soon as the 
 sphere is full. The reading should then commence without 
 a moment's delay, the temperature being announced for each 
 degree, and, together with the time, recorded by the operator 
 attending the chronograph, precisely as before stated with 
 reference to the process of ascertaining the time of cooling. 
 The result of tin- oxperinieut with the ai)iiaratus under con- 
 sideration will be found by inspecting the annexed tables C 
 together with the diagram attached to the illustration, see 
 Fig. 4, in which the ordinates of the curve a h' b represent 
 the observed time for each degree of differential tempera- 
 ture, while the ordinates of the curve a c' c represent the 
 time that would elapse if the rate of heating were in exact 
 accordance with Newton's doctrine. The small amount of 
 the discrepancy shown by the diagram and table, at differ- 
 ential temperatures exceeding 75° F., proves that the energy 
 varies in accordance with dynamical laws, whether heat be 
 parted with by radiation towards a cooler body, or whether 
 heat be received from a radiating surrounding medium of 
 higher temperature. The perceptible discrepancy at low dif- 
 ferential temperature exhibited in the diagram, namely, the 
 observed times being shorter than the theoretical times, is 
 owing to the unavoidable conduction of heat from the boiler 
 to the cold central sphere through the connecting tubes 
 
34 BADIANT HEAT. chap. ii. 
 
 / and VI. Obviously the prolonged period of heating conse- 
 quent on low differential temperature renders the effect of con- 
 duction appreciable, as indicated by the diagram. Our experi- 
 ments, then, establish the fact that until the emissive power 
 is changed by disturbing causes, the energy developed by 
 radiation increases in the exact ratio of the differential tem- 
 peratures. 
 
 It is proper to mention that one of the principal objects of 
 my investigations relating to the velocity of cooling by radia- 
 tion, has been that of disproving the correctness of the assumj)- 
 tion Avhich certain eminent physicists have based on Dulong 
 and Petit's estimates of the increase of radiant enei-gy deve- 
 loj^ed at high temperatures. Radiation, in a dynamic point 
 of view, being merely transmission of mechanical j^ower, it 
 will be evident, on reflection, that we need only ascertain the 
 number of thermal units developed in a given time, by a 
 given amount of radiating surface maintained at a certain tem- 
 perature, in order to determine the intensity which Dulong 
 and Petit endeavored to deduce from the velocity of cooling 
 shown by the contraction of mercury and other fluids. It 
 must be inferred, from their having adopted a method so un- 
 satisfa,ctory, that these physicists overlooked the fact that 
 radiant intensity is most accurately measured by ascertaining 
 the amount of thermal energy developed in a given time ; 
 and that they deemed it impossible to determine, l)y direct 
 means, tlie velocity of cooling of incandescent bodies. It can- 
 not be supposed that such skilful experimentalists as Diilong 
 and Petit questioned the practicability of suspending an in- 
 
UHAi'. II. EAniATlOX AT 1HI-FEI:EM TEMrEh'ATCRES. X> 
 
 candescent body within a vaciiuni surrounded by a coolin'^ 
 medium. Why, then, it may be asked, did they not resort 
 to that obvious method whicli, if adopte'd, would at once 
 have convinced them of tlie fallacy of tlioir formula assifrn- 
 ing a fabulous rate of velocity of cooling to incandescent 
 bodies ? We must infer that they did not deem it practi- 
 cable to construct an instrument by which the temperature 
 of the enclosed incandesctMit body could be accui'atclv mea- 
 sured. It will be urged in defence of Dulong and Petit's 
 method, that the time occupied in cooling cannot be ascer- 
 tained e.xactly unless an instrument can be devised capable 
 of showing the temperature of incandescent liodies. This 
 objection, apparently valid since we possess no reliable and 
 delicate pyrometer, falls to the ground before the fact that, 
 at a constant distance, the temperatures impai-ted to a ther- 
 mometer by radiation are proportionate to the temperatures 
 of the radiator ; and that consequently the velocity of cooling 
 of a radiator at any temperature whatever may be ascertained 
 by a distant thermometer as con-ectly as by one in actual con- 
 tact. Before entering on a description of the method, before 
 referred to, of measuring the radiant energy at high tempei-a- 
 tures by ascertaining the amount of thermal energy developed 
 in a given time by a given area, I will now l)riefly describe 
 an instrument constructed in accordance with the fact just 
 mentioned, that the teinperatures of a recipient of radiant heat 
 are proportionate to the temperatures of the radiator. The 
 illustration on Plate G, Fig. 5, represents a veitical section 
 and top view of the instrument referred to, by means of 
 
36 EAIUANT HEAT. CHAP. II. 
 
 ^vliich the velocity of cooling of an incandescent l)ody may- 
 be ascertained M-ith perfect accuracy. Description : a a, sphe- 
 rical vessel composed of thin copper, coated with lamp-black 
 on the inside and provided with a cover h fitting air-tight 
 against the top side of a ring secured to the upper part of 
 the spherical vessel, the cover and the ring being ground to- 
 gether in order to dispense with packing, c c, an open cylin- 
 drical vessel through which a constant stream of cold water 
 is circulated by means of a force-pump and flexible tubes 
 attached at d and e, these tubes communicating with a cis- 
 tern in which water is maintained at a constant temperature 
 of 33° F. A" is a solid sphere of cast iron suspended by means 
 of a lug formed at the upper part of the sphere, secured under 
 the cover b as shown by the drawing. A brass stopple ff, 
 fitting air-tight in a conical socket formed in the before-men- 
 tioned ring, supports a Casella thermometer /. The cover b 
 is provided with a vertical handle b' in order to facilitate the 
 operation of inserting the solid sphere A', after being heated 
 and suspended under the cover. The air is exhausted from 
 the spherical vessel by means of a large air-pump, a stop- 
 cock being inserted in the exhaust pipe at /. The mode of 
 operation will be readily understood by the following expla- 
 nation. The solid cast-iron sphere being brought to nearly 
 white heat in an air-furnace, is removed by suitable tongs 
 and suspended under the cover b, which latter is quickly put 
 in position over the opening of the spherical vessel. The 
 air-pump is put in rapid motion immediately after putting 
 on the cover ; a few strokes of the water-pump being required 
 
CHAP. II. EADIATWS AT DIFFERENT TEMVEFxATUREa. 37 
 
 to fill the cistern r c to the height shown in the drawing. 
 These operations being performed with due diligence, practice 
 has shown that the temperature of the incandescent sphere 
 will not fall below 1,GUU" F. before the vacuum is sufficiently 
 complete to admit of recording the indicated temperatures. 
 The thermometer / being secured firmly in the stopple <j, may 
 of course be inserted very quickly, an important circumstance, 
 since the bulb, in order to save time, should be previously 
 heated to nearly maxinmm temperature. We have already 
 pointed out that, agreeably to the laws which govern the 
 transmission of radiant heat, the temperatures produced by 
 radiation at constant distances are proportionate to the tem- 
 peratures of the radiating body. It will be readily perceived, 
 therefore, that the thermometer / will correctly indicate the 
 raie of cooling of the suspended incandescent sphere. But 
 in order to prove the fallacy of Dulong and Petit's estimate 
 of the rate of cooling at high temperatures, we must also 
 ascertain the temperature of the sphere itself. Bearing in 
 mind that the temperature of the recipient of radiant heat 
 is proportionate to that of the radiator, it ^vill be perceived, 
 on reflection, that we can previously determine the ratio be- 
 tween the temperature of the sphere A' and that of the record- 
 ing themometer /. Evidently this latio may be ascertained 
 at temperatures which admit of the emplo^Tuent of ordinary' 
 mercurial thermometers, hence the determination may be very 
 exact. It will be readily understood that the thermometer 
 intended for our investigati.m may be placed at such a dis- 
 tance from a radiating spheiical vessel containing mercuiy 
 
38 EADIAXT HEAT. cuap. it. 
 
 maiiitaiued at a temperature of 400", tliat its indication shall 
 be exactly 100°. In other words, the distance between the 
 radiator and the recording theiTQometer may be such that the 
 temperature of the former shall be exactly four times greater 
 than the temperature of the latter. Now, if we remove the 
 spherical vessel referred to and substitute an incandescent 
 sphere whose diameter corresponds with the outside diameter 
 of the said vessel, the indication of the recording thermometer 
 multiplied by 4 will show the temperatiire of the incandes- 
 cent sphere. It is hardly necessary to point out, that much 
 time would be wasted in the tedious process of determining 
 experimentally the distance between the recording thermo- 
 meter and the radiator, necessary to cause an indication of 
 the former exactly one-fourth of the temperature of the latter. 
 Obviously a less symmetrical coefficient will answer nearly 
 as well as the one mentioned. Let us therefore ascertain, 
 approximately, how near the recording thermometer may be 
 placed fi'om an incandescent sphere of the intended size, in 
 order that the mercuiy in the bulb may not be brought to 
 boiling heat. Having determined the distance of the thermo- 
 meter as stated, we may, without further experimenting, con- 
 struct our instrument as shown by the illustration. The sphe- 
 rical vessel containing mercury being then introduced and a 
 vacuum fomied within the surrounding enclosure, the tempe- 
 rature of the said spherical vessel, compared with the tempe- 
 rature indicated by the recording thermometer, will of course 
 determine the coefficient of the instrument. The following 
 brief explanation of the procedure will suffice. Suppose that 
 
CH.U'. n. RADIATION AT DIFFERENT TEMPEIiArURES. :iO 
 
 the temperature of the mercury in the spherical vessel is 409° 
 while that indicated by the recording thermometer is UM". 
 The ratio of temperature between the radiator and the record- 
 ing thermometer will then be as 409 to 98 ; and hence the 
 
 .„ , 409 
 coemcient sought will he — = 4.173. Consequently, if we 
 
 suspend the incandescent sphere K within the enclosure as 
 before directed, and tind by observation that the recordin*' 
 thermometer indicates 401°, we then learn that the tempera- 
 ture of the enclosed incandescent body, at the moment of 
 observation, is 4.173 X 401 = 1,673° F. The means thus 
 afforded of measuring, with great exactness, the temperature 
 and radiant intensity of incandescent bodies, cannot fail to 
 facilitate future thermic investigations. 
 
 The result of our experiments with the suspended incan- 
 descent sphere, \n\\ be found recorded in the annexed set t)f 
 tables, the first series, marked D, being constructed to the 
 Fahrenheit scale, while the second series E has been reduced 
 to the Centigrade scale. The fii-st column in Table D shows 
 the temperature of the enclosiire, the tenipei-ature of the radi- 
 ating sphere being inserted in the second column. The excess 
 of tem2:)erature of the radiator over that of the surrounding 
 cold vessel will l)e fVnuid in the thu'd column, the time actu- 
 ally occupied in reducing the temperatuie of the radiator 10" 
 F. being entered in the fourth column. The time of cooling, 
 agreeably to the Newtonian law, will l)e found in the fifth 
 and last column. The mode of constructing the ta1)lrs will 
 be readily comprehended by referring to previous explaiia- 
 
40 
 
 BADIAXT HEAT. 
 
 D 
 
 Table showing the Kate 
 
 OF Cooling of an 
 
 1 
 Incandescent 
 
 Solid Spheue suspended within a Cold Enclosuue. | 
 
 1 
 
 
 
 c'_o 
 
 
 -S 
 1^ 
 
 B 9 
 
 el 
 
 h 
 
 c o 
 
 
 h 
 
 
 
 
 o 
 
 £"(3 
 tJO o 
 
 Bll 
 e5 
 
 i^ o 
 || 
 H 
 
 II 
 
 ^g 
 
 1^ " S 
 
 o 
 
 Eh 
 
 'FM. 
 
 °Fnh. 
 
 ° Ji-'iA. 
 
 Seea. 
 
 Seas. 
 
 • Fah. 
 
 'Fuh. 
 
 ' Fah. 
 
 Seed. 
 
 Sees. 
 
 34 
 
 1600 
 
 1566 
 
 22.24 
 
 21.97 
 
 34 
 
 1320 
 
 1286 
 
 30.77 
 
 26.65 
 
 34 
 
 1590 
 
 1556 
 
 22.48 
 
 22.11 
 
 34 
 
 1310 
 
 1276 
 
 31.17 
 
 26.85 
 
 34 
 
 1580 
 
 1546 
 
 22.72 
 
 22.25 
 
 34 
 
 1300 
 
 1266 
 
 31.58 
 
 27.06 
 
 34 
 
 1570 
 
 1536 
 
 22.96 
 
 22.39 
 
 34 
 
 1290 
 
 1256 
 
 32.00 
 
 27.27 
 
 34 
 
 1560 
 
 1526 
 
 23.21 
 
 22.54 
 
 34 
 
 1280 
 
 1246 
 
 32.42 
 
 27.48 
 
 34 
 
 1550 
 
 1616 
 
 23.46 
 
 22.68 
 
 34 
 
 1270 
 
 1236 
 
 32.85 
 
 27.70 
 
 34 
 
 1540 
 
 1506 
 
 23.72 
 
 22.83 
 
 34 
 
 1260 
 
 1226 
 
 33.28 
 
 27.92 
 
 34 
 
 1530 
 
 1496 
 
 23.98 
 
 22.98 
 
 34 
 
 1250 
 
 1216 
 
 33.72 
 
 28.15 
 
 34 
 
 1520 
 
 1486 
 
 24.25 
 
 23.13 
 
 34 
 
 1240 
 
 1206 
 
 34.16 
 
 28.38 
 
 34 
 
 1510 
 
 1476 
 
 24.52 
 
 23.28 
 
 34 
 
 1230 
 
 1196 
 
 34.61 
 
 28.61 
 
 34 
 
 1500 
 
 1466 
 
 24.80 
 
 23.44 
 
 34 
 
 1220 
 
 1186 
 
 35.06 
 
 28.84 
 
 34 
 
 1490 
 
 1456 
 
 25.08 
 
 23.60 
 
 34 
 
 1210 
 
 1176 
 
 35.52 
 
 29.08 
 
 34 
 
 1480 
 
 1446 
 
 25.37 
 
 23.76 
 
 34 
 
 1200 
 
 1166 
 
 35.99 
 
 29.32 
 
 34 
 
 1470 
 
 1436 
 
 25.66 
 
 23.92 
 
 34 
 
 1190 
 
 1156 
 
 36.47 
 
 29.57 
 
 34 
 
 1460 
 
 1426 
 
 25.96 
 
 24.08 
 
 34 
 
 1180 
 
 1146 
 
 36.95 
 
 29.82 
 
 34 
 
 1450 
 
 1416 
 
 26.26 
 
 24.25 
 
 34 
 
 1170 
 
 1136 
 
 37.44 
 
 30.08 
 
 34 
 
 1440 
 
 1406 
 
 26.57 
 
 24.42 
 
 34 
 
 1160 
 
 1126 
 
 37.94 
 
 30.34 
 
 34 
 
 1430 
 
 1396 
 
 26.88 
 
 24.59 
 
 34 
 
 1150 
 
 1116 
 
 38.44 
 
 30.61 - 
 
 34 
 
 1420 
 
 1386 
 
 27.20 
 
 24.77 
 
 34 
 
 1140 
 
 1106 
 
 38.95 
 
 30.88 
 
 34 
 
 1410 
 
 1376 
 
 27.52 
 
 24.95 
 
 34 
 
 1130 
 
 1096 
 
 39.47 
 
 31.15 
 
 34 
 
 1400 
 
 1366 
 
 27.85 
 
 25.13 ' 
 
 34 
 
 1120 
 
 1086 
 
 40.00 
 
 31.43 
 
 34 
 
 1390 
 
 1356 
 
 28.18 
 
 25.30 
 
 34 
 
 1110 
 
 1076 
 
 40.54 
 
 31.71 
 
 34 
 
 1380 
 
 1346 
 
 28.52 
 
 25.48 
 
 34 
 
 1100 
 
 1066 
 
 41.08 
 
 32.00 
 
 34 
 
 1370 
 
 1336 
 
 28.87 
 
 25.67 1 
 
 34 
 
 1090 
 
 1056 
 
 41.63 
 
 32.30 
 
 34 
 
 1360 
 
 1326 
 
 29.23 
 
 25.86 1 
 
 34 
 
 1080 
 
 1046 
 
 42.19 
 
 32.60 
 
 34 
 
 1350 
 
 1316 
 
 29.60 
 
 26.05 
 
 34 
 
 1070 
 
 1036 
 
 42.76 
 
 32.91 
 
 34 
 
 1340 
 
 1306 
 
 29.98 
 
 26.25 
 
 34 
 
 1060 
 
 1026 
 
 43.34 
 
 33.22 
 
 34 
 
 1330 
 
 1296 
 
 30.37 
 
 26.45 
 
 34 
 
 1050 
 
 1016 
 
 43.93 
 
 33.54 
 
CHAP. II. EAIHJ lloX AT DIFFEliEXT TEMl'KUMrREa. 
 
 41 
 
 D 
 
 T.\BLF, SIIOWINO THF 
 
 Rate of Cooling 
 
 OF AN 
 
 [NCAN DESCENT 
 
 Solid Sphere suspended within a 
 
 Cold Enclosuke. j 
 
 o 
 E u 
 
 Hi 
 "3 
 
 1" 
 
 it 
 
 o 
 
 2. 
 
 H 
 
 u 
 
 H 
 
 £•1 
 
 b* i 
 
 S.1 
 
 so . 
 ■2 = 1 
 
 Observed time cool- 
 ing enclosed 
 sphere ten degs. 
 
 2 
 
 1 
 
 1 
 1^ 
 
 'Fah. 
 
 'Fah. 
 
 •Fiih. 
 
 S»a. 
 
 8ta. 
 
 'Fah. 
 
 -Fah. 
 
 'Fah. 
 
 Stet. 
 
 Stct. 
 
 34 
 
 1040 
 
 lOOG 
 
 44.53 
 
 33.80 
 
 34 
 
 760 
 
 726 
 
 72.05 
 
 46.42 
 
 34 
 
 1030 
 
 996 
 
 45.15 
 
 34.19 
 
 34 
 
 750 
 
 716 
 
 73.60 
 
 47.04 
 
 34 
 
 1020 
 
 986 
 
 45.79 
 
 34.53 
 
 34 
 
 740 
 
 706 
 
 75.33 
 
 47.68 
 
 34 
 
 1010 
 
 976 
 
 46.45 
 
 38.87 
 
 34 
 
 730 
 
 696 
 
 77.06 
 
 48.34 
 
 34 
 
 1000 
 
 966 
 
 47.12 
 
 35.22 
 
 34 
 
 720 
 
 686 
 
 78.85 
 
 49.01 
 
 34 
 
 990 
 
 956 
 
 47.81 
 
 35.58 
 
 34 
 
 710 
 
 676 
 
 80.70 
 
 49.71 
 
 34 
 
 980 
 
 946 
 
 48.52 
 
 35.94 
 
 34 
 
 700 
 
 666 
 
 82.61 
 
 50.43 
 
 34 
 
 970 
 
 936 
 
 49.24 
 
 36.31 
 
 34 
 
 690 
 
 656 
 
 84.58 
 
 51.16 
 
 34 
 
 960 
 
 926 
 
 49.98 
 
 36.69 
 
 34 
 
 680 
 
 646 
 
 86.61 
 
 61.92 
 
 34 
 
 950 
 
 916 
 
 50.74 
 
 37.08 
 
 34 
 
 670 
 
 636 
 
 88.71 
 
 52.70 
 
 34 
 
 940 
 
 906 
 
 51.53 
 
 37.48 
 
 34 
 
 660 
 
 626 
 
 90.88 
 
 53.50 
 
 34 
 
 930 
 
 896 
 
 52.35 
 
 37.89 
 
 34 
 
 650 
 
 616 
 
 93.12 
 
 64.33 
 
 34 
 
 920 
 
 886 
 
 5r20 
 
 38.30 
 
 34 
 
 640 
 
 606 
 
 95.43 
 
 65.18 
 
 34 
 
 910 
 
 876 
 
 54.08 
 
 38.72 
 
 34 
 
 630 
 
 596 
 
 97.81 
 
 66.07 
 
 34 
 
 900 
 
 866 
 
 64.99 
 
 39.15 
 
 34 
 
 620 
 
 586 
 
 100.26 
 
 66.98 
 
 34 
 
 890 
 
 856 
 
 55.93 
 
 39.59 
 
 34 
 
 610 
 
 576 
 
 102.78 
 
 67.92 
 
 34 
 
 880 
 
 846 
 
 56.90 
 
 40.04 
 
 34 
 
 600 
 
 566 
 
 105.37 
 
 68.90 
 
 34 
 
 870 
 
 836 
 
 57.91 
 
 40.51 
 
 34 
 
 590 
 
 556 
 
 108.03 
 
 69.91 
 
 34 
 
 8C0 
 
 826 
 
 58.96 
 
 40.98 
 
 34 
 
 580 
 
 546 
 
 110.76 
 
 60.95 
 
 34 
 
 850 
 
 816 
 
 60.05 
 
 41.47 
 
 34 
 
 570 
 
 636 
 
 113.56 
 
 62.03 
 
 34 
 
 840 
 
 806 
 
 61.18 
 
 41.97 
 
 34 
 
 560 
 
 626 
 
 116.43 
 
 63.15 
 
 34 
 
 830 
 
 796 
 
 62.36 
 
 42.48 
 
 34 
 
 550 
 
 616 
 
 119.37 
 
 64.30 
 
 34 
 
 820 
 
 786 
 
 63.59 
 
 43.00 
 
 34 
 
 540 
 
 506 
 
 122.38 
 
 65.50 
 
 34 
 
 810 
 
 776 
 
 64.87 
 
 43.53 
 
 34 
 
 530 
 
 496 
 
 125.46 
 
 66.75 
 
 34 
 
 800 
 
 766 
 
 66.20 
 
 44.08 
 
 34 
 
 520 
 
 486 
 
 128.61 
 
 68.05 
 
 34 
 
 790 
 
 756 
 
 67.58 
 
 44.65 
 
 34 
 
 510 
 
 476 
 
 131.83 
 
 69.40 
 
 34 
 
 780 
 
 746 
 
 69.01 
 
 45.22 ' 34 
 
 500 
 
 466 
 
 135.13 
 
 70.80 
 
 34 
 
 770 
 
 736 
 
 70.50 
 
 45.81 1 
 
 
 
 
 
42 
 
 BABIANT HEAT. 
 
 E 
 
 Table showing the Rate 
 
 OF Cooling of 
 
 AN Incandescent 
 
 Solid Sphere 
 
 suspended avithin a Cold Exclosvre. 
 
 i 
 
 i- 
 
 o 5? 
 
 3 
 
 a 
 
 i- 
 
 o ?J 
 
 o 
 
 gJ 
 
 S.S 
 
 S-a S? 
 
 "i 
 
 ts 
 
 a§ 
 
 Or-J M 
 
 .&•& 
 
 5& 
 a-i 
 
 P'g'o, 
 ^ ^^ ^ 
 
 •Ill 
 
 S.S.S 
 o 
 
 'i'i 
 
 3& 
 
 Hi 
 
 III 
 
 g; teg 
 
 1^ 
 
 
 P4 
 
 
 
 1^ 
 
 o 
 
 fi 
 
 ° Cont. 
 
 ° Cent. 
 
 Seoe. 
 
 Sees. 
 
 ° Cmit. 
 
 " Cent. 
 
 Sees. 
 
 Sece. 
 
 890 
 
 771.12 
 
 22.18 
 
 21.94 
 
 735 
 
 716.12 
 
 30.63 
 
 26.58 
 
 885 
 
 776.12 
 
 22.39 
 
 22.06 
 
 730 
 
 711.12 
 
 30.99 
 
 26.76 
 
 880 
 
 761.12 
 
 22.61 
 
 22.19 
 
 725 
 
 706.12 
 
 31.36 
 
 26.95 
 
 875 
 
 756.12 
 
 22.83 
 
 22.32 
 
 720 
 
 701.12 
 
 31.73 
 
 27.13 
 
 870 
 
 751.12 
 
 23.05 
 
 22.44 
 
 715 
 
 696.12 
 
 32.11 
 
 27.32 
 
 865 
 
 746.12 
 
 23.27 
 
 22.57 
 
 710 
 
 691.12 
 
 32.49 
 
 27.61 
 
 860 
 
 741.12 
 
 23.49 
 
 22.70 
 
 705 
 
 686.12 
 
 32.87 
 
 27.71 
 
 855 
 
 736.12 
 
 23.73 
 
 22.83 
 
 700 
 
 681.12 
 
 33.26 
 
 27.91 
 
 850 
 
 731.12 
 
 23.96 
 
 22.97 
 
 695 
 
 676.12 
 
 33.65 
 
 28.12 
 
 845 
 
 726.12 
 
 24.20 
 
 23.11 
 
 690 
 
 671.12 
 
 34.05 
 
 28.32 
 
 840 
 
 721.12 
 
 24.44 
 
 23.25 
 
 685 
 
 666.12 
 
 34.45 
 
 28.53 
 
 835 
 
 716.12 
 
 24.69 
 
 23.39 
 
 680 
 
 661.12 
 
 34.86 
 
 28.74 
 
 830 
 
 711.12 
 
 24.94 
 
 23.53 
 
 675 
 
 656.12 
 
 35.27 
 
 28.95 
 
 825 
 
 706.12 
 
 25.20 
 
 23.67 
 
 670 
 
 651.12 
 
 35.69 
 
 29.17 
 
 820 
 
 701.12 
 
 25.46 
 
 23.81 
 
 665 
 
 646.12 
 
 36.11 
 
 29.39 
 
 815 
 
 696.12 
 
 25.73 
 
 23.96 
 
 660 
 
 641.12 
 
 36.54 
 
 29.61 
 
 810 
 
 691.12 
 
 26.00 
 
 24.11 
 
 655 
 
 636.12 
 
 36.98 
 
 29.84 
 
 805 
 
 686.12 
 
 26.28 
 
 24.26 
 
 650 
 
 631.12 
 
 37.42 
 
 30.07 
 
 800 
 
 681.12 
 
 26.56 
 
 24.41 
 
 645 
 
 626.12 
 
 37.87 
 
 30.30 
 
 795 
 
 676.12 
 
 26.84 
 
 24.57 
 
 640 
 
 621.12 
 
 38.32 
 
 30.54 
 
 790 
 
 671.12 
 
 27.13 
 
 24.72 
 
 635 
 
 616.12 
 
 38.78 
 
 30.78 
 
 785 
 
 766.12 
 
 27.42 
 
 24.88 
 
 630 
 
 611.12 
 
 39.24 
 
 31.03 
 
 780 
 
 761.12 
 
 27.71 
 
 25.04 
 
 625 
 
 606.12 
 
 39.71 
 
 31.28 
 
 775 
 
 756.12 
 
 28.01 
 
 25.20 
 
 620 
 
 601.12 
 
 40.19 
 
 31.53 
 
 770 
 
 751.12 
 
 28.31 
 
 25.36 
 
 615 
 
 596.12 
 
 40.67 
 
 31.79 
 
 765 
 
 746.12 
 
 28.62 
 
 25.53 
 
 610 
 
 591.12 
 
 41.16 
 
 32.05 
 
 760 
 
 741.12 
 
 28.93 
 
 25.70 
 
 605 
 
 586.12 
 
 41.65 
 
 32.31 
 
 755 
 
 736.12 
 
 29.25 
 
 25.87 
 
 600 
 
 581.12 
 
 42.15 
 
 32.58 
 
 750 
 
 731.12 
 
 29.58 
 
 26.04 
 
 595 
 
 576.12 
 
 42.66 
 
 32.85 
 
 745 
 
 726.12 
 
 29.92 
 
 26.22 
 
 590 
 
 571.12 
 
 43.18 
 
 33.13 1 
 
 740 
 
 721.12 
 
 30.27 
 
 . 26.40 
 
 585 
 
 566.12 
 
 43.71 
 
 33.42 j 
 
CHAP. II. L'AIU Alios AT lilFFi:i!i:ST rilMVEUATlUES. 
 
 43 
 
 E 
 
 Table siiowino the Ratk 
 
 OF Cooling of 
 
 AN InCAN 
 
 DESCIiNT 
 
 SOLII 
 
 ) Sl'UEKIC 
 
 SUSl'KXDKl) WITHIN A L'OLU AWci.OSl'JiJ-:. 
 
 £ 
 If 
 
 o 
 
 lis 
 
 "•si 
 
 la 
 
 S 3 
 
 ill 
 
 5 ■" 
 
 2 
 
 Eh 
 
 li 
 
 n 
 
 11 
 
 E ° . 
 
 is 
 
 HI 
 is: 
 
 o 
 
 5 
 
 'Cmt 
 
 'Cent. 
 
 S«ea. 
 
 Stm. 
 
 •On/. 
 
 °Cmt. 
 
 Stet. 
 
 8k». 
 
 580 
 
 561.12 
 
 44.25 
 
 33.71 
 
 425 
 
 406.12 
 
 71.19 
 
 46.08 
 
 575 
 
 556.12 
 
 44.80 
 
 34.01 
 
 420 
 
 401.12 
 
 72.61 
 
 40.03 
 
 670 
 
 551.12 
 
 45.37 
 
 34.31 
 
 415 
 
 396.12 
 
 74.08 
 
 47.20 
 
 665 
 
 646.12 
 
 46.95 
 
 34.61 
 
 410 
 
 391.12 
 
 76.60 
 
 47.78 
 
 660 
 
 541.12 
 
 46.54 
 
 34.92 
 
 405 
 
 386.12 
 
 77.17 
 
 48.37 
 
 555 
 
 636.12 
 
 47.16 
 
 35.24 
 
 400 
 
 381.12 
 
 78.78 
 
 48.98 
 
 660 
 
 531.12 
 
 47.77 
 
 36.56 
 
 395 
 
 376.12 
 
 80.44 
 
 49.60 
 
 645 
 
 520.12 
 
 48.41 
 
 36.89 
 
 390 
 
 371.12 
 
 82.14 
 
 60.24 
 
 640 
 
 521.12 
 
 49.06 
 
 36.22 
 
 385 
 
 366.12 
 
 83.89 
 
 60.90 
 
 535 
 
 616.12 
 
 49.72 
 
 36.56 
 
 380 
 
 361.12 
 
 86.69 
 
 51.58 
 
 630 
 
 611.12 
 
 50.40 
 
 36.91 
 
 376 
 
 356.12 
 
 87.66 
 
 52.27 
 
 625 
 
 500.12 
 
 51.10 
 
 37.26 
 
 370 
 
 351.12 
 
 89.47 
 
 62.98 
 
 620 
 
 601.12 
 
 51.82 
 
 37.62 
 
 365 
 
 346.12 
 
 91.44 
 
 63.71 
 
 515 
 
 496.12 
 
 52.66 
 
 37.99 
 
 360 
 
 341.12 
 
 93.47 
 
 64.46 
 
 610 
 
 491.12 
 
 53.32 
 
 38.36 
 
 365 
 
 336.12 
 
 95.55 
 
 65.23 
 
 505 
 
 486.12 
 
 54.11 
 
 38.74 
 
 350 
 
 331.12 
 
 97.69 
 
 56.02 
 
 600 
 
 481.12 
 
 54.93 
 
 39.13 
 
 345 
 
 326.12 
 
 99.89 
 
 56.84 
 
 495 
 
 476.12 
 
 56.78 
 
 39.63 
 
 340 
 
 321.12 
 
 102.15 
 
 57.68 
 
 490 
 
 471.12 
 
 66.65 
 
 39.94 
 
 335 
 
 316.12 
 
 104.46 
 
 58.66 
 
 486 
 
 466.12 
 
 57.65 
 
 40.35 
 
 330 
 
 311.12 
 
 106.83 
 
 59.46 
 
 480 
 
 461.12 
 
 58.48 
 
 40.77 
 
 325 
 
 306.12 
 
 109.26 
 
 60.37 
 
 475 
 
 466.12 
 
 59.44 
 
 41.20 
 
 320 
 
 301.12 
 
 111.74 
 
 61.32 
 
 470 
 
 451.12 
 
 60.44 
 
 41.64 
 
 315 
 
 296.12 
 
 114.28 
 
 62.30 
 
 465 
 
 446.12 
 
 61.47 
 
 42.10 
 
 310 
 
 291.12 
 
 116.87 
 
 63.31 
 
 460 
 
 441.12 
 
 62.54 
 
 42.66 
 
 306 
 
 286.12 
 
 119.62 
 
 64.36 
 
 465 
 
 436.12 
 
 63.65 
 
 43.03 
 
 300 
 
 281.12 
 
 122.23 
 
 65.44 
 
 450 
 
 431.12 
 
 64.80 
 
 43.51 
 
 296 
 
 276.12 
 
 125.00 
 
 66.56 
 
 445 
 
 426.12 
 
 65.99 
 
 44.00 
 
 290 
 
 271.12 
 
 127.83 
 
 67.72 
 
 440 
 
 421.12 
 
 67.22 
 
 44.50 
 
 285 
 
 266.12 
 
 130.71 
 
 68.92 
 
 435 
 
 416.12 
 
 68.60 
 
 46.01 
 
 280 
 
 261.12 
 
 133.65 
 
 70.62 
 
 430 
 
 411.12 
 
 69.82 
 
 45.64 
 
 275 
 
 256.12 
 
 
 
44 BABIANT HEAT. CHAP. ii. 
 
 tions. It vriU he seen that tlie time occupied in cooling tlie 
 radiating sphere through a range of ten degrees fonns the 
 basis, the temperature being of course deduced from the indi- 
 cation of the thermometer /. The advocates of Dulong and 
 Petit's doctrine will be surprised to find by our tables that 
 the incandescent sphere enclosed in a vacuum surrounded by a 
 cooling medium maintained at a very low temperature, requires 
 upwards of twenty seconds to cool five degrees Centigrade, 
 ^vhile agreeably to Duloug's formula the stated reduction of 
 intensity takes place during an interval occupying only a very 
 small fraction of one second. Again, our tables show that 
 the fall of temperature from 1,600° F. to 500° F. requires 
 the considei'able lapse of ninety-seven minutes, a fact ^\hich 
 alone proves that the celebrated formula of Dulong is grossly 
 eri'oneous. 
 
 We have before adverted to the fact that physicists have 
 siq:)posed that the rate of cooling of a body approaching white 
 heat cannot be ascertained practically, because we possess no 
 reliable instrument for measuring high temperatures. MM. 
 Dulong and Petit, apparently impressed with this idea, con- 
 fined their researches, as before stated, to temperatures below 
 that of boiling mercury, imagining that by observing the 
 rate of cooling up to a differential temperature of 240° C. 
 they would be enabled to establish a law^ which would deter- 
 mine radiant energy for all intensities. Unfortunately, phy- 
 sicists have accepted the result of those researches, Dulong's 
 formula being now the guide which the student is taught to 
 follow in calculating the energy of radiant heat. Probably 
 
CHA1-. II. BAVIATWS AT DIFFEUE}iT TEMPERATURES. 45 
 
 uo doctrine iu physics has occasioned such serious miscon- 
 ception as that propounded by MM. Dulong and Petit. The 
 advance of every branch of knowledge connected with radiant 
 hoiit has been retarded by the adoption of their doctrine 
 ren-ardin"- its transmission. Among other important matters, 
 the question relating to the temperature of the sun has be 
 come seriously entangled by its adoption. It has led such 
 eminent men as Pouillet to assume that the emission of heat 
 at the surface of the sun is so rapid that the ascertained enor- 
 mous development of 300,000 thermal units in a minute, on 
 one square foot, may be accounted for by the accepted doc- 
 trine, even supposing the temperature of the surface of the 
 photosphere to be only 1,461° C. A glance at the preceding 
 Table E, which is based on actual trial, shows the velocity of 
 cooling to be so moderate at the high temperature of 890° C, 
 
 that the fall only amounts to — ^ X (890 - 885) = 13.5° C. 
 •^ 22.18 
 
 in one minute, instead of 1,870.6° C. for the stated tempera- 
 ture, shown by Dulong and Petit's formula \' = 2.037 (1.0077' 
 — 1). It is sui-prising that the tme character of this pal- 
 pably erroneous foi-mula, so often applied by engineei-s and 
 physicists, has never been subjected to practical investigation. 
 A positive test might have been applied in vaiiuus ^vays, 
 although not with that degree of exactness which is attain- 
 able by the means we have just described. The fallacy of 
 the assumption, that an increase of a few hundred degrees 
 of temperature suffices to augment the rate of cooling a thou- 
 sand fold, becomes self-evident if \ve consider that an increased 
 
46 RADIANT HEAT. chap. ii. 
 
 rate of cooliug menus u proportionate increase of the capabi- 
 lity of commuuicating mecliauical energy. Practical engineers 
 who have observed the short time required to raise a mass 
 of red-hot metal to the melting point, are familiar with the 
 fact that the heat imparted to the metal by the combustibles 
 during the short interval necessary to reach the point of fusion, 
 from that of red heat, is wholly insufficient to communicate 
 the enormous amount of energy assumed by those who accept 
 MM. Duloug and Petit's formula. This practical mode of test- 
 ing the correctness of the accejjted doctrine demands serious 
 reflection. Let us consider that bright-red heat indicates a 
 temperature of 1,000° C, while welding or white heat indi- 
 cates 1,500° C. Now, according to the formula V = 2.037 
 (1.0077'- 1), the velocity of cooling at 1,000° is represented 
 by a fall of temperature of one degree in the short time of 
 
 — -— - of a minute, while the velocity of coolius; at 1,500° C, 
 4,356 J fc. ) ' 
 
 assigned by the same formula, will be at the rate of one de- 
 gree in the exceedingly brief period of of a minute ; 
 
 ^ ^ -^ ^ 202,710 
 
 hence the energy imparted to the heated metal ^\\n\e its tem- 
 perature is being raised from 1,000' to 1,500° — viz., 500° C. — 
 
 .,, , 202,710 
 will be "yrrr" = 46 times greater than the energy imparted 
 
 during the longer period required to attain bright-red heat. 
 Let us also consider that, owing to the diminished differential 
 temperature between the burning combustible and the metal, 
 the latter will absorb proportionably less heat from the former 
 while attaining white heat than before red heat is reached; 
 
CHAP. n. nADTATTOy AT DIFFERENT TEMPEliATURES. 47 
 
 thus the fallacy of the formula becomes yet more apparent. 
 A\"e have already stated that the rate of cooling at high tem- 
 peratures may be determined with perfect accuracy, by mea- 
 suring the number of therin;il units wliich a certain amount 
 of radiating surface develops in a given time. It will be 
 shown in Chapter XIII. that fused iron maintained at a tem- 
 peratuie of 2,940° F. above the surrounding air, develops 
 I.oi;] tliernial units per minute u^Km an area of one square 
 foot ; the means employed to ascertain this fact being a calo- 
 rimeter floating on the surface of the fused metal, nearly in 
 contact with the same. -The illustration shown on Plate 6, 
 Fig. 6, represents a vertical section of a calorimeter for mea- 
 suring the radiant enei'gy of incandescent metals. It is placed 
 on the top of a block of cast iron brought to white heat, in 
 an air-furnace, the bottom of the instrument being nearly in 
 contact with the top of the incandescent metal. The nature 
 of the device, and the mode of conducting the experiment, 
 being precisely the same as shown with great minuteness in 
 Chapter XIII., it w ill only be necessary to observe, with refe- 
 rence to the apparently improtected condition of the heater, 
 that the initial temperatiire of the water contained in the same 
 is equal to the atmospheric temperature, while the powerf'ul 
 radiation from the incandescent block, together with the inter- 
 vention of the double casing, effectually prevents cooling l)y 
 external cuirents of air. Again, the duration of the exi^eri- 
 ments being restricted to a few minutes, the refrigerating in- 
 fluence of the surrounding air will be practically inappreciable. 
 The result of luimerous experiments conducted in order to 
 
48 RADIANT HEAT. chap. il. 
 
 ascertain the amount of I'adiant energy developed at various 
 differential tenaperatures from 100° F. to 2,900° F., will be 
 found by inspecting the annexed Table F, which shows the 
 number of thermal units developed in one minute by a radi- 
 ating surface of one square foot. The ordinates of the curves 
 <i I) (see diagrams Fig. 7 and Fig. 8 attached to the illustra- 
 tion Plate 7) rejjresent the radiant energy developed by 
 the differential temperatures marked on the vertical base-line 
 upon which the ordinates have been projected. It should be 
 stated that the thermal units represented by the ordinates 
 in Fig. 7 correspond with the energy developed by raising 
 the temperatxire of a quantity of water weighing 1 lb. avoir- 
 dupois 1° F. ; while the units (" caloris ") expressed by the 
 ordinates in Fig. 8 correspond with the energy developed by 
 I'aising the temperature of 1 kilogramme of water 1° Centi- 
 grade. The spaces between the ordinates Fig. 8, of course, 
 mark intervals of 100" Centigrade, the spaces between the 
 ordinates in Fig. 7 marking intervals of 100° Fahrenheit. 
 Referring to Table F, it mil be seen that the number of 
 thermal units entered in the third column is that develojied 
 during one minute on an area of one square foot ; the cor- 
 responding temperature of the radiating body being entered 
 in the second column. The fourth column shows the incre- 
 ment of dynamic energy for each 100° increase of tempera- 
 ture, also exj)ressed in thei-mal units. The fifth column con- 
 tains the rate of increment, assuming the minimum energy 
 as unit. Eegarding the diagrams Figs. 7 and 8, let us clearly 
 understand that the length of the ordinates of the curves a 
 
CHAP. II. EAI'IATIOX AT DIFFERENT TEMPERATURES. 
 
 4t) 
 
 Ti^ 
 
 Table showing teie Mechanic 
 
 AL Energy developed by 
 
 x' 
 
 11a 1)1 A XT I 
 
 E.\T AT DII'FEIIENT INTENSITIES. 
 
 
 t* ho 
 = a 
 
 g| 
 
 i i 
 
 It 
 51 
 
 (5 
 
 £P.S 
 
 c a.'-' 
 "3 II 
 
 a 
 
 £ c 3 
 
 — ■>->. 
 
 c w^ 
 
 I'M 
 
 1 = 
 
 £'3 
 
 ^ 1 
 
 it 
 
 c — 
 1 o 
 
 ti 
 
 (5 
 
 ll 
 Vi 
 
 ^o — c: 
 
 > 3 ° 
 
 c - 
 
 i 
 
 .2 a-" 
 
 a 
 
 E-S c 
 
 e E = 
 
 C.2 = 
 
 4.-5 £ 
 ■5 E 5 
 
 •Fah. 
 
 • Fah. 
 
 Thermal 
 unite. 
 
 
 T^aMo. 
 
 'Fah. 
 
 •FcA. 
 
 Thermal 
 vniti. 
 
 Thermal 
 uniU. 
 
 nat4o. 
 
 CO 
 
 2900 
 
 980.2 
 
 81.7 
 
 12.38 
 
 60 
 
 1500 
 
 209.5 
 
 27.9 
 
 4.22 
 
 CO 
 
 2800 
 
 898.5 
 
 77.2 
 
 11.70 
 
 CO 
 
 1400 
 
 181.6 
 
 25.0 
 
 3.79 
 
 CO 
 
 2700 
 
 821.3 
 
 72.8 
 
 11.03 
 
 60 
 
 1300 
 
 156.6 
 
 22.3 
 
 3.38 
 
 CO 
 
 2600 
 
 748.5 
 
 68.5 
 
 10.38 
 
 60 
 
 1200 
 
 134.3 
 
 19.8 
 
 3.00 
 
 CO 
 
 2500 
 
 680.0 
 
 64.3 
 
 9.74 
 
 60 
 
 1100 
 
 114.5 
 
 17.5 
 
 2.65 
 
 CO 
 
 2400 
 
 615.7 
 
 60.2 
 
 9.12 
 
 CO 
 
 1000 
 
 97.0 
 
 15.4 
 
 2.33 
 
 CO 
 
 2300 
 
 555.5 
 
 56.2 
 
 8.51 
 
 CO 
 
 900 
 
 81.6 
 
 13.5 
 
 2.05 
 
 60 
 
 2200 
 
 499.3 
 
 52.3 
 
 7.92 
 
 60 
 
 800 
 
 68.1 
 
 11.8 
 
 1.79 
 
 60 
 
 2100 
 
 447.0 
 
 48.5 
 
 7.35 
 
 60 
 
 700 
 
 56.3 
 
 10.3 
 
 1.5C 
 
 60 
 
 2000 
 
 398.5 
 
 44.8 
 
 6.79 
 
 60 
 
 COO 
 
 46.0 
 
 9.1 
 
 1.38 
 
 60 
 
 1900 
 
 353.7 
 
 41.2 
 
 6.24 
 
 60 
 
 500 
 
 36.9 
 
 8.2 
 
 1.24 
 
 60 
 
 1800 
 
 312.5 
 
 37.7 
 
 5.71 
 
 CO 
 
 400 
 
 28.7 
 
 7.6 
 
 1.15 
 
 CO 
 
 1700 
 
 274.8 
 
 34.3 
 
 5.20 
 
 CO 
 
 300 
 
 21.1 
 
 7.3 
 
 1.10 
 
 60 
 
 1600 
 
 240.5 
 
 31.0 
 
 4.70 
 
 CO 
 CO 
 
 200 
 100 
 
 18.8 
 6.6 
 
 7.2 
 
 1.09 
 
 • Cct 
 
 tiffrade. 
 
 
 
 * Centigrade. 
 
 15 
 
 1600 
 
 242.9 
 
 36.0 
 
 11.61 
 
 15 
 
 800 
 
 48.5 
 
 11.3 
 
 3.64 
 
 15 
 
 1500 
 
 206.9 
 
 32.1 
 
 10.35 
 
 15 
 
 700 
 
 . 37.2 
 
 9.3 
 
 3.00 
 
 15 
 
 1400 
 
 174.8 
 
 28.8 
 
 9.29 
 
 15 
 
 600 
 
 27.9 
 
 7.4 
 
 2.39 
 
 15 
 
 1300 
 
 14C.0 
 
 25.5 
 
 8.22 
 
 15 
 
 500 
 
 • 20.5 
 
 5.7 
 
 1.84 
 
 ! 15 
 
 1200 
 
 120.5 
 
 22.3 
 
 7.19 
 
 15 
 
 400 
 
 14.8 
 
 4.C 
 
 1.48 
 
 15 
 
 1100 
 
 98.2 
 
 19.5 
 
 6.29 
 
 15 
 
 300 
 
 10.2 
 
 3.7 
 
 1.19 
 
 1 ^^ 
 
 1000 
 
 78.7 
 
 16.4 
 
 5.29 
 
 15 
 
 200 
 
 6.5 
 
 3.4 
 
 1.09 
 
 1 15 
 
 900 
 
 62.3 
 
 13.8 
 
 4.46 
 
 15 
 
 100 
 
 3.1 
 
 
 
50 RADIANT HEAT. chap. ii. 
 
 b represent the nuiii];er of thermal units actually transferred 
 from one square foot of radiating surface to the fluid con- 
 tained in the calorimeter, during one minute, the temjjerature 
 of the radiator being that marked on the vertical line. It 
 should be particularly noticed that, while the energy trans- 
 ferred at 100° C. is 3.1 thermal units (caloris), see Table F, 
 it amounts to 242.9 such units at 1,600° C. ; hence the energy 
 
 242.9 
 •vvill be -; — — = 78.3 times greater at a differential "tempe- 
 3.1 
 
 rature of 1,600° C. than at 100° C. Newton's law shows 
 
 that the radiant energy augments in the ratio of 1,600 : 100 
 
 = 16 : 1. It will thus be seen that the actual increase of 
 
 7o.O , . - • T 1 
 
 energy is = 4.89 times greater tlian that assigned by 
 
 the doctrine the correctness of which our investigations tend 
 to prove. The fact should not be overlooked that the stated 
 discrepancy refers to the maximum intensity of overheated 
 fused cast iron just before the internal molecular arrangement 
 is broken up and the metal dissipated. Besides, we have 
 already pointed out that the expansion of metals is accompa- 
 nied by molecular change within the mass, augmenting the 
 energy of radiation. Nor should the fact be lost sight of 
 that Newton's doctrine takes no cognizance of such mole- 
 cular change or disturbance within the heated body. The 
 insignificance of the apparent error of the Newtonian law 
 referred to will be seen by a practical application of the 
 rival theory of MM. Dulong and Petit. Let us compare 
 the difference of energy produced at the extremes of 100° C. 
 
CHAP. II. UADIATIOX AT l)ll-FEliEyT TEMPERATURES. •''1 
 
 and 1,500° C. ditferential temperature established by their re- 
 searches. According to the tables, accepted by Prof. Stewart 
 and others, contained in the second part of Duloug and Petit's 
 famous \\ ork, " The Laws of Refrigeration," the rate of co(.)l- 
 ing at a differential temperature of 100^ C. is 2°.30 C. in 
 one minute (the surrounding medium being maintained at 
 the freezing point of water) ; while at a differential tempe- 
 rature of 240° C. the rate is stated to be 10°.()9 C. Applying 
 these rates to the fonnula of Dulong and Petit, it will be 
 found that when the differential temperature is 1,500° C. the 
 fall will be 202,710° C. in one minute. The radiant energy 
 
 202,71(» 
 parted with at 1,500° C. will accordingly be — ■ = 88,1 3y 
 
 times greater than at 100 C. But our tables and diagrams, 
 based on actual trial, show that the radiant energy at 1,500' C. 
 
 is only - — '— = 66.7 times greater than at 100° C. Hence the 
 
 radiant energy at 1,500° C, agreeably to Dulong and Petit's 
 
 theoiT, will l)e — '—— = 1,321 times higher than that estab- 
 •'' 66.7 
 
 lished by our elaborate practical investigation. 
 
CHAPTER III. 
 
 INTENSITY OP SOLAR RADIATION. 
 
 The illusti-iitlon shown on Plate 8 represents a vertical 
 section of an instrument constnicted for ascertaining, by a 
 new and exact method, the intensity of solar radiation at the 
 surface of the eai-th, specially arranged for revolving obser- 
 vatories. Sir John Herschel's definition of the word actino- 
 meter — " an instrument for measuring the intensity of heat in 
 the sun's rays " — warrants the adoption of that term. 
 
 The caiises which modify the intensity of solar radiation 
 are chiefly : the position of the eai-th in its orbit, the sun's 
 zenith distance — on which depends the depth of the atmosjihere 
 to be penetrated by the rays — and vapors in the atmosphere. 
 The temperature of surrounding objects which radiate towards 
 substances exposed to the sun's rays, and the heat abstracted 
 from such substances by currents of air, present serious dis- 
 turbing elements, rendering an accui-ate determination of the 
 radiant intensity by ordinary thei'mometers practically impos- 
 sible. It is haiilly necessary to point out that solar inten- 
 
CHAP. III. iyTi:NSITy OF SOLAR KADIATION. 5:J 
 
 sity cannot be satisfactorily ascertaiuetl by tlio old niethoil 
 of deducting the temperature of a thermometer in the .shade 
 from that of another thermometer exposed to the sun. The 
 investigations of Daniell relating to the sun's radiant lu-at, 
 fiequently referred to in works on meteorology, conducted in 
 the latitude of London, where the depth of the atmo.sphere at 
 noon, during the summer solstice, is 0.57 greater than on the 
 ecliptic, merit serious consideration. The subjoined table con- 
 tains the result of his observations on solar radiation through- 
 out a day in the month of June. 
 
 A glance at this table sLows that, according to the adopted 
 method of determining the intensity of solar radiation by de- 
 ducting the temperature indicated by a thermometer in the 
 shade from the temperature attained in the sun, the ladiant 
 heat is considerably less before than after noon. The diffe- 
 rential temperature, or solar intensity, at 9 A.jr., according to 
 this table, is 25°, while at 3 p.ji., with an equal zenith distance 
 and equal depth of atmosphere to penetrate, the solar intensity 
 is stated to be 62°, thus exhibiting the enornu)US difference of 
 27°. An explanation of the causes of the extraordinary errora 
 of Daniell's table is scarcely needed, but attention should be 
 called to the gross imperfection of such a mode of detennin- 
 infj solar radiation as that of noting the different indications 
 of shaded and exposed theiTnometere. During the early stages 
 of my investigation relating to the mechanical properties of 
 the sun's radiant heat, I adopted this mode of ascertaining 
 the temperature produced by solar radiation ; but notwith- 
 standing numerous expedients lesoited to in order to prevent 
 
54 
 
 BADIANT UEAT. 
 
 Time of obser- 
 vation. 
 
 TEMPERATURE, A.M. 
 
 Differential 
 
 temperatui'e or solar 
 
 intensity. 
 
 In the sun. 
 
 In the shade. 
 
 'Fah. 
 
 °Cmt. 
 
 ° Fah. 
 
 °Cmt. 
 
 ' Fah. 
 
 ' Cent. 
 
 9.00 
 
 93 
 
 33.88 
 
 68 
 
 20.00 
 
 25 
 
 13.88 
 
 9.30 
 
 103 
 
 39.44 
 
 69 
 
 20.55 
 
 34 
 
 18.89 
 
 10.00 
 
 111 
 
 43.88 
 
 70 
 
 21.11 
 
 41 
 
 22.77 
 
 10.30 
 
 119 
 
 48.33 
 
 71 
 
 21.66 
 
 48 
 
 26.67 
 
 11.00 
 
 124 
 
 SI. 11 
 
 71 
 
 21.66 
 
 53 
 
 29.45 
 
 11.30 
 
 125 
 
 51.66 
 
 72 
 
 22.21 
 
 53 
 
 29.45 
 
 12.00 
 
 129 
 
 53.88 
 
 73 
 
 22.77 
 
 56 
 
 31.11 
 
 1. 
 
 ° 
 
 «H .2 
 
 TEMPERATURE, P.M. 
 
 Differential 
 
 temperature or solar 
 
 intensity. 
 
 In the sun. 
 
 In the shade. 
 
 'Fah. 
 
 ° Cent. 
 
 'Fah. 
 
 °Omi. 
 
 'Fah. 
 
 ° Cent. 
 
 12.30 
 
 132 
 
 55.55 
 
 74 
 
 23.33 
 
 58 
 
 32.22 
 
 1.00 
 
 141 
 
 60.55 
 
 74 
 
 23.33 
 
 67 
 
 37.22 
 
 1.30 
 
 140 
 
 60.00 
 
 75 
 
 23.88 
 
 65 
 
 36.12 
 
 2.00 
 
 143 
 
 61.66 
 
 75 
 
 23.88 
 
 68 
 
 37.78 
 
 2.30 
 
 138 
 
 58.88 
 
 76 
 
 24.44 
 
 62 
 
 34.44 
 
 3.00 
 
 138 
 
 58.88 
 
 76 
 
 24.44 
 
 62 
 
 34.44 
 
 3.30 
 
 132 
 
 55.55 
 
 77 
 
 25.00 
 
 55 
 
 30.55 
 
 4.00 
 
 124 
 
 51.11 
 
 76 
 
 24.44 
 
 48 
 
 26.67 
 
 4.30 
 
 123 
 
 50.55 
 
 77 
 
 25.00 
 
 46 
 
 25.55 
 
 5.00 
 
 112 
 
 44.44 
 
 76 
 
 24.44 
 
 36 
 
 20.00 
 
 5.30 
 
 106 
 
 41.11 
 
 75 
 
 23.88 
 
 31 
 
 17.23 
 
 6.00 
 
 100 
 
 i 
 
 37.77 
 
 73 
 
 22.77 
 
 27 
 
 15.00 
 
CHAP. in. INTEXSITT OF SOLAR RADIATION. 55 
 
 the thermometers from being unduly influenced by the I'adiant 
 beat of the air and surrounding objects, I failed to .secure 
 satisfactory results. The most important point — the ctjntml- 
 ling the irregular action of the surrounding air, wliirli ;itl'crt> 
 the exposed as well as the shaded thermometer — having pre- 
 sented obstacles which no mechanical airangement whatever 
 could overcome, I adopted the method of wholly excluding 
 the atmosphere. By this expedient the bull) of the thermo- 
 meter within the instiniment becomes surrounded by ether 
 only ; hence the energy transmitted by the sun's rays deter- 
 mines the temperature freed from atmospheric influence. It 
 will be objected that the thermometer cannot be applied 
 within a vacuum without the employment of some transpa- 
 rent covering, and that, conseqixently, the energy of the rays 
 will suffer considerable loss before reaching the bulb. This 
 objection is readily met by applying a thin lens of about 50 
 ins. focus, inserted at such a distance tVom the bulb that the 
 gain effected by concentrating the rays will exactly balance 
 the loss of calorific energy attending their passage through 
 the lens. It is evident, however, that a plane crystal might 
 be employed in place of the lens, provided its al)Sorptive 
 power were known. I have accordingly constnicted an appa- 
 ratus for measuring the loss of calorific energy attending the 
 passage of the sun's rays through plane crystals of adequate 
 thickness to resist the atmospheric pressure when the air is 
 withdrawn from the interior chamber of the instrument. It 
 should be mentioned that many of the observations recorded 
 in this work have been made with an actinometer provided 
 
56 BADIAXT HEAT. CHAP. ill. 
 
 with a plane crystal, tlie absorptive power of wliicli amounts 
 to 0.066. Veiy little trouble has been experienced in record- 
 ing observations made with this instrument, the indicated dif- 
 ferential temperatures having simply been multiplied by the 
 coefficient 1.066. 
 
 Careful inspection of our illustration, together with the 
 foregoing explanations regarding the nature of the lens and 
 its substitute, and the object of applying the recording ther- 
 mometer within a vacuum, render a minute description of the 
 detail of the device unnecessary. Like the solar calorimeter, 
 Plate 10, particularly described in Chapter V., the actino- 
 meter is attached to a table, the face of which is kept per- 
 pendicular to the sun during observations. It is also provided 
 with a graduated arc and stationary index, similar to those 
 applied to the solar calorimeter, by means of which the sun's 
 zenith distance may be ascertained at every instant. The 
 chamber containing the bulb of the thermometer is 4f ins. 
 in diameter, plated with polished nickel and surrounded with 
 a double casing, through which a current of water is circu- 
 lated by means precisely like those employed in the solar 
 calorimeter, the vacuum being also produced in a similar 
 manner, and flexible tubes employed for connecting the in- 
 strument with the stationary pump. The cistern which sup- 
 plies the circulating water is kept at a constant tempera- 
 ture of 60° F., and, in order to secure perfect accuracy, the 
 thermometer employed for regulating this temperature is so 
 applied that the return current from the actinometer to the 
 cistern circulates round its bulb. A thin metallic screen of 
 
CHAP. in. INTEXSITV OF SOLAh' i:.\l>IATI<)X. 5'! 
 
 aiimilar form, supported by four columus, and plated witli 
 silver, protects the instrument from the sun's radiant heat, 
 for the puqiose of economizing the cooling medium required 
 to keep the circulating water at tlie proper temperature. 
 The opening in the screen corresponds with the size of the 
 lens. The bulb of the thermometer is 3 ins. in length, in 
 order to expose a relatively large surface to the action of 
 the solar rays ; the proportion of heating surface to the con- 
 tents of the bulb being thus much greater than in ordinary 
 thermometers. The upper half of the bulb is coated with 
 l;iiii2)ldack, the lower half being exposed to the action of 
 the reflected raya fi'om the bottom of the chamber, in such 
 a manner that the radiation of this lower bright half of the 
 bulb is neutralized by reflected heat. Much pains has been 
 bestowed in order to attain this desirable object. Before 
 making an observation with the actinometer, the vacuum 
 gauge should be inspected, and the water from the cistern 
 be permitted to run freely throxigh the. casing for several 
 minutes until the temperature of the return current an<l that 
 indicated by the enclosed thermometer correspond. The ob- 
 servatory being then turned to the sun, and the declination 
 table on which the actinometer stands adjusted to the proper 
 angle, the cover over the lens should be removed. The 
 height of the mercurial column of the enclosed thermometer 
 will then, in due time, indicate Avith absolute certainty the 
 intensity of the sun's radiant heat, independent of atmospheric 
 temperature and other disturbing causes which lender the 
 indications of solar intensity by common thermometers mere 
 
58 RADIANT MEAT. chap. hi. 
 
 approximations. It will be evident that, since the instru- 
 ment is kept at a constant temperature of 60°, it continually 
 radiates heat of that intensity towards the bulb of the en- 
 closed thermometer ; hence the zero of the thermoraetric 
 scale of the actlnoraeter will mark 60° above Fahrenheit's 
 zero. Obviously, the point reached by the mercurial column 
 of the enclosed thermometer above the stated zero on expo- 
 sure to the sun's rays can only be attained in virtue of the 
 power of unaided solar radiation. In supjiort of this asser- 
 tion it may be well to repeat the explanation already given, 
 namely, that the bulb is surrounded by ether alone, freed 
 fi'om all disturbing influences of ponderous matter, and that 
 the heat which determines the zero of the actinometer is sup- 
 plied by radiation from the instalment itself. The important 
 question of actual intensity with reference to the adopted 
 " absolute " zero will be discussed in Chapters IX. and XVI. ; 
 but it may be well to state in this place that the actino- 
 meter merely shows the thermometric interval of solar inten- 
 sity on Fahrenheit's scale, without reference to the position 
 of that interval on a scale which commences at the accepted 
 " absolute zero." I regard this absolute zero, however, as an 
 ignis-fatvv8, retreating as fast as we approach it. 
 
 It should be observed that all the actinometric observa- 
 tions contained in this work have been made in lat. 40 deg. 
 42 min., thus only 17 deg. 12 min. from the trojjic of Cancer. 
 The depth of atmosphere so near the tropics being, at mid- 
 summer, only 0.047 greater than on the ecliptic, while the 
 sun's zenith distance on lat. 40 dea:. 42 min. duriuc the winter 
 
CHAP. III. lA'TEXSITl' OF SOLAR HADIATION. 59 
 
 solstice is only 2 cleg. 18 min. less thau at the pole at mid- 
 summer, I have lieen enabled to determine the maximum 
 intensity of solar radiation for all latitudes from the equator 
 to the ])ole. The diagram represented on PI. 9, Fig. 1, shows 
 the relations of atmospheric depth and solar intensity for each 
 degree of zenith distance from the vertical to the 75th dei^ree. 
 A brief description will suffice to render this diagram readily 
 understood. The ordinates between the curve e a and the 
 base line / <y exhibit the true proportions of the depth of the 
 atmosphere penetrated by the rays from the vertical to 75 deg. 
 zenith distance ; while the ordinates of the curve c a indi- 
 cate tlie relative intensity of the sun's I'adiaut heat at the 
 summer solstice for each degree of the sun's zenith distance 
 from the vertical to 75 deg. The straight line bah the 
 tangent of the cuives e a and c a. It will be seen by closely 
 examining these curves, and the ordinates resting on the base 
 line / g (comparing the same with the figures in the tables 
 of this chapter), that the intensities of solar radiation var}' 
 nearly in the inverse ratio of the cube roots of the atmospheric 
 depth for all zenith distances not exceeding 75 deg. The ordi- 
 nates between the irregular line d d d and the base line /' <j 
 show the solar intensity for each degree of zenith distance 
 from 23 deg. to 75 deg., ascertained by actinometric observa- 
 tion during a day in the month of August when the sun was 
 obscured by cirri of average density. AVith refei-ence to the 
 solar engine this irregular line d d d possesses great interest, 
 as it indicates the available solar energy, foi' mechanical [lur- 
 poses, during a day A\hen the sun is partially obscured. 
 
60 BABIANT HEAT. chap. hi. 
 
 The engineer will regard, this diagram as a solar indicator- 
 card, the space below the line d d d representing the avail- 
 able power, while the sj)ace contained between that line and 
 the curve c a indicates the loss. For the purpose of elucida- 
 tion, the Xorth Pole, together ^vith the cities of Edinburgh, 
 London, Paris, and. New York, have been introduced on this 
 diagram, their positions having reference solely to the depth 
 of atmosphere and solar intensity during the summer solstice. 
 The annexed Tables A, B, and C show the relative de^Jth of 
 atmosphere and maximum solar intensity at midsummer fiir 
 each degree of the sun's zenith distance from the vertical to 
 75 deg. 
 
 Referring to the second part of the diagram, it will be 
 found that Fig. 2 contains a delineation of the graduated 
 plate specially mentioned in Chapter V. This plate is fur- 
 nished with a movable radial index, to enable the observer 
 to ascertain quickly the depth of atmosphere corresponding 
 with observed zenith distances. The gradiiated plate is con- 
 structed to a scale of 24 miles to the inch, the curvature 
 of the earth's surface and the atmospheric boundary (sup- 
 posed to extend 42 miles above the earth) being accurately 
 laid down according to the said scale. The vei'tical depth 
 of the atmosjDhere, it will be seen, has been divided into 100 
 equal parts, the same graduation having been introduced on 
 the movable index. Accordingly, by placing this index at 
 angles corresponding with the observed zenith distance, the 
 intersection of its upper edge with the top line of the atmo- 
 sphere will show the proportion of diagonal and vertical 
 
IXTEXSITY OF SOLA R ItA DIA TION. 
 
 fil 
 
 A 
 
 TaRLE SllOWIXG THE InTEXSITY OF 
 
 Solar U. 
 
 VDIATIOS ANM) TUE 
 
 Atmospheric Depth for given Zenith Distances. 
 
 
 Atmospheric 
 depth. 
 
 Increment 
 
 of atmospheric 
 
 deptti. 
 
 Maximum 
 
 intensity. 
 
 Observed intensity 
 
 during 
 
 a partially cloudy day. 
 
 Dtg. 
 
 Ktlativf. 
 
 Jielallrt. 
 
 • Fah. 
 
 •Cdiit. 
 
 •Fah. 
 
 • Cent. 
 
 
 
 1.000 
 
 0.000 
 
 67.20 
 
 37.33 
 
 
 
 1 
 
 1.000 
 
 0.000 
 
 67.19 
 
 37.33 
 
 
 
 
 2 
 
 1.000 
 
 0.000 
 
 67.18 
 
 37.32 
 
 
 
 
 3 
 
 1.001 
 
 0.001 
 
 67.16 
 
 37.31 
 
 
 
 
 4 
 
 1.002 
 
 0.002 
 
 67.14 
 
 37.30 
 
 
 
 
 5 
 
 1.003 
 
 0.003 
 
 67.11 
 
 37.28 
 
 
 
 
 G 
 
 1.005 
 
 0.005 
 
 67.07 
 
 37.26 
 
 
 
 
 7 
 
 1.007 
 
 0.007 
 
 67.02 
 
 37.24 
 
 
 
 
 8 
 
 1.010 
 
 0.010 
 
 66.97 
 
 37.21 
 
 
 
 
 9 
 
 1.013 
 
 0.013 
 
 66.91 
 
 37.17 
 
 
 
 
 10 
 
 LOIG 
 
 0.016 
 
 66.84 
 
 37.13 
 
 
 
 
 11 
 
 1.019 
 
 0.019 
 
 66.77 
 
 37.09 
 
 
 
 
 12 
 
 1.023 
 
 0.023 
 
 66.69 
 
 37.05 
 
 
 
 
 13 
 
 1.027 
 
 0.027 
 
 66.60 
 
 37.00 
 
 
 
 
 14 
 
 1.031 
 
 0.031 
 
 66.51 
 
 36.95 
 
 
 
 
 15 
 
 1.030 
 
 0.036 
 
 66.41 
 
 36.89 
 
 
 
 
 16 
 
 1.041 
 
 0.041 
 
 66.30 
 
 36.83 
 
 
 
 
 17 
 
 1.046 
 
 0.046 
 
 66.19 
 
 36.77 
 
 
 
 
 18 
 
 1.051 
 
 0.051 
 
 66.07 
 
 36.70 
 
 
 
 
 19 
 
 1.057 
 
 0.057 
 
 65.94 
 
 36.63 
 
 
 
 
 20 
 
 1.063 
 
 0.063 
 
 65.80 
 
 36.56 
 
 
 
 
 21 
 
 1.070 
 
 0.070 
 
 65.66 
 
 36.48 
 
 
 
 
 22 
 
 1.077 
 
 0.077 
 
 65.51 
 
 36. 4C 
 
 
 
 
 23 
 
 1.085 
 
 0.085 
 
 65.36 
 
 36.31 
 
 5 
 
 0.5 
 
 33.05 
 
 24 
 
 1.093 
 
 0.093 
 
 65.20 
 
 36.22 
 
 58.4 
 
 32.44 
 
 25 
 
 1.102 
 
 0.102 
 
 1 
 
 65.03 
 
 36.13 
 
 57.7 
 
 - 32.04 
 
62 
 
 BABIANT HEAT. 
 
 B 
 
 Table siio\vi\g the Intensity of Solak Eadiation 
 
 \NI) THE 
 
 Atmusi'iiekic Depth for given Zenith Distances. 
 
 ^3 
 
 Atmospheric 
 depth. 
 
 Increment 
 
 of atmospheric 
 
 depth. 
 
 Maximum 
 
 intensity. 
 
 Observed intensity 
 
 during 
 
 a partially cloudy day. 
 
 Dig. 
 
 Relalviie. 
 
 JielatiTi. 
 
 ' Fah. 
 
 °Cmt. 
 
 ° Fah. 
 
 'CeiU. 
 
 26 
 
 1.111 
 
 0.111 
 
 64.85 
 
 36.03 
 
 58.4 
 
 32.44 
 
 27 
 
 1.121 
 
 0.121 
 
 64.67 
 
 35.93 
 
 57.5 
 
 31.95 
 
 28 
 
 1.132 
 
 0.132 
 
 64.48 
 
 35.82 
 
 57.7 
 
 32.05 
 
 29 
 
 1.141 
 
 0.141 
 
 64.29 
 
 35.72 
 
 57.7 
 
 32.05 
 
 30 
 
 1.152 
 
 0.152 
 
 04.08 
 
 35.60 
 
 57.8 
 
 32.11 
 
 31 
 
 1.104 
 
 0.164 
 
 63.87 
 
 35.49 
 
 56.9 
 
 31.61 
 
 32 
 
 1.176 
 
 0.176 
 
 63.66 
 
 35.36 
 
 56.9 
 
 81.61 
 
 33 
 
 1.189 
 
 0.189 
 
 63.43 
 
 35.24 
 
 56.7 
 
 31.50 
 
 34 
 
 1.203 
 
 0.203 
 
 63.20 
 
 35.11 
 
 57.0 
 
 31.66 
 
 35 
 
 1.217 
 
 0.217 
 
 62.96 
 
 34.98 
 
 56.8 
 
 81.55 
 
 36 
 
 1.232 
 
 0.232 
 
 62.72 
 
 84.84 
 
 56.4 
 
 31.38 
 
 37 
 
 1.248 
 
 0.248 
 
 62.47 
 
 34.70 
 
 56.2 
 
 31.22 
 
 38 
 
 1.265 
 
 0.265 
 
 62.21 
 
 34.56 
 
 55.8 
 
 81.00 
 
 39 
 
 1.283 
 
 0.283 
 
 61.94 
 
 34.21 
 
 56.1 
 
 31.17 
 
 40 
 
 1.302 
 
 0.302 
 
 61.67 
 
 34.26 
 
 55.0 
 
 30.55 
 
 41 
 
 1.322 
 
 0.322 
 
 61.39 
 
 34.11 
 
 54.8 
 
 80.44 
 
 42 
 
 1.342 
 
 0.342 
 
 61.10 
 
 38.95 
 
 54.8 
 
 80.44 
 
 43 
 
 1.363 
 
 0.363 
 
 60.81 
 
 38.78 
 
 54.9 
 
 30.50 
 
 44 
 
 1.884 
 
 0.384 
 
 60.51 
 
 33.62 
 
 54.6 
 
 80.34 
 
 45 
 
 1.406 
 
 0.406 
 
 60.20 
 
 38.45 
 
 54.7 
 
 30.40 
 
 46 
 
 1.431 
 
 0.431 
 
 59.88 
 
 38.27 
 
 54.5 
 
 30.28 
 
 47 
 
 1.457 
 
 0.457 
 
 59.56 
 
 83.09 
 
 54.4 
 
 30.22 
 
 48 
 
 1.485 
 
 0.485 
 
 59.23 
 
 32.91 
 
 53.4 
 
 29.66 
 
 49 
 
 1.514 
 
 0.514 
 
 58.89 
 
 82.72 
 
 53.0 
 
 29.44 
 
 50 
 
 1.545 
 
 0.545 
 
 58.54 
 
 32.52 
 
 58.2 
 
 29.55 
 
Ii\TE.\SITV OF iiOLAh' KADIATION. 
 
 U3 
 
 c 
 
 Taulk siio\vix(; tiii; In'tkn'sity of Solak U 
 
 VDIATION 
 
 AND TirE 
 
 Atmospiikuic Dki'tu foh given Zenith Distances. 
 
 II 
 
 Atmospheric 
 depth. 
 
 Incrempiil 
 
 of atruospheric 
 
 depth. 
 
 Maximum 
 
 iiUensit)'. 
 
 Obscrve<l intensity 
 
 during 
 
 a partially cloudy day. 
 
 Dtg. 
 
 lUlatite. 
 
 Belulirt. 
 
 • Fah. 
 
 •Cent. 
 
 'Fah. 
 
 'Cent. 
 
 51 
 
 1.577 
 
 0.577 
 
 58.18 
 
 32.32 
 
 50.7 
 
 28.16 
 
 5:2 
 
 1.0J2 
 
 0.GJ2 
 
 57.81 
 
 32.12 
 
 50.6 
 
 28.11 
 
 m 
 
 1.648 
 
 0.648 
 
 57.44 
 
 31.91 
 
 45.8 
 
 25.44 
 
 64 
 
 1.686 
 
 0.686 
 
 57.05 
 
 31.70 
 
 45.4 
 
 25.22 
 
 55 
 
 1.726 
 
 0.726 
 
 56.66 
 
 31.48 
 
 44.6 
 
 24.77 
 
 56 
 
 1.769 
 
 0.796 
 
 56.25 
 
 31.25 
 
 44.6 
 
 24.77 
 
 57 
 
 1.815 
 
 0.815 
 
 55.82 
 
 31.02 
 
 47.0 
 
 26.11 
 
 58 
 
 1.864 
 
 0.864 
 
 55.39 
 
 30.77 
 
 48.0 
 
 26.66 
 
 69 
 
 1.916 
 
 0.916 
 
 54.94 
 
 30.52 
 
 47.3 
 
 26.27 
 
 60 
 
 1.970 
 
 0.970 
 
 54.47 
 
 30.26 
 
 47.4 
 
 26.32 
 
 61 
 
 2.037 
 
 1.037 
 
 53.99 
 
 29.99 
 
 46.4 
 
 25.77 
 
 62 
 
 2.098 
 
 1.098 
 
 53.48 
 
 29.72 
 
 46.8 
 
 26.00 
 
 63 
 
 2.164 
 
 1.164 
 
 52.96 
 
 29.42 
 
 4G.S 
 
 26.00 
 
 64 
 
 2.235 
 
 1.235 
 
 52.41 
 
 29.12 
 
 46.4 
 
 25.77 
 
 65 
 
 2.312 
 
 1.312 
 
 51.85 
 
 28.81 
 
 46.1 
 
 25.50 
 
 66 
 
 2.398 
 
 1.398 
 
 51.26 
 
 28.48 
 
 45.0 
 
 25.00 
 
 67 
 
 2.490 
 
 1.490 
 
 60.63 
 
 28.13 
 
 42.6 
 
 23.66 
 
 68 
 
 2.591 
 
 1.591 
 
 49.96 
 
 27.70 
 
 43.1 
 
 23.94 
 
 69 
 
 2.701 
 
 1.701 
 
 49.24 
 
 27.36 
 
 43.2 
 
 24.00 
 
 70 
 
 2.821 
 
 1.821 
 
 48.43 
 
 26.93 
 
 42.8 
 
 23.77 
 
 71 
 
 2.952 
 
 1.952 
 
 47.67 
 
 26.46 
 
 41.9 
 
 23.27 
 
 72 
 
 3.097 
 
 2.097 
 
 46.72 
 
 25.96 
 
 40.4 
 
 22.44 
 
 73 
 
 3.255 
 
 2.255 
 
 45.72 
 
 25.40 
 
 33.5 
 
 18.61 
 
 74 
 
 3.428 
 
 2.428 
 
 44.61 
 
 24.79 
 
 36.3 
 
 20.16 
 
 76 
 
 3.624 
 
 2.624 
 
 43.39 
 
 24.11 
 
 32.4 
 
 18.00 
 
64 RABIAXT HEAT. chap. hi. 
 
 atmospheric depth. The impoi'taiit rehntioiis of solar inten- 
 sity, zenith distance, atmospheric depth, and hititude, being 
 exhibited by the diagram under consideration, let us now 
 consider the mode adopted in constructing the same. 
 
 The data indispensable in constructing the curve c a, the 
 ordiuates of which represent the maximum solar intensity for 
 given zenith distances when the earth is in aphelion, first claim 
 our attention. It will be perceived, on reflection, that, owing 
 to the vai'ying intensity resulting from change of distance be- 
 tween the sun and the earth, the temjierature produced by 
 solar radiation varies fi'om day to day with the altered posi- 
 tion of the earth in its orbit. Consequently, observed defi- 
 ciency of solar intensity at any given zenith distance does 
 not always furnish, as supposed by certain meteorologists, a 
 correct indication of the amount of absorption caused by the 
 presence of vapor in the atmosphere. 01)viously, the observed 
 deficiency of intensity may result partially from the earth's 
 proximity to the aphelion. It will be seen by reference to 
 Chapter IV. that for equal zenith distance a diminution of 
 temperature of 4°. 66 F. takes jilace during the summer sol- 
 stice, compared with the intensity of the radiant heat at mid- 
 winter. Consequently, if we omit to make a proper allow- 
 ance for the difference of intensity resulting from the sun's 
 distance at the time of observation, no satisfactory record can 
 be produced. It is hardly necessary to point out that, unless 
 we establish some fixed position of the earth in its orbit, as a 
 zero, it will be impossible to construct tables of varying solar 
 intensity cajjable of being employed as a means of correction. 
 
OHAi'. 111. INTENSITY OF tSOLAR RADIATION. 65 
 
 I have accordingly adopted the aphelion as the controlling 
 zero, all my tables relating to this subject having reference 
 to maadmum solar inteimtij when the earth is furthest from 
 the svn. The reader will iind the question of solar distance 
 fully discussed in the next chapter. 
 
 Regarding the observations which have furnished the data 
 on which our diagram has been constructed and the annexed 
 tables calculated, it \\ill be necessar)' to state, for the infor- 
 mation of those who are not familiar with the subject, that 
 observations of maximum solar intensity should be continued 
 during a series of yeai-s. It sometimes happens that an entire 
 season elapses without a single opportunity for a satisfactory 
 observation presenting itself. Records of solar observations 
 are therefore exceedingly irregular ; during some seasons reli- 
 able observations may be made fi'om day to day ; then again 
 a considerable period intervenes during which the work must 
 be wholly suspended. Indeed, the difficulties inseparable from 
 investigations of maximum solar energy can hardly be exag- 
 gerated. A complete record requires that the sun should be 
 perfectly clear while we observe the temperature produced 
 by the rays for each degree of the sun's zenith distance, upon 
 each degree of latitude, for each day in the year. Now, 
 observations continued during a century would not suffice 
 to produce such a record ; hence I have had recoui-se to 
 the graphic method, projecting from a base line drawn on a 
 large diagram ordinates representing the maximum intensity 
 shown by each satisfactory observation. The position of the 
 ordinates thus projected, it is scarcely necessarj'^ to observe, 
 
66 BADIANT HEAT. CHAP. ill. 
 
 depends on the zenith distance under which the respective 
 observations are made, while their length will be deter- 
 mined by the observed maximum temperature, to be marked 
 on the diagram in accordance with a fixed scale. As the 
 recoi'ded investigations extend from the vertical to V5 deg. 
 zenith distance, it will be evident that the base line should 
 be divided into 75 equal parts, each division representing 
 one degree of zenith distance. Tor each successful observa- 
 tion an ordinate will be projected at a position on the base 
 corresponding with the zenith distance under which the suc- 
 cessful observation has been made ; the length of the ordi- 
 nate being marked off according to the fixed scale mentioned. 
 It will be perceived, therefore, that, at the termination of a 
 series of observations, the base line on the diagram, with its 
 75 equal divisions, representing degrees of zenith distance, 
 subdivided into minutes, will be studded with, a number of 
 perpendicular lines of unequal length, placed at irregular dis- 
 tances. In accordance with the rules of the graphic system, 
 the terminations of the several irregularly-spaced perpendi- 
 cular lines will then be connected by a curved line which, 
 if uniform and nearly parabolic when completed, proves that 
 the observations have been accurate. Should it, however, 
 contain breaks, or should its curvature not be gradually in- 
 creasing with increased zenith distance, in accordance with 
 an ascending series, fresh observations must be made at zenith 
 distances embracing the defective portions. Again, it may 
 happen that in connecting the terminations of the ordinates 
 considerable gaps present themselves ; in other words, that 
 
CHAP. III. INTENSirV OF SOLAE RADIATION. C7 
 
 want of observatiou occurs for several succeeding degrees of 
 zenith distance. These gaps must be filled by fresh observa- 
 tions, unless the completed parts of the curve on both sides 
 of the gap, when extended over it, meet in such a manner as 
 to produce a consistent curve. Our small diagram (Plate 9) 
 has been copied from a large diagram constructed in accor- 
 dance -^vith the foregoing explanation of the plan adopted. 
 The scale employed in laying down the observed tempera- 
 tures, viz., marking oft" the length of the ordinates, being 
 sufiiciently large to admit of sho\ving fractious of a degree 
 of Fahrenheit, the temperatures entered in the Tables A, B, 
 and C, for given zenith distances, will be found very pre- 
 cise. In view of the foregoing explanation of the procedure 
 resorted to, the reader need not be reminded that the tem- 
 peratures appearing in the tables have not all been deter- 
 mined by actual observation. Agreeably to the graphic 
 system, several of these temperatures have been determined 
 by measuring the height of the ordinates in the diagram. 
 It needs no demonstration, however, to prove that intensities 
 ascertained by measurement, in the manner pointed out, are 
 as reliable as if ascertained directly by the actinometer, pro- 
 vided the curve be perfect which connects the termination 
 of the ordinates obtained by actual observation. It should 
 be stated that, during a period embracing many yeai-s, but 
 few favorable opportunities have been neglected of verifying 
 the coiTectness of the preceding tables. No material discre- 
 pancies having been obsei-\-ed, future investigations are not 
 likely to lead to any important modifications of these tables. 
 
68 RADIANT HEAT. chap. hi. 
 
 Possibly observations conducted on the table-lands of India 
 might show somewhat higher intensities for given zenith dis- 
 tances ; but tables modified agreeably to such observations 
 would not be as useful for meteorological investigations con- 
 cerning the effects of solar intensity in America and Europe 
 as those here presented. It may be mentioned that it was 
 my original intention to extend the observations of zenith dis- 
 tance and solar intensity from the 75th deg. to the horizon; 
 l)ut exjierience has shown that both the eastern and western 
 horizon, viewed fi'om my present observatory, are so seldom 
 free from clouds and haze that too long a time would elapse 
 before the investigations of atmospheric absorption at extreme 
 zenith distance could be completed. It will be found, how- 
 ever, on mature reflection, that the most important meteoro- 
 logical phenomena connected with solar heat are confined 
 within the vertical segment of 150 deg., which embi'aces the 
 result of my actinometric observations. 
 
 With reference to the maximum intensity of solar radia- 
 tion on each degree of latitude, it will be evident that, since 
 we jiossess an accurate knowledge of the relation of zenith " 
 distance and temperature, we can readily determine the maxi- 
 mum intensity for different latitudes during the summer sol- 
 stice. For instance, we know, by referring to the preceding 
 Table B, that, when the earth is in aphelion and tlie zenith 
 distance is 43 deg., the temperature produced by solar radi- 
 ation is 60°.81 F. ; hence we know that this temperature 
 marks maximum solar intensity on the Arctic Circle, the 
 latter being 43 deg. from the ecliptic at the summer sol- 
 
INTENSITY OF SOL Alt RADIATION. 
 
 69 
 
 1 
 Table D, showing the 
 
 Temperature produced by 8oi,ar IUdi- 
 
 ATION AT NOOX FOR 
 
 EACH Degree of Latitude whkn the 
 
 EaIITH is IX Al'HKLIOX. NORTHERN IIeMISPHEIIE. 
 
 Equator 
 
 Lat. 
 
 Solar intensity at 
 noon. 
 
 Lat. 
 
 Solar intensity St 
 noon. 
 
 Veg. 
 
 •Fah. 
 
 •Cent. 
 
 Dtg. 
 
 'Fah. 
 
 ' Ctnt. 
 
 
 
 65.30 
 
 36.28 
 
 24 
 
 67.20 
 
 37.33 
 
 
 1 
 
 65.4") 
 
 36.36 
 
 25 
 
 67.19 
 
 37.32 
 
 
 2 
 
 65.60 
 
 36.44 
 
 26 
 
 67.18 
 
 37.32 
 
 
 3 
 
 65.75 
 
 36.52 
 
 27 
 
 67.17 
 
 37.31 
 
 
 4 
 
 65.89 
 
 36.60 
 
 28 
 
 67.14 
 
 37.30 
 
 
 5 
 
 66.02 
 
 36.68 
 
 29 
 
 67.10 
 
 37.28 
 
 
 6 
 
 66.15 
 
 36.75 
 
 80 
 
 67.05 
 
 37.25 
 
 
 7 
 
 66.27 
 
 36.82 
 
 31 
 
 66.99 
 
 37.21 
 
 
 8 
 
 66.39 
 
 36.88 
 
 32 
 
 66.93 
 
 37.18 
 
 
 9 
 
 66.49 
 
 36.94 
 
 83 
 
 66.87 
 
 37.15 
 
 
 10 
 
 66.58 
 
 36.99 
 
 34 
 
 66.80 
 
 37.11 
 
 
 11 
 
 66.66 
 
 37.03 
 
 35 
 
 66.73 
 
 37.07 
 
 
 12 
 
 66.73 
 
 37.07 
 
 36 
 
 66.66 
 
 37.03 
 
 
 13 
 
 66.80 
 
 37.11 
 
 37 
 
 66.58 
 
 36.99 
 
 
 14 
 
 66.87 
 
 37.15 
 
 38 
 
 66.49 
 
 36.94 
 
 
 15 
 
 66.93 
 
 37.18 
 
 39 
 
 66.39 
 
 36.88 
 
 
 16 
 
 66.99 
 
 37.21 
 
 40 
 
 66.27 
 
 36.81 
 
 
 17 
 
 67.05 
 
 37.25 
 
 41 
 
 66.15 
 
 36.75 
 
 
 IS 
 
 67.10 
 
 37.28 
 
 42 
 
 66.02 
 
 36.68 
 
 
 19 
 
 67.14 
 
 37.30 
 
 43 
 
 65.89 
 
 36.60 
 
 
 20 
 
 67.17 
 
 37.31 44 
 
 65.75 
 
 36.52 
 
 
 21 
 
 67.18 
 
 37.32 , 45 
 
 65.60 
 
 36.44 
 
 
 22 
 
 67.19 
 
 37.32 
 
 46 
 
 65.45 
 
 36.36 
 
 
 23 
 
 67.20 
 
 37.33 
 
 47 
 
 65.30 
 
 36.28 
 
 Tropic of Cancer 
 
 23.30 
 
 97.20 
 
 37.33 
 
 48 
 
 65.13 
 
 36.18 
 
70 
 
 BABIANT HEAT. 
 
 Table E, SHOwiNa the Temperatoke produced by Solar Kadi- 
 
 ATioN AT Noon for each Degree of Latitude when the 
 
 Eakth is in Aphelion. Northern Hemisphere. 
 
 
 Lat. 
 
 Solar intensity at 
 noon. 
 
 Lat. 
 
 Solar intensity at 
 noon. 
 
 Deg. 
 
 'Fall. 
 
 ° Cani. 
 
 Deg. 
 
 °Fah. 
 
 " Cent. 
 
 49 
 
 64.95 
 
 86.08 
 
 69 
 
 59.76 
 
 33.20 
 
 
 50 
 
 64.77 
 
 35.98 
 
 70 
 
 59.42 
 
 33.01 
 
 
 51 
 
 64.58 
 
 35.88 
 
 71 
 
 59.06 
 
 32.81 
 
 Greenwicli 
 
 51.28 
 
 52 
 
 64.48 
 64.38 
 
 35.82 
 35.77 
 
 72 
 73 
 
 58.69 
 58.31 
 
 32.61 
 32.39 
 
 
 
 53 
 
 64.17 
 
 35.65 
 
 74 
 
 57.92 
 
 32.18 
 
 
 54 
 
 63.96 
 
 35.53 
 
 75 
 
 57.52 
 
 31.95 
 
 
 55 
 
 63.74 
 
 35.41 
 
 76 
 
 57.10 
 
 31.72 
 
 
 56 
 
 63.51 
 
 35.28 
 
 77 
 
 56.67 
 
 31.48 
 
 
 57 
 
 63.28 
 
 35.15 
 
 78 
 
 56.24 
 
 31.24 
 
 
 58 
 
 63.04 
 
 35.02 
 
 79 
 
 55.79 
 
 30.99 
 
 
 59 
 
 62.79 
 
 34.88 
 
 80 
 
 55.32 
 
 30.73 
 
 
 60 
 
 62.53 
 
 34.74 
 
 81 
 
 54.84 
 
 30.46 
 
 
 61 
 
 62.25 
 
 34.58 
 
 82 
 
 54.35 
 
 30.19 
 
 
 62 
 
 61.96 
 
 34.42 
 
 83 
 
 53.84 
 
 29.91 
 
 
 63 
 
 61.65 
 
 34.25 
 
 84 
 
 53.32 
 
 29.62 
 
 
 64 
 
 61.34 
 
 34.08 
 
 85 
 
 52.78 
 
 29.32 
 
 
 65 
 
 61.03 
 
 33.91 
 
 86 
 
 52.23 
 
 29.02 
 
 
 66 
 
 60.72 
 
 33.73 
 
 87 
 
 51.68 
 
 28.71 
 
 Arctic Circle 
 
 66.30 
 
 60.57 
 
 33.65 
 
 88 
 
 51.11 
 
 28.39 
 
 
 07 
 
 60.41 
 
 33.56 
 
 89 
 
 50.52 
 
 28.07 
 
 
 68 
 
 60.09 
 
 33.38 
 
 90 
 
 49.01 
 
 27.73 
 
CUAP. III. INTENSITY OF SOLAR RADIATION. 71 
 
 stice. Again, the North Pole being 6C cleg. 30 rain, from the 
 ecliptic at tlie same time, we learn, by referring to Table C, 
 that the inaxiinum solar intensity at the said pole will be 
 y0°.95 F., that being the temperature produced by solar radi- 
 ation when the zenith distance is GG dejj. 30 rain. 
 
 The Tables D and E contain the raaxiniura solar intensities 
 for all latitudes frora the Equator to the North Pole, deter- 
 mined agreeably to the foregoing explanation. The difference 
 of atmospheric density towards the pole calls for a trifling 
 correction of the temperature entered in the table; but the 
 data not being sufficiently well known, I have deemed it best 
 to present the theoretical temperatures without correction. 
 
 As far as ascertained by means of the actinometer, there 
 is an appreciable difference in the sun's energy for corre- 
 sponding zenith distances early in the morning and late in 
 the afternoon, which cannot be traced to any adequate phy- 
 sical cause. I have accordingly attem2:)ted to explain the dis- 
 crepancy on the ground that the oi-bital motion of the earth 
 occasions a very considerable advance towards, and retreat 
 from, the solar wave early a.m. and late \\y\. The subject 
 A\ill be readily understood by reference to Fig. 3, which 
 represents a section of the earth through the plane of the 
 ecliptic, the line d e indicating the orbit, and the straight 
 arrow the earth's course, while the curved arrow shows the 
 direction of rotation ; a h c, etc., represent the sun's rays, 
 the orbital velocity during a definite period being represented 
 by/i/ and h I. Let us assume that the latitude of the point 
 /', on the earth's surface, is such that the prolongation of the 
 
72 RADIANT HEAT. chap. hi. 
 
 ray ah to g makes h g tliree times longer than / g. It will 
 now be evident ttat tie ray a li, AvLicli lias been arrested 
 at h, must, while the earth advances from / to g, continue its 
 course at a rate three times greater than the earth's orbital 
 velocity, in order to reach g simultaneously with/". Assuming 
 the mean distance of the earth fi'om the sun to be 91,430,000 
 miles, the orbital velocity will be 96,120 ft. per second; hence 
 the ray a 7i, to keep up with the retreating western quarter 
 
 of the globe, must move at the rate of 288,360 ft. per second. 
 The advancing eastern quarter obviously imparts a retrograde 
 movement to the solar wave, consequently the ray m I will, 
 on grounds already set forth, be pushed towards the sun at 
 the rate of 288,360 ft. per second. We have thus established 
 a difference of advancing and retreating velocity exceeding 
 600,000 ft. per second for the lower altitudes, which unques- 
 tionably interferes with the regularity of the solar wave, and 
 thereby tends to disturb the uniformity of the intensity of the 
 sun's radiant heat toAvards evenino-. Meteorolo2:ists will account 
 
CHAP. III. INTEXSITY OF SOLAIi RADIATION. 73 
 
 for the observed diiniiiution by pointing to the fact that during 
 sunshine — without wliich the actinometer cannot be used — 
 the atmosphere, in most localities, gradually becomes charged 
 with vapor as the day advances ; and that dust and other 
 light dry particles are carried up into the atmosphere by the 
 ascending heated current of air, thus obstructing the sun's 
 rays. These plausible reasons lose their force if we consider 
 that, during the season most favorable for actinometric obser- 
 vations, the vapors are held fast within icy boundaries, and 
 that the dust is buried under the snow. 
 
 The extraordinary velocity of light — nearly 1,700 times 
 greater than the velocity shown by the foregoing demonstra- 
 tion — will be urged as a reason why the disturbance of the 
 solar wave could not be practically appreciable. This objec- 
 tion cannot be deemed valid unless it can be shown that the 
 dynamic energy imparted by solar heat is not partially the 
 result of arresting the motion of the rays. The following 
 facts connected with the subject demand serious considera- 
 tion. Owing to the orl)ital motion of the earth, the lens of 
 a solar calorimeter, while exposed to the radiant heat, sweeps 
 across the path of the sun's rays at the rate of 96,120 ft. 
 per second ; hence the fluid contained within the insti-ument 
 receives the energy of a countless number of rays following 
 each other in an inconceivably rapid succession. 
 
 Pouillet, having ascei-tained the number of thermal units 
 imparted to the water in his pyrlieliometer of ."^.93 ins. dia- 
 meter, imagined that he had measured only the energy of 
 the rays contained in a pencil of 11.9 square inches section; 
 
74 BABIANT HEAT. chap. hi. 
 
 whereas, in reality, lie had, at the end of his experiment of 
 five minutes' duration, subjected his instrument to the action 
 of the entire number of rays contained in a passing pencil or 
 sunbeam, the section of wliich we ascertain by multiplying the 
 orbital advance of the earth during five minutes, 28,836,000 ft., 
 by the diameter of the pyrheliometer, 0.305 ft. 
 
CHAPTER IV. 
 
 PERIODIC VARIATION OF THE INTENSITY OF SOLAR 
 RADIATION. 
 
 The preceding cliapter has made the reader familiar with 
 the construction of the aetinometer, and with the leading 
 results of the investigations conducted by means of that in- 
 strument, relating to the variations of temperature consequent 
 on the sun's vaiying zenith distance. Let us now con.sider 
 the variation of solar intensity consequent on the varying 
 distance between the sun and the earth from day to day. 
 Meteorologists, in recording the temperature produced by 
 solar radiation, have hithei-to taken no notice of the position 
 of the earth in its orliit at the time of making their obser- 
 vations. At the commencement of my investigation of the 
 mechanical properties of solar heat, I committed the same 
 oversight ; but finding that the result of my observ'-ations fre- 
 quently presented discrepancies that coidd not be accounted 
 for on the gi'ound of diiferent zenith distance and presence 
 of vapor in the atmosphere, I was led to examine systema- 
 ticall\- and very oaiefuljy the effect on solar intensity pro- 
 
t6 BABIANT HEAT. chap. iv. 
 
 ducecl by tlie variation of the sun's distance from the earth. 
 Sir John Herschel, in " Outlines of Astronomy," says, regard- 
 ing the eifect of the sun's varying distance : " The angular 
 velocity of the earth in its orbit is not uniform, but varies 
 in the inverse ratio of the square of the sun's distance — that 
 is, in the same precise ratio as his heating power. The mo- 
 mentary supply of heat, then, received by the earth in eveiy 
 point of its orbit varies exactly as the momentary increase 
 of its longitude ; from which it obviously follows that equal 
 amounts of heat are received from the sun in passing over 
 equal angles round it, in whatever part of the ellijjse those 
 angles may be situated." As regards the temperature deve- 
 loped by the sun at different periods, the author of the " Out- 
 lines of Astronomy" calculates that there is a difference of 
 one-fifteenth ; but, in judging of the effect of this difference on 
 the tempeiature produced by solar radiation, he says : " We 
 have to consider as our unit, not the number of degrees above 
 a purely arbitrary zero point (such as the freezing point of 
 water or the zero of Fahrenheit's scale) on which a thermometer 
 stands on a hot summer day, as compared with a cold winter 
 one, but the thermometric interval between the temj)eratures 
 it indicates in the two cases, and that it would indicate did 
 the sun not exist, which there is good reason to believe would 
 be at least as low as 239° lelow zero of Fahrenheit. And as a 
 temperature of 100° F. ahovezevo is no uncommon one in a fair 
 shade exposure under a sun nearly vertical, we have to take 
 one-fifteenth of the sum of these intervals (339°), or 23° F., 
 as the least variation of tenq)erature under such (.■ircumstances 
 
CUAP. IV. rEliWDW VARIATION OF HOLAE EADIATION. 77 
 
 wbicli can reasonably be attributed to the actual variation of 
 tlie sun's distance." Adopting the stated calculation, with- 
 out reflecting on the erroneous grounds ujion which it rests, 
 I introduced corrections of the obsei"ved temperatures accord- 
 ing to the supposed augmentation of radiant intensity, viz., 
 23° F. when the eaiih is in perihelion. The application of 
 these corrections proved that the discrepancies in my records, 
 before adverted to, amounted to only one-fifth of that which, 
 agreeably to Sir John Ilerschel's theoiy, ought to have ap- 
 peared. Fully convinced, however, that the difference of solar 
 intensity resulting from the variation of distance between the 
 sun and the earth was the ti-ue cause of the irregularity and 
 breaks in the curve which I had constructed according to 
 the observed temperatures, I availed myself of eveiy favor- 
 able opportunity to ascertain the maxinuun intensity produced 
 during the summer and winter solsticcn. It will be Avell to 
 state, for the information of those who have not paid special 
 attention to the subject, that mean remdts of observation are 
 inadmissible in records intended to establish solar energy. 
 The superior intensity ascertained at a single observation 
 will set aside the result of previous observations continued 
 for many years. It will be well also to correct the prevail- 
 ing erroneous supposition that solar intensity cannot be ascer- 
 tained during the summer months, owing to vapor in the 
 atmosphere. There are short intervals at all seasons when 
 polar winds prevail, during which the sun is perfectly clear. 
 Those who have paid close attention to this matter will say 
 that thev have seen as brij^ht a solar disc in August as in 
 
78 RADIANT BEAT. chap. it. 
 
 January. This fact has heen repeatedly verified by the indi- 
 cations of my sohir calorimeter, an unerring test, since it 
 records the number of units of heat developed by the sun 
 in a given time on a given area. Obviously the smallest 
 increase of the absoi-ptive power of the atmosphere will be 
 detected by this method. It will be perceived from the fore- 
 going remarks that the records of solar intensity connected 
 with the solstices, kept during a series of years, possess no 
 matei'ial interest regarding the question at issue — namely, the 
 true maximum difference of intensity of solar radiation when 
 the earth is in aphelion and in perihelion. The reader, there- 
 fore, instead of being called upon to examine an extended 
 record of observations, will simply have his attention directed 
 to the fact that January 7, 1871, the earth being then, of 
 course, very near perihelion, the temperature produced by 
 solar radiation, indicated by the actinometer, reached 57°.25 
 F. at noon; the zenith distance being 63 deg. 15 min. Re- 
 fen-ing to the table of temperatures for given zenith distances 
 (Chap. III.), it will be seen that the temperature produced by 
 solar radiation when the earth is in apJielion is 52°.84 F. at 
 a zenith distance of 63 deg. 15 min. ; consequently, an aug- 
 mentation of solar intensity of 57.25 — 52.84 = 4°.41 F. takes 
 place when the earth is in perihelion. The result of my acti- 
 nometric observations, coiitinued through a series of years, 
 recorded in Chap. III., shows that when the earth is in aphe- 
 lion the maximum solar intensity on the ecliptic is 67°.20 F. 
 The law of inverse squares being true for spherical radiators 
 and for radiating circular discs subtending small angles, the 
 
cn.\i>. IV. PERIODIC VAIilATIoy OF SOLAR RADIATION. 79 
 
 stated iutensity of solar radiation when the earth is in a2)lie- 
 lion enables us to determine with absolute cei-tainty wliat 
 temiDeralure will lie produced when the earth is in perihe- 
 lion. The ratio of the earth's distance from the sun at the 
 two opposite points of the orbit, in aphelion and in perihe- 
 lion, being 218.1 : 210.0, while the temperature produced by 
 solar radiation during the suininei- sol.-<tiee, as stated, is G7°.2 
 F., the radiant iutensity during the winter solstice will be 
 
 218.1' X 67.2 
 
 = 7l''.86 F. The temperature produced by solar 
 
 210.9 ^ ^ -^ 
 
 radiation at the surface of the earth will thus, agreeably to 
 the laws which govern the transmission of radiant heat, be 
 71.86 - 67.20 = 4°.66 F. higher when the earth is nearest 
 the sun than when furthest from it. The aetinometric obser- 
 vation, Jan. 7, 1871, having established a ditVereiitial teiiipe- 
 ratiu-e of 4°.41 F., it will thus be seen that a discrepancy 
 exists amounting to 0°.25 F. That is, the observed solar iu- 
 tensity, 57°.25, Jan. 7, was 0°.25 less than the calculated in- 
 tensity. A closer agreement between computed and observed 
 increment of solar intensity consequent on the eccentricity of 
 the earth's orbit could not reasonably be exjiected. Besides, 
 the record shows that, although the sun was exce]itionally 
 clear on the day mentioned, there was a perceptible mist 
 round the solar disc, indicating that the full radiant powei- 
 was not transmitted to the actinometer. In constructing the 
 tables appended to this chapter I have, therefoi-e, based my 
 calculations of diurnal variation of solar intensity on the dif- 
 ferential temperature, -4". 06 F., determined by computations 
 
80 BADIANT EEAT. chap. iv. 
 
 founded ou tlie distance of the earth from the suu at the 
 opposite points of the orbit. Referring to the tables, it ^vill 
 be seen that the maximum temperature produced by solar 
 radiation has been entered for each day throughout the year ; 
 also the increment of solar intensity for each day, consequent 
 on the varying distance of the sun. The principal object 
 of the tables being that of enabling the meteorologist to 
 ascertain to what extent the result of his observations of 
 solar radiation is influenced by the distance of the sun, it 
 will be evident that some zero having a fixed relation to 
 the position of the earth in its orbit should be adopted, in 
 order to render comparisons possible. Accordingly, the ap- 
 pended tables have reference to the maximum temperature 
 produced by solar radiation when the earth is in aphelion. 
 The utility of adopting a fixed zero will be seen by the fol- 
 lowing explanation : Suppose that we find by observation, 
 during bright sunshine, January 20, that, under a zenith dis- 
 tance of 68 deg., the actinometer indicates 54°.5 F. Suppose, 
 also, that our records of solar intensity during summer show 
 that, June 15, the actinometer indicated 49°.9 F. at equal zenith 
 distance — viz., 68 deg. Leaving the influence of solar distance 
 out of sight, the inference would be that the diminished inten- 
 sity of 54.5 — 49.9 = 4°.6 F., observed June 15, was owing to 
 the presence of vapor in the atmosphere. Secchi and others, 
 who suppose that the absorptive power of the atmosphere 
 called forth by vapor prevents the full development of solar 
 radiation at all times during summer, would not hesitate to 
 ascribe the observed diminution of radiant intensity to that 
 
CHAP. IV. PEKIODIC VARIATION OF HOLAU RADIATION. «1 
 
 cause. But a gluiu-e at our table at once ilisclascs the true 
 cause of the observed difference of solar intensity under equal 
 zenith distance in January and in June. Consulting- the table 
 for Jaiuiaiy, and running the eye down the column headed 
 " Increment," the temperature 4°.6 will be found opposite the 
 date 20tli in the tirst column. Accordingly, the feebleness 
 of snlar j-adiation observed in the middle of June, instead of 
 being caused by atmospheric absorption, is solely due to the 
 increased distance of the earth from the sun. It should be 
 observed that the assumed temperature of 49°.9 F., at a zenith 
 distance of GS deg. in the middle of June, is not imagiuarj', 
 having frequently been observed during my investigations of 
 solar energy. Again, temperatures exceeding 54° F. have been 
 observed during mid-winter at a zenith distance of 68 deg. 
 
 It will be proper to remind meteorologists accustomed to 
 observe the intensity of solar radiation, hence familiar with 
 the extraordinaiy discrepancy of observations made with ordi- 
 nary " solar radiation thermometers," that the invariably con- 
 sistent indications, and the freedom from conHicting results, 
 in my actinometric observations of solar energy, are chiefly 
 due to the fact that the bulb of the recording thermometer 
 is enclosed within an exhausted vessel, maintained at a con- 
 stant temperature of 60° F. during obser\'a(ions. Hence, ^\•he- 
 ther the investigation be conducted in calms or high winds, 
 or whether the thermometer marks 100 in the shade or the 
 temperature of the air be below zero, no material error is 
 possible. This fact merits special consideration on the part 
 of those who have questioned the possibility of determining 
 
82 BADIANT MEAT. chap. iv. 
 
 practically to \\'liat degree tlie temperature produced by solar 
 radiation is increased when the earth is in perihelion. And 
 those who have adopted Herschel's conclusion, that " 23° Fah- 
 renheit is the least variation of temperature which can reason- 
 ably be attributed to the actual variation of the sun's distance," 
 will do well to contrast the reasoning of the great astronomer 
 with the reasoning which assigns 5° Fahrenheit as the utmost 
 increment of solar intensity in the southern hemisphere, con- 
 sequent on the proximity of the luminary during our mid- 
 winter. 
 
 Regarding the construction of the appended tables of solar 
 intensity, the following explanation and recapitulation will 
 suffice : Having determined the maximum solar intensity when 
 the earth is in aphelion, in accordance with the data furnished 
 in Chap. III., the maximum intensity in perihelion was deter- 
 mined by computation in the manner already pointed out. 
 The result, as we have seen, has been fully corroborated by 
 actinometric observation. The maximum increment of tempe- 
 rature produced by solar radiation when the earth is in peri- 
 helion having been fixed at 4°.66 F., the increment of tem- 
 perature for each day throughout the year was determined 
 by inverting the ratio of the square of the sun's distance from 
 day to day. 
 
 We have already pointed out that investigations of solar 
 intensity which do not take cognizance of the influence of the 
 sun's distance are of no value. Obviously, the difference of 
 temperature produced by solar radiation furnishes an infal- 
 lible index of atmospheric absorption, provided we make due 
 
CHAP. IV. PERIODIC VAIUATIOX OF tiOLAIi RADIATION. 83 
 
 allowance for the sun's zenith distance and the position of the 
 earth in its orbit at the time of making our observation. But, 
 unless suoli allowance be made, the inference we draw from the 
 indicated temperatures will prove wholly erroneous. 
 
 Respecting the mode of applying the tables, it will be 
 evident, on reflection, that, if we desire to ascertain whether 
 vajiors are present in the atmosphere, the temperature con- 
 tained in the column headed " Increment," for the day on 
 which the observation is made, should be deducted from 
 the temperature indicated by the actinometer. Suppose, for 
 instance, that the indicated solar intensity, Aug. 26, is 58° F. 
 at a zenith distance of 40 deg. Deducting from 58' the incre- 
 ment of temperature entered in the table for Aug. 26 — viz., 
 0''.96 — consequent on the diminished distance fi-om the sun, 
 we obtain 58 — 0.96 = 57^.04. Now, if we consult the table 
 of temperatures for given zenith distance (see page 62, Chap, 
 in.), it will be found that the maximum solar temperature 
 for a zenith distance of 40 deg. should be 61°.67. Our obser- 
 vation of Aug. 26, therefore, shows that the vapors contained 
 in the atmosphere have absorbed 61.67 — 57.04 = 4°.63 of 
 the sun's radiant heat. The advantage of this positive mode 
 of ascertaining the absorptive power — i.e., the presence of vajtor 
 in the atmosphere — meteorologists cannot fail to appreciate. 
 
 The close agreement between the computed and observed 
 increment of the temperature produced by solar radiation when 
 the earth is in perihelion, before adverted to, calls for special 
 notice in this place, since it proves the correctness of the acti- 
 nometric observations recorded in Chap. III., on which our 
 
8-i BABIANT HEAT. chap. iv. 
 
 determination of the varying intensity of the sun's radiant 
 heat is based. It will be evident that, if our determination 
 of maximum solar intensity when the earth is in aphelion 
 (based solely on our observations) were incorrect, the infal- 
 lible test of applying the law of inverse squares would at 
 once expose the error. Now, this test has been applied ; we 
 have inverted the squares of the relative aphelion and peri- 
 helion distance of the earth, and we find that in the latter 
 position an increment of 4°. 6 6 F. results from the proximity 
 of the sun. Our ohservations show an increment of 4°.41 dif- 
 ference = 0°,25 Fahrenheit. 
 
cuAP. IV. rEnioinc vaiuatiox of solai; hadiatios. 
 
 T. 
 
 IBLE SHOWING THE MAXIMUM INTENSITY 
 
 OF Solar Radiation 
 
 
 
 
 ON 
 
 THE Ecliptic. 
 
 
 
 
 1 
 
 JAyUARY. 
 
 FEBRUARY. 
 
 Maximum. 
 
 Increment. 
 
 Maximum. 
 
 Increment. 
 
 • Fuh. 
 
 ° Ctnt. 
 
 'Fah. 
 
 • Cmt. 
 
 i • Fah. 
 
 'Ctnt. 
 
 •Fah. 
 
 'Cmt 
 
 1 
 
 71.94 
 
 39.97 
 
 4.66 
 
 2.59 
 
 71.59 
 
 39.77 
 
 4.31 
 
 2.39 
 
 2 
 
 71.94 
 
 39.97 
 
 4.60 
 
 2.59 
 
 71.57 
 
 39.76 
 
 4.29 
 
 2.38 
 
 3 
 
 71.94 
 
 39.97 
 
 4.66 
 
 2.59 
 
 71.54 
 
 39.74 
 
 4.26 
 
 2.36 
 
 4 
 
 71.93 
 
 39.97 
 
 4.65 
 
 2.59 
 
 71.52 
 
 39.73 
 
 4.24 
 
 2.35 
 
 5 
 
 71.93 
 
 39.97 
 
 4.65 
 
 2.59 
 
 71.50 
 
 39.72 
 
 4.22 
 
 2.34 
 
 6 
 
 71.92 
 
 39.96 
 
 4.64 
 
 2.58 
 
 71.47 
 
 39.71 
 
 4.19 
 
 2.33 
 
 7 
 
 71.92 
 
 39.96 
 
 4.64 
 
 2.58 
 
 71.45 
 
 39.69 
 
 4.17 
 
 2.31 
 
 8 
 
 71.91 
 
 39.95 
 
 4.63 
 
 2.57 
 
 71.42 
 
 39.68 
 
 4.14 
 
 2.30 
 
 9 
 
 71.91 
 
 39.95 
 
 4.63 
 
 2.57 
 
 71.40 
 
 39.67 
 
 4.12 
 
 2.29 
 
 10 
 
 71.90 
 
 39.94 
 
 4.62 
 
 2.56 
 
 71.38 
 
 39.65 
 
 4.10 
 
 2.27 
 
 11 
 
 71.90 
 
 39.94 
 
 4.62 
 
 2.56 
 
 71.35 
 
 39.64 
 
 4.07 
 
 2.26 
 
 12 
 
 71.89 
 
 39.93 
 
 4.61 
 
 2.55 
 
 71.32 
 
 39.62 
 
 4.04 
 
 2.24 
 
 13 
 
 71.88 
 
 39.93 
 
 4.60 
 
 2.55 
 
 71.29 
 
 39.60 
 
 4.01 
 
 2.22 
 
 14 
 
 71.87 
 
 39.92 
 
 4.59 
 
 2.54 
 
 71.27 
 
 39.59 
 
 3.99 
 
 2.21 
 
 15 
 
 71.86 
 
 39.92 
 
 4.58 
 
 2.54 
 
 71.24 
 
 39.58 
 
 3.96 
 
 2.20 
 
 16 
 
 71.85 
 
 39.91 
 
 4.57 
 
 2.53 
 
 71.20 
 
 39.56 
 
 3.92 
 
 2.18 
 
 17 
 
 71.84 
 
 39.91 
 
 4.56 
 
 2.53 
 
 71.17 
 
 39.54 
 
 3.89 
 
 2.16 
 
 IS 
 
 71.82 
 
 39.90 
 
 4.54 
 
 2.52 
 
 71.14 
 
 39.52 
 
 3.86 
 
 2.14 
 
 19 
 
 71.81 
 
 39.90 
 
 4.53 
 
 2.52 
 
 71.11 
 
 39.50 
 
 3.83 
 
 2.12 
 
 20 
 
 71.80 
 
 39.89 
 
 4.52 
 
 2.51 
 
 71.08 
 
 39.49 
 
 3.80 
 
 2.11 
 
 21 
 
 71.78 
 
 39.88 
 
 4.50 
 
 2.50 
 
 71.05 
 
 39.47 
 
 3.77 
 
 2.09 
 
 22 
 
 71.77 
 
 39.87 
 
 4.49 
 
 2.49 
 
 71.02 
 
 39.45 
 
 3.74 
 
 2.07 
 
 23 
 
 71.76 
 
 39.87 
 
 4.48 
 
 2.48 
 
 70.98 
 
 39.43 
 
 3.70 
 
 2.05 
 
 24 
 
 71.74 
 
 39.86 
 
 4.46 
 
 2.47 
 
 70.95 
 
 39.42 
 
 3.67 
 
 2.04 
 
 25 
 
 71.72 
 
 39.84 
 
 4.44 
 
 2.46 
 
 70.02 
 
 39.40 
 
 3.64 
 
 2.02 
 
 2G 
 
 71.70 
 
 39.83 
 
 4.42 
 
 2.45 
 
 70.88 
 
 39.38 
 
 3.60 
 
 2.00 
 
 27 
 
 71.68 
 
 39.82 
 
 4.40 
 
 2.44 
 
 70.85 
 
 39.36 
 
 3.57 
 
 1.98 
 
 28 
 
 71.67 
 
 39.81 
 
 4.39 
 
 2.43 
 
 70.82 
 
 39.34 
 
 3. .54 
 
 1.90 
 
 20 
 
 71.65 
 
 39.80 
 
 4.37 
 
 2.42 
 
 
 
 
 
 3(J 
 
 71.63 
 
 39.79 
 
 4.35 
 
 2.41 
 
 
 
 
 
 31 
 
 71. Gl 
 
 39.78 
 
 4.33 
 
 •J. -10 
 
 
 
 
 
8G 
 
 EADIANT UEAT. 
 
 Table siiowinr the Maximum: Intensity 
 
 OF Solar Radiation 
 
 
 
 
 ON 
 
 THE Ecliptic. 
 
 
 
 
 
 MARCH. 
 
 APRIL. 
 
 Maximum. 
 
 Increment. 
 
 Maximum. 
 
 Increment. 
 
 °Fah. 
 
 ' Cent. 
 
 ° Fah. 
 
 ' Cent. 
 
 'Fah. 
 
 ' Cent. 
 
 ° Fah. 
 
 ' Cent. 
 
 1 
 
 70.79 
 
 39.33 
 
 3.51 
 
 1.95 
 
 69.60 
 
 38.67 
 
 2.32 
 
 1.29 
 
 2 
 
 70.75 
 
 39.31 
 
 3.47 
 
 1.93 
 
 69.56 
 
 38.64 
 
 2.38 
 
 1.26 
 
 3 
 
 70.72 
 
 39.29 
 
 3.44 
 
 1.91 
 
 69.52 
 
 38.62 
 
 2.24 
 
 1.24 
 
 4 
 
 70.08 
 
 39.27 
 
 3.40 
 
 1.89 
 
 69.48 
 
 38.60 
 
 2.20 
 
 1.22 
 
 5 
 
 70.65 
 
 39.25 
 
 3.37 
 
 1.87 
 
 69.44 
 
 38.58 
 
 2.16 
 
 1.20 
 
 6 
 
 70.61 
 
 39.23 
 
 3.33 
 
 1.85 
 
 69.40 
 
 38.56 
 
 2.12 
 
 1.18 
 
 7 
 
 70.58 
 
 39.21 
 
 3.30 
 
 1.83 
 
 69.36 
 
 38.53 
 
 2.08 
 
 1.15 
 
 8 
 
 70.54 
 
 39.19 
 
 3.26 
 
 1.81 
 
 69.32 
 
 38.51 
 
 2.04 
 
 1.13 
 
 9 
 
 70.50 
 
 39.17 
 
 3.22 
 
 1.79 
 
 69.28 
 
 38.49 
 
 2.00 
 
 1.11 
 
 10 
 
 70.46 
 
 39.14 
 
 3.18 
 
 1.76 
 
 69.24 
 
 38.47 
 
 1.96 
 
 1.09 
 
 11 
 
 70.42 
 
 39.12 
 
 3.14 
 
 1.74 
 
 69.20 
 
 38.44 
 
 1.92 
 
 1.06 
 
 12 
 
 70.38 
 
 39.10 
 
 3.10 
 
 1.72 
 
 69.16 
 
 38.42 
 
 1.88 
 
 1.04 
 
 13 
 
 70.35 
 
 39.08 
 
 3.07 
 
 1.70 
 
 69.13 
 
 38.40 
 
 1.85 
 
 1.02 
 
 14 
 
 70.31 
 
 39.06 
 
 3.03 
 
 1.68 
 
 69.09 
 
 38.38 
 
 1.81 
 
 1.00 
 
 15 
 
 70.27 
 
 39.04 
 
 2.99 
 
 1.66 
 
 69.05 
 
 38.36 
 
 1.77 
 
 0.98 
 
 16 
 
 70.23 
 
 39.02 
 
 2.95 
 
 1.64 
 
 69.01 
 
 38.34 
 
 1.73 
 
 0.96 
 
 17 
 
 70.19 
 
 38.99 
 
 2.91 
 
 1.61 
 
 68.97 
 
 38.32 
 
 1.69 
 
 0.94 
 
 18 
 
 70.15 
 
 38.97 
 
 2.87 
 
 1.59 
 
 68.93 
 
 38.29 
 
 1.65 
 
 0.91 
 
 19 
 
 70.12 
 
 38.95 
 
 2.84 
 
 1.57 
 
 68.89 
 
 38.27 
 
 1.61 
 
 0.89 
 
 20 
 
 70.08 
 
 38.93 
 
 2.80 
 
 1.55 
 
 68.85 
 
 38.25 
 
 1.57 
 
 0.87 
 
 21 
 
 70.04 
 
 38.91 
 
 2.76 
 
 1.53 
 
 68.81 
 
 38.23 
 
 1.53 
 
 0.85 
 
 22 
 
 70.00 
 
 38.89 
 
 2.72 
 
 1.51 
 
 68.78 
 
 38.21 
 
 1.50 
 
 0.83 
 
 23 
 
 69.96 
 
 38.87 
 
 2.68 
 
 1.49 
 
 68.74 
 
 38.19 
 
 1.46 
 
 0.81 
 
 24 
 
 69.92 
 
 38.84 
 
 2.64 
 
 1.46 
 
 68.70 
 
 38.17 
 
 1.42 
 
 0.79 
 
 25 
 
 69.88 
 
 38.82 
 
 2.60 
 
 1.44 
 
 68.66 
 
 38.14 
 
 1.38 
 
 0.76 
 
 26 
 
 69.84 
 
 38.80 
 
 2.56 
 
 1.42 
 
 68.63 
 
 38.12 
 
 1.35 
 
 0.74 
 
 27 
 
 69.80 
 
 38.78 
 
 2.52 
 
 1.40 
 
 68.59 
 
 38.10 
 
 1.31 
 
 0.72 
 
 28 
 
 69.76 
 
 38.75 
 
 2.48 
 
 1.37 
 
 68.55 
 
 38.08 
 
 1.27 
 
 0.70 
 
 29 
 
 69.72 
 
 38.73 
 
 2.44 
 
 1.35 
 
 68.52 
 
 38.06 
 
 1.24 
 
 0.68 
 
 80 
 
 69.68 
 
 38.71 
 
 2.40 
 
 1.33 
 
 68.48 
 
 38.04 
 
 1.20 
 
 0.66 
 
 31 
 
 69.64 
 
 38.60 
 
 2.36 
 
 1.31 
 
 
 
 
 
cuAP. IV. rEiiioDia VAitiATJoy of solak riAniAiwy. 
 
 87 
 
 Table siiowin-g tiik Maximum Isteksity 
 
 OF SOLAE 
 
 Kadiatiox 1 
 
 
 
 
 ox 
 
 rnrc Ecliptic. 
 
 
 
 
 "a 
 
 MAY. 
 
 jm'E. 
 
 Maximum. 
 
 Increment. 
 
 Maximum. 
 
 Increment. 
 
 •Fall. 
 
 • Ctnt. 
 
 'Fah. 
 
 •Ctnt. 
 
 •Fah. 
 
 • Cent. 
 
 •Fak. 
 
 •Cent. 
 
 1 
 
 68.45 
 
 38.03 
 
 1.17 
 
 0.65 
 
 67.58 
 
 37.54 
 
 0.30 
 
 0.16 
 
 2 
 
 68.41 
 
 38.01 
 
 1.13 
 
 0.63 
 
 67.56 
 
 37.53 
 
 0.28 
 
 0.15 
 
 3 
 
 68.38 
 
 37.99 
 
 1.10 
 
 0.61 
 
 67.54 
 
 37.52 
 
 0.26 
 
 0.14 
 
 4 
 
 68.35 
 
 37.97 
 
 1.07 
 
 0.59 
 
 67.52 
 
 37.51 
 
 0.24 
 
 0.13 
 
 5 
 
 68.32 
 
 37.95 
 
 1.04 
 
 0.57 
 
 67.51 
 
 37.50 
 
 0.23 
 
 0.12 
 
 6 
 
 68.28 
 
 37.93 
 
 1.00 
 
 0.55 
 
 67.49 
 
 37.49 
 
 0.21 
 
 0.11 
 
 7 
 
 68.25 
 
 37.92 
 
 0.97 
 
 0.54 
 
 67.47 
 
 37.48 
 
 0.19 
 
 0.10 
 
 8 
 
 68.22 
 
 37.90 
 
 0.94 
 
 0.52 
 
 67.46 
 
 37. -18 
 
 0.18 
 
 0.10 
 
 9 
 
 68.19 
 
 37.88 
 
 0.91 
 
 0.50 
 
 67.44 
 
 37.47 
 
 0.10 
 
 0.09 
 
 10 
 
 68.16 
 
 37.87 
 
 0.88 
 
 0.49 
 
 67.43 
 
 37.40 
 
 0.15 
 
 0.08 
 
 11 
 
 68.13 
 
 37.85 
 
 0.85 
 
 0.47 
 
 67.42 
 
 37.45 
 
 0.14 
 
 0.07 
 
 12 
 
 68.10 
 
 37.83 
 
 0.82 
 
 0.45 
 
 67.40 
 
 37.44 
 
 0.12 
 
 0.06 
 
 13 
 
 68.06 
 
 37.81 
 
 0.78 
 
 0.43 
 
 67.39 
 
 37.44 
 
 0.11 
 
 0.06 
 
 14 
 
 68.03 
 
 37.79 
 
 0.75 
 
 0.41 
 
 67.38 
 
 37.43 
 
 0.10 
 
 0.05 
 
 15 
 
 68.00 
 
 37.78 
 
 0.72 
 
 0.40 
 
 67.37 
 
 37.43 
 
 0.09 
 
 0.05 
 
 16 
 
 67.97 
 
 37.76 
 
 0.69 
 
 0.38 
 
 67.36 
 
 37.42 
 
 0.08 
 
 0.04 
 
 17 
 
 67.94 
 
 37.74 
 
 0.60 
 
 0.36 
 
 67.35 
 
 37.42 
 
 0.07 
 
 0.04 
 
 18 
 
 67.91 
 
 37.73 
 
 0.63 
 
 0.35 
 
 07.34 
 
 37.41 
 
 0.06 
 
 0.03 
 
 19 
 
 67.89 
 
 37.72 
 
 0.61 
 
 0.34 
 
 67.33 
 
 37.40 
 
 0.05 
 
 0.02 
 
 20 
 
 67.86 
 
 37.70 
 
 0..58 
 
 0.32 
 
 67.32 
 
 37.40 
 
 0.04 
 
 0.02 
 
 21 
 
 67.84 
 
 37.09 
 
 0.56 
 
 0.31 
 
 67.32 
 
 37.40 
 
 0.04 
 
 0.02 
 
 22 
 
 67.81 
 
 37.67 
 
 0.53 
 
 0.29 
 
 67.31 
 
 37.39 
 
 0.03 
 
 0.01 
 
 23 
 
 67.79 
 
 37.66 
 
 0.51 
 
 0.28 
 
 67.30 
 
 37.39 
 
 0.02 
 
 0.01 
 
 24 
 
 67.77 
 
 37.65 
 
 0.49 
 
 0.27 
 
 67.30 
 
 37.39 
 
 0.02 
 
 0.01 
 
 25 
 
 67.75 
 
 37.64 
 
 0.47 
 
 0.26 
 
 67.29 
 
 37.38 
 
 0.01 
 
 0.00 
 
 26 
 
 67.72 
 
 37.62 
 
 0.44 
 
 0.24 
 
 67.29 
 
 37.38 
 
 0.01 
 
 0.00 
 
 27 
 
 67.70 
 
 37.61 
 
 0.42 
 
 0.23 
 
 67.28 
 
 37.38 
 
 0.00 
 
 0.00 
 
 28 
 
 67.67 
 
 37.59 
 
 0.39 
 
 0.21 
 
 67.28 
 
 37.38 
 
 0.00 
 
 0.00 
 
 29 
 
 67.65 
 
 37.58 
 
 0.37 
 
 0.20 
 
 67.28 
 
 37.38 
 
 0.00 
 
 0.00 
 
 30 
 
 67.63 
 
 37.57 
 
 0.35 
 
 0.19 
 
 67.28 
 
 37.38 
 
 0.00 
 
 0.00 
 
 31 
 
 67.60 
 
 37. .55 
 
 0.32 
 
 0.17 
 
 
 
 
 
88 
 
 RADIANT HEAT. 
 
 T. 
 
 BLE SHOWING TUB MaXIMUSI INTENSITY 
 
 Of 80LA1 
 
 . Radiation 
 
 
 
 
 ON 
 
 THE Ecliptic. 
 
 
 
 
 1 
 
 JULY. 
 
 AUGUST. 
 
 Maximum. 
 
 Increment. 
 
 Maximum. 
 
 Increment. 
 
 ' FaJi. 
 
 ° Cent. 
 
 ° Fah. 
 
 ° Cent. 
 
 ° Fah. 
 
 ° Cent. 
 
 • Fah. 
 
 ° Cent. 
 
 1 
 
 67.28 
 
 37.88 
 
 0.00 
 
 0.00 
 
 67.58 
 
 87.54 
 
 0.30 
 
 0.16 
 
 2 
 
 67.28 
 
 37.88 
 
 0.00 
 
 0.00 
 
 67.60 
 
 87.55 
 
 0.32 
 
 0.17 
 
 3 
 
 07.28 
 
 37.38 
 
 0.00 
 
 0.00 
 
 67.63 
 
 87.57 
 
 0.35 
 
 0.19 
 
 ■1 
 
 67.28 
 
 87.38 
 
 0.00 
 
 0.00 
 
 67.65 
 
 37.58 
 
 0.87 
 
 0.20 
 
 5 
 
 67.28 
 
 87.38 
 
 0.00 
 
 0.00 
 
 67.67 
 
 37.59 
 
 0.89 
 
 0.21 
 
 6 
 
 67.28 
 
 87.38 
 
 0.00 
 
 0.00 
 
 67.70 
 
 37.61 
 
 0.42 
 
 0.23 
 
 7 
 
 67.29 
 
 87.38 
 
 0.01 
 
 0.00 
 
 67.72 
 
 37.62 
 
 0.44 
 
 0.24 
 
 8 
 
 67.29 
 
 87.38 
 
 0.01 
 
 0.00 
 
 67.75 
 
 37.04 
 
 0.47 
 
 0.26 
 
 9 
 
 67.30 
 
 37.39 
 
 0.02 
 
 0.01 
 
 67.77 
 
 37.65 
 
 0.49 
 
 0.27 
 
 10 
 
 67.30 
 
 87.39 
 
 0.02 
 
 0.01 
 
 67.79 
 
 37.66 
 
 0.51 
 
 0.28 
 
 11 
 
 67.31 
 
 87.89 
 
 0.03 
 
 0.01 
 
 67.81 
 
 37.67 
 
 0.53 
 
 0.29 
 
 12 
 
 67.32 
 
 37.40 
 
 0.04 
 
 0.02 
 
 67.84 
 
 87.69 
 
 0.56 
 
 0.31 
 
 13 
 
 67.32 
 
 37.40 
 
 0.04 
 
 0.02 
 
 67.86 
 
 37.70 
 
 0.58 
 
 0.32 
 
 14 
 
 67.33 
 
 87.41 
 
 0.05 
 
 0.08 
 
 67.89 
 
 37.72 
 
 0.61 
 
 0.34 
 
 15 
 
 67.34 
 
 37.41 
 
 0.06 
 
 0.08 
 
 67.91 
 
 37.73 
 
 0.63 
 
 0.35 
 
 16 
 
 67.35 
 
 37.42 
 
 0.07 
 
 0.04 
 
 67.94 
 
 37.74 
 
 0.66 
 
 0.36 
 
 17 
 
 67.36 
 
 37.42 
 
 0.08 
 
 0.04 
 
 67.97 
 
 37.76 
 
 0.69 
 
 0.88 
 
 18 
 
 67.37 
 
 37.48 
 
 0.09 
 
 0.05 
 
 68.00 
 
 37.78 
 
 0.72 
 
 0.40 
 
 19 
 
 67.38 
 
 37.43 
 
 0.10 
 
 0.05 
 
 68.03 
 
 87.79 
 
 0.75 
 
 0.41 
 
 20 
 
 67.39 
 
 37.44 
 
 0.11 
 
 0.06 
 
 68.06 
 
 37.81 
 
 0.78 
 
 0.43 
 
 21 
 
 67.40 
 
 37.44 
 
 0.12 
 
 0.06 
 
 68.10 
 
 87.88 
 
 0.82 
 
 0.45 
 
 22 
 
 67.42 
 
 37.45 
 
 0.14 
 
 0.07 
 
 68.13 
 
 37.85 
 
 0.85 
 
 0.47 
 
 23 
 
 67.43 
 
 37.46 
 
 0.15 
 
 0.08 
 
 68.16 
 
 87.87 
 
 0.88 
 
 0.49 
 
 24 
 
 67.44 
 
 37.47 
 
 0.16 
 
 0.09 
 
 68.19 
 
 37.88 
 
 0.91 
 
 0.50 
 
 25 
 
 67.46 
 
 37.48 
 
 0.18 
 
 0.10 
 
 68.22 
 
 87.90 
 
 0.94 
 
 0.52 
 
 26 
 
 67.47 
 
 37.48 
 
 0.19 
 
 0.10 
 
 68.25 
 
 37.92 
 
 0.97 
 
 0.54 
 
 27 
 
 67.49 
 
 87.49 
 
 0.21 
 
 0.11 
 
 68.28 
 
 87.93 
 
 1.00 
 
 0.55 
 
 28 
 
 67.51 
 
 37.50 
 
 0.28 
 
 0.12 
 
 68.32 
 
 37.95 
 
 1.04 
 
 0.57 
 
 29 
 
 67.52 
 
 37.51 
 
 0.24 
 
 0.13 
 
 68.35 
 
 37.97 
 
 1.07 
 
 0.59 
 
 30 
 
 67.54 
 
 87.52 
 
 0.26 
 
 0.14 
 
 68.38 
 
 87.99 
 
 1.10 
 
 0.61 
 
 31 
 
 67.56 
 
 37.58 
 
 0.28 
 
 0.15 
 
 68.41 
 
 38.01 
 
 1.13 
 
 0.63 
 
IV. PERIODIC rAlilArWN OF SOLAI! umumios. 
 
 so 
 
 Table showing the 
 
 Maximum I.ntensity of Solar 
 
 Radiation- 
 
 
 
 
 ON the Ecliptic. 
 
 
 
 
 i 
 
 1 
 SEPTEMBER. 1 
 
 OCTOBER. 
 
 Maximum. 
 
 Increment. 
 
 Maximum. 
 
 Increment. 
 
 'rah. 
 
 'Cmi. 
 
 'Fali. 1 
 
 • On*, j 
 
 'Fah. 
 
 ' Cmt. 
 
 •Fah. 
 
 •0«i<. 
 
 1 
 
 68.45 
 
 38.03 
 
 1.17 
 
 0.65 
 
 69.«0 
 
 38.67 
 
 2.32 
 
 1.29 
 
 2 
 
 68.48 
 
 38.04 
 
 1.20 
 
 0.66 
 
 69.64 
 
 38.69 
 
 2.36 
 
 1.31 
 
 3 
 
 68.52 
 
 38.06 
 
 1.24 
 
 0.68 
 
 69.68 
 
 38.71 
 
 2.40 
 
 1.33 
 
 4 
 
 68.55 
 
 38.08 
 
 1.27 
 
 0.70 
 
 69.72 
 
 38.73 
 
 2.44 
 
 1.35 
 
 5 
 
 68.59 
 
 38.10 
 
 1.31 
 
 0.72 
 
 69.76 
 
 38.75 
 
 2.48 
 
 1.37 
 
 6 
 
 68.63 
 
 38.12 
 
 1.35 
 
 0.74 
 
 69.80 
 
 38.78 
 
 2.52 
 
 1.40 
 
 7 
 
 68.66 
 
 38.14 
 
 1.38 
 
 0.76 
 
 69.84 
 
 38.80 
 
 2.56 
 
 1.42 
 
 8 
 
 68.70 
 
 38.17 
 
 1.42 
 
 0.79 
 
 69.88 
 
 38.82 
 
 2.60 
 
 1.44 
 
 9 
 
 68.74 
 
 38.19 
 
 1.46 
 
 0.81 
 
 69.92 
 
 38.84 
 
 2.64 
 
 1.40 
 
 10 
 
 68.78 
 
 38.21 
 
 1.50 
 
 0.83 
 
 69.96 
 
 38.87 
 
 2.68 
 
 1.49 
 
 11 
 
 68.81 
 
 38.23 
 
 1.53 
 
 0.85 
 
 70.00 
 
 38.89 
 
 2.72 
 
 1.51 
 
 12 
 
 68.85 
 
 38.25 
 
 1.57 
 
 0.87 
 
 70.04 
 
 38.91 
 
 2.76 
 
 1.53 
 
 13 
 
 68.89 
 
 38.27 
 
 1.61 
 
 0.89 
 
 70.08 
 
 38.93 
 
 2.80 
 
 1.55 
 
 14 
 
 68.93 
 
 38.29 
 
 1.65 
 
 0.91 
 
 70.12 
 
 38.95 
 
 2.84 
 
 1.57 
 
 15 
 
 68.97 
 
 38.32 
 
 1.69 
 
 0.94 
 
 70.15 
 
 38.97 
 
 2.87 
 
 1.59 
 
 16 
 
 69.01 
 
 38.34 
 
 1.73 
 
 0.96 
 
 70.19 
 
 38.99 
 
 2.91 
 
 1.61 
 
 17 
 
 69.05 
 
 38.30 
 
 1.77 
 
 0.98 
 
 70.23 
 
 39.02 
 
 2.95 
 
 1.64 
 
 18 
 
 69.09 
 
 38.38 
 
 1.81 
 
 1.00 
 
 70.27 
 
 39.04 
 
 2.99 
 
 1.66 
 
 19 
 
 69.13 
 
 38.40 
 
 1.85 
 
 1.02 
 
 70.31 
 
 39.06 
 
 3.03 
 
 1.68 
 
 20 
 
 69.16 
 
 38.42 
 
 1.88 
 
 1.04 
 
 70.35 
 
 39.08 
 
 3.07 
 
 1.70 
 
 21 
 
 69.20 
 
 38.44 
 
 1.92 
 
 1.06 
 
 70.38 
 
 39.10 
 
 3.10 
 
 1.72 
 
 22 
 
 69.24 
 
 38.47 
 
 1.96 
 
 1.09 
 
 70.42 
 
 39.12 
 
 3.14 
 
 1.74 
 
 23 
 
 69.28 
 
 38.49 
 
 2.00 
 
 1.11 
 
 70.46 
 
 39.14 
 
 3.18 
 
 1.70 
 
 24 
 
 69.32 
 
 38.51 
 
 2.04 
 
 1.13 
 
 70.50 
 
 39.17 
 
 3.22 
 
 1.79 
 
 25 
 
 69.36 
 
 38.53 
 
 2.08 
 
 1.15 
 
 70.54 
 
 39.19 
 
 3.26 
 
 1.81 
 
 26 
 
 69.40 
 
 38.56 
 
 2.12 
 
 1.18 
 
 70.58 
 
 39.21 
 
 3.30 
 
 1.83 
 
 27 
 
 69.44 
 
 38.58 
 
 2.16 
 
 1.20 
 
 70.61 
 
 39.23 
 
 3.33 
 
 1.85 
 
 28 
 
 69.48 
 
 38.60 
 
 2.20 
 
 1.22 
 
 70.65 
 
 39.25 
 
 3.37 
 
 1.87 
 
 29 
 
 69.52 
 
 38.62 
 
 2.24 
 
 1.24 
 
 , 70.08 
 
 39.27 
 
 3.40 
 
 1.89 
 
 30 
 
 69.56 
 
 38.64 
 
 2.28 
 
 1.26 
 
 70.72 
 
 39.29 
 
 3.44 
 
 1.91 
 
 31 
 
 
 ... 
 
 
 ... 
 
 ' 70.75 
 
 39.31 
 
 3.47 
 
 1.93 
 
90 
 
 lUBIAXT HEAT. 
 
 Table showikg the Maximum Intensity 
 
 OF Solar Eadiation 
 
 
 
 
 ox 
 
 THE Ecliptic. 
 
 
 
 
 si 
 
 A^ormiBEB. 
 
 DECEMBER. 
 
 Maximum. 
 
 Increment. 
 
 Maxiiniim. 
 
 Increment. 
 
 'Fah. 
 
 ° Cent. 
 
 ° Fah. 
 
 • Cent. 
 
 ° Fah. 
 
 ° Cent. 
 
 'Fah. 
 
 " Cent. 
 
 1 
 
 70.79 
 
 39.33 
 
 3.51 
 
 1.95 
 
 71.63 
 
 39.79 
 
 4.85 
 
 2.41 
 
 2 
 
 70.82 
 
 39.34 
 
 3.54 
 
 1.96 
 
 71.65 
 
 39.80 
 
 4.37 
 
 2.42 
 
 3 
 
 70.85 
 
 39.36 
 
 3.57 
 
 1.98 
 
 71.67 
 
 39.81 
 
 4.39 
 
 2.43 
 
 4 
 
 70.88 
 
 39.38 
 
 3.60 
 
 2.00 
 
 71.68 
 
 39.82 
 
 4.40 
 
 2.44 
 
 i) 
 
 70.92 
 
 39.40 
 
 3.64 
 
 2.02 
 
 71.70 
 
 39.83 
 
 4.42 
 
 2.45 
 
 6 
 
 70.95 
 
 39.42 
 
 3.67 
 
 2.04 
 
 71.72 
 
 39.84 
 
 4.44 
 
 2.46 
 
 7 
 
 70.98 
 
 39.43 
 
 3.70 
 
 2.05 
 
 71.74 
 
 39.86 
 
 4.46 
 
 2.47 
 
 8 
 
 71.02 
 
 39.45 
 
 3.74 
 
 2.07 
 
 71.76 
 
 39.87 
 
 4.48 
 
 2.48 
 
 9 
 
 71.05 
 
 39.47 
 
 3.77 
 
 2.09 
 
 71.77 
 
 39.87 
 
 4.49 
 
 2.49 
 
 10 
 
 71.08 
 
 39.49 
 
 3.80 
 
 2.11 
 
 71.78 
 
 39.88 
 
 4.50 
 
 2.50 
 
 11 
 
 71.11 
 
 39.50 
 
 3.83 
 
 2.12 
 
 71.80 
 
 39.89 
 
 4.52 
 
 2.51 
 
 12 
 
 71.14 
 
 39.52 
 
 3.86 
 
 2.14 
 
 71.81 
 
 39.90 
 
 4.68 
 
 2.52 
 
 13 
 
 71.17 
 
 39.54 
 
 3.89 
 
 2.16 
 
 71.82 
 
 89.90 
 
 4.54 
 
 2.52 
 
 14 
 
 71.20 
 
 39.56 
 
 3.92 
 
 2.18 
 
 71.84 
 
 89.91 
 
 4.56 
 
 2.53 
 
 15 
 
 71.24 
 
 39.58 
 
 3.96 
 
 2.20 
 
 71.85 
 
 39.91 
 
 4.57 
 
 2.53 
 
 10 
 
 71.27 
 
 39.59 
 
 3.99 
 
 2.21 
 
 71.86 
 
 89.92 
 
 4.58 
 
 2.54 
 
 17 
 
 71.29 
 
 39.60 
 
 4.01 
 
 2.22 
 
 71.87 
 
 39.92 
 
 4.59 
 
 2.54 
 
 18 
 
 71.32 
 
 39.62 
 
 4.04 
 
 2.24 
 
 71.88 
 
 39.93 
 
 4.60 
 
 2.55 
 
 19 
 
 71.35 
 
 39.64 
 
 4.07 
 
 2.26 
 
 71.89 
 
 39.93 
 
 4.61 
 
 2.55 
 
 20 
 
 71.38 
 
 39.65 
 
 4.10 
 
 2.27 
 
 71.90 
 
 39.94 
 
 4.62 
 
 2.56 
 
 21 
 
 71.40 
 
 39.67 
 
 4.12 
 
 2.29 
 
 71.90 
 
 39.94 
 
 4.62 
 
 2.56 
 
 22 
 
 71.42 
 
 39.68 
 
 4.14 
 
 2.30 
 
 71.91 
 
 39.95 
 
 4.63 
 
 2.57 
 
 23 
 
 71.45 
 
 39.69 
 
 4.17 
 
 2.31 
 
 71.91 
 
 39.95 
 
 4.63 
 
 2.57 
 
 24 
 
 71.47 
 
 39.71 
 
 4.19 
 
 2.33 
 
 71.92 
 
 39.96 
 
 4.64 
 
 2.58 
 
 2o 
 
 71.50 
 
 39.72 
 
 4.22 
 
 2.34 
 
 71.92 
 
 39.96 
 
 4.64 
 
 2.58 
 
 20 
 
 71.52 
 
 39.73 
 
 4.24 
 
 2.35 
 
 71.93 
 
 39.97 
 
 4.65 
 
 2.59 
 
 27 
 
 71.54 
 
 39.74 
 
 4.26 
 
 2.36 
 
 71.93 
 
 39.97 
 
 4.65 
 
 2.59 
 
 28 
 
 71.57 
 
 39.76 
 
 4.29 
 
 2.38 
 
 71.94 
 
 39.97 
 
 4.66 
 
 2.59 
 
 29 
 
 71.59 
 
 39.77 
 
 4.31 
 
 2.39 
 
 71.94 
 
 39.97 
 
 4.66 
 
 2.59 
 
 30 
 
 71.01 
 
 39.78 
 
 4.83 
 
 2.40 
 
 71.94 
 
 39.97 
 
 4.66 
 
 2.59 
 
 31 
 
 
 
 
 
 71.94 
 
 39.97 
 
 4.00 
 
 2.59 
 
CHAPTER V. 
 
 MECHANICAL ENEKGY OF SOLAR RADIATION. 
 
 The mechanical energy developed by solar radiation cannot 
 be accurately determined unless we possess means of ascertain- 
 ing (1) the energy actually called forth by the sun's radiant 
 heat at the surface of the earth, and (2) the energy lost during 
 the passage of the rays through our atmosphere. The illus- 
 tration shown on Plate 10 represents a solar calorimeter, 
 an instrument constnicted for measuring the energy actually 
 developed near the earth's surface. An instrument for mea- 
 suiing the amount of radiant heat absorbed by the atmosphere, 
 the actinometer, has been fully described in Chapter III. Evi- 
 dently, if we possess reliable means of ascertaining the heat 
 developed on a given area at the surface of the earth, and 
 that lost by atmospheric absoi-ption, we can state positively 
 what amount of dynamic energy is developed on a given area 
 by solar radiation at the boundary of the terrestrial atmo- 
 sphere. And since we know the tnie relation between the 
 semi-diameter of the sun and the distance from the sun's centre 
 
92 L'AliIAXT HEAT. chap. V. 
 
 to the earth, we can calciUate the exact degree of dispersion 
 of the solar rays on reaching the atmospheric boundary. Ac- 
 cordingly, we possess all the elements necessary to compute 
 the amount of mechanical energy transmitted by radiation 
 from a given area of the surface of the sun. Now, it will 
 be found, on referring to Chap. VI., that the sun emits heat 
 of equal energy in all directions ; hence we are enabled to 
 estimate the total amount of mechanical power developed and 
 transmitted by the sun as a motor. 
 
 Sir John Herschel and M. Pouillet conceived the idea, 
 nearly at the same time, of measuring the energy of solar 
 radiation by exposing a given quantity of water, presenting 
 a given area, to the sun's rays. Having ascertained the ele- 
 vation of temperature of the water acquired in a given time, 
 and added the energy suj)posed to be lost by atmospheric 
 absorption, together with the loss consequent on the disper- 
 sion of the rays, they computed the total dynamic energy 
 developed by the sun. Herschel employed a small, stationary, 
 open vessel, which he termed an actinometer ; while Pouillet 
 resorted to a close, movable vessel, the well-knowai " pyrhe- 
 liometre." The simple instrument devised by the great Eng- 
 lish astronomer, althoiigh very defective and incapable of 
 furnishing exact data, demands particular notice, since it 
 was employed in the ftrst investigation of a practical nature 
 intended to solve the important problem of solar energy. 
 The result of the investigation of Herschel, it will be remem- 
 bered, startled the Avorld by the inconceivable magnitude of 
 solar energy which it disclosed. The actinometer, agreeably 
 
CHAP. V. MECHANICAL ENERGY OF SOLAR RADIATION. 9o 
 
 to the following lucid statement, furnished by the distin- 
 guished designer himself, consisted of " a light cylindrical 
 vessel of tinned iron, open at the top, 3f inches in diameter 
 and 2.4 inches in depth, weighing 1,069 grains, nearly filled 
 with water moderately darkened by a slight admixture of 
 ink. This vessel was placed on a light wooden support, 
 covered with cotton cloth, and touching it only in a narrow 
 ring (to avoid the communication of heat by conduction), in 
 the interior of an iron cylinder of much larger diameter, to 
 protect it from wind and external radiation, the upper part 
 of which was covered by an iron plate ^vell protected from 
 sunshine by several separate diaphragms of paper laid lightly 
 one over the other thereon. This plate had a circular aper- 
 tui'e somcAvhat wider than that of the tin cylinder, and ver- 
 tically over it, centre corresponding to centre. The mouth 
 of the tin cylinder was covered with a circle of stiff paper, 
 having an aperture exactly circular and concentric with the 
 cylinder, so as to admit a vertical or nearly vertical sunbeam 
 somewhat less in section than the vessel, and loholly incident 
 on the surface of the contained liquid. This cover also pro- 
 jected over the exterior of the cylinder on all sides, so as 
 to prevent any ray from striking ou its outside, even when 
 the uj^per iron plate was removed from the exterior vessel. 
 Lastly, to cut off effectually all lateral radiation from the 
 region of sky near the sun, a paper diaphragm but very little 
 more in aperture than the mouth ol' the tin cylinder was laid 
 concentrically on the upper iron plate and its diaphragms. 
 Plunged into the liquid, and resting on the bottom when not 
 
94 BABIANT HEAT. chap. V. 
 
 in use, was a circular plate of mica '6\ inches in diameter, 
 attaclied to a light rod of reed 0.1 inch in diameter, for the 
 purpose of completely stirring and mixing the strata of the 
 liquid by one or two up and down movements. When thus 
 prepared, the whole apparatus was placed in the sunshine 
 at noon, or somewhat before, and so adjusted that, on the 
 admission of sunlight, a narrow ring of light surrounded con- 
 centrically the aperture in the diaphragm of the tin cylinder 
 beneath, which was carefully watched during the progress 
 of the experiment, and kept unaltered. These arrangements 
 being made, the sun was shaded off, the temperature of the 
 lit[uid (after stirring by the mica plate) taken by an exceed- 
 ingly delicate and sensitive thermometer by Crichton, and 
 again after a certain noted number of minutes. The shade 
 being then removed, the sun was allowed to shine into the 
 aperture on the liquid for ten minutes. During this exjio- 
 sure, the liquid was three times stirred by the mica plate, 
 allowing five seconds for each stirring, and shading the aper- 
 ture during that operation (which, of course, was not counted 
 as part of the ten minutes' exposure). The temperature was 
 now again taken, and, after remaining shaded again a certain 
 noted number of minutes, finally once more. The mean of 
 the minutely change of temperature, deduced from the shade 
 observations, being obtained, was applied as a correction (in 
 all cases a very small one) to the minutely elevation of tem- 
 perature in the sun exposure ; and thus the true effect of the 
 Sim was concluded." 
 
 The " pyrheliometre " having been described in nearly all 
 
CHAP. V. MECHANICAL EXEROT OF SOLAR RADIATION. 95 
 
 recent works on solar energy, and accurately delineated in 
 Pouillet's ":^lement8 de Physique" (Tome II., Paris, 1856), 
 it will only be necessaiy to remind the reader that the vessel 
 exposed to the sun, composed of polished silver, contains 100 
 grammes of water. The diameter is 1 decimetre and the depth 
 15 millimetres, the top exposed to the sun's rays being coated 
 with lamp-black. The radical defect of Pouillet's instrument 
 is that it cannot be used during winter when the tlierniometer 
 is below the freezing point, as warm water would have to be 
 used, in which case the loss of heat by radiation and convec- 
 tion would be so great as to render the task futile of accu- 
 rately measuring the force of solar radiation. This defect of 
 Pouillet's method is the more serious as the heat of the sun 
 is most intense during the winter solstice for given zenith 
 distances, on account of the diminished distance between the 
 sun and the earth, and because the sky is generally clearer 
 on a cold winter's day than during the heat of summer when 
 the air is charged with vapor. 
 
 The loss of heat by radiation, in the pyrheliometre ; the 
 loss of heat by convection, accelerated by currents of air ; the 
 absence of adequate means for circulating the fluid contained 
 within the heater ; the rade method of keeping the instrument 
 pei-pendicular to the sun wiih the hand, not to mention tlu- 
 disturbing influence of respiration and the radiation from the 
 operator's body, are self-evident defects. Nor can we pass 
 unnoticed the want of any direct means of ascertaining the 
 depth of the atmosphere through which the radiant heat 
 passes at the moment of measuring its energy. I need scarcely 
 
96 BADIAXT HEAT. en A p. v. 
 
 point out that computations based on latitude, date, and exact 
 time are too complex and tedious for investigations in which 
 the principal element, the depth of the atmosphere, is con- 
 tinually changing. 
 
 It will be well to state that the solar calorimeter, and 
 all my instruments constructed for investigating the mecha- 
 nical properties of solar heat, are attached to a vibrating 
 table applied within a revolving observatory, supported on 
 horizontal journals and provided with a declination movement 
 and a graduated arc. Consequently, the sun's zenith distance 
 may at all times be ascertained by mere inspection, a very 
 great convenience in an investigation which at every instant 
 is dependent on the changing depth of the atmosphere through 
 \\hich the solar rays pass. As this depth bears a fixed rela- 
 tion to the sun's zenith distance, it may of course be accurately 
 determined by noting the position of the fixed index on the 
 graduated arc ; but, as already pointed out, there is no time 
 during investigations of this kind for computations. I have, 
 therefore, constructed a graduated scale provided with a mov- 
 able radial index, which, by being brought to the division 
 corresponding with the observed zenith distance, shows the 
 depth of atmosphere (see diagram in Chap. III., Plate 9). 
 It is proper to observe that, in constructing this scale, I have 
 assumed the earth to be a perfect sphere of 3,956 miles radius. 
 The error resulting from this assumption is, however, so trifling 
 that the described graphic method of ascertaining the dejith 
 of the atmosphere may, without appreciable error, be employed 
 for all latitudes. The solar calorimeter consists of a double 
 
CHAP. V. MLCIIASIVAL KyEnay OF no LA n 1;A1>IATUk\. 'J7 
 
 vessel, cyliudriciil at the bottom uikI conical tit the top, an 
 8-iii. lens being inserted at the wide end in the manner shown 
 by the illustration. The interior is lined with polished silver, 
 the space between the two vessels being closed at the toji and 
 l)ottom Ijy means of perforated rings, as shown in the trans- 
 verse section. The object of these perforations is that of dis- 
 tributing equally a current of water to be circulatc<l thrcmgh 
 the s]>ace between the vessels. Nozzles are ajipliiMJ at the 
 top and bottom of the external vessel, of suitable form to 
 admit of small flexible pipes being attached. A stop-cock 
 with coupling-joint is applied at the bottom, comnmnicating 
 witli the interior chamber of the calorimeter and connected 
 with an air-pump, for exhausting the same. A cylindrical 
 vessel, ^\•ith closed ends, composed of ])olislied silver, is secured 
 in the lower part of the interior chamber, and provided with 
 a conical nozzle at the top, through which a thermometer is 
 inserted from -without. Within the lower part of this cylin- 
 drical vessel a centrifugal paddle-wheel is applied, surrounded 
 by a cylimhii'al casing divided into two compartments by a 
 circular diaphragm. The lower compartment contains four 
 i-adial wings, or paddles, the diaphragm being perforated in 
 the centre. The said paddle-wheel revolves on a vertical axle, 
 which passes through a stuffing-box applied at the bottom of 
 the surrounding vessels, the rotary motion being imparted by 
 means of a pulley secured to the lower end of the axle. The 
 operation of this wheel, designed to proipote perfect circula- 
 tion of the fluid witliin the cylindrical vessel when charged, 
 is quite peculiar. It will be readily understood that the 
 
98 RADIANT HEAT. CHAP. X. 
 
 centrifugal action produced by the rotation of tlie paddles 
 Avill draw in water downwards through the central perfora- 
 tion of the diaphragm, and force the same into the annular 
 space round the casing of the wheel ; thus an upward current 
 ^vill be kept \\]) through this annular space uniform on all 
 sides. The circulating water, after reaching the top of the 
 heater, will then retui-n, first entering the open end of the 
 casing of the wheel, and ultimately the central perforation 
 of the diaphragm. I have been thus particular in describing 
 this system of pi-omoting uniform cii'culation, because a correct 
 indication of the mean temj)erature of the water contained 
 within the vessel subjected to the action of the concentrated 
 rays, is the all-important condition on which depends the 
 accuracy of the determination of the number of thermal units 
 developed by the radiant heat. It only remains to be pointed 
 out that the lens, which is so proportioned as to admit a sun- 
 beam of 53.45 sq. ins. of section, is j)laced at such a distance 
 from the heatei' that when the concentrated rays reach the 
 upper end (coated with lamp-black) they are confined to an 
 area of 3.35 sq. ins., viz., -j^ of the sectional area of the pencil 
 of rays which enters the lens. 
 
 It will be obvious that the concentration of the radiant 
 heat on an area of only one-sixteenth of that of the section 
 of the pencil of rays admitted to the instrument removes 
 a very difiicult disturbing element from the investigation — 
 namely, the great amount of heat radiated by the blackened 
 surface of the heater, which in the pyrheliometre is 16 times 
 greater for a given amount of radiant heat than in the solar 
 
Cll.vi'. V. MEVUASIVAL ESKUGY OF SOLAR It AVIATION. 911 
 
 caloiimt'ter. But this is not all ; while the extensive black- 
 ened sui-face of the former is exposed to currents of air, the 
 disturbing effect of wliicli can neither be controlled nor com- 
 puted, error arising from convection is wholly removed from 
 the latter, because the reduced blackened surface oi the vessel 
 exposed to the solar rays receives the concentrated radiant heat 
 within a vacuum. The loss of heat at the bottom and sides 
 of Pouil let's instrument, caused liy convection and currents of 
 air, is likewise wholly removed in the solar calorimeter \)\ 
 the expedient of operating within a vacuum. It ^vill be seen, 
 therefore, that the loss from these causes has been wholly obvi- 
 ated in this instrument, while the loss occasioned by radiation 
 from tlie blackened surface which receives the concenti-ated 
 radiant heat has been reduced to a mere fi'action. It may be 
 contended, however, that tlie loss by radiation of the heater 
 against the interior surface of the calorimeter, although minute, 
 is yet appreciable, and that some heat will be lost by conduc- 
 tion at the points where that vessel joins the surrounding 
 chambers. Even these trifling sources of erroi', it will l^e seen 
 presently, have been removed by the new method. A force- 
 pump and a cistern containing water maint<ained at a constant 
 temperature of 60° F. are arranged near the calorimeter. By 
 means of this ]iump and the flfxilile pipes before referred to, 
 a constant ciin-ent is kept up through the space between the 
 internal and external casings of the instrument ; hence the 
 materials composing the lattei- may quickly be brought to 
 the same temperature as the cii'culating watei'. The process 
 of measuring the radiant energy is conducted in the following 
 
100 KABJANT HEAT. CHAP. v. 
 
 maimer: The tliciiiKuiieter being witlulrawn, tlie cvliiidrical 
 vessel <>i' licatiT is cliarged with distilled \vater of a tempe- 
 rature of al)<)iit 45° F., after whieli the thermometer is again 
 inserted and the instrument exposed to the sun, the paddle- 
 wheel ])eing kept in motion. The indication of the thermo- 
 meter must then be watched, and the time accurately noted 
 Avhen the mercurial column marks 50° on the scale, the obser- 
 vation continuing until the thermometer marks 70°, at Avhieh 
 point tile time is again accurately noted. The experiment 
 being then concluded, the lens should be covered. The cir- 
 culating water being kept at a constant temperature of 60° 
 F., it scarcely needs explanation that, during the elevation of 
 the temperature of the water from 50° to 60°, the instrument 
 radiates fowarJn (lie lieater / and that, while the temperature 
 rises from 60° to 70°, the heater radiates towanh the instni- 
 mevt. In each case the amount of heat radiated and received 
 is almost inappreciable, since the vessel containing the water 
 to be heated and the surrounding vessel are composed of 
 highly polished metal. The amounts of gain and loss of heat 
 by conduction at the points Avliere the heater is joined to the 
 external vessel, if appreciable, evidently balance each otlier 
 in the same manner as the gain and loss by radiation. 
 
 The weight of distilled water at 60° contained in the 
 heater, and the weight and specific heat of the materials 
 which compose its parts, being ascertained, tlie number of 
 thermal units necessary to elevate the temperature of the 
 whole 20° F. may be readily calculated. To this must be 
 added the percentage of calorific energy lost dui'iiig the pas- 
 
CHAi-. V. MEVIIAMCAL EKEUCY OF SOLAE TiAI>TATION. 101 
 
 sage of the sun's rays tluough the lens. The sum will repre- 
 sent a permanent coefficient foi- each particular instrument. 
 Obviously, the indication of the solar calorimeter will not be 
 less reliable during winter in a northern latitude, with the 
 mercury at zero, than duiing summer within the tropics, when 
 the thermometer marks 100° in the shade. Noi- must it be 
 supposed that the same difficulty presents itself in ascei-tain- 
 ing the loss of energy of the rays of heat as that involved 
 in a determination of the retardation A\hich rays of light 
 suffer during their passage through a lens. In order to deter- 
 mine the former, we have only to comjiare the units of heat 
 developed by the direct action of a pencil of rays of a given 
 section with the number of units developed by another pencil 
 of equal section, acting during an equal interval and at the 
 mme time, through the lens the retaiding influence of which 
 we desire to ascertain. 
 
 The weight of water contained in the heater of the solar 
 calorimeter employed during the investigations refen-ed to in 
 this work is 0.8125 11). avoirdupois, the weight of the materials 
 composing the heater, paddle-wheel, and other parts l>eing 
 0.20.8 lb. As the specific heat of these materials is 0.12"), it 
 will be evident that 0.125 X 0.298 = 0.0;?72 lb. should be 
 added to the weight of water contained in the heater. Ac- 
 cordingly, the total weight will amount to 0.8125 + 0.0.'i72 
 = 0.8497 lb. The elevation of temperature in the heater 
 l)eing fi.ved at 20^ F., it will be seen that the dynamic enei-gy 
 developed during each experiment will ;tmount to 20 X 0.8497 
 = 1G.994 tlwrmal units, besides the energy absorbed l)y the 
 
102 RADIANT HEAT. citai'. V. 
 
 lens, wliicTi, agreea1)]y to actual trial, amounts to very nearly 
 0.10. Consequently, 0.10 X 16.994 + 16.994 = 18.6934 ther- 
 mal nnits represent a permanent coefficient of energy for the 
 particular instrument referred to. Let us clearly understand 
 that, at the conclusion of each experiment, whatever be the 
 time occupied in attaining the stii)ulated 20° F., the stated 
 amount of energy, viz., 18.6934 thermal units, has been deve- 
 lojied. We are, therefore, enabled to determine the amount 
 of mechanical energy developed by solar radiation at the sur- 
 face of the earth by observing the time occupied in attaining 
 the stipulated temperature, and then dividing the coefficient 
 of energy by the time thus observed. But it will be per- 
 ceived, on reflection, that, in order to solve the important 
 problem of solar emission, the following conditions must be 
 fulfilled at the time of conducting the experiment : (1) The 
 sun must be perfectly clear. (2) The position of the earth 
 in the orbit must be knoAvn in order to enable lis to deter- 
 mine the distance of the sun and the consequent dispersion 
 of the rays during the observation. (3) The sun's zenith 
 distance must be known, since the loss of radiant energy 
 by absorption depends on the depth of atmosphere pene- 
 trated by the rays. The second and third of these condi- 
 tions are of course readily met ; but the first condition can 
 only be fulfilled by repeating the observations during a 
 series of years whenever the sun is exceptionally clear. The 
 writer feels confident that, by having adopted this system, 
 the problem of solar emission has been satisfactorily solved. 
 An account of the observations successively made being devoid 
 
cu.U'. V. MEVUAyiCAL ENEliGT OF SOLAR RADIATION. 103 
 
 of interest, it will be suflicieut to state that the observed 
 maximum solar intensity occurred March 7, 1871, the sun 
 beiuo- then so clear that the before-mentioned amount of 
 18.6934 thermal units was developed in 10 min. 0.5 sec, 
 
 hence — '- = 1.8678 units per minute. The sectional 
 
 10.00833 ^ 
 
 area of the pencil of rays entering the solar calorimeter wa.", 
 as already stated, 0.37187 square foot. Consequently, if we 
 reduce the foregoing elements to the usual standard — one 
 square foot of area acted tipo7i by the sun in one minute — 
 it will be found that, on the occasion I'eferred to, an energy 
 of 5.03 units of heat per minute was developed by a pencil 
 of solar rays of 1 square foot section. The mean zenith dis- 
 tance during the experiment was 46 deg. 5 rain., -while the 
 position of the earth in the orbit was such that the sun's rays 
 suffered a dispersion of 45,400 to 1. Keferring to the table of 
 temperatures for given zenith distances (see page 62), it will 
 be found that the radiant intensity at 46 deg. 5 min. zenith dis- 
 tance is diminished in the ratio of 67°.2 : 59°.85. The energy 
 developed by oui- calorimeter during the experiment was, of 
 coui-se, reduced in the same proportion. Introducing, then, 
 the necessaiy correction for the stated loss caused by zenith 
 distance — i.e., atmospheric absorption— the tnie energy deve- 
 loped by the radiation at the surface of the earth during the 
 
 67.20 X 5.03 .^^ .^ . , p^ 
 
 experiment was — — = o.64 units per minute. Ke- 
 ferring to Chap. III., it will be found that the temperature 
 produced by solar radiation at the boundary of the terrestrial 
 
104 B AVIAN! HEAT. chap. V. 
 
 atiuospliere is U.207 greater tbau that developed near the siir- 
 faee of the earth ; in other words, the eneigy absorbed b}- 
 the atmosphere is to that transmitted to the eai-th as 0.207 : 
 0.793. Consequently, the energy develojied by solar radia- 
 tion at the boundary of the atmos^phere, March 7, 1871, was 
 
 5.64 WIT- !■ , J5 P 
 
 = 7.11 thermal units ou one square toot of suriace ; 
 
 0.793 ^ 
 
 while the dispersi<,)u of the rays on that day was in the ratio 
 of 45,400 to 1. It needs no demonstration to prove that, 
 according to this ratio of dispersion of the rays, the energy 
 emanating from one square foot of the photosphere must 
 heat 45,400 square feet of surface at the boundary of the 
 teri'estrial atmosphere. Our investigation having shown that 
 solai" radiation develops an energy of 7.11 units to the square 
 foot on entering the terrestrial atmosphere, it follows that 
 solar emission amounts to 45,400 X 7.11 = 322,794 thermal 
 units in one minute for each square foot of the photosphere. 
 In view of the completeness of the means adopted in mea- 
 suring the energy developed, and the ample time which has 
 been devoted to the determination of maximum intensity, it 
 is not probable that future labors will change the result of 
 our investigation. The continuous shrinking of the sun will 
 produce a perceptible diminution of the radiant energy trans- 
 mitted to the earth in the course of a few hundred centuries, 
 but the emissive energy for a given area of the sun \vill remain 
 constant for millions of years, since the intensity developed 
 by the falling mass will increase inversely as the square of 
 its distance from the solar centre, thus balancing the dinii- 
 
CHAP. T. MKcn.iMCAL i:.\i:i;ay oFsoLAi: i:ai>i.\ti<)x. loo 
 
 uutiou of energy consequent on the rediicecl fall of the 
 mass. 
 
 The illustration sliowu on Platte 11 represents a vertical 
 section of a portable solar calorimeter, in all essential features 
 similar to the instrument descrihed in the present chapter, 
 the only material ditVerence l)eing that of employing a W^- 
 actimj circulating \vheel within the heater. Referring to the 
 illustration, it will lie seen that the instrument is placed on 
 an ordinary table, a weight being suspended under the same 
 for actuating the circulating wheel. The cylindrical chamber 
 which contains the heater moves on a hinge secured to a 
 circular bed-plate i^rovided with cogs, turning round a ver- 
 tical pivot fastened to the top of the table, the inclination 
 being regulated by a tangential screw. A horizontal pinion, 
 geared into the cogged bed-plate referred to, enables the ope- 
 rator to follow the diurnal motion, while the tangential screw 
 enables him to regulate the inclination of the lens with refe- 
 rence to the sun's declination. Appropriate sights are applied 
 to the front side of the cylindrical chambei', showing when 
 its axis points tow-ards the sun's centre, while a graduated 
 (piadrant indicates the zenith distance at all times. It will 
 l)e found, by inspecting the illustration, that the axle of the 
 bai-rel actuated by the motive weight is connected by a train 
 of cog-w^heels to the shaft of the circulating wheel within 
 the heater. The perfect regularity of rotation imparted to 
 this wheel, and the consequent perfectly uniform circulation 
 kept up within the heater, dispenses with the necessity of 
 exhausting the air from the cylindrical chamber, on the fol- 
 
100 RADIANT HEAT. CHAP. v. 
 
 lowing grounds : The beat imparted by the air \vithin the 
 chamber during the fii'st half of the experiment balances the 
 heat absorbed during the second half. There is a difference, 
 but too small to cause an appreciable error. It may be men- 
 tioned that the portable solar calorimeter thus described was 
 originally constructed for ascertaining the dynamic energy 
 developed by solar radiation on the plains of India and in 
 xViisti'alia.. 
 
CHAPTER VI. 
 
 THERMAL ENERGY TRANSMITTED TO THE EARTH I'.Y 
 
 RADIATION EROi[ DIFFERENT PARTS OF THE 
 
 SOLAR SURFACE. 
 
 Pere Secchi, in the second edition of " Le Soleil," pub- 
 lislied at Paris, 1875, calls special attention to tlie result of 
 his early investigations of the force of radiation emanating 
 from different regions of the sun's surface, reiterating with- 
 out modification his former opinions regarding the absorption 
 of the radiant heat by the solar atmosphere. It will be well 
 to bear in mind that the plan adopted by the Italian physi- 
 cist in his original researches, on which his present opinion is 
 based, was that of projecting the sun's image on a screen, 
 and then, by means of thermopiles, measuring the tempera- 
 ture at different points. The serious defects inseparable from 
 this method of measuring the intensity of the radiant heat 
 I need not point out, nor Avill it be necessary to urge that 
 a correct determination of the energy transmitted calls for 
 direct observation of the temperature produced by the rays 
 projected towards the earth. Accordingly, on taking up that 
 
108 EAVIANT HEAT. CHAP. VI. 
 
 brancli of my investigations of radiant licat which relates to 
 the difference of intensity transmitted from different parts of 
 the sun's surface, T adopted the method of direct observation. 
 The progress was slow at the beginning, owing to the neces- 
 sity of constructing an astronomical apparatus of unusual 
 dimensions ; but having devised means which rendered the 
 employment of any desirable focal length easy, the work has 
 progressed rapidly. An instrument of 17.7 metres (58 feet) 
 focal length, erected to conduct preliminary experiments, has 
 proved so satisfactory that the construction of one of 30 metres 
 focal length, which I supposed to be necessaiy, has been dis- 
 pensed with. Considering that the apparent diameter of the 
 sun at a distance of 17.7 metres from the observer's eye is 
 162.4 millimetres, even when the earth is in aphelion, the 
 efficacy of the instrument employed might have been antici- 
 pated. The nature of the device will be readily comprehend eil 
 by the following explanation: Suppose a telescopic tube 17.7 
 metres long, 1 metre in diameter, devoid of object-glass and 
 lenses, and mounted equatorially, to be closed at both ends 
 by metallic plates or diaphragms, at right angles to the tele- 
 scopic axis ; suppose the diaphragm at the upper end to be 
 perforated with two circular apertures 200 millimetres in 
 diameter, situated one above the other in the vertical line, 
 3 GO millimetres from centre to centre ; and suppose a third 
 circular perforation whose area is one-fifth of the apparent 
 area of the solar disc — viz., 72,6 millimetres diameter — to be 
 made on either side of the vertical line ; suppose, lastly, 
 that the diaphragm wliich closes the lower end of the tube 
 
CH.VP. vr. UNEQUAL UADIATIOX OF SOLAR DISC. 109 
 
 be pei-forated witli three snmll apertures luilliinetres in dia- 
 meter, whose centres correspond exactly with the centres of 
 the three large perforations in tlie upper diaphragm. The 
 tube being then dii-ccted tuwaiils tlie sun, and actinometers 
 applied below the thi'ee small apertures in the lower dia- 
 phragm, it will be evident that two of tliese instruments will, 
 after due exposure to a clear sun, indicate maximum solar 
 iutensiity, say 3;")° C, while the actinometer applied in line 
 with the perforation whose area is one-fifth of the apparent 
 
 area of the solar disc will indicate — = 7° C, unless the 
 
 central portion of the solar disc radiates more powerfully 
 towards the earth than the rest, in which case a higher inten- 
 sity than 7° C. "will be indicated by the actinometer referred 
 to. It will be readily understood that the solar rays entering 
 thi'ough the perforations at the upper end of the tube con- 
 verge at the low^er end and pass through the small perfora- 
 tions, causing maximum indication of the focal actinometera 
 as stated. Now, suppose that a circular plate, the area of 
 which is exactly four-fifths of the appai'ent area of the sun — 
 viz., 145.2 millimetres diameter — be inserted concentrically in 
 either of the two large perforations of the diaphragm at the 
 top of the telescopic tul)e. The apparent diameter of the 
 sun being, as before stated, 1G2.4 millimetres, it will be per- 
 ceived that the inserted plate will only pai-tially exclude the 
 solar radiation, and that the rays from a zone 1' 42" wide 
 will pass outside the said plate, converging in the form of 
 a hollow cone at the lower end of the tulte, and there enter 
 
110 RADIANT HEAT. CHAP. VI. 
 
 the respective actiuometer. The iudicatiou of the latter will 
 then show the thermal energy transmitted by radiation from 
 a zone whose mean width extends 49" from the sun's border. 
 It should be particulai'ly observed that the three focal acti- 
 nometers employed will be acted upon simtdtaneously by the 
 converged I'ays, (1) from the entire area of the solar disc, 
 (2) from a central region containing one-fifth of the area, 
 and (3) from a zove at the border containing also one-fifth 
 of the area of the solar disc. It is scarcely necessary to 
 point out that an accurate comparison of the intensity of the 
 radiant heat emanating from the central pai't and from the 
 sun's border calls for simtdtaneovs observation, in order to 
 avoid the errors resulting from change of zenith distance and 
 variation of atmospheric absorption during the investigation. 
 The great advantage of obtaining also a simultaneous indica- 
 tion of the intensity transmitted by radiation from the entire 
 solar disc is self-evident, since this indication serves as an 
 effectual check on the observed intensities emanating from 
 the centre and from the harder. The latter obviously must 
 be less, while the former must be greater, for a given area, 
 than the indication of the focal actiuometer which receives 
 the radiation of the entire solar disc. 
 
 The foregoing demonstration, based on h}'pothesis, having 
 established the possibility of ascertaining by direct observa- 
 tion the temperature produced by the rays projected from 
 certain parts of the solar surface, let us now examine the 
 means actually employed. An observer on the 40th deg. 
 latitude, stationed on the north side of a building 28 metres 
 
CHAP. VI. UXEQUAL FADTATTOX OF SOLA!.' DfSC. m 
 
 liigli, pointing east and west, can just see the sun pass the 
 meridian, duiing the sunmier s<il.«tice, if he occu2>ies a posi- 
 tion about 8 metres fi'om such buihling. Kow, if an opaque 
 screen, perforated by a circuhir opening .".1."') niillinieti-es in 
 diameter, be placed on the top of tlie su])po.';ed building, the 
 entire solar disc may be seen through the same, pro\nded it 
 faces the sun at right angles. But if the pei-foration in the 
 said screen be 140 millimetres in dianietci-, only one-fifth of 
 the area of the solar disc will be seen. And if the screen be 
 removed and a circular plate 280 millimetres in diameter put 
 in its place, the observer, ranging himself in line with the plate 
 and the sun's centre, can see only a narrow border 1' 42" of 
 the solar disc. Obviously the screen placed on the io-p of 
 the building might be perforated lil:e the upjier diaphragm 
 of the supposed telescopic tube, and a plate resembling the 
 lower diaphragm, secured by apjiropriate means near the 
 ground, might be made to support the focal aotinometers in 
 such a manner that their axes pass through the centres of 
 the perforations of the screen above the building. It is 
 hardly necessarj' to state that the plate supporting the aoti- 
 nometers should be attached to some mechanism capable of 
 imjiarting to it a parallactic movement, during the observa- 
 tion, corresponding with the sun's declination and the earth's 
 diurnal motion, and that some adequate mechanism should 
 be employed for regulating the position of the perforated 
 screen and adjusting the focal distance in accordance vaih 
 the change of the subtended angle consequent on the vary- 
 ing distance from the sun. It will be evident that, since 
 
113 liADIANT HEAT. chap. vi. 
 
 the first-iiauieJ mecluiiiisiu rests on tLe ground, wliile tlie 
 latter is secured to a massive building, far greater steadi- 
 ness will be attained by our simple and comj^aratively inex- 
 [)ensive device tlian by employing a telesco^iic tube of the 
 most perfect construction mounted equatorially. 
 
 Witli reference to tlie influence of diffraction, it should 
 be stated tliat, before deteiiuining the size of the screens in- 
 tended to shut out cei-tain parts of the soLnr disc during the 
 investigation, the amount of inflection of the sun's rays was 
 carefully ascertained. Two distinct methods were adopted : 
 (1) measuring the additional amount of heat transmitted to 
 the focal thermometers in consequence of the inflection of 
 the rays ; (2) increasing the tlieoretical size of the screens 
 until the effect of inflection was overcome and the luminous 
 rays completely excluded. Regarding the first-named method 
 of ascertaining the diffraction, it is important to mention that 
 the temperature transmitted to the focal actinometers by the 
 inflected radiation which passes outside of the theoretically 
 determined screens is not proportionate to the inflection ascer- 
 tained by the process of enlargement referred to. This cir- 
 cumstance at first rendered the investigation somewhat com- 
 plicated, but it soon became evident that the discrepancy was 
 caused by tlie comparatively small inflection of the invislhh 
 heat rays. It will Ije seen jiresentlj^ that the radiant heat 
 which passes outside of the screens in consequence of diffrac- 
 tion is considerably less than that which -would be transmitted 
 to the focal actinometers if the calorific rays ^vel•e sulijected 
 to an amount of inflection corresponding with the enlargement 
 
CUAP. VI. UNEQUAL HAI'IATIOX OF SOLAR DISC. 113 
 
 of the screens beyoucl the tbeuietical dimeusious uecessaiy to 
 excliule tLe luminous rays. 
 
 Let us first consider the metht)d of ascertaining the inflec- 
 tion of the rays by measuring the additional amount of heat 
 transmitted to the focal actinometers. Fig. 1 (see Plate 12) 
 represents the solar disc, a being the focal actinometer exposed 
 to the converged rays, a' a' representing an imaginary plane 
 situated 17.7 metres from a, at which distance the section of 
 the pencil of converging rays will be 162.4 millimetres in 
 diameter, provided the earth is near aphelion. Fig. 2 also 
 represents the solar disc, and c the actinometer exposed to 
 the converged rays ; but a perforated screen b' b' is intei-posed, 
 the perforation being of such a size that only the rays pro- 
 jected by the central half of the solar disc (indicated by the 
 circle h />) pass through the same and reach the focal actino- 
 meter. The screen b' b' being situated 17.7 metres from c 
 M-heu the earth is in the position before referred to, the said 
 perforation must be 114.83 millimetres in diameter, in order 
 that the lines b x' c may be straight. Fig. 3 likewise repre- 
 sents the solar disc, its area being divided into two concentric 
 halves by the circle d d ; but, in place of a perforated screen, 
 an opaque circular screen d' is introduced at the same distance 
 from the focal actinometer as in Fig. 2 ; consequently, the lines 
 d y' f ^vill be straight. Now, if the actinometei-s a, c, and / 
 be exposed to the converged solar radiation simultaneously 
 and during an eqval interval of time, c and / receiving the 
 heat from one half of the solar disc (the former from the 
 central and the latter from the sun-ounding half), the tempe- 
 
114 RADIANT HEAT. chap. vi. 
 
 ratures of c and / added together should coiTesjDoud exactly 
 with the temperature transmitted from the entire solar disc 
 to a. Observation, however, shows that the temperature of 
 c and /together is 0.091 greater than the temperature imparted 
 to a. Hence an increase of temperature of nearly one-eleventh 
 is produced by the inflection of the calorific rays, one-half 
 being the result of the bending of the rays vnthin the per- 
 foration of the screen h' h', the other half resulting from the 
 bending outside of the screen d'. The increment of tempe- 
 rature being thus known, the degree of inflection may be 
 easily determined by drawing a circle x x round the circle 
 
 7 7 • •,-...-, „ 0.091 
 
 b, covermg an additional area of = 0.0455; and by 
 
 inscribing a circle y y within d d, covering an area of 0.0455 
 less than the area of d d. It will be perceived, on reflection, 
 that X x' b represents the angle of inflection of the calorific 
 rays within the perforation of the screen b' b', and that d y' y 
 represents the angle of inflection outside of the screen d'. 
 Demonstration shows that the former angle measures 14".57, 
 while the latter measui-es 14".86, the mean being ]4".7l. 
 Having thus determined the inflection resulting fi-om invi- 
 sible radiation, let us now ascertain the inflection of the 
 luminous rays. As before stated, the apparent diameter of 
 the sun at a distance of 17.7 metres from a given point is 
 162.4 millimetres when the luminary is fui-thest from the 
 earth. Now, our investigation shows that a screen 167 milli- 
 metres in diameter hardly sufiices to exclude the luminous 
 
 1 ^1 ■ ■ a .■ 167 - 162.4 
 rays ; hence their inflection amounts to = 2.3 
 
CHAP. VI. UXEQUAL EADIATIOy OF SOLAR DISC. 115 
 
 millimetres in a length of 17.7 metres. Their angle of inflec- 
 tion \vill therefore be 26".81, against 14".71 for the dark rays. 
 We have thus incidentally established the fact that the inflec- 
 tion of the luminous and calorific rays differs nearly in,' the 
 same proportion as the calorific energies of the invisible and 
 visible portions of the solar spectrum. 
 
 The illustration on Plate 13 represents a top view (see 
 Fig. 15) and a transverae section (see Fig. 17) of the paral- 
 lactic mechanism employed in the investigation. The leading 
 feature of the device is that of attaching three actinometers, 
 f, Ji, and (/, to a plate which may be set at any desired 
 inclination, and capable of being moved simultaneously at 
 right angles to, and in a direction parallel vpith, the meri- 
 dian. The mode of effecting this movement will be readily 
 understood by the following description, reference being had 
 to the illustration : a is a screw, the threads of which are 
 foi-med to a pitch of three-eighths of an inch, placed hori- 
 zontally and at right angles to the meridian, the ends turn- 
 ing in bearings bolted to a substantial frame /; ?>, supported 
 by legs resting on a solid stone foundation. A radial arm 
 c, the position of which is regulated by a graduated quadrant 
 c', is fastened to the end of the screw a; the latter being 
 by that means prevented fi-om turning round, d J, arms con- 
 nected by a cylindrical socket d', which slides freely back 
 and fonvards on the screw. The said socket is prevented 
 from turning round the screw by the application of a square 
 key a', fitted accurately into a rectangular longitudinal groove 
 formed in the side of the screw, e e, plate sliding between 
 
116 RADIANT HEAT. chap. vi. 
 
 appropriate guide-rods secured to the ujiper side of tlie arms 
 d d, motion being imparted to this plate by a micrometric 
 screw e'. The actinometers /, (/, and h are attached to a 
 plate I; bolted to the top of e e. The sliding socket d' is 
 moved along the main screw by a milled nut /, held against 
 the end of the said socket by a forked piece I' fastened to 
 the arm d and acting on a collar formed at the small end 
 of the milled nut. It scarcely needs explanation that, by 
 turning this nut, the sliding socket d' may be made to move 
 along the main screw in either direction, thereby imparting 
 motion to the plate Js which supports the three actinometers. 
 Nor will it be necessary to demonstrate that, by turning the 
 micrometric screw <?', the said plate may be moved at right 
 angles to the main screw a. Consequently, the mechanism 
 thus described enables us to move the actinometers with 
 great regularity and precision across the meridian, and in a 
 direction parallel with it. By means of a flexible tube m m, 
 which connects the surrounding casings of the three actino- 
 meters, a stream of water is circulated through the latter 
 during observations. An ordinary force-pump is employed 
 for this purpose, attached to a capacious cistern containing 
 water maintained at a constant temperature. The centres 
 of the actinometers / and (/ are 160 centimetres apart, a 
 colored glass n being attached to the jilate 7c in a direct 
 line vsdth, and equidistant from, the stated centres. An eye- 
 piece is applied to the colored glass w, below the plate k 
 The actinometers are provided with conical apertures on the 
 upper side, through which the thermometers are introduced 
 
cn.vp. VI. Uy EQUAL RADIATION OF SOLAR DISC. 117 
 
 when the instrument is in operation. In the illustration these 
 apei-tures are closed by conical plugs provided with globular 
 handles. Fig. 16 represents a square bar /•, to which three 
 circular discs, composed of sheet brass, are attached by means 
 of deep and thin arms, foi-med as sho^vn by the drawing. 
 The discs /' and </' are placed 160 centimetres apart, fi-om 
 centre to centre, corresponding exactly with the distance 
 between the axes of the actinometers / and ff ; while the 
 centre of the disc s con-esponds with the centre of the eye- 
 piece below the colored glass n. The square bar r is held 
 horizontally and at right angles to the meridian by a sub- 
 stantial bracket secured to the top of some building of ade- 
 quate height ; the angular position of the bar being such 
 that it corresponds with the sun's zenith distance. The fi-ame 
 b h, which .supports the parallactic mechanism, rests on a level 
 stone foundation, its distance from the building referred to 
 depending on the season and the latitude of the place of 
 observation. Assuming that the bar ;•, which supports the 
 discs f, s, and ;/', has been correctly placed, and that the 
 parallactic mechanism occupies a proper position on the 
 ground, it will then be found that, when the sun passes 
 the meridian, the disc s will throw a small round shadow 
 covering the colored glass n. Obviously, an operator lying 
 on his back under the frame which supports the instniment, 
 with his right hand turning the milled nut I, and his left 
 luuul turning the micrometric screw e', will be enabled to 
 impart a simultaneous right-angular movement to the acti- 
 ii(»nit'tei-s. Now, the diameter of the flisc s- is such that, 
 
118 RADIANT HEAT. chap. ti. 
 
 when brouglit in line mth the sun and the eye-jiiece below 
 n, only the extreme edge of the solar disc is seen through 
 the colored glass. The operator, therefore, by careful mani- 
 pulation, may readily keep the eye-piece and the disc 5>in 
 line with the solar centre. My original design was that of 
 actuating the parallactic mechanism by clock-work ; but, 
 warned by the frequent failures of astronomers to keep the 
 sun accurately in focus even during the short period of an 
 eclipse, I adopted the safer method of operating by hand. 
 The distance between the centres of the discs /' and g' cor- 
 responding exactly with the distance between the axes of the 
 actinometers / and g, both being equidistant from the axis 
 of the eye-piece, it will be evident that the centres of the 
 discs /' and g' wlW always coincide with the axes of their 
 respective actinometers directed towards the- solar centre, pro- 
 vided the operator manipulates the instrument so carefully 
 that the sun is kept accurately in focus ; in other words, 
 that no distortion is suffered to take place of the annular 
 face or narrow border of the sun seen through the colored 
 glass. It hardly needs explanation that the actinometer h 
 is at all times exposed to the full energy of the converging 
 rays from the sun. 
 
 As a detailed account of the result of the investigation 
 would occupy too much space, the leading jjoiuts only will 
 be presented. The observations have all been made at 
 noon, the duration of the exposure to the sun having been 
 limited to seven minutes, during which period the actino- 
 meters are moved, by the parallactic mechanism, through a 
 
CUAP. VI. UNEQUAL RADIATION OF SOLAR 1)180. 119 
 
 distance of about 55 centimetres, from west to east. The 
 intensity of the radiant heat imparted to the actinometers 
 has been recorded by the observers at the termination of 
 the fourth, fifth, sixth, and seventh minute, the exact 
 moment for reading off being indicated by a chronograph. 
 The relative intensities transmitted by radiation from the 
 centre and from the border of the solar disc first claim 
 
 our attention. Fig. 6 re2:)resents the solar disc covered l)y 
 a circular screen 145.25 millimetres in diameter, excluding 
 the rays excepting from a narrow zone, the mean width of 
 which is situated 49" from the border of the photosphere. 
 Fig. 7 shows a screen excluding the solar rays excepting 
 from the central portion, the area of which is precisely eqval 
 to the area of the narrow zone in Fig. 6. The following 
 table shows the intensities transmitted to the actinometers 
 
130 BABIANT HUAT. chap. vi. 
 
 during an observatiou, August 25, 1875, the radiation from 
 the solar disc being then excluded in the manner shown in 
 Fis:s. 6 and 7. 
 
 ime. 
 
 Uentral portion. 
 Cent. 
 
 £>orner. 
 
 Cent. 
 
 Rate of difference. 
 2.19 
 
 4' 
 
 3°.28 
 
 2°.19 
 
 aas - »■<=''' 
 
 5' 
 
 3°. 56 
 
 2°.37 
 
 2.49 
 
 6' 
 
 3°. 73 
 
 2°.49 
 
 2.60 
 
 7' 
 
 3°.88 
 
 2°. 60 
 
 3.88 - '■'^^ 
 Mean = 0.667 
 
 It should be particularly observed that this table records 
 the result of four distinct observations ; nor should it be over- 
 looked that although the intensities vary greatly for each 
 observation, in consequence of the continued . exposure to the 
 sun, yet the rates shovdng the difference of the intensity of 
 the rays transmitted from the border, inserted in the last 
 column, is pi-actically the same for each observation, the dis- 
 crepancy betvreen the highest and the lowest rate being only 
 0.004. It should be mentioned that all my instruments for 
 measuring radiant heat referred to in this work have been 
 graduated to the Fahrenheit scale, which practically is more 
 exact than the Centigrade, owing to its finer divisions. For 
 the benefit of the majority of readers the observed tempera- 
 tures have been reduced to Centigrade scale before being 
 entered in our tables. Persons practically acquainted ^vith 
 
CHAP. VI. UNEQUAL BADIATION OF SOLAR DISC. 121 
 
 the difficulty of ascertaining tlie intensity of solar radiation 
 will be sui-prised at the exactness and consistency of the 
 indications of our instruments. This desirable exactness has 
 been attained by surrounding the actinometers with ^\•ater- 
 jackets, which communicate with each other by connecting 
 pipes, through which a steady stream of water is circulated. 
 By this expedient the chambers containing the bulbs of the 
 several thermometers are maintained with critical nicety at 
 equal temperature — an inexorable condition when the object 
 is to determine differential temperature with great exactness. 
 Apart from this, the chambers which contain the bulbs of 
 the thermometers are air-tight, the radiant heat being admitted 
 through a small aperture at the top of the chamber, covered 
 by a thin crystal. 
 
 Referring to the preceding table, it ■will be seen that the 
 intensity transmitted by radiation from the sun's border, repre- 
 sented in Fig. 6, is 0.667 of the intensity transmitted from 
 the central region represented in Fig. 7, the area of each being 
 precisely alike. From the stated intensity must be deducted 
 the heat imparted to the actinometer by the inflection of the 
 calorific rays. The circumference of the perforation of the 
 screen sho^vn in Fig. 7 being exactly one-half of the circum- 
 ference of the screen in Fig. 6, while the central region radi- 
 ates more powerfully than the border, fully one-half of the 
 inflected radiation from the border will be balanced by the 
 inflected radiation emanating fi-om the central region. Agree- 
 ably to the previous demonstration relating to Figs. 2 and 3, 
 it will be seen that the unbalanced inflection amounts to 
 
122 BABIANT HEAT. chap. VI. 
 
 0.029 ; hence tlie radiation transmitted from the border zone 
 will be 0.667 - 0.029 = 0.638 of the intensity of radiation 
 transmitted fi'om the central region. We have thus shown 
 by a reliable method that the intensity of the rays directed 
 toAvards the earth from the border zone suffers a diminution 
 of 1.000 - 0.638 = 0.362 of the intensity of the radiation 
 emanating from the central region. But the mean depth of 
 the solar atmosphere of the border zone, in the direction of 
 the earth, is 2.551 greater than the vertical depth, while the 
 mean depth over the central region referred to is only 0.036 
 greater than the vertical depth of the solar atmosphere. It 
 will be evident that if the law of retardation were known, 
 the foregoing figures would enable us to determine the absorj)- 
 tive power of the solar atmosphere. Concerning this law, it 
 should be mentioned that in the first edition of " Le Soleil," 
 page 264, the author assumes that the absorption of the calo- 
 rific rays by the atmosphere " augments in proportion to the 
 secant of the zenith distance " ; in other words, as the depth 
 of the atmosphere penetrated by the rays. Consequently, if 
 this assumption be correct, the absorption by the solar atmo- 
 
 0.362 „ , 
 
 sphere cannot exceed = 0.144 of the radumt 
 
 ^ 2.551 - 0.036 
 
 heat emanating from the photosphere. It will be found, on 
 
 referring to the revised edition of " Le Soleil," Vol. I., p. 212, 
 
 that P6re Secchi makes the following statements regarding 
 
 the absorptive power of the solar atmosphere : (1) " At the 
 
 centre of the disc — that is to say, perpendicularly to the 
 
 surface of the photosjihere — the absorption arrests about I, 
 
CHAP. VI. UNEQUAL liADIATIOX OF SOLAR DISC. 123 
 
 or, more exactly, tWt, of the total force." (2) "The total 
 action of the absorbing envelope on the hemisphere visible 
 from the sun is so great that it allows only iSV of the total 
 radiation to pass, the remainder, namely, n'oV, being absorbed." 
 It is unnecessary to criticise these figures presented by the 
 Eoman astronomer, as a cursory inspection of our table and 
 diagrams is sufficient to show the fallacy of his computa- 
 tions. Besides determining the absoi-ptive power of the solar 
 atmosphere, another impoi-tant problem may be solved by 
 accurately measuring the intensity of the radiation emanat- 
 ing fi'om various parts of the disc — namely, that relating to 
 the sun's emissive power in different directions. In order 
 to decide this question, I have adopted the plan of measur- 
 ing the energy of the radiant heat transmitted from zones 
 crossing the solar disc at right angles, as shown in Figs. 10 
 and 11. Repeated observations having shown that the acti- 
 nometers are equally affected by the radiation from these 
 zones, each of which occupies an arc of 30 deg., containing 
 one-third of the area of the disc, the inference is irresistible 
 that the sun emits heat of equal intensity in all directions. 
 It should be borne in mind that, agreeably to our method, 
 the radiations from these zones are observed simultaneously — 
 a fact tending to prove that our conclusions cannot be erro- 
 neous. The arrangement exhibited in Figs. 10 and 11 hardly 
 needs explanation. Eef erring to Fig. 10, it will be seen that 
 two segmental screens are employed excluding the radiant 
 heat, excepting from the zone, wliioh is inirallel with the 
 sun's equator. Similar screens are employed (see Fig. 11) 
 
124 BABIANT HEAT. chap. VI. 
 
 for excludiug tlie rays excepting from the zone j)ai"allel "with 
 the sun's polar axis. The curvatures of the segmental screens, 
 it shoiild be observed, have been struck to a radius of ninety- 
 millimetres, in order to cut off effectually the inflected radi- 
 ation from the sun's border. Obviously diffraction has not 
 called for any correction of our observations relating to this 
 part of the investigation, since the inflected radiation from 
 the equatorial zone exactly balances the inflected radiation 
 from the polar zone. As already stated, repeated observa- 
 tions show that the radiant energies transmitted to the acti- 
 nometers from the two zones are identical. 
 
 The observations relating to the temperature of the polar 
 regions, represented in Figs. 8 and 9, at first led to the sup- 
 position that the rays projected from the north pole of the 
 sun transmit a perceptibly greater energy to the actinometers 
 than the rays from the opposite pole. Subsequent observa- 
 tions having positively established the fact that the polar and 
 equatorial zones transmit equal intensities, it became evident 
 that some other cause than difference of temperature within 
 the polar regions influenced the actinometers. The only valid 
 reason that could be assigned in explanation of the anomaly 
 being the considerable angle subtended, and the consequent 
 difference of zenith distance of the opposite poles of the sun, 
 my table of maximum solar intensity for given zenith dis- 
 tances (prepared from data collected during a series of years) 
 was consulted, in order to ascertain the influence of zenith 
 distance. The observations indicating a higher temperature 
 at the north pole, it should be mentioned, had been made 
 
CHAP. VI. UNEQUAL HABIATIOX OF SOLAR DISC. 125 
 
 while the sun's zenith distance I'anged between 32 deg. and 
 33 deg. at noon. Now, the table i-eferred to shows that there 
 is a difference of radiant intensity of G3°.G3 — G3°.40 =: 0°.23 
 F. between the stated zeiiitli distances. Tlie mean angle sub- 
 tended by the sun being fully thirty-two minutes, it will thus 
 be seen that, owing to the absoi-ptive power of the terresti-ial 
 atmosphere, the radiant intensities transmitted from the oppo- 
 site poles of the luminary dift'ei' consideiably. The magni- 
 tude of this difference, adequate to explain the discrepancy 
 under consideration, need not excite surprise if we consider 
 that thirty-two minutes of zenith distance involves an addi- 
 tional depth of more than half a mile of atmosphere to be 
 penetrated by the rays projected towards the actinometer from 
 the south pole of the sun. The foregoing facts show the neces- 
 sity of taking the difference of zenith distance between the 
 opposite poles into account in making exact observations of 
 the sun's polar temjierature, especially at the lower altitudes, 
 where the secant of the zenith distance increases rapidly. 
 
 Regarding the calorific energy of the radiation emanating 
 from the border of the sun, the following brief statement 
 presents facts of considerable importance hitherto unknown. 
 Several observations during the early part of the investigation 
 pointed to the fact that increased energy is transmitted to 
 the actinometers by radiation from the sun's border. Again, 
 considerable irregularity was observed in the progressive dimi- 
 nution of the force of radiation towards the circumference of 
 the solar disc. It has already been shown that the radiation 
 from the border zone, 1' 42" wide, occupying one-fifth of the 
 
126 
 
 RADIANT HEAT. 
 
 area of the solar disc, transmits 0.638 of the intensity trans- 
 mitted from an equal area at the centre of the disc. Of 
 course it will be supposed that the rate of the diminution 
 of intensity within the zone thus ascertained is much greater 
 near the border of the photosphere than at the middle of 
 the zone. Such, however, is by no means the case, notwith- 
 standing the assumption of physicists that the heat transmitted 
 by radiation from the border is very feeble. In order to test 
 
 the truth of the indications referred to, sho\ving considerable 
 radiant energy at the border of the photosphere, a very careful 
 investigation was made, Sept. 9, 1875, by means of screens 
 excluding the rays from the solar disc, as shown in Figs. 12 
 and 13. The diameter of the screen represented in Fig. 12, 
 being 154.06 millimetres, covered nine-tenths of the area of 
 the disc ; while the screen shown in Fig. 18, being 145.25 
 millimetres, covered four-fifths of the disc. It will be well 
 to mention that the dimensions of the screens referred to 
 
CHAP. TI. UNEQUAL RADIATION OF SOLAR DISC. 127 
 
 correspond to the angle subtended by the sun Avben the 
 earth is in aphelion. Accordingly, the distance between the 
 actinometers and the screens was adjusted previous to the 
 obseiTation — that is, shortened — in order to comiiensate for 
 the increase of the angle subtended by the sun. Agreeably 
 to the stated dimensions of the screens, it will be found that 
 the zone represented in Fig. 13 is 1' 42", Avhile the zone in 
 Fig. 12 is 49".6. The mean width of the latter is conse- 
 quently situated only 2J:".3 from the border of the photo- 
 sphere. 
 
 The following table shows the intensities transmitted 
 to the actinometers from the zones represented in Figs. 12 
 and 13: 
 
 ^" Cent^' ^"' ^^ °^ Difference. 
 
 „ " 1.333 
 
 ''■''" 1:1 = '■''' 
 
 1°.583 ^^ = 0.652 
 
 2.425 
 
 7' 2°.485 1°.666 ]^ = 0.670 
 
 2.485 
 
 Mean = 0.660 
 
 The rate of diflference inserted in the last colmnn, it will 
 be noticed, is not qiiite so consistent as in the table recording 
 the observations made Aug. 25. The discrepancy is, however, 
 not material, the difference between the lowest and the mean 
 rate being 0.008. It will be seen, on inspecting the registered 
 intensities, that the border zone represented in Fig. 12, whose 
 
 Time. 
 
 Cent. 
 
 4' 
 
 2°.011 
 
 6' 
 
 2°. 248 
 
 6' 
 
 2°. 425 
 
128 'RADIANT HEAT. chap. vi. 
 
 area is only oue-lialf of tlie area of tlie zone in Fig. 13, trans- 
 mits 0.660 of tlie intensity of tlie latter. This, at fii'st sight, 
 indicates an extremely disproportionate transmission of heat 
 from the narrow border zone ; but it should be considered 
 that the inflected radiation imparts relatively more heat to 
 the actinometer exposed to the radiation from the narrow 
 zone than from the wide zone. It will be readily under- 
 stood that, since the inflection of the calorific rays is 14".7, 
 the first-mentioned actinometer receives radiant heat from 
 14".7 + 49".6 = 64".3 ; while the actinometer exposed to the 
 radiation from the wide zone receives heat from 1' 42" + 14".7 
 = 116".7. Consequently, the radiant heat emanating from 
 
 64" 3 
 
 the narrow zone will be -— — 0.551 of that transmitted 
 
 116".7 
 
 from the wide zone, hence somewhat more than one-half. 
 Our investigation therefore proves that the radiant heat 
 transmitted from the narrow border zone represented in 
 Fig. 12 is 0.660 - 0.551 = 0.109 more intense than that 
 transmitted from the zone represented in Fig. 13, although 
 the mean distance of the latter is twice as far from the 
 border of tbe photosphere as the mean distance of the for- 
 mer. The singular fact thus revealed can only be accounted 
 for by supposing that internal radiation is not incompatible 
 -^^-ith the constitution of the photosphere, and by adopting 
 Lockyer's views expressed in the Senate House at Cambridge, 
 1871, that "the photosphere must be a something suspended 
 in the solar atmosphere." Let a h, Fig. 14, represent a sec- 
 tion of the " suspended " photosphere, and d c, g f, I'ays pro- 
 
CIIAP. VI. UNEQUAL RADIATION OF SOLAR DISC. 129 
 
 jected towards the eartli. Agreeably to the cuiulitious men- 
 tioned, and in view of the fact that the force of radiation from 
 incandescent gases presenting equal areas varies nearly as 
 their depth, we are Avarranted in concluding that, since the 
 depth (I d' is greater than g g', the radiant heat transmitted 
 from the photosphere by the ray d c Avill be greater than 
 that transmitted by the ray g f. It should be observed that 
 the energy transmitted towards the earth by d c suffers a 
 greater diminution than the energy transmitted by g f, in 
 consequence of the greater depth of the solar atmosphere 
 penetrated. Hence the augmented energy estal)lished by our 
 
 investigation does not show the full amount of the increase 
 of radiant heat transmitted from the border of the sun. 
 
 Having thus briefly stated the result of my observations, 
 it \n\\ be proper to mention that, before undertaking a sys- 
 tematic investigation of the difference of theraial energy trans- 
 mitted to the earth by radiation from different parts of the 
 solar sui-face, I examined thoroughly the merits of Laplace's 
 famous demonstration relating to the absorptive power of the 
 sun's atmosphere, proving that only t^ of the energy deve- 
 loped by the sun is transmitted to the earth. The demon- 
 stration being based on the assumption that the sun's rays 
 
130 
 
 BABIANT HEAT. 
 
 emit energy of eqiial intensity in all directions, my iuitiary 
 step was tLat of testing practically the trutli of that propo- 
 sition. It lias been asserted that Laplace did not propound 
 the singular doctrine involved in such a proposition ; I 
 therefore feel called upon, before proving its unsoundness, 
 to quote the words employed by the celebrated mathemati- 
 cian (see " M^canique Celeste," Tom. IV. page 284). Having 
 called attention to the fact that any portion of the solar 
 
 disc, as it approaches the limb, ought to appear more bril- 
 liant because it is viewed under a less angle, Laplace adds : 
 " Car il est naturel de penser que chaque point de la sur- 
 face du soleil renvoie une lumiere ^gale dans tous les sens." 
 Let a h c d in the above diagram. Fig. 18, represent pai-t of 
 the border of the sun, and h «, c d, small equal arcs ; a a', 
 h I)', c c', d d', being parallel rays projected towards the earth. 
 Laplace's theory asserts that, owing to the concentration of 
 the rays, the radiation emanating from the portion d c trans- 
 mits greatei' intensity towards the earth than h a, in the pro- 
 
CHAP. VI. UNEQUAL RADIATION OF SOLAR DISC. 131 
 
 poilion oi c d io f c. The proposition is thus stated in 
 " Mecanique Celeste " : " Call d the arc of a great circle of 
 the sun's surface, included between the luminous point and 
 the centi'e of the sun's disc, the sun's radius being taken fur 
 unity ; a very small poi-tiou, a, of the surface being removed 
 to the distance 6 from the centre of the disc, will appear to 
 be reduced to the space a cos. d; the intensity of its light 
 must therefore be increased in the ratio of unity to cos. 0." 
 
 In order to disprove the correctness of the stated demon- 
 stration, I have measui-ed the relative thermal energy of rays 
 projected in different directions from an incandescent metallic 
 disc by the method minutely described in Chap. XL The 
 follo\ving brief description will, however, be necessary in this 
 place, reference being had to the illustration adverted to 
 in the said Chapter XI. (see Plate 21). Fig. 2 represents a 
 section of a conical vessel covered by a movable semi-spherical 
 top, the vessel being siirrounded by a jacket through which 
 water may be circulated. A revolving disc a a, composed 
 of cast iron, the back being semi-spherical and protected l)y 
 fire-clay, is suspended across the top of the conical vessel, 
 supported by horizontal journals attached at opposite sides. 
 The angular position of the disc is regulated by a radial 
 handle b, connected to one of the journals, the exact incli- 
 nation to the vertical line being ascertained by means of a 
 graduated quadrant d. An instrument c, capable of indicat- 
 ing the intensity of the radiant heat transmitted by the incan- 
 descent disc, is applied at the bottom of the conical vessel. 
 The movable cover and its lining of fire-clay being removed, 
 
132 BADIANT HEAT. CHAP. vi. 
 
 the cast-iron disc is lieated in an air-furnace to a tempera- 
 ture of 1,800° F. It is then removed by appropriate tongs, 
 and susj)ended over the conical vessel, the lining and cover 
 being quickly replaced. The temperature shown by the instru- 
 ment at the bottom of the conical vessel, resulting from the 
 action of the radiant heat of the disc, is then recorded for 
 every tenth degree of inclination. The investigation, it may 
 be briefly stated, shows that the temperature imparted by 
 radiation to the recording instrument is exactly as the sines 
 of the angles of inclination of the disc. Hence, at an incli- 
 nation of 10 deg. to the vertical line, the temperature imparted 
 to the thermometer is scarcely i of that imparted when the 
 disc faces the thermometer at right angles ; yet in both cases 
 an equal amount of surface of an equal degree of incandes- 
 cence is radiating towards the instrument. Laplace and his 
 followers have evidently overlooked this important and some- 
 what anomalous fact proving that radiation emanating from 
 heated bodies is incapable of exerting full energy in more 
 than one direction. Our practical experiments -ttath the 
 revolving incandescent disc have thus fully demonstrated 
 the truth of the proposition intended to be established — 
 namely, that the rays emanating from incandescent planes 
 do not transmit heat of equal intensity in all directions, the 
 energy transmitted being, as stated, proportionate to the sines 
 of their angle of inclination to the radiating surface. 
 
 The next step in the preliminary investigation was that 
 of measuring the radiant energy transmitted in a given direc- 
 tion by an incandescent solid metallic sphere. For this pur- 
 
CHAP. Ti. USEQUAL RADIATION OF SOLAR DISC. 133 
 
 pose I employed a double conical vessel similar to the one 
 already described, the incandescent sphere being suspended 
 over the conical vessel in the manner minutely described in 
 Chap. XII. A brief explanation will, however, be necessary 
 here, reference being had to the diagram on Plate 24, repre- 
 senting four spheres. Figs. 3, 4, 5, and G. Each sphere is divided 
 into four zones, A, B, C, and D, occupying unequal arcs, but 
 containing equal convex areas. Semi-spherical screens, com- 
 posed of non-conducting substances, were ajiplied below each 
 sphere, provided with annular openings arranged as shown 
 in the diagram. Through these annular openings the radiant 
 heat from the incandescent zones D, C, B, and A was trans- 
 mitted to the thermometers /, g, Ji, and k respectively. Pere 
 Secchi and other followers of Laplace will be siirprised to 
 learn that when the suspended sphere was maintained at a 
 temperature of 1,800° R, the radiation fi-om the zone C, Fig. 
 4, imparted a temperature of 27°.49 F. to the thermometer 
 g ; while the radiation from the zone A, Fig. 6, imparted 
 only 6°.19 F. to the thermometer l\ Let iis bear in luind 
 that the radiating surface I m oi the zone A is equal to the 
 radiating surface p q oi the zone C. The stated great dif- 
 ference of temperature produced by the radiation from zones 
 of equal area funiishes additional proof that Laplace based 
 his remarkable analysis on false premises. " The sun's disc 
 ought to appear more brilliant towards the border because 
 viewed imder a less angle," we are told by the great analyst. 
 The instituted pi-actical tests, however, prove positivel}- that 
 the energy of the rays projected from the border of an incan- 
 
134 BADIAWI EEAT. chap. vi. 
 
 descent sphere is greatly diminished because viewed under a 
 less angle from the point occupied by the recording thermo- 
 meter. 
 
 The result of the experiment with the revolving incan- 
 descent disc shows that if the small arc h a, in Fig. 18, be 
 reduced until the field represented by h' a' becomes equal 
 to the field represented by c' d', the radiant energy trans- 
 mitted through each of those fields will be alike ; the reason 
 being that the number of rays of diminished intensity passing 
 through c' d' will be as much greater than the number of 
 rays of maximum intensity passing through h' a' as c d is 
 greater than the reduced h a — f c. It should be observed 
 that c 6? is so small that we may, without appreciable error, 
 regard it as a straight line, and / c as the sine of the angle 
 c d f. It follows from this demonstration that if the solar 
 atmosphere exerted no retarding influence, the radiant heat 
 transmitted towards the earth would be alike for equal areas 
 of the solar disc ; more correctly, for areas subtending equal 
 angles, since the receding part of the solar sui-face is at a 
 greater distance from the earth than the central part. 
 
 Encouraged by the results of the instituted practical tests 
 showing the actual intensity transmitted by radiant heat ema- 
 nating from incandescent spheres and inclined discs, I devised 
 the method before described, proving positively that the polar 
 and equatorial regions of the solar disc transmit radiant heat 
 of equal intensity to the earth, and that the sun emits heat 
 of equal energy in all directions. Accepting Secchi's doctrine 
 relating to the retardation suffered by calorific rays in passing 
 
CHAP. VI. UNEQUAL RABIATIOX OF SOLAR DISO. 135 
 
 through atmospheres — namely, that the diminution of energy 
 is as the depth penetrated b}- the rays — I liave showii, by the 
 easy calcuhxtion before presented, that the absorption by the 
 sohir atmosphere cannot exceed Vo*uV of the radiant energy 
 emanating from the photosphere. 
 
 Concerning the plan resorted to by the Director of the 
 Roman Observatory and others of investigating the sun's 
 image instead of adopting the method of direct observations, 
 I will merely observe, in addition to what has already been 
 stated, that the information contained in the several works 
 of the Roman astronomer furnishes the best possible guide 
 in judging of the efficacy of image-investigation. Let us select 
 his account of the investigations conducted between the 19th 
 and 23d of March, 1852. Having pointed out that in these 
 experiments it was impossible to approach within a minute 
 of the edge of the sun, and that during a later observation 
 — date not mentioned — he had approached ^vithin a minute, 
 the investigator obseives : " But at this extreme limit, even 
 making use of the most accurate means of observation, we 
 find difficulties Avhicli it is impossible to overcome com- 
 pletel)'." In addition to this emphatic expression regarding 
 the difficulties encountered, the author adds : " Moreover, it 
 is impossible to study the edge alone, for the unavoidable 
 motions of the image do not admit of its being retained at 
 exactly the same point of the pile ; we have, therefore, been 
 unable to push the exactness as far as we hoped, and we 
 have discontinued the pursuit of these researches, although 
 the results obtained are quite interesting." (See revised edi- 
 
136 BABIANT HEAT. CHAP. vi. 
 
 tion of " Le Soleil," Vol. I. p. 205.) It is needless to iustitute 
 a comparisou between a system of Avhicli its founder speaks 
 so despondiugly and one wliicli enables us to push our inves- 
 tigations to the extreme limit of the solar disc, admitting 
 of entire zones being viewed at once, instead of only small, 
 isolated spots. 
 
 The foregoing demonstration, showing that the solar atmo- 
 sphere absorbs 0.144 of the heat radiated, it should be remem- 
 bered, is based on the assumption that the retardation is as 
 the depth penetrated by the rays. In view of the fact that 
 projectile force diminishes inversely as the square of the 
 depth of the medium penetrated, we are, of course, not com- 
 pelled to accept the stated assumption. Adopting the dyna- 
 mical law relating to projectile motion, referred to, it will 
 be found that the retardation, instead of being 0.144, will be 
 ( 3 55i-'oo3ii)^ ~ 0.057. But even this apparently small amount 
 of absorption of the radiant intensity cannot be satisfactorily 
 accounted for, since the mechanical energy developed by the 
 sun is at least 322,000 thermal units per minute upon an area 
 of one square foot. Consequently, 322,000 X 0.057 X 772 = 
 14,169,288 foot-pounds represent the mechanical equivalent of 
 tU-o of the radiant energy emanating from each square foot of 
 the photosphere. Owing to the high temperature and conse- 
 quent lightness, no conceivable work performed within the 
 solar atmosphere can satisfactorily account for the disappear- 
 ance, every minute, of such an amount of energy. It is there- 
 fore demonstrable that we have not underrated the absorptive 
 power. 
 
CHAPTER VII. 
 
 THE SOURCE OF SOLAR ENERGY. 
 
 Sir AVilliaji Thomson, iu his celebrated paper on the 
 Mechanical Energies of the Solar System, read before the 
 Royal Society of Edinburgh, April, 1854, puts the question : 
 "AVhat is the soiu'ce of mechanical eueigy, drawn upon Ijy 
 the sun, in emitting heat, to be dissipated in space ? " Having 
 very briefly examined the question, he adds : " We see, then, 
 that all theories which have yet been proposed, as well as 
 every conceivable theory, must be one or other, or a combi- 
 nation, of the following three: (1) That the sun is a heated 
 body losing heat. (2) That the heat emitted from the sun 
 is due to clieniical action among materials originally belong- 
 ing to the mass, or that the sun is a great fire. (3) That 
 meteoi-s falling into the sun give rise to the heat which he 
 emits." The second and third of these suppositions having 
 been disposed of l)y more recent investigations, let us con- 
 fine our discussion to the first hypothesis, that the sun is a 
 heated body losing heat, regarding which the eminent phy- 
 sicist remarks: "In alluding to theories of solar heat in a 
 foiTuer communication to the Ro\al Society I pointed out 
 
 137 
 
138 BADIANT HEAT. chap. vii. 
 
 tliat the first hypothesis is quite untenable. In fact, it is 
 ueuionstrable that, unless the sun be of matter inconceivably 
 more conductive for heat, and less volatile, than any terres- 
 trial meteoric matter we knoAV, he would become dark in 
 two or tliree minutes, or days, or months, or years, at his 
 present rate of emission, if he had no source of energy to 
 draw from but primitive heat. This assertion is founded on 
 the supposition that conduction is the only means by which 
 heat could reach the sun's surface from the interior, and 
 perhaps recpiires limitation. For it might be supposed that, 
 as the sun is no doubt a melted mass, the brightness of his 
 surface is constantly refreshed by incandescent fluid rushing 
 from below to take the place of matter falling upon the 
 surface after becoming somewhat cooled and consequently 
 denser — a process which might go on for many years with- 
 out any sensible loss of brightness. If we consider, however, 
 the whole annular emission at the present actual rate, we 
 find, even if the sun's thermal capacity were as great as 
 that of an equal mass of water, that his mean temperature 
 would be lowered by about 3° Cent, in two years." It -would 
 appear from this reasoning that Sir William Thomson had 
 overlooked Laplace's important nebular theory which leads 
 to the conclusion that the sun's heat is the result of con- 
 densation caused by gravitation. Obviously this condensa- 
 tion is progressing at the present time as fast as the mass 
 cools, the process being the same now as millions of years ago, 
 the heat generated by the condensation becoming gradually 
 intensified in the inverse ratio of the radius of the contract- 
 
CHAP. VII. Till-: aorucK of holai; enei;gy. inn 
 
 ing mass. Let us consider, liowever, that the matter com- 
 posing this contracting mass siitfei-s a proportionate reduction 
 of velocity ; hence the emission of heat becomes constant 
 for a given area of the soLir suiface, notwithstanding the 
 augmentation of intensity. But, wliile the emission from a 
 given area is constant, the bulk of the cooling mass is con- 
 tinually diminishing. The important consequences of this 
 diminution of bulk will l^e otnisidered presently. Of course 
 it will not be necessary to prove that the sun is continu- 
 ally becoming smaller, since all incandescent bodies shrink 
 rapidly if permitted to radiate freely, the rate being nearly 
 proportional to the degree of incandescence. Our task, there- 
 fore, will be confined to a simple demonstration showing 
 that the emission of 322,000 thermal units per minute on 
 each squai-e foot of solar surface, established in Chaj)ter V., 
 is capable of being developed by the contraction of the mass, 
 in accordance "with the nebular theory of Laplace. At first 
 sight it would appear that no probable amount of contrac- 
 tion of the solar mass could develop, by gravitation towards 
 the centre, an amount of mechanical energy of 322,000 X 
 772 = 248,584,000 foot-pounds per minvte for each square 
 foot of the surface of the sun. Yet so A-ast is the amount 
 of matter covered by the insignificant area of 144 square 
 inches of the solar surface — in other words, such is the con- 
 tents of a spherical pyramid the base of which is one square 
 foot, and whose length is equal to the sun's radius — that a 
 veiy small amount of contraction suffices to develop, l>y gra- 
 vitation towards the solar centre, the stated enormous mecha- 
 
140 BADIANT HEAT. chap. vii. 
 
 uical enei'gy. It will be readily understood that the energy 
 developed by the shrinking of a spherical pyramid whose sides 
 are sectors of the great circle of the sun M'ill represent cor- 
 rectly the relative energy produced by the shrinking of the 
 entire solar mass. Hence, if Ave can determine the amount 
 of longitudinal contraction of the supposed spherical pyramid 
 requisite to produce, by gravitation tow^ai'ds the centre of the 
 mass, a mechanical energy of 248,584,000 foot-pounds per 
 minute, we need not enter into any further computation, since 
 a corresponding contraction of the sun's radius will develop 
 for every sqxiare foot of his surface a like energy. 
 
 Let I K S, Fig. 1 (see Plate 14), represent the great 
 circle of the sun, a m a' the spherical pyramid referred to, 
 and Fig. 2 the said pyramid drawn to a larger scale, its axis 
 being divided into ten equal parts. It is proposed to ascer- 
 tain what extent of longitudinal contraction of the spherical 
 pyramid a m a' is necessary to produce an amount of djaiamic 
 energy corresponding with that developed by the radiation 
 from 1 sq. ft. of the solar surface in a given time. The in- 
 vestigation will be facilitated and more readily comprehended 
 if we compute the amount of energy developed by a definite 
 contraction of the sun's radius, say one foot. Let us, there- 
 fore, suppose that the surface a a', the distance of which is 
 
 '- X 5,280 = 2,250,821,760 ft. from m, has fallen through 
 
 a space of one foot, the intermediate points b, c, d, etc., par- 
 ticipating proportiouably in the fall. Assuming that the solar 
 mass remains homogeneous during the contraction, it follows 
 
CHAP. Til. THE SOURCE OF SOLAU EXERGT. 141 
 
 from Newton's demonstration (" Principia," Lib. I. Prop. 
 LXXIII.) that, since a particle just within the circumference 
 of the sphere at a is ten times further from the centre m 
 than a partick- at /, the foniicr will be attracted towards ni 
 with ten times greater force than the latter. It will be per- 
 ceived, on reflection, that, for a given movement towards the 
 centre, the quantity of matter put in motion at a will be 
 greater than at J, in the ratio of the squares of a a', and / 
 1, or 100 ; 1. Hence, in accordance with the demonstration 
 referred to, a given radial depth of the solar mass at a will 
 exert a force towards w 10 X 100 = 1,000 times greater than 
 an equal radial depth at I. But in computing the dynamic 
 energy developed by the shrinking of the sun, it must be 
 borne in mind that a particle at a falls through a distance 
 ten times greater than a particle at I. The length of the 
 onlinates of the curve j) (, Fig. 3, representing the ratio of 
 dynamic energy developed at the respective distances from 
 the sun's centre, has been calculated accordinglj'. A cui-sory 
 examination of Fig. 2 can scarcely fail to lead to the suppo- 
 sition that the mass composing the smaller sections of the 
 sjiherical i)}ramid near the centre of the sphere Avill be 
 attracted by the larger mass composing the sections near 
 the circumference. Newton has disposed of this question 
 by a geometrical demonstration which, considering the foiiii 
 t)f the attracting mass, and the e.xtreme complication arising 
 from the varying direction and unequal magnitude of the 
 attracting forces, may be I'egarded as one of the most ele- 
 gant of his masteily demonstrations of inqxirtant -propositions 
 
143 RADTANT HEAT. chap. Tii. 
 
 and tlieorems. Unless it can be proxed that a jiarticle at P 
 is not attracted by any portion of the mass contained within 
 the external spherical superficies IKS and the interior sphe- 
 rical superficies P j^, we must assume that the mass compos- 
 ing the sections near the base of the spherical pyramid will 
 exert the disturbing attraction before alluded to. Hence 
 our demonstration of the enei'gy produced by the attraction 
 of the matter within the sun, during shrinking, falls to the 
 ground unless it can be shown that every particle compos- 
 ing the spherical pyramid is in perfect repose as regards the 
 attraction exerted by exterior particles. The great geometer 
 thus establishes that repose : " Let II I K L be a spherical 
 superficies and P a corpuscle placed -within. Through P let 
 there be drawn to this superficies the t\vo lines II K, I L, 
 intersecting very small arcs II I, K L ; and because the tri- 
 angles H P I, L P K are homogeneous, those arcs will be 
 proportional to the distances H P, L P, and any particles 
 at H I and K L of the spherical superficies, terminated by 
 right lines passing through P, will be in duplicate ratio of 
 those distances. Therefore the forces of these particles ex- 
 erted upon the body P are equal between themselves. For 
 the forces are as the particles directly, and the squares of the 
 distances inversely. And these two ratios compose the ratio 
 of ecpiality. The attractions, therefore, being made equally 
 towards contrary parts, destroy each other ; and, by a like rea- 
 soning, all the attractions through the whole spherical super- 
 ficies are destroyed by contrary attractions. Therefore the 
 body P will not be any way impelled by those attractions." 
 
cUAi>. VII. THE SOVUCE OF SOLAU ESEEGY. Ul! 
 
 Sir Isaac Newton, iu his demoustratious relatiug to sphe- 
 rical bodies, supposed these to be composed of an infinite 
 number of spherical superficies, the thickness of which he 
 thus defines: "By the superficies of whlcli I here imagine 
 the solids composed, I do not mean superficies purely mathe- 
 matical, but orbs so extremely thin that their thickness is as 
 nothing ; that is, the evanescent orbs of which the sphere will 
 at last consist when the number of the orbs is increased, and 
 their thickness diminished without end." 
 
 Referring to Fig. 3, it should be p;irticularly observed 
 that the ordinates of the curve 2) t do not indicate the force 
 exerted by mere attraction. As already stated, their length 
 represeuts the dynamic energy developed at the indicated 
 distances from the solar centre. Consequently, the energy 
 actually developed by the shrinking of the mass is repre- 
 sented by the superficies o p t, while the rectangle o p u t 
 represents the energy that would be called forth if the force 
 exei-ted at every point of the axis of the spherical pyramid 
 were the same as that exerted at a a'. Having already pointed 
 out the manner of determining the length of the ordinates of 
 the curve p> t, it will suffice to state that their mean length 
 is 0.20015 oi p ; hence the superficies o p t \& 0,20015 of 
 the superficies o p u t. 
 
 Let us no^v consider whether the want of homogeneity 
 of the solar mass will materially affect the calculated amount 
 of energy developed by the gravitating force during the sun's 
 shrinking. Referring to the diagram Fig. 3, it will be seen 
 that the energy exerted by a given small amount of contj-actiou 
 
144 EADIANT MEAT. chap. vii. 
 
 at a section equidi.stant between ni and a — viz., at /' 5 — will 
 
 625 
 be, as shown by the length of the ordinates, = iV of 
 
 that exerted by a like amount of contraction at a a' ; and 
 that, since m f is one-half of m a, the energy developed by 
 the contraction of the mass contained within the spherical 
 pyramid / in 5 amounts to only -h of that developed by the 
 contraction of the mass contained in the spherical pyramid 
 a m a'. Now, the volume of the spherical pyramid / m 5 
 represents that of a sphere the diameter of which is one-half 
 of the sun, while the spherical pyramid (/ m a' represents 
 the volume of the entire solar mass. The energy resulting 
 from gravitation during the contraction of the central sphe- 
 rical mass P p being thus only ^V of the energy resulting 
 from gravitation during the contraction of the spherical mass 
 I K S, it will be perceived that the degree of density of the 
 matter near the sun's centre will not materially affect the 
 result of our calculations founded on j^ei'fect homogeneity. 
 
 We may now proceed to ascertain the amount of dynamic 
 energy produced by the assumed contraction of one foot of 
 the axis of the sj^herical pyramid a m a'. Having already 
 demonstrated that the said energy will be 0.20015 of that 
 produced by the gravitation of a homogeneous mass, the sec- 
 tion of which is one square foot, extending from a a' to m, 
 it only remains to determine the weight of one cubic foot at 
 the surface of the siin. The weight of the solar mass being 
 85.6 lbs. per cubic foot, while the sun's attraction is 27.2 times 
 greater than terrestrial attraction, the weight of one cubic 
 
cuAP. VII. TUi: sorncE of solar EXEHOY. 145 
 
 foot of the solar sui-face will be 27.2 X 85.() = 2,328.3 lbs. ; 
 multipl}nng this weight by the sun's radius, expressed in feet, 
 we have 2,328.3 X 2,250,821,000 = 5,240,586,000,000, which 
 product nniltiplit'd by 0.20015 shows that the gravitating 
 energy of the matter contained in the spherical pyramid, ex- 
 erted during a longitudinal contraction of one foot, amounts 
 to 1,048,900,000,000 foot-pounds. Dividing this latter product 
 by the ascertained solar emission of 248,584,000 foot-pounds 
 per minute, it will be seen that the mechanical energy pro- 
 duced by the shrinking of one foot of the sun's radius is suf- 
 ficient to make good the power lost by solar emission during 
 a period of 4,219.5 minutes. If we then divide- this quutieiit 
 in the minutes in a year, 525,960, it will be found that a fall 
 of 124.65 feet of the solar surface per annum must take place 
 in order to sustain the present emission of heat. At this rate 
 of shrinking the diameter of the sun will be reduced Tshm in 
 the coui-se of 1,805 years. It has already l)een observed that 
 the intensity of the radiant heat will not diminish with the 
 diminished size of the sun. On the contraiy, for a given 
 area of the solar surtace, the dynamic energy produced by 
 a triven rate of shrinking will be increased, since the ma.ss 
 remains the same, while the attraction is invei-sely propor- 
 tional to the distance from the centre. Rut the rate will 
 diminish w ith the contraction of the sphere ; hence a shrink- 
 ing of f'oth of the sun's diameter, instead of occupying 1,000 
 X 1,805 = 1,805,000 yeai-s, will require somewhat more than 
 2,000,000 yeai-s. At the end of that period the gravitating 
 energy will continue to develop, as at present, an amount 
 
146 EADIANT HEAT. chap. vii. 
 
 of dynamic energy i-epresented by 322,000 tLennal units per 
 
 niiunte for eacli supeificial foot ; but tLe radiating sm-face 
 
 — i.e., the area of tlie solar disc — will liave diminished in 
 
 the ratio of nearly 10' to 9". 
 
 The present maxinuim temperature produced by solar 
 
 radiation on the ecliptic when the earth is in aphelion being 
 
 67.2 deg. (see Chap. III.), while the intensity of radiant heat 
 
 diminishes as the area of the radiating surface, it follows that 
 
 at the end of 2,000,000 years from the present time the tro- 
 
 9" X 67 2 
 pical solar intensity Avill be reduced to — = 54.4 deg. 
 
 Falirenheit. The result of the elaborate investigations of solar 
 intensity described in the preceding chapter proves the cor- 
 rectness of the foregoing calculations based on the a^'ea of the 
 solar disc, and disposes of the opinion held by some physi- 
 cists that thei'e is no established relation between the dia- 
 meter of the sun and the transmitted energy. It was found, 
 during the investigation referred to, that, in shutting out the 
 radiation from the external zones of the sun and exposing 
 an actinometer to the I'ays emanating from a circular ai'ea at 
 the centre measuring 380,000 miles in diameter, the intensity 
 of the radiant heat was reduced to one-third of that trans- 
 mitted to another actinometer exposed to the radiation from 
 the entire solar disc. Can we douT)t, tlieu, that the future 
 diminution of the diameter of the sun will cause a corre- 
 sponding diminution of the transmitted energy ? Adopting 
 the same mode of calculating the solar intensity for past ages 
 as the foregoing calculation of future solar intensity, it will 
 
CHAP. VII. THE SOURCE OF SOLAR ESEIiGY. 147 
 
 1)6 tV)mul tliat the teiii])erature proiluced Ijy sular radiation 
 :?,000,000 years ago (owing to tbe greater diameter of the sun 
 
 . 11' X 67.2 
 
 at that period) must have been nearly ; = 81 des:. 
 
 I / ''10 ° 
 
 l''ali. within the tropics. Concerning this^great intensity of 
 the radiant heat, and the consequent high atmospheric tempe- 
 rature, we are justified in assuming that increased evaporation 
 of the sea, and coTTPsponding humidity of the atmosphere, mol- 
 lified the apparently destructive temperature, calling forth the 
 luxuriant flora ^\•hich geology has made us acquainted with. 
 The computed diminution of solar intensity, 67 deg. — 54 deg. 
 = 13 deg., during the next 2,000,000 years, will be deemed 
 extravagant by those who do not bear in mind that we must 
 base our computation on the assumption that a contimiouts 
 power will be exerted during the stated period capable of 
 developing, as at present, the stupendous energy of 248 mil- 
 lions of foot-pounds in a single minute for each square foot 
 of the surface of a sphere whose diameter exceeds 850,000 
 miles. Persons speculating on the cause of solar energy will 
 do well to consider that this inconceivable amount of work 
 cannot be pei-fonned with a less expenditure than the motive 
 energy developed by the fall of a mass equal to the mass 
 contained in the sun. But a continiioits development of such 
 an amount of energy is obviously impossible, since the dis- 
 tance is limited through Avhich the mass can fall. Now, the 
 foregoing demonstration enables us to determine the said limit 
 \ni\\ sufiicient exactness to prove that, although the efficiency 
 of the great motor during the past may be measured by lain- 
 
148 RADIANT MEAT. chap. vii. 
 
 dreds of niillious of yeurs, its future efficiency will be of com- 
 paratively brief duration. 
 
 Statements frequently made relating to the permanency of 
 solar heat, based on the assertion that no diminution has been 
 observed during historic times, have no weight in view of our 
 demonstration showing that a shrinking of A of the sun's dia- 
 meter can only reduce the intensity from 81 deg. to 67.2 deg., 
 difference = 13.8 deg., in the course of two millions of years. 
 This period being 500 times longer than "historic times," it 
 will be seen that the diminution of the temperature produced 
 
 13 8 
 by solar radiation has not exceeded — rr = 0.027, or tV deg. 
 ■' 500 
 
 Fah., since the erection of the Pyramids. 
 
 It will be proper to notice, before concluding our brief 
 investigation of the source of solar energy, that the develop- 
 ment of heat by the shrinking of the sun, however fully 
 demonstrated, leaves the important question unanswered : 
 How is the heat generated by gravitation within the mass 
 transmitted to the surface ? If the matter within the sun 
 is a perfect conductor of heat — a very improbable supposi- 
 tion — that fact alone furnishes a satisfactory answer. Imper- 
 fect conductivity, on the other hand, calls for other means 
 of transmitting the energy from within to the surface. What 
 those means 'are presents a problem suscejitible of positive 
 demonstration. The api)lication of cold at the surface of 
 any heated gaseous fluid, or the reduction of temperature of 
 siich a fluid by radiation upwards, invariably produces a ver- 
 tical circulation within the heated mass, the particles cooled 
 
CHAP. VII. TtlE auLltVE OF HOLAR ESKViGl'. 149 
 
 ilesceiuliug, iu virtue of tlieir incieased sipecific gravity. Evi- 
 ileiitly a uiimber i)f particles desceudiug oue after the other 
 will produce a downward vertical current of greater specific 
 gravity than the rest of the fluid. Now, as this current, com- 
 posed of comparatively cold and heavy particles, descends, 
 it displaces a corresponding bulk of heated fluid, which, since 
 there is no unoccupied space below, must rise to the surface. 
 Descending and ascending currents of nearly uniform magni- 
 tude and velocity will thus be established in the heated fluid 
 mass, provided no disturbing force be applied, causing ai;ita- 
 tion either at the suiface or within. It needs no proof to 
 show that in case disturbing force be applied the refularitN 
 of the distribution of descending and ascending currents 
 within the fluid ceases, and that the established vertical 
 circulation, instead of being alike at all points, becomes 
 divided into groups. Nor is it necessary that the distuib- 
 ing force should be of great magnitude. Obviously a con- 
 tinued uniform distribution of the descending and ascending' 
 currents through the heated mass calls for perfect aljsence 
 of disturbing influences of any kind. Xow, within tlie sun 
 the descending and ascending masses of heated matter are 
 influenced by numerous disturbing causes. (1) The particles 
 composing the descending currents, possessing the vis viva 
 due to the angular velocity of the sun's surface, gradually 
 encounter the particles of less angular velocity composing 
 the ascending columns. Conflicting motions will thus be 
 produced, resulting in an inci-ea-sed angular velocity of the 
 interior of the solar mass ; w hile {^2) the particles of the 
 
150 EADIANT HEAT. chap. vn. 
 
 asceudiug cuiTeiitj<, wliicli start witli a slow angular velo- 
 city, will be gradually impelled by tlie surrounding mass 
 during their ascent ; tlie energy thereby absorbed causing a 
 lagging of the entire solar mass towards the surface. Of 
 course there is an exchange of angular vis viva between 
 the descending and ascending currents, but obviouslj^ there 
 will be a loss, productive of perceptible lagging at the sur- 
 face of the rotating mass. (3) The attraction of the plane- 
 tary masses ^\■ill seriously disturb the vertical circulation Ijy 
 alternately impelling and retarding one or the other of the 
 descending and ascending columns of heated matter, thus 
 occasioning great irregularity. (4) The periodic change of 
 position of the centre of gravity of the aggregate planetaiy 
 mass must necessarily produce a periodic maximum and 
 minimum disturbance of the descending and ascending cur- 
 rents ^vithIn the solar mass. It will be evident that the 
 frerpient near approach and consequent powerful attraction 
 of Venus greatly complicates the question of maximum dis- 
 turbance. (5) The rotation of the sun, it should be parti- 
 cularly observed, tends to mollify the disturbing influence of 
 planetary attraction on the vertical circulation, since, owing 
 to this rotation, the descending and ascending motion of the 
 heated matter within the solar mass is successively relieved 
 from maximum disturbing iufluence twice in twenty-five 
 days. It may be sho^vn that, but for this frequent check, 
 the power exercised by planetary attraction would aug- 
 ment, and fatally derange the internal circulation indispen- 
 sable to the regular performance of the functions assigned 
 
CHAP. vii. THE SOURCE OF SOLATi ESEliGT. 151 
 
 to the sun. (0) It neetl.s no cxiilmiation that a continuous 
 disturbance and consequent cessation of circuhition at certain 
 jioints of the solar .surface would produce permanent dark 
 spots. Bearing in mind the enormous amount of the regular 
 emission of heat demonstrated in Chap. V. (;522,000 thermal 
 units per minute on a square foot of the solar surface), it 
 becomes evident that any considerable diminution of the 
 ^"PPb' o^ energy from within, consequent on deranged cir- 
 culation, will at once produce a great fall of temperature at 
 the surface of the photosphere, and a corresponding diminu- 
 tion of the temperature of the contiguous solar atmosphere, 
 accompanied by a sudden condensation and do\\n-rush ovei- 
 the regions of t>bstructed circulation. Considering the in- 
 creased Aveight of the condensed matter put in motion l)y 
 the sun's powei-ful attraction, we can readily imagine that 
 the photosphere, suspended over the solar mass, as supposed 
 by Lockyer, may be pierced by the descending column, and 
 an opening formed, exposing that part of the solar mass 
 which, for want of circulation and supply of heat from 
 within, has lost intensity and radiant power. It should be 
 observed that, although we have no knowledge of the con- 
 stitution of the photosphere, we may assert positively that 
 its radiant power is derived from the underlying solar mass, 
 and that, therefore, any diminution of energy of the lattci-, 
 occa.sioned by disturbed circulation, will at once diminish the 
 tenqierature and radiant power of that portion of the photo- 
 sphere which is situated above the obstruction. (7) The 
 descending and ascending columns of heated matter between 
 
152 EABIANT HEAT. chap. vii. 
 
 the solar centre and tlie poles, being acted upon almost at 
 right angles by planotarv attraction, remain at all times nearly 
 undisturbed ; hence only a few dark spots, of small size, form 
 in the polar regions. 
 
 Regai'diug the permanency of solar radiation, the forego- 
 ing explanations show that the system of veitioal circulation, 
 upon which depends the efficiency of the sun as a motor, 
 may become deranged. The consequence of this precarious 
 feature of the scheme is self-evident, if we consider that the 
 present solar emission is dependent on a given rate of contrac- 
 tion of the solar mass. Should that contraction be checked 
 by interrupted circulation, the development of heat will also 
 l)e checked, and, consequently, the intensity of solar radiation 
 become inadequate to sustain animal and vegetable life, as 
 now organized, on our planet. History informs us that the 
 luminary has at certain epochs partially failed to pei'foi'm 
 its functions. Hei-schel mentions, in his " Outlines of Astro- 
 nomy," that " in the annals of the year a.d. 536 the sun is 
 said to have suffered a great diminution of light, which con- 
 tinued fourteen months. From October a.d. 626 to the fol- 
 lowing June a defalcation of light to the extent of one-half 
 is recorded ; and in a.d. 1547, during three days, the sun is 
 said to Lave been so darkened that stars were seen in the 
 day-time." Again, the glacial periods, the ascertained abrupt 
 termination and recurrence of which puzzles the geologist, 
 point to periodical derangement of the solar mechanism iu 
 past ages. 
 
CHAPTER VIII. 
 
 RAl)lAriX(i FOWKli AXD DEI'l'lI OF THE SOLAE 
 ATMOSFllEKE. 
 
 Tin: illustration shoA^Ti on Plate 15 represents an instini- 
 nient constructed for the purpose of asccrtnininy the radiant 
 [lower of the sohir atmosphere, and for measuring its depth, 
 the leading feature i>f the device being that of shutting out 
 the rays h\m\ tlic pliutosjihere during the investigation. Evi- 
 dently the expedient of shutting out tlie photi'sphere while 
 examining the effect produced by the i'a}s emanating from 
 the solar envelope calls for means by which the sun may be 
 kept accurately in focus during the period i-equired to com- 
 jilete the observations. The main features of tlie instrument 
 being clearly shown by the illustration, a brief description 
 ■will be sufficient to explain its detail. A jiarabolic reilector, 
 applied for the purpose of concentrating the rays of the solar 
 atmosphere, is inserted in the cavity of a conical dish of cast 
 iron, secured to tlie top of a table suspended on two hori- 
 zontal journals, and revolving on a vertical axle. The latter, 
 slightly taper, turns in a oastirou socket, which is bushed 
 
 Hi 
 
154 HADIAST HEAT. CHAP. vill. 
 
 witli lii'ass and supj)orte(l l)y tlirt'c legs stepped on a trian- 
 gnlar l)ase, resting on fi'letion-rollers. The liori/.ontal jour- 
 nals referred to turn in bearings attached to a rigid Lar of 
 wrought iron situated under the table, fii-ndy secured to the 
 upper end of the vertical axle. The horizontal angular posi- 
 tion of the taltle is adjnsted by a screw operated by the 
 small hand-wheel a, the inclination being regulated by an- 
 other screw turned by the hand-wheel h. A gradiuxted quad- 
 rant, e, is attached to the end of the table in order to 
 afford means of ascertaining tlie sun's zenith distance at any 
 moment. The index cl, which marks the degree of inclina- 
 tion, is stationary, being secured to the rigid bar before 
 described. The rays from the photosphere are shnt out l)y 
 a circular disc _/', composed of sheet metal turned to exact 
 size, and supported by three diagonal rods of steel. These 
 rods are secured to the circumference of the conical dish by 
 screws and adjustal:)le nuts in such a manner that the centre 
 of the disc / may readily be brought in a direct line "with 
 the axis of the reflector. The mechanism adopted for adjust- 
 ing the position of the tal)le by the hand-wheels a and h 
 requires no explanation ; but the device which enables the 
 operator to ascertain when the axis of the reflector is pointed 
 exactly towards the centre of the sun demands particular 
 notice. A shallow cylindrical box (j, provided ^\ith a flat 
 lid and open at the bottom, excepting a narroAV flange extend- 
 ing rouud the circumference, is firmly held by two columns 
 secured to the top of the table. A convex lens of 26 ins. 
 focus is inserted in the (!}liudricul box, the narrow flange 
 
cuAi'. VIII. i;.\i>iATi\(! I'owKi: or sola I! ATMosriii:!;!-:. 155 
 
 meutioued affoidiug necessary supixirt. 'llir li<l is perfo- 
 rated by two opeuings at right angles, (iji.") in. wide, l\5 
 ins. long, forming a cross, tlie lens being so adjusted that 
 its axis passes tlirough the central point of intersection of 
 the cross. The face of the table being tnrned at right 
 angles to the sun, or nearly so, it will be evident that the 
 rays passing through the pei-forations and through the lens 
 will produce, at a certain distance, a brilliantly illuuiiuat.Ml 
 cross of small size and sharp outline. A piece of ivory, on 
 which parallel lines are drawn intersecting each other at 
 right angles, is attached to the top of the table in such a 
 position that the centre of intei-section of the said lines coin- 
 cides with the axis of the lens. This axis being parallel 
 with the line passing through the centre of the disc / and 
 the focus of the reflector, it Avill be perceived that the ope- 
 rator, in directing the table, has oidy to bring the illuminated 
 cross within the intersecting parallel lines on the piece of 
 ivory. Ample practice has sho^vn that by this arrangement 
 an attentive person can easily keep the disc /' accurately in 
 line with the focus of the reflector and the centre of the 
 sun during any desirable length of time. The absence of 
 any perceptible motion of the column of the f<x-al thermo- 
 meter employed during the e.xperiiiK iits furnishes the l^est 
 evidence that the sun's rays have been effectually shut out 
 by the intervening disc /; which, it should be remembered, 
 is only large enough to screen the aperture of the reflector 
 from the rays projected l»y the photosj.here. It is woithy 
 of observation that the lightness of the adopted mechanism 
 
156 RADIAST HEAT. CHAP. viii. 
 
 renders exact iuljustiueut easy, since screws of small diameter 
 and tine pitch may be employed. It is hardly necessary to 
 point out that the table represented by our illustration is 
 admirably adapted for actinometric observations, since, apart 
 from its perfect parallactic motion, it is provided with a cpiad- 
 rant and index sho\\ing the sun's zenith distance. 
 
 Fig. 2 (see Plate 16) represents a vertical section of the 
 parabolic reflector before adverted to, inserted in the cast-iron 
 dish attached to the parallactic table. This I'etiector consists 
 of a solid wrought-iron ring lined -with silver on the inside, 
 turned to exact form and highly polished. An annular plate 
 U.5 ins. internal diameter is secured to the top of the wrought- 
 iron ring, in order to jjrevent effectually all rays projected by 
 the photosphere from reaching the reflector. 
 
 The important cpiestion whether the solar envelope pos- 
 sesses an appreciable radiant power, and whether the high 
 temperature of the attenuated matter of which it is composed 
 exercises any marked influence on the sun's radiant energy, 
 may unquestionably be answered practically by means of the 
 instrument thus described. I have accordingly conducted an 
 investigation based on the expedient of concentrating the heat- 
 rays of the solar atmosphere by the parabolic reflector men- 
 tioned, in such a manner that only the heat-rays, if such there 
 be, from the chromosphere and exterior solar envelope are 
 reflected, while the rays from the photosphere are effectu- 
 ally shut out. Fig. 1 shows the general ai'rangemeut; /' a' 
 represents the exterior of the photosi^here, and (j' Ii the boun- 
 dary of the surrounding solar atmosphere ; 7t / is the circulai' 
 
cuAP. VIII. L'ADiATisa I'owKi; OF s<n..\i: .\'nrosrin:nr:. 157 
 
 metallic disc, supported above the parabolic reflectoi' before 
 described, and marked f in the ilbistration on Plate 15. This 
 disc is exactly !<• ins. in diaiiifter, placed 5.'^.7'> ins. above 
 the base line <^ o; the latter (•oincidini;- \\illi the top of the 
 parabolic reflector. The stated distance l)et\veen the disc 
 and the top of the reflector obvionsly varies considerably 
 uith the seasons. Assumincf that the investimition takes 
 place when the sun subtends an angle of 32 min. 1 sec, 
 and making proper allowance for diffraction, the disc k 7, if 
 placed 53. 7() inches from a o, will throw a shadow of fully 
 i*.5 ins. diametei' ; hence, if /' o be 9.5 inches, objects in the 
 plane </ c, placed within /' o, will ])e effectually shut out 
 from the rays projected by the photosidiere, while they will 
 be fully exposed to the rajs (/' g and 7/ //, emanating from 
 the chromosphere and outer strata of the solar enve]o[)e. It 
 is evident, therefore, that a parabolic reflector of ])ro[)er size 
 placed immediately below /' n will concentrate the radiant 
 heat, if any, transmitted by the rays f f and <j' ;/, and the 
 intermediate rays. It has already been stated that an an- 
 nular plate 9.5 ins. internal diameter is secured to the top 
 of the reflector to i)revent effectually any rays projected by 
 the photosphere from reaching the same. 'J'he prolongation 
 of the rays f f — (/' 'J :uid /( n — a' o are shown by dotted 
 lines /*, g and n, o in Fig. 2 ; also the reflected rays directed 
 towards the bulb of the focal thermometer, marked resjiec- 
 tively /\ o' and g\ it'. 
 
 The following brief account is deemed sufficient to ex])laiii 
 the mode of conductiiiir the investigation: TuriiiiiLr the reflec- 
 
158 EADIAXT HEAT. chat. vin. 
 
 tor towards tlie sun, w itliuut applying' thu disc l /, it will 
 be found that a naiiow zone of dazzling \vliite light is pro- 
 duced on the black bulb of the focal thermometer p, the 
 mercurial column commencing to rise the moment the rays 
 strike the reflecting surface. With a perfectly clem' sky, the 
 column has been found to reach 320 deg. Fall, in 35 sec. 
 The screen !• / being applied after cooling the thermometer, 
 a zone of feeble gray light appears on the black bulb nearly 
 as deep as the one produced by the rays froin the photo- 
 sphere, but situated somewhat lower. The column of the 
 focal thermometer invariably remains stationary, excejiting 
 the oscillation which always takes place when a thermo- 
 meter is subjected to the influence of the currents of air 
 unavoidable in a place exposed to a powerfid sun. It is 
 proper to remark that, owing to the stated oscillation, it 
 cannot be positively asserted that no heating whatever has 
 been produced by the reflection and concentration of the 
 raj's which fonu the zone of gray light adverted to. But 
 the recorded oscillations prove absolutely that the heating 
 does not exceed 0.5 deg. Fah. 
 
 Assuming that a temperature of 0.5 deg. Fah. has actu- 
 ally been produced by the reflected concentrated heat ema- 
 nating from the solar envelope, the following calculation \\ ill 
 show that the energy thereby established is too insignificant 
 to exercise anj- appireciable influence on the sun's radiant 
 power. Theoretically, the temperature transmitted to the 
 bulb of the focal thermometer by the reflection of the rays 
 / and 0, Fig. 2, is as the foreshortened illuminated area of 
 
CH A I". VIII. RAl>fATfX(l I'OWEli Ol-i^OL.yn ATMOSl'IlKUi:. loH 
 
 the ivrtector to the aiva of tlie zone of light pimliU'L-d on 
 the bulb. Obviously these areas bear nearly the same rela- 
 tion to each othei" as the squares of f or u' to the square of 
 the radius of the bulb ji. The length of _/' being 4.77 ins., 
 while the radiu.s of the bulb is 0.12.") in., calculation shows 
 tliat the temperature transmitted by the ray / would be 
 increased 1,45G times if the reflectoi' did not absorli anv 
 heat. Allowing that (>.7l^ of the heat is reflected, the aug- 
 mentation of intensity by concentration will amount to 0.72 
 X 1,456 = 1,048 times the temperature transmitted by the 
 rays / and o. The records of the oscillations of the mercu- 
 rial column during the experiments show, as stated, that 
 the temperature resulting from concentration caiHH)t exceed 
 0.5 deg. ; hence the temperatuie transmitted by the rays 
 emanating from tlie heated matter of the solar envelope will 
 
 only amount to — ^ = U.U0047 deg. F;ili. The recorded 
 
 observations having been made when the sun's zenith dis- 
 tance was 32 deg. 15 min., a correction for loss occasioned 
 by atmospheric absorption amounting to 0.2G will, hinvever, 
 be necessaiy. This correction being made, it will be found 
 that the heat actually transmitted by the I'ays from the solar 
 enveloj^e during the experiment referred to did not exceed 
 0.00059 deg. Fah. — a fact which completely disposes of 
 Secchi's remarkable assumption that the high temperature 
 of the photosphere is owing to the " radiation received from 
 all the transparent strata of the solar envelope" (see his 
 letter to Xatun, published June 1, 1671). 
 
IGO HABIAXT HEAT. chap. viii. 
 
 Having tlius positively established the fact that no appre- 
 ciable heat is transmitted to the earth by the radiant power 
 of the solar atniosj^here, and thereby disposed of Secchi's 
 erroneous assumptions, and proved the unsoundness of the 
 views entertained by other physicists that " the regions near 
 the sun augment the radiant energy transmitted by the lumi- 
 nary," let us now consider the probable weight and depth 
 of the solar atmosphere. The investigation will be greatly 
 facilitated by instituting a comparison between the sun's 
 envelope and the terrestrial atmosphere, and by adoj)ting 
 as a basis in our calculation the fact, established in Chap. 
 X., that the temperature at the surface of the photosphere, 
 and hence that of the . contiguous solar atmosphere, exceeds 
 4,000,000 deg. Fah. The fallacy of Dulong and Petit's for- 
 mula relating to the rate of cooling of incandescent matter 
 at high temperatures, and its consequent inapplicability to 
 the question of solar temperature, having been fully demon- 
 strated in Chap. II., Avhile our actinometric observations 
 recorded in Chap. III. have established the intensity of 
 solar radiation at the boundary of the terrestrial atmosphere, 
 we possess, it should be borne in mind, the elements neces- 
 sary to prove the correctness of the assumed high degree of 
 solar temperature. As before stated, our investigation wall 
 be simplified by comparing the solar and terrestrial atmo- 
 spheres ; hence the following mode of solving the important 
 problem : The increase of the volume of atmospheric air, 
 under constant pressure, being directly proportional to the 
 increment of temperature, while the coefficient of expansion 
 
CHAP. VIII. RADIATIXG POWER OF SOLAR ATMOHI'IIERE. ICl 
 
 is 0.00203 deg. for 1 deg. of Faliiviiheit, it will he found 
 by calculation that 3,272,000 deg. Fall, (that being the mean 
 temperature of the solar atmosphere) communicated to the 
 terrestrial atmosphere would reduce its density to Wrj of 
 the existing density. Accordingly, if we assume that the 
 height of our atmosphere is only 42 miles, the elevation of 
 temperature mentioned would cause an expansion increasing 
 its height to 6,643 X 42 = 270,000 miles. This calculation, 
 it should be observed, takes no cognizance of the diminution 
 of the earth's attraction at great altitudes, which, if taken 
 into account, would considerably increase the estimated 
 height. Let us now snj^pose the atmosphere of the sun to 
 be replaced by a medium similar to the terrestrial atmo- 
 sphere raised to the temperature of 3,272,000 deg. Fah., and 
 containing the same quantity of matter as the terrestrial 
 atmosphere for coiTesponding area of the solar surface. Evi- 
 dently the attraction of the sun's mass would, under these 
 conditions, augment the density and weight of the supposed 
 atmosphere nearly in the ratio of 27.9 : 1 ; hence its height 
 
 , , , , , 279,006 
 
 would be reduced to ^^ ^ = 10,000 miles. But if the 
 27.9 
 
 atmosjihere thus increased in density by the sun's superior 
 attraction consisted of a compound gas, principally hydro- 
 gen, say 1.4 times heavier than pure hydrogen (the specific 
 weight of ^vhich is only tV of that of atmospheric air), the 
 height would be 10 X 10,000 = 100,000 miles. The pres- 
 sure exerted by this supposed atmospliere at the surface of 
 the photosphere would obviously be 14.7 X 27.9 = 410 lbs. 
 
162 E AVIAN T HEAT. chap. yiii. 
 
 per sq. in. nearly. It will be observed that our computa- 
 tions are based on a solar attraction of 27.9, instead of the 
 recent estimate of 27.2. The foregoing calculations prove 
 that, unless the depth greatly exceeds 100,000 miles, and 
 unless it can be shown that the mean temperature is less 
 than 3,272,000 deg. Fah., the important conclusion must be 
 accepted that the solar atmosphere contains an exceedingly 
 small quantity of matter. Now, the assumed mean teuqie- 
 rature, 3,272,000 deg. Fah., so far from being too high, w\Y[ 
 be found to be underrated. It will be seen, on reference to 
 Chap. X., that the temperature at the surface of the photo- 
 sphere, determined in accordance Avith well-ascertained ele- 
 ments, somewhat exceeds 4,035,000 deg. Fah. Consequently, 
 as the diminution of intensity caused by the dispersion of 
 the rays is inversely- as the convex area of the photosphere 
 and that of the sj)here formed by the boundaiy of the solar 
 envelope — namely, as 1.52 : 1 — imder the supposition that the 
 depth of the solar atmosphere is 100,000 miles, the tempera- 
 
 •1. , 4,035,000 
 
 ture at the said boundary will be = 2,654,000 deer. 
 
 • 1.52 ° 
 
 The true mean, therefore, will be 3,344,800 deg., instead of 
 
 3,272,000 deg. Fah. — a difference which leads irresistibly to 
 
 the inference that either the sun's atmosphere is more than 
 
 100,000 miles in depth, or it contains less matter than the 
 
 terrestrial atmosphere for corresponding area of the solar 
 
 surface. The ratio of diminution of the density of the gases 
 
 composing the solar atmosphere at succeeding altitudes is 
 
 represented by Fig. 3, in Avhich the length of the oi'di- 
 
CHAP. VIII. RADIATIM! POWER OF tiOLAlt ATMOHl'll EliE. IC.'i 
 
 nates of tlie curve a d h sLows the degree of teuuity at 
 definite points above the photosphere. This curve has been 
 constructed agreeably to the theory tliat the densities at dif- 
 ferent altitudes, or, what amounts to tlie same, the weight of 
 the masses incumbent at succeeding points, decreases in geo- 
 metrical progression as the height above the base increases 
 in arithmetical progression. The vertical line a c has been 
 divided into 42 equal parts, in order to facilitate compari- 
 sons with the terrestrial atmosphere, supposed to be 4:2 miles 
 deep, the relative density of which, at corresponding heights, 
 is obviously as correctly represented by our diagram as that 
 of the solar atmosphere. It is true that, owing to the greater 
 height of the latter compared with the attractive force of the 
 sun's mass, the upper strata of the terrestrial atmosphere will 
 be relatively more powerfully attracted than the upper strata 
 of the vastly deeper solar atmosphere. The ordinates of the 
 curve a d h will therefore not represent the density quite 
 correctly in both cases. The discrepancy, however, resulting 
 from the I'elatively inferior attraction of the sun's mass at 
 the boundary of its atmosphere will be very nearly neutral- 
 ized by the increased density towards that boundary, conse- 
 quent on the great reduction of temperatui-e — fully 1,380,000 
 deg. Fah. — caused by the dispersion of the solar rays before 
 entering space. It may be Avell to state that, in representing 
 the relative height and pressure of the terrestrial atmosphere, 
 a c in our diagram indicates 42 miles, while h c indicates a 
 pressure of 14.7 lbs. per s(p in. ; and that, in representing 
 the solar atmosphere, a c indicates 100,000 miles, anil l> c 
 
164 BADIANT HEAT. chap. vm. 
 
 410 lbs. per stp in. Beariug in miud tlie liigb temperature 
 and the exceedingly small specific gravity of the matter 
 composing tlie solar atmosphere, the exti'eme tenuity of the 
 higher regions, indicated by the ordinates sho-wn in the dia- 
 gram, will be readily comprehended. Calculation shows that 
 towards the assumed boundary the density of the solar atmo- 
 sphere is so far reduced that it contains -only to^Wo of the 
 quantity of matter contained in an equal volume of atmo- 
 sphere at the surface of the earth. 
 
 The diminution of intensity consequent on the increased 
 depth of the solar atmosphere through which the calorific 
 rays pass, which are projected towards the earth from the 
 receding surface of the photosphere, having been considered 
 in Chap. VI., it will only be necessary to mention in this 
 place that Fig, 4 represents the sun, and its atmosphere 
 extending i of the semi-diameter of the photosphere, ni h, 
 c g, etc., being the rays projected towards the earth. The 
 depth of the solar atmosphere at a distance of ii of the 
 radius from the centre of the luminary, it will be seen, 
 amounts to only 2.0012 of that of the vertical depth. It is 
 hardly necessary to observe that the radiant energy trans- 
 mitted by the ray c d will be to the energy transmitted by 
 a ^ as the sine of the angle f c d to unity. 
 
 The foregoing reasoning demonstrates that the solar atmo- 
 sphere, owing to its enormous temperature, may reach a height 
 of 100,000 miles and yet not contain more matter on a given 
 area of the sun than the terrestrial atmosphere on an equal 
 area. I have endeavored to verify this important conclusion 
 
CHAP. VIII. TiATUATISG POWEU OF .SOLA h' ATMOSI-IJIJI.'E. Ifio 
 
 piactifiilly, and for that purpose resorted to the expedient 
 of enlarging the disc/ (see illustration on Plate 15) until the 
 spectrum disappear which is formed on the focal thermo- 
 meter by the concentration of the rays emanating from the 
 sun's atmosphere. It should be stated that the original object 
 of the instrument illustrated was merely that of ascertaining 
 whether the incandescent matter contained in the solar atmo- 
 sphere transmits radiant heat of sufficient energy to admit of 
 thenuometric measurement. But the appearance of a sjiec- 
 trum on the bulb of the focal thermometer, after shutting 
 out the rays from the photosphere, suggested the expedient 
 of substituting for the thermometer a small cylindrical stem 
 of metal, coated with lamp-black, in order to ascertain Avith 
 some degree of i>recision what amount of enlargement of the 
 disc /' is necessaiy to exclude the focal spectrum, as bj- that 
 means the depth of the sun's atnicsphere might be measured. 
 The result of the observation proves that while a disc of 10 
 ins. diameter effectually shuts out the rays from the photo- 
 sphere, an enlargement of about 0.15 inch of the radius of 
 the disc is necessaiy to exclude completely the observed spec- 
 trum from the focal stem. Now, the distance between the 
 spectrum and the disc being 53.7 ins., it will be found by 
 calculation that the stated enlargement of the disc corresponds 
 with an angular distance of 9' 45"; hence, assuming the radius 
 of the photosphere to be 426,000 miles, the depth of the mea- 
 surable part of the solar envelope cannot be less than 255,000 
 miles. 
 
CHAPTER IX. 
 
 THE FEEBLENESS OE SOLAR RADIATION DEMONSTRATED. 
 
 It is a remarkable fact tliat some of tlie most prominent 
 scientists entertain wholly incorrect views regarding the sun's 
 radiant intensity. Apparently, they are not aw^are that the 
 temperature produced by unaided solar radiation is fully 300 
 deg. Fahrenheit below the freezing-point of water. Sir John 
 Herschel, in discussing the increase of the intensity of solar 
 radiation consequent on the reduced distance from the sun 
 when the earth is in perihelion, presents the following views : 
 " In estimating the effect of any additional fraction, as ont- 
 jifteenth, of solar radiation on temperature (this fraction being 
 determined by applying the law of inverse squares to the dimi- 
 nution of the sun's distance when the earth is in perihelion), 
 we have to consider as our unit, not the niimber of degrees 
 above a purely arbitrary zero-point — such as the freezing- 
 point of water or the zero of Falirenheit's scale — on which 
 a thermometer stands in a hot summer day, as compared 
 with a cold winter one, but the thermometric interval between 
 
CHAP. IX. THE FEEBLENESS OF SOLAE liADIATIOX. 167 
 
 t/ie temperature it indicates in the two cases and that which 
 
 it would indicate did the sun not exist, which there is good 
 
 reason to believe would be at least as low as 239° below 
 
 zero of Fahrenheit. And as a temperature of 100° above 
 
 zero is no uncommon one in a fair shade exposui-e under a 
 
 sun nearly vertical, we have to take one-fifteenth of the sum 
 
 /239 + 100\ 
 
 of these intervals ( ) = 23^ Fahrenlieit, as the least 
 
 \ 15 ' 
 
 variation of temperature under siich circumstances Avhich can 
 reasonably be attributed to the actual variation of the sun''s 
 distancey It will be observed that the foregoing quotation 
 has partially appeared in a preceding chapter, yet it could 
 not be omitted in this place without rendering the demon- 
 stration incomplete. Considering that "absolute zero" (ascer- 
 tained in the meantime) Is 460° below the zero of Fahrenheit 
 instead of 239°, as supposed by Herschel, it will be seen 
 that, according to his doctrine, the increase of temperature 
 resulting from the su)\!s proximity irhen the earth is in peri- 
 
 helion shoidd be = 37° F. Referring to Chap. 
 
 15 
 
 IV., it will be found that solar intensity at the atmospheric 
 boundaiy is 90°.72 F. during the winter solstice, and that the 
 increase of the radiant intensity at that time, owing to the 
 sun's proximity, is 4°.66 F., besides the loss of energy, 0.207, 
 caused by atmospheric absorption, together 5°.88 F., instead 
 of 37° F. This extraordinary discrepancy is the result of Sir 
 John Hei-schel's misapprehension of solar intensity. He sup- 
 poses, as we have seen, that the enei-gy required to raise the 
 
168 EADIANT HEAT. chap. IX. 
 
 temperature from absolute cold to Fabreiilielt's zero, added to 
 the energy uecessary to raise tlie tempei'ature from tbat zero 
 to the j)oiut reached on the Fahrenheit scale, indicates the 
 tnie intensity of the radiant heat ; hence 460', in addition to 
 the 90°.72 before mentioned, together 550°.72 F., instead of 
 90°.72 F. Referring again to Chap. IV., it will be found 
 that the intensity of solar radiation when the earth is in 
 
 n T • o.o ■.. 1 n MT . 'J0.72 + 84.84 
 
 aphelion is 84 .84 b. ; hence the mean will be 
 
 2 
 
 = 87°.78 F. Now, the mean distance of the earth from the 
 sun's centre being 91,430,000 miles, it will be perceiv^ed that 
 solar intensity at that distance cannot exceed 87°.78 Fali- 
 renheit. In order to show the practical result of this deter- 
 mination of solar intensity, let us suppose that an air-ther- 
 mometer, suiTounded by some permanent gas maintained at 
 a temperature of 100° F. above absolute zero, be exposed to 
 the sun, and that the side of the bulb exposed to the sun's 
 rays is nearly flat, while the back is semi-spherical and effec- 
 tually protected by non-conducting substances. It needs no 
 demonstration to prove that the said thermometer will indi- 
 cate 100 + 87.78 = 187°.78 F. above absolute zero, or 492 
 — 187.78 = 304°.22 F. below the freezing-point of water, 
 although exposed to the full energy of the sun's rays. Per- 
 sons assigning a high teinjierature to the surface of the moon 
 will do well to consider the important fact thus established. 
 A moment's reflection will convince them that, but for the 
 accumulation of heat effected by the intervention of the ter- 
 restrial atmosphere, water could not exist in a fluid state, 
 
ciiAP. IX. THE FEEBLENESS OF SOLAR UADIATION. 1G9 
 
 aud that even the vertical rays of the sun within the tro- 
 pics would not possess sufficient power to retain mercury in 
 a fluid state. 
 
 It has been asserted by physicists that the differential 
 temperature shown by thermometers, however judiciously 
 arranged, does not furnish a reliable indication of solar 
 energy, on the ground that when the supposed maximum 
 intensity has been reached the temperature of the bulb 
 Italances that of the solar heat, thus preventing further in- 
 crement. The radiating power of the heated bulb, it is 
 urged, remains undiminished, while the differential tempera- 
 ture between the same and the surrounding medium is at its 
 maximum ; consequently promoting maximum loss of energy 
 by radiation. In order to test practically the merits of this 
 plausible argument, and in order to determine the tine inten- 
 sity of solar radiation, I have constructed the instrument 
 illustrated on Plate 17. The delineation represents a ver- 
 tical section through the centre line of the instrument. The 
 leading feature of the device is that of applying a hollow 
 revolving sphere (composed of very thin copper) within an 
 exhausted cylindrical vessel. This revolving sphere, coated 
 with lamp-black inside and outside, is exposed to the sun's 
 rays admitted through a thin crystal covering the open end 
 of the exhausted vessel, the diameter of the crystal being 
 equal to that of the sphere. The exhausted cylindrical vessel, 
 it will be seen by inspecting the illustration, is surrounded by 
 an external casing, water of a given temperature being circu- 
 lated through the intervening space. The sphere is caused 
 
170 BADIANT HEAT. OHAP. IX. 
 
 to revolve by means of a small hand-wheel attached to a 
 hollow stem connected with the sphere; the said stem tm-n- 
 ing in an air-tight stufBng-box applied on the upper side of 
 the exhausted vessel. A thermometer is inserted through the 
 hollow stem, the bulb, coated with lamp-black, occupjdng a 
 central position within the sphere. Obviously the thermo- 
 meter participates in the rotary motion of the sphere when 
 turned by the hand-wheel. The instrument is supported on 
 columns secured to a table provided with parallactic move- 
 ment for the purpose of pointing the axis of the exhausted 
 cylindrical vessel towards the solar centre. The water circu- 
 lated through the external casing of the instrument is main- 
 tained at a constant temperature of 60° F., a vacuum being 
 kept up within the internal vessel. As the thermometer 
 does not fit air-tight in the hollow stem, it will be evident 
 that the pressure of the air mthin the revolving sphere will 
 at all times balance the external atmospheric pressure. Let us 
 now institute a comparison between the instrument described 
 and the ordinary thermometer. 
 
 The convex area of the revolving sphere being four times 
 greater than the area of its great circle, the latter being equal 
 to the area of the pencil of rays admitted through the crystal, 
 it will be evident that the refrigerating surface of the sphere 
 is foiir times greater than the sectional area of the solar rays 
 which supply the radiant heat. Accordingly, if the surround- 
 ing cylindrical vessel be permitted to radiate freely towards 
 the centre, the temperature retained by the revolving sphere 
 thus exposed to cold radiation from all points will be only 
 
CHAP. IX. THE FEEBLENESS OF SOLAIi RADIATION. 171 
 
 one-fourtL of the temperature capable of being imparted by 
 the pencil of rays to the face of a flat disc composed of some 
 non-conducting substance. It may be stated, in further expla- 
 nation of the foregoing demonstration, that, since the surround- 
 ing exhausted vessel is maintained at a constant temperature 
 of 60° .F., it will radiate heat of that energy towards the 
 sphere. An exchange will consequently take place which Avill 
 prevent maximum temperature being attained by the sphere 
 unless the radiant energy transmitted by the solar rays enter- 
 ing the instrument be four times greater for equal area than 
 the I'adiant energy of the heat-rays projected by the sphere 
 towards the cold enclosure. It follows from this important 
 proposition that the temperature acquired by the revolving 
 sphere represents only one-fourth of the intensity of the 
 radiant energy actually passing through the crystal of the 
 exhausted cylindrical vessel. It will be I'eadily perceived 
 that the temperature of the air vrithin the revolving sphere 
 correctly represents the temperatiare of the metal composing 
 the same ; also, that the metal itself, owing to its almost 
 perfect conductivity, will become uniformly heated all over. 
 The inserted thermometer, therefore, will show the tempera- 
 ture of the revolving sphere sufficiently near for the object 
 in view. As already demonstrated, only one-fourth of the 
 radiant heat entering through the ciystal is retained by the 
 sphere ; hence the thermometer will indicate only one-fourth 
 of the actual intensity of the sun's rays. Accordingly, if we 
 multiply the indication of the thermometer within the revolv- 
 ing sphere by 4, we ascertain the true solar intensity, less the 
 
ira BADIANT HEAT. chap. ix. 
 
 heat absorbed by the crystal covering the exhausted vessel. 
 The result of careful observation has proved the soundness 
 of the foregoing reasoning and demonstration. The conclud- 
 ing investigation, instituted when the zenith distance was 30 
 deg. 50 niin., established the fact that while the standard aeti- 
 nometer indicated a solar intensity of 54°.57 F., the thermo- 
 meter within the revolving sphere indicated a differential 
 temperature of 13°.3 F. According to our theory, it should 
 
 54.57 
 have indicated — '- — = 13°. 64 F., thus showing a deficiency 
 
 of 0°.34. If, however, a correction be introduced adding the 
 proportion of heat absorbed by the crj'^stal — viz., 0.066 — the 
 stated deficiency of temperature will be more than balanced. 
 This apj)arent inaccuracy is occasioned by the radiation of 
 the crystal towards the sphere, and by the diminution of 
 the radiating surface of the surrounding vessel at the point 
 where the crystal is inserted. The last-mentioned source of 
 error may be easily ascertained, as it dejiends on the solid 
 angle formed by straight lines drawn from the circumference 
 of the crystal to the centre of the sphere. The deficiency 
 of radiating surface ascertained by that process amounts to 
 0.012. Proper allo^vance having been made on account of 
 these sources of error, it was foimd during the concluding 
 investigation referred to that the temperatui'e retained by 
 the sphere exposed to the solar heat is exactly one-fourtli of 
 the temperature imparted and retained by the bulb of an 
 actinometer simultaneously exposed to the sun. It is im- 
 portant to observe that the difference between the intensity 
 
CHAP. IX. THE FEEBLENESS OF SOLAH UADTATTOX. 173 
 
 of the radiant heat of the sun and the tempei'ature of the 
 metal composing the sphere was 54°.57 - 13''.64 = 40°.93 F. 
 during the investigation ; while the temperature of the bull) 
 of the actinometer exposed to the sun, at the same time, bal- 
 anced the intensity of the solar heat. This fact completely 
 refutes the assertions of certain physicists before referred to. 
 Regarding the actual iiitensitij of solar radiation, no further 
 demonstration is needed to show that the temperature indi- 
 cated by the thermometer within the revolving sphere, mul- 
 tiplied by 4, determines the true energy of the sun's radiant 
 heat at the surface of the earth ; not, however, including the 
 energy lost by atmospheric absoqition. The close agreement 
 between the indication furnished by the instrument thus exa- 
 mined and the indications of the actinometer described in 
 Chap. III. proves, it is satisfactory to observe, the reliable 
 character of the tables of solar temperature contained in 
 that chapter constructed in accordance with our actinome- 
 trie observations. It remains to be noticed that, before the 
 conception of a dry revolving bull), I constructed an instru- 
 ment for determining solar intensity, in which a large sta- 
 tionary hulb, filled with water and provided with an internal 
 rotating paddle-wheel, was employed. The illustration on 
 Plate 18 represents a section through the vertical plane of 
 that instrument. Before giving a description of the same, it 
 will be well to present an outline of the reasoning Avhich 
 led to its construction. 
 
 Suppose a small spherical body of perfect conductivity 
 and radiating power to be suspended witliiu a large enclo- 
 
174 BABIANT HEAT. CHAP. ix. 
 
 sure provided Tvath ca perforation in the direction of tLe 
 sun, of sufficient size to admit a pencil of rays of the same 
 diameter as the spherical body. Suppose, also, that the tem- 
 perature of the said body is 64° F. higher than the temjje- 
 rature of the enclosure and the air contained within the 
 same. The convex area of a sphere being four times greater 
 than the area of its great circle, while the area of the great 
 circle of the sphere which we have imagined corresj)onds 
 exactly with the sectional area of the pencil of rays entering 
 through the perforation of the enclosure, it will be evident 
 that tbe supposed excess of temperature, 64°, cannot be 
 maintained unless the radiant energy of the sun's rays be 
 four times greater for corresponding area than the radiating 
 energy of the sphere. It will also be evident that, if the 
 assumed excess of temperature of the sphere gradually falls 
 while exposed to the sun's rays, until it is reduced to 16° 
 above the temperature of the enclosure, then the intensity 
 of the sun's rays cannot be more than 4 X 16 = 64°. Bear- 
 ing in mind that the section of the pencil of . rays which 
 transmits the energy is only 0.25 of the convex area of the 
 radiating sphere which receives the beat and, in turn, radi- 
 ates that heat towards the enclosure, we cannot question the 
 correctness of the deduction that, assuming the rays to be 
 parallel, the intensity of the radiant energy which enters 
 through the perforation of the enclosure is four times greater 
 than the radiant energy wbicli the sphere parts with. Again, 
 the intensity of the radiant heat emanating from solid bodies 
 being a correct index of dynamic energy, it -svill be perceived 
 
CHAP. IX. THE FEEBLENESS OF SOLAR liADIAIJON. 175 
 
 that the energy transmitted to the enclosure by the radiating 
 sphere at a differential temj)erature of 16° will be exactly 
 balanced by the energy transmitted by the pencil of rays at 
 64° entering through the perforation of the enclosure, the area 
 of which is 0.25 of the convex area of the sphere. 
 
 The demonstration thus presented, it will be admitted, 
 establishes the important fact that the temjierature produced 
 by solar radiation is four times higher than the differential 
 temperature of a black sphere, composed of materials of per- 
 fect conductivity, exposed to the sun, and permitted to radiate 
 freely towards an enclosure of a unifonn temperature. 
 
 Let us now examine the instrument before referred to, 
 shown by out illustiation, which represents a section through 
 the central vertical plane : I- j9 is a spherical vessel com- 
 posed of copper, charged with water and coated with lamp- 
 black on the outside, suspended within a spherical enclosure 
 0. The latter is provided with a circular opening a h, to 
 which a cylindrical tniuk a g is attached, the spherical enclo- 
 sure as well as the trunk being coated with lamp-black on 
 the inside, as shown by the black tint in the illustration. 
 A thermometer provided with a cylindiical bulb is inserted 
 into the spherical vessel, and also a rotating paddle-wheel, 
 operated by an axle passing through a water-tight stuffing- 
 box. A cylindrical vessel r r, filled with water, surrounds 
 the spherical enclosure and trunk, nozzles being applied at 
 the top and bottom, to which flexible tubes are attached for 
 circulating a cuiTent of cold water through the vessel. The 
 instrument is mounted within a revolving observatory, and 
 
176 BADIANT HEAT. chap. ix. 
 
 attacLed to a table turning on horizontal journals, and pro- 
 vided witli appropriate meclianism, by means of wliicli it 
 may be directed at right angles to the sun. It mil be evi- 
 dent that, if the axis of the cylindrical trunk a g be pointed 
 accurately towards the centre of the sun, the sphere Tc 2> will 
 receive the whole radiant energy of the rays within the tan- 
 gential lines h f and p g, the sectional area of the pencil of 
 rays, as before stated, being 0.25 of the convex area of the 
 sphere. It will be evident also that, owing to the opening a l> 
 of the spherical enclosure, the sphere h p will not be acted 
 upon by the full amount of refrigeration that would be pro- 
 duced by the radiation of a continuous enclosure. Agreeably 
 to the theory of exchanges, the deficiency will, however, not 
 be great, since the side / « of the tnink a g will radiate as 
 powerfully towards the sphere as a portion of the spherical 
 enclosure corresponding with the angular distance determined 
 by the radial lines a e and / c. But the convex surface of 
 the segment e d, depending on the angle subtended by f^ c 
 and g c, will obviously be subjected to far less refrigeration 
 than an equal surface on the opposite side of the sphere. 
 Regarding the exact amount of deficient refrigeration conse- 
 quent on the opening in the enclosure at a h referred to, it 
 will be perceived, on reflection, that the radiation of the 
 enclosure towards the semi-spherical surface presented to the 
 sun will be reduced in the exact proportion which the area 
 e d bears to the entire convex area of the semi-sphere. It 
 will be seen, therefore, that although the condition coupled 
 with our proposition has not been fully complied with — 
 
CHAP. IX. THE FEEBLENESS OF SOLAli RADIATION. 177 
 
 namely, that tlie enclosure should Le of great extent com- 
 pared with the size of the radiating sphere— yet the enclo- 
 sure of our instrument and the comparatively lai-ge opening 
 at a h will not materially aflfect the refrigerating influence 
 to which the sphere is subjected. Besides, the known soliil 
 angle subtended by the radial lines f c and g c enable us to 
 calculate the amount of deficient radiating surface presented 
 by the enclosure. 
 
 Several experiments have been made simultaneously with 
 tliis instrument and a standard actinometer, in order to ascer- 
 tain the precise relation betAvecn the temperature ti'ansmitted 
 by the sun's rays to the radiating sphere and to the actino- 
 meter. Both instruments have invariably been attached to 
 the same parallactic table during the investigation ; conse- 
 queiitly the energy of the radiant heat transmitted to each 
 has been precisely alike. Respecting the instituted tests, it 
 will suffice to record the result of an experiment conducted 
 at noon, October 20, 1871, the solar radiation on that day 
 being of nearly average intensity, while the sun's zenith dis- 
 tance, 51 deg. 40 min., was also near an average. Observa- 
 tions made at equal intervals of 5 min., from 11 hours 55 
 min. A.M. to 12 hours 30 min., showed that the radiating 
 sphere of the instrument, the contents of which was effec- 
 tually agitated by the internal paddle-wheel, attained a tem- 
 perature of precisely 75°, while the enclosure was maintained 
 at a constant temperature of 61°.3. Accordingly, an increase 
 of temperature of 13°.7 above that of the enclosure was pro- 
 duced by the solar radiation acting freely on the sphere, 
 
178 BADIANT HEAT. chap. ix. 
 
 the actiuometer, at the same time, indicating a temperature 
 
 of 51°.36. Now, agreeably to our theory, the temperature 
 
 n 51.36 
 of the sphere ought to have been only — - — = 12 .84, thus 
 
 showing a discrepancy of 13.7 - 12.84 = 0°.86 F. It has 
 already been explained that the sphere does not receive a 
 full amount of refrigeration, in consequence of the opening 
 in the enclosure necessary to admit the cylindrical trank ; 
 hence the temperature of the sphere ought to exceed that 
 which our theory has established. The observed difference, 
 0°.86, is, however, greater than it should be in accordance 
 with the relative magnitude of the convex surface e d and 
 the area of the sphere. But, referring to the illustration, it 
 will be seen that, at the point where the thermometer is 
 inserted, a considerable area of the sphere is not subjected 
 to any radiation from the enclosure, nor at the point where 
 the axle of the paddle-wheel enters. Adding these areas to 
 that of e d, calculation shows that the amount of cold radi- 
 ation prevented from acting on the sphere accounts very 
 nearly for the discrepancy of 0°.86 F. before referred to. 
 We are, therefore, warranted in stating that the temperature 
 indicated by the actinometer during the experiments has 
 j)roved to be exactly foiir times higher than that indicated 
 by the thermometer inserted in the sphere exposed to the 
 radiant power of the sun's rays and to the refrigerating in- 
 fluence of the enclosure. The soundness of our theory has 
 thus been fully proved, and, consequently, additional evi- 
 dence furnished of the correctness of the determination of 
 
CHAP. i.\. THE FEEBLENESS OF SOLAli RADIATION. 179 
 
 solar intensity by means of the actinomeiev described in 
 Chap. III. No further proof is needed in support of the 
 demonstration already presented, sho^^ang that the tempera- 
 ture produced by solar radiation, instead of being, as Sir 
 John Herscliel supposed, equal to the maximum shade tem- 
 perature within the tropics added to the temperature of the 
 Fahrenheit zero above absolute zero — viz., 100 + 460 = 5G0° 
 F. — scarcely reaches 88° F. at a distance of 91,430,000 miles 
 from the solar centre. 
 
 Concerning the radiant heat which reaches the distant 
 planets of the solar system, the stated discrepancy is of vital 
 importance. Were it true that the intensity of the sun's 
 radiant heat is 560° F. at the distance mentioned, the rays 
 on reaching Jupiter's atmosphere would be capable of de- 
 veloping a temperature of — — ^ = 20°.7 F. We can readily 
 
 imagine that the atmosphere of the giant planet might, by 
 
 some system of accumulation, raise this temperature to such 
 
 a degree that organisms like those of the earth might be sus- 
 
 88 
 tained. But can the insignificant temperature of — = 3°.2 F. 
 
 transmitted to Jupiter's atmosphere be sufficiently elevated by 
 the process of accumulation to sustain animate and vegetable 
 organizations resembling those of our planet ? The stated 
 low temperature need excite no sui-prise if we reflect on the 
 fact that the sun, as seen from the boundary of the atmo- 
 sphere of Jupiter, is no larger thiin an orange viewed at a 
 distance of one hundred feet. As seen from Saturn, the size 
 
ISO J?ADIANT TIEAT. CHAP. ix. 
 
 of the sun is that of a musket-ball at a distance of fifty feet 
 from the observers eye ; while the transmitted solar heat 
 scarcely develops a temperature of 1° F. where it enters 
 Saturn's atmosphere. Speculations regarding the habitabi- 
 lity of the distant planets are futile, in view of the insuffi- 
 cient radiant intensity of solar emission established by the 
 actinometric observations recorded in this work, and by the 
 adopted tests proving their reliability. 
 
CHAPTER X. 
 
 TEMPERATURE OF THE SOLAR SURFACE. 
 
 The illustration on Plate 19 represents an instrument for 
 ascertaining tlie temperature of the surface of the sun. At first 
 sight it Avill appear futile to undertake the construction of an 
 instrument capable of indicating temperature at a distance 
 exceeding 90,000,000 miles ; but in view of the fact that the 
 sun has been weighed by an instrument consisting principally 
 of four leaden balls less than one foot in diameter, the attemjjt 
 cannot justly be deemed absurd. The reader will remember 
 that in the celebrated Cavendish experiments, afterwards re- 
 peated by Baily and others, the weight of the earth — on 
 which the weight of the sun is based — was ascertained by 
 measuring the attraction exerted by spheres of lead weigh- 
 ing 174 lbs. The delicate nature of the experiment may be 
 infen-ed fi-om the fact that the ascertained attractive force 
 was found to be only nVir of a grain. The illustrated instiu- 
 ment, the solar pyrometer, by means of which the tempei'atui'e 
 of the sun has been measured, involves no such nicety. 
 
 181 
 
182 BABIANT HEAT chap. x. 
 
 Before entering on a description of the solar pyrometer, 
 it will be necessary to call attention to tlie demonstration in 
 Chap. I., showing that the law relating to radiating spheres 
 is also applicable to concave spherical radiators, if the sub- 
 stances exposed to their radiant heat be placed in their foci. 
 The demonstration referred to also proves that the tempera- 
 ture produced by the radiant heat transmitted by concave 
 radiators of equal temperatures and curvature, at equal dis- 
 tances, is directly as their areas. Melloni and Leslie's experi- 
 ments, conducted in the presence of the disturbing influences 
 of atmosjjheric air, not being sufficiently accurate to warrant 
 their being cited in support of the correctness of the stated 
 relation between areas and temperatures, the construction of 
 the pyrometer has been so modified as to enable us to prove, 
 independently of the demonstrations in the preceding chapter 
 referred to, that, under the stated conditions, the temperatures 
 correspond exactly with the areas. 
 
 Our illustration represents a longitudinal section through 
 the vertical plane, and a photographic perspective view of the 
 pyi'ometer. It will be seen, by inspecting the longitudinal sec- 
 tion, that the instrument is composed of four principal parts : 
 (1) A heater consisting of a cylindrical vessel with spherical 
 bottom and open top, supported by an ordinary stove, the fire- 
 chamber of which it partially entei's. Enlargements resem- 
 bling truncated cones with concave spherical ends are formed 
 near the middle of the heater. The latter is pai-tially filled 
 with water, as shown in the illustration. (2) A conical vessel, 
 surrounded by a double casing, secured to the base of the 
 
CHAP. X. TEMPEHATUBE of the SOLAB SriiFACE. 183 
 
 large conical enlargeiueut of the heater. (3) A cylindrical 
 vessel secured to the small end of the enlai-gement, likewise 
 surrounded by a double casing. (4) An ordinaiy stove, into 
 wliicli the lower end of the heater is inserted. The curva- 
 ture of the spherical concavity at the base of the large conical 
 enlargement of the heater is stioick to a radius of 18 inches, 
 its diameter being 10 inches, hence presenting an area of 78.84 
 square inches. The opposite spherical concavit}-, the ladius of 
 which is 9 ins. (its diameter being 5 ins.) presents an area of 
 
 78.84 
 
 — - — = 19.51 square inches. Thermometers are applied at 
 
 the/(9Ci of the spherical concavities, their stems being placed 
 as shown in the illustration, in order that the bulbs may 
 present unobstructed semispheres towards the radiatoi-s. It 
 is hardly necessary to observe that thermometers intended to 
 measure the intensity of radiant heat should be protected so 
 that those parts of their bulbs which are not acted upon by 
 the heat-rays emanating from the radiators may not lose their 
 heat by radiation or convection. The cylindrical as well as 
 the conical chamber of the pyrometer containing the ther- 
 mometers are connected by suitable tubes with an air-pump, 
 by which the air is withdrawn ; a current of water being 
 circulated through the double casings when the instrument 
 is in operation. With reference to the heater, it should be 
 observed that, being open at the top, the water it contains 
 will always be maintained at a constant temperature when 
 the furnace is in action. 
 
 It ma}- be briefly stated that the principle of the pyro- 
 
1S4 BADIANT HEAT. chap. x. 
 
 meter is tliat of ascertaiiiiiig solar intensity by comparing 
 tlie temperature transmitted liy a concave spherical radiator 
 (if 10 ins. diameter to a tliermometer placed at a distance 
 of IS ins. from its face, ■\^itll tlie temjjerature produced by 
 the radiant Leat emanating from an incandescent sphere of 
 832,584 miles in diameter, at a distance of 91,430,000 miles. 
 The radiant heat in both cases is transmitted through ether; 
 in the former to the sui-face of the bulb of the enclosed ther- 
 mometer ; in the latter to the boundary of the earth's atmo- 
 sphere. The law which governs the transmission of radiant 
 heat through space is as absolute as the law of gravitation, 
 whatever be the distance ; hence it is indisputable that the 
 solar pyrometer in which the radiant heat acts at a distance 
 of 18 inches is as competent to determine the temperature 
 of the sun as the Cavendish leaden spheres acting at a dis- 
 tance of 8.85 inches to determine his weight. The chances, 
 however, of an exact determination are greatly in favor of 
 the pyrometer. In the first place, while the area of the con- 
 cave radiator of the j^yrometer is to the area of the great 
 circle of the sun as 1 : 2,871 X 10", the weight of the leaden 
 ball employed in the Cavendish experiments is to the weight 
 of the sun as 1 : 2,367 X 10"; thus showing a difference of 
 1 : 824,500,000 in favor of the pyrometer. Besides, the ele- 
 ment of distance through which the radiant and the gravi- 
 tating forces act is in favor of the pyrometer, in the ratio of 
 18 to 8.85. But these considerations, however important on 
 account of the greater difference of the magnitudes involved, 
 may be considered unimportant in comparison with the direct- 
 
CUAI>. X. TEMPJiKATUIiE Of THE HOLAU HUEFAVE. 185 
 
 ness of the means by Avliich the solar pyrometer solves the 
 problem, contrasted with the indirectness, exceeding compli- 
 cation, and nicety involved in the Cavendish experiments. 
 In the solar pyrometer we only require a correct indication 
 of the tempeiature of the radiating concave spherical surface, 
 and of the temperature transmitted to its focus ; together 
 with au accurate measurement of the distance of that focus, 
 and of the area of the radiating suiface. These points being 
 readily determined, while the relative distance and diameter 
 of the sun and the temperature produced by solar radiation 
 at the boundary of the terrestrial atmosphere are known, we 
 may enter upon and carry out our computation without intro- 
 ducing a single correction. How different the Cavendish expe- 
 riment, with its nimierous disturbing elements depending on 
 barometric and thermometric conditions and changes, influenc- 
 ing a gravitating force amoimting to only rAir of a grain ! 
 An account of the almost insuperable difliculties which were 
 surmounted in those remarkable experiments, which for inge- 
 nuity, care, and perseverance stand unequalled in the annals 
 of physics, would be out of place here ; yet the foregoing 
 brief allusion to experiments which satistactorily determined 
 the weight of the earth, and thereby the weight of the sun, 
 has been deemed appropriate as a contrast. The dii-ectness, 
 facility, and cei-tainty of measuring solar temperature by the 
 means we are now considering A\oixld scarcely be appreciated 
 without calling to mind the method adopted for ascei-taining 
 the sun's weight. 
 
 Referring to the construction of the solar pyrometer and its 
 
186 JiABIANT HEAT. CHAP. X. 
 
 apparently ponderous cliaractei', it Avill be ^\ ell to bear in niiml 
 tliat tlie indispensable condition in this instrument of maintain- 
 ing a constant temperature of the couipHratively large con- 
 cave spherical radiator is not easily fullilled. Kotliing short 
 of an o^yen heater containing a fluid Avhich readily evaporates, 
 and the application of an excess of heating power, will effec- 
 tually accomjUish the object in view. Evidently the loss 
 occasioned by radiation cannot be exactly made good by the 
 most delicate mechanical contrivance ; but by applying an 
 excess of heat in the furnace, the fluid which regulates the 
 temperature of the radiator will be prevented from falling 
 below the boiling point ; and since the heater is open, the 
 steam formed will caiiy off superfluous heat, and thus main- 
 tain the fluid at the desired unifoi'm temperature. The 
 exhaiisted chambers which contain the thei'mometers must 
 of course be maintained at a constant temperature, the least 
 fluctuation being fatal to accurate indication of the intensity 
 of the heat transmitted by the radiators. In order, there- 
 fore, to keep up the necessaiy constant temperatui'e during 
 the investigation, a current of water has been circulated 
 throuo-h the double casinos which surround the exhausted 
 chambers. By this expedient, in connection with the per- 
 fectly uniform temperatui'e maintained in the heatei', it has 
 been easy to ascertain with critical nicety the temperature 
 produced by the radiant heat transmitted from the spherical 
 radiator to its focus. Eegarding the area and curvature of 
 the radiators, accurate workmanship alone will insure A\hat 
 is requisite ; but the position t>f the thermometei', the placing 
 
CHAP. X. TEMrEnATrin-: or tiik solar srnFACE. isr 
 
 the bull) at tlie proper distance with reference to the focus, 
 demands some consideration. Obviously it would not be 
 correct to place the centre of the bulb in the focus of the 
 radiator, as that would bring the face of a bulb of i in. 
 diameter ^ in. in ailvance of said focus ; nor would it be 
 proper to carry the bulb so far back that its face would 
 intei"sect the focus. The focal distance being 18 ins., it will 
 be found that placing the bulb half way between these two 
 positions will cause an error of fully 0.007. Conflicting indi- 
 cation, it should be oliserved, is unavoidable, since every part 
 of the exposed half of the convex surface of the bulb can- 
 not be equidistant from the face of the concave i-adiator ; 
 but this notwithstanding, there is a distance at which the 
 indication of the thermometer will be precisely the same as 
 if its entire contents were concentrated in the focus of the 
 radiators. This position has been practically determined. 
 
 The hitherto accepted doctrine, that the intensity of radi- 
 ant heat is directly as the area of the radiatoi-s, for equal 
 distances, has been shoA\n, in a previous chapter, to be fal- 
 lacious, because the radiant heat transmitted from the boun- 
 daries of plane radiators becomes enfeebled by distance and 
 the conserpient dispersion of the heat-rays, in the ratio of 
 the squares of the distance between the I'adiator and the 
 recipient of the radiant heat. The solar pyi-ometer having 
 been constrnctetl before I had satisfactorily demonstrated 
 that the intensity of the radiant heat transmitted from con- 
 cave spherical surfaces is directly as the areas of such ladi- 
 ators, it was deemed necessary to esta1)lish the correctness 
 
188 BADIANT HEAT. chap. x. 
 
 of that assumption ; hence the solar pyrometer has been 
 modified as before mentioned. The lesser radiator attached 
 to the conical enlargement of the heater, and the cylindrical 
 chamber enclosing the same, were accordingly added to the 
 instrument. It has already been stated that the area of the 
 spherical radiatoi' within the conical chamber is exactly four 
 times greater than that of the opposite radiator, and that the 
 radius of the curvature of the latter is one-half of the radius of 
 the former. The demonstration contained in Chap. I., before 
 referred to, has established the fact that in concave sphe- 
 rical radiators presenting equal areas the radiant heat trans- 
 mitted is in the inverse ratio of the square of the distances 
 if the substance exposed to the radiant heat be placed in 
 the focus of the radiator. It follows from this demonstra- 
 tion that, for equal area, the intensity of the radiant heat 
 transmitted to the focus of the lesser radiator will be four 
 times greater than the intensity of the radiant heat trans- 
 mitted to the focus of the large radiator. But the area of 
 the latter is exactly four times greater than the area of the 
 former, while the thermometers in both chambers are exposed 
 to the radiation of surfaces heated by the same medium, and 
 therefore of precisely equal temperatures. At the same time 
 these thermometers ividiate against surfaces maintained at a 
 constant temperature, by the reliable expedient of employ- 
 ing a powerful continuous current of water. Consequently, 
 the enclosed thermometers, although exposed to radiators of 
 different area, should indicate precisely equal temperature. 
 Actual trial having shown that such is the case, the correct- 
 
CHAP. X. TEMPEEATUBE OF THE SOLAR SUIiFACE. 180 
 
 ness of the foregoing assumption must be accepted as fully 
 established. 
 
 I will now biieriy advert to the result of an experiment 
 made with the solar pyrometer while the atmospheric pres- 
 sure balanced 29.91 inches column of mercury, the tem- 
 perature of the water in the heater being then precisely 
 212°. Apart from having thus insured a definite indication 
 of heat applied to the concave radiatoi-, the temperature of 
 the current of cold water circulated throuu-h the casinir ■'^ur- 
 rounding the exhausted chaml>ers did not fluctuate in the 
 least during the experiment, the thernionieter inserted in the 
 exit-pipe of the casings continuing to indicate steadily 48°.l 
 F. The circulation of cold water having been kept up fully 
 half an hour previous to the experiment, it is hardly neces- 
 sary to state that, before the fire ^\■as applied in the furnace, 
 the enclosed thermometer, the surrounding chamber, the watei- 
 contained in the heater, and the radiator all indicated 48°. 1. 
 The fuel in the furnace having been ignited, and the water 
 in the heater brought to boiling-point, the temperature X)f 
 the spherical radiator was observed to increase from 48°. 1 
 to 212°, difference = 163°.9 ; the temperature of the focal 
 thermometer at the same time rising from 48°. 1 to G0°.3, 
 difference = 12°.2. 
 
 It results, from previous demonstrations (see Chap. I.), 
 that the temperature of spherical radiatin-s transmitting e(iual 
 intensities to their foci are invereely as the square of the 
 sines of half of the angles which they subtend — that is, the 
 angles formed by the axis of the radiator and the heat-rays 
 
190 BADIANT HEAT. cuxv. x. 
 
 projected from tlie oireiinifereiice to the focus. Conseqiiently, 
 as the spherical radiator of the solar pyrometer, the differen- 
 tial temperature of which is 163°.9, transmits to its focus an 
 intensity of 12°.2, -we are enabled to calculate what tem23e- 
 I'ature the sun must possess in order to transmit an intensity 
 of 12°.2 to the boundary of our atmosphere. The mean angle 
 subtended by the sun being 32 min. 1 sec. during the expe- 
 riment, while that subtended by the radiator of the pyi'ometer 
 was 32 deg. 15 min., it follows that the ratio of the square 
 of the sines of half these angles will be 1 : 3,567.7. Accord- 
 ingly, the sun, in order to produce by its radiant heat a tem- 
 perature of 12°.2 at the boundary of the atmosphere of the 
 earth, must possess a temperature 3,567.7 times greater than 
 that of the sjiherical radiator of the pyrometer. This latter 
 temperature being 163°.9, that of the sun cannot be less than 
 3,567.7 X 163.9 = 584,746° in order to transmit an intensity 
 corresponding ^\ith a thermometric interval of 12°.2 on the 
 Fahrenheit scale. But solar intensity at the boundary of our 
 atmosjDhere, as shown by our actinometric observations (see 
 
 S4 S4 
 Chap. III.), is 84°.84 ; hence —^ — = 6.95 times greater than 
 
 that transmitted by the radiator of the pyrometer to its focus. 
 The temperature of the sun, therefore, cannot be less than 
 6.95 X 584,746 = 4,063,984 deg. Fah. 
 
 It will be recollected that the demonstration in a preced- 
 ing chapter established with as much certainty as any propo- 
 sition in the "Principia" that the temperature produced by 
 the radiant heat transmitted l)y a sphere of uniform tempe- 
 
•-•UAP. X. TEMrEUATUIiE OF TUE tiVLAU HUliFACE. I'-'l 
 
 ratiire at tlie sui-fufe is to the teuipeniture of tlie splieie 
 itself inversely as the si|Uaie of the ladiiis to the square 
 of the distance fioni the centre to the point exposed to the 
 radiant heat. The distance between the earth and the sun, 
 at the snnnuer solstice, being such that the angle subtended 
 by the latter is 31 niiu. 32 sees., the ratio of distance and 
 radius will l)e 218.1 : 1 ; hence the ratio of the squares, 
 47,5G7 : 1. Consequently, the temperature of the sun must 
 be 47,507 times greater than the temperature produced by 
 solar radiation at the boundary of the earth's atmosphere. 
 That temperature being, as before stated, S4°.84, the sun's 
 temperature cannot be less than 47,507 X 84°.84 = 4,035,584 
 deg. Fah. Thus the previously demonstrated temperature of 
 4,063,984 deg. Fah., based on the indications of the solar 
 pyrometer and the angles subtended by its radiator, ditl'ers 
 only 0.007 from the computations just presented. The 
 methods by means of which these results have been reached 
 diftering entirely, both being based on sound physical and 
 mathematical principles, we cannot doubt the correctness of 
 the determination. Nor can it be questioned that the actual 
 temperature of the surface of the sun, at the point of maxi- 
 mum intensity, is still higher, since the lays in passing 
 through the solar atmosphere suffer considerable loss of 
 energy, as shoA\n in Chap. VI. 
 
 Let us now consider briefly the extraordinary diversity 
 of views entertained by scientists regarding the temperature 
 of the sun. In view of the fact that all practical data 
 necessary to solve the problem are known, it is sur[irising 
 
193 BABIANT HEAT. CHAP. X. 
 
 that any diflference of opiuiou sliould exist on the subject. 
 Zolluer apparently rejects the positive evidence of high 
 sohir temperature furnished by the fact that the sun's rays, 
 after having suffered dispersion in the ratio of 4G,000 to 1, 
 and penetrated the terrestrial atmosphere, are capable of 
 developing a temperature of nearly 70° F. on the ecliptic. 
 It will be remembered that he published, some time ago, 
 an elaborate demonstration, founded on the height of the 
 solar prominences, showing that the sun's temperature does 
 not exceed 70,000° C. Secchi, on the other hand, asserted, 
 in his original work on the sun, that, owing to the acces- 
 sion of energy received by radiation from the outer layers 
 of the solar atmosphere, the temperature of the surface of 
 the photosphere is fully 140 times gi-eater than the tempe- 
 rature announced by Zollner. Let us first notice the investi- 
 gations of the Italian astronomer. In his work " Le Soleil," 
 published at Paris, 1870, he presents calculations showing that 
 the temperature of the solar surface is at least 10,000,000° C. 
 Prof. Newcomb, in a review of the M'ork referred to, pub- 
 lished in Nature, showed that, if the temperature reached 
 ten million degrees of Centigrade, as asserted by the author 
 of "Le Soleil," the earth would speedily be converted into 
 vapor. In ansAver to this objection, Pere Secchi urged, " that 
 a body may have a veiy high temperature and yet radiate 
 very little," contending "that a thermometer dipped inside 
 the solar envelope in contact with the photosphere" would 
 indicate the temperature mentioned. " This high tempera- 
 ture," he observes, " is really a virtual temperature, as it is 
 
CHAP. x. TEMPEliATUliE OF THE soLM; nUliFAGE. 103 
 
 the amount of radiation received from all the transparent 
 strata of the sohir envelope, and this body at the outer 
 shell must certainly be at a lower temperature." What 
 information is intended to be conveyed by the statement 
 that 10,000,000" C. "is really a virtual temperature," on the 
 ground that it is " the amount of radiatiou received from all 
 the transparent strata outside of the photosphere," we can 
 only conjecture. 
 
 Our demonstrations, based on the indication of the solar 
 pyrometei-, ha\e shown that the supjjosed thermometer, if 
 brought in contact with the photosphere, cannot possibly 
 indicate the enormous temperature of 10,000,000° C. assumed 
 by the Italian plnsicist. The assertion that " a body may 
 have a very high temperature and }et radiate but very 
 little," were it correct with reference to the photosphere, 
 does not aflect the question. It is of no consequence whe- 
 ther the photosphere belongs to the class of active or slug- 
 gish incandescent radiators imagined by the distinguished 
 savant; the temperature of the radiant surface, not its capa- 
 city to radiate more or less copiously, is the problem to be 
 solved. 
 
 Very recently Pere Seechi, much to the surprise of those 
 who had accepted his estimate of solar temperatui-e published 
 in " Le Soleil," has changed his views completely. In an ela- 
 borate essay presented to the Academy of Sciences at Paris 
 he underrates the intensity of solar energy more than he for- 
 merly overestimated it. Apprehensive that a synopsis would 
 fail to give a correct idea oi the rcmarkalde demonstration 
 
194 BADIANT HEAT. CHAP. X. 
 
 by wliicli tlie author of " Le Soleil " now reverses all his 
 previous notions on the subject, and in order to furnish a 
 complete exposition of the untenable character of the hypo- 
 thesis tending to discredit the result of my labors, I will 
 present without abridgment the essential points of his com- 
 munication to the French Academy, jjublished in " Comptes 
 Eendus," Tome LXXVIII., No. 11 : " During last summer 
 (1873) I made some experiments in order to determine the 
 relation of the radiation of the sun to that of the electric 
 light, in the hope of solving the question of solar tempera- 
 ture. This source of light was selected, because its inten- 
 sity differs the least from that of the sun. Hence I expect 
 to harmonize the conflicting opinions regarding the law of 
 radiation existing among the followers of Kewton and those 
 of Dulong and Petit. In estimating the t^vo radiations I have 
 used the thernioheliometre, the same apparatus described in 
 my work ' Le Soleil.' This instrument, in spite of the objec- 
 tions made to it [by the wiiter of this work], seems to me 
 appropriate, particularly for determining mere diiferences, as 
 in this case. Let I^ and I,, be the absolute intensities of 
 the radiations of the sun and of the charcoal points ; 0, and 
 Be the excess of temperature of the black thermometer above 
 that of the surrounding medium, in the cases of the solar and 
 the electric radiations ; a and 6 the apparent diameters of the 
 radiating surfaces, viewed from the centre of the black thermo- 
 
 d, I, tang.' 6 6, tang.' a 
 
 meter, and we have — = ~ ^^^ whence 1, = 1^ -— rz • 
 
 ' d^ I, tang.- a t^c. tang. 6 
 
 It is very difficult to determine practically the radiating 
 
CHAP. X. TEMrERATUliK OF TlIK SOLA I; SURFACE. 105 
 
 surface of the charcoal points. The point is generally very 
 lirilliant, but beyond this the incandescence decreases very 
 lapidly; besides, the arc between them has a very different 
 I'adiation. We have tried to determine the sui-face of the 
 radiating parts of the charcoal points 1iy comparing their 
 dimensions with those of glass tul)es placed in their imme- 
 diate vicinity, and estimating the distance at which a thin 
 wire of platinum commenced to melt without touching tliera. 
 We have thus obtained an almost rectangular surface, equal 
 to that of a circle of 1 centimetre in diameter; besides, the 
 radiation from the parts outside of this limit was intercepted 
 by diaphragms. The pile consisted of 50 elements (Bunseu) 
 immersed in fresh nitric acid. The diameter of the ele- 
 ments was n".12 and their height (i°.20. The electrodes were 
 short and very tliii-k ; the current was so intense that the 
 insulating plates of an apparatus of Foucault were fused 
 almost immediately, and an iron wire of 1 millimetre in dia- 
 meter and 2".50 of length \vas constantly kept at "white- 
 heat." The author having explained that these data are 
 vague, proceeds : " Having placed the thermoheliometre at 
 a level with the charcoal points and the black thermometer 
 at a distance of 0".395, I found after half an hour a dif- 
 ference of 3°.03 between the temperature of the surrounding 
 medium and the lilack thermometer. During sevei'al days of 
 July, about noon, I determined, with the same instrument, 
 the temperature produced by solar radiation. I found a dif- 
 ference of IT^.ST, allowing for zenith distance. By sul)stitut- 
 ing these amounts in the previous fornuda and calcidatiug 
 
196 BADIANT HEAT. CHAP. X. 
 
 the diameters a and d according to the dimensions and dis- 
 tances of the radiating atmosphere, ^ve ha^■e I, = I^ X 36.468, 
 making the intensity of solar radiation ;36A times greater than 
 that of the charcoal points. This estimate, however, falls 
 short of the actual temperature ; for we know that the cor- 
 rection for atmosjiheric absorption is too small. Mr. Soret 
 has found on the Mont Blanc 21°.18 ; at the upper limit of 
 our atmosphere that amount would probably be about 27°. 
 These two amounts would give respectively : For 21°.13 : I^ 
 = I„ X 44.36 ; for 27°.00 : I, = I, X 56.66. 
 
 These results differ materially from those obtained by 
 other observers. Apj^rehending that some unknown cause in 
 my electi'ic light might produce an excessive error, I com- 
 jjared it with the light from a stearin candle. I found that 
 it equalled 1,450 common candles, showing the intensity of 
 an ordinary good pile. In another series of expei'iments, in- 
 stituted when the pile had worked for some time, I found 
 Ij = lo X 47.5, Avhich result differs very little from that 
 obtained by adopting the temperature of 21°. 13 observed by 
 Mr. Soret. Thus, if we accept this temperature of 21°. 13, 
 which is incontestably below the real one, and supposing the 
 temperature of the radiating surface of the charcoal points to 
 be 3,000° — an amount by no means exaggerated, since that 
 ]5art of the platin^un exposed to the heat was fused — and if 
 we estimate radiation as proportional to temperature, \ve 
 obtain 133,780° as the potential temperature of the sun. 
 This amount may be raised even to 169,980°, by adopting 
 the temperature of 27° produced by solar radiation." 
 
CHAP. X. TEMPEBATUIiE OF THE SOLAR SURFACE. 197 
 
 The result, then, of Pore Seechi's Latest researches shows 
 
 that the potential temperature of the sun is 1;{3,78U' C, which 
 
 he thinks may l)e raised even to 169,980° C. Accordingly, 
 
 , . , . , , . . 10,000,000° C. 
 
 Ills loriiier coiniJiitatitin ot solar inteiisitv was 
 
 '■ ■ 169,980" C. 
 
 = 59 times hii^du'r than his jiresent. Referring to Chap. 
 XIII., it will lie found that I'diiillet, whose estimate of solar 
 intensity at the boundary of the terrestrial atmosphere is 
 nearly identical with that which my actinometric observa- 
 tions have established, but who bases his computations on 
 Duloiig and Petit's erroneous formula, arrives at the con- 
 clusion that the temperature of the sun is from 1,461° to 
 1,761° C. (mean = 1,611° C.) M. Vicaire, adopting, like 
 Pouillet, Dulong's law, states, in his paper presented to the 
 Preiich Academy, that the temperature deduced from that 
 law is between 1,400° and 1,500° C. Sainte-Claire-Deville 
 concludes his essay on solar temperature by the announce- 
 ment that " solar temperature will not lie found far removed 
 from 2,500° to 2,800° C." It is very important to observe 
 that no difference of opinion exists regarding the fh/uatuic 
 inevfjij developed by the sun. All physicists accept Pouillet's 
 computation showing that each square foot of the solar sur- 
 face develops about 300,000 thermal units per minute. It will 
 be asked ho\v Pouillet could reconcile such an eiu)rmous deve- 
 lopment of energy with the insignificant intensity represented 
 by 1,611° C. A satisfactory answer will be found in Chap. II. 
 An examination of ^Ir. Poxc's tabic of temperatures in Chap. 
 XIII. will also suggest a satisfactory answer. This table 
 
lOR FABFANT HEAT. chap. x. 
 
 slioAvs that, agreeal)ly to Diiloiig's foi'imila, the radiant energy 
 of a body raised to a temperature of 2,520° F. is 4,600 times 
 greater than the radiant energy developed by a body raised 
 to a temjDerature of 60° F. above that of the atmosphere. It 
 is needless to enter into any further discussion showing that 
 the low solar temperature, 1,611° C, assumed by Pouillet, 
 results from the adoption of the enormous emissive power 
 of radiators announced by Dulong and Petit. It should be 
 borne in mind that no peculiar property has been attributed 
 to solar radiation, by Pouillet or other scientists, distinguish- 
 ing it from the radiation of such metallic substances as those 
 on which Dulong and Petit experimented. Consequently, if 
 we can show l)y j^i'actical test that the radiation of fluid 
 metals raised to a temperature of 1,611° C. develops only a 
 small fraction of the energy assigned by Dulong's formula 
 to that temperature, we prove conclusively that the method 
 is fallacious Avhich Pouillet and others have adopted in deter- 
 mining the temperatui'e of the solar surface. Now, the result 
 of the calorimetric measurement of the radiant energy of fused 
 and overheated iron, recorded in Chap. XIII., has established 
 in the most positive manner that the emissive power at a 
 temperature of 3,000° - 60 = 2,940° F. (1,633° C.) above that 
 of the atmosphere amounts to only 1,013 thermal units per 
 minute upon an area of one square foot. Pouillet's notions 
 of solar emission being based on the assumption that a tem- 
 perature of 1,011° C. is capable of developing 300,000 thermal 
 units, while actual trial shows that only 1,013 units are deve- 
 loped by the radiation of fused metal at even a liigher tem- 
 
OHAI'. X. TEMrKRATUliE OF I'UE HOLAU HVRFAL'E. I'.iO 
 
 pfiature, we are compelled to reject liis estiiuato of st>lar 
 temperature as wholly erroneous. 
 
 It sliould be observed that our precise determinatitiu of 
 solar energy by means of the calorimeter descriljed in C'liap. 
 V. shows that 322,000 thermal units are developed in una 
 minute by each square foot of the solar surface. Notwith- 
 standing this enormous development of dynamic energy, the 
 emissive power of the sun is relatively less than that of 
 fused cast iron ; a fact which tends to prove that the sun's 
 radiant heat emanates from incandescent gases. A brief ana- 
 lysis of this important matter will be appropriate in tliis 
 place. The radiant energy or emissive power developed by 
 the sun being 322,000 thermal units per minute upon each 
 square foot of surface, while boiling iron at a temperature 
 of 2,940° F. develops 1,013 units upon one square foot, 
 during one minute, it follows that the emissive power of 
 
 322 000 
 
 the sun is only '- = 318 times greater than that of 
 
 •' 1,013 
 
 iron at the stated temperature. But the temperature of the 
 
 . 4,030,000 
 solar surface being at least 4,036,000°, its intensity is ^ 
 
 = 1,373 times greater than that of boiling iron. Sir Isaac 
 
 Newton supposed it to be 2,000 times greater than that of 
 
 red-hot iron ; a i-emarkable agreement. The emissive power 
 
 1,373 
 of frised iron is consequently = 4.3 .tmies greater tlian 
 
 318 
 
 that of the solar surface at equal temperature. Now, there is 
 
 no terrestrial incandescent substance, whether solid or liquid. 
 
200 BADIANT HEAT. CHAP. X. 
 
 whose emissive power is not more than oiie-fourtli of that of 
 iron at oorrespoiicling temperatures ; lieiiee it is reasonable to 
 infer that solar radiation emanates from incandescent gases. 
 
 Tlie question of relative radiant power of solids and gases 
 having presented itself at the beginning of my investigations 
 of radiant heat, I constructed the apparatus illustrated on 
 Plate 20, in order to ascertain the temperature produced by 
 the i-adiatiou of incandescent gases. The illustration repre- 
 sents two vertical sections of the apparatus (see Figs. 2 and 3) 
 and a perspective view (see Fig, 1). Before entering on a 
 description, it will be proper to state that the device resorted 
 to was intended to produce a column of incandescent gas of 
 uniform density supplied with oxygen at every point within 
 the burning mass. This condition, it was supposed, could 
 only be fulfilled by employing a centrifugal blower forcing 
 a current of atmospheric air verticall}- upAvards through a 
 mass of easily-ignited combustibles, divided into small pieces, 
 placed on a horizontal grate. Fig. 1 represents a conical fur- 
 nace, 2)rovided with a grate applied at the contracted lower 
 portion, admitting of a free passage of the air between the 
 Ijars at every point. A capacious chamber is formed under 
 the grate. Into which air is forced by an ordinary centrifugal 
 blower. The internal portion of the furnace is contracted 
 towai'ds the top, as shown at h in Fig. 2, terminating with 
 a square opening, over which is placed a square trunk a, 
 corresponding exactly with the said opening. The furnace 
 being charged with combustibles which readily ignite, it 
 will be evident that a moderate speed of the blower will, 
 
CHAP. X. tempeeatvue of the solah svbface. 201 
 
 soon after ignitiou, fill the square trunk with a dense flamo 
 of perfectly uniform temperature throughout, contact with 
 the exterior atmosphere being wholly prevented, while the 
 air which supports the combustion is subdivided almost 
 inflnitely, and uniformly disjDersed, through the mass of 
 burning fuel. A chimney, the section of which is equal to 
 that of the contracted part of the furnace, being applied 
 above the square trunk, any tendency to pressure and accu- 
 mulation in the same will be effectually prevented. A dense 
 riame of uniform temperature having thus been obtained, its 
 radiant power has been ascertained by the folio-wing device : 
 A conical vessel h, open at the large end, surrounded Avith a 
 water-jacket of cylindrical form, shown in Fig. 2, is secured 
 to the square trunk, a circular opening e, shown in Fig. 3, 
 being ftu-med in the side of the latter, corresponding w^ith 
 the open end of the conical vessel. Referring to Fig. 2, it 
 will be seen that a perforated diaphragm d (composed of 
 polished silver) is inti-oduced near the small end of the 
 conical vessel. A thermometer is applied near the circular 
 perforation of the diaphragm, the bulb being placed exactly 
 in the centre line of the vessel. An opening /, siuTOunded 
 by a short conical tube, covered wnth a piece of mica, affords 
 a view of the interior of the conical vessel. The water- 
 jacket was supplied from the street-main, a constant stream 
 being kept up during experiments. The application of a 
 chimney of large diameter above the square flame-trunk, 
 and the covei-ing of the short conical tube with mica, as 
 stated, in order to prevent currents of heated air or gas 
 
202 EADIANT HEAT. CHAP. x. 
 
 from circulating tlirougla tlae conical vessel, have contribi:ted 
 to secure the desired result — viz., a disc of flame of uniform 
 
 r 
 
 brightness, the color varying with the speed of the blower. i 
 
 It might be supposed that the high temperature of the flame ij 
 
 would at once destroy the square trunlc. Such, ho^\"ever, is '^ 
 
 not the case, the trunk being made of plate-iron only iV in. ,j 
 
 thick, the radiation of which is so rapid that the gases com- 
 posing the flame cannot communicate the heat as fast as it 
 is carried ofE by external radiation. The top of the furnace 
 at the point where the flame is concentrated and conducted 
 into the sqi;are trunk, being exposed to intense heat, is lined 
 with fire-clay. It should be borne in mind that the apparatus 
 is exposed to a high temperature only while the blower is 
 in operation, the motion being stoj^ped as soon as the internal 
 thermometer reaches maximum indication. 
 
 It will be noticed by those Avho have paid attention to 
 the demonstration in Chap I. that, unless the radiant surface 
 forms a sj)herical concavity, the focus of which coincides 
 Avith the centre of the bulb of the recording thermometer, 
 the indication Avill not be exact. The flame-disc being ci/-- 
 cular, this objection may be overcome by removing the ther- 
 mometer from the flame to such a distance that the mean 
 length of the heat-rays directed to the bulb corresj)onds 
 with the radius of a concave radiator of the same diameter 
 as the flame-disc. For the sake of ready comparison, the 
 diameter of this disc and the focal distance of the record- 
 ing thermometer have the same relative propoi'tions as in 
 our solar pyrometer. 
 
CHAP. X. TEMPEBATUBE OF THE SOLAR SUEFACE. 203 
 
 The result of the iuitiary experiments with tlie ajipuratiis 
 thus described proved that the temperature transmitted to 
 the focal thermometer by the radiation of the flame-disc 
 ^vas relatively the same as in the solar pyrometer. It was 
 inferred from this fact that the radiant power of a dense 
 flame with active combustion kept up through its entire 
 mass is the same, for eqiial temperatures, as the radiant 
 power of metallic substances. Further investigation, how- 
 ever, disclosed the fact that this unexpected result -was 
 owing to the circumstance that the steam emanating from 
 the burning fuel during the experiment had entered the 
 conical vessel, and, reaching the bulb of the recording ther- 
 mometer, elevated the temperature very considerably. The 
 presence of steam had not been overlooked, but it was sup- 
 posed that the rapid circulation of cold water through tlie 
 jacket surroimding the conical chamber would produce in- 
 stant condensation. It is much to be regretted that before 
 the disturbing influence of the presence of steam in the 
 chamber containing the focal thermometer had been dis- 
 covered, the results of the initiary experiments had been 
 published in several mechanical journals. The main fea- 
 ture of the apparatus is, however, not without merit, as 
 probably no better method could be devised for produc- 
 ing a dense flame within which active combustion is being 
 kept up at every point. In combination with the thermo- 
 electric pile, should a reliable mode of calibration hereafter 
 be devised, there is reason to suppose that the illusti-ated 
 machine may prove very useful. In the meantime, I have 
 
304 BADTANT HEAT. chap. x. 
 
 instituted numerous experiments to ascertain tlie radiant 
 power of flames as compared with that of metallic sub- 
 stances. The result in every instance proves the feebleness 
 of the radiation of incandescent gases compared with incan- 
 descent solid substances of equal temperature ; but until the 
 conclusion of tlie investigation, definite statements must be 
 deferred. 
 
 Regarding the constitution of the solar surface, the tem- 
 perature of which Ave are now considering, the relative feeble- 
 ness of its emissive power compared with that of fused iron, 
 shown by our investigations, leads irresistibly to the conclu- 
 sion that we are dealing with an incandescent gas. The 
 constitution of the solar surface, however, has nothing to 
 do with its temperature. But, obviously, our endeavor to 
 ascertain the constitution of the sun will prove futile until 
 the temperature at its surface be first established. This will 
 be readily admitted. Suppose that the temperature of the 
 sun's surface is only 1,600° C, as Pouillet tells us, and that 
 the solar atmosphere extends to the moderate height of 
 100,000 miles above the photosphere. It may be shown, by 
 an easy calculation, that the specific gravity of the gases near 
 the solar surface Avould under these conditions, owing to the 
 low temperature, the depth of the superincumbent mass, and 
 the great attraction of the sun's mass, exceed tJiat of fused 
 iron. It will be evident, therefore, that, until the tempera- 
 ture of the surface shall have been established, investigations 
 relating to the constitution of the interior mass and the sur- 
 rounding atmosphere cannot lead to any safe conclusions. 
 
CHAP. X. TEMPEEATUEE OF THE SOL.IE SURFACE. 205 
 
 I now propose to slaow, by a demonstration whioli can- 
 not be objected to, that the high temperature established 
 by the indications of the sohxr pyrometer really exists on 
 the solar surface. Astronomers, while admitting their inabi- 
 lity to compute the degree of temperature imjiarted to the 
 surface of the planets of our solar system, OAving to the 
 unknown properties of their atmospheres in retaining the 
 heat received from the sun, have no hesitation in assigning 
 the exact degree of solar intensity transmitted to the atmo- 
 spheric boundary of each planet, compared with that ti'ans- 
 mitted to the boundary of our atmosphere. Sir John Herschel, 
 in treating of the planet Mercury, does not admit that any 
 donbt exists as to the relative degree of solar intensity to 
 which its atmosphere is subjected. The mean radius of the 
 orbit of this planet being to that of the earth as 38 : 100, 
 he tells us that the temperature produced by the sun's rays 
 on reaching the atmosphere of Mercury is nearly seven 
 times greater than the temperature produced by solar radi- 
 ation at the boundary of the terrestrial atmosphere. The 
 following extract from the "Outlines of Astronomy" sho^vs 
 the confidence which Sir John Herschel places in the appli- 
 cation of the law of' inverse squares to the determination 
 of solar energy at given distances : " The intensity of solar 
 radiation is nearly seven times greater on ^fercury than 
 on Earth, and on Neptune 900 times less ; the proportion 
 of the two extremes being that of upwards of 5,600 : 1. 
 Let any one figure to himself the condition of our globe 
 were the sun to be septupled, to say nothing of tlie greater 
 
20G BABIANT HEAT. chap. x. 
 
 ratio ! or were it diminislied to a seventli, or to a OOOth ! 
 It is true that, owing to the remarkable difference between 
 the properties of radiant heat as emitted from bodies of 
 very exalted temperature as the sun, and as from such as 
 we commonly teim tvai'in, it is very possible that a dense 
 atmosphere surrounding a planet, while allowing the excess 
 of solar heat to its surface, may oppose a powerful obstacle 
 to its escape, and that thus the feeble sunshine on a remote 
 planet .may be retained and accumulated on its surface." 
 
 No doubt Pouillet would have felt as little hesitation as 
 Herschel in determining by the application of the law of 
 inverse squares the temperature produced by solar radiation 
 on Mercury. Chapter I. contains an elaborate demonstration, 
 proving the correctness of this law with reference to radiat- 
 ing spheres uniformly heated at the surface. 
 
 The above diagram represents the orbits of the Earth, 
 Venus, and Mercury, their relative mean distance from the 
 sun being correctly drawn. The orbit of an imaginary body 
 I, revolving at a distance of 10,000,000 miles from the solar 
 
CHAP. X. TEMrEi:ATVliE-&F THE SOLAR SURFACE. 207 
 
 centre, has also been introduced In the diagram, for the pur- 
 pose of demonstrating that a body revolving round the sun 
 at that distance would be exposed to a temperature gieatly 
 exceeding that which Pouillet assigns to the solar suirface. 
 Astronomers agreeing that the law of inverse squares holds 
 for all distances, whether it be that of Neptune, which is 
 30 times further from the sun than the earth, or that of 
 Mercury, whose distance from the luminary is less than -bV of 
 that of Neptune, we are Avarranted in applying that law to 
 the body I, shown in our diagram, supposed to revolve at 
 a distance of ten millions of miles from the centre of the 
 sun. Let us then calculate what degree of solar intensity 
 this imaginary body will be subjected to. Our calculations, 
 obviously, must be based on the temperature produced by 
 solar radiation at the boundary of the earth's atmosphere ; 
 hence it will be necessary fii-st to establish that temperature. 
 Assuming that it will prove more satisfactory to persons 
 having obtained their knowledge from standard physical 
 works, I will leave out of sight the result of my own acti- 
 nometiic observations, and adojit those of Pouillet, the dif- 
 ference, besides, being quite unimportant. In his " Elements 
 de Physique," published at Paris, 1856, second volume, this 
 savant states, with reference to the amount of heat given 
 out by the sun : " In the vesical passage the atmosphere 
 absorbs at least 0.21 of the incidental heat, and at most 
 0.27, beyond which the sky ceases to be serene ; I should 
 add, however, that the 28th of June, on which day the 
 absorption ■was 0.27, a light white veil was perceptible in 
 
aOS BABIANT HEAT. chap. X. 
 
 the sky." The observed and commouly accepted maximum 
 solar intensity on the ecliptic being 68° R, it will be found, 
 by adding the loss of heat caused by atmospheric absorp- 
 tion — viz., 0.21, assumed by Pouillet — that the mean tempe- 
 rature produced by solar radiation at the boundary of the 
 atmosphere is fully 86° F. The mean distance of the earth 
 being in round numbers 92 millions of miles, while that of 
 the imaginary body I is 10 millions of miles, from the sun, 
 the solar intensity to which the latter -would be exposed is 
 to that transmitted to the boundary of the terrestrial atmo- 
 sphere as 10' : 92' = 1 : 84.6. Pouillet's calculations, and 
 common observations, having, as before stated, established 
 the intensity of the sun's rays on entering our atmosphere 
 to be nearly 86° F., the foregoing analogy proves that the 
 supposed body I will be subjected to a radiant intensity of 
 86° X 84.6 = 7,275° F. We have thus shown by a method 
 the correctness of which cannot be disproved that the radiant 
 heat emanating from the sun (a body the temperature of 
 Avhich, Pouillet informs us, is under 3,000° F.) develops an 
 intensity of 7,275° Fahrenheit at a distance of ten millions of 
 miles. Well-informed persons will not dispute the correctness 
 of the foregoing demonstration, nor ask further evidence of 
 the erroneous character of Secchi's recent speculations or the 
 fallacy of Zollner's and Pouillet's computations assigning a 
 low temperature to the solar surface. 
 
CHAPTER XI. 
 
 RADIATION FllOM INC.LNDESCENT PLANES. 
 
 Some eminent scientists have supposed that the surface 
 of an incandescent body projects rays of equal energy in 
 all directions. Laplace, having full confidence in the cor- 
 rectness of this assumption, founded upon it the demonstra- 
 tion adverted to in a previous chapter, proving that the 
 radiant energy which emanates from the receding surface 
 of the sun possesses greater intensity than that emanating 
 fi-om the central regions of the luminary. But actual obser- 
 vation having shown that the radiant energy from the sun's 
 border, so far from being more intense, is considerably less 
 than fi-om its centre, the persistent mathematician was driven 
 to the alternative of proving that the retardation produced 
 by the greater depth of the sun's atmosphere towards the limb 
 neutralizes the assumed increase of intensity of the radiant 
 heat. How satisfactorily the dexterous analyst proves the 
 startling proposition will be found on referring to " M^ca- 
 nique Celeste," Tome IV. pp. 284-288 : the result of his 
 
 a» 
 
210 BABIANT HEAT. chap. xi. 
 
 demonstration leading to tlie monstrous assumption that the 
 solar atmosphere absorbs H of the entire energy emanating 
 from the radiant surface. Evidently Laplace did not regard 
 solar radiation as molecular action ca^iable of being con- 
 verted into mechanical energy, or he would have perceived 
 the impossibility of H being absorbed by the solar atmo- 
 sphere. It is not intended to enter on a criticism of the 
 famous demonstration, but the question is so intimately con- 
 nected with the subject under consideration that a refei'ence 
 to the main points is called for, showing on what grounds 
 the conclusion was based that, but for the retardation pro- 
 duced by the solar atmosphere, the I'adiant energy of the 
 luminary would be increased towards the border. If we 
 admit the correctness of Laplace's assumption that the inten- 
 sity of radiation increases mth the obliquity of the radiant 
 surface and the increased number of rays contained in a 
 given section, we must also admit that the radiant energy 
 from the regions near the sun's border will be greatly en- 
 hanced. And since it has been found, by actual observation, 
 that no increase of intensity takes place, the inference cannot 
 be resisted that the retardation produced by the solar atmo- 
 sphere actually neutralizes the increased intensity occasioned 
 by obliquity. Accordingly, the retardation may be deter- 
 mined by calculating the increase of intensity corresponding 
 with the obliquity and consequent crowding of the rays. 
 But this calculation, it is evident, will not show the full 
 extent of retardation, since not only is there no inci'ease, 
 but a considerable diminyiion of intensity towards the sun's 
 
CHAP. XI. KADIATIOX Fi;O.M IXVAXDESCBXT PLAXES. 211 
 
 border. Hence, the amount of retardation determined agree- 
 ably to the doctrine tliat the radiant intensity is increased 
 by the obliquity of the rays will be still further augmented. 
 The reader will perceive, from this exi^osition, on Avhat erro- 
 neous grounds Laplace's enunciation is based, that "if the 
 sun were stripped of its atmosphere, it would appear twelve 
 times as luminous." 
 
 Tlie foregoing reference to doctrines promulgated nearly 
 a century ago, when solar radiation was but imperfectly 
 undei*stood, will be deemed inappropriate by those who do 
 not bear in mind that the highest authorities of the present 
 time advocate similar doctrines. Referring to Chapter VI., 
 it will be seen that Pfere Secchi, who has devoted more 
 time to the investigation of the subject than any one else, 
 presents calculations intended to prove that the retardation 
 offered by the solar atmosphere to the passage of the rays 
 is so great that only a fraction of the radiant heat enters 
 space. He sums xip his investigation by the following posi- 
 tive statement : " 1st. At the centre of the disc, perpendicu- 
 larly to the surface of the photosphere, the absorption arrests 
 about I, more exactly ^, of the total energy. 2d. The total 
 action of the absorbing envelope of the visible hemisphere 
 of the sun is so great that it allows only t^V of the entire 
 radiation to pass, the remainder — that is to say, -i-A — being 
 absorbed." Pei-sons accustomed to compare mechanical equi- 
 valents, especially those who possess practical knowledge of 
 the amount of mechanical power developed by the radiant 
 heat emitted by incandescent bodies at definite temperatures. 
 
312 RADIANT HEAT chap. xr. 
 
 positively reject all assumptions involving any considerable 
 loss of radiant energy by absorption in a medium perpetu- 
 ally exposed to the radiator. Nor will the assertion that 
 the radiant heat is converted into molecular motion within 
 the solar envelope be accepted by any j)ersou comprehend- 
 ing that the mechanical energy capable of being developed 
 by the heat-rays projected from the photosphere must enter 
 space less only the amount of actual toovTe performed during 
 the passage through that envelope. The investigations con- 
 ducted by means of the solar calorimeter described in Chap. 
 V. have shown that the dynamic energy develo^jed by the 
 sun's radiant heat on entering the earth's atmosphere amounts 
 to 7.11 thermal units per minute upon an area of one square 
 foot, while the dispersion of the rays is in the ratio of 1 : 45,400 ; 
 hence, each square foot of the photosphere emits, as shown 
 in the chapter referred to, 322,000 thermal units per minute. 
 Secchi says that only i of the heat emitted passes through 
 the sun's atmosj)here. Accordingly, 7 X 322,000 = 2,254,000 
 thermal units per minute are absorbed. Noav, the develop- 
 
 33,000 
 ment oi one horse-power requires = 42.7 units per 
 
 minute ; hence the energy supjiosed to be absorbed repre- 
 sents a mechanical force, continually acting, amounting to 
 
 2,254,000 
 
 — -— - — = 52,700 horse-power for each square foot of the 
 
 surface of the photosphere. Considering that the sun is sur- 
 rounded by highly attenuated gases, containing a very small 
 quantity of matter, Secchi's assumption that the stated enor- 
 
CHAP. XI. UABIATTOy FTtOM INCANDESCENT PLANES. 213 
 
 luous amount of energy is absorbed by tlie sun's atmosphere 
 is utterly at variance Anth tlie laws of mechanics. The fore- 
 . going discussion has been deemed appropriate in this place, 
 in order to t-all attention to the importance of ascertaining 
 the true energy of heat-rays projected from incandescent sur- 
 faces at acute angles. If we can prove by positive practical 
 means that the assumption is false which asserts that radia- 
 tors emit rays of equal energy in all directions, we destroy 
 the foundation on which the theory rests which has led to 
 the conclusion that only i of the energy developed by the 
 sun jienetrates its atmosphere. 
 
 The illustration on Plate 21, referred to in Chapter VI., 
 repi-osents a vertical section and top view of an inverted 
 conical vessel, the bottom of which is concave, the top 
 being open and pi'ovided with a wide flange. A revolv- 
 ing semi-sphi'rical disc of cast iron, flat on the under side, 
 is suspended on two transverse axles above the open end of 
 the conical vessel, the axles turning in appropriate bearings 
 resting on the top of the wide flange before mentioned. A 
 lever handle is secured to one of the transverse axles, for 
 the purjiose of placing the disc at any desired angle, the 
 degree of inclination l)eing indicated by a graduated quad- 
 rant applied as shown in the illustration. The conical vessel 
 is surrounded by a jacket, a stream of water being circulated 
 through the intervening space during experiments. The incan- 
 descent revolving disc is protected against loss of heat on 
 the top bv a non-conducting covering composed of fire-clay, 
 so arranged that it may be quickly applied and removed. A 
 
214 BABIANT BEAT. chap. xt. 
 
 semi-spherical water-jacket is a})plie(I aljove the revolving disc, 
 to protect the same from the disturbing influence of currents 
 ■ of air. It will be found, on examining the illustration, that 
 the water-jacket referred to is placed on the top flange of 
 the conical vessel, without fastening; hence it may be taken 
 away and replaced in a few seconds. The jacket surround- 
 ing the conical vessel being maintained at a constant tempe- 
 rature by a current of ^vater, the air in the lower part will 
 also be maintained at a constant temperature. Obviously, the 
 heated air at the top cannot descend to the bottom ; conse- 
 quently, the bulb of the recording thermometer will be influ- 
 enced only by the radiation of the surrounding vessel, and 
 by the radiant heat which the incandescent disc transmits. 
 It is hardly necessary to mention that the lower half of the 
 bulb is protected by a non-condiicting covering. In view of 
 the foregoing explanation, it will be evident that the mea- 
 surement of the intensity of the radiant heat projected from 
 the incandescent disc towards the bulb of the thermometer, 
 at different angles of inclination, will be as reliable as if the 
 air were exhausted from the conical vessel. In either case, 
 the temperature of the surrounding vessel which radiates 
 towards the thermometer, being deducted from the tempe- 
 rature indicated by the same, shows the intensity of the 
 radiant heat transmitted to the bulb. It may be contended 
 that the upper part of the latter loses a small amount of 
 heat by convection attending the presence of air ^vithin the 
 vessel. Assuming that the loss of heat from that cause is 
 appreciable, this loss will be proportionate to the intensity 
 
CHAP. XI. BADIATION FROM nCAXDESCEST PLANES. 215 
 
 of tlie lieat traiisiiiittecl during eacli experiment ; hence it 
 cannot affect the relative difference of intensity for different 
 degrees of inclination of the incandescent disc. 
 
 Dui'iug experiments, the apparatus is placed near an air- 
 furnace, hose being attached to the nozzles of the external 
 casing for circulating a constant stream of water through the 
 intervening space. The furnace having been charged with 
 combustibles capable of producing a steady fire, and heated 
 to the requisite degree, the disc is inserted. Having remained 
 in the furnace until the color of the metal approaches bright 
 orange, the disc is quickly withdrawn and placed over the 
 open conical vessel, supported by the axles shown in the top 
 view of the illustration. 
 
 Agreeably to the theory, the correctness of which we are 
 going to disprove, the incandescent disc, placed at the incli- 
 nation shown in the illustration, \\\\\ transmit a higher tem- 
 perature to the thermometer than if it were placed at a 
 greater angle to the vertical line ; the reasons assigned for 
 this assumption being that the same number of radiating 
 points are presented by the disc, and the same number of 
 rays of equal energy emitted in either position, while in the 
 former they are more concentrated than in the latter. The 
 stated assumption involves the proi^osition tliat parallel I'ays 
 projected at an acute angle, fi'om a given number of radiating 
 points, transmit greater intensity than an equal number t»f 
 parallel rays projected at a less acute angle to the radiant 
 surface. That this proposition, although untenable, is very 
 plausible, will be seen by reference to Fig. I (see diagram 
 
216 BADIANT BEAT. chap. xi. 
 
 Plate 22). Let a b represent the inclined radiant siu-face, 
 and a c h the several radiating points projecting heat-rays 
 towards the sjjaces d f and Tc g. The number of radiating 
 points and the number of heat-rays projected being alike in 
 each case, while the space represented by ^ ^ is only one- 
 third of that represented by d f, it must be admitted, if we 
 assume all rays to possess equal energy, that the concentra- 
 tion of heat within h g is, three times greater than within 
 d f. In other words, that a given area within h g receives 
 three times more heat than an equal area within d f. This 
 apparently correct view of the question, and its application 
 to spheres, led Laplace astray in his demonstration concern- 
 iug solar intensity. In the next chapter, which, as already 
 stated, will be devoted to the consideration of radiant heat 
 ti'ansmitted from incandescent spheres, the influence of the 
 spherical form on radiant intensity will be fully considered. 
 Li the meantime, we must admit that the demonstration con- 
 tained in Fig. 1 is unanswerable under the stijjulated condition 
 that all heat-rays emitted by a radiator possess equal energy. 
 Our task, therefore, will be to show, lyractically, that the stated 
 condition is based upon false assumptions, tiaviug already 
 made ourselves acquainted with the apparatus constructed for 
 this purpose, we may at once proceed to consider the results 
 of the experiments which have been instituted. It will be 
 evident that, owing to the high temperature of the revolv- 
 ing disc, it will cool very rapidly after being removed from 
 the furnace and placed in position over the conical vessel, 
 and that the recording thermometer, however sensitive, will 
 
CiiAi'. XI. BAVIATION FROM INCANDESCENT PLANES. 217 
 
 require so long a time before reacliiug maxiiuuiii iiulication 
 tliat only oiie inclination of the disc can be experimented on 
 at a time, thus rendering reheating indispensable for each 
 change of angle. The number of changes of inclination 
 during the investigation have, therefore, been limited to ten, 
 beginning -with 90 deg. and ending with 10 deg. inclination 
 to the vertic:il line. It will be evident that the high tem- 
 perature renders it practically impossible to impart exactly 
 the same degree of incandescence at each operation. I have, 
 thei-efore, resorted to the expedient of maintaining the furnace 
 at a uiiiforni temperature, and to expose the disc to the 
 action of the heat duiing an equal interval of time for each 
 operation. This method, though not precise, has conclusively 
 established the fact that the temperatiire transmitted to the 
 thei'uiometer by the radiant heat varies in the exact ratio 
 of the sines of the mean of the angles formed by the face 
 of the disc and a line drawn from its centre through the 
 centre of the bulb. The result of an experiment made with 
 great care will be found recorded by the diagram Fig. 5, in 
 which the oi'dinates of the curve a b represent the sines of 
 the angles formed by the disc and the lines mentioned, the 
 ordinates of the irregular line c d e representing the tempe- 
 ratui'o transmitted to tlie recording thernionietei'. The figures 
 inserted below the base line /' g show the number of degrees 
 of inclination corresponding with the sine represented by each 
 ordinate, while the figures above the curve a h sho\v the dis- 
 crepancy between the calculated and the actn;d temperature 
 transmitted to the thermometer. It will be found ku inspec- 
 
218 EABIANT HEAT. chap. XI. 
 
 tion that the mean diiference of the actual and the calcu- 
 lated temperature ahove the curve is 1.94°, that below the 
 same being 1.08° ; hence the mean discrepancy is only 0.86° 
 Fah. Considering the difficulty of imparting an equal tem- 
 perature at each operation during the experiments, this dis- 
 crepancy between the calculated and the actual temperature 
 transmitted by the radiation of the incandescent disc is unim- 
 portant. We are warranted, therefore, in adopting the con- 
 clusion that the temperatures vary exactly as the sines of 
 the angles of inclination of the radiant surface. It has been 
 deemed proper, in view of the great importance of this con- 
 clusion, and in order to render the subject clearly undei'stood, 
 to introduce Figs. 4 and 5 combined, showing the several 
 angular positions of the incandescent disc during the inves- 
 tigation. Dotted lines, it Avill be seen, have been introduced, 
 connecting these angular positions with the corresponding 
 ordinates of the curve a h. A mere glance at the geome- 
 trical representation contained in Figs. 4 and 5 will show 
 that the temperatures indicated by the ordinates of the curve 
 a h correspond exactly with the sines of the angles of incli- 
 nation of the disc. Bearing in mind the facts thus established, 
 let us again refer to Fig. 1, in which the space h g \s one- 
 third of the space d f. We are now enabled to demonstrate 
 that the heat transmitted to a given area within the former 
 is only one-third of the heat transmitted to an equal area 
 within the latter. Laplace and his followers, assuming the 
 reverse to be the case — viz., that the temperature within h g 
 will be three times higher than within d f — their estimate of 
 
CHAP. XI. RADIATION FliOM INCANDESCENT PLANES. 219 
 
 the radiant intensity of inclined surfaces will obviously be 
 too higb in tlie inverse ratio of the sines of angles of incli- 
 nation. The consequence of this grave mistake, with reference 
 to the radiant power of incandescent spherical bodies, will be 
 demonstrated in the next chapter, containing a record of the 
 temperatures developed by the heat-rays projected in a given 
 direction from different zones of a metallic sphere raised to 
 a hii^li decree of incandescence. 
 
CHAPTER XII. 
 
 RADIATION PROM INCANDESCENT SPHERES. 
 
 The question wlietlier equal areas at different points of 
 tlie solar surface transmit equal energy towards the earth 
 has engaged the attention of several eminent scientists. It 
 was mentioned in the previous chapter, on radiation from 
 inclined incandescent planes, that the author of " Mecanique 
 C61este," finding by observation that equal areas of the sun 
 do not transmit equal energies (the central portion transmit- 
 ting, in opposition to his reasoning, much greater intensity 
 than those near the border), explains the matter Ijy sho^ving 
 that the solar atmosphere retards the passage of the rays, 
 causing a great diminution of the energy of the radiant heat 
 projected from the border of the sun towards the earth. It 
 but seldom happens thfit questions of a cosmical nature admit 
 of being decided by actual experiment, the present being 
 one of the rare instances in which experimental tests may 
 be resorted to. Evidently, if the great retardation of energy 
 towards the border, demonstrated by Laplace, is caused solely 
 by the obstruction encountered during the passage of the rays 
 
 S30 
 
CHAP. xii. RADIATION FBOil INCANDESCE};! SrUEliES. 221 
 
 througL tlie atmosplieie suriouiuling the sun, the receding 
 surface of an incandescent spherical body not sxirrounded by 
 a retarding mediuni will transmit the supposed intensified 
 radiant heat undiminished. The illusti'ation on Plate 28 
 represents an apparatus by means of Avliich it has been 
 clearly demonstrated that, notwitlistanding the absence of 
 a retarding mediuiu round an incandescent sphere, the sup- 
 posed increase of radiant energy resulting from the obli- 
 quity of the heat-rays projected by the receding surface 
 does not take place. The said illustration shows a vertical 
 section and toj) view of a conical vessel sui'i'ounded 1)\ a 
 water-jacket, and in other respects consti'ucted as the appa- 
 ratus descril)ed in the preceding chapter. The top flange 
 of the conical vessel now under consideration is, however, 
 provided with a groove, the bottom of which supports a 
 solid sphere of cast iron, in the manner shown in the illus- 
 tration. Belo^v the sj)here are inserted two semi-cylindrical 
 screens of different diameter, each composed of two thin 
 plates of iron, the intervening space between these plates 
 being filled with a fire-proof non-conducting substance. It 
 will be seen, on carefully inspecting the illustration, that the 
 external screen is annular as Avell as semi-spherical, while the 
 central screen consists of a concave disc ; hence an annular 
 opening is formed ])etween each pair of screens. Supposing 
 the cast-iron sphere to be heated before being placed in the 
 position represented, it will be evident that the thermometer 
 at the bottom of the conical vessel will only receive the 
 radiant heat transmitted Ijy tlie heat-rays projected towards 
 
232 BABIANT HEAT. chap. xii. 
 
 the bulb throngli tlie anmilar opening formed bet^\'een the 
 two screens. It will be readily understood that, by employ- 
 ing screens of different proportions, zones containing equal 
 convex areas, but occupying different positions, may be made 
 to radiate towards the thermometer, and that by this means 
 the radiant intensity transmitted from any portion of the 
 spherical surface may be ascertained. Consequently, we are 
 enabled to test pi'actically the truth of the assertion that, 
 but for the intervention of the sun's atmosphere, the reced- 
 ing solar surface would, owing to the increased number of 
 rays contained within a given section, transmit au increased 
 radiant intensity towards the earth. It may be urged against 
 our device that atmospheric air intervenes between the incan- 
 descent sphere and the recording thermometer. A moment's 
 consideration, however, will show that the consequent retar- 
 dation is practically inappreciable. It has been established 
 in preceding chapters that the retardation sustained by the 
 sun's rays in passing through our atmosphere amounts to 
 0.207 on the ecliptic, while solar intensity at the boundary 
 of the terrestrial atmosphere is very nearly 85° F. Conse- 
 quently, the loss of radiant heat hardly reaches 18° F. in 
 passing through 28,800 feet of atmospheric air of maximum 
 density. The radiant heat of our experimental a|)paratus 
 being transmitted through a depth of only 2 feet, tlie retard- 
 ing influence of the air intervening between the ladiatiug 
 sphere and the bulb of the recording thermometer \vill be 
 
 2 X 18° 
 '^^^y ~^^o~oa77 ~ 0.0012° F. We may, therefore, without appre- 
 
CHAP. XII. EADIAl'lON FEOM IXCAXDESCENT SPHERES. 223 
 
 ciable error, assume that no retarding medium surrounds the 
 experimental incandescent sphere. The princi2)al features of 
 our apparatus having thus been exphiiued, and the nietliod of 
 solving the problem under consideration pointed out, we may 
 now proceed to consider the result of the experiments Avhich 
 have been instituted. In order to facilitate comparison, the 
 lower half of the sphere visible from the centre of the bulb 
 of the recording thermometer (see Fig. 6 in the diagram Plate 
 24) has been divided into four zones, A, B, C, and D, contain- 
 ing equal areas. It w^ill be seen, on inspecting the arrangement 
 of screens shown in the diagram, that no part of the surface 
 of the sphere excepting that contained within the parallel 
 lines defining each zone is capable of radiating towards the 
 thermometer, all the rest being shut out by the screens. Ob- 
 viously, the latter can be so proportioned that the radiant 
 heat from any part of the lower half of the sphere may be 
 projected towards the bulb. Figs. 3, 4, 5, and 6 in the dia- 
 gram show the arrangement of screens adopted in our expe- 
 riments, by means of which the transmitted radiant power 
 of each of the zones has been ascertained. The dimensions 
 of the several screens have been determined by drawing 
 radial lines from the centre of the bulb of the thermometer 
 to the points where the termination of the zones intei"sect 
 the circumference of the sphere. The subject will be most 
 readily understood by referring to Fig. 4, which exhibits 
 zone C. The screens being made to terminate \vhere they 
 meet the radial lines p, g and q, g, it will be seen that an 
 annular opening p q is formed, pennitting all heat-rays to 
 
m RADIANT HEAT. chap. xil. 
 
 pass wLicli are projected from the zone C in tlie direction 
 of the Inilb of the thermometer. A simihir arrangement 
 permits the radiant heat from zone B, in Fig. 5, to act on 
 tlie thermometer, lleferring to Fig. 3, it ^\•ill be found that 
 only one screen, perforated in the centre, is required to shut 
 out the radiant heat from the thi'ee upper zones, C, B, and 
 A ; while in Fig. 6 the radiation from the three lower zones, 
 D, C, and B, is shiit out by a single central screen, the cir- 
 cumference of which is defined by the radial lines m, I: It 
 should be borne in mind that, although the several screens 
 are represented by single lines in the diagram, they are, as 
 already explained, composed of double plates, a fire-proof 
 non-conducting substance being inserted between the two, 
 the object of which is self-evident. 
 
 Referring to the demonstration contained in the previous 
 chapter relating to the diminution of energy of heat-rays 
 projected at an acute angle to the radiant surface, it will be 
 seen, on mere inspection, that the upper zones represented in 
 <iur diagram, though containing an equal area with the lower 
 zones, cannot possibly transmit the same temperature as the 
 latter. The advocates of the views expressed in " Mecanique 
 Celeste " Avill learn with surprise that, notwithstanding the 
 absence of an intervening retarding medium, so great is the 
 difference of energy communicated that, while the zone D, 
 of the experimental incandescent sphere, transmits a tempe- 
 rature of 42°. 5 to the thermometer, the zone A transmits only 
 4°. 7. The latter zone being further from the thermometer 
 than the foiiuer, a correction is, however, necessary on account 
 
ciiAi". XII. EM'IATWN FROM IXCANDESCEST SPUEIiES. 225 
 
 of the iiicieaseJ dLspersiun of tlie Leat-rays before reaching 
 the bulb. This correction being made, the true ratio of tem- 
 perature transmitted by the zones D and A \\ ill be 42°.50 : 
 6°. 19. Consequently, the heat-rays projected fi'om the lower 
 zone of the incandescent sphere towards the bulb of the 
 thermometer transmit nearly seven times higher temperature 
 than the heat-rays iwnn the upper zone. The amount of 
 radiant surface being alike in each zone, while the tempei'a- 
 ture of the sphere is uniform throughout, it will be admitted 
 that owv practical test has clearly demonstrated the feeble- 
 ness of the he;it-rays projected from the border of an incan- 
 descent sphere towards a given point. It is hardly necessary 
 to add that each zone has called for a separate experiment, 
 rendering reheating of the sphere indispensable for each. 
 The same e.xpedieiit has, therefore, been resorted to, in order 
 to insure an ecpial degree of temperature during each experi- 
 ment, as in the case of the incandescent inclined disc described 
 in the previous chapter. Of course, it has been found iinprac- 
 ticalde to impart an ecpial temperature to the sphere at each 
 operation ; ])ut this difficulty has been satisfactorily overcome 
 by establishing a mean, as in determining the intensity of 
 the radiant heat transmitted by the inclined disc refenvd 
 to. Besides, the result has been checked by computing the 
 degree of temperature capable of being transmitted to the 
 recording therinometei' by each zone, in accordance with the 
 relation wliicli the intensities l^ear to the angles formed by 
 the radiating surface and the heat-rays projected towards the 
 centre of the bulb. Before giving an account of our exjie- 
 
226 EABIANT HEAT. chap. xii. 
 
 riments, let us demonstrate theoretically what temperature 
 each zone ought to communicate to the thermometer, in con- 
 formity with the fact established by the experiment recorded 
 in the preceding chapter, that the intensity of the radiant 
 heat transmitted by an incandescent disc is directly propor- 
 tional to the sines of the angles formed by the projected 
 heat-rays and the radiating surface. In order to simplif}^ 
 the demonstration, the several zones have been divided into 
 halves by dotted lines (see Fig. 7); radial lines being drawn 
 to the thermometer at z from the points of intersection of the 
 dotted lines referred to and the circumference of the sphere. 
 Tangential lines, (/ f, c ii, h x, and a y, have also been drawn 
 from the said points of intersection. It will be evident, on 
 considering the properties of spherical zones, that the radial 
 lines d z, c z, b z, and a z represent the mean direction of 
 the heat-rays projected by each zone respectively towards z. 
 Hence the sines of the angles t d z, to c z, x b z, and y a z 
 will determine the amount of radiant heat transmitted towards 
 z by each of the zones D, C, B, and A. Calculation shows 
 that if the sine of the angle t d z he represented by unity, 
 the sines of the other angles, in the order presented, will be 
 0.671, 0.384, and 0.121, while the experiments which have 
 been made show tliat the zone D transmits a temperature 
 of 42°.50 to the recording thermometer. Consequently, the 
 zones C, B, and A ought to transmit respectively 28 .50, 
 16°.31, and 5°, 16 to the thermometer at z. The accompany- 
 ing table shows to what extent the actual temperatures 
 transmitted by the incandescent sphere differ from the stated 
 
CHAP. XII. EADIATIOX FROM INCANDESCEUT SPHERES. 
 
 227 
 
 computed temperatures. It should be observed tliat no di- 
 rect comparison can be based upon tlie temperatures entered 
 in the fourth column, since the heat-rays projected by the 
 several zones are subjected to different degrees of dispersion, 
 owing to the unequal distance from the thermometer. Due 
 allowance being made for the dispersion of the rays, in con- 
 formity with the elements fiu-nished in Fig. 7, the consequent 
 augmentation of temperature has been added, and the cor- 
 rected values entered in the fifth column of the table. The 
 
 1 
 
 2 
 
 ■s 
 
 4 
 
 5 
 
 6 
 
 § 
 
 Mean angle of 
 projection. 
 
 Comparative 
 sine. 
 
 Obserrcd 
 temperature. 
 
 Corrected 
 temperature. 
 
 Computed 
 temperature. 
 
 Dei/, ^'n. 
 
 Proportion. 
 
 • Fah. 
 
 'Fah. 
 
 °Fah. 
 
 D 
 C 
 B 
 A 
 
 58 
 
 34 40 
 
 19 
 
 5 55 
 
 1.000 
 0.671 
 0.384 
 0.121 
 
 42.5 
 24.2 
 10.1 
 
 4.7 
 
 42.50 
 
 27.49 
 
 12.82 
 
 6.19 
 
 42.50 
 
 28.50 
 
 16.31 
 
 5.16 
 
 computed temperatures will be found in the sixth column. 
 It will be imagined, at first sight, that the figures entered in 
 the table indicate a serious discrepancy between the observed 
 and the computed temperature. That such is not the case 
 \n\\ be found on referring to Fig. 8, in which the ordinates 
 of the regular curve a h represent the computed temperatures, 
 while the ordinates of the irregular curve a d c represent 
 the observed temperatures. Obviously, the computed and 
 the observed energies transmitted by the radiation of the 
 
238 BABIANT HEAT. chap. xii. 
 
 incandescent sjiliere are tnily represented by the supei-ficies 
 contained between the base / g and the curves a h and a d c 
 respectively. Calculation shows that these superficies are as 
 1.000 to 0.945. Considering the small amount of this dis- 
 crepancy, in connection with the difliculty of bringing the 
 heated sphere to an equal degree of incandescence duiing 
 each experiment, Ave are warranted in asserting that the in- 
 stituted test has proved conclusive, and that the inaccuracy 
 of the doctrine promulgated in " Mecanique Celeste," regard- 
 ing the radiant energy transmitted by the rays projected from 
 the receding sm-face of an incandescent sphere, has been fully 
 demonstrated. 
 
CHAPTER XIII. 
 
 RADIATION FROM FUSED IRON. 
 
 The illustration on Piute 25 represents a calorimeter 
 originally constructed to demonstrate practically the fallacy 
 of the statements contained in certain papers read before 
 the Academy of Sciences at Paris by Messrs. Saiute-Claire- 
 Deville and M. E. Vicaire. These physicists assert that the 
 temperature of the solar surface does not exceed that pro- 
 duced by the combustion of organic substances. Their rea- 
 soning being based on the law of radiant heat established 
 by Dulong and Petit, I instituted, soon after the publication 
 of the papers referred to, a series of experiments on a very 
 large scale, in order to test thoroughly the correctness of 
 that law AN-ith reference to radiation at high temperatures. 
 The nature of these experiments will be seen by the fol- 
 lowing brief description : An iron vessel, lined with fii'e- 
 clay on the inside, was filled with fused cast-iron obtained 
 from a cupola furnace in wliicli the metal had been raised 
 to a temperature exceeding .'5,000' F. by the process of over- 
 
230 RADIANT HEAT. chap. xiii. 
 
 heating. On tlie surface of this fused mass, the weight of 
 which exceeded 7,000 pounds, the calorimeter represented 
 in the illustration was floated while registering the dynamic 
 energy developed by the radiant heat of the metal. Sir 
 Isaac Newton, -whose sagacity perceived that radiation ^vith 
 i-eference to mechanics is simply transmission of energy, 
 assumed that the quantity of heat lost or gained by a body 
 in a given time is j)i'oportional to the difference between 
 its temperature and that of the surrounding medium. Some 
 eminent scientists, however, accepting the conclusions and for- 
 mula of Dulong and Petit (see Chap. II.), assert positively 
 that the stated assumption is incorrect. The important fact 
 appears to have been overlooked that the investigations insti- 
 tuted by those experimentalists have in reality established 
 only the degree of conductivity of the radiators employed, 
 under certain conditions, but by no means their true radiant 
 energy at high temperatures. Sainte-Claire-Deville and M. E. 
 Vicaire, therefore, commit a serious mistake in assuming that 
 the quantity of heat transmitted by the radiation of incan- 
 descent bodies, at very high temj^eratures, has been determined 
 by theii' celebrated coimtrymen. The fact may properly be 
 adverted to in this place that the relation between the time 
 of cooling and the quantity of heat transmitted by I'adiation 
 which Dulong and Petit established, misled Pouillet regarding 
 the temperatui'e of the solar surface, which he computed at 
 l,4Gr C, or at most 1,761° C. It will be well to bear in 
 mind that Pouillet had himself ascertained with considerable 
 accuracy the temjaerature produced by solar radiation on the 
 
CRAP. XIII. EADIATIOS FEOM FUSED IliON. 231 
 
 surface of the eartli, aud also the retardation suffered during 
 tlie passage of the rays through the terrestrial atmosphere. 
 He was, therefore, able to demoustrate that the dynamic 
 energy developed by solar heat amounts to nearly 300,000 
 thermal units per minute for each square foot of the surface 
 of the sun. Considering the imperfect means employed by 
 Pouillet, his " pyrheliometer," the near approach to exactness 
 of his determination of solar energy is remarkable. 
 
 Temperature being a true index of molecular and mecha- 
 nical energy, conclusively established by the exact relation 
 bet\veen the degree of heat and the expansive force of per- 
 manent gases under constant volume, it is surprising that 
 Pouillet did not perceive that an intensity of 1,461° C. or 
 1,761° C. could not possibly develop on a single square foot 
 of surface the enormous energy represented by 300,000 ther- 
 mal units per minute. M. Vicaire, adopting, like Pouillet, 
 Dulong's formula, states in the paper presented to the French 
 Academy that " an increase of 600° is sufficient to increase 
 the radiation a hundredfold," and that Pouillet has verified 
 Dulong's law to more than 1,000°. " Supposing," he observes, 
 "that beyond this temperature the law ceases to be true, it 
 cannot be absolutely remote from the tnith for the tempera- 
 tures of from 1,400° to 1,500°, \\hioh we deduce by adopting 
 the law." Saiute-Claire-Deville concludes his essay on solar 
 temperature thus : " In accordance with my fii-st estimate, I 
 believe that this temperature will not be found far removed 
 from 2,500° to 2,800°, the numbers which result from the 
 experiments of M. Bunsen aud those published long ago by 
 
232 BABIANT HEAT. chap. xiii. 
 
 M. Debray and myself." The Frencli scientists tlien agree 
 that the temperature of the surface of the sun does not 
 exceed the intensity produced by the combustion of organic 
 substances, tlieir grounds for this assumption being, as we 
 have seen, Dulong's formula relating to the velocity of cool- 
 ing at high temperatures. But Dulong and Petit, apart from 
 the imperfections of the means which they adopted, did not 
 carry tlieir investigations practically beyond the temperature 
 of boiling mercury. It will be seen, on reference to Chap. 
 II., that their formula relating to high temperatures is mere 
 theory, the unsoundness of which the investigation now under 
 consideration has established in the most conclusive manner 
 by the insignificance of the radiant power developed by a 
 mass of fused metal presenting an area of 900 superficial 
 inches, 30 inches deep, raised to a temperature of 3,000° F. 
 Before describing the instrument employed in determining 
 the radiant energy developed by the stated unprecedentedly 
 large mass, I deem it impoi-tant to point out briefly the con- 
 dition of the fluid metal during the experiments. In the first 
 place, the temperature was sufficiently high to produce an 
 intense white light, luminous rays of great brilliancy being- 
 emitted by the radiant surface during the trial ; (2) the bulk 
 of the fused mass being adequate, the intensity of radiation 
 was sustained ^\•itllout appreciable diniinutic:)n during the time 
 required for observation. The temperature being higher than 
 that which the French investigators assign to the sui-face of 
 the sun, Avliile the bulk, as stated, was sufficient to maintain 
 the temperature of the fused mass, it may reasonably be 
 
CHA r. XIII. EA DIA HON FliO.U FUSED IRON. 833 
 
 asked why an area of one square foot of our experimental 
 lumiuous radiator should not emit as much heat in a given 
 time as an oqual area of tlie solar suiface, if the tempera- 
 tui-e of the latter he that assumed by Pouillet ? It is hardly 
 necessary to observe that an increase of tlie dimensions of 
 the radiating mass of metal to ,any extent whatever could 
 not augment the intensity or add to the dynamic energy 
 developed by a given area. It has been shown in previous • 
 chapters that, agreeably to Dulong's erroneous formula, the 
 emissive power of a metallic radiator raised to a temperature 
 of 3,000° F. considerably exceeds Pouillet's estimate of solar 
 emission. 
 
 Let us now briefly examine the illustrated calorimeter, 
 constructed for asceiiaining the mechanical energy developed 
 by the radiation of a fused mass of cast-ii'on raised to the 
 temperature of 3,000° F. Fig. 1 represents a vertical section, 
 and Fig. 2 a perspective view ; a is a cylindrical boiler, 
 having a flat bottom, composed of thin sheet-iron 0.012 in. 
 thick, coated witli lamp-black. The vertical part of this 
 boiler is surrounded by a concentric casing Z», the interven- 
 ing space being filled with a fii'e-proof non-conducting sub- 
 stance. A horizontal wheel revolving on a vertical axle f7, 
 and provided with si.\ radial paddles attached to a perforated 
 disc c c, is applied within the boiler. An open cylindrical 
 trunk g is secured to the perforated disc which supports 
 the paddles. The vertical axle passes through the top of 
 the lioiler, a conical pinion being secured to its upper termi- 
 nation. By means of a cog-wheel h, attached to the hoi-izontal 
 
234 BABIANT HEAT. OHAP. xin. 
 
 axle Ic, and geared into the conical jsinion, rotary motion is 
 communicated to tlie paddles. The centrifugal action of tlie 
 latter will obviously cause a rapid and uniform circulation of 
 tlie water contained in tlie boiler — indispensable to prevent 
 tlie intense radiant lieat of the fused metal from burning 
 the bottom. The boiler and mechanism thus described are 
 secured to a raft I I, composed of fire-bricks floating on the 
 top of the fluid metal. By this means it has been found 
 practicable to keep the bottom of the boiler at a given dis- 
 tance, very near the surface of the fused mass, while, by 
 moving the raft from point to point during the observation, 
 irregular heating, resulting from the reduction of tempera- 
 ture of the surface of the metal under the bottom of the 
 calorimeter, has been prevented. The radiant heat emanat- 
 ing from such a large body of fused metal being too intense 
 to admit of the axle k being turned dii-ectly by hand, an 
 intervening shaft of considerable length, provided with a 
 crank-handle at the outer end, has been employed for keep- 
 ing up the rotation of the paddle-wheel during the trial. It 
 is scarcely necessary to mention that the intervening shaft 
 should be coupled to the gear-work by means of a "uni- 
 versal joint," to admit of the necessary movement of the 
 raft from point to point on the surface of the liquid metal. 
 The experiment, repeated several times, has been conducted 
 in accordance with the following programme : The boiler 
 being charged with pure water, the paddle-wheel should be 
 turned at a moderate speed while observing the temperature 
 of the fluid, the thermometer employed for this purpose being 
 
CHAP. xni. RADIATION FliOM FUSED IRON. 235 
 
 introduced through an opening m at the top of the boiler. 
 The temperature being ascertained, the calorimeter should 
 be placed on the raft as quickly as possible, and the time 
 noted. As soon as vapor is observed to escape through the 
 opening at m, the instninient must be instantly removed, the 
 time again noted, and the temperature of the ^vater in the 
 boiler ascertained. It will be well to keep the paddle-wheel 
 in motion until the last observation has been concluded. 
 
 The temperature of the fused metal having been as high 
 during our experiments as that of the solar surface, accord- 
 ing to the computation of Pouillet and his followers/ while 
 the thin substance composing the bottom of the calorimeter 
 has been brought almost in contact with, and consequently 
 received the whole energy transmitted by, the radiant sur- 
 face, the reader will be anxious to learn what amount of 
 dynamic energy has been developed by the radiation of the 
 metal in a given time on a certain area. The desired infor- 
 mation is contained in the following brief statement : Allow- 
 ance being made for heat absorbed by the materials composing 
 the paddle-wheel, etc., the instituted test shows that the tem- 
 perature of a quantity of water weighing 10 lbs. avoirdupois 
 has been elevated 121° F. in 164 seconds (2.73 min.), the area 
 exposed to the radiant heat being 63 sq. in. Hence a dynamic 
 
 10 X 121 IM , ^_ , , ., . - 
 
 enerev X = 1,013 tliennal units per min. has 
 
 ^^ 2.73 63 '■ 
 
 been developed by the radiation from 1 sq. ft. of the surface 
 
 of the fused metal maintained at 3,000° F., against 300,000 
 
 units developed by tlie radiation of 1 sq. ft. of the solar sur- 
 
336 BABIANT HEAT. chap. xiii. 
 
 face, the temperature of Avliicli, agreeal^ly to the calculations 
 of the Frencli physicists, is less than that of our experimental 
 radiator. Comment is iinnecessarj. 
 
 Some advocates of Dulong and Petit's theory explain the 
 enormous discrepancy which our investigation of the radiant 
 power of fused metal discloses, by showing that their law 
 I'elates to the velocity of cooliug, and not to the amount of 
 dynamic energy parted Avith. That Pouillet regards the law 
 as referring to dynamic energy is evident, or he would not 
 have attempted to establish by computations based upon it 
 that an energy represented by 300,000 thermal units could 
 be developed in one minute upon an area of one square foot 
 by a body whose temperature is under 1,700° C. Before 
 concluding our discussion, let us consider how Dulong and 
 Petit's law is regarded by practical engineers, who are more 
 interested in its infallibility than students of natural philo- 
 sophy. Mr. Box, in his " Practical Treatise on Heat," pub- 
 lished 1868, says, with reference to Dulong and Petit's 
 investigations, in which he evidently places full confidence : 
 " Dulong has given rules which agree well with experi- 
 ments up to a difference of temperature of 468° F. This 
 rule is a very difficult one to apply, but it may be put in 
 such a form as to give a ralio by which calculation by the 
 simple rule may be easily corrected. The rule thus becomes : 
 
 124.72 X 1.0077* X (1.0077' - 1) ^ . 
 
 — = R", in which t = the 
 
 temperature of the absorbent, or recipient of radiant heat ; 
 T = the excess of temperature of the radiating body in degrees 
 
cn.\i'. xiri. 
 
 liADIATIOX FBOM FUSED IJIOX. 
 
 237 
 
 Centigrade ; and E," = the i-jitio <>f loss of heat under the 
 given temperatures." The author of tlie " Practical Treatise 
 on Heat" then proceeds to construct a table of temperature 
 and ratio of loss by cooling ; but before presenting the same 
 
 Ml!. Box's Taulh ov tuk Katio of Loss of Hf.at at very Higu 
 Tempek.vtures, by the Formula of Dllon'g. 
 
 Temperatui-e of the heated body. 
 
 li. 
 
 S S l^ 
 
 Temperature of 
 
 the body 
 
 above that of 
 
 the air. 
 
 Ratio of heat at diffe- 
 rent temperatures. 
 
 RadiatioD. 
 
 Contact of 
 air. 
 
 'Fah. 
 
 'Fah 
 
 •FoA. 
 
 Satio. 
 
 BaUo. 
 
 490 
 
 60 
 
 60 
 60 
 00 
 60 
 60 
 60 
 00 
 60 
 60 
 60 
 
 450 
 
 540 
 
 720 
 
 840 
 
 1080 
 
 1260 
 
 1440 
 
 1620 
 
 1800 
 
 2160 
 
 2520 
 
 3.10 
 4.19 
 
 7.17 
 12.68 
 23.01 
 42.70 
 80.67 
 154.5 
 . 299.7 
 1159.0 
 4604.0 
 
 1.980 
 2.085 
 2.230 
 2.378 
 2.450 
 2.540 
 2.620 
 2.693 
 2.760 
 2.880 
 2.985 
 
 600 
 
 780 
 
 900 Red, just visible 
 
 1140 " " 
 
 1320 Dull red 
 
 1600 Dull cherry red 
 
 1680 Cherry red 
 
 1860 Clear red 
 
 2220 Clear orange 
 
 2580 White, bright 
 
 
 to his readers he prefi.xes the following observation : " This 
 table shows that with a radiant body at a clear red-heat of 
 1,800° the loss is about 300 1 times the amount due by the 
 simple formula, and at a l^right white-heat of 2,580 it rises 
 
238 BADIANT HEAT. chap. xiii. 
 
 to 4,604 ! ! times tliat amount." If any doubt existed on the 
 subject, tlie author's emphatic exchimations furnish unques- 
 tionable evidence that he is not aware of the fact that he is 
 propagating a mischievous doctrine, and that he regards the 
 stated extraordinary ratios as true measures of the amount 
 of dynamic energy parted with at high temperatures. 
 
 The fallacy of Dulong and Petit's formula relating to high 
 temjjeratures having been conclusively demonstrated in Chap. 
 II., I have deemed it unnecessary to examine the calculations 
 based on that formula contained in the papers presented by 
 Messrs. Sainte-Claire-Deville and M. E. Vicaire to the Aca- 
 demy of Sciences at Paris, referred to at the commencement 
 of this chapter. 
 
CHAPTER XIY. 
 
 RADIANT HEAT MEASUKED BY THE THERMO-ELECTRIC 
 METHOD. 
 
 Melloni asserts, in " La Tberinoclirose," that the calorific 
 energies imparted to a thermopile are as the arcs through 
 which the needle of the galvanometer sweeps, until the deflec- 
 tion exceeds 13 degrees. This assumption being at variance 
 with the principles of dynamics, its correctness calls for a 
 thorough investigation before it can be accepted. Intending 
 originally to employ the thermo-electric method for ascer- 
 taining the difference of the radiant energy transmitted by 
 the sun's rays from diiierent portions of the solar disc, I 
 carefully investigated the subject, and found, by experi- 
 mental test, that Melloni's law is not correct. Theoretical 
 demonstration pointed to the fact that, for deflections not 
 exceeding 15 deg., the calorific energy impai-ted to the pile 
 by radiant heat is very nearly as the square root of the 
 versed sine of the angle of deflection from zero. It may 
 be briefly stated that, having previously resorted to various 
 
240 
 
 BADIANT HEAT. 
 
 expedients for testing roughly tlie relicability of the assump- 
 tion that the energy is as the arc up to thirteen deg. deflec- 
 tion — the result of the test in each instance proving decidedly 
 unfavorable to Melloni's doctrine — I undertook the construc- 
 tion of a special ajjparatus for calibrating the galvanometer 
 applied to my thermopile. By means of this apparatus the 
 energy developed for different deflections of the needle from 
 zero to 35 deg. has been accurately determined. Before 
 describing the new device, it will be proper to examine 
 
 Melloni's method of calibrating galvanometers, described in 
 the work referred to; especiall}- since its supposed correct- 
 ness has induced several eminent physicists to accept the 
 assumption that the energies are as the arcs swept by the 
 needle from zero to 13 deg. deflection. " Two vessels V V 
 (see Fig. 17) are half filled -with quicksilver, and connected 
 by two short wires, separately, with the terminations G G 
 of the galvanometer. The vessels and wires, arranged as 
 shown, will not change the action of the instrument ; the 
 thermo-electric current between the pile and the galvano- 
 
CHAP. XIV. TUEBMO-ELEVTinc 21EASUEEMENT. 241 
 
 meter being freely kept up as before. But if we establish 
 a communication between the two vessels by means of the 
 wire F, a portion of the current will pass through this wire 
 and then return to the pile. The quantity of circulating elec- 
 tricity in the galvanometer will then be diminished, while 
 the deflection of the needle will be reduced. Suppose that 
 by this expedient we have diminished the galvanometric 
 deviation to one-fourth or one-fifth — viz., that the needle 
 indicating 10 or 12 degrees, by the power of a constant 
 source of heat located at a given distance from the pile, 
 recedes 2 or 3 degrees when part of the current is diverted 
 by the outside wire. If we then cause the source of heat 
 to act at various distances, and observe in each case the 
 inaxinuini deflection and the least deflection, we obtain the 
 necessary data for determining the ratio between the deflec- 
 tion of the needle and the energy causing that deflection. 
 To make the matter better understood, and to give at the 
 same time an example of the manner of operating, let us 
 take the numbers bearing on the application of the method 
 to one of the thermo-multipliers. Let the outside circuit be 
 interrupted, and the source of heat located at an adequate 
 distance fi'om the pile to deflect the needle not moi-e than 
 5 degrees. The wii'e being then passed from V to V, the 
 needle falls to 1.5; the connection between the vessels being 
 again interrupted, and the source of heat placed near enough 
 to produce the following deflections in succession : 5°, 10°, 15°, 
 20°, 25°, 30°, 35°, 40°, 45°. Applying the same wire between 
 V and V after each deflection, we obtain the followiua: ener- 
 
243 EADIAKT MEAT. chap. xiv. 
 
 gles: 1°.5, 3°, 4°.5, 6°.3, 8°.4, 11°.2, 15°.3, 22°.7, 29°.7. Sup- 
 posing the energy equal to unity Avhicli is necessary to cause 
 tlie needle to describe arcs corresponding with each of the 
 first degrees of the galvanometer, we then have the number 
 5 as an expression of the energy corresponding to the initial 
 observation. The other energies are readily ascertained by 
 
 5 
 
 the relations : 1.5 : 5 = a? = — a = 3.333, where a indicates 
 
 lo 
 
 the deflection when the outside circuit is closed. Obviously, 
 any diminished current is to the total current to which it 
 corresponds as any other diminished current to its corre- 
 sponding total current. Hence, 5, 10, 15.2, 21, 28, and 37.3 
 are the energies corresponding to the deflections 5°, 10°, 15°, 
 20°, 25°, and 30°. In this presentation it will be seen that 
 the energies are nearly proportional to the arcs up to about 
 15 degrees ; but beyond this deflection the proportionability is 
 at an end, the discrepancy augmenting with the arcs." The 
 energies at intermediate degrees, it is stated, are readily ascer- 
 tained by calculation or by the graphic method, the latter 
 being assumed to be sufficiently precise for the purpose. The 
 accompanying table exhibits the corresponding deflections and 
 energies determined by Melloni in accordance with the fore- 
 going demonstration : 
 
 The constructor of the table observes that no notice has 
 been taken of deflections under 13 deg., since the energies 
 within that deflection are, as he supposes, correctly repre- 
 sented by the arcs swept by the needle. It is hardly neces- 
 sary to call attention to the unsatisfactory nature of the fore- 
 
ni.vi'. XIV. 
 
 TllKUMO-ELECTlilC ME. 1 6' L HEMES T. 
 
 243 
 
 going iiit'tliod of leturinug part of the electi'ic current by the 
 -w'nv F, fur the purpose of ascertaining the calorific energies 
 imparted to the pile by the radiant heat emanating I'roin 
 the radiator. Unless we adopt some ]iositive means of mea- 
 suring the intensity of the heat to which the face of the 
 pile is subjected at the instant of observing the deflection 
 
 Melloni's Table, showing the Relation betweek 
 
 Deflection 
 
 AND Energy. 
 
 
 u 
 
 Energy ex- 
 erted. 
 
 
 Energy ex- 
 erted. 
 
 Deflection of 
 needle. 
 
 
 13.0 
 
 13.0 
 
 19.0 
 
 19.8 1 25.0 
 
 28.0 
 
 14.0 
 
 14.1 
 
 20.0 
 
 21.0 1 26.0 
 
 29.7 
 
 15.0 
 
 15.2 
 
 21.0 
 
 22.3 
 
 27.0 
 
 31.5 
 
 16.0 
 
 16.3 
 
 22.0 
 
 23.5 
 
 28.0 
 
 33.4 
 
 17.0 
 
 17.4 
 
 23.0 
 
 24.9 
 
 29.0 
 
 35.3 
 
 18.0 
 
 18.G 
 
 24.0 
 
 26.4 
 
 30.0 
 
 37.3 
 
 of the needle of the galvanometer, the relation of deflection 
 and calorific energy cannot be accurately determined. Now, 
 the demon-stration contained in C'liap. I. proves that the inten- 
 sity of radiant heat transmitted through a given space by a 
 circular radiatoi' of known diameter and temperature may be 
 determined with positive accuracy. Accordingly, the method 
 of calibrating galvancmietei-s, which I am going to lay before 
 the reader, is based on the stated demonstration i:)roving that 
 a correct knowledge of form, distance, and temperature of 
 
244 EADIANT HEAT. 
 
 CHAP. XIV. 
 
 the radiator enables us to ascertain, with absolute precision, 
 the degree of calorific energy imparted to the thermo-electric 
 pile during the investigation. The following brief descrip- 
 tion of the apparatus represented by our illustration (see PI. 
 26) will suffice to give a clear idea of the same : t, table 
 having a longitudinal parallel groove, 6 ins. wide, 2 ins. deep, 
 formed on the top. h, sliding wooden block, 12 ins. square, 
 1^ ins. thick, provided with a j)arallel projection below cor- 
 responding Avith the groove in the table, and admitting of 
 the block sliding freely from end to end. A vertical plate 
 s, 20 ins. high, 12 ins. wide, is secured to the sliding block. 
 p represents a thermo-electric pile placed on the top of the 
 table. The vertical plate s is perforated in the centre, for 
 the purpose of supporting a cylindrical boiler r, -1 ins. in 
 diameter, provided with an open trunk on the tojj, through 
 which a thermometer is inserted. The end of this boiler 
 pointing towards the pile is concave, while the opposite end 
 is flat ; a spirit-lamp being applied ixnder the same, sup- 
 ported by the sliding block h. A scale, divided into 100 
 parts of one inch each, is attached to the side of the table, 
 the zero of this scale coinciding with a pei-pendicular line 
 draAvn from the face of the thermo-electric pile ^>. The 
 extreme j)oiut of the concavity of the boiler r being in line 
 Avith the front side of the plate s, while the zero of the 
 scale, as stated, is in line with the face of the pile, it -will 
 be seen that the distance through which the radiant heat 
 acts may be regulated by simply moving the sliding block 
 to any desired division on the scale. A metallic screen s', 
 
CHAi\ XIV. TUEUMO-ELECTRIC MEASUREMENT. 245 
 
 2(t ins. high, 12 ins. wide, platrd with polished silver, is 
 attaclied to the vei-tical plate s, in order to prevent the 
 radiant heat of the latter from acting on the thermo-electric 
 pile. The metallic screen is provided with a central perfo- 
 ration, 4 ins. in diameter, the centre of which coincides 
 with the prolongation of the axis of the cylindrical boiler. 
 Regarding the temperature of the lattei-, it will be seen that, 
 by applying the spirit-lamp, as already stated, the water may 
 be kept constantly at the boiling point, since any excess of 
 heat above that point will be carried off by the steam allowed 
 to escape through the open trunk which contains the thex'- 
 mometer. Accordingly, the thermo-electric pile j) will at all 
 times be subjected to a definite radiant intensity depending 
 on its distance from the radiator /■; while the graduated 
 scale attached to the side of the table t enables the experi- 
 menter to regulate that distance rapidly and accurately. In 
 accordance with the law governing the transmission of heat, 
 before referred to, the temperature imparted to the thermo- 
 electric pile p will bear the same relation to the differential 
 temperature of the boiler as the square of the radius of the 
 semi-spherical end of the latter bears to the square of the dis- 
 tance of the same from the face of the pile. Consequently, 
 when the boiler is jilaced as shown in the illustration, the 
 temperature transmitted by radiation may be ascertained by 
 the following calculation : Assuming that the thermometer 
 inserted through the open tinink at the top of the boiler 
 indicates 212° F., and that the tepmeratui'e of the surround- 
 ing air is 70 F., the intensity of the radiant heat emanating 
 
246 BABIAKT HEAT. chap. xiv. 
 
 from tLe boiler will be 212 - 70 = 142° F. Now, as the 
 position of tLe block b is siicli tliat the concavity r coincides 
 A\itli tlie aOtli division on the scale, tlie distance between tlie 
 face of tlie pile and the radiating surface will be 50 inches, 
 while the I'adius of the concavity is 2 inches. The tempe- 
 rature imjjarted by the radiation emanating from the boiler 
 
 2' X 142 
 will, therefore, amount to = 0°.227 F. The calorific 
 
 ' ' 50' 
 
 energy transmitted by the I'adiator at all other distances may 
 of course be determined by a similar process of computation. 
 Having theoretically determined the intensity of the radiant 
 heat for each division on the scale, the existing relation 
 between the deflection and the computed energy will be 
 ascertained simply by observing the corresponding position 
 of the needle of the galvanometer. Our task, however, is 
 that of ascertaining the calorific energy corresponding with 
 arcs not of varying length swept by the needle of the galva- 
 nometei', but arcs each measuring shs of a circle commencing 
 at the galvauometric zero. Evidently, arcs and energies are 
 not directly comparable ; hence, we must ascertain, experimen- 
 tally, what calorific energy corresponds with the first degree, 
 or the first half degree, of deflection of the needle from zero. 
 It has already been stated that the radiant energy emanating 
 from the concave face of the boiler is 142° F. when the tem- 
 perature of the surrounding air is 70° F. Hence, agreeably 
 to the foregoing process of calculation, the temperature im- 
 parted to the pile when the boiler is placed at the extreme 
 end of the scale — ^■iz., 100 inches from the face of the pile 
 
CHAP. XIV. THEUMO-ELECTEIC MEASUEEMEXT. 247 
 
 2' X 142 
 
 — will be — = 0°.057 F. Now, such are the propor- 
 
 100' ' ^ '■ 
 
 t'umii of tlic illu.^trated apparatus that, when the dilVereiitial 
 teiii2)eratiii-e nf tlie Ixiiler is 142° F. and the eoneave face 
 of the boiler coincides Avith the 100th division of the scale, 
 the detlection of the needle is 30' from zero. The forefjoins; 
 demonstrations and reasoning being deemed sufficiently expla- 
 natory, Ave may now consider the diagram attached to the 
 illustration, Plate 20. The length of the ordinates of the 
 curves hf and !> (/represent the relation between the energies 
 imparted to the pile and the arcs swept by the needle ; the 
 figures marked on the vertical base-line a c denoting the degrees 
 of deflection from zero. In other Avords, the ordinates of the 
 curve h d shoAv the observed deflections for each degree from 
 zero, while the oi'dinat(>s of the curve h f show the developed 
 energy. It Avill be seen, therefore, that the portions of the 
 ordinates Avhich are contained between the two ctii'ves lepre- 
 sent the excess of developed energy above that of the observed 
 deflection. Considering that the length of the ordinates of 
 the curve b f have been detenuined in accordance with the 
 well-established laws governing the transmission of radiant 
 heat, while the length of the ordinates of the curve h d is 
 the result of actual observation, the correctness of the ascer- 
 tained rehition cannot be qiiestioned. Conserpiently, a mere 
 inspection of the diagram suffices to show the fallacy of 
 Melloni's assumption that, Avithin a deflection of 13 degrees, 
 the arc swept by the needle and the energy imparted to the 
 pile correspond exactly. Obviously, by comparing the inter- 
 
348 
 
 IIABIANT HEAT. 
 
 CHAP. XIV. 
 
 '1 
 
 'able a, 
 
 snowiNfi 
 
 THE ReL 
 
 \TION BETWEEN 
 
 Energy 
 
 \ND 
 
 
 
 Deflection at Depinite Distances. 
 
 
 
 2- 
 
 
 .2 1 
 
 II 
 So 
 O 
 
 "o 
 
 ^1 
 
 i 
 
 a 
 
 a 
 o 
 
 If 
 
 II 
 go 
 
 Inches. 
 
 ' Fah. 
 
 Relative. 
 
 Beg. 
 
 IncTies. 
 
 ° Fah. 
 
 Relative. 
 
 Deg. 
 
 100 
 
 0.057 
 
 0.50 
 
 0.50 
 
 34 
 
 0.491 
 
 4.31 
 
 3.80 
 
 95 
 
 0.063 
 
 0.55 
 
 0.60 
 
 33 
 
 0.521 
 
 4.58 
 
 4.00 
 
 90 
 
 0.070 
 
 0.61 
 
 0.70 
 
 32 
 
 0.555 
 
 4.87 
 
 4.30 
 
 85 
 
 0.078 
 
 0.68 
 
 0.80 
 
 31 
 
 0.591 
 
 5.19 
 
 4.50 
 
 80 
 
 O.OSS 
 
 0.77 
 
 0.90 
 
 30 
 
 0.631 
 
 5.54 
 
 4.90 
 
 75 
 
 0.101 
 
 0.88 
 
 1.00 
 
 29 
 
 0.675 
 
 5.92 
 
 5.30 
 
 70 
 
 0.116 
 
 1.01 
 
 1.10 
 
 28 
 
 0.724 
 
 6.36 
 
 5.60 
 
 65 
 
 0.134 
 
 1.17 
 
 1.20 
 
 27 
 
 0.779 
 
 6.84 
 
 6.00 
 
 60 
 
 0.158 
 
 1.38 
 
 1.40 
 
 26 
 
 0.840 
 
 7.37 
 
 6.40 
 
 55 
 
 0.188 
 
 1.64 
 
 1.60 
 
 25 
 
 0.909 
 
 7.98 
 
 6.90 
 
 50 
 
 0.227 
 
 1.99 
 
 1.80 
 
 24 
 
 0.986 
 
 8.65 
 
 7.30 
 
 49 
 
 0.236 
 
 2.07 
 
 1.90 
 
 23 
 
 1.073 
 
 9.42 
 
 8.00 
 
 48 
 
 0.246 
 
 2.16 
 
 2.00 
 
 22 
 
 1.173 
 
 10.30 
 
 8.90 
 
 47 
 
 0.257 
 
 2.26 
 
 2.10 
 
 21 
 
 1.288 
 
 11.30 
 
 9.90 
 
 46 
 
 0.268 
 
 2.35 
 
 2.20 
 
 20 
 
 1.420 
 
 12.46 
 
 10.90 
 
 45 
 
 0.280 
 
 2.46 
 
 2.30 
 
 19 
 
 1.573 
 
 13.80 
 
 12.00 
 
 44 
 
 0.293 
 
 2.57 
 
 2.40 
 
 18 
 
 1.753 
 
 15.38 
 
 13.10 
 
 43 
 
 0.307 
 
 2.69 
 
 2.50 
 
 17 
 
 1.965 
 
 17.24 
 
 14.40 
 
 42 
 
 0.322 
 
 2.82 
 
 2.60 
 
 16 
 
 2.219 
 
 19.46 
 
 16.30 
 
 41 
 
 0.338 
 
 2.96 
 
 2.70 
 
 15 
 
 2.525 
 
 22.14 
 
 18.50 
 
 40 
 
 0.355 
 
 3.11 
 
 2.90 
 
 14 
 
 2.898 
 
 25.42 
 
 20.40 
 
 89 
 
 0.373 
 
 3.27 
 
 3.00 
 
 13 
 
 3.361 
 
 29.48 
 
 23.00 
 
 38 
 
 0.393 
 
 3.45 
 
 3.00 
 
 12 
 
 3.944 
 
 34.59 
 
 25.90 
 
 37 
 
 0.415 
 
 3.64 
 
 3.10 
 
 11 
 
 4.694 
 
 41.18 
 
 29.00 
 
 36 
 
 0.439 
 
 3.85 
 
 3.30 
 
 10 
 
 5.680 
 
 49.82 
 
 32.00 
 
 35 
 
 0.464 
 
 4.07 
 
 3.50 
 
 9 
 
 7.012 
 
 61.51 
 
 35.00 
 
CHAP. XIV. THEEMO-ELECTEW MEASUEEMENT. 249 
 
 vening space between the curves b J and b f on tlie 13th 
 ordinate, and the length of that ordinate between the base-line 
 a c and curve b d, we obtain a definite idea of the magnitude 
 of the error involved in Melloni's doctrine that deflection and 
 enei'gy correspond until the position of the needle marks thir- 
 teen degrees from zero. It is important to observe that a 
 scale, corresponding with the scale of inches marked on the 
 side of the table represented in the illustration, has been 
 introduced parallel with the vertical base-line a c in the 
 diagram. This expedient enables us to make a direct com- 
 parison between the position of the concave radiator and 
 the deflection of the needle of the galvanometer, marked on 
 the vertical base-line a c. It wull be seen, for instance, that, 
 when the deflection of the needle is 35 degrees, the radiator 
 is placed 9 inches from the face of the pile. 
 
 Let us now consider briefly the manner of conducting the 
 experiments which have enabled us to constioict the diagram 
 referred to and the accompanying table A. The deflections 
 of the needle of the galvanometer and the energies imparted 
 by radiation to the pile beyond the 50th division being very 
 small, the observations between that division and the termi- 
 nation of the scale have been confined to spaces of 5 inches 
 each, as will be seen on reference to the table mentioned. 
 But from the 50th to the 9th division the observations have 
 been made for each inch. Accordingly, 41 distinct experi- 
 ments were instituted while advancing the boiler from the 
 fiftieth to the ninth division of the scale attached to the 
 side of the table. It is important to mention that, in order 
 
250 RADIANT HEAT. chap. xiv. 
 
 to allow tlie pile to cool effectually, the sliding block h, 
 together with the boiler and screens s and s', were removed 
 into an adjoining room, and the needle of the galvanometer 
 brought to perfect rest at zero, for each observation. It may 
 be mentioned that the final investigation was carried out 
 under very favorable conditions, the temperature of the sur- 
 rounding air fluctuating so slightly that the differential tem- 
 perature of the boiler did not vary one degree during the 
 experiment. The mode of constructing Table A, which exhi- 
 bits the relation between energy and deflection at definite 
 distances, will be readily understood. The distance between 
 the concave radiator and the face of the pile, it will be 
 seen, has been entered in the first column, while the tempe- 
 rature transmitted to the pile by radiation has been entered 
 in the second column. The determination of the temperature 
 referred to is effected by the following simple arithmetical 
 process : Multiply the squai'e of the I'adius of the concave 
 surface r by the differential temperature of the boiler, and 
 divide the product by the square of the distance between 
 the radiator and the pile ; the quotient — entered in the second 
 column of the table — expresses the intensity transmitted to the 
 pile by the heat emanating from the radiator. For example, 
 the intensity of the transmitted radiant heat, at a distance 
 
 2' X 142 
 of 19 inches, -will be — — = 1°.573 F. Eef erring to the 
 
 table, it Avill be found that the temperature thus ascei'tained 
 is recorded in the second column, opposite the distance 19 
 entered in the first column. The mode of ascertaining the 
 
CHAP. XIV. TnEBMO-ELECTRIV MEASUREME^'T. 251 
 
 temperature transmitted by tlie radiator to the thermo-electric 
 pile at each poiut of the scale being thus fully explained, 
 let us now consider the relative amount of energy repre- 
 sented by the temperature entered in the second column of 
 our table. It has already been pointed out that the tempe- 
 rature imparted to the pile by the radiant heat, and definite 
 arcs — say degrees — swept by the needle of the galvanometer, 
 are not comparable quantities ; hence cannot be determined 
 by calculation. We must therefore, as already pointed out, 
 have recourse to the experimental process in ascertaining the 
 relation of the temperature transmitted and the deflection of 
 the needle when the radiator is at maximum distance from 
 the thermo-electric pile. Repeated trials have shown that, 
 when the radiator is placed at a distance of 100 inches 
 from the face of the pile, the ratio between the deflection 
 of the needle — measured by arcs containing tJt of a circle ; 
 and the temperature transmitted to the pile — measured by 
 degrees of Fahrenheit, is as 0.50 to 0.57. The energies 
 inserted in the third column of the table have been deter- 
 mined in accordance with the stated relation of temperature 
 and deflection of the needle of the galvanometer, while the 
 deflections of the needle entered in the fourth column have 
 been determined by observation. It will be seen, by inspect- 
 ing the latter, that the energy exceeds that indicated by the 
 deflection of the needle, for all distances between the 55th 
 division on the scale and the pile ; the energy at that point 
 being 1.64, while the deflection is only 1.60. Between the 
 100th and 60th divisions of the scale the observed deflec- 
 
252 
 
 BABIANT HEAT. 
 
 CHAP. XIV. 
 
 tions, it mil be noticed, are irregular, slightly exceeding tlie 
 energies. This ii'regiilarity is occasioned by the sensitiveness 
 of tlie instrument when the radiator is far from the pile. It 
 only remains to call attention to Table B, exhibiting the final 
 result of our elaborate investieation of the thermo-electric 
 
 Table B, showing the Eelation between Deflection 
 AND Energy. 
 
 Deflection of 
 needle. 
 
 Energy ex- 
 erted. 
 
 Deflection of 
 needle. 
 
 Energy ex- 
 erted. 
 
 Deflection of 
 needle. 
 
 Energy ex- 
 erted. 
 
 Deg. 
 
 Relative. 
 
 Deg. 
 
 Relative. 
 
 Deg. 
 
 Relative. 
 
 1 
 
 2 
 
 3 
 
 4 
 
 6 
 
 6 
 
 7 
 
 8 
 
 9 
 
 10 
 
 11 
 
 12 
 
 0.89 
 
 2.17 
 
 3.38 
 
 4.59 
 
 5.77 
 
 6.86 
 
 8.00 
 
 9.15 
 
 10.33 
 
 11.53 
 
 12.75 
 
 14.00 
 
 13 
 
 14 
 
 15 
 16 
 17 
 18 
 19 
 20 
 21 
 22 
 23 
 24 
 
 15.28 
 16.58 
 17.90 
 19.24 
 20.60 
 21.98 
 23.39 
 24.84 
 26.34 
 27.90 
 29.53 
 31.25 
 
 25 
 26 
 27 
 28 
 29 
 30 
 31 
 32 
 33 
 34 
 35 
 
 33.07 
 35.00 
 37.05 
 39.24 
 41.58 
 44.20 
 47.00 
 50.00 
 53.30 
 57.00 
 61.23 
 
 method of measuring radiant heat. It will be seen, on care- 
 fully examining this table, that the deflection of the needle of 
 the galvanometer at the termination of the first degree exceeds 
 the energy transmitted by the radiator towards the thermo- 
 electric pile in the ratio of 100 to 89, difference = 0.11 ; 
 
CHAP. XIV. THEBMO-ELECTBIC MEASXTREMENT. 253 
 
 an uninipoii:aiit irregularity already adverted to. Beyond 90' 
 from zero tlie energy becomes greater than the deflection in 
 a constantly increasing ratio as the arcs swept by the needle 
 augment. Accordingly, when the needle has moved through 
 an arc of 13 degrees, the energy is greater than the deilec- 
 tion in the ratio of 15.28 to 13.00, instead of being exactly 
 balanced, as stated by Melloni. 
 
CHAPTER XY. 
 
 THE THEEMOHELIOMETER. 
 
 The calculations presented by Pere Secchi in his work 
 " Le Soleil," relative to the intensity of solar radiation and 
 the temperature of the sun, being based on the indications 
 of his thermoheliometer, I have carefiilly examined the pro- 
 perties of this unique device, delineated on page 267 of the 
 work referred to. The accompanying illustration (see Fig. 1) 
 represents a longitudinal section of the same through the ver- 
 tical plane. A B and C D are two concentric cylinders sol- 
 dered one to the other ; they form a kind of boiler, the 
 annular space being filled with water or oil at any tempe- 
 rature. A thermometer t passes through a tube, across the 
 annular space, to the axis of the cylinder ; it receives the 
 solar rays introduced through a diaphragm m •??, the opening 
 of which is very little larger than the bulb of the thermo- 
 meter. A thick glass v closes the back part of the instrument, 
 and admits of ascertaining whether the thermometer is placed 
 in a direct line with the pencil of rays. The interior cylinder 
 
 254 
 
THE TEEBMOEELIOMETEB. 
 
 255 
 
 and the thermometer t are coated with hmip-Llack. A second 
 thermometer t' shows the temperature of the annular space, 
 and consequently that of the enclosure. The A\hole apparatus 
 is mounted on a support having a j)arallactic movement, to 
 facilitate following the diurnal motion of the sun. The appa- 
 ratus being exposed to the sun, it will be found, on observing 
 the two theimometers, that their difference of temperature 
 increases gradually, and that in a short time it ends by being 
 constant. 
 
 Before pointing out the peeuliaiities of the contrivance 
 thus described by Pcre Secchi, it will be instructive to 
 examine his " solar intensity apparatus," manufactured by 
 Casella, represented in Fig. 2. The manufacturer publishes 
 the following statement regarding this instmment ; " Two 
 thermometers are here kept immersed in a fluid at any 
 temperature, and a third surrounded by the same conditions, 
 but not immersed, is exposed to the rays of the sun. The 
 
256 BABIANT HEAT. CHAr. XT. 
 
 increase of temperature tlius obtained is found to be tLe 
 same, irrespective of the temperature of tlie fluid ■wliicli sur- 
 rounds it." No one acquainted with the principles which 
 govern the transmission of heat within circuhiting fluids can 
 fail to observe that the thermometers applied above the 
 central tube will not furnish a reliable indication of the 
 temperature of the fluid below the same, nor of any portion 
 of the contents of the annular space towards the bottom. 
 Apart fi'om this defect, it will be j)erceived that an upward 
 current of atmospheric air will sweep the under side of the 
 external cylinder, causing a reduction of temperature of the 
 fluid confined in the lower half of the annular sjjace. Again, 
 the heat radiated by the bulb of the thermometer exposed 
 to the sun will elevate the temperature of the air within 
 the central tube, and consequently produce an internal circu- 
 lation tending to heat the upper part of the fluid contained 
 in the annular space. The effect of the irregular heating 
 and cooling thus adverted to will be considered after an 
 examination of the result of some observations recoi'ded in 
 Tal)les A and B, which I conducted at different times 
 during the month of September, 1871. In order to insure 
 an accurate position, the instrument during these observa- 
 tions was mounted in a revolving observatory upon a table 
 turning on declination axes provided with appropriate mecha- 
 nism and declination circle. An actinometer being attached 
 to the same table, the true intensity of the radiant heat, as 
 well as the sun's zenith distance, were recorded simultane- 
 ously with the indications of the Secchi instrument furnished 
 
TUE THERMOUELIOMETEB. 
 
 257 
 
 Table A 
 
 , SHOWING 
 
 TUE Hesult of Observations made with 
 
 Secchi's Tiieumoheliometer, 
 
 manufactured by Casella. 
 
 SEPTEMBER S. 
 
 Ill 
 
 External casing. 
 
 3i 
 p 
 
 if 
 
 tS.a 
 
 Upper 
 thermometer. 
 
 Lower 
 thermometer. 
 
 Mean. 
 
 •F<A. 
 
 • Fa^. 
 
 'Fah. 
 
 'Fah. 
 
 'Fah. 
 
 Dtg. 
 
 83.5 
 
 j 
 
 76.0 
 
 70.0 
 
 73.0 
 
 10.5 
 
 33.0 
 
 84.2 
 
 77.0 
 
 71.5 
 
 74.2 
 
 10.0 
 
 
 8.5.5 
 
 70.0 
 
 74.2 
 
 76.6 
 
 8.8 
 
 32.50 
 
 86.0 
 
 83.5 
 
 74.5 
 
 79.0 
 
 7.0 
 
 
 89.0 
 
 84.0 
 
 75.5 
 
 79.7 
 
 9.2 
 
 33.0 
 
 90.5 
 
 85.0 
 
 76.5 
 
 80.7 
 
 9.7 
 
 
 92.0 
 
 85.5 
 
 78.0 
 
 81.7 
 
 10.2 
 
 33.10 
 
 93.0 
 
 86.5 
 
 79.0 
 
 82.7 
 
 10.2 
 
 
 94.0 
 
 87.8 
 
 80.0 
 
 83.9 
 
 10.1 
 
 33.21 
 
 94.5 
 
 89.0 
 
 81.5 
 
 85.2 
 
 9.2 
 
 
 95.5 
 
 90.0 
 
 82.5 
 
 86.2 
 
 9.2 
 
 33.32 
 
 9G.5 
 
 90.5 
 
 83.5 
 
 87.0 
 
 9.5 
 
 
 98.0 
 
 91.5 
 
 84.5 
 
 88.0 
 
 10.0 
 
 33.44 
 
 99.0 
 
 92.0 
 
 85.0 
 
 88.5 
 
 10.5 
 
 
 100.0 
 
 93.0 
 
 80.0 
 
 89.5 
 
 10.5 
 
 33.56 
 
 101.0 
 
 93.5 
 
 86.5 
 
 90.0 
 
 11.0 
 
 
 101.5 
 
 94.0 
 
 87.0 
 
 00..-) 
 
 11.0 
 
 34.8 
 
 93.1 
 
 86.9 
 
 79.7 
 
 83.3 
 
 9.80 
 
 33.24 
 
a58 
 
 BADIANT HEAT. 
 
 Table B, 
 
 SHOWING THE 1!esult of Obskuvatioxs -mape with 
 
 Secch] 
 
 'S TUEKMOIIELIO.METEH, MANUFACTURED BY CaSELLA. 
 
 SEPTEMBER 6. 
 
 lis 
 
 O o (D 
 
 External easing. 
 
 3 2 
 
 II 
 
 ^1 
 
 Upper 
 thermometer. 
 
 Lower 
 thermometer. 
 
 Mean. 
 
 °i?aA. 
 
 'ITah,. 
 
 ' Fah. 
 
 " Fah. 
 
 " Fah. 
 
 Deg. 
 
 94.0 
 
 88.0 
 
 81.5 
 
 84.7 
 
 9.7 
 
 35.56 
 
 95.5 
 
 88.5 
 
 83.0 
 
 85.7 
 
 9.7 
 
 
 96.5 
 
 89.5 
 
 84. 5 
 
 87.0 
 
 9.5 
 
 35.41 
 
 97.5 
 
 90.0 
 
 85.0 
 
 87.5 
 
 10.0 
 
 
 98.0 
 
 90.0 
 
 85.0 
 
 87.5 
 
 10.5 
 
 35.26 
 
 98.5 
 
 90.5 
 
 85.5 
 
 88.0 
 
 10.5 
 
 
 99.0 
 
 90.5 
 
 85.7 
 
 88.1 
 
 10.9 
 
 35.11 
 
 100.0 
 
 91.0 
 
 86.5 
 
 88.7 
 
 11.2 
 
 
 100.3 
 
 91.0 
 
 87.0 
 
 89.0 
 
 11.3 
 
 34.56 
 
 100.3 
 
 91.2 
 
 87.5 
 
 89.3 
 
 11.0 
 
 
 100.5 
 
 91.5 
 
 88.0 
 
 89.7 
 
 10.8 
 
 34.41 
 
 98.2 
 
 90.1 
 
 85.3 
 
 87.7 
 
 10.45 
 
 35.33 
 
 SEPTEMBER 27. 
 
 78.5 
 
 64.0 
 
 64.0 
 
 64.0 
 
 14.5 
 
 44.0 
 
 79.0 
 
 65.0 
 
 64.0 
 
 64.5 
 
 14.5 
 
 
 79.5 
 
 65.0 
 
 64.5 
 
 64.7 
 
 14.7 
 
 44.55 
 
 79.5 
 
 63.0 
 
 65.0 
 
 64.0 
 
 15.5 
 
 
 79.5 
 
 64.0 
 
 65.0 
 
 64.5 
 
 15.0 
 
 45.51 
 
 79.0 
 
 64.5 
 
 65.0 
 
 64.7 
 
 14.2 
 
 
 79.0 
 
 64.5 
 
 65.5 
 
 65.0 
 
 14.0 
 
 46.48 
 
 79.0 
 
 64.5 
 
 65.5 
 
 65.0 
 
 14.0 
 
 
 79.0 
 
 65.0 
 
 65.5 
 
 65.2 
 
 13.8 
 
 47.46 
 
 79.1 
 
 64.4 
 
 64.9 
 
 64.65 
 
 14.45 
 
 45.16 
 
CUAI-. XV. lUE TUEliMOUEl.lOMETEIi. 259 
 
 by Casella. Let us fii-st couskler the tabulated observations 
 of September 2, recorded at equal intervals of three minutes. 
 The indication of the two theruK)nieters immersed in the 
 tUiid contained in the annular space first claims our attention, 
 since the temperature of this fluid is 'the principal element in 
 Pere Secchi's original computations of solar temperature. It 
 will be seen, on referring to the secimd and third cohnnns of 
 the table, that, while the upper thermometer indicates a 
 mean temperature of SG^'.O, the lower one shows only 79°.."), 
 difference = 7°.4. This great discrepancy of temperature at 
 differeut points of the upper portions of the anmdar space 
 at which, owing \.o the inclined pt)sition of the concentric 
 tubes, something like uniformity ought to exist, suggests a 
 still gi'eater discrepancy of temperature at the under side 
 towards the lower termination of the tubes. In addition, 
 therefore, to the observed irregularity of temperature at the 
 upper pai-t, shown by the table, no indication whatever is 
 furnished of the temperature of the fluid in the annular 
 space below the central tube, nor towards the termination 
 at either side. Obviously, then, no accurate computation 
 can be made of the degree of refrigeration to which the 
 central thermometer is exjoosed by the radiation from the 
 cold Itlackened surface of the internal tube, every part of 
 which, as we have seen, possesses a different temperature 
 compared A\ith the rest, consequently transmitting radiant 
 energj" of different intensity. It will be found practically 
 impossible, therefore, to detemiine the true differential tem- 
 perature of the contents of the bulb exposed to the sun's 
 
260 MADIANT HEAT. chap. xv. 
 
 rays and the fluid contained in the annular space. Hence, 
 the differential temperature entered in the table, the result 
 of comparing the indications of the thermometers, is mani- 
 festly incorrect. It will be found, also, by reference to 
 Table A, that, while the mean temperature imparted to the 
 central thermometer by the sun's rays is 93°. 1, the mean 
 temperature of the fluid in the annular space is 83°. 3. Con- 
 sequently, the intensity of solar radiation established by the 
 instrument is only 93°.l — 83°.3 = 9°.80 F. Now, the sun 
 during the recorded experiment of September 2 was excep- 
 tionally clear, the mean indication of the actinometer while 
 the experiment lasted being 60°.05, thus sho\\ing that the 
 
 energy developed was only — '- — = 0.16 of the true radiant 
 ^ "^ 60.05 
 
 intensity. The mean zenith distance, it may be mentioned, 
 was only 33 deg. 24 min. during the exjieriment. Agree- 
 ably to the table of temperatures (see Chap. III.), the maxi- 
 mum solar intensity for the stated zenith distance is 63°.35 ; 
 thus we find that the sun, as stated, was exceptionally clear 
 while the trial took place, which resulted in developing the 
 trifling intensity of 9°.80 F. The result of the experiments 
 conducted September 6, recorded in Table B, it will be 
 seen, was nearly the same as that just referred to, the mean 
 temperature indicated by the thermometer exposed to the 
 sun being 98°.2, while the mean of the two thermometers 
 immersed in the fluid was 87°.7 ; hence the differential tem- 
 perature 98°.2 - 87°.7 = 10°.45. The mean temperature of 
 solar radiation during the experiment, ascertained by the 
 
CUAP. XV. TIU-: TUEimOBELIOiLETHB. 261 
 
 aotinometer, was 5'J°.75, the zeuitb distance being 35 cleg. 33 
 
 mill. Consequently, the intensity indicated September G was 
 
 ^ 10.45 
 only zrrzz = ^'-1" of the true enei'icy of the sun's radiant 
 
 lu'at, against O.IG during the previous experiment. It will 
 be observed that the fluctuation of the differential tempera- 
 ture was much greater SejJtenibcr 2 tliaa duiiiig the succeed- 
 ing experiment, owing, no doubt, to the influence of currents 
 of air produced by a strong breeze on the first occasion, the 
 I'evolving observatory being partially open on the side pre- 
 sented to the sun during observations. 
 
 With reference to the small differential temperature indi- 
 cated by the Secchi instniment manufactured by Casella, it 
 may be urged that it is not intended to sho\v the true inten- 
 sity of solar radiation on the earth's suiface, but simjjly a 
 means of deteiTuining solar temperature. Granted that such 
 is the object, yet the extreme irregularity of the temperature 
 of the fluid within the aniuilar space shows that the instru- 
 ment is unreliable — a fact established beyond contradiction 
 by an experiment instituted September 27, on which occasion 
 water of a uniform temperature was circulated through the 
 annular space. This was effected by gradually charging the 
 intervening space from the top, and carrying off the waste 
 at the bottom, holes having been drilled in the external 
 casing for that purpose. The result of this conclusive expe- 
 riment is recorded at the foot of Table B. It \vill be found, 
 on reference to the figures, that the mean difference of the 
 two thermometers immersed in the fluid was only 64°.9 — 
 
262 EADIAM MEAT. chap. xv. 
 
 04:°.4 = 0.5°, while tlie mean differential temperature was 
 augmented to 79".! — 64:°.65 = 14°.45, against 9°.80 on tlie 
 :2nd of September, althougli the zenitli distance was greater, 
 and the solar intensity less ; circumstances which ought to 
 have diminished the indicated intensity. It is needless to 
 enter into any further discussion of the demerits of the 
 instrument represented in Fig. 2. We maj' uo\v return to 
 the consideration of the device delineated in Fig. 1, copied 
 from " Le Soleil." It will be seen that the material diffe- 
 rence of construction is that of applying only one thermo- 
 meter for ascertaining the temperature of the fluid in the 
 annidar space. Possibly this single thermometer may indi- 
 cate approximately the mean tempei'atui'e of the uppei' and 
 lower portions of the fluid above the central tube ; but it 
 furnishes no indication of the temperature below, nor at 
 either extremity of the annular space. The iuaderpiacy of 
 the means adopted for ascertaining the temperature of the 
 intei'nal sui-face which radiates towards the bulb of the cen- 
 tral thermometer having thus been pointed out, it Mill be 
 well to consider whether the expedient of passing a stream 
 of \vater of nearly uniform temperature through the annular 
 space will insure trustworthy indication. In order to deter- 
 mine this question, I have constructed two instruments, in 
 strict accordance with the delineation in Fig. 1, excepting 
 that in one of these the concentric cylinders are considerably 
 enlarged, the annular space, however, remaining unchanged. 
 Ex|ieriments with the two instruments prove that the enlarge- 
 ment does not materially influence the indications, provided 
 
THE THEBMOUELJOMETEIi. 
 
 203 
 
 Avatcr of a uniform teniperahue be circulated tlirougli the 
 aiiiiuhir hpacf. But tliese exjicrinients liave deuioustiatcd 
 that the size of tlie Lull) of the thcnnouieter e.xpo.sed to the 
 
 Table C, showing the Kesult of employing diffehknt 
 
 TlIEKVO- 
 
 METEKS, THE Bvi.IIS OF WIIICI 
 
 ARE OF 
 
 UNEQU. 
 
 lL Diam 
 
 :ti:k. 
 
 
 J) 
 
 (tmctcr of B 
 
 nib, 0.30 
 
 Tmh. 
 
 
 
 1 ij-iuch tube. 
 
 5S 
 
 3-inch tulie. 
 
 c 2 
 
 
 
 
 
 
 
 Siui. 
 
 Fluid. 
 
 Diff. 
 
 
 Sun. 
 
 Fluid. 
 
 Diff. 
 
 S!^ 
 
 •Fah. 
 
 'Fah. 
 
 'Fah. 
 
 Z)«(7. 
 
 •Fah. 
 
 'Fah. 
 
 = Fah. 
 
 Dtg. 
 
 74.0 
 
 no.o 
 
 14.0 
 
 50.32 
 
 77.5 
 
 62.1 
 
 15.4 
 
 49.54 
 
 74.5 
 
 60.3 
 
 14.2 
 
 50.24 
 
 78.5 
 
 62.3 
 
 16.2 
 
 50.30 
 
 75.0 
 
 60.7 
 
 14.3 
 
 50.16 
 
 79.0 
 
 62.5 
 
 16.5 
 
 50.12 
 
 75.5 
 
 61.0 
 
 14.4 
 
 50.08 
 
 79.0 
 
 63.0 
 
 IC.O 
 
 50.21 
 
 76.0 
 
 61.0 
 
 15.0 
 
 50.01 
 
 79.0 
 
 63.0 
 
 16.0 
 
 50.30 
 
 75.0 
 
 60.6 
 
 14.4 
 
 50.16 
 
 78.6 
 
 62.6 
 
 16.0 
 
 50.12 
 
 
 Ih 
 
 <i meter of II 
 
 idh, 0.58 
 
 Inch. 
 
 
 
 83. C 
 
 02. G 
 
 21.0 
 
 49.54 
 
 79.2 
 
 CO.! 
 
 19.1 
 
 50.32 
 
 85.5 
 
 63.0 
 
 22.5 
 
 50.30 
 
 81.0 
 
 60.3 
 
 20.7 
 
 50.24 
 
 8G.4 
 
 63.4 
 
 23.0 
 
 50.12 
 
 82.5 
 
 60.7 
 
 21.8 
 
 50.16 
 
 86.7 
 
 63.5 
 
 23.2 
 
 50.21 
 
 82.7 
 
 60.7 
 
 22.0 
 
 50.08 
 
 87.7 
 
 63.7 
 
 23.0 
 
 50.30 
 
 83.0 
 
 61.0 
 
 22.0 
 
 50.01 
 
 85.9 
 
 63.2 
 
 22.5 
 
 50.12 
 
 81.7 
 
 60.6 
 
 21.1 
 
 50.16 
 
 sun cannot lie chauLicd witliout influencing the differential 
 temperature most mateiially. This ^vill be seen by refer- 
 ence to Table C\ uhicli records the result of e.viierinients 
 
2(i4 RADIANT HEAT. chap. xv. 
 
 with different thermometers, and tubes of different diameters, 
 conducted October 17. As on previous occasions, the instru- 
 ments, in order to insure accurate position, were attached to 
 the deoliuation table arranged within the revolving observa- 
 toiy. The bulbs of the thermometers employed were veiy 
 nearly sj^herical, their diameters being respectively 0.30 and 
 0.58 inch. The upper division of Table C, which records the 
 experiment with the small bulb exposed to the sun, estab- 
 lishes, it will be seen, a differential temperature of 14°.4 
 for the instrament having the Ij-in. central tube, and 16° 
 for the one having the 3-in. central tube. Referring to the 
 knver division of the same table, it ^vill be seen that, Avhen 
 the thermometer with the large bulb is exposed to the sun, 
 the differential temperature reaches 22°.5 in the instrument 
 containing the \\-m central tube, and 21°.l in the one having 
 the 3-in. tube. We thus find that, by doubling the diameter 
 of the bulb of the thermometer exposed to the sun, all other 
 things remaining unchanged, an augmentation of the differen- 
 tial temperature amounting to nearly one-third takes place. 
 This fact proves the existence of inherent defects fatal to 
 the device delineated in Fig. 1, rendering the same wholly 
 unreliable. 
 
 Agreeably to the doctrine of exchanges, the diameter of 
 the bid!) is an element of no moment, since the internal radi- 
 ation towards the same — -jprovided its teinperature he nnifonn 
 — depends solely on the temperature and angular distances 
 of the I'adiating points of the enclosure. Infallibility of the 
 thermoheliometer has evidently been taken for granted on 
 
CHAP. XV. THE TIIEBMOHELIOMETER. 205 
 
 the strength of the suuiuluess of this ductriue, as Ave fiud 
 no allusion to the size of the bull) in M. Soret's account of 
 his observations of solar intensity on ]Mont Blanc ; nor does 
 yiv. AVaterston, Avho employed a similar instrument during 
 his obsei'vatious in India, advert to the dimensions of the 
 bulb of the thermometer exposed to the sun. These i>hysi- 
 cists apparently overlook the fact that, ^vhile the entire convex- 
 area of the bulb is exposed to what may be considered the cold 
 radiation from the enclosure, only one-half receives radiant 
 heat from the sun. This circumstance would be unimportant 
 if the heat thus received were instantly transmitted to every 
 part ; but the bulb and its contents are slow conductors, 
 while the conducting power diminishes nearly in the iuvei-se 
 ratio of the square of the depth. Consequently, by increas- 
 ing the diameter, the parts of the bulb opposite to the sun 
 \\\\\ receive considerably less heat relatively in a given time 
 than if the diameter be diminished. 
 
CHAPTER XVI. 
 
 BAEOMETRIC ACTINOMETEE. 
 
 The bulb of tlie tLermometer is cliarged mth air ; tlie 
 intensity of the radiant heat determined by the pressure in 
 the bulb ; the height of the mercurial column indicates the 
 pressure ; a fixed graduated arc and movable index show the 
 sun's zenith distance ; the graduation of the scale of tempe- 
 rature effected without exposing the bulb to heat or cold ; 
 the sun's zenith distance and the intensity of the i-adiant 
 heat observed simultaneously ; the bi;lb, being placed within 
 a vacuum, is not exposed to the disturbing influence of atmo-" 
 spheric currents ; the vessel surrounding the bulb maintained 
 at a constant temperature ; the atmospheric temperature does 
 not affect the indication of radiant intensity ; the quantity of 
 matter contained in the bulb is exceedingly small compared 
 with the convex area exposed to the solar rays ; suitable 
 mechanism, operated by two small hand-wheels, enables the 
 observer to follow the diurnal motion and sun's declination ; 
 the instrument is portable. 
 
cH.vr. XVI. BAUOMEIRIC AVTlSOMETElt. 267 
 
 Au accurate deteimiiiatioii uf the iuteusity of solar heat 
 calls for a thermoiueter capable of indicating the temperature 
 produced by radiation after a very brief exposure to the 
 radiant heat. It has been pointed out in previous chapters 
 that the sluggish action of an ordinary thermometer renders 
 it \\ holly unfit to measure the temperature produced by solar 
 radiation at any given zenith distance, since the diurnal motion 
 is so rapid that before an equilibrium can be established 
 between the heat received by the bidb and the heat radiated 
 by the same, the zenith distance is materially changed. Con- 
 sequently, the temperatures indicated by common thermome- 
 ter before noon are too low, Avhile in the afternoon the indi- 
 cation is too high, for the zenith distance at any given instant 
 of time. In order to ascertain to \\\\i\i extent the ordinaiy 
 thermometer is defective as a means of measuring the sun's 
 radiant heat, consequent on the slow expansion of the con- 
 tents of the bulb, I have conducted a series of experiments 
 which show that a thermometer surrounded by a vessel kept 
 at a constant temperature and exposed to the i-adiation of a 
 steady gas-flame, requires from 20 to 25 minutes before the 
 mercurial column becomes stationaiy. Consequently, the 
 rapid change of the sun's zenith distance, especially early 
 in the morning and late in the afternoon, presents a difficulty 
 which renders oidinary theriuonieters useless for measuring the 
 intensity of solar heat at given zenith distances. But a far 
 greater defect inseparable from the oidinaiy form of thermo- 
 meter remains to be noticed, namely : the section of the pencil 
 of rays which imparts the radiant heat is less than the con- 
 
368 RADIANT HEAT. chap. xvi. 
 
 vex area of the bulb exposed to tlie sun. This circumstauce 
 presents a serious difficulty, calling for delicate counteracting 
 expedients (see Chap. III.), since although that half of the 
 bulb which is turned away from the sun may be protected 
 by a non-conducting substance, the other half on which the 
 sun acts exposes a radiating surface twice as great as the 
 sectional area of the acting pencil of I'ays, because the convex 
 area of the bulb is four times greater than the area of its 
 greatest section. It should be observed, however, that by 
 employing a long bulb of cylindrical form, the inherent defect 
 of the common thermometer thus pointed out may be miti- 
 gated in the ratio of about 4 to 3, since the plane which 
 passes through the axis of a cylinder bears a greater pro- 
 portion to its convex area than the area of the great circle 
 of a sphere to its convex area. Some scientists contend that, 
 agreeably to the law of exchanges, no actual loss of heat is 
 sustained by the excess of radiating surface of the bulb over 
 the sectional area of the pencil of I'ays which imparts the 
 radiant heat. A moment's consideration will dispose of this 
 unsound doctrine. Admitting that by means of non-conduct- 
 ing substances the half of the bulb opposite to the sim may 
 be effectually protected against loss of heat by radiation, 
 the other half which is turned towards the luminary will 
 constitute a refrigerator as well as a heater. Now, the effi- 
 ciency of refrigerators and heaters is as their areas, all other 
 things being alike ; hence a sj^herical bulb of 0.8 in. diameter 
 (the convex area of Avhich is 2 square inches) presents a 
 radiating surface exactly 1 square inch towards the sun, 
 
CUA1-. XVI. BMlVMETiac ACTIXOMETEE. 2G9 
 
 while the section of tlie peucil of tittlar rays Avliicli the bulb 
 intercepts contains only 0.5 square inch. The lays lieing 
 thus distributed over an area twice as great as their section, 
 the mean intensity of the radiant heat imparted to the bulb 
 will be diminished one-half. But, while the hemispherical 
 sui-face of the IniUi turned towards the sun is thus rendered 
 inefficient by the dispersion of the rays, its efficiency as a 
 vcfriyc-rato)', remaining unimpaired, carries off the lieat with 
 full eneig}' towards the surrounding cold medium. As our 
 demonstration relates fmly to the defect consequent on the 
 sectional area of the ])encil of rays being less than the area 
 which receives the solar heat, I have not noticed the serious 
 loss caused by cold cun-ents of air circulating louud the 
 exposed half of the bulb. The self-evident chai-acter of the 
 foregoing explanation renders theoretical deductions, from the 
 law of exchanges, unnecessary to establish the fact that ordi- 
 naiy thermometei-s cannot furnish correct indications of the 
 temperature produced by solar radiation. 
 
 Before entering on a detailed description of the instru- 
 ment illustrated on Plate 27, it will be instnictive to con- 
 sider what proportion of the indicated temperature results 
 from vnaided solar Itcat when a substance surrounded by 
 the atmosphere is exposed to the sun's rays ; and whether 
 the observed increment of tenqierature above that of the 
 suiTonnding air occupies a fixed position on the themio- 
 metric scale, or whether it rises and falls with the inciease 
 and diminution of the atmospheric temperature. Suppose 
 that we place a circular disc, composed of some black non- 
 
270 BABIA^^T HEAT. chap. xvi. 
 
 conducting substance, at the bottom of a veiy deep cyliu- 
 di'ical vessel, kept at a uniform temperature, ■whose axis 
 points towards the sun, and "which is provided "with an open- 
 ing opjjosite the disc — the diameter of the disc and opening 
 being alike. Suppose also that, by some adequate device, 
 a perfect vacuum is kej^t up in the said vessel, and that the 
 axis of the black circular disc is directed towards the solar 
 centre. I maintain that the temperature acquired by the 
 suiiace of the supposed disc — less the temperature of the 
 surrounding vessel — furnishes an accurate indication of the 
 real intensity of the sun's radiant heat. It will be perceived, 
 on due reflection, that the differential temperature thus ascer- 
 tained, which furnishes a true measure of the I'adiant inten- 
 sity of solar heat, cannot occupy a fixed position on the 
 thermometric scale. It rises and falls by increase and dimi- 
 nution of the temperature of the supposed surrounding vessel. 
 The investigations and observations refei'red to in Chap. III., 
 conducted during a series of years, it should be borne in 
 mind, have established the fact that during the summer 
 solstice the differential temperature produced by solar radia- 
 tion, indicated by my actinometer, is fully 66° F. in the 
 latitude of New York, when the sky is perfectly clear and 
 the zenith distance 18 degrees. The question will be asked : 
 Is there any limit to the rise and fall of the thermometric 
 interval of 66°, consequent on changes of the temperature 
 of the surrounding medium? As the leading j^oints con- 
 nected with this question have been fully discussed in Chap. 
 IX., I will merely observe that the result of numerous obser- 
 
CHAP. XVI. BAllOMETEIC ACTINOMETEB. 271 
 
 rations, and the trial of various expedients resorted to in 
 order to determine the limit, show that the movement is not 
 limited. The extent of the fall, as far as ascertained, is so 
 considerable that Ave must infer that, but for the intir\ention 
 of our atmosphere and the accumulation of heat lesultinj^^ 
 from that intervention, the sun's unaided radiant heat woidd 
 not be sufficient to prevent the sui-face of the earth hum 
 falling several hundred degrees below the freezing point of 
 water. No reliable experiments have yet been instituted to 
 ascertain the extent of the upward movement of the thermo- 
 metric interval under discussion, when the solar rays are 
 admitted into an incandescent enclosure. 
 
 The following general description will enable the reader 
 to form a cori-ect idea of the nature of the barometric actino- 
 meter. The leading feature of the instrument is that of em- 
 ploying a bulb (see Plate 27, Fig. 1) composed of very thin, 
 hard plate-metal, charged with dry atmospheric air, in place 
 of a bulb containing some liquid substance. The lower part 
 of this air-bulb is heuiis^dierical, Avhile the upper part, ex- 
 posed to the sun, consists of a circular plate with a slight 
 upward curvature, the extreme diameter being 2.75 inches. 
 The hemispherical jxirt of the bulb is plated with nickel on 
 both sides, highly polished and thoroughly pi-otected by a 
 non-conducting external covering, A\"hile the t(->p plate, ex- 
 posed to the radiant heat, is coated with lampblack on the 
 outside, the inside presenting a rough surface. It Avill be 
 evident fi"om this description that the form of the bulb of 
 the barometric actinometer fulfils the condition inseparable 
 
273 FADIANT HEAT. chap. xti. 
 
 from devices intended to fumisli accurate indication of the 
 intensity of solar radiation — namely, that the area of the face 
 Avliich receives the radiant heat should not exceed the area 
 of the section of the pencil of rays admitted to the instru- 
 ment. The air-bulb, as already stated, is i^laced near the 
 bottom of an exhausted cylindrical vessel, 8 inches long, 
 3 inches diameter, surrounded by a double casing through 
 which water, kept at a constant temperature, is circulated. 
 An ordinary force-pump is employed for this purpose, 
 receiving its supply from a small portable cistern ; the 
 water thus circulated through the double casinij beinff 
 returned to the c-istern by the action of the pump. The 
 upper end of the cylindrical vessel is closed by a plate of 
 glass 0.12 inch thick, forming an air-tight joint. Tlie aper- 
 ture in the ring which secures the glass to the cylinder is 
 2.75 ins. in diameter, corresponding exactly with tliat of the 
 bulb. The exhausted vessel is held in position by a trans- 
 verse axle secured to its bottom (see Fig. 2), turning in 
 appropriate bearings supported by columns resting on the 
 circular bed-plate of the instrument. The transverse axle 
 mentioned is provided with a central perforation ^vhich, by 
 means of minute passages, communicates M'ith the interior of 
 the bullj, conmiunicating also with a close mercurial cistei-n f 
 formed at the lower end of the barometric tulje, as represented 
 in the illustration. The height of the column in this baro- 
 metric tube, it is scarcely necessary to observe, furnishes an 
 exact index of the expansive force of tlie air witliin the bulb, 
 and hence the intensity of the radiant heat to Avhich it is sub- 
 
CHAP. xvi. BArxOMETBIC ACTiyoMETER. 273 
 
 jt'Cted. It is iiiipoitiiiit to observe, reganliiig the graduatiou 
 of the scale of temperature, that the height of the mercurial 
 column alone will not determine the length of the degrees. 
 Certain obvious corrections must be made, especially on ac- 
 count of the loss of energy of the sun's radiant heat in 
 passing through the crystal which covers the exhausted tube. 
 This loss may be readily ascertained by employing an appa- 
 ratus by means of which the absorptive property of the 
 crystal is tested before the scale of temperature is graduated. 
 Referring to the mechanism represented in the illustration, 
 it will be seen that the inclination of the exhausted cylin- 
 der is regulated by a tangential screw a, while the adjust- 
 ment called for by the earth's diurnal motion is effected 
 by a small pinion b actuating the cogged base-plate which 
 supports the instrument. The mechanism thus described, 
 together with a perforated sight c and an index-plate d, 
 enables the operator to direct the tube accurately towards 
 the sun. The centre line of the index (/ attached to the 
 upper end of the exhausted chamber shows the zenith dis- 
 tance on the graduated quadrant at all times by raei'e inspec- 
 tion. It should be mentioned that, when the barometric 
 actinoiueter is arranged within a revolving observatory pro- 
 vided with a declination table, the mechanism just described 
 for obtaining a parallactic movement is wholly dispensed 
 ^vith. Meteorologists will do well to adopt such an insti-u- 
 ment in all important observations, since its simultaneous 
 indication of solar intensity and zenith distance enables them 
 to determine the relative amount of vapor jiresent in the 
 
274 EADIANT HEAT. ' chap. xvi. 
 
 atmosphere ^yith a degree of precision probably unattainable 
 by any other means. The sensitive nature of the instrument 
 will be readily comprehended if we consider that while the 
 surface of the bulb, 2.75 ins. diameter, amounts to 5.93 square 
 ins. (the section of the pencil of rays which imparts the 
 radiant heat contains also 5.93 superficial inches), the quan- 
 tity of matter to be heated is only a small fraction of that 
 contained in an ordinary thermometer whose bulb is only 
 0.6 in. diameter. Besides, the latter bulb receives heat from 
 a pencil of rays of only 0.28 square in. section. Apart froni 
 this important difference in favor of the barometric actino- 
 meter, the radiating surface of spherical bulbs exposed to 
 solar radiation, as already stated, is twice as great as the 
 section of the pencil of rays Avhich imparts heat to the same. 
 
CHAPTER XVII. 
 
 CONDUCTIVITY OF MERCURY. 
 
 It was shown in Chapter XV. that the radiant intensity 
 of the sun cannot be accurately ascertained by the thermo- 
 heliometer employed by Pere Secchi, owing, among other 
 causes, to the imperfect conductivity of the mercury in the 
 bulb exposed to the sun. ^leteorologists are not generally 
 aware of the fact that the conducting power of mercuiy is 
 so imperfect as to affect materially the correctness of the 
 indication of mercurial thermometei-s, Deschanel being quoted 
 in support of the opinion that mercury is a very good con- 
 ductor. Prof. Everett, in a recent translation of the works 
 of the author mentioned, assumes that the conductivity of 
 quicksilver in the bulb of a thermometer is the same as a 
 vessel " with thin metallic sides containing water which is 
 stirred" (see Prof. Everett's translation of "Deschanel's Natu- 
 ral Philosophy," Part II., pp. 245-387). The subject is so 
 intimately connected with the detennination of solar tempe- 
 rature and solar energy that it has become necessaiy to 
 
276 
 
 RADIANT HEAT. 
 
 CHAP. XVII. 
 
 settle tbe question by some reliable, practical test. I have 
 accortlingly constructed the apparatus illustrated ou Plate 
 28, by means of wliicli the relative conducting j)ower of a 
 colunm of copper and one of mercury has been ascertained 
 with ci'itical nicety. Before entering on a description, it will 
 be instructive to point out that the heat communicated to 
 the bulb of a thermometer by solar radiation is transmitted 
 to its contents chiefly by convection; hence that the altitude 
 
 of the sun during the observation influences the accuracy of 
 the indication. This will be readily comprehended. Fig. 2 
 represents the bulb of a thermometer exposed to the rays 
 Avhen the sun's zenith distance is 65 deg. ; Fig. 3 I'epresenting 
 the bulb when the zenith distance is 18 deg. 23 min., the 
 latter being the minimum at the observatory of the Roman 
 College, where the thermoheliometer described in Chajiter 
 XV. has been long emjiloyed for the purpose of ascertaining 
 
CHAP. XVII. COl^DUCTIVITY OF MEBCURT. 277 
 
 the inten.sity of solar radiation. Referring to Fig. 2, it will 
 be seen that the blank crescent c, whose varying thickness 
 indicates veiy nearly the relative amonnt of heat imparted 
 at each point of the spherical surface pi-eseiited towards the 
 sun, occupies an almost vertical position. The mercury con- 
 tained within the space indicati-d by the said crescent, hav- 
 ing its specific gravity reduced by the heat transmitted by 
 the solar rays, will ascend ; w hile tlie mercury on the f)pposite 
 side, which retains its specilic gravity, will descend ; thus a 
 circulation will be established by means of which the heat 
 received from the sun will be gi'adually communicated to the 
 entire mass of mercur}^ in the bulb. But when the latter 
 is exposed to the sun's rays under a zenith distance of about 
 18 deg., as shown in Fig. 3, the heated mass (jf mercury con- 
 tained within the crescent a has so slight an im-lination that 
 scarcely any circulation takes place. Consequently, if it can 
 be sho\\Ti practically that mercury is incapable of transmit- 
 ting heat from particle to particle with sufficient velocity, 
 it will be evident that thermometei's and thermoheliometers 
 with spherical bulbs are worthless as means of measuring 
 maximum intensity of solar radiation. It will be perceived 
 that if the bulb in Fig. 3 be surrounded by an enclosure, as 
 in the thei'iuoheliometer, the mercury contained witliin the 
 space indicated by the crescent h will radiate far less heat 
 towards such enclosure than the mercury within the opposite 
 heated crescent a. It will also be perceived that, by increas- 
 ing the size of the bulb, the transmission of heat from a to h 
 will be retarded unless the conductivity of mei-cury ]>e pei-- 
 
278 RADIANT HEAT. chap. xvii. 
 
 feet. Heuce the size of the bulb is an element aflFecting the 
 accuracy of the indication — a circumstance fatal to the em- 
 ployment of a sjiherical bulb in the thermoheliometer. 
 
 The nature of the illustrated apparatus constructed for 
 the determination of the conductivity of mercury will be 
 readily understood by the following description: Fig. 1, 
 Plate 28, represents a longitudinal section through the verti- 
 cal plane, a is a boiler, with a flat bottom and semicircular 
 ends, supported on two columns / and g, resting on the 
 bottom of the cisterns c and d. The column / is composed 
 of wrought copper plated with silver, highly polished. The 
 column g consists of a cylindrical vessel of glass open at 
 the top, filled with mercury, and surrounded with a socket 
 h, composed of polished silver. The cisterns c and d, sup- 
 ported on non-conducting substances, are plated with polished 
 silver, and provided with funnel-shaped openings at the top, 
 through which thermometers are inserted. These cisterns, 
 as well as the columns / and g, are surrounded with non- 
 conducting coverings j9, ^p and o, o. A lamp h is applied 
 between the cisterns for heating the water in the boiler. It 
 is scarcely necessary to observe that the polished silver plat- 
 ing of the copper column, and the polished silver socket 
 round the mercurial column, are intended to prevent loss of 
 heat by I'adiatitm, while the coverings before mentioned are 
 intended to prevent loss of heat by convection attending 
 atmospheric currents. The inside diameter of the cylindrical 
 vessel g, it should be noticed, is 0.5 in., corresponding exactly 
 with the diameter of the copper column f, the top of which 
 
CHAP. XVII. COXDVCTIVITT OF MEBVUEY. 279 
 
 is on a level with that of the mercurial column. The lines 
 Ic I and m n are in the same horizontal plane, their distance 
 below the upper ends of the columns / and g being precisely 
 2 inches. 
 
 The object of the apparatus being that of comparing the 
 conductivity of mercuiy to that of some other metal, copper 
 has been selected, as its conducting property is better known 
 than that of any other. The leading feature of the arrange- 
 ment will be comprehended by a mere glance at the illustra- 
 tion. An equal amount of heat being applied to the top of 
 each column, it is intended to show by the elevation of the 
 temperature of the water in the cisterns c and d what rela- 
 tion exists between the conductivity of mercury and copper. 
 Regarding the application of the heat, it will be evident that 
 an equal amount must infallibly be imparted to each column 
 if the lamp be sufficiently powerful to keep the water in a 
 state of continuous ebullition. Obviously the heat from the 
 lamp, if urged, will cause a rapid upward motion of the 
 water in the middle of the boiler, and a correspondingly 
 rapid descending current at each end. Accordingly, lateral 
 currents, varpng in velocity with the strength of the flame, 
 applied under the boiler, mil fl.ow inwards over the upper 
 ends of the columns / and g. 
 
 Several experiments have been made under different baro- 
 metric pressure and different atmospheric temperature, yet 
 the results as regards the comparative conductivity of mer- 
 cuiy and copper have proved to be very nearly alike in all. 
 The accompanying tables record the result of the last trial, 
 
380 
 
 BADIANT HEAT. 
 
 CHAP. XVII. 
 
 Table I. — Coppeb Column. 
 
 
 
 
 '^ 
 
 A 
 
 
 n 
 
 
 o 
 
 .3 
 
 ±3 
 
 ^a C 
 
 o 
 
 J^ ® 
 
 6 
 
 3 ^ S 
 
 35 a 
 
 
 
 0J« 3 
 
 tec— 
 
 s 
 
 a .s 
 
 
 gjS. 
 
 
 til 
 
 
 H 
 
 o 
 
 1 
 
 
 1 ' 
 
 1 ^ 
 
 Min. 
 
 'Fah. 
 
 °Fah. 
 
 Therm, mitts. 
 
 °J?aA. 
 
 Therm, units. 
 
 TAerm. uni^. 
 
 
 73.50 
 
 
 
 138.50 
 
 
 
 0.5 
 
 75.15 
 
 1.65 
 
 0.143 
 
 136.85 
 
 0.143 
 
 104.873 
 
 1.0 
 
 77.25 
 
 3.75 
 
 0.326 
 
 134.75 
 
 0.183 
 
 134.208 
 
 1.5 
 
 80.14 
 
 6.64 
 
 0.577 
 
 131.80 
 
 0.251 
 
 184.078 
 
 2.0 
 
 83.84 
 
 10.34 
 
 0.898 
 
 128.10 
 
 0.321 
 
 235.415 
 
 2.5 
 
 88.14 
 
 14.64 
 
 1.272 
 
 123.80 
 
 0.374 
 
 274.284 
 
 3.0 
 
 92.81 
 
 19.31 
 
 1.678 
 
 119.00 
 
 0.406 
 
 297.752 
 
 3.5 
 
 97.67 
 
 24.17 
 
 2.100 
 
 114.25 
 
 0.422 
 
 309.486 
 
 4.0 
 
 102.56 
 
 29.06 
 
 2.525 
 
 109.44 
 
 0.425 
 
 311.686 
 
 
 
 Table II 
 
 . — Mercurial Column. 
 
 
 H 
 
 
 .3 
 3 • 
 
 a a " 
 «_2 
 
 1 
 i s 
 If 
 
 be A 
 
 III 
 
 Si 
 
 r3 
 
 IS 
 1 ^ 
 
 ^3 
 
 c .'c 
 
 Min. 
 
 °J?'aA. 
 
 °^aA. 
 
 Therm. «»j's. 
 
 - /'-a/i. 
 
 T/ierm. units. 
 
 T/ierm. units. 
 
 
 73.50 
 
 
 
 138.50 
 
 
 
 0.5 
 
 73.52 
 
 0.02 
 
 0.002 
 
 138.48 
 
 0.002 
 
 1.466 
 
 1.0 
 
 73.56 
 
 0.06 
 
 0.005 
 
 138.44 
 
 0.003 
 
 2.200 
 
 1.5 
 
 73.64 
 
 0.14 
 
 0.012 
 
 138.36 
 
 0.007 
 
 5.133 
 
 2.0 
 
 73.75 
 
 0.25 
 
 0.022 
 
 138.25 
 
 0.010 
 
 7.334 
 
 2.5 
 
 73.90 
 
 0.40 
 
 0.035 
 
 138.10 
 
 0.013 
 
 9.534 
 
 3.0 
 
 74.08 
 
 0.58 
 
 0.051 
 
 137.92 
 
 0.016 
 
 11.734 
 
 3.5 
 
 74.28 
 
 0.78 
 
 0.068 
 
 137.72 
 
 0.017 
 
 12.467 
 
 4.0 
 
 74.50 
 
 1.00 
 
 0.087 
 
 137.50 
 
 0.019 
 
 13.934 
 
CHAP. XVII. COXDUCTIVITY OF MEUCUUY. . 281 
 
 comlucted as carefully as practicable. The lieailiiigs of the 
 several columus explain so clearly the object of the tables 
 that it will only be necessary to ^ate that the energy in- 
 serted iu the fourth column is the energy developed from the 
 beginning of the experiment 
 
 Referring to Table I., it will be seen that at the termina- 
 tion of 4 minutes from the commencement of the experiment, 
 the temperature of the water in the cistern v bad increased 
 29'.0G, the differential temperature being then 212° - 102°.56 
 = 109°.44 F. During the same period an amount of dynamic 
 enero-y represented by 2.525 thermal units had been trans- 
 mitted past the line l 1, communicated to (1) the water in 
 the cistern ; (2) the part of the copper column immersed ; 
 (3) the metal composing the cistern ; (4) the immersed part 
 of the thermometer. But, while the entire energy transmitted 
 past the line I- I during the 4 minutes thus amounted to 2.52.3 
 units, the rate of transmission was actually 0.850 unit per 
 minute at the termination of the fourth minute. This appa- 
 rent discrej>ancy was caused by the heat absorbed Ijy that 
 part of the column which extends above the line h I, the 
 temperature at the commencement of the experiment being 
 the same as that of the surrounding air, 73'.50. Referring 
 to Table II., it will be seen that the euerg>=- transmitted 
 through the mercurial column past the line m n, during 4 
 minutes, was only 0.087 unit against 2.525 units for the 
 copper column, although the differential temperatui-e of the 
 water in the cistern ^^ was 137°.50 - 109'.44 = 28^06 higher 
 than in the cistern c. Accordingly, the conductivity of the 
 
282 RADIANT HEAT. chap. xvii. 
 
 2.526 
 copper composing the column /' has proved to be ;: = 
 
 29.06 times greater than the conductivity of tlie mercury of 
 the column g, notwithstanding the higher differential tempe- 
 rature to which the latter was exposed. It will be observed 
 that the glass, 0.02 in. thick, composing the cylindrical vessel 
 which contains the mercury, will conduct some heat down- 
 ward, tending to increase the temperature in the cistern d. 
 This tendency, however, will be balanced by the loss of 
 heat occasioned by the radiation of the glass cylinder, since 
 the application of the polished silver socket and the non- 
 conducting covering cannot wholly prevent the refrigerating 
 action of the surrounding air. It is important to observe, 
 regarding the loss of heat from the latter cause, that the cis- 
 terns, previous to trial, are charged with water of the same 
 temperature as the atmosphere. No\v, considering that the 
 increment of temperature in the cistern d does not averao-e 
 more than 0°.40 above that of the atmosphere during the 
 trial, it will be evident that the amount of error caused by 
 radiation will be quite inappreciable. We are therefore 
 warranted in concluding that the conductivity of mercury, 
 determined by the increment of temperature in cistern d, 
 and by the dynamic energy transmitted past the line m n, 
 cannot be far from correct. It will be asked why columns 
 of such small diameter have been employed. The principal 
 object has been that of presenting a sectional area in the 
 mercurial column g corresponding as nearly as possible to 
 the size of the bulb of an ordinary thermometer. The inves- 
 
CHAP. XVII. COi\J)VCTIViri' OF MEIiCVliY. 283 
 
 tigation, then, has couclu.sively established the fact that mer- 
 cury transmits heat from particle to particle too slowly to 
 effect a sufficiently rapid indication of mercurial thermome- 
 ters provided with splierical bulbs ; and that, when the heat 
 is applied from above, the indication of such thermometers 
 is wholly um'eliable. 
 
 A subject of great interest presents itself in connection 
 with the rate of transmission of energy exhibited in the 
 sixth column of Table I. It will be seen that, although the 
 copper column / is only 0.5 in. in diameter = 0.19635 sq. in. 
 section, the rate of transmission at the termination of the 
 fourth minute is 0.850 unit per minute. Reducing this 
 amount to the usual standard of one square foot, it will be 
 
 144 
 
 found that the energy developed is X 0.850 = 623 
 
 "'•' ^ 0.19635 
 
 thermal units per minute for a sectional area of one square 
 
 foot. It will be observed that this extraordinary amount of 
 
 • ,, ,, . .623 
 
 energy (theoretically capable of exerting = 14.5 horse- 
 
 42.7 
 
 power) is called forth by the moderate differential tempera- 
 ture of 212° - 102°.56 = 109°.44 F. Now, let us compare 
 the stated energy of 623 thermal units per minute to that 
 produced by the radiation of a metallic surface coated with 
 lamp-black, and maintained at a temperature of 212" within 
 an enclosure of 102^ Actual trial shows that, under these 
 conditions, the radiant energy emanating from the face of a 
 plate composed of copper, containing 144 sq. ins., scarcely 
 reaches 6 theraial units per minute. Our experiment has 
 
284 BABIANT HEAT. chap. xvir. 
 
 therefore incidentally established the fact that, under the 
 stated conditions, a plate of wrought copper 2 ins. in thick- 
 ness is capable of transmitting by conduction from one side 
 to the other, in a given time, an amount of mechanical energy 
 more than one huntlred times greater than the mechanical 
 energy transmitted by the radiation of the same plate during 
 an equal interval of time. 
 
CHAPTER XYIII. 
 
 INCANDESCENT CONCAVE SPHEPJCAL HADIATOK. 
 
 The illustration on Plate 29 represents an apparatus 
 constructed for the purpose of j^roving the correctness of 
 the indications furnished by the solar pyrometer described 
 in Chapter X. Fig. 1 is a side elevation and Fig. 2 an 
 end view of the apparatus. Objections have been raised 
 against the solar pyrometer on account of the low tempera- 
 ture employed. It is contended that, unless the radiator is 
 raised to the temperature of incandescence, emitting luminous 
 rays, the radiant heat transmitted to the focus will not fui'- 
 nish an indication capable of determining the temperatui-e 
 of distant incandescent bodies. The reader is aware that 
 the idea of ascertaining the temperature of the sun by the 
 indications of a surface coated with lamp-black, maintained 
 at only boiling heat, has lieen deemed absurd by certain 
 pliysicists. Secchi, in a letter to Nature, says : " Very few 
 indeed \vill allow tliat which ]Mr. Ericsson takes for granted, 
 that the radiating power of the solar materials may be com- 
 
 3^ 
 
286 RADIANT HEAT. chap, xviii. 
 
 pared to tLat of pure lamj^-black, as lie assumes." Numerous 
 experiments, lio\vever, sliow tliat, relatively, there is no appre- 
 ciable difference between tlie energy of the dark heat-rays 
 emanating from a metallic radiator of low temperature, pre- 
 senting a thoroughly disintegrated or a blackened surface, 
 and the energy of heat-rays accompanied by a light emanat- 
 ing from an incandescent metallic radiator. The temperature 
 transmitted by the radiant heat to the focus is, in each case, 
 directly proportional to the temperature of the radiant sur- 
 face. An air thermometer placed in the focus of a concave 
 spherical radiator composed of ice, and surrounded with very 
 cold substances, say 100° below zero, will furnish an indica- 
 tion by which the temj^erature of distant incandescent bodies 
 may be ascertained with as much certainty as by employing 
 a radiator heated to such a degree as to emit luminous rays. 
 It scarcely needs explanation that my reason for constructing 
 the solar pyrometer with a radiator kept at the low tempe- 
 rature of boiling water is that of admitting of operating 
 within a vacuum, besides rendering it possible to measure 
 the temjjeratures with positive exactness. No doubt the 
 instrument could be so arranged that the metallic radiator 
 might be maintained at a temperature considerably above 
 that of incandescence (Sir Humphry Davy, it will be remem- 
 bered, fixed the temperature of incandescence at 812°); but 
 we lack accurate means of measuring the intensity when 
 metals are brought to white heat or bright orange. Nor 
 would anything be gained by resorting to a mode of con- 
 struction involving both complication and uncertainty, since 
 
UHAI-. xviii. INCANDESCENT SPHERIOAL RADIATOR. 287 
 
 (laik heat-rays, with refereuce to temperature, in no manner 
 differ fi'oni heat-rays accompanied by light. As already 
 stated, no irregularity h;is been observed by me in the fall 
 of the temperature of an incandescent radiator, and that of 
 the focal thermometer exposed to the I'adiaut heat, while the 
 color gradually changes fx'om bright orange to black. On 
 the contrary, the temperatures of the radiator, and the reci- 
 pient of the radiant heat, continue to bear the same relation 
 to each other during both cooling and heating. The times, 
 compared mth the increment or diminution of intensity, 
 differ a little; but, as stated, the proportion between the 
 temperature of the radiant surface and that transmitted to 
 the focus continues as nearly uniform as practical test can 
 show. 
 
 The radiator of the instrument represented by our illus- 
 tration on PI. 29 consists of a solid cylindrical block h, com- 
 posed of cast iron, 10 ins. diameter, 6 ins. long, placed hori- 
 zontally on a pedestal, the fi-ont end forming a spherical 
 concavity a h c of 18 ins. radius, precisely like the radiator 
 of the solar pyrometer described in Chap. X. The under side 
 of the cylindrical block is provided with a square projection 
 corresponding with two guide pieces on the top of the 
 pedestal shown on Fig. 2, intended to facilitate the operation 
 of placing the block rapidly in a proper position after having 
 been heated in an aii--furnace. A focal thermometer J, simi- 
 lar to the one employed in the solar pyrometer, is secured 
 to a bent arm attached to the front side of the pedestal ; 
 the distance b cl between the centre of the bulb and the face 
 
BABJANT HEAT. 
 
 CHAP. XVIII. 
 
 a b c oi the concave spherical radiator being also precisely 
 as in the sohar p}'rometer. 
 
 The accouipauyiug table exhibits the result of a trial of 
 the ajjparatus, conducted at a mechanical establishment in 
 New York possessing air-furnaces well adapted for the in- 
 vestia:ation : 
 
 Appearance of 
 radiator. 
 
 Temperature of radiator. 
 
 is, 
 
 <l-2 
 
 Temperature of focal 
 thermometer. 
 
 Actual. 
 
 Diflerential. 
 
 Actual. 
 
 Differential. 
 
 
 ' Fah. 
 
 ° Fah. 
 
 ° Fah. 
 
 ° Fah. 
 
 'Fah. 
 
 Light orange.... 
 
 Deej) orange 
 
 Br't cherry red.. 
 Pull cherry red. 
 Dull cherry red. 
 Dull red heat. . . . 
 
 Mean 
 
 2190 
 
 2010 
 1830 
 1650 
 1470 
 1290 
 
 2149.3 
 1969.3 
 1789.3 
 1609.5 
 1429.8 
 1249.0 
 
 40.7 
 40.7 
 40.7 
 40.5 
 40.2 
 41.0 
 
 178 
 173 
 166 
 150 
 144 
 130 
 
 137.3 
 132.3 
 125.3 
 115.5 
 103.8 
 89.0 
 
 1740 
 
 1699.87 
 
 40.63 
 
 157.83 
 
 117.2 
 
 
 The following brief account of the manner of conducting 
 the trial \vill show the simple nature of the investigation. 
 Tlie solid radiator, Ijefore being 2)laced on the pedestal, Mas 
 heated in the air-furnace to very nearly white heat, and 
 then, by means of tongs, quickly removed from the furnace 
 and placed in the position shown by the illustration. The 
 focal thermometer was then closely observed, its indication 
 being recorded when the radiator had cooled so as to pre- 
 sent a color of light orange. The indications during the 
 succeeding stages of brightness and color of the incandescent 
 
ciiAi'. Win. Ii\CAM)J-:SCEM l^U'llKlilCAL RMHATOR. 280 
 
 radiator were in like iiiaiiucr iveoi'ded. The teiiipeniture of 
 
 the sui'rouucling air \\as> observed siiniiltaiieously with that 
 
 of the focal thermoiuetei'. The time Avhich elapsed between 
 
 tlie fii'st and last observation entered in the table was 29 
 
 minutes. It will be seen that the temperature transmitted 
 
 by the radiator to tLe focal thermometer was recoi'ded at 
 
 six different stages of incandescence, the color presented by 
 
 the radiant surface determining the time foi' olxservation. 
 
 The mean temperature of the radiator during the experiment 
 
 was 1,740° F. Deducting the mean atmospheric temperature, 
 
 40°. 63 F., the actual mean differential temperature of the 
 
 radiant surface, the lumint)us heat rays of ^vhicll acted on 
 
 the focal thermometer, was 1,699°.37. The mean temperature 
 
 transmitted to the focal thermometer exposed to the radiant 
 
 heat being 157°.83, while the atmospheric temperature, as 
 
 already stated, was 40°.63, we find that a temperature of 
 
 117°.2 was imparted to the focal thermometer by a radiant 
 
 intensity of 1,699°. 37. It will be recollected that in the 
 
 solar pyrometer a differential radiant intensity of 163°.9 
 
 transmitted a temperature of 12°.2 to the focal thermometer; 
 
 12°.2 
 hence ■ = 0.074 of the temperature of the i-adiator was 
 
 1 63.9 
 
 117.2 
 transmitted to its focus, against ^^^j^jT^r;: = 0.069 in the ap- 
 
 paratus under consideration. Conse([uently, 0.074 — 0.06i* 
 = 0.00.5 less heat, relatively, is transmitted by the incan- 
 descent radiator than by the comparatively cool radiator of 
 the solar pyi'ometer. As this small discrepancy can readily 
 be accounted for, the result of the instituted test fully estab- 
 
290 BABIANT HEAT. chap, xvili. 
 
 lislies the truth of the doctrine which forms the basis of 
 the sohu- pyrometer — namely, that the cakirific energy of both 
 dark and luminous heat rays is directly proportional to the 
 temperature of the radiant surface. The cause of the dis- 
 crepancy adverted to will be readily comprehended by the 
 following explanation relating to the solar pyrometer. The 
 heat imparted by the radiant to the recipient surface is 
 transmitted through ether alone; therefore neither the radi- 
 ator nor the bulb of tlie focal thermometer are subjected to 
 any loss by convection ; while the incandescent concave radi- 
 ator, as well as its focal thermometer, are exposed to the 
 refrigerating influence of the atmospheric air. Obviously 
 the heated bulb of the thermometer will caiise an up^vard 
 curi'ent of air, which, acting on its face, reduces its tempe- 
 rature and indication, while the intense heat of the radiator 
 tends to augment the said current. Again, the rapid suc- 
 cession of cold particles passing over the intensely heated 
 surface of the radiator will inevitably diminish the energy 
 of the radiant heat, since the molecular motion within the 
 heated mass cannot instantly restore the loss to which the 
 molecules at the siirface are continually l:)eing subjected by 
 the cold current. The diminution of radiant energy from 
 this cause, though not great, will be appreciable, and, added 
 to the loss of heat, to which the bulb of the focal thermo- 
 meter is subjected, satisfactorily accounts for the discrepancy 
 adverted to; at the same time showing the necessity of 
 carrying on investigations relating to radiant heat within a 
 vacuum. 
 
CHAP. XVIII. IXCAXl>i:SCi:XT arJlERICAL RADIATOn. 291 
 
 Let us now calculate the temperature of the sun agree- 
 ably to the indications furnished by the incandescent ladiatov 
 of the illustrated apparatus, without reference to the iiidica- 
 tioiis of tlie iiistrunu'iit, tlio relialjility of wliicli we are dis- 
 cussing. But in place of basing our calculations on the 
 angle subtended by the sun from the earth, and the angle 
 subtended by the concave spherical radiator from its focus, 
 let us detenniiie tlie solar tennuTature ou the basis of areas 
 and distances alone. This method will be more satisfactory 
 to practical men than the one which takes no direct cogni- 
 zance of areas and distances. Assuming the sun's diametei' 
 to be 852,584 miles, the area of the great circle will be 
 15,912,029 X 10" sq. ft. The diameter of the spherical radi- 
 ator being 10 ins. and the radius 18 ins., its face presents 
 80.06 sq. ins. = 0.556 sq. ft. Accoi'dingly, the sun's area 
 is 28,620,377 X 10" times greater than the area of the con- 
 cave face of the radiator. The mean distance between the 
 sun and the earth is 91,430,000 miles, or 482,750,400,000 ft. ; 
 the distance between the radiator and its focus is 1.5 ft. 
 The radiant heat of the sun, theiefore, acts through a dl.s- 
 tance 321,833,600,000 times greater than the radiant heat 
 of the incandescent radiator. We have demonstrated in 
 Chap. I. that the temperature transmitted to the foci of 
 concave spherical radiators of equal area is inversely as the 
 square of their radii ; and we have sho\vn that, owing to 
 the great distance of the sun, every part of his face may, 
 without material error in our computations, be considered as 
 equidistant fi-oni the earth. Hence, if we square and invert 
 
292 BABIANT HEAT. 
 
 CHAP, xvrii. 
 
 the lief ore-mentioned distances tbrongli whicli the I'.-idiant 
 heat acts, we ascertain that for equal intensity and equal 
 area the incandescent radiator will transmit 10;^, 57(5,800 X 
 10" times higher temperature to its focus than that trans- 
 mitted by the sun to the Iwundary of the earth's atmosphere. 
 But the area of the sun, as we have stated, is 28,620,377 X 
 10" times greater than the area of the radiator; hence for 
 equal inlensity the radiant heat transmitted to the focus of 
 
 . , ^^ .,, - 103,576,866 X 10" 
 
 the latter will be ^.^ '^^^ ^ ^^^,., = 3,618.99 times greater 
 28,620,3vY X 10 ° 
 
 than that transmitted by the sun. It will be readily seen, 
 
 on reflection, that unless the temperature of the sun is 
 
 3,018.99 times greater than that of the incandescent radiator, 
 
 it cannot transmit to the atmospheric boundary the same 
 
 temperature as that transmitted by the radiator to its focus, 
 
 viz., 117°.2 F. The temperature produced by solar radiation 
 
 when the earth is in aphelion is, however, only 84°.84: at 
 
 the said boundary; hence the sun's temperature need be 
 
 , 3,618.99 X 84.84 
 
 onlv — — = 2,619.25 times greater than that of 
 
 117.2 ^ 
 
 the incandescent radiator (1,099°.37), in order to cause an 
 
 elevation of 84°.84 on the Fahrenheit scale at the boundary 
 
 of the earth's atmosphere. Multiplying 1,099°.37 by 2,619.25, 
 
 we find that the indication of the incandescent concave 
 
 spherical radiator of our illustrated device proves the sun's 
 
 temperatxire to be 4,451,924° F. It will be seen, on referi-ing 
 
 to Chap. X., that the calculations based on the indications 
 
 of the solar pyrometer prove the sun's temjserature to be 
 
cnAP. XVIII. TNCJXPESCEyT SriTEHICAL HADTATOn. 293 
 
 only 4,063,984\ The cause of this discrepancy of 0.087 has 
 already been explained, viz., diminution of the radiant energy 
 of the incandescent radiator, produced by currents of cold 
 air sweeping over its face ; together with the loss of heat 
 to wJiich the unprotected bulb of the focal therinonieter is 
 subjected by the refrigerating effect of the surrounding at- 
 mosphere. Making due allowances for these losses — insepa- 
 i-al)le from conducting the e.xpei'inient in the presence of 
 atmospheric influence — it will be found that the indications 
 furnished by an incandescent concave spherical radiator 
 assign very iieaily the same temperature to the sun as the 
 comparatively cold radiator of the solar pyrometer. The 
 objection, then, urged against this instrument, that its tem- 
 jterature is not high enough, cannot be maintained in view 
 of the fact which we have established, that the intensity 
 deduced from its indication is not att'ected by employing an 
 incandescent radiator in place of one raised to merely boiling 
 heat. 
 
CHAPTER XIX. 
 
 REFLECTIVE POWER OF SILVER AND OTHER METALS. 
 
 Desciianel informs us that the reflection of calorific rays 
 has been satisfactorily determined hj the investigations of 
 Melloni, Laprovostaye, and Desains, and that these physicists 
 have practically ascertained the reflective power of polished 
 silver and other metals. He states, also, that Laprovostaye 
 and Desains have shown that, "contrary to wliat was pre- 
 viously supposed, the reflecting power varies according to the 
 source of heat." "Thus," he adds, "the reflecting power of 
 polished silver, which is 0.97 for rays from a Locatelli lamp, 
 is only 0.92 for solar rays." It follows from this announce- 
 ment that while the loss of radiant energy is only 1.00 — 
 0.97 = 0.03 Avhen the rays emanate from the lamp mentioned, 
 it is 1.00 — 0.92 = 0.08 when the rays from the sun are 
 reflected, xiccordingly, the loss of energy attending reflec- 
 tion will be nearly three times greater for solar heat than 
 for artificial heat. That such a difference does not exist is 
 known to all persons conversant with reflectors. Besides, a 
 
CHAP. MX. BEFLECTIVi: rOWEi: OF rOLlHUED METALS. 295 
 
 momeut's consideration of the properties of calorific rays suf- 
 fices to show the untenable character of the proposition, 
 compelling us to reject Laprovostaye's and Desains's investi- 
 gation as unrt'lial>le, although generally accepted by plij- 
 sicists. Moreover, the table of reflective power of various 
 metals presented by Deschanel as the result of the celebrated 
 investigation furnishes additional evidence of its unreliable 
 character. Let us select from Laprovostaye and Desains's 
 tabular statement the relative reflecting capacity of silver 
 and brass. The former is represented to be 0.97, while the 
 latter is 0.93 ; consequently, the reflective powers of silver 
 aii<l brass are supposed to be in the ratio of 1.000 : 978, 
 
 22 
 difference = —--—' Manufacturei's of reflectors practicallv 
 1000 ^ •' 
 
 acquainted with the subject are aware that silver possesses 
 a far greater reflective power compared with brass than that 
 indicated by the stated proportion. 
 
 The leading feature of my solar engine being that of 
 developing mechanical power by concentrating the sun's 
 radiant heat, the efficacy of various metals for this purpose 
 engaged my attention during the early stages of my labora 
 connected with utilizing solar energy for the production of 
 motive power. Taking for granted the correctness of Lapro- 
 vostaye's and Desains's statement of relative efficacy, I con- 
 structed reflectoi-s composed solely of brass, in oi-der to avoid 
 the great cost of silver-plating. The difference of reflecting 
 
 22 
 power asrreeablv to the researches referred to beincr 
 
•Zd6 BADIAXT HEAT. chap. xix. 
 
 ill favor of silver, I simply enlarged the reflectors iii that 
 
 proportion, expecting to produce the same result as with 
 
 22 
 
 silver-i>lated metals preseutins; less area. It proved, 
 
 ^ ^ ° 1000 ^ ' 
 
 however, im the tirst trial, that the reflective capability of 
 Ijrass is far inferior, and that the required temperature 
 could not be produced by brass I'eflectors ; the machine, con- 
 sequently, failing to operate as designed. A thorough in- 
 vestigation capable of determining the true reflective power 
 of various metals, therefore, became necessary; but before 
 commencing experiments, the method resorted to by Melloni, 
 Laprovostaye, and Desains, referred to by Deschanel, Avas 
 carefully examined. The said method is thus described in 
 his " Elementary Treatise on Natural Philosophy " : " The 
 substance under investigation is placed upon a circular plate, 
 Avliich is graduated round the circumference. The thermo- 
 electric pile is carried by a horizontal bar which turns about 
 a pillar supporting the circular plate. This bar is so ad- 
 justed as to make the reflected rays impinge upon the pile, 
 the adjustment being made by the help of the divisions 
 marked on the circular plate. In making an observation, 
 the bar is iirst placed so as to coincide with the prolonga- 
 tion of the said principal bar, and the intensity of direct 
 radiation is thus observed. The pile is then placed so as 
 to receive the reflected rays, and the ratio of intensity thus 
 obtained to the intensity of direct radiation is the measure 
 of the reflecting power." The accompanying sketch, Fig. 3, 
 represents a top view of the arrangement referred to in the 
 
CHAP. XIX. DEFLECTIVE POWEE OF POLISHED METALS. 
 
 297 
 
 above explanation. a represents a Locatelli lamp, /> the 
 graduatetl circular plate on wliicb is placed the polislied 
 reflecting substance b', and c the tbermo-electrie pile, d 
 shows a cluster of parallel calorific ra3's projected from the 
 lamp, through a perforation in the .screen f, toward.-; the 
 polished substance b'. The face of this polished substance 
 being jdaced at an angle of 45 deg. to the course of the 
 cluster of rays d, it will he evident that the I'ays will be 
 deflected at the same angle ; hence the cluster will be di- 
 
 rected as shown by h, ultimately striking the face of the 
 jiile e. The latter, it should be mentioned, is protected 
 against radiation from the lamp by a screen g. The tempe- 
 rature imparted by the deflected radiant heat having been 
 recorded, the pile is allowed to cool, and then placed in the 
 position c', after which the polished substance // is removed. 
 As the radiant heat transmitted by the parallel rays d ema- 
 nating from the lamp will now act directly on the face of 
 the pile, a higher temperature will obviously be produced 
 than when the pile occupied the position c. The difference 
 
298 BADIANT HEAT. chap. xix. 
 
 of temperature thus produced, Ave are told, indicates exactly 
 the amount of loss of radiant intensity attending the reflec- 
 tion of the rays by the polished substance b'. Laprovostaye 
 and Desains, as before stated, foimd that the heat trans- 
 mitted directly is to that reflected as 1.000 to 0.0V8, hence 
 the loss of energy = 0.022. The assumption of such perfect 
 reflective poAver being palpably erroneous, let us examine 
 carefully the adopted method in order to detect the cause of 
 the false deduction. Fig. 4 represents a top view of Lapro- 
 vostaye and Desains's arrangement, already described, but 
 drawn to a larger scale than in Fig. 3, similar letters of 
 reference being employed in both figures, d d represents 
 the cluster of calorific rays transmitted by the lamp to the 
 polished substance V, and by it reflected, as shown by li,. 
 towards the face of the pile c. It will be evident that the 
 section of the reflected central cluster of calorific rays de- 
 pends upon, and corresponds with, the face of the pile c, 
 provided the perforation in the screen / be sufiiciently large. 
 It will also be evident that the annular cluster Te ^, the 
 external diameter of which depends on the size of the per- 
 foration of the screen /, will be projected towards the 
 polished substance I'. Owing to defraction, the stated an- 
 nular cluster will expand to the size I I before reaching the 
 substance h', and thus the rays become dispersed over a 
 considerable portion of the angular face of b'. Consequently, 
 the latter will become heated, and, since calorific rays radi- 
 ate in all directions, a certain amount of heat will be trans- 
 mitted to the pile wholly independent of that propagated by 
 
CHAP. XIX. DEFLECTIVE roWEB OF POLTSFTED METALS. 200 
 
 the defected central cluster of rays represented by //. But, 
 after taking away the polished substance b', and moving the 
 pile to the position c' (shown in Fig. 3), the pile will obvi- 
 ously receive heat only from the central cluster d d. AVe 
 have thus demonstrated that more heat will be imparted to 
 the pile when placed in the position c than when placed at 
 c' in line with the cluster of rays d d, since in the latter case 
 the calorific energy due to the section of rays corresponding 
 with the area of the face of the pile can alone be transmitted 
 to the same. Apart from this cause of error, it should be 
 observed that, assuming the distance a b', Fig. 3, to be three 
 times greater than b' c', the diverging radiation from the 
 heated metal of the lamp will subject the substance b' to a 
 greater increase of temperature than that to which the pile 
 is subjected when at c', in tlie ratio of 4' to 3'. The effect of 
 this in causing undue transmission of heat to the pile, Avhen 
 placed at c, needs no explanation. Several other causes of 
 error inseparable from Laprovostaye and Desains's method 
 of determining the reflective power of metals might be 
 sho^^-n. Among these, the inaccuracy resulting from the un- 
 certain power of the lamp in developing a uniform amount 
 of heat during the experiments may be mentioned. In 
 view of the foregoing, it is evident that the thermo-elec- 
 tric method is unfit to determine accurately the reflective 
 power of different substances. And it may be demonstrated 
 that, unless the various substances under examination be 
 exposed simultaneously to a common source of heat, the rela- 
 tive power of reflecting calorific raj-s possessed by the same 
 
300 BADIANT BEAT. chap. xix. 
 
 cannot te ascertained witli perfect accuracy by any method 
 whatever. The instrument illustrated on PL 30 has been 
 constructed in accordance with the condition thus presented, 
 the source of heat emjjloyed being solar radiation. The fol- 
 lowing somewhat elaborate description and explanation have 
 been deemed necessary to point out clearly its peculiar fea- 
 tures. Fig. 1 represents a vertical section of the instrument 
 and the table to which it is attached, the latter being pro- 
 vided with parallactic mechanism by means of which its 
 face is kept at right angles to the sun during investiga- 
 tions. Fig. 2 shows a top view of the instrument as seen 
 from a point situated in the prolongation of its axis, at 
 right angles to the face of the table. « a is a conical re- 
 flector composed of cast iron, the sides of which are accu- 
 rately turned to an angle of 45° to its axis, a flat bottom 
 being attached provided with a central hub. An axle h is 
 firmly keyed in the said hub, extending both above and be- 
 low the same. The lower part of the axle turns in a boss 
 formed on opposite sides of a cross-piece c c, the latter being 
 supported by two columns bolted to the parallactic table. 
 A hand-wheel d, secured to the axle b, enables the operator 
 to turn the conical reflector during experiments. Four seg- 
 mental heaters /, g, Ji, and k, precisely alike, composed of thin 
 sheet metal, are secured to the bottom of the reflector, at 
 equal distance from the centre, with intervening spaces, as 
 shown in the drawing ; these spaces to be filled with some 
 non-conducting substance. Each heater is provided with a 
 conical socket on the top, into which a perforated cork is in-' 
 
CHAP. XIX. BEFLECTIYE POWER OF POLISHED METALS. 301 
 
 serted for the purpose of supporting thermometei-s entering 
 the fluid, as shown in the sectional representation. It is im- 
 portant to observe that, before being attached to tlie reflector, 
 (^ach heater should be fille<l with water of a given tempera- 
 ture and accurately \veighed. In case of any difference of 
 weight, small quantities of soft solder should be gradually 
 applied, until the deficient weight is made good and the 
 four charged heatera balance each other exactly. Regarding 
 the construction of the reflector represented by the illus- 
 tration, it remains to be stated that four segmental plates 
 f, </', 7i', and k', composed respectively of silver, brass, nickel, 
 and steel, precisely alihe in size and form, should be sol- 
 dered to the inside of the cone n n. The plates being 
 thus secured to the conical surface, the reflector should be 
 put before a rotating polishing machine, for the purpose of 
 having an equal polish imparted to each plate. 
 
 Referring to Fig. 2, jt will be seen that the polished seg- 
 mental plates, the reflective power of which it is intended 
 to measure, are attached opposite to the heaters. Thus the 
 plate f, composed of silver, is placed exactly opposite to 
 the heater // the brass plate g' opposite to the heater g ; 
 the remaining plates h' and Tc', composed respectively of 
 nickel and steel, being likewise placed opposite to their cor- 
 responding heatei*s h and h. 
 
 Referring to the vertical section of the instrument shown 
 in Fig. 1, supposed to be turned towards the sun, it will be 
 seen that the solar rays m, n, indicated hy dotted lines, are 
 reflected by the polished segmental plates towards the semi- 
 
30a RADIANT HEAT. chap. xix. 
 
 circular faces of the heaters. The areas of the polished 
 plates being alike, while the Aveight aud areas of the lieat- 
 ers presented to the reflected solar heat are also alike, it 
 follows that if the reflective power of the several plates 
 does not vary, the thermometers inserted in the heaters will 
 show an equal increase of temperature during equal inter- 
 vals of time. But should the segmental plates not possess 
 an equal power of reflecting the energy transmitted by the 
 solar rays, the thermometers will, after a brief exposure to 
 the sun, indicate different temperatures. Obviously the dif- 
 ferential temperature reached by each thermometer in the 
 same time will furnish a true indication of the relative re- 
 flective power of the sevei'al segmental plates, j)rovided 
 there are no disturbing causes affecting the heaters un- 
 equally. Before examining this important point, it will be 
 proper to observe that the internal diameter of the conical 
 reflector at n ii is 2.5 times greater than the external dia- 
 meter of the heaters at o o; hence the radiant intensity 
 transmitted to their faces will be 2.5 times greater than 
 the direct radiant intensity conveyed by the rays m n. It 
 shoiild be particularly observed that while the effect of this 
 concentration of heat will influence the four heaters alike, 
 it greatly increases the sensitiveness of the instrument, since 
 the thermometers "will be siibjected to a radiant energy 2.5 
 greater than that produced by the direct action of the sun's 
 rays, less the loss occasioned by the imperfect reflective 
 power of the segmental plates, together with the losses of 
 heat caused by convection and radiation. It will be seen, 
 
CHAP. XIX. BEFLECTIYE POWER OF POLISHED METALS. 303 
 
 on inspecting the vertical section in Fig. 1, tliat the lowei' 
 heater will be subjected to the cooling influence of the up- 
 ward current of the air caused by the heat imparted by 
 contact with the lower side of the heater. The upper 
 heater, on the other hand, ^\•ill not be subjected to any ap- 
 preciable loss of heat by convection, other than that which 
 is shared by the lower heater. It is hardly necessary to 
 point out that the conical reflector itself cannot be main- 
 tained at a uniform temperature all over, owing to the pow- 
 erful upAvard current of air unavoidable in a place exposed 
 to the sun's rays. The apparently insuperable difiiculties 
 thus presented, and nuuij^ others of minor importance, are 
 overcome by the simple expedient of causing the reflector to 
 revolve during the investigation. The rotary motion is ef- 
 fected, as already stated, by means of the hand-wheel d, 
 \\\\'\c\i is kept revolving at the rate of about six turns per 
 minute while the heaters are being exposed to the reflected 
 and concentrated solar heat. It needs no demonstration to 
 prove that by this expedient all disturbing influences will 
 affect the four heaters alihe, thereby neutralizing the same 
 as completely as if no disturbance existed. The nature of 
 the device having been thus minutely described, let us 
 now consider the mode of managing the same during the 
 investigation. (1) Before turning the reflector towards the 
 sun, bring the parallactic table to a horizontal position. 
 
 (2) Remove the corks and thermometers, and fill the heat- 
 ers with water of the same temperature as the atmosphere. 
 
 (3) Keplace the corks and thermometers, and bring the 
 
301 ItABIAyT MEAT. chap. xix. 
 
 table to an inclined position corresponding with the sun's 
 zenith distance. (4) In order to ensure a perfect mixture 
 of the particles of water in the heaters, turn the reflector 
 for about two minutes. (5) The rotation being then stop- 
 ped, note carefully the indication of the thermometers, and 
 keep a separate record for each heater and corresponding 
 segmental plate. (6) The initial temperatures having been 
 thus ascertained, and the parallactic table and reflector 
 turned to the sun, the hand-wheel should be kept in con- 
 tinuous motion, at the I'ate before mentioned, for a period 
 of ten minutes. (7) At the expii-ation of the stated time 
 the motion to be suflficiently reduced to admit of reading 
 the thermometers, two observers being emplo}ed for that 
 purpose. If the investigation be conducted at noon, dniing 
 the winter season — the most favorable for accurate observa- 
 tion — it will be well to continue the experiment for fifty 
 minutes, the indications of the thermometers being noted at 
 the termination of eveiy other minute, after the expiration 
 of the fii'st ten minutes before adverted to. It might be sup- 
 posed that an observation at the expiration of the 10th and 
 50th minutes Avould suflice to ascertain the dift'ereutial tem- 
 perature imparted by the reflected and concentrated rays to 
 each heater; and that by the observed difference of indica- 
 tion the comparative reflective power of the several seg- 
 mental plates might be accurately determined. But it will 
 be perceived, on due consideration, that in case of defects, 
 such as imperceptible leaks or impei'fect circulation, the re- 
 corded temperature would lead to a false determination of 
 
CHAP. XIX. EEFLECTIVE rOWEli OF FOLiaUED METALS. 305 
 
 the comparative reflective properties. Numerous observa- 
 tions at definite intervals during tlie experiment will evi- 
 dently serve as effectual checks. Of course, if equal reflective 
 power of the respective plates results from each of these 
 observations, then the accuracy of the determination may be 
 regarded as absolutely certain. Regarding the initial tempe- 
 rature of the heatei-s, it will be perceived that if at starting 
 the indications of all the thermometers corresponded, the 
 comparison of the reflective power of the plates would be 
 very simple ; but experience has shown that a perfectly equal 
 temperature at starting is impracticable. Consequently, each 
 plate, with its corresponding heater and thermometer, calls 
 for a separate record. 
 
 It is worthy of notice that investigations intended to deter- 
 mine the dynamic energy of solar radiation are rendered very 
 diflioult, because the heat acts on the upper part of any fluid 
 intended to absorb it. Artificial means must therefore be 
 employed to cause circulation, or mixture, of the particles 
 within the mass exposed to the solar rays. The difficulty 
 attending artificial circulation, and the complicated character 
 of the means necessary to promote a thorough mixture of 
 particles when a fluid is heated from above, has been shown 
 in Chapter V. Nor can it be denied that, notwithstanding 
 the elaborate character of those means, the object has been 
 but partially attained. In the present instance, however, 
 the circulation or mixture of heated and colder particles of 
 the fluid within the heaters is absolutely perfect ; since, for 
 each revolution of the reflector, the colder particles are trans- 
 
306 RADIANT HEAT. CHAP. XIX. 
 
 ferred from the buttuiu to the top of the vessel, tmd vice 
 versd. 
 
 Before presenting the result of our iuvestigatiou of the 
 comparative reflective power of silver aud othei' metals, it 
 will be necessary to explain the nature of the annexed tables. 
 We have just stated that, owing to the impossibility of ensur- 
 ing a uniform initial temperature, each of the four polished 
 plates attached to the inside of the conical reflector has called 
 for a separate record. The tables, accordingly, contain four 
 distinct divisions, headed respectively silver, hra^is:, niclcel, and 
 steel ; each division having a different initial temperature, viz., 
 silver, G5°.7 ; brass, 66°. 1 ; nickel, 65°. 4 ; and steel, 6.j°.l F. As 
 the temperatures recorded in the table commence at the ter- 
 mination of the toitli minute from starting, the oi-iginal initial 
 temperatures have not been entered, excepting that of the 
 heater exposed to the solar rays reflected by the silver plate, 
 Avhicli forms the standard of comparison. The headings of 
 the several columns of the tables being sufficiently explana- 
 tory, it will only be necessary to point out the mode of deter- 
 mining the relative reflective power of the segmental plates. 
 Let us select the nickel plate. Eeferring to the table, it will 
 be found that, at the termination of the 30th minute from 
 starting, the thermometer inserted in the heater li indicated 
 89°.5, while the initial temperature in that heater was 65°.4 ; 
 hence an increase of 89.5 — 65.4 = 24°.l during 30 minutes. 
 During the same period the temperature of the heater /, sub- 
 jected to the reflected solar rays from the silver plate /'', it 
 will be seen, by referring to the table, increased 96.4 — 65.7 
 
CH vi'. XIX. liEFLECriVE I'OWEU OF POLISHED METALS. 
 
 307 
 
 
 I'able siiowino Tin 
 
 liKLATIVi 
 
 Reflective Powei 
 
 OF 
 
 
 
 Polished Metals. 
 
 
 
 a 
 
 
 SILVES. 
 
 
 
 SRASS. 
 
 
 -si 
 
 If 
 
 ■si 
 
 11 
 5| 
 
 li 
 
 3| 
 
 ll 
 
 O H 
 
 IB 
 
 £ '5 
 
 ■gas 
 
 •7i 
 
 Jftn. 
 
 ' F^ih. 
 
 'Fah. 
 
 °I-ah. 
 
 'Fah. 
 
 'Fall. 
 
 Ratio. 
 
 10 
 
 78.8 
 
 13.1 
 
 65.7 
 
 77.4 
 
 11.6 
 
 0.885 
 
 12 
 
 80.9 
 
 15.2 
 
 65.7 
 
 79.6 
 
 13.4 
 
 0.881 
 
 14 
 
 83.1 
 
 17.4 
 
 65.7 
 
 81.5 
 
 15.4 
 
 0.885 
 
 16 
 
 8o.0 
 
 19.3 
 
 65.7 
 
 83.2 
 
 17.1 
 
 0.885 
 
 IS 
 
 87.1 
 
 21.4 
 
 65.7 
 
 85.1 
 
 19.0 
 
 0.887 
 
 20 
 
 88.6 
 
 22.9 
 
 65.7 
 
 86.4 
 
 20.3 
 
 0.886 
 
 22 
 
 90.2 
 
 24.5 
 
 65.7 
 
 87.7 
 
 21.6 
 
 0.881 
 
 24 
 
 91.6 
 
 25.9 
 
 65.7 
 
 89.2 
 
 23.1 
 
 0.891 
 
 26 
 
 93.3 
 
 27.6 
 
 65.7 
 
 90.8 
 
 23.7 
 
 0.858 
 
 28 
 
 94.9 
 
 29.2 
 
 65.7 
 
 92.2 
 
 26.1 
 
 0.894 
 
 30 
 
 96.4 
 
 30.7 
 
 65.7 
 
 93.5 
 
 27.4 
 
 0.892 
 
 32 
 
 97.6 
 
 31.9 
 
 65.7 
 
 94.3 
 
 28.2 
 
 0.884 
 
 34 
 
 98.4 
 
 32.7 
 
 65.7 
 
 95.4 
 
 29.3 
 
 0.896 
 
 36 
 
 99.7 
 
 34.0 
 
 65.7 
 
 96.5 
 
 30.4 
 
 0.894 
 
 38 
 
 100.8 
 
 35.1 
 
 65.7 
 
 97.3 
 
 31.2 
 
 0.888 
 
 40 
 
 102.0 
 
 36.3 
 
 65.7 
 
 98.4 
 
 32.3 
 
 0.889 
 
 42 
 
 103.4 
 
 37.7 
 
 65.7 
 
 99.5 
 
 33.4 
 
 0.886 
 
 44 
 
 104.0 
 
 38.3 
 
 65.7 
 
 100.0 
 
 33.9 
 
 0.885 
 
 46 
 
 104.7 
 
 30.0 
 
 65.7 
 
 100.5 
 
 34.4 
 
 0.882 
 
 48 
 
 105.3 
 
 39.6 
 
 65.7 1 
 
 100.9 
 
 34.8 
 
 0.878 
 
 50 
 
 105.9 
 
 40.2 
 
 65.7 1 
 
 101.4 
 
 35.3 
 Mean 
 
 0.878 
 
 = 0.885 
 
308 
 
 RADIANT HEAT. 
 
 CHAP. XIX. 
 
 
 Table showing the Relative Eeflective Power of 
 
 
 
 Polished Metals. 
 
 
 
 H 
 
 NICKEL. 
 
 STEEL. 
 
 la 
 
 c 1 
 
 Increase of 
 temperature. 
 Initial, C5°.4. 
 
 £ '3 
 
 || 
 
 c a 
 
 las" 
 
 *5 c; to 
 
 Min. 
 
 ' Fall. 
 
 °Fah. 
 
 Ratio. 
 
 ° Fah. 
 
 ° Fah. 
 
 Ratio. 
 
 10 
 
 75.5 
 
 10.1 
 
 0.771 
 
 74.1 
 
 9.0 
 
 0.687 
 
 12 
 
 77.2 
 
 11.8 
 
 0.776 
 
 75.6 
 
 10.5 
 
 0.684 
 
 14 
 
 78.8 
 
 13.4 
 
 0.770 
 
 77.0 
 
 11.9 
 
 0.683 
 
 16 
 
 80.3 
 
 14.9 
 
 0.772 
 
 78.4 
 
 13.3 
 
 0.689 
 
 18 
 
 82.0 
 
 16.6 
 
 0.776 
 
 79.9 
 
 14.8 
 
 0.691 
 
 20 
 
 83.2 
 
 17.8 
 
 0.777 
 
 81.1 
 
 16.0 
 
 0.699 
 
 22 
 
 84.5 
 
 19.1 
 
 0.779 
 
 82.3 
 
 17.2 
 
 0.702 
 
 24 
 
 85.7 
 
 20.3 
 
 0.783 
 
 83.4 
 
 18.3 
 
 0.706 
 
 26 
 
 8G.7 
 
 21.3 
 
 0.772 
 
 84.2 
 
 19.1 
 
 0.692 
 
 28 
 
 87.6 
 
 22.2 
 
 0.760 
 
 85.9 
 
 20.8 
 
 0.712 
 
 30 
 
 89.5 
 
 24.1 
 
 0.785 
 
 87.1 
 
 22.0 
 
 0.716 
 
 32 
 
 90.4 
 
 25.0 
 
 0.783 
 
 87.7 
 
 22.6 
 
 0.708 
 
 34 
 
 91.4 
 
 26.0 
 
 0.795 
 
 88.5 
 
 23.4 
 
 0.715 
 
 36 
 
 92.3 
 
 26.9 
 
 0.791 
 
 89.5 
 
 24.4 
 
 0.717 
 
 38 
 
 93.2 
 
 27.8 
 
 0.792 
 
 90.1 
 
 25.0 
 
 0.712 
 
 40 
 
 94.2 
 
 28.8 
 
 0.793 
 
 91.0 
 
 25.9 
 
 0.713 
 
 42 
 
 95.3 
 
 29.9 
 
 0.799 
 
 91.8 
 
 26.7 
 
 0.708 
 
 44 
 
 95.7 
 
 30.3 
 
 0.791 
 
 92.5 
 
 27.4 
 
 0.715 
 
 46 
 
 96.1 
 
 30.7 
 
 0.787 
 
 92.8 
 
 27.7 
 
 0.710 
 
 48 
 
 97.0 
 
 31.6 
 
 0.797 
 
 93.3 
 
 28.2 
 
 0.712 
 
 50 
 
 97.5 
 
 32.1 
 Mean = 
 
 0.798 
 
 94.0 
 
 28.9 
 Mean 
 
 0.718 
 
 = 0.786 
 
 = 0.709 
 
CHAP. XIX. EEFLECTIVE I'OWEE OF POLISHEB METALS. 309 
 
 = iJO.T. Consequently, the reflective power of silver is to 
 that of nickel as 30°.7 to 24M = 1.000 to 0.785. Now, it 
 will be fmiiul, on inspecting the table, that the reflective 
 power of nickel, determined by the mean of twenty-five dis- 
 tinct observations, is 0.786 against the stated 0.785. This 
 insignificant difference furnishes positive evidence of the reli- 
 able character of the investigation. The important question 
 of comparative reflective power of silver, brass, nickel, and 
 steel may therefore be regarded as permanently settled ; while 
 the reflective power of all other metals may be determined 
 in like maniici-, by simply detaching the brass, nickel, and 
 steel plates from the conical reflector and substituting others 
 in their places ; the silver plate, of couree, remaining as a 
 standard of comijarison. It has already been stated that, 
 according to Laprovostaye and Desains's investigation, the 
 reflective power of brass is to that of silver Jis 0.978 to 
 1.000, difference = 0.022, while our exact method of measur- 
 ing the energy lost by reflection shows that brass possesses 
 
 885 
 
 a reflecting power of only of that of silver, difference 
 
 ° '■ •' 1000 
 
 1.000 — 0.885 = 0.115 against 0.022. Considering the precision 
 which distinguishes all Laprovostaye and Desains's investi- 
 gations, this extraordinary discrepancy proves indisputably 
 that the exact loss of calorific energy attending the reflection 
 of radiant heat cannot be ascertained by the thermo-electric 
 method. 
 
CHAPTER XX. 
 
 RAPID-INDICATION ACTINOMETER. 
 
 Physical observatories cannot be regarded as fully equip- 
 ped unless provided with actiiiometers })y means of wliieli 
 the intensity of solar radiation may be quickly ascertained, 
 say in the course of one minute. The thermo-electric pile 
 is of no avail for this pur^iose, from varioiis reasons: (1) 
 The temperature produced by solar radiation is too high to 
 be correctly registered by the deflection of the galvanometric 
 needle ; (2) The degree of actual temperature inferred from 
 the ai'c described by the needle is mere guess-work at the 
 high temperature developed by a vertical sun ; (3) The 
 disturl)ing effects of movable masses of iron, the jiresence of 
 ^vhich are unavoidable; (4) The rotation of the obsei-vatory 
 and consequent irregular change of the magnetic meridian. 
 
 The illustration on PI. 31 represents a vertical section of 
 an instrument constructed for the special purpose of ascer- 
 taining the intensity of the radiant heat after a very brief 
 exposure to the sun's rays. The leading feature of the de- 
 
cu.U>. XX. UAriD-lSUlCATWy ACTlSOMElEli. 311 
 
 viit' is iliat of coueeutratiiig the rays before reaching the 
 bull) ut' tlie thermometer employed to measure the iuteusity. 
 It will be leadily understood that by employing a lens of 
 [H'opiT t'nrm the degree of concentration may be such that 
 in the space of sixty seconds the mercurial column of the 
 thermometer subjected to the concentrated I'ays will rise to 
 the same height as the column of another thermometer ex- 
 posed to the direct influence of the sun's radiant luat (hiring 
 a period sufficient to produce maximum indication. Hence, 
 assuming that a proper degree of concentration has been at- 
 tained ill the iifw iustnimeiit, its exposure to the suii during 
 sixty seconds will obviously funiisli an indication of the 
 intensity of the sun's rays as C(jrrectly as the actinometer 
 described in Chap. III., whatever be the zenith distance or 
 other conditions at the time of making the observation. The 
 concentration of the sun's rays, it should be observed, nuist 
 take place within an exhausted vessel maintained at a con- 
 stant temperature corresponding with that of the actinometer 
 referred to. It is hardly net'cssaiy to point out that it 
 would be impossible to determine theoretically the precise 
 distance between the lens employed to concentrate the rays 
 and the thermometer ^\hich receives the accumulated heat. 
 The instrument should therefore be so constructed that the 
 lens may be readily moved away from or towards the 
 thermometer during observations. This property of the in- 
 strument under consideration may be regarded as another 
 leading feature indispensable to render accurate adjustment 
 practicable. Referring to the illustration, it will be seen 
 
312 BADIAM HEA T. CHAP. xx. 
 
 tliat tlie exhausted vessel euclosing the recording tLeniiometer 
 is surroiiuded by an external casing, the form of which, like 
 the internal vessel, is cylindrical. The intervening space is 
 tilled with water maintained at a constant temperature by a 
 similar process of circulation as that adopted in the actino- 
 nieter described in Chap. III. Couplings for attaching the 
 circulating tubes are applied at the bottom and top of the 
 external casing, as shown in the illustration. The inclina- 
 tion of the cylindrical vessel is regulated by a tangential 
 screw ^\'orking through a nut turning in bearings attached to 
 the side of the external vessel. By means of the graduated 
 quadrant represented in the illustration, the sun's zenith 
 distance may be ascertained by mere inspection, at all times, 
 during observations. A journal applied under the exhausted 
 vessel coinciding with the centre of the graduated quadrant, 
 and turning in appropriate bearings, suppoi"ts the instrument. 
 These bearings are secured to the top of a vertical column 
 resting on a revolving circular plate, to which a small pinion 
 is applied, as shown in the illustration. This j)inion, intended 
 to be operated by hand, works into cogs formed at the cir- 
 cumference of the circular bed-plate attached to the top of 
 a substantial table. The tangential screw, it will be seen, 
 is provided with a small hand-wheel at the lower end, 
 enabling the operator to give any desired inclination to the 
 instrument, while the small hand-wheel of the pinion, before 
 referred to, enables him to follow the earth's diurnal motion. 
 Let us now examine the mode of regulating the distance 
 between the lens and the thermometer which is employed to 
 
CHAP. XX. EAriD-IXDICATIOX ACTIXOMETER. 313 
 
 show the intensity of the concentrated radiant heat. We 
 have already pointed out that it is impossible to determine 
 theoretically the said distance. Of course, it may be calcu- 
 lated approximately, but not sufficiently near to dispense 
 with means admitting of a very considerable movement of 
 the lens up and down during the process of adjustment. In 
 order to meet this condition, the lens is inserted in a per- 
 forated piston fitting air-tight in the exhausted cylindrical 
 vessel ; the latter being bored out accurately like the barrel 
 of an air-pump. The adjustment of the position of the lens, 
 it will be seen by referring to the illustration, is effected by 
 screws secured in the piston, the nuts employed for raising 
 the same acting against substantial brackets bolted to the 
 top riange at the upper end of the exhausted cylinder. Con- 
 venient means being thus arranged for raising or lowering 
 the lens in the exhausted cylinder, it will be perceived on 
 reflection that although the thermometer subjected to the 
 action of the concentrated heat is graduated in the ordinary 
 manner, its indication may be made to correspond exactly 
 with that of an actinometer exposed to the direct action of 
 the sun's rays, provided the piston be placed correctly, viz., 
 at such a distance that the energy of the convei'ged rays 
 under the lens be sufficient to raise the mercurial column of 
 the enclosed thermometer as high during an interval of sixty 
 seconds as that of an actinometer exposed to the direct solar 
 heat during a period sufficient to produce maximum indica- 
 tion. It will lie admitted that but for the introduction of 
 a movable lens it would be impracticable to construct an 
 
314 BABIANT HEAT. chap. xx. 
 
 instrument furnishing maximiim indication of solar intensity 
 during tlie exact interval of sixtij seconds. The expedient, 
 however, of inserting the lens in a piston capable of being 
 l^laced at any desirable distance in order to subject the bulb 
 of the enclosed thermometer to any requisite temperature, it 
 Avill be perceived, renders that adjustment easy which is 
 necessary to secui'e a given indication in a stipulated time. 
 
 It has been pointed out in previous chapters that, owing 
 to the rapid change of zenith distance, and the consequent 
 variation of the depth of atmosphere penetrated by the sun's 
 rays, the indication of an actinometer is too loAV during the 
 forenoon and too high in the afternoon. It will be obvious, 
 therefore, that the lens-instrument under consideration should 
 be adjusted when the sun passes the meridian, since the acti- 
 nometer employed for comparison then indicates correctly the 
 solar intensity at the moment of making the observation. 
 The adjustment is effected in the following manner : Having 
 placed the lens-instrument by the side of the standard acti- 
 nometer selected for comparison, the casings or water-jackets 
 of the two should be connected, by means of flexible tubes, 
 in such a manner that the outlet pipe of the actinometer is 
 made to communicate with the inlet pipe of the lens-instru- 
 ment. Both instruments having been tui'ned towards the sun, 
 and the usual connections to the cistern and hand-pump of the 
 actinometer having been completed, the refrigerating current 
 should be circulated through the jackets until the actinometer 
 indicates maximum differential temperature. In the mean- 
 time, the operator attending to the circulation should screen 
 
CHAP. XX. RAPID-INDICATION ACTINOMEXmi. 315 
 
 the lens-iustrunient from tlie sun by a disc of pasteboard. A 
 second operator, as soon as maximum temperature lias been 
 reached by the actinometer, starts the chronograph and calls 
 time, the pasteboard screen being at once lowered. At the 
 termination of sixty seconds, time is again called, and the 
 protecting screen instantly raised. Let us now suppose that 
 the actinometer has continued to indicate a differential tem- 
 perature of 5G° during the experiment, and that the thermo- 
 meter of the lens-instrument indicates only oS"" at the end 
 of sixty seconds' exposure to the concentrated solar a-adiation. 
 It needs no explanation to show that the observed difference 
 of 56 — 52 = 4° is owing to the want of adequate concen- 
 tration, and that the discrepancy will be remedied simply 
 by moving the lens further from the enclosed thermometer. 
 This is effected by turning each of the nuts at the top of 
 the exhausted cylinder, say once round, thereby increasing 
 the distance. The chronograph should again be started, the 
 screen removed, and the operation already described repeated 
 as quickly as possible, and the result recorded. Should it 
 now be found that the enclosed thermometer indicates 54°, 
 another turn of the adjusting nuts should be given, and the 
 operation repeated a third time. Due diligence being exer- 
 ci^^ed by the operators, the proper position of the lens may 
 thus be determined before a change of the sun's zenith dis- 
 tance takes place sufficient to affect the indication of the 
 standard actinometer to an extent preventing accurate adjust- 
 ment. In view of the fact that the form of the lens, and 
 its distance from the thermometer which indicates the inten- 
 
316 BABIANT HEAT. chap. xx. 
 
 sity of the concentrated radiation, may be approximately 
 determined by calculation, it is evident that tlie lens might 
 be stationary and the recording thermometer graduated by 
 repeated exposure to the sun during intervals of 60 seconds, 
 so as to correspond with the indication of a standard actiuo- 
 meter. I have constructed small lens-actinometers on this 
 plan, useful for ordinary observations, but for physical obser- 
 vatories and investigations requiring perfect accuracy the 
 instrument having a movable lens, as represented by the 
 illustration in PL 31, is far preferable. 
 
CHAPTER XXI. 
 
 SOLAR RADIATION AND DIATHERMANCY OF FLAMES. 
 
 The readers of " Coiiiptes Rendus " are a^vare tLat Pere 
 Secchi addressed a letter to the Academy of Sciences at 
 Pans (see "Comp. Rend.," tome Ixxiv., pp. 26-30), contain- 
 ing a review of my communications to JVatvre, publislied 
 July 13, October 5, and November 16, 1871, in wliicli he 
 questions the correctness of my published reports containing 
 tabulated statements of the temperatm-e produced by solar 
 radiation. His reason for questioning the reliability of my 
 tables appears to rest on the supposition that my instruments 
 do not furnish correct indications. "It is astonishing," he 
 says, " that Mr. Ericsson should find with his instrument a 
 higher stationaiy temperature in winter than in summer. 
 This (even bearing in mind the greater proximity of the sun 
 in winter) makes me think that there must be something 
 very singular in his apparatiis, making all its indications 
 deceptive. Even under the beautiful sky of Madrid has M. 
 Rico y Sinobas found, in December, for the solar radiation, 
 12 div., 19 by his actinometer, and, in June, 25 div., 56." 
 
318 BADIANT HEAT. chap. xxi. 
 
 P^re Secclii ought to liave 2:)erceived "tliat there must be 
 sometliiug very singular " in the actinometer emj)loyed by the 
 Spanish physicist, or it could not have indicated an intensity 
 twice as high in June as in December. Obviously, a correct 
 actinometer will indicate a higher temperature during the 
 ^vinter solstice than at midsummer f<.)r equal zenith distance. 
 The instrument employed at Madrid, if Secchi's figures are 
 correct, must therefore be founded on utterly erroneous prin- 
 ciples. In North America, in lat. 40 deg. 42 min. (the 
 latitude of Madrid is 40 deg. 24 min.), solar intensity at noon 
 during the latter part of June is 64°. 5 when the shy is 
 clear, while at noon during the latter part of December the 
 temperature under similar atmospheric conditions reaches 
 57°.70. But observations made in the morning or evening 
 during the month of June, at the hour when the sun's alti- 
 tude is the same as at noon in December, show that the 
 intensity of the radiant heat in June is only 53°. 08, against 
 57°.49 F. in December. Actual observations have thus 
 established the fact that for corresjyonding zenith disfauce the 
 temperature produced by the radiant heat Avhen the earth has 
 nearly reached perihelion, is 57°.49 — 53°.0S = 4°.41 higher 
 than at midsummer. Referring to Chap. IV., it Avill be seen 
 that, owing to the greater proximit}' of the sun, the increase 
 of absolute intensity of solar radiation is 4°.GG F. during the 
 winter solstice. Pere Secchi will do well to examine the sub- 
 ject more carefully, and make himself better acquainted with 
 the character of the investigations which have led to an exact 
 determination of the temperature produced by solar radiation. 
 
CHAP. xxi. SOLAR rADIATIOX AXV DIATnEEMANOT. 319 
 
 The readers of " Comptes Rendus " who have examined the 
 review referred to, ignoraut of the articles in Nature, the 
 contents of which Pere Secchi criticises, will be surprised to 
 learn that I have not, as the revieAver asserts, questioned the 
 power of vapo)' to diminish solar intensity. Having stated 
 the result of numerous observations of the sun's radiant pow er 
 at corresponding zenith distance, and proved that the tempe- 
 rature during midwinter is higher than at midsunuuer, I made 
 the following remark in Kature, Nov. 1(5, 1S71 : "In the face 
 of such facts it is idle to contend that the temperature pro- 
 duced by solar radiation under coiTesponding zenith distance 
 and a clear skij varies from any other cause than the vaiying 
 distance between the sun and the eartli." It is absurd to 
 suppose that a person having devoted many years to the 
 investigation of solar radiation should deny the retarding 
 influence of vapor, since not one observation in a hundred 
 indicates maximum solar intensity, owing to the presence of 
 vapor in the atmosphere. 
 
 The following brief description of the actinometer which 
 Pere Secchi supposed to be constructed on erroneous prin- 
 ciples was inserted in my reply to his criticism published in 
 " Comptes Rendus," before referred to, in hopes that, on learn- 
 ing that there is not anything "very singular" in my appa- 
 ratus, lie would have seen fit to witlidraw Iiis statement 
 (piestioning the correctness of my observations relating to 
 solar intensity : " The principal part of the instrument con- 
 sists of an air-tight cylindrical vessel, the axis of which is 
 directed towards the sun, the upper end l)eing provided with 
 
320 BADIANT HEAT. chap. xxi. 
 
 a thin lens covering an aperture 3 ins. in diameter. The 
 bnllj and part of the stem of a mercurial thermometer is 
 inserted through the u^'per side, at right angles to the axis, 
 a small air-pump being applied for exhausting the air fi-om 
 the cylindrical vessel. The latter is surrounded liy a casing 
 through which water is circulated by means of an ordinary 
 force-pump and ilexible tubes, connected with a capacious 
 cistern containing water kept at a constant temperature of 
 (50° F. The bulb of the thermometer is cylindi'ical, 3 ins. 
 long, its contents bearing a very small proportion to its 
 convex area. The up2:)er half is coated with lamp-black, 
 while the lower half of the bulb is effectually protected 
 against loss of heat from undue radiation. The diminution 
 of energy attending the passage of the sun's rays through the 
 lens is made good by the concentration effected by its cur- 
 vature ; hence the true energy of the radiant heat trans- 
 mitted will be shown by the expansion of the mercury in 
 the bull). The inclination of the lattei", it should be ob- 
 served, promotes a rapid upward current of the contents on 
 the top side, and a corresponding downward current t>n the 
 lower side, thereby rendering the indication prompt and 
 trust\vorthy. The water in the surrounding casing being 
 maintained at a constant temperature of. 60° F., it will be 
 evident that the zero of the thermometric scale of the actino- 
 meter must correspond with the line which marks G0° on 
 the Fahrenheit scale. It scarcely needs explanation that the 
 height reached by the mercurial column after turning the 
 instrument to^vards the sun will be due wholly to solar 
 
CHAP. XXI. SOLAIi FABIATIoy AXD BIATllEUMAliGY. 321 
 
 eneig}', since the i-adiatiuii of tlie exbaiisted vessel toAvartls 
 the bulb of the thermometer is only capable of raising the 
 column to the actinometric zero (60° Fahr.) " 
 
 The readers of Xature will remember that one of my 
 articles reviewed by Pere Secchi contains a demonstration 
 accompanied by several diagrams, proving that the radiant 
 heat emitted by the chromosphere and outward strata of the 
 solar envelope is inappreciable at the surface of the earth. 
 It will be remembered, also, that the mode adopted in set- 
 tling the fpiestiou whether the solar atmosphere is capable of 
 emitting heat rays of appreciable energy was that of shutting 
 out the rays from the photosphere and collecting those from 
 the chromosphere and envelope, in the focus of a parabolic 
 reflector. Scarcely any heat being produced, notwithstand- 
 ing the great concentration l^y the reflector, I proved the 
 fallacy of Pere Secchi's remarkable assumption, that the high 
 temperature at the surface of the photosphere is caused by 
 radiation "received from all the transparent strata of the 
 solar envelope." It is surprising that, notwithstanding the 
 completeness and positive nature of my demonstration, no 
 allusion whatever is made to the same in a review profess- 
 ing to scrutinize the subject critically. Ignoring the evidence 
 furnished by actual trial in proof of the extreme feebleness 
 of the radiating power, the revieAver proceeds to state "that 
 the outward strata might be less hot, and that the effect 
 which we measure is the aggregate of the quantities of heat 
 whicli are added, emanating from the vaiious transparent 
 strata." Ilow tlie outward coldei' strata cause an elevation 
 
332 BADIANT EEAT. chap. xxi. 
 
 of temperature by their I'adiatiou toward tlie solar surface is 
 not explained, but reference is made to tbe result of an ex- 
 periment with three small flames in support of the assertion 
 that the high temperature of 10,000,000° C. assigned to the 
 surface of . the sun is owing to radiation received from all 
 the transparent strata surrounding the photosphere. "A very 
 simple experiment," the reviewer states, " made at my request 
 by P. Provenzali, has shown that, if a heating of 2.5 deg. 
 can be obtained with one flame, "with two flames placed one 
 before the other 4.5 deg. are obtained ; with three flames, 5.4 
 deg. — a result easily foreseen, for everybody knows that 
 flames are transparent." 
 
 My practical demonstration establishing the feebleness of 
 the radiating power of the matter composing the solar enve- 
 loj^e having received no considei'ation, while the reviewer, in 
 support of his singular theory of solar temperature, points to 
 the result of the rude experiment conducted by Pere Pro- 
 venzali, I have deemed it necessary to show that the trans- 
 parency of flames is too imperfect to warrant the inferences 
 drawn. 
 
 The illustration on Plate 32 represents an apparatus by 
 means of which the exact degree of transparency of a series 
 of flames has been ascertained. 
 
 Description : h, conical vessel, open at the top, the bottom 
 communicating with a cylindrical chamber / by a narrow pas- 
 sage, the whole being enclosed in an exterior vessel c, charged 
 with water kept at a constant temperature, precisely as in the 
 actinometer. A thermometer is applied near the bottom of 
 
CHAP. XXI. SOLAK liADIArWX AND DIATIIEIiMANOY. 323 
 
 the cyliiulrieal chamber, the centre of the bulb coinciding 
 with the prolongation of the axis of the conical vessel. A 
 gas-pipe J, provided with a series of vertical burners, is firmly 
 secured to an inclined table in a position parallel to the axis 
 of the conical vessel. The burners are provided with caps in 
 order to admit of any desirable number of jets being ignited 
 at one time, ^^']len gas of ordinary pressure is admitted into 
 the pipe d, the side view of the flame will be as indicated 
 by the dotted lines at m m, the thickness of each flame being 
 nearly 0.20 in., while the width, shown by the dotted lines 
 ?i 11, somewhat exceeds 3 ins. from point to point. It will 
 be observed that the prolongation of the axis of the conical 
 vessel upwards passes through the central portion of the 
 flames at the point of maximum thickness and intensity. 
 Supposing that the instrument (attached to a table pro- 
 vided with parallactic mechanism) is directed towards the 
 sun, it will be evident that all the rays of a pencil, the 
 section of which corresponds with that of the bulb of a 
 thermometer, will pass through the flames before reaching 
 the said bulb. Now, the temperature of the flames at the 
 point pierced by the solar rays is fully 2,000° F., while the 
 actual intensity of the rays does not exceed 60°. It is hardly 
 necessary to observe that the illustrated device enables us 
 to ascertain whether the solar rays thus entering at a diffe- 
 j-ential temperature 1,9J:0° lower than that of the incan- 
 descent gas have their intensity augmented or diminished 
 (luring the passage through the heated medium. But before 
 we can determine this question, it will be necessary to ascer- 
 
324 
 
 BADIAKT HEAT. 
 
 CHAP. XXI. 
 
 tain what temperature is comiimnicated to tlie tliermometer 
 by the radiant energy of tlie flames alone. Accordingly, a 
 series of experiments have been made, the result of -which 
 is recorded in the annexed table. 
 
 The instrument turned away from 
 the sun. 
 
 The insti-ument directed towards the sun. 
 
 -a. 
 
 |2i 
 
 1 
 
 i . 
 
 ■si 
 
 a 
 .1 
 
 III 
 as 
 
 Temperature produced 
 
 by the sun's rays 
 
 acting directly on the 
 
 bulb. 
 
 Temperature produced 
 
 by the sun's rays 
 
 passing through "the 
 
 flames. 
 
 Increment of 
 
 temperature attending 
 
 the passage of the solar 
 
 rays through the 
 
 flames. 
 
 
 Inches. 
 
 °Fah. 
 
 ' Fah. 
 
 'Fah. 
 
 -Fah. 
 
 1 
 
 2 
 3 
 4 
 5 
 6 
 7 
 8 
 9 
 10 
 
 24.8 
 23.8 
 22.8 
 21.8 
 20.8 
 19.8 
 18.8 
 17.8 
 16.8 
 15.8 
 
 1.76 
 2.88 
 3.80 
 4.58 
 5.24 
 5.84 
 6.38 
 6.91 
 7.40 
 7.90 
 
 21.60 
 21.61 
 21.62 
 21.63 
 21.64 
 21.65 
 21.66 
 21.67 
 21.68 
 21.69 
 
 21.90 
 22.20 
 22.49 
 22.75 
 22.99 
 23.32 
 23.43 
 23.63 
 23.82 
 24.00 
 
 0.30 
 0.69 
 
 0.87 
 1.12 
 1.35 
 1.57 
 1.77 
 1.96 
 2.14 
 2.31 
 
 The nature of the investigation T\dll be readily under- 
 stood by the following explanation : The instrument being 
 turned away from the sun and the upper flame m ignited, 
 while the external casing c is kept at a constant temperatui'e 
 
CHAP. XXI. SOLAB FADTATIOX AXD DIATEEBMANOY. 325 
 
 of 60°, the coluinii of the thermometer at / slowly rises to 
 Gl°.70. The radiant heat, therefore, of a single Hanie pro- 
 duces a differential temperature of 01°. 70 — 00 = 1°.7G. The 
 second Hanie being ignited, the temperature rises to G2°.88, 
 thus increasing the differential temperature to 2°.88. The 
 ignition of the third flame augments the differential tempe- 
 rature to 3°.80. The remaining flames being ignited in regu- 
 lar order downwards, their combined radiant energy elevates 
 the temperature to G7°.90. Deducting the teniperaturi^ of 
 the enclosure c (60°), the trial shows that, although the 
 single flame at the maximum distance from the bulb is capable 
 of producing a differential tempei-ature of 1''.70, the energy 
 of the ten flames together produces only 7°.90. This fact 
 fui-nishes conclusive evidence of the imperfect transparency 
 of the flames. Assuming that the heat rays are capable of 
 pas.sing freely through tlie incandescent medium, it will be 
 perceived that the entii'e series of flames should produce a 
 differential temperature of 1.76 X 10 = 17''. 6, showing a re- 
 tardation of 17.6 — 7.9 = 9°.7. And if we take into account 
 the diminished distance of the lower flames from the bulb 
 of the thermometer, it will ])e found that the actual retarda- 
 tion greatly exceeds this compntation. We have thus de- 
 monstrated that flames are not transparent, as supposed l)y 
 Pere Secchi ; consequently, the inferences drawn from the 
 experiment to which the distinguished savant refers in his 
 letter to the French Academy of Sciences are Avholly un- 
 wari'nntable. 
 
 Having disposed of the question of transparency, and 
 
33G FADIANT HEAT. CHAP. xxi. 
 
 ascertained the degree of temperature commimicated to the 
 tliermometer by tlie radiant energy of tlie flames alone, let 
 us now suppose that the instrument has been turned towards 
 the sun. The temperature produced by the combined energy 
 of solar radiation and the radiation of the flames, after direct- 
 ing'' the instrument towards the luminary, will be found I'e- 
 corded in the fifth column of the table. Dispensing with a 
 detailed record of the energy transmitted for each flame 
 separately, let us at once consider the effect produced by 
 passing the sun's rays through the entire series. It has 
 already been stated that the radiation of all the flames com- 
 bined imparts a differential temperature of 7°.90 to the 
 thermometer. By reference to the table, it will be seen that 
 the temperature produced by the sun's rays is 21°.69 when 
 the flames are extinguished. Consequently, the temperature 
 produced after lighting the whole series ought to be 21.69 
 + 7.90 = 29°.59 instead of 24°. 00, since solar heat, under 
 analogous conditions, is capable of increasing the temperature 
 of substances, whatever be their previous intensity. Our 
 experiment, therefore, furnishes additional evidence of the 
 imperfect transparency of flames. But notwithstanding this 
 want of transparency, it will be found on referring to the 
 table that an augmentation of temperatui'e of 24.00 — 21. G9 
 = 2°.31 takes place while the comparatively cold solar rays 
 pass through the incandescent medium. This exti'aordinary 
 fact points to an increase of molecular' energy within the 
 incandescent gas, although its temperature is fully 1,900° 
 higher than the intensity of the sun's rays. 
 
CHAPTER XXII. 
 
 CONSTANCY OF ROTATIOX OF THE EARTH INCOMPATIBLE 
 WITH SOLAR INFLUENCE. 
 
 Laplace's demonstration, showing that the axial rotation 
 of the eartli is not affected by atmospheric currents and 
 similar motions caused by solar heat, has been accepted by 
 physicists as incontrovertible. The German mathematician, 
 Dr. Mayer— celebrated for his demonstration establishing the 
 equivalent of heat— says in a discourse on that branch of 
 celestial mechanics which relates to the effect produced by 
 contrary atmospheric currents : " The final result of the ac- 
 tion of these opposed influences is, as regards the rotation of 
 the earth, according to well-known mechanical principles = ; 
 for these currents counteract each other, and therefore cannot 
 exert the least influence on the axial rotation of the earth. 
 This important conclusion was proved by Laplace." "The 
 same," he adds, "holds good foi- eveiy imaginable action 
 which is caused by the radiant heat of the sun, or by the 
 heat which reaches the surface from the earth's interior, 
 
338 EADIAKT HEAT. chap. xxii. 
 
 A\lietlier the action be in the aii', in the water, or on the 
 land. The effect of every single motion produced by these 
 means on the rotation of the globe is exactly compensated by 
 the effect of another motion in an oj^posite direction, so that 
 the resultant of all these motions is, as far as axial rotation of 
 the globe is concerned, = 0." I propose to show that this con- 
 clusion is fallacious, and that the sun's radiant heat develops 
 forces capable of diminishing perceptibly the earth's rotary 
 velocity ; and that unless the retarding influences of solar 
 heat, the existence of which I am going to establish, are 
 counteracted by some cosmical force of which we have no 
 knowledge, the earth's rotary velocity will be considerably 
 reduced in the course of time. 
 
 There are two classes of force produced by solar heat, 
 capable of retarding the axial rotation, differing, however, 
 entirely as regards ultimate results. The first class includes 
 animate exertion, mechanical force produced by heat de- 
 veloped by the combustion of organic substances, and the 
 resistances of abraded solid matter transferred from its ori- 
 ginal position by the waters of rivers flowing towards the 
 equator. The forces thus enumerated, it will be shown, retard 
 the rotary velocity of the globe in all cases when they remove 
 weight to a greater distance from the axis of rotation, i.e., 
 expand the circle of gyration, thereby diminishing the num- 
 ber of revolutions performed in a given time. Obviously, 
 the vis viva of the rotating mass Avill remain iTudiminished, 
 as the centre of gyration is merely removed to a greater 
 distance from the axis of rotation. Accordingly, the axial 
 
CHAP. XXII. THE EAETU-S AXIAL ROTATION. 329 
 
 rotation, though checked, cau never be stopped by the class 
 of retarding influences thus pointed out. The second class, 
 however — which comprises the retardation produced by the 
 atmospheric air during its course from the polar to the e(pia- 
 torial regions, and the retardation caused by the waters which 
 flow towards the equator to restore the quantity lost by the 
 powerful evaporation within the tropics — not only diminishes 
 the rotary velocity, but, at the same time, deprives the earth 
 of so creat an amount of vis viva that the axial rotation 
 must ultimately cease, unless some exterior compensating 
 force exists — a supposition at variance ^nth the principles of 
 mechanics. 
 
 Let us now briefly examine the nature of the retarding 
 influences of the first-named class, which, as stated, are un- 
 attended by any loss of the earth's vis viva — namel)', animate 
 force and mechanical energy, resulting fi'om the combustion 
 of organic substances when expended in raising weight to 
 remain permanently in an elevated position ; and the retarda- 
 tion caused by solid matter carried towards the equator. 
 Before entering on this examination, it will be instinctive to 
 test by some familiar illustration the correctness of Mayer's 
 assumption, that "every imaginable action affecting the rota- 
 tion of the globe is exactly compensated by the effect of 
 another motion in an opposite direction." A great variety of 
 instances might be mentioned in which the development of 
 mechanical energy, productive of heat, counteracts the rotary 
 motion of the earth, and deprives it permanently of a certain 
 amount of vis viva. Suppose, for instance, a locomotive 
 
330 BADIANT HEAT. chap. xxii. 
 
 train weigliing 400 tons to be started from the Mestern 
 terminus of a railway running from west to east. Suppose, 
 also, that when this train has acquired a velocity of 50 ft. 
 per second, it encounters another similar train which is at 
 rest. The result of such an encounter, in a dynamic point 
 of vie^v, is now Avell understood. Apart from a small amount 
 of energy absorbed in overcoming the cohesive force between 
 the particles composing the materials fractured by the con- 
 cussion, the encounter will develop an amoimt of heat 
 corresponding with the vis viva of the arrested train. It 
 scarcely needs explanation that in putting the train in motion 
 from the terminus eastward the rails— /.(>., the surface of the 
 earth — will, in consequence of the adhesion between the 
 wheels and the rail, be pushed westward ; hence in a direc- 
 tion contrary to the earth's rotation. The amount of dynamic 
 energy which the train thus imparts to the earth in an oppo- 
 site direction to that of rotation may be readily ascertained 
 by multiplying the arrested weight by the height necessary 
 to produce a velocity of 50 ft. per second — namely, 39 ft. ; 
 hence 400 X 2,240 X 39 = 34,944,000 foot-pounds. Deduct- 
 ing the small amount of energy Avhieh favors the earth's 
 rotation called forth by the rolling friction and adhesion of 
 the wheels of the stationary train, during the short retro- 
 grade motion attending the concussion, it will be found that 
 the earth loses an amount of vis viva of fully 34,000,000 
 foot-pounds. The assumption of Dr. Mayer, based on the 
 theory of Laplace, that the resultant of all imaginable mo- 
 tions as regards the earth's axial rotation is = 0, has thus 
 
CHAP. XXII, THE EAliTWS AXIAL L'OTATIOK. 331 
 
 been proved to be untenable. It is not intended to question 
 Laplace's conclusion as regards the existence of a compen- 
 sating effect; he was mistaken only as to its nature — a mis- 
 take, however, of ijaraiiuniiit importance, as we have shown 
 that the compensation for the lost energy, in the case pre- 
 sented, is the generation of a certain amount of heat Avhich, 
 in less than three hours after the concussion, if the sky be 
 clear, radiates into space, leaving the earth minus 34,000,000 
 foot-pounds of vis viva. The important fact should not be 
 overlooked that the retardation thus established is the result 
 of solar energy stored in the combustibles of the locomotive 
 furnace. Numerous instances of a similar nature micrht be 
 mentioned in support of the assertion that the earth is sub- 
 jected to retarding influences and loss of vis viva by me- 
 chanical motions on the earth's surface \vhich result in the 
 production of heat radiated into space. But all these are 
 insignificant compared with the stupendous amount of retarda- 
 tion caused by the conversion of mechanical energy into heat 
 within the opposing atmospheric currents circulating between 
 the equatorial and polar regions. In connection with this 
 proposition, it will be proper to remark that our knowledge 
 of the convertibility of mechanical energy and heat — in other 
 words, the convertibility of mechanical and molecular energy 
 — has completely upset Laplace's demonstration, on which 
 physicists have based their assumption, that the rotaiy 
 velocity of the earth cannot be affected by the sun's radiant 
 heat. 
 
 Let us now examine, separately, those forces produced by 
 
333 BABIANT HEAT. chap. xxii. 
 
 solar lieat wliicli tend to check tlie earth's rotary velocity 
 by removing weight from the axis of rotation — i.e., expanding 
 the circle of gyration — and those vt^hich occasion a diminution 
 of the rate of axial rotation without disturbing the balance 
 of the rotating mass. The first class : Animate or muscular 
 enei'gy, and the force generated by heat from the combustion 
 of organic matter, controlled by the hiiman mind, both re- 
 sulting indirectly from the sun's radiant heat. That the hand 
 and intellect of men have caused a disturbance of the position 
 of the earth's centre of gp-ation will be deemed a startling 
 assertion, yet it cannot be controveiied in view of the fol- 
 lowing facts. The millions of tons of matter contained in 
 the Pyramids, removed to a greater distance from the axis 
 of rotation by the muscular exertion of the ancient Egyptians, 
 disturbed the previous balance of the rotating mass, causing 
 a tendency to check the earth's rotary velocity and to increase 
 the length of day. Nor can it be questioned that if London 
 had not been built, and if the building materials of Paris yet 
 remained in the Catacombs, the sun would rise earlier than 
 it now does, though the diifereuce woiild be small beyond 
 computation. The aggregate of the weight removed from 
 below, and piled above the crust of the globe by the hand 
 of man, is, however, so great that figures are competent to 
 express the extent of the consequent retai'dation of the axial 
 rotation, while the divisions of our common instruments for 
 measuring distance are sufficiently minute to indicate the 
 expansion of the earth's circle of gyration caused by the 
 transfer of matter under consideration. A first-class modern 
 
CHAP. XXII. TEE EAETU'S AXIAL BOTATION. 333 
 
 cit}', for instance, contains upwards of 100,000 bouses; each 
 house contains on an average 400 tons of mineral matter ; 
 hence the total weight of brick, earth, or stone removed fi-om 
 below the surface to a considerable height above the earth's 
 surface exceeds 40,000,000 tons — a mere fractidu compar('d 
 with the weight of the whole of human hal)itations and 
 other stnictures raised above the surface of the earth chiefly 
 by muscular eifort. Let us add the weight of materials 
 raised from mines to an increased distance from the axis of 
 rotation, by animate exertion and by mechanical force con- 
 trolled by intellect. 
 
 An element of greater importance, connected with the 
 fii-st class of retarding influences produced by the sun's ra- 
 diant heat, next claims our attention — namely, the solid and 
 sedimentaiy matter detached by the abrasion of i-ain-water, 
 and afterwards conveyed by the currents of rivers to a po- 
 sition nearer the equator ; hence removed to a greater dis- 
 tance from the axis of rotation. The question whether any 
 estimate can be made of the aggregate weight of matter, 
 the original position of which is being changed during defi- 
 nite periods by the caiise referred to, is by no means so 
 ditticult to answer as might appear without due considera- 
 tion. It is tnie, we do not know what quantity of water 
 or sediment is carried towards the equator by the several 
 rivei"s; but we can compute with sutficient exactness the ex- 
 tent of the river basins. Accordingly, if we could estab- 
 lish a mean of discharge per square mile of some very 
 extensive basin comprising all the varieties of climate and 
 
334 BADIANT HEAT. chap. xxii. 
 
 soil, tlie question could be satisfactorily answered. Fortu- 
 nately, one of tlie longest rivers on tlie globe, the Missis- 
 sippi, wliicli drains tlie greatest extent of sui-face witb but 
 two important exceptions, Las been carefully surveyed by a 
 corps of Topograpliical Engineers, by order of tlie United 
 States Government. Not only lias this great river been 
 thus carefully examined, but the basin it drains comprises 
 ex^ery variety of soil and climate — its source being among 
 sno\vs and lakes frozen during the greater portion of the year, 
 while the outlet is near the tropics. That the Mississippi 
 basin represents the average of the liver systems of both 
 hemispheres has been established by the fact that, although 
 the rain-gauges at its northern extremity show only 13 ins. 
 for twelve months, those of its southern boundary reach 
 66 ins., with every possible gradation of jwecipitatiou in the 
 intermediate space. In addition to this important circum- 
 stance, the basin covers 21 deg. of latitude and 35 deg. of 
 longitude, or 1,460 miles by 1,730 miles ; hence comprising 
 an area greater than the entire European Continent west of 
 the rivers Vistula and Pruth. It may be confidently as- 
 sumed, therefore, that the Mississippi basin represents the 
 average discharge of Avater and sediment so nearly that cal- 
 culations based thereon, applied to the river systems of 
 botli hemispheres — excepting some of the northern Asiatic 
 and American livers — will exhibit a general result differing 
 but slightly from what would be established if each river 
 had been examined. 
 
 The elaborate repoi't of General Humphreys to the Bu- 
 
CUAP. XXII. THE EAirrWS AXIAL liOTATlON. 335 
 
 reau of Topographical Eugiueers, AVasliington, shows that 
 the average quantity of earthy matter carried into the Gulf 
 of ]\Iexico, partly suspended in the water and partly pushed 
 along the bottom of the river by the current, amounts for 
 each twelve months to 903,100 millions of pounds. This 
 enormous weight of matter is contributed l>y numerous 
 large branches and upwards of 1,000 small tributaries. The 
 mean distance of the streams along which the sediment is 
 carried in its course to the sea exceeds 1,500 miles. The 
 distance which determines the amount of force tending to 
 check the earth's rotation is, howevei', cousidei'ably shorter. 
 
 The maps of the Mississippi River basin accompanying 
 the report referred to show that its centre is situated 7 deg. 
 10 min. west of the mouth of the main river, and 11 deg. 
 15 min. north of the same, in latitude 40 deg. 15 min. It 
 will be found on inspecting the section of the earth (see 
 PI. 33) that, agreeably to the stated Latitude, the centre 
 of the Mississippi basin rotates in a circle of 15,784,782 ft. 
 radius ; hence its velocity round the axis of the globe is 
 1,147.90 ft. per second. The month of the river, it will be 
 found on calculation, rotates in a circle of 18,246,102 ft. 
 radius, with a circumferential velocity of 1,326.89 ft. per 
 second. Comparing these velocities, it ^\ill be seen that an 
 increased circumferential velocity of 178.99 — say 179 — ft. 
 per second is imparted to the water and to the sediment- 
 ary matter w^hich it conveys during the course from the 
 centre of the basin to the mouth of the ii\er. As l)efore 
 stated, the annual discharge of earthy matter at the mouth 
 
o.]G EABIAM UEAT. chap. xxii. 
 
 of the river is 903,100 millions of pounds. The centre of 
 the basin^lat. 40 deg. 15 min. — being 2,401, oi'O ft. nearer 
 to the axis of rotation than the mouth of the river iu lat. 
 20 deg. min., it will be found that tlie increase of rotary 
 velocity is 179 ft. per second, as already stated — a rate ac- 
 quired by a fall of 500.6 ft. The elements are thus fur- 
 nished for determining with exactness the amount of 
 retardation attending the change of position of the abraded 
 matter during its transfer from the basin to the mouth of 
 the river. Multi2->lying 903,100 millions l)y 500.6, Ave ascer- 
 tain that the counteracting force amounts to 452,000,000,- 
 000,000 foot-pounds annually = 452 X 10" foot-pounds in a 
 century. The earth's present vis viva being 18,875,361 X 
 10'' foot-pounds (to be demonstrated hereafter), it is easy to 
 calculate that the retardation occasioned by the stated re- 
 acting energy called forth by the sedimentary mattei' which 
 is carried to the ocean by the Mississippi will amount to 
 lu ^^oo of a second in a century. In view of this small frac- 
 tion of time, it will be well to remind the reader that the 
 retardation of the earth's rotary velocity, inferred from the 
 apparent acceleration of the moon's mean motion, now gene- 
 rally admitted by astronomers, is somewhat under 12 seconds 
 in a century. Insignificant as this retardation appears to be, 
 it calls for a constant reacting force of 455,000,000,000 foot- 
 pounds per second, as will be shown in the course of our in- 
 vestigation. Dividing this amount by the adopted standard 
 of a horse-power — viz., 550 foot-pounds per second — it will 
 be found that a constant energy represented by 827,000,000 
 
CHAP. XXII. THE EAinir^ AXIAL nOTATION. 337 
 
 horse-po\yer, exerted in a contrary direction to that of rota 
 tion, is necessary to check the rotary motion to the extent 
 mentioned — viz., sUs^ — tsW of a revolution in the course 
 of a century. Accordingly, 720,000 years, nearly, ^\ ill elapse 
 before one entire revolution shall liave been lost, notwith- 
 standing the existence of a constantly retarding force of 
 455,000 niillions of foot-pounds per second. M'e can i-cadily 
 ascertain the aggregate of this force during the long period 
 mentioned, if we multiply the same by the number of revo- 
 lutions of the earth per annum, alid tlie number of seconds 
 for each revolution ; thus, 455 X 10" X 365.24 X 86,400 X 
 720,000 = 103,379,867 X 10" foot-pounds. By dividing this 
 amount of energy in the earth's vis viva, 18,875,361 X 10"' 
 foot-ponnds, we ascertain that the stated enormous retarda- 
 tion overcome in the course of 720,000 years amounts to 
 only ts\ts of the present rotary vis viva of our jjlanet. Pro- 
 liably no other mode of presenting the subject could give 
 so clear an idea of the vastness of the mechanical energy 
 developed by the axial rotation of a sphere 8,000 miles in 
 diameter, whose specific gravity is 2\ times that of granite, 
 revolving at a rate of one revolution in 24 hom*s. Let us 
 bear in mind that the retardation produced by the sedi- 
 mentaiy matter carried to the Gulf of ]\Iexico by the ^lissis- 
 sippi, and the precipitation which causes the abrasion of the 
 solid matter and the currents by which it is conveyed, are 
 the direct results of the sun's radiant heat. 
 
 \\\\\\ reference to the taldes (see pages 342-353), it 
 sliould be stated that the amounts of the retarding force 
 
338 BABIANT HEAT. chap. xxii. 
 
 entered in the last two columns but one are based on tbe data 
 furnished by the examinations of the great Western river — viz., 
 that 1 lb. of solid matter and 1,350 lbs. of water per second 
 are carried to the sea for every 40.08 sq. miles of basin. 
 All other particulars necessary in computing the retarding 
 energy exerted by each river, separately for the two hemi- 
 spheres, will be found in the tables. The mode adopted in 
 determining the area drained by each river and tributaries 
 will be readily comprehended by the following explanation : 
 The extent of the several river basins, 136 in all in both 
 hemispheres, has been ascertained from the best maps extant ; 
 the boundaries of the basin being determined by drawing 
 a line on the map, and dividing the territory equally be- 
 tween the source of each river and tributaries and those of 
 adjoining basins. The boundaries being thus defined, the 
 areas have been calculated in English statute miles ; the 
 latitude and longitude of the centre of each basin being de- 
 termined at the same time. By supposing the earth to be a 
 perfect sphere 7,912.41 miles in diameter, according to Sir 
 John Herschel's determination, the calculations have been 
 rendered extremely simple. This will be seen by reference to 
 the section of the earth before referred to, which contains 
 all the elements for computing the rotary velocity of the 
 centres of the river basins and of the outlets of the rivers. 
 These velocities have been entered in the tables separately 
 for each river basin; also the retardation, expressed in foot- 
 pounds per second, caused by the increase of rotary velocity 
 during the transfer of the sedimentary matter from the cen- 
 
CH A I". XXII. THE EAinWS AXIAI. liOTATION. 339 
 
 tre of the basiu to tlie luoutli of the river. The last coluinn 
 but one of the tables contains the result of computations of 
 the amount of retardation occasioned, by the volume of water 
 which conveys the sedimentary matter — a subject to be con- 
 sidered under a separate head hereafter. 
 
 It should be observed that, o\ving to their trifling in- 
 fluence on the earth's rotation, and in order to save space, 
 all the Englisli and Scotch river basins whose sediment is 
 transferred in the direction of the equator have been en- 
 tered together in the tables ; the rivers of Ireland likewise. 
 But in computing the loss or gain of energy, each river 
 basin has been calculated by itself, the amount of retarda- 
 tion entered being the result of the whole quantity of sedi- 
 ment transferred towards the equator by the several small 
 basins referred to. Accordingly, the area which is entered 
 in the table represents the total. The river basins of 
 Sweden and Norway, being very numerous and unimportant, 
 have also, in some districts, been entered together in tlie 
 tables like those of Great Britain. Finally, the narrow- 
 coast districts, in both hemispheres, have been computed 
 and inserted in the table in a similar manner. 
 
 The quantities of sedimentary matter discharged by the 
 Indus, Ganges, and Brahmapooti-a, being known with toler- 
 able accuracy from actual observation, have not been com- 
 puted according to the standard furnished by the Mississippi, 
 which, as before stated, is 1 lb. of sediment per second for 
 every 40.08 sq. miles of basin. Besides, local circumstances, 
 sucli as the heated waters and profuse evaporation of the 
 
340 RADIANT BEAT. CHAP. xxil. 
 
 Bay of Bengal, and the powerful condeusatiou attending 
 the close vicinity of the Himalaya Mountains, render the 
 Gauges quite exceptional. 
 
 Respecting the African rivers, none of which have been 
 entered in the tables, it may be briefly stated that they 
 luive no material influence on the earth's rotation, from the 
 fact that the two principal rivers, the Nile and the Niger, 
 flow in opposite directions — the former towards the pole 
 and the latter towards the equator. There is, however, 
 considerable dift'ereuce of latiti;de, productive of an in- 
 creased retarding influence of the Nile ; but this cannot be 
 far from balanced by the greater quantity of sedimentary 
 matter brought down by the Niger, as pi'oved by its delta 
 of 240 miles of coast. The general course of the other im- 
 portant rivers of Africa — the Senegal, Zambesi, and the 
 Oi'ange River — is so nearly parallel with the equator that 
 they exercise no appreciable influence on the axial rotation 
 of the earth. 
 
 Australia, being drained by rivers the courses of which 
 are directed to all points of the compass, consequently exer- 
 cising no appreciable influence as regards the earth's i-otary 
 motion, has likewise been excluded from the tables. It 
 should be observed that the basin of the important river 
 Goolwa and its tributaries (excepting the Callewatta) is 
 almost on the same parallel with the mouth of the main 
 river; hence scarcely any retarding force is produced, uot- 
 Avithstanding the great extent of basin drained by the 
 Goolwa. The Amazon, which drains more than two mil- 
 
CHAP XXII. THE EAETirS AXIAL ROTATION. 341 
 
 limis of Siiuare miles, strikingly illustrates the trifling in- 
 tlueuce on the earth's rotary velocity of rivers the centres 
 of whose basins are nearly on the same parallel \vith their 
 outlets, the enormous mass of solid matter carried to the 
 ocean by this river — the greatest ou the globe — exerting a 
 retarding influence of only 70,000 foot-pounds per second. 
 
 The asrorreorate of solid matter removed from its orii^inal 
 position by the river S3'stems of both hemispheres, and carried 
 towards the equator — consecpiently removed to a greatei' dis- 
 tance from the axis of rotation — exerts, as shown by the 
 tables, a retai-ding influence of 39,894,658 foot-pounds per 
 second. If we multiply this amount by 86,400 seconds, we 
 learn that for each revolution the earth has to overcome 
 a retarding energy represented by 3,4-46,898,451,200 foot- 
 pounds ; l^ut the effect of this retardation as regards the 
 length of the century cannot be properly considered until we 
 have investigated the second class of force before adverted 
 to, viz., that force which destroys tlie earth's vis viva Avith- 
 out disturbing the position of its centre of gyration. We 
 have, however, proceeded far enough with our investigation 
 to show the fallacy of the accepted doctrine of compensation 
 lelative to the energies which affect the earth's rotary velo- 
 city. We have clearly shown that constancy of rotation of 
 the earth is incompatible with solar influence. 
 
349 
 
 BAD J ANT HEAT. 
 
 CHAP. XXII. 
 
 elvers flowing towards the equator. 
 Hemisphere. 
 
 Eastern 
 
 Name op Rivee oe Disteict. 
 
 Anadir , 
 
 N. W. coast Sea of Kamtschatka 
 
 Penshina 
 
 W. coast Sea of Okhotsk 
 
 Yama 
 
 Tawi and Kowa 
 
 Ochola 
 
 Ud 
 
 Tollmen 
 
 Yaloo 
 
 Sira Muren 
 
 Chanton 
 
 Pei-ho 
 
 Min Kiang 
 
 Han Kiang 
 
 Tcke Kiang 
 
 Hoang Ho 
 
 Sang Koi 
 
 Coast Eivers, Gulf of Tonqnin 
 
 Menam Kong 
 
 Irawady 
 
 Bralimapootra 
 
 Ganges above Gliazepoor 
 
 Ganges below Gliazepoor 
 
 Rivee Basin. 
 
 Area. 
 
 Sq. MUm. 
 
 119,200 
 
 52,000 
 
 50,100 
 
 32,600 
 
 15,500 
 
 13,300 
 
 21,500 
 
 25,400 
 
 20,000 
 
 21,200 
 
 78,700 
 
 29,800 
 
 124,000 
 
 25,400 
 
 26,200 
 
 144,400 
 
 448,200 
 
 76,300 
 
 26,000 
 
 329,500 
 
 205,000 
 
 379,000 
 
 187,100 
 
 237,200 
 
 
 Deg. M. 
 
 65 50 
 6142 
 63 20 
 62 00 
 60 45 
 60 25 
 60 48 
 55 25 
 43 20 
 
 41 10 
 
 42 40 
 41 00 
 38 40 
 
 26 20 
 24 18 
 24 10 
 36 50 
 22 40 
 20 12 
 16 38 
 22 35 
 29 12 
 
 27 25 
 27 30 
 
 East 
 
 of 
 outlet. 
 
 Deg. M. 
 
 10 
 20 
 
 05 
 
 05 
 
 05 
 
 West 
 
 of 
 outlet. 
 
 135 
 
 05 
 130 
 10 
 40 
 2 35 
 
 15 
 
 1 06 
 
 30 
 
 2 45 
 
 1 15 
 
 3 50 
 10 25 
 
 2 28 
 150 
 
 3 15 
 30 
 
 5 25 
 5 12 
 
 622 
 
 720 
 
 681 
 
 713 
 
 742 
 
 750 
 
 741 
 
 8G2 
 
 1,105 
 
 1,144 
 
 1,117 
 
 1,146 
 
 1,186 
 
 1,362 
 
 1,384 
 
 1,386 
 
 1,216 
 
 1,402 
 
 1,426 
 
 1,455 
 
 1,403 
 
 1,326 
 
 1,346 
 
 1,347 
 
 3 
 
 4 
 5 
 6 
 
 7 
 8 
 
 9 
 
 lO 
 
 II 
 
 12 
 13 
 
 ■4 
 IS 
 16 
 
 17 
 18 
 
 19 
 
 20 
 21 
 22 
 
 23 
 •24 
 
CHAP. XXII. 
 
 THE EAh'TWS AXIAL DOTATION. 
 
 343 
 
 2 
 
 3 
 4 
 5 
 6 
 
 7 
 8 
 
 9 
 
 10 
 
 II 
 
 12 
 
 13 
 
 14 
 15 
 i6 
 
 17 
 i8 
 
 19 
 20 
 21 
 22 
 23 
 24 
 
 Rivers 
 
 FLOWING TOWARDS 
 
 THE Equator. 
 
 Eastern 
 
 
 
 Hemisphere. 
 
 
 Mouth of Kiveb, 
 
 Retardation. 
 
 ■§ 
 
 5"^ 
 
 i -§* 
 
 
 
 
 3 
 
 3 
 
 
 1 1 
 
 By sediment 
 
 By water. 
 
 Total. 
 
 Deg. if. 
 
 Fett ptr 
 etcond. 
 
 Feet per 
 second. 
 
 Foot-pounds per 
 teoond. 
 
 Foot-pound t per teoond. 
 
 Foot-povnde per tecond. 
 
 64 30 
 
 054 
 
 32 
 
 47,680 
 
 64,368,000 
 
 64,415,680 
 
 GO 45 
 
 742 
 
 22 
 
 9,828 
 
 13,267,800 
 
 13,277,628 
 
 61 46 
 
 719 
 
 38 
 
 28,256 
 
 38,145,600 
 
 38,173,856 
 
 61 20 
 
 729 
 
 16 
 
 3,260 
 
 4,401,000 
 
 4,404,260 
 
 59 40 
 
 767 
 
 25 
 
 3,786 
 
 5,111,100 
 
 5,114,886 
 
 59 40 
 
 767 
 
 17 
 
 1,503 
 
 2,029,050 
 
 2,030,553 
 
 59 45 
 
 765 
 
 24 
 
 4,838 
 
 6,-531,300 
 
 6,536,138 
 
 55 20 
 
 864 
 
 2 
 
 40 
 
 54,000 
 
 54,040 
 
 42 65 
 
 1,112 
 
 7 
 
 . 385 
 
 519,700 
 
 520,085 
 
 40 10 
 
 1,161 
 
 17 
 
 2,396 
 
 3,234,600 
 
 3,236,996 
 
 40 50 
 
 1,149 
 
 32 
 
 31,480 
 
 42,498,000 
 
 42,529,480 
 
 39 10 
 
 1,178 
 
 32 
 
 11,920 
 
 16,092,000 
 
 16,103,920 
 
 38 25 
 
 1,187 
 
 1 
 
 48 
 
 64,800 
 
 64,848 
 
 26 00 
 
 1,364 
 
 2 
 
 40 
 
 54,000 
 
 54,040 
 
 23 20 
 
 1,395 
 
 13 
 
 173 
 
 233,550 
 
 233,723 
 
 22 53 
 
 1,399 
 
 13 
 
 9,530 
 
 12,865,500 
 
 12,875,030 
 
 33 45 
 
 1,263 
 
 47 
 
 386,797 
 
 522,175,950 
 
 ."^22, 502,747 
 
 20 32 
 
 1,423 
 
 21 
 
 13,143 
 
 17,743,050 
 
 17,756,193 
 
 20 00 
 
 1,427 
 
 1 
 
 10 
 
 13,500 
 
 13,510 
 
 10 Co 
 
 1,496 
 
 41 
 
 216,399 
 
 292,138,650 
 
 292,355,049 
 
 16 10 
 
 1,459 
 
 56 
 
 251,125 
 
 339,018,750 
 
 339,269,875 
 
 22 00 
 
 1,408 
 
 82 
 
 5,187,040 
 
 7,002,504,000 
 
 7,007,691,040 
 
 22 00 
 
 1,408 
 
 62 
 
 1,463,818 
 
 1,976,154,300 
 
 1,977,0!8,1]8 
 
 22 UO 
 
 1,408 
 
 61 
 
 2,694,633 
 
 3,637,754,550 
 
 3,640,449,183 
 
344 
 
 BABIANT HEAT. 
 
 CHAP. XXII. 
 
 Rivers flowing towards the Equator. Eastern 
 Hemisphere. 
 
 Name of River or District. 
 
 EiVER Basin. 
 
 Area. 
 
 Sq. Miles. 
 
 Dig. M. 
 
 East 
 
 of 
 outlet. 
 
 West 
 
 of 
 outlet. 
 
 Deg. it 
 
 fc-.t 
 
 Khaladaing 
 
 Braining and Coyli' 
 
 Maliannddy 
 
 N. W. coast Bay of Bengal . . . . 
 
 Godavery 
 
 Kislina 
 
 W. coast Bay of Bengal 
 
 Penaar 
 
 Palar 
 
 Cauvery 
 
 Tapty 
 
 Nurbndda 
 
 Sukernnlly 
 
 Jahu 
 
 Indns 
 
 Helmund 
 
 N. coast of Arabi;in Sea.. ...... 
 
 N. E. coast of Persian Gulf. . . . 
 
 S. coast of Arabia 
 
 Euphrates and Tigris 
 
 Kour 
 
 Oural 
 
 Don 
 
 Emba district 
 
 27,400 
 
 31,000 
 
 41,300 
 
 24,000 
 
 133,000 
 
 104,900 
 
 8,300 
 
 17,900 
 
 23,700 
 
 25,000 
 
 23,600 
 
 21,600 
 
 29,700 
 
 48.300 
 
 346,300 
 
 124,000 
 
 93,000 
 
 66,000 
 
 96,000 
 
 282,100 
 
 95,400 
 
 126,300 
 
 195,400 
 
 37,800 
 
 21 15 
 
 22 12 
 21 02 
 19 15 
 19 35 
 16 45 
 15 45 
 
 14 45 
 12 52 
 11 45 
 
 21 15 
 
 22 14 
 
 23 27 
 26 25 
 33 30 
 32 42 
 26 30 
 29 00 
 
 15 20 
 35 05 
 39 55 
 50 04 
 48.56 
 47 50 
 
 2 60 
 
 3 45 
 
 25 
 
 2 20 
 5 20 
 
 3 25 
 
 1 00 
 45 
 
 2 01 
 1 48 
 
 3 00 
 
 20 
 126 
 3 55 
 32 
 
 3 25 
 
 4 35 
 
 18 
 
 1 30 
 1 20 
 153 
 
 50 
 4 35 
 3 32 
 
 1,416 
 1,406 
 1,418 
 1,434 
 1,431 
 1,455 
 1,462 
 1,469 
 1,481 
 1,487 
 1,416 
 1,406 
 1,394 
 1,360 
 1,267 
 1,278 
 1,359 
 1,329 
 1,465 
 1,243 
 1,165 
 975 
 998 
 1,020 
 
 9 
 
 lO 
 
 II 
 
 12 
 
 13 
 14 
 15 
 i6 
 
 17 
 i8 
 19 
 20 
 21 
 22 
 
 23 
 
 24 
 
CHAP. XXII. 
 
 thl: j!:AiiTjrs axial uotatwn. 
 
 345 
 
 
 RlVEUS 
 
 FLOWING TOWARDS 
 
 THE Equatou. Eastern 
 
 
 
 
 IlEillSPUEKE. 
 
 
 
 Mouth of Riteb. 
 
 Retabdation. 
 
 
 
 II 
 
 
 By sediment 
 
 By water. 
 
 Total 
 
 
 Z>W. if. 
 
 F*tt ptr 
 etcond. 
 
 Feet per 
 second. 
 
 Footpoundt per 
 troond. 
 
 Foot-poundt ptr steonit. 
 
 Footpounds per teoond. 
 
 I 
 
 20 30 
 
 1,423 
 
 7 
 
 527 
 
 711,450 
 
 711,977 
 
 2 
 
 21 16 
 
 1,416 
 
 10 
 
 1,209 
 
 1,632,150 
 
 1,633,359 
 
 3 
 
 20 28 
 
 1,423 
 
 5 
 
 403 
 
 544,0.50 
 
 544,4.53 
 
 4 
 
 18 20 
 
 1,442 
 
 8 
 
 600 
 
 810,000 
 
 810,600 
 
 5 
 
 IG 50 
 
 1,454 
 
 23 
 
 27,498 
 
 37,122,300 
 
 37,149,798 
 
 6 
 
 10 10 
 
 1,459 
 
 4 
 
 656 
 
 885,600 
 
 886,256 
 
 7 
 
 L) 30 
 
 1,464 
 
 2 
 
 13 
 
 17,550 
 
 17,563 
 
 8 
 
 14 35 
 
 1,470 
 
 1 
 
 7 
 
 9,450 
 
 9,457 
 
 9 
 
 12 40 
 
 1,482 
 
 1 
 
 9 
 
 12,150 
 
 12,159 
 
 lO 
 
 10 55 
 
 1,492 
 
 5 
 
 244 
 
 329,400 
 
 329,644 
 
 1 1 
 
 21 10 
 
 1,417 
 
 1 
 
 9 
 
 12,150 
 
 12,159 
 
 12 
 
 •21 48 
 
 1,4]0 
 
 4 
 
 135 
 
 182,250 
 
 182,385 
 
 13 
 
 22 30 
 
 1,403 
 
 9 
 
 943 
 
 1,273,050 
 
 1.273,993 
 
 14 
 
 24 52 
 
 1,378 
 
 18 
 
 6,110 
 
 8, 248, {500 
 
 S,2.j4,610 
 
 15 
 
 24 00 
 
 1,388 
 
 121 
 
 5,140,969 
 
 6,940,308,1.50 
 
 6,945,449,119 
 
 i6 
 
 31 52 
 
 1,290 
 
 12 
 
 6,975 
 
 9,416,250 
 
 9,423,225 
 
 17 
 
 25 15 
 
 1,374 
 
 15 
 
 8,184 
 
 11,048,400 
 
 11,056,584 
 
 i8 
 
 27 50 
 
 1,343 
 
 14 
 
 5,040 
 
 6,816,150 
 
 6,821,199 
 
 •9 
 
 13 50 
 
 1.475 
 
 10 
 
 3,744 
 
 5,054,400 
 
 .5,058,144 1 
 
 1 
 
 20 
 
 30 05 
 
 1,314 
 
 71 
 
 5.')5,525 
 
 749,958,750 
 
 750,514,276 ' 
 
 21 
 
 39 52 
 
 1,166 
 
 1 
 
 37 
 
 49,950 
 
 49,987 
 
 22 
 
 47 02 
 
 1,035 
 
 60 
 
 177,609 
 
 239,772,150 
 
 239,949,759 
 
 23 
 
 47 06 
 
 1,034 
 
 39 
 
 98,921 
 
 133,.543,3.50 
 
 133,642,271 
 
 24 
 
 46 50 
 
 1,039 
 
 19 
 
 5,330 
 
 7,19.5,500 
 
 7,200,830 
 
346 
 
 EADIANT HEAT. 
 
 CHAP. XXII. 
 
 KlVEItS FLOWING TOWARDS THE EqUATOU. 
 
 Hemispheke. 
 
 Eastern 
 
 NAiiE OF River or District. 
 
 River Basin. 
 
 Sq. Miles. 
 
 Deg. M. 
 
 East 
 
 of 
 outlet. 
 
 West 
 
 of 
 outlet. 
 
 «£ 
 
 Volga 
 
 Kuban 
 
 Rioni 
 
 Syrian Mediterranean rivers.. 
 
 Jordan district 
 
 School! 
 
 Axon and otliers 
 
 Minder 
 
 Sarabat 
 
 Dnieper 
 
 Bug and Indul 
 
 Dneister 
 
 Maritza 
 
 Striman 
 
 Vardar 
 
 Drin 
 
 Donau 
 
 Tornea and Kcmi district 
 
 Wester Botten 
 
 Wester Noniaud 
 
 River Dalil district 
 
 Clara and Gotlia Elv 
 
 Malar district 
 
 Eastern part of South Sweden 
 
 557,700 
 
 34,600 
 
 17,900 
 
 8,800 
 
 12,400 
 
 12,000 
 
 23,600 
 
 9,900 
 
 8,300 
 
 181,000 
 
 27,400 
 
 30,500 
 
 23,700 
 
 13,000 
 
 12,300 
 
 9,500 
 
 293,100 
 
 41,700 
 
 36,900 
 
 26,200 
 
 20,300 
 
 16,700 
 
 9,400 
 
 .5,800 
 
 55 00 
 44 50 
 42 38 
 33 40 
 32 20 
 47 40 
 47 32 
 
 47 40 
 
 48 22 
 5125 
 
 49 05 
 48 32 
 42 00 
 4128 
 
 41 20 
 
 42 10 
 40 25 
 67 20 
 65 50 
 63 30 
 6116 
 59 40 
 59 45 
 
 56 55 
 
 10 
 
 2 55 
 
 50 
 
 1 10 
 35 
 
 15 
 
 2 00 
 
 1 35 
 1 15 
 
 25 
 
 45 
 
 33 
 121 
 2 08 
 22 
 30 
 010 
 
 9 50 
 05 
 2 55 
 158 
 2 00 
 
 110 
 100 
 
 871 
 
 1,077 
 
 1,118 
 
 1,265 
 
 1,285 
 
 1,023 
 
 1,026 
 
 1,023 
 
 1,009 
 
 947 
 
 905 
 
 1,006 
 
 1,129 
 
 1,138 
 
 1,141 
 
 1,126 
 
 1,047 
 
 585 
 
 622 
 
 678 
 
 730 
 
 767 
 
 765 
 
 829 
 
 14 
 15 
 i6 
 
 17 
 i8 
 
 19 
 20 
 21 
 
 22 
 
 23 
 24 
 
CHAP. XXII. 
 
 THE EAirrirs axial hotatwn. 
 
 347 
 
 
 RiVEKS 
 
 FLOWING TOWARDS 
 
 THE EliDATOK. ] 
 
 ] ASTERN 
 
 
 
 
 1Ii:misi'iiehk. 
 
 
 
 Mouth of River. 
 
 Retardation. 
 
 
 4 
 3 
 
 ^1 
 
 
 By sediment. 
 
 By water. 
 
 Total 
 
 
 j>ta. if. 
 
 Fttt ptr 
 mcond. 
 
 Fest per 
 seootul. 
 
 Foct-poundH per 
 tfcond. 
 
 Foot-poundu per second. 
 
 Fooi-pound» per second. 
 
 I 
 
 46 10. 
 
 1,052 
 
 181 
 
 7,137,020 
 
 9,634,085,100 
 
 9,642,122,120 
 
 2 
 
 44 42 
 
 1,080 
 
 3 
 
 121 
 
 163,350 
 
 103,471 
 
 3 
 
 42 15 
 
 1,124 
 
 6 
 
 251 
 
 338,850 
 
 339,101 
 
 4 38 22 
 
 1,268 
 
 3 
 
 30 
 
 40,500 
 
 40,530 
 
 5 31 48 
 
 1,290 
 
 6 
 
 121 
 
 163,350 
 
 103,471 
 
 6 46 45 
 
 1,041 
 
 18 
 
 1,518 
 
 2,049,300 
 
 2,050,818 
 
 7 46 40 
 
 1,042 
 
 16 
 
 2,360 
 
 3,180,000 
 
 3,188,360 
 
 8 47 30 
 
 1,026 
 
 3 
 
 35 
 
 47,250 
 
 47,285 
 
 9 ! 48 18 
 
 1,010 
 
 1 
 
 3 
 
 4,050 
 
 4,053 
 
 lO 46 40 
 
 1.042 
 
 95 
 
 638,115 
 
 801,4.5,5,250 
 
 862,093,365 
 
 11 46 42 
 
 1,042 
 
 47 
 
 23,646 
 
 31,922,100 
 
 31,94.%740 
 
 12 46 20 
 
 1,049 
 
 43 
 
 22,029 
 
 29,739,1.50 
 
 29,761,179 
 
 13 ' 40 4i5 
 
 1,148 
 
 19 
 
 3,342 
 
 4,511,700 
 
 4,51.5,042 
 
 14 ' 40 55 
 
 1,148 
 
 10 
 
 507 
 
 684,4.50 
 
 684,957 
 
 15 1 40 32 
 
 1,154 
 
 13 
 
 812 
 
 1,090,200 
 
 1,097,012 
 
 16 41 40 
 
 1,135 
 
 9 
 
 302 
 
 407,700 
 
 408,002 
 
 17 45 15 
 
 1,069 
 
 22 
 
 55,396 
 
 74,784,000 
 
 74,839,996 
 
 18 1 65 52 
 
 621 
 
 36 
 
 21,111 
 
 28,499,850 
 
 28,-520,961 
 
 19 i 64 40 
 
 650 
 
 28 
 
 11,301 
 
 15,256,3.50 
 
 15,267,651 
 
 20 ] 62 20 
 
 705 
 
 27 
 
 7,460 
 
 10,071,000 
 
 10,078,460 
 
 21 60 35 
 
 750 
 
 20 
 
 3,172 
 
 4,282,200 
 
 4,285,372 
 
 22 D7 48 
 
 809 
 
 42 
 
 11,506 
 
 15,533,100 
 
 1.5, .544, 606 
 
 23 ' 50 25 
 
 773 
 
 8 
 
 235 
 
 317,2.^0 
 
 317,485 
 
 24 
 
 1 56 15 
 
 844 
 
 15 
 
 510 
 
 ' 688, .500 
 
 689,010 
 
348 
 
 EABIANT HEAT. 
 
 CHAP. XXII. 
 
 klvers plowing towards the equator. eastern 
 Hemisphere. 
 
 Name op Riveb or District. 
 
 River Basin. 
 
 Sq. Miles. 
 
 5° g 
 3 " 
 
 East 
 
 of 
 outlet. 
 
 Deg. M. 
 
 West 
 
 of 
 outlet. 
 
 Western part of South Sweden. 
 Glommen and Lauven district.. 
 
 Southern part of Norway 
 
 E. coast of Adriatic 
 
 Grulf of Taranto and Ionian Sea. 
 Western part of South Italy. . . 
 
 Tiber 
 
 Arno and W. coast Central Italy 
 
 Etsch district 
 
 Po 
 
 Rhone 
 
 Ter and Slobregat 
 
 Ebro 
 
 Guadalaviar 
 
 Jucar 
 
 Segura 
 
 Guadalquiver 
 
 Gnadiana 
 
 Caldoa 
 
 Tflgus 
 
 Duero 
 
 Minho 
 
 Rivers of Great Britain 
 
 Rivers of Ireland 
 
 6,600 
 
 21,300 
 
 9,500 
 
 17,200 
 
 7,500 
 
 11,900 
 
 7,200 
 
 8,900 
 
 12,000 
 
 28,100 
 
 38,000 
 
 11,200 
 
 33,200 
 
 8,900 
 
 8,100 
 
 8,200 
 
 20,000 
 
 25,700 
 
 7,100 
 
 28,900 
 
 38,700 
 
 10,600 
 
 46,015 
 
 16,224 
 
 57 30 
 60 52 
 59 18 
 43 50 
 
 40 00 
 
 41 15 
 
 42 42 
 
 43 25 
 46 15 
 45 25 
 45 42 
 42 45 
 42 06 
 40 10 
 
 39 15 
 38 10 
 
 37 40 
 
 38 25 
 37 48 
 
 40 00 
 
 41 30 
 
 42 41 
 
 42 
 
 20 
 
 30 
 17 
 32 
 
 41 
 
 2 02 
 
 2 00 
 20 
 
 3 50 
 3 05 
 115 
 
 04 
 15 
 
 25 
 
 34 
 3 10 
 
 25 
 2 20 
 
 54 
 115 
 
 1 20 
 
 817 
 740 
 776 
 1,096 
 1,164 
 1,142 
 1,116 
 1,103 
 1,050 
 1,065 
 1,061 
 1,115 
 1,127 
 1.161 
 1,176 
 1,194 
 1,202 
 1,190 
 1,200 
 1,164 
 1,139 
 1,117 
 
 '3 
 14 
 
 f5 
 i6 
 
 '7 
 [8 
 
 '9 
 
 20 
 
CHAP. XXII. 
 
 THE EAETWS AXIAL llOTATION. 
 
 349 
 
 
 
 lllVEItS 
 
 I'LOWING TOWARDS 
 
 THE Equator. Eastern 
 
 
 
 
 
 Hemisphere. 
 
 
 
 Mouth of Riyeb. 
 
 Retabdatiox. I 
 
 
 
 
 3 
 
 
 
 By sediment. 
 
 By water. 
 
 Total 
 
 
 />«». j^. 
 
 Ftttper 
 ttcond. 
 
 Feet per 
 itecoiui. 
 
 Foot-pounds per 
 tteond. 
 
 Font-pounde per second. 
 
 Foot-pounds per eeeond. 
 
 I 
 
 56 40 
 
 835 
 
 18 
 
 835 
 
 1,127,250 
 
 1,128,085 
 
 2 
 
 50 10 
 
 779 
 
 39 
 
 12,658 
 
 17,088,300 
 
 17,100,958 
 
 3 
 
 58 10 
 
 801 
 
 26 
 
 2,320 
 
 3,132,000 
 
 3,134.320 
 
 4 
 
 43 30 
 
 1,102 
 
 6 
 
 241 
 
 325,350 
 
 325,591 
 
 5 
 
 39 30 
 
 1,172 
 
 8 
 
 188 
 
 253,800 
 
 253,988 
 
 6 
 
 40 40 
 
 1,152 
 
 10 
 
 407 
 
 630,450 
 
 630,017 
 
 7 
 
 41 45 
 
 1,133 
 
 17 
 
 814 
 
 1,008,900 
 
 1,000,714 
 
 8 
 
 43 00 
 
 1,111 
 
 8 
 
 223 
 
 301,050 
 
 301,273 
 
 9 
 
 45 00 
 
 1,074 
 
 24 
 
 2,700 
 
 3,045,000 
 
 3,647,700 
 
 lO 
 
 44 50 
 
 1,077 
 
 12 
 
 1,412 
 
 1,906,200 
 
 1,907,612 
 
 11 
 
 43 25 
 
 1,103 
 
 42 
 
 26,182 
 
 35,345,700 
 
 35,371,882 
 
 12 
 
 42 40 
 
 1,117 
 
 2 
 
 18 
 
 24,300 
 
 24,318 
 
 13 
 
 41 02 
 
 1,146 
 
 10 
 
 4,681 
 
 6,319,350 
 
 6,324,031 
 
 14 
 
 39 20 
 
 1,175 
 
 14 
 
 681 
 
 010,350 
 
 920,031 
 
 >5 
 
 39 00 
 
 1,180 
 
 4 
 
 51 
 
 08,850 
 
 68,901 
 
 i6 
 
 38 06 
 
 1,195 
 
 1 
 
 3 
 
 4,050 
 
 4,053 
 
 17 
 
 36 42 
 
 1,218 
 
 16 
 
 2,000 
 
 2,700,000 
 
 2,702,000 
 
 i8 
 
 36 48 
 
 1,216 
 
 26 
 
 6,785 
 
 9,159,750 
 
 9,166,535 
 
 19 
 
 37 40 
 
 1,202 
 
 2 
 
 11 
 
 14,850 
 
 14,861 1 
 
 20 
 
 38 60 
 
 1,183 
 
 19 
 
 4,075 
 
 5,501,250 
 
 5,505,325 
 
 21 41 15 
 
 1,142 
 
 3 
 
 135 
 
 182,250 
 
 182,385 
 
 22 42 28 
 
 1,121 
 
 4 
 
 66 
 
 80,100 
 
 80,106 
 
 23 
 
 
 
 1,181 
 
 1,594,350 
 
 1,505, .531 
 
 24 
 
 
 
 
 021 
 
 1,243,350 
 
 1,244,271 
 
350 
 
 RADIANT HEAT. 
 
 CHAP. XXII. 
 
 RiVEKS FLOWIXG TOWARDS THE EQUATOR. AVeSTERN 
 
 Hemisphere. 
 
 Naste of River or District. 
 
 RivEE Basin. 
 
 Sg. Mltu. 
 
 Deg. M. 
 
 East 
 
 of 
 
 outlet. 
 
 West 
 
 of 
 outlet. 
 
 Deg. M 
 
 Bastard, Pentecost 
 
 Sagueiiay 
 
 St. John, Penobscot 
 
 Kennebec, Androscoggin, Saco 
 
 Merrimack 
 
 Connecticut 
 
 Hudson, Housatonic 
 
 Delaware 
 
 Susquelianna 
 
 Potomac, Rappahannoclv 
 
 .James 
 
 Roanoke, Tar 
 
 Santee, Neuse 
 
 Savannah and others 
 
 Alabama district 
 
 Mississippi 
 
 Colorado, Brazos, 'J'rinidad . . 
 
 Rio del Norte 
 
 Colorado 
 
 E. coast Gulf of California. . . 
 
 33,000 
 
 27,800 
 
 34,400 
 
 13,000 
 
 7,100 
 
 10,800 
 
 16,000 
 
 11,500 
 
 27,200 
 
 16,000 
 
 13,100 
 
 18,800 
 
 44,100 
 
 ,^1,200 
 
 128,700 
 
 ,244,000 
 
 191,200 
 
 -2V:^SM){) 
 
 267,000 
 
 140,000 
 
 50 50 
 49 02 
 46 40 
 44 45 
 43 20 
 43 20 
 42 30 
 41 00 
 41 14 
 89 00 
 37 32 
 36 36 
 34 40 
 32 51 
 31 30 
 40 55 
 31 12 
 29 50 
 36 04 
 28 06 
 
 02 
 04 
 03 
 
 25 
 
 1 06 
 1 10 
 
 05 
 2 45 
 
 1 00 
 20 
 38 
 
 52 
 
 1 15 
 1 32 
 1 40 
 1 25 
 1 05 
 
 7 10 
 1 .50 
 6 40 
 
 959 
 996 
 1,042 
 1,078 
 1,105 
 1,105 
 1,120 
 1,146 
 1,142 
 1,180 
 1,204 
 1.220 
 1,249 
 1,276 
 1,295 
 1,148 
 1,300 
 1,318 
 1,228 
 1,340 
 
CHAP. XXII. 
 
 THE KARTWa AXIAL ROTATION. 
 
 351 
 
 
 ItlVElis FLUWINO TOWARDS 
 
 1 
 
 TlIK EcjU.VTOU. WeSTEBN 
 
 
 
 Hemispuekh. 
 
 
 
 Mouth of Riveb. 
 
 Rktaedahon. 
 
 
 •3 
 1 
 
 
 s 1 
 
 By sediment. 
 
 By water. 
 
 Total. 
 
 
 7)^. .v. 
 
 Fttt per 
 tecond. 
 
 Faet per 
 
 »tCOHd, 
 
 Foot-poundt per 
 ercottd. 
 
 Fool-pounde per teeond. 
 
 Foot-pounds per ueond. 
 
 I 
 
 49 08 
 
 994 
 
 35 
 
 15,790 
 
 21,316,500 
 
 21,332,290 
 
 2 
 
 48 00 
 
 1,017 
 
 21 
 
 4,788 
 
 6,463,800 
 
 6,468,588 
 
 3 
 
 45 20 
 
 1,068 
 
 26 
 
 9,081 
 
 12,259,8.50 
 
 12,268,431 
 
 4 
 
 43 55 
 
 1,094 
 
 16 
 
 1,300 
 
 1,755,000 
 
 1,756,300 
 
 5 
 
 42 48 
 
 1,114 
 
 9 
 
 225 
 
 303,750 
 
 303,975 
 
 6 
 
 41 10 
 
 1,143 
 
 38 
 
 6,091 
 
 8,222,850 
 
 8.228,941 
 
 7 
 
 40 42 
 
 1,152 
 
 32 
 
 6,400 
 
 8,640,000 
 
 8,646,400 
 
 8 
 
 39 35 
 
 1,172 
 
 26 
 
 3,036 
 
 4,098,600 
 
 4,101,636 
 
 9 
 
 39 32 
 
 1,172 30 
 
 9,561 
 
 12.907,350 
 
 12,916,911 
 
 lO 
 
 38 02 
 
 1,197 ' 17 
 
 1,806 
 
 2,488,100 
 
 2,439,906 
 
 1 1 
 
 36 52 
 
 1,215 
 
 11 
 
 619 
 
 835,6.50 
 
 836,269 
 
 12 
 
 35 52 
 
 1,221 
 
 11 
 
 888 
 
 1.198,800 
 
 1,199,688 
 
 '3 
 
 33 42 
 
 1,264 
 
 15 
 
 3,881 
 
 5,239,350 
 
 5,248.231 
 
 '4 
 
 31 52 
 
 1,290 
 
 14 
 
 3,917 
 
 5,287,950 
 
 5.291.867 
 
 •5 
 
 30 15 
 
 1,312 
 
 17 
 
 14, .543 
 
 19,633,0.50 
 
 19,647.-593 
 
 i6 
 
 29 08 
 
 1,327 
 
 179 
 
 14,323,668 
 
 19,336,951,800 
 
 19,351,275,468 
 
 17 
 
 28 40 
 
 1,333 
 
 33 
 
 81,354 
 
 109,827,900 
 
 109,909,254 
 
 i8 
 
 25 30 
 
 1.371 
 
 53 
 
 236,457 
 
 319,216,950 
 
 319,453,407 
 
 •9 
 
 32 15 
 
 1.285 
 
 57 
 
 338,890 
 
 457,501,500 
 
 457,840,390 
 
 20 
 
 27 00 
 
 1,353 13 
 
 9,240 
 
 12,474,000 
 
 12.483,240 
 
352 
 
 RADIANT HEAT. 
 
 CHAP. XXII. 
 
 ElVEi;S FLOWING TOWARDS THE EQUATOR. WeSTEKX 
 
 Hemisphere. 
 
 Najie of Utter or Disteict. 
 
 ■RiTER Basin. 
 
 
 Sq. itiles. 
 
 Deo- J/. 
 
 East 
 
 of 
 outlet. 
 
 Dea. M. 
 
 West 
 of 
 
 outlet. 
 
 Kalamatli and Tsashtl 
 
 Columbia 
 
 Frazer 
 
 Simpson and Frances district. 
 
 Atna or Copper 
 
 Bolsas, Yopez, Verde 
 
 Sirano district 
 
 San Juan de Mcaragua. . . . 
 
 Sacramento 
 
 Paraliyba and Grande 
 
 Ciara, Croayhn 
 
 Jaguaribe 
 
 Belmoute and Doce dist.. . . 
 
 Paraliyba (Sontli) 
 
 San Francisco 
 
 Paranahyba 
 
 Maranhao and Itaqieura. . . 
 
 Gurupy and Turyassu 
 
 Tocantins 
 
 Amazon 
 
 42,900 
 
 283,400 
 
 138,300 
 
 53,600 
 
 34,900 
 
 71,. 500 
 
 24,700 
 
 23,700 
 
 33,000 
 
 37,600 
 
 33,700 
 
 19,900 
 
 114,400 
 
 35,800 
 
 247,600 
 
 157,300 
 
 59,600 
 
 49,000 
 
 376,000 
 
 2,236,000 
 
 42 00 
 
 45 52 
 
 5128 
 
 56 40 
 
 62 12 
 
 18 00 
 
 14 00 
 
 12 00 
 
 40 00 
 
 6 40 
 
 5 08 
 
 5 50 
 
 17 25 
 
 22 00 
 
 14 00 
 
 8 05 
 
 5 00 
 3 20 
 
 10 10 
 
 6 20 
 
 1 15 
 
 7 02 
 
 2 00 
 05 
 05 
 10 
 
 40 
 
 1 20 
 
 08 
 
 1 50 
 05 
 
 30 
 
 1 48 
 
 2 12 
 7 00 
 2 30 
 
 40 
 100 
 
 1 00 
 1250 
 
 1,129 
 1,058 
 946 
 835 
 708 
 1,445 
 1,474 
 1,486 
 1,164 
 1,509 
 1,513 
 1,511 
 1,449 
 1,408 
 1,475 
 1,504 
 1,513 
 1,516 
 1,495 
 1,510 
 
 / 
 8 
 
 9 
 
 lO 
 
 II 
 
 12 
 
 '3 
 14 
 IS 
 i6 
 
 17 
 i8 
 
 19 
 20 
 
CHAP. XXII. 
 
 THE EARTH'S AXIAL ROTATION. 
 
 353 
 
 
 
 KlVERS 
 
 PLOWING TOWARDS TUE EqDATOR. WeSTERK 
 
 Hemisphere. 
 
 
 Mouth of Riveb. 
 
 Retakdation. 
 
 
 1 
 
 3 
 
 II 
 
 i -S" 
 
 By sediment. 
 
 By water. 
 
 Total. 
 
 
 D«(7. JA. 
 
 Fe«t per 
 tecowL. 
 
 Fat per 
 second. 
 
 Foot-pounds par 
 gecond. 
 
 Fool-pound » per aecorui. 
 
 Foot-pounds per second. 
 
 I 
 
 41 45 
 
 1,133 
 
 4 
 
 268 
 
 361,800 
 
 362,068 
 
 2 
 
 45 48 
 
 1,059 
 
 1 
 
 111 
 
 149,850 
 
 149,961 
 
 3 
 
 49 02 
 
 996 
 
 50 
 
 135,050 
 
 182,317,500 
 
 182,452,650 
 
 4 
 
 55 10 
 
 871 
 
 36 
 
 27,135 
 
 36,632,250 
 
 36,659,385 
 
 5 
 
 60 20 
 
 752 
 
 44 
 
 26,393 
 
 35,630,550 
 
 35,656,943 
 
 6 
 
 16 42 
 
 1,455 
 
 10 
 
 2,788 
 
 3,763,800 
 
 3,766,588 
 
 7 
 
 13 30 
 
 -1,477 
 
 3 
 
 86 
 
 116,100 
 
 116,186 
 
 8 
 
 11 15 
 
 1,490 
 
 4 
 
 148 
 
 199,800 
 
 199,948 
 
 9 
 
 37 50 
 
 1,200 
 
 36 
 
 16,706 
 
 22,553,100 
 
 22,569,806 
 
 lO 
 
 5 40 
 
 1,512 
 
 3 
 
 132 
 
 178,200 
 
 178,332 
 
 1 1 
 
 3 10 
 
 1,517 
 
 4 
 
 211 
 
 284,850 
 
 285,061 
 
 12 
 
 4 30 
 
 1,514 
 
 3 
 
 70 
 
 94,500 
 
 94,570 
 
 J3 
 
 17 00 
 
 1,453 
 
 4 
 
 715 
 
 965,250 
 
 965,965 
 
 H 
 
 21 30 
 
 1,413 
 
 5 
 
 349 
 
 417,150 
 
 417.499 
 
 '5 
 
 10 SO 
 
 1,494 
 
 19 
 
 34,912 
 
 47,131,200 
 
 47,166,112 
 
 i6 
 
 2 58 
 
 1,517 
 
 13 
 
 10,382 
 
 14,015,700 
 
 14,026,082 
 
 17 
 
 2 50 
 
 1,517 
 
 4 
 
 373 
 
 503,520 
 
 503,893 
 
 i8 
 
 1 00 
 
 1,519 
 
 3 
 
 172 
 
 232,200 
 
 232,372 
 
 19 
 
 2 00 
 
 1,518 
 
 23 
 
 77,738 
 
 104,946,200 
 
 105,023,938 
 
 20 
 
 1 25 < 
 
 1,519 
 
 9 
 
 70,993 
 
 95,840,550 
 
 95,911,543 
 
354 BADIANT HEAT. chap. xxii. 
 
 The last column of the preceding tables contains the 
 amount of retardation caused by the waters of rivers flowing 
 towards the equator. The computation of the retarding 
 energy being based on the Aveight of water discharged and 
 the increase of rotary velocity acquired during the transfer 
 from the source to the outlet, no question can be raised as 
 to the existence of the retardation entered in the tables ; but 
 Avhether compensating energies are called forth by the I'eturn- 
 ing vapors before the condensation takes place which results 
 in the precipitation on the river basins, demands careful 
 consideration. Dr. Mayer, in his discourse previously ad- 
 verted to, positively asserts that, agreeably to the demonstra- 
 tion of Laplace, based on abstract mechanical principles, the 
 compensating energy corresponds in every instance with the 
 amount of retardation to which the rotaiy motion of the' 
 globe may be subjected. Admitting this conclusion to be 
 correct, Ave must assume the adequacy of the vapojs Avhich 
 rise within the tropics to restore, during theii" transfer to 
 the temperate and polar regions, the loss of ?'/s viva occa- 
 sioned by the condensed Avater Avhich, in the form of rivers, 
 flows towards the equator. ObA'iously, such restoration of 
 energy could only be effected by friction or pressure of the 
 vapors against projections on the earth's surface directly, t)i' 
 through the interA^ention of particles of the atmospheric air 
 put in motion by the A^apors. The Astronomer-Eoyal of 
 Sweden, in an elaborate demonstration presented to the 
 Eoyal Academy of Sciences at Stockholm, in refutation of 
 my assertion that solar influence is capable of diminishing 
 
CHAP. XXII. THE EAETWH AXIAL liOTATION. 365 
 
 perceptibly the rotary velocity of the earth, thus states the 
 case: "The globe aud its atmosphere constitute a combiuecl 
 system iu luotiou, iu which no part can by any outside cause 
 be disturbed in its relative position without the motion of 
 the entire system being thereby influenced. Consequently, a 
 body of air which, for instance, is carried from the direc- 
 tion of the equator towards either of the poles must, by 
 degrees, positively part with the excess of rotary vis viva 
 which it possessed at the commencement of the motion, 
 compared with the rotary velocity of the region to which 
 it has been transferred, and must impart the entire surplus 
 to the earth undiminished, the rotation of which must con- 
 sequently ha accelerated by this current of air ; and, con- 
 versely, a current of air of contrary direction, or from either 
 of the poles towards the equator, must produce retardation. 
 No dilierence can exist iu this respect between a curi'ent of 
 water aud a carrent of air." The Swedish astronomer, like 
 Laplace, thus puts the \vhole question in a nutshell, assert- 
 ing that air and water, water and air, may circulate in any 
 manner whatever between the equator and the poles, aud 
 between the poles and the equator, without influencing the 
 axial rotation of the globe. It is very true that the earth, 
 with its rivers aud atmosphere, constitute a " combined sys- 
 tem in nidtion"; but we must not lose sight of the import- 
 ant fact that an outside energy — the sun's radiant heat — is 
 being continually exerted, A\liich interferes Avith the motions 
 within that combined system. Accordingly, no argument 
 can prove the correctness of the statement laid before the 
 
356 RADIANT MEAT. 
 
 CHAP. XXII. 
 
 Royal Academy of Sciences at Stockholm short of a posi- 
 tive demoustratiou showing that particles of vapor corre- 
 sponding in weight with the water dischargeil by some 
 river — say the Mississi^jpi — are capable of imparting by 
 friction against the earth's surface, directly or through the 
 agency of the atmosphere, a rotary force exactly balancing 
 the retarding energy which we have established. 
 
 The advocates of the theory of compensation, \vhile ad- 
 mitting that they cannot furnish any 'practical evidence of 
 the truth of their doctrine, assert that the subject is not 
 susceptible of experimental test. It would, indeed, be a 
 fruitless task to undertake the construction of anemometers, 
 or similar instruments, showing that the pressure and friction 
 of the particles of the retui'ning vapor, exerted directly or 
 through atmospheric intervention against the surface of the 
 Mississip25i river basin, from west to east, are capable of com- 
 pensating the established retardation of 19,336,000,000 foot- 
 pounds per second. 
 
 The admitted impossibility of proving by direct measure- 
 ment the existence of compensating force has suggested the 
 resort to some indirect method. I have accordingly con- 
 structed an instrument which practically demonstrates the 
 truth of the following proposition, on which the solution of 
 the problem unquestionably depends : The retarding influ- 
 ence produced by currents of water, confined wdthin channels 
 which convey a given weight in a given time, from the pole 
 to the equator of a rotating sphere, cannot be compensated 
 by ojjposite currents of vapor transferring an equal weight 
 
CHAP. XXII. THE EARTH'S AXIAL liOTATlOX. 357 
 
 in equal time over the surface of tbe f^aid sphere from its 
 equator to its pole. 
 
 The illustration on Plate 34 represents the instrument 
 adverted to; but before entering on a description, it -will be 
 well to define clearly the problem intended to be solved by 
 experimental demonstration. The rotary velocity of the sur- 
 face of the eai-th, for instance, on the 45th parallel is 1,074 
 ft. per second, that of the equator 1,519 ft. per second ; 
 hence the water of a river flowing from lat. 45 deg. to the 
 equator will have its velocity round the axis of the earth 
 increased 1,519 — 1,074 = 445 ft. per second. It needs no 
 demonstration to sbow that the expenditure of energy neces- 
 sary to produce tliis increase of rotaiy velocity will cause 
 the earth to rotate at a diminished rate ; the amount of 
 retarding force being readily ascertained by multiplying the 
 weight of water transferred by the height necessary to gene- 
 rate a velocity of 445 ft. per second, viz., 3,094 ft. Conse- 
 quently, each pound of water transferred from lat. 45 deg. 
 to the equator demands the expenditure of a dynamic enei'gy 
 of 3,094 foot-pounds. The question now presents itself, 
 Avhether a pound of water evaporated on the equator, and 
 returned in the form of vapor to lat. 45 deg., can, during 
 the return movement, impart a rotary energy of 3,094 foot- 
 pounds to the earth. Of coui-se the vapor, on leaving the 
 equator, possesses a rotary velocity of 1,519 ft. per second, 
 while the sui-face of the earth in lat. 45 deg. rotates, as be- 
 fore stated, at a rate of only 1,074 ft. per second. It will 
 be evident, therefore, that during the return movement the 
 
358 RADIANT HEAT. chap. xxir. 
 
 vapor, by contact with, the eai'tli, will have its rotary velocity 
 diminished in the ratio of 1,519 : 1,074. On purely theo- 
 retical considerations, it must be admitted that this contact, 
 1)}' ^vhich the returning pound of vapor has its rotary velo- 
 city diminished 445 ft. per second, will restore to the earth 
 the whole of the energy which was previously expended in 
 augmenting the speed of the pound of water fi'om 1,074 to 
 1,519 ft. per second during its transfer from lat. 45 deg. to 
 the equator. But practice shows that rotary motion cannot 
 be imparted to cylindrical or spherical bodies, however 
 rough their surface may be, by currents of air or steam, with- 
 out great loss of mechanical energy. Conversely, currents 
 of air or steam cannot be produced by the action of similar 
 rotating bodies Avithout a corresponding loss of mechanical 
 energy. 
 
 Practical engineers familiar with these facts fully appre- 
 ciate the difficulty of instituting experiments intended to 
 determine exactly what amount of force is expended in caus- 
 ing rotary motion by currents, and what amount of force is 
 developed hj cui-rents produced by rotating bodies, as sup- 
 posed. The illustrated dynamic register, Plate 34, obviates 
 this difficult comparison between energy expended and de- 
 veloped, by the simple expedient of applying heat and cold 
 in such a manner that the retarding influence of a current 
 of water flowing from the pole to the equator acts simul- 
 taneously with the accelerating influence of an opposite 
 current of vapor, transferring equal Aveight in equal time, 
 from the equator towards the pole. The detail of the in- 
 
CHAP. XXII. THE EAIiTirS AXIAL nOTATWN. 359 
 
 stiumeut will be understood by the following description : 
 Fig. I. slioAvs a section of a hollow sphere 6.25 ins. dia- 
 meter, composed of very thin brass partially filled Avith a 
 light non-conducting substance, made to revolve on its ver- 
 tical axis ; the upper half being covered with a light semi- 
 spherical casing, extending a short distance below the hori- 
 zontal central plane of the sphere. A cylindrical cistern, 
 provided with a flat cover, is attached to the top of the 
 semi-spherical casing. The mode of supporting the lower 
 pivot on which the sphere turns, as also the axle at the top, 
 Avill be seen by reference to the drawing. Rotary motion 
 is imparted to the sphere by a horizontal toothed rack (see 
 top view. Fig. II.) working into the teeth of a wheel at- 
 tached to the vertical axle; the moti^■e power consisting of 
 a weight suspended by a light cord passing over a pulley 
 and secured to the rack. A circular gas-pipe, provided with 
 a series of burners, surrounds the sphere some distance be- 
 low its centre. Referring to Fig. II., it will be seen that 
 the guide-pieces which support the horizontal rack, and 
 through which it slides, act as stops which regulate the ex- 
 tent of the movement. It should be particularly noticed 
 that the arrangement is such that when the motion is 
 checked by the - right-hand stop the last cog of the rack 
 has just slipped out of the cog-wheel, thus allowing the 
 sphere to turn freely. The extent of motion of the rack is 
 O.lJ^G ft., and the weight exactly 2 lbs. It mil be seen, 
 therefore, that the motive force is 0.1 8G X 2 = 0.372 foot- 
 pound, or 2 X 7,000 X 0.18G = 2,604 foot-grains. Deduct- 
 
360 RADIANT HEAT. chap. xxii. 
 
 ing the loss by friction — 64 foot-gi'ains- — tlie effective motive 
 power will be 2,540 foot-grains. The axis of the sphere 
 being exactly vertical, there is obviously no friction what- 
 ever at the tipper bearing after the slipping of the last cog 
 of the rack, while the lower pivot presents a mere point 
 of hardened steel to the step under it; hence the over- 
 coming the atmospheric resistance against the outside of the 
 sphere and the cistern may be considered as the only work 
 to be performed by the stated available motive power of 
 2,540 foot-grains. It only remains to be noticed that when 
 the sphere is to be put in motion the rack is geared into 
 the cog-wheel and brought up against the left-hand stoj), as 
 represented in the drawing, the check-lever (Fig. IV.) being 
 at the same time placed in the position shown by the 
 dotted lines. The moment for starting, indicated by the 
 chronometer, having ariived, the check-lever is brought to 
 the horiiiontal position as quickly as possible, in order to 
 prevent di'agging at the moment of lil^erating the toothed 
 rack. As shown by the illustration, a small quantity of 
 water is confined within the surrounding casing of the 
 sphere, thus forming an aqueous belt round its equator ; 
 the polar cistern being filled with water. 
 
 The object of the instrument having been clearly set 
 forth, it scarcely needs explanation that the device is in- 
 tended to show that when the heat of the gas-flames is 
 applied to the aqueous belt, causing evaporation, while con- 
 densation is effected by the cold water in the polar cistern, 
 the motive energy (2,540 foot-grains) will be incapable of 
 
CUAI'. XXII. THE EAETWS AXIAL BOTATION. 3C1 
 
 turulug the sphere as fast aud as long as when heat and 
 refiigeratiou are not applied. This assumption, it will be 
 perceived, is in direct opposition to the views held l)y the 
 Asti'onomer Royal at Stockholm and other followers of La- 
 place, who contend that "no difference can exist" as regards 
 the effect on the axial rotation of the globe between currents 
 of water and currents of aeriform matter transferring equal 
 Meight in equal time. 
 
 The mode of conducting the experiment with the dynamic 
 register will be readily undei-stood b}' the following explana- 
 tion: The polar cistern is charged \\\i\i boiling water, and 
 the gas-flames applied for a fe^v minutes until the water 
 round the equator is brought near boiling heat. The gas is 
 then shut off, and the toothed rack geared aud afterwards 
 locked by the check-lever. The chronometer being then 
 carefully observed, the check-lever should be quickly pushed 
 down when the hand marks exact time. The motive weight, 
 being thus liberated, puts the sphere in moticm through the 
 inteiTention of the rack and cog-wheel, the time elapsing 
 between the commencement of the movement and the slip- 
 ping of the last tooth of the rack occupying about one 
 second. The observation of the chronometer should con- 
 tinue, in order to ascertain the exact time when the sphei'e 
 is brought to rest. In the meantime, the number of turns 
 must be accurately coiinted. The fii'st experiment being con- 
 cluded, the sphere is again put in motion, as before, without 
 changing the water in the polar cistern or applying the gas- 
 flames, the object of employing heat before starting being 
 
362 EABIANT HEAT. chap. xxii. 
 
 merely that of expaudliig the sjahere to proper dimensions. 
 The experiment having been repeated six times, the 'Dieuji 
 of time occupied and number of turns performed, resulting 
 from the expended energy of 2,540 foot-graius, should be 
 determined ^vith the utmost precision. The procedure will 
 then be changed : the polar cistern ^vill be charged with 
 cold water, and the gas-ilames applied and kept burning. 
 The sphere, under these altered conditions, is again started, 
 but not until boiling temperature in the equatorial belt has 
 been attained and evaporation commenced. The experiment, 
 as before, will be repeated six times, and the mean of time 
 and the number of turns ascertained. 
 
 The law of compensation relating to solar influence on 
 the axial rotation of the earth, expounded by Dr. Mayer, 
 is evidently strictly applicable to the dynainic register, since 
 the equatorial belt of the rotating sphere is being continu- 
 ally heated, while the polar region is being exposed to con- 
 tinuous refrigeration, vapor being thus foi'med at the equator, 
 and currents produced which condense on reaching the cold 
 semi-sj^herieal covering over the pole. The water thus 
 formed, divided into small sti'eams, flows back on the sur- 
 face of the sphere to the equator, where it is again converted 
 into vapor; hence a continued circiilation of opposite cur- 
 rents of vapor and water wall be kept up. It sliould be 
 particularly observed that the vapor in its passage toAvards 
 the pole not only acts against the sxirface of the sphere, 
 but also against the inside of the semi-spherical covering, 
 thereby affording a double chance of imparting motion to 
 
CHAP. XXI f. THE EAETWS AXIAL ROTATION. 363 
 
 the rotating mass. But this notwithstanding, the exjieri- 
 meuts have .shown tliat the retanlimj energy of the con- 
 densed water flowing in small streams from the jwle to the 
 equator on tlie surface of the sphere, greatly exceeds the 
 accelerating energy imparted by the excess of rotary velocity 
 of the vapor in its course towards the pole, and the conse- 
 quent friction of its particles against the sui-faces of the 
 sphere and the semi-spherical casing. Agreeably to Dr. 
 Mayer's conclusion.s, founded on the theory of Laplace, the 
 opposite currents which result from high temperature on 
 the equator, and the refrigeration over the temperate zone 
 and the poles, cannot affect the axial rotation of the globe. 
 " The effect of every single motion by these means on the 
 rotation of the globe," he saj-s, "is exactly compensated by 
 the effect of another motion in an opposite direction." Nor 
 can the Swedish Astronomer, as we have seen, perceive any 
 difference between currents of water and currents of ai-ri- 
 form matter. In direct opposition to the conclusions of these 
 physicists, our expeiiments prove that, although the weight 
 transferred fi-om the pole to the equator of the sphere of 
 the dynamic register is precisely the same as the weight 
 which is transferred in the oi)posite direction, the contact 
 and friction of the particles of vapor against the surfaces of 
 the convex and concave spheres is incapable of restoring the 
 loss of vis viva consequent on imparting rotary motion to 
 the particles of water transferred fi"om the pole to the 
 equator. 
 
 Too much space would be occupied by a detailed ac- 
 
364- BABIANT HEAT. CHAP. xxii. 
 
 count of the experiments whicli Lave been made with the 
 dynamic register; hence only the most important facts bear- 
 ing directly on the question will be presented. The number 
 of turns of the rotating sphere produced by the nioti\-e 
 force of 2,540 foot-grains has been 660.5, ocoup3ing 10 inin. 
 37 sec, the barometer at the time indicating 29.8, the tem- 
 perature of the surrounding atmosphere being 62° F. The 
 mean of the force expended for each turn will therefore 
 
 2,540 
 amount to — -7— = 3.84 foot-grains. It will be asked, in 
 060.5 
 
 view of this insignificant motive power, chiefly expended 
 in overcoming the atmospheric resistance against the rotat- 
 ing sphere and cistern, how the excess of retarding energy 
 of the condensed water flowing over the surface of the 
 sphere, from the pole to the equator, can possibly be mea- 
 sured. The answer is that we need not consider the amount 
 of energy developed by the motive Aveight ; we merely count 
 the number of turns and note the time required to bring 
 the sphere to a state of rest from the moment of stai'ting, 
 the gas-flames being kept burning and the refrigerating 
 medium retained in the polar cistern during the observations. 
 Then, removing the cooling medium and replacing the same 
 with boiling water, we again put the sphere iu motion, count 
 the number of turns, and note the time. The result of this 
 change of procedure will be, as shown l^y our experiments, 
 that the sphere will run much longer and perform a greater 
 number of turns — a startling fact, since the motive energy 
 of 2,540 foot-grains has not been increased. To practical 
 
CHAP. SXii. TEE ICAHTn'S AXIAL liOTATION. 366 
 
 minds the explanation will at once suggest itself, that 
 because there is an expenditure of lient while condensation is 
 kept up, which ceases when the refrigerating medium is 
 withdrawn, some additional work is being performed Avhile 
 the cold medium I'emains at the pole. Now, what is the 
 nature of this Avork ? Evidently the condensed water, while 
 flowing from the pole to the equator, has its rotary speed 
 successively increased corresptmding with that of the sur- 
 face of the sphere ; hence work must be performed ^vhi]e 
 refrigeration is kept up at the pole. Satisfactory as this 
 explanation appeai-s, it is met by the cardinal objection 
 that, since force cannot be anniliilated, the o^iposite current 
 of vapor, which simultaneously transfers an equal weight 
 from the equator to the pole of the rotating s^^here, must, 
 by friction or contact of some kind, positively return the 
 whole of the mechanical energy expended in augmenting 
 the rotary velocity of the particles of water moving in a 
 contrary direction. This n(>twithstanding, we must accept 
 the fad proved by the dynamic I'egister, that a certain 
 amount of mechanical energy disappears when the rotating 
 sphere is subjected to the action of differential temperatures. 
 There was a time when we could not account foi- such dis- 
 appearance of energy, but — thanks to the labors of Joule 
 and Mayer — the mechanical theoiy of heat has thrown light 
 on the subject. The theoreti^'al deductions of Laplace have 
 lost their potency. We no longer confine ourselves to the 
 balance anil rule in measui'ing the result of exi^eiidcd force. 
 Joule and Mayer have taught us to consult also the titer- 
 
366 BADIANT HEAT. CHAP. xxiT. 
 
 mometer during our iuvestigatious. Bearing in mind, then, 
 ■what the new theoiy of heat teaches, the disappearance of 
 mechanical energy during the experiments with the dynamic 
 register ceases to l)e a puzzle. Close investigation shows 
 that the heat resulting from the arrested motion of the cir- 
 culating vapor, which, on leaving the aqueous belt, possesses 
 a rotary velocity equal with that of the circumference of 
 the sphere, represents A-ery nearly an equivalent of the 
 observed loss of energy, the diiference being made up by 
 heat generated by the particles of the circulating vajjor as 
 they successively impinge against the minute projections of 
 the surface of the convex and concave spheres. Obviously, 
 the heat thus generated is carried off l)y the cold semi- 
 spherical casing surrounding the sphere of the dynamic 
 register, precisely as heat produced by analogous motions 
 Avithiu the terrestrial atmosphere is carried off by radiation 
 into space. In either case the heat lost is an equivalent of 
 the mechanical energy abstracted from the rotating sphere. 
 
 Illustrations and descriptions have been prepared explana- 
 tory of important modifications of the dynamic register deline- 
 ated on PI. 34, adojited in order to control the irregular 
 resistance of the atmospheric air against the rotating sphere, 
 unavoidalde in employing gas-flames for heating the equa- 
 torial belt; but the subject having already occupied too 
 much space, I now propose to state only the result of the 
 experiments which have been made with the modified instru- 
 ment, the dimensions of which, it should be observed, have 
 been considerably increased, the motive power, however, 
 
CUAI". xxii. THE EAlilWH AXIAL liOTATION. 367 
 
 remaining uucliangod. It is iscarcely necessary to remark 
 that a complete tlenumstratiou and record of an investiga- 
 tion of this complicated nature would present an array of 
 figures inadmissible in this work. The diagram on PI. oS 
 has, therefore, been devised to dispense with tigures ; the 
 relations of time, velocity, and resistance being jjiesented in 
 sueli a manner that, among other facts, the aniduul of mechani- 
 cal enei'gy which disappears duiing the experiment ma}' be 
 ascertained by mere inspection. For the [)urpose of saving 
 space and facilitating direct comparison, this diagram has, 
 moreover, been so arranged that the iccord of the experi- 
 ments in which heat and refrigeration have been employed 
 is placed on the same base-line with the record of the 
 experiments in which difference of temperature was pre- 
 sented. The divisions on the l)ase-liue a h mark the time 
 of rotation, the large spaces indicating minutes and the 
 smaller divisions 10 sec. each. The length of the ordinates 
 of the curve c b resting on the base-line represents the 
 number of turns performed in a given time when the rotat- 
 ing sphere is not subjected to the action of heat and 
 refrigeration; while the length of the ordinates of the curve 
 d e represents the number of turns when heat and cold are 
 being applied. It will be readily perceived that, for 
 instance, the ordinate between 1 and the curve c b repre- 
 sents the number of turns per minute at the commencement 
 of the second minute, while the ordinate 2 represents the 
 number of turns per minute at the commencement of the 
 third minute, and so on for all the other ordinates. 
 
368 RABIAXT HEAT. chap. xxil. 
 
 The 'permanent frictiou of the instrument — i.e., the friction 
 of tlie pivot on -which the sphere turns — being practically 
 inappreciable, it will be evident that the resistance ojjposing 
 the rotation will vary in the ratio of the square of the 
 velocities. Hence, as the respective ordiuates between the 
 cui'ves c 1) and d e and the base-line represent the velocities, 
 it will only be necessary to square these ordinates in order 
 to determine the exact amount of resistance to the periods 
 indicated bj^ the divisions on the base. Accordingly, the 
 ordinates mentioned have been prolonged in the ratio of their 
 squares, the curves / Z> and g e being the result of this pro- 
 longation. Obviously, the lengths of the ordinates of these 
 curves resting on the line a h represent accurately the amount 
 of resistance opposed to the rotation of the sphere at the 
 times indicated by their intersection with that line. The 
 rate of velocity — i.e., the mimber of turns per minute per- 
 formed by the sphere at the commencement and at the 
 termination of each minute — will be found by referring 
 to the figures marked on the vertical lines / a and I h. 
 Thus, for instance, the rate of A^elocity at the termination of 
 the second ininute is 75.4 turns Avhen refrigeration is vot 
 applied, while tlie rate is 68.0 when the cooling medium is 
 applied at the pole. As might be expected from the irregu- 
 lar nature of the external resistance opposed to the rotating 
 mass, the curves / b and g e do not correspond with any of 
 the conic sections. The available motive power of 2,540 
 foot-grains expended during the experiment is represented by 
 the supei-ficies / a h, the energy developed being represented 
 
CHAP. XXII. TBE EAliTWS AXIAL ROTATION. 369 
 
 by the superficies g a e. Assuuiing the former to be 1.000, 
 tlie latter, as shown by our tliagraui, will be 0.763, differ- 
 ence = 0.237; hence the amount of lost energy is 0.237 X 
 2,540 = 601.98 foot-grains. Now, if the weight of ^\•ater 
 which is condensed at the pole and returned to the equator, 
 inulti])lied by the height necessaiy to generate the rotary 
 velocity acquired during the transit, should amount to 601.98 
 foot-grains, the fact will be established that the current of 
 vapor has not, during its passage from the equator to the 
 pole, restored any of the energy abstracted from the sphere 
 by the current of water flowing in the contraiy direction. 
 The quantity of water condensed and returned to the equa- 
 torial belt being readily ascertained by observing the incre- 
 ment of temperature of the contents of the polar cistern, it 
 is easy to sho\v that the energy abstracted from the rotating 
 mass by the water thus transferred from the pole to the 
 equator corresponds so nearly with the differential mechanical 
 energy represented by the superficies f g e b, that the com- 
 pensation resulting from the tangential force exerted by the 
 particles of the currents of vapor against the sui-face of the 
 sphere of the dynamic register is inappreciable; precisely as 
 we find that the compensating tangential force of the cur- 
 rents of vapor Avliich sweep over the basin of the Mississippi 
 from west to east (neuti-alized by the currents which pass 
 from east to Avest) is an inappreciable fraction of the retard- 
 ing energy of 19,836,000,000 foot-pounds per second, exerted 
 by the water which the Mississippi canies in the direction 
 of the equator. 
 
370 BABIANT HEAT. chap. xxil. 
 
 Having thus anal3^zed the opposing energies called forth by 
 the waters flowing toAvards the equator, and of the I'eturning 
 vapors, the condensation of which replenishes the river basins, 
 we may now enter on a computation of the aggregate amount 
 of the retarding energy, and the consequent diminution of 
 the rotary velocity, of the earth, caused by the rivers enume- 
 rated in the preceding table. The total of the retarding force 
 entered in the column next the last, it will be found, amounts 
 to 53,857,788,300 foot-pounds per sec, which sum, multiplied 
 by 86,400 sec, shows that the earth has to overcome a 
 resistance of 4,653,313 X 10° foot-pounds during each revolu- 
 tion. Multiplying this resistance by 36,524 days, we ascertain 
 that the retarding energy of the water transferred in the 
 direction of the equator by the entire Southern river systems 
 of both hemispheres amounts to 16,995,760,069 X 10" foot- 
 pounds in a century. Now, in order to determine the dimi- 
 nution of rotary velocity consequent on this counteracting 
 energy, it will be indispensable to compute the earth's rotary 
 vis viva. The elements necessary in this computation are : 
 volume, time of revolution, specific gravity, and the position 
 of the centre of gyration of the rotating mass. The two 
 first-named elements are known mth desirable accuracy ; the 
 third element, specific gravity, has been ascertained Avith 
 tolerable accuracy; but the position of the centre of gyration, 
 which depends on the internal temperature of the globe and 
 the disposition of its constituent parts, has not yet been 
 determined. Physicists assume that the density of the globe 
 increases towards the centre in arithmetical progression ; but 
 
CHAP. XXII. THE EAirnra aaial uotation. 371 
 
 this assuraption is not sustained by sound reasoning. Our 
 space not admitting of discussing this complicated question 
 at length, let us merely consider the leading fact, that, at a 
 distance of only -h of the earth's radius = 1,0-14,400 ft. from 
 the surface, the weight of a superincumbent mass of fused 
 granite will exceed 900,000 lbs. to the sq. in. = 60,000 
 atmospheres. Under this pressure the weight of air will be 
 70 times that of water, and 3.5 times that of the heaviest 
 metals. Gold, at the point of fusion, is 7 times heavier than 
 fused granite, while neither of these solids loses more than 
 Toir of specific gravity at melting heat — a fact which proves 
 conclusively that high temperature of metals and minerals is 
 not incompatible with great density. Hence fused granite, 
 in the earth's interior, may be many times heavier than the 
 cold mineral at the surface. Unless, therefore, we are pre- 
 pared to dispute the assumption that fused granite, under a 
 pressure of 900,000 lbs. to the sq. in., will have its specific 
 gravity doubled — involving a density less than one-third of 
 fused gold not subjected to compression — we must admit 
 that the specific gravity of the earth at the depth of A of 
 the radius is so great that, if the density, as physicists have 
 assumed, increases in arithmetical progression towards the 
 centre, our planet Avould be many times heavier than it is. 
 We are compelled, therefore, to reject the accepted theory, 
 more especially as the stated enormous pressure consequent 
 on superincumbent weight takes placer at only ^ of the 
 earth's radius below the surface. 
 
 In accordance with the foregoing reasoning, our compu- 
 
373 RADIANT HEAT. chap. xxii. 
 
 tation of the earth's rotary vis viva will be based ou the 
 assumption that the mass is homogeneous. It is true that 
 the specific gravity at the surface is somewhat less than 
 one-half that of the entii'e mass ; but we have sho^vn that 
 at a depth of iv of the radius from the surface the density 
 is so great that if it continued to augment in arithmetical 
 progression, the specific gravity of the globe would far exceed 
 that which has been determined by careful investigation. 
 Nor should we lose sight of the important fact that the 
 temperature corresponding with the compression produced 
 by the superincumbent weight is so great that the compo- 
 nent parts of the centi'al mass may be as light as pumice, 
 notwithstanding the enormous external pressure. Conse- 
 quently, it may be satisfactorily demonstrated that the 
 earth's circle of gyration extends considerably beyond, in 
 place of being within, that of a Jiomogeneous sphere, agree- 
 ably to the accepted theory of augmented density towards 
 the centre. In our computations, however, we will assume 
 that the circle of gyration is that corresponding -with homo- 
 geneity, which, in accordance with the property of spheres, 
 is 0.6325 of the great circle. Sir John Herschel's determi- 
 nation shows that the mean diameter of the earth, considered 
 as a perfect sphere, is 7,912.41 statute miles, or 41,777,524 ft. ; 
 hence, if we assume the specific gravity to be 5.5, we can 
 readily calculate that the weight is 1,308,608 X 10" lbs. 
 Multiplying the equatorial velocity — 1,519.07 ft. per second 
 — by 0.6325, we ascertain that the mean rotary velocity of 
 the entire mass of the earth is 960.81 ft. per second — a rate 
 
CHAP. XXII. TEE EABTE'S AXIAL rxOTATION. 373 
 
 acquired by a fall of 14,424 ft. The earth's rotary vis viva 
 will accordingly amount to 14,424 X 1,308,008 X 10" = 
 18,875,361 X 10" foot-pounds. The mind being utterly in- 
 capable of conceiving this stupendous energy without com- 
 parison with mechanical energies of less magnitude, let us 
 ascertain to what extent it will be diminished by the 
 retardation exhibited in the tables pre^^ously presented — 
 namely, 16,995,760,069 X 10'° foot-pounds, exerted in the 
 course of a century by the southern river systems of both 
 hemispheres. Dividing the stated retarding energy in the 
 
 , , . . , 18,875,361 X 10" ^ , , 
 
 earths v^s viva, thus: i^. 995^760^069 X lO"' ''^ ^""'^ '^^"*' 
 
 notwithstanding the enormous amount of retardation exerted 
 in a centuiy, only 1 1 1 e' u 00 of the rotary energy of the 
 earth will be destroyed in that time. And if we multiply 
 the fraction thus presented by 10,000, we learn that at the 
 end of 1,000,000 years the rotary energy of the earth \\\[\ 
 be only i\hiis less than at present ! By no other compari- 
 son, probably, than the one we have instituted could we 
 clearly comprehend the magnitude of 18,875,361 X 10" foot- 
 pounds of mechanical energy. 
 
 Let us now calculate the effect of the tabulated resist- 
 ance on the earth's rotary velocity with reference to time. 
 The retardation observed by astronomers being, as before 
 stated, about 12 sec. in a century, our object will be to 
 ascertain how far this retardation may be attributed to the 
 counteracting energy under consideration. Multiplj'ing, then, 
 the number of seconds in a century, 3,155,673,600, by the 
 
374 EABIANT KEAT. CHAP. xxii. 
 
 retarding energy of 53,857,T80,300 foot-pounds per second, 
 entered in the table, we establish the fact before adverted 
 to, that the total retardation is 16,995,760,069 X 10'° foot- 
 ponnds in one centniy. Dividing this retardation in the vis 
 viva, it Avill be seen that the earth loses mus ^ a^a^a of its 
 rotary energy in the course of 100 years ; but in calculating 
 the time corresponding with this loss, we have to consider 
 that the velocities are as the square root of the forces, and 
 that consequently the rotary velocity will not be reduced as 
 rapidly as the rotary energy. Evidently, if the diminution 
 of energy and velocity corresponded exactly, the retardation 
 of the earth's rotary motion during one century would be 
 
 3,155,673,600 ^„,^ , -^ ^ . , ui ^i i 
 
 — = 2.8414 sec. But, m accordance with the laws 
 
 1,110,592,343 
 
 of motion referred to, the diminution of velocity during the 
 century will be in the ratio of the square roots of the earth's 
 vis viva at the beginning and at the termination of that 
 period. Now, this ratio being readily computed, as we know 
 the amount of energy lost in one centuiy, while the time in 
 seconds is also known, we are enabled to show, by an easy 
 calculation, that the earth suffers a retardation of 1.42071 
 sec. Adding the retardation occasioned by the tabulated 
 sedimentary matter = 0.00105 sec, ascertained in the manner 
 explained, the total retardation of the earth's rotary velocity 
 in a century, at the presevt epoch, will be 1.42176 sec. The 
 vastness of the rotary vis viva of the earth having already 
 been discussed, it will not be necessary to offer any expla- 
 nations with reference to the insignificance of the stated 
 
CUAP. XXII. THE EARTWkS AXIAL liOTATION. 375 
 
 retardation in comparison witli the uiagnitutle of the counter- 
 acting energy exerted by the water and sediment of the entire 
 river system presented in our tables. 
 
 "We have now to consider the influence on the eailh's 
 rotary energy exercised by rivers, the course of which is in 
 the direction of the poles. Evidently river water running 
 from the equator Avill have its motion round the axis of 
 rotation continually diminished as it reaches the noi-thcrn 
 parallels ; hence rotary energy will be imparted to the earth 
 by all rivers flowing towards the poles. At first sight, it 
 will be imagined that the energy thus imparted will neutra- 
 lize the retarding force exerted by the waters transferred 
 towards the equator. Certain physical causes, however, pie- 
 vent the imparted energy from restoring any of the earth's 
 lost vis viva. The subject will be most readily conq)rehend- 
 ed by an examination of the nature of the neutralizing force 
 exerted by the following great rivers, namely, the Lena, 
 Yenesei, Obi, and Mackenzie, which furnish the princij)al 
 amount of water discharged into the Arctic Ocean. These 
 rivers drain an area of 3,840,000 sq. miles, the latitude of 
 the centre of their basins and their outlets being very 
 nearly in the same parallel. The mean of the former is 59 
 deg. 30 min., that of the latter G9 deg. 56 min. Accordingly, 
 the mean circumferential velocity of outlet is 421.18 ft. per 
 second, while that of the centre of basin is 770.95 ft. per 
 second. It will be seen, therefore, that a diminution of rotary 
 velocity of 770.95 - 521.18 = 249.77, say 250 ft. per second, 
 takes place during the transfer of the water from the centre 
 
376 BADIANT HEAT. chap. xxii. 
 
 of tlie basins of these rivei'S to tlieir outlets. Now, a velocity 
 of 250 ft. per second is produced by a fall of 976.5 ft, lience 
 each pound of M'ater discharged into the Arctic Ocean by 
 the before-named rivers •will impaii; a mechanical energy of 
 976.5 foot-pounds. Apart from this powerful neutralizing 
 force of a given weight, the quantity of water ti-ansferred is 
 so great, owing to the vast extent of the basins, that, uotAvith- 
 standing the moderate precipitation in high latitudes, the 
 rotary energy imparted to the earth ^^•ill balance the retarda- 
 tion of the 136 rivers entered in our tables. It scarcely 
 requires explanation that the stated enormous force exerted 
 by the water transferred by the great northern rivers is owing 
 to the rapid diminution of rotary velocity in approaching the 
 pole ; a single degree of latitude at the point where, for 
 instance, the river Lena dischai'ges into the Arctic Sea having 
 a greater fall than te7i degrees have within the tropics. It 
 would be waste of time, however, to compute the exact 
 amount of energy imj)arted to the earth by the Arctic livers, 
 as will be seen by the following examination of the subject. 
 Unquestionably, if the supposed pound of water, on entering 
 the Arctic Ocean, at once evaporates and ascends into the 
 atmosphere, we must admit that an impulse of 976.5 foot- 
 pounds has been imparted to the earth by its transfer from 
 the centre of the river basin; but if it should be found that, 
 in place of evaporating on entering the cold polar sea, the 
 pound of water commences a retrograde motion towards the 
 equator through Behring's Straits or through the wide channel 
 bet\\'een Norway and Greenland; and if we should find, also, 
 
CHAP. XXII. THE EAllTWS AXIAL ROTATION. 377 
 
 that wLeu it crosses the 59 deg. 30 rain, parallel (the same 
 as that of the centre of the river basin) it has not yet been 
 converted into vapor, we must then admit that the whole 
 of the energy imparted to the earth by the ajyjyroacJi towards 
 the axis of rotation, during the original transfer to the polar 
 sea, has been completely neutralized by the retardation con- 
 sequent on the retreat from the axis of rotation during the 
 southerly course to the last-mentioned latitude. Following 
 our pound of water during the continuation of the motion 
 towards the equator, we may discover that it has not changed 
 its form into vapor, even when reaching latitude 47 deg. 
 45 min., at which point the circumferential velocity is exactly 
 250 ft. per second greater than that of the centre of the 
 basin from whence the motion proceeded. In that case, not 
 only has the imparted energy been neutralized, but a retarda- 
 tion of 976.5 foot-pounds has been called forth by the pound 
 of water, the course of which may possibly continue until 
 it mixes wdth the warm water -within the tropics. Let us 
 guard against confounding the movement of the water dis- 
 charged into the Arctic Sea by the northern rivei-s with the 
 currents produced by the combined influence of lunar attrac- 
 tion, wnds, differential oceanic temperature, and solar attrac- 
 tion. It has long been recognized that the water poured 
 into the Arctic Sea by the great Asiatic rivers is the result 
 of condensation of vapors raised by the sun within or near 
 the tropics. A corresponding amount of water must, there- 
 fore, be returned from the polar sea, or its surface would be 
 elevated, and that of the tropical seas suffer a proportionate 
 
378 BABIANT HEAT. chap. xxii. 
 
 depression. The reader cannot fail to perceive the important 
 bearing of these facts on the question of retardation of the 
 earth's rotary velocity. 
 
 The result of the experiments with the dynamic register 
 proves that the rotary motion possessed by the \'apors on 
 leaving the equatorial seas may be almost entirely destroyed 
 by being converted into heat during their course towards the 
 basins of the northern rivers ; hence imparting no perceptible 
 tangential force to the earth. Accordingly, the return to the 
 tropical seas of the water which is continually being dis- 
 charged by the northern rivers into the polar seas will, on 
 account of the increased velocity round the axis of rotation 
 imparted during the southern course, subject the earth to an 
 amount of retardation far exceeding that produced by rivers 
 flowing towards the equator. It may be asked, under these 
 circumstances, why the latter rivers have been tabulated and 
 their inferior retarding energy calculated. The rivers flow- 
 ing in the direction of the poles have been examined, tabu- 
 lated, and their counteracting energy calculated ; but the 
 question of attendant retardation of rotary velocity cannot 
 properly be entertained until certain other counteracting in- 
 fluences shall have been examined. The publication of the 
 table containing the southern rivers has been deemed neces- 
 sary as a 2Mini (Vappui facilitating demonstrations intended 
 to establish the fact that, independently of the counteracting 
 force of the tidal wave (hitherto greatly overestimated), the 
 retarding energy called forth by the evaporation within the 
 tropics, and the consequent condensation and precipitation 
 
CHAP. XXII. THE EAUTira AXIAL ROTATION. 379 
 
 in the temperate zones, fully account for the retardation of 
 the earth's rotary velocity — 12 seconds in a century — inferred 
 fi'om the apparent acceleration of the moon's mean motion. 
 
 The fact being well knoA\Ti, through European and Ame- 
 rican publications, that the results attained by the employ- 
 ment of the instruments described in the foregoing chapters 
 rank among the most important scientific achievements of 
 the latter part of the first century of the American Republic, 
 I expected that the Centennial Commissioners would invite 
 me to display those instruments during the Exhibition, and 
 consequently caused tlie same to be repolished and arranged 
 for the great occasion. It is proper to state that, although 
 not wanted by the Commissioners, the time spent in prepar- 
 ing the collection lor exliiljition lias not been wasted, since 
 it is my intention to present the same to the Smithsonian 
 Institution, after the completion of certain investigations. 
 
CHAPTER XXIII. 
 
 DISTANCE INSTRUMENT,* 
 
 FOK MEASURING DISTANCES AT SEA. (SEE PLATE 36.) 
 
 This instrument is principally intended for the use of the 
 naval officer iu measuring the distance of an enemy's ship, 
 to enable him to elevate his guns with precision. Modern 
 naval tactics being principally based on distant firing, an 
 accurate knowledge of the object to be aimed at becomes 
 indis^jensable. Any device for obtaining it based on any 
 process of calculation is evidently out of the question, con- 
 sidering that a single minute will bring two approaching 
 vessels, moving at a rate of ten knots, full a quarter of a 
 mile nearer each other. In firing beyond point-blank range, 
 therefore, seconds are precious in determining the elevation 
 of the guns. Accordingly, nothing will ausAver short of an 
 instrument which, by a single observation and the reading 
 off at sight, tells the distance. The instrument under con- 
 sideration meets these somewhat severe conditions perfectly, 
 
 • This instniment formed part of the original equipment of the steamship Princeton. 
 
CHAP. xxm. ' 
 
 DISTANCE INSTRUMENT. 381 
 
 as will be seen by tlic following explanation. An observer 
 stationed on the maintop or cross-tree of a ^hip, k)oking at 
 ji vessel— say a mile off— will perceive that liis line of vision, 
 directed to the horizon, passes over the point marked by the 
 water-line of the hull, and he will also perceive that, as the 
 vessel approaches, that point appears to sink lower and 
 lower below the line directed to the horizon; in other words, 
 the angle formed by that line and the line directed to the 
 water-line of the approaching vessel continually increases. 
 On the other hand, if the vessel recedes, it \\\\\ be found 
 gradually to diminish. Now, the observer's eye (see the 
 plate before referred to) being placed at a definite height 
 above the level of the sea, and his line of vision directed to 
 the horizon being the tangent of the earth's curvature pass- 
 ing through the definite point a, it follows that the angles 
 h a c and (/ a c, respectively, will determine the distance of 
 h and g from h, when vertical to a. Now, the earth's cui-va- 
 ture is constant, the height of a above the level of the sea 
 is knoAvn, and the angles h a c, etc., may be readily mea- 
 sured b}- reflectors and the graduated arc. But there is no 
 time for mathematical computation. 
 
 The Distance Instrument, it will be seen by the following 
 description, performs the required computation with unemng 
 certainty whilst the observer measures the angle, and it 
 exhibits the result the instant he has pei-formed his part. 
 Every one familiar with nautical instniments will, by an 
 attentive inspection of the drawing, readily understand its 
 principle and operation; a very brief description will there- 
 
382 DISTANCE INSTRUMENT. chap, xxiii. 
 
 fore suffice. A is an ordinary reiiector, as used in quadrants ; 
 B, the object-glass, and C, tlie sight by which the angles // a e, 
 etc., are measured ; D, a spindle, to the end of which the 
 reflector is firmly attached ; E, lever for turning said spindle ; 
 F, sliding-nut made to move freely up and down in a slot 
 at the lower end of lever E ; G, thumbscrew worldng in the 
 sliding-nut ; H, pinion on said screw, working into cogs cut 
 in the circumference of a graduated index-plate J ; K, socket 
 sliding on the main stem of the instrument and supporting 
 the frame and centre of the revolving index-plate. Before 
 noticing the operation of the instniment, it will be necessary 
 to point oxit the manner of graduating the index-plate. 
 Considering the extent of the point-blank range of naval 
 ordnance, it is evident that no distance under 400 yards need 
 be measured : supposing, therefore, that Ti is 400 yards dis- 
 tant fi'om h, it will be seen that the operation or I'ange of 
 the instrument will be limited to the angle A a c, which thus 
 determines the extent of movement or vibration of the lever 
 E. This movement again determines the pitch of the thumb- 
 screw and the relative diameter of the pinion, it being evident 
 that the extreme vibration of lever E should not produce 
 more than one revolution of the index-j)late. Any con- 
 venient-sized index-plate being selected, a scale graduated 
 into feet or yards is then constructed, corresponding to the 
 circumference of the plate ; the mode of dividing the scale 
 being as follows: In the first place, a base-line of 100 feet 
 (represented hj a b in the diagram) is supposed, and the 
 tangent a c determined accordingly. The known curve b d c 
 
CHAP. XXIII. DISTANCE IXSTFUilENT. 383 
 
 is tlieu divided into spaces h (/, </f,fe, etc., of 10 yards eacli, 
 comraenciug at a point, I>, 400 yards from b. The sines of 
 the angles 7i a c, g a c, etc., are next calculated and marked 
 on the before-mentioned scale, and from thence ultimately 
 transferred to the curved scale on the index-plate. The fol- 
 lowing directions for using the instrument are deemed sufK- 
 cient : Turn the thumbscrew until the line maiked '' horizon " 
 is placed directly under the fixed index. Then adjust the 
 oljject-glass by means of the set screw M, so that the real 
 and reflected horizons come in a line. This adjustment being 
 made, the instrument is ready for use, and need not be 
 readjusted unless disturbed. The process of measuring the 
 distance consists simply in turning the thumbscrew G until 
 the reflected water-line of the object observed is brought in 
 a line watli the real horizon seen through the object-glass. 
 The point on the scale of the index-plate placed directly 
 under the fixed index shows the distance desired. It must 
 be conceded, on theoretical considerations, that if the base- 
 line be previously known and the instniment made to cor- 
 respond thereto, the measurement cannot fail to be accurate ; 
 but such is the nature of this base-line that it cannot be 
 previously known ; accordingly, the base-scale L has been 
 introduced, by which the instrument may at all times be 
 made to conform to the variable height of the base. It 
 is evident that an increase of altitude would render the 
 scale of the index-plate too short, and, on the other hand, 
 too long if the altitude be diminished. It is also evident 
 that by sliding the index-plate up, the effect of which 
 
384 DISTANCE TNSTBUMENT. chap, xxiii. 
 
 will be to shorten the lever E, any diminution of the 
 base may be compensated for, and the index-plate remain 
 very nearly correct. The sliding the index down would in 
 like manner compensate for any increase of the base. On 
 mathematical considerations, it is obvious, however, that this 
 mode of compensating for variations of the base cannot be 
 carried very far. Index-plates of different graduations will, 
 therefore, be employed to suit the height of the masts of 
 different classes of vessels, and the base-scale only resorted 
 to for compensation to meet irregularities occasioned by 
 altered draught of water, consequent on diminution of ammu- 
 nition, stores, etc. At first sight, it would appear that the 
 base employed in this instrument is not sufficiently definite 
 or accurate ; on due consideration, however, it will be found 
 to be fully as definite as required. 
 
 In the first place, the height of the maintop, cross-tree, or 
 other point of a shiji above the bottom of the keel may be 
 ascertained to an inch, and, when once known, may be re- 
 corded, as well as tonnage, length, beam, etc. Secondly, the 
 draught of water amidships is ahvays known to a careful 
 commander, within two inches or less. Tlie draught of water, 
 being deducted from the height above the keel, establishes 
 the altitude above the water-line. The height of the obser- 
 ver's eye — ordinarily five feet six inches — being next added, 
 determines the base Avithin an inch or two. So far, then, 
 the accuracy is all that can be desired for practical pur- 
 poses. The effect of the rolling of the ship, Avhich at sea 
 always takes place to some extent, next demands attention. 
 
cuAi'. xxiii. uiiiTAMi: lysxnuMEyr. 385 
 
 It would be au extreme ciise to suppose the observer tossed 
 through uu arc of 20 feet whikt taking au observation — viz., 
 10 feet on each side of the vertical line. On cakuhition, it 
 will be found that sucdi oscillati(tu would t>nly produce a 
 depression of six inches at the loiccd point. Finally, the 
 rising and falling of the sliip deserves to be noticed. The 
 vertical movement of the inidaliip hoilij, being at all times 
 surprisingly small, \vill be found quite unimportant at times 
 when the Distance Instrument is likely to be wanted. Again, 
 as each observation only i-equires a feAV sect)nds, it may be 
 frc(piently repeated. It is proper to add that an error <»f 
 6 inches in a base of 100 feet, and which will not ordi- 
 narily occur, only causes an error of distance of nine yards 
 in a mile. 
 
CHAPTER XXIY. 
 
 THE STEAM FIEE-ENGINE. 
 (SEE PLATE 37.) 
 
 The Mechanics' Institute of New York offered its great 
 gold medal, in January, 1840, as a prize for the best plan 
 of a steam fire-engine. Having several years previously de- 
 signed such machines in England, among which may be 
 mentioned the steam fire-engine employed during the mem- 
 orable fire at the Argyle Rooms in London, in 1830 (the 
 first time fii-e had ever been extinguished by the mechanical 
 power called forth by fire), I had no difficulty in producing 
 plans complying with the conditions of the Mechanics' In- 
 stitute in a manner warranting the award of the prize 
 offered. 
 
 The following description — reference being had to the 
 illustration on PI. 37 — shows the detail of the steam fire- 
 engine thus accepted by the Mechanics' Institute of New 
 York. A, double-acting force-pump, composed of gun-metal, 
 
CHAP. XXIV. THE STEAM FIHE-EXGINE. 387 
 
 firmly secured to the carriage-frame by four strong brackets 
 cast on its sides ; a a, suction-valves ; a' a', suction-passages 
 leading to tlie cylinder; a", clianibur containing the suction- 
 valves, to which chamber are connected suction-pipes a'" a'", 
 the hose being attached to the latter by screw-couplings in 
 the usual manner, and closed by the ordinary screw-cap. 
 Tlie delivery-valves and passages at the top of the force- 
 pump are similar to the suction-valves and passages men- 
 tioned. B, the air-vessel, composed of copper, its form being 
 spherical ; l> h, delivery-pipes, to which the hose is attached. 
 "Wlu'u only one jet is required, the opposite pipe may be 
 closed by a screw-cap, as usual. The piston of the force- 
 pump is provided with double leather packing, the piston- 
 rod being nuide of copper. C, boiler, constructed on the 
 liiineiple of the oi'dinary locomotive boiler, containing an 
 adequate number of tubes of suitable diameter. The top of 
 the steam-chamber and the horizontal part of the boiler is 
 covered w'ith wood as usual, in order to prevent loss of heat 
 by radiation, c, fire-door ; c', ash-pan ; c", box attached to 
 end of boiler, enclosing the exit of the tubes. The hot air 
 from the tubes entering this box is passed off through a 
 smoke-pipe c'", the exit of which makes a half-spiral turn 
 round the air-vessel, as shown in the illustration, e"", iron 
 brackets, riveted to the boiler and supported by the carriage- 
 frame. C'*, a wrought-iron brace, bolted to the carriage-frame, 
 for supporting the horizontal part of the boiler. E, vertical 
 ])ipe attached to the top of the steam-chamber, containing 
 a conical steam-valve e, and also the safety-valve e' / e", 
 
388 THE STEAM FIBE-ENGINE. chap. xxiv. 
 
 regulating screw and handle, connected to the steam-valve, 
 for admitting or shutting off the steam ; <?'", induction-pipe 
 for conveying the steam from the boiler. F, doul)le-actin_f 
 working cylinder, provided -with steam-passages and slide- 
 valve of the usual construction ; it is firmly secured to the 
 carriage-frame by means of brackets, in the same manner as 
 the force-pump. /, eduction-pipe, for cariying oft" the steam 
 to the atmosphere ; /', piston, provided -with metallic pack- 
 ing on Barton's plan ; /", piston-rod of steel, attached to 
 the piston-rod of the force-j)ump) by means of the cross-head 
 G, composed of Avrought iron ; both piston-rods being in- 
 serted into the said cross-head and secured by keys, g, 
 tappet-rod, attached to the cross-head, for moving the slide- 
 valve of the steam-cylinder by means of the nuts g' </'. The 
 latter may be placed at any desirable position on the tajipet- 
 rod, to ensure a regular action of the valve. A short axle 
 of wrought iron, turning in bearings attached to the cover 
 of the steam-cylinder, operates the steam-valve by appro- 
 priate levers acted ujion alternately by the nuts g' g'. I, 
 feed-pumjj for supplying the boiler, provided with spindle- 
 valves on the ordinary plan, its suctiou-pijie communicating 
 with the valve-chamber of the force-pump, the feed-^witer 
 being carried into the horizontal part of the Ijoiler. /, 
 plunger of the force-pump, composed of gun-metal or copper, 
 attached to the cross-head G. 
 
 Before proceeding with the description, it will be neces- 
 sary to state that, although the efficiency of the " steam-blast " 
 was well ascertained at the time, the prevailing opinion that, 
 
CHAP. XXIV. 77//: STEAM I'lHE-EXdlXE. 389 
 
 in (••niiiiKHi witli the Int-oiuotive engine, red-hot sparks would 
 eiiiauatc fr<iiii a steam tire-engine, and prove dangerous among 
 houses in a elose-lmilt eity, I was compelled to resort to 
 the employment of a blowing apparatus for generating the 
 neeessarily large (piantity of steam indispensable to render 
 the machine efficient. 
 
 J, blowing ap[iaratus, consisting of a square wooden 
 bo.\, with panelled si<les, in which a thin s(piare piston _/' is 
 applied, leather packing being employed to make air-tight 
 joints. J, circular holes through the sides for admitting 
 atmospheric air into the box, these holes being covered on 
 the inside by pieces of leather or India-rubber cloth acting 
 as valves, j", similar holes through the top of the box for 
 passing off the air, at each stioke of the i)iston, into the 
 receiver or regulator K, which is provided with a movable 
 top I: The latter is com])osed of wood, joined by leather 
 to the upper part of the box, a thin sheet of lead being 
 attached thereto, in order to keep up a certain pressure of 
 air in the regulator. k\ channel made of sheet-iron, attached 
 to the lilowing apparatus, communicating freely with the 
 regulator K. To the said channel is connected a conducting- 
 pipe, marke<l by dotted lines in the longitudinal section of 
 the engine, for conveying the air from the receiver into the 
 a-sh-pan under the furnace of the boiler at l-'. The con- 
 ducting-pipe, it should be ol)served, passes along the inside 
 ()f the carriage-frame on either side. 
 
 L L, parallel iron rods, to which the piston of the blow- 
 ing apparatus is attached. These rods work thiough guide- 
 
390 THE STEAM FIBE-EKGINE. chap. xxiv. 
 
 brasses / I, and may be attached to the cross-bead G l)y keys 
 at I' v. The holes at the end of the said cross-head admitting 
 these rods are sufficiently large to allow a free movement 
 "whenever it is desirable to woi'k the blowing apparatus inde- 
 pendently of the engine. M, spindle of wrought iron, placed 
 transversely, turning in bearings fixed under the carriage- 
 frame. To this spindle are fixed two crank-levers m m, 
 which, by means of two connecting-rods m' m', give motion 
 to the rods L L attached to the piston of the blowing appa- 
 ratus. 
 
 N, crank-lever secured to the end of spindle M, which, 
 by ineans of a connecting-rod applied to a crank-pin, fixed 
 in the hub of the carriage-wheel, on the outside, will com- 
 municate motion to the blowing apparatus whenever the 
 carriage is in motion. By this simple expedient a powerful 
 combustion will be kept up in the boiler furnace while the 
 engine is on its way to the place of conflagration. By de- 
 taching the connecting-rod referred to, and applying a hand- 
 lever to N, the blowing apparatus may be operated by 
 manual labor. The carriage-frame should be made of oak, 
 plated with iron all over the outside ; the top plate to have 
 small recesses, as shown in the drawing. The lock of the 
 carriage, axles, and springs to be made as usual, only differ- 
 ing by having the large springs suspended below the axle. 
 The carriage-wheels to be constructed on the suspension prin- 
 ciple ; spokes and outside of rim to be made of wrought 
 iron, very light. 
 
CHAPTER XXY. 
 
 THE STEAM-SHIP PRINCETON. 
 (SEE PLATES 38, 39, AND 40.) 
 
 In the sixty-sixth year of the American Rejiublic — 1842 
 — the first steam ship-of-wai- provided -witli a submerged j^ro- 
 peller and steam machinery located below water-line was 
 built at Philadelphia. Complete success attended this under- 
 taking, as will be seen from the following extract from a 
 speech delivered in the Senate of the United States by Sena- 
 tor Mallory, of Florida, May 14, 1858 : 
 
 "In 1839 Congress authorized the construction of three 
 war-ships. In 1840 the Secretaiy of the Kavy, in obedience 
 to that law, ordered two to be constructed. There were two 
 plans at that time. The question of whether steam could or 
 could not be successfully applied to war-vessels had not then 
 been solved ; the fear of danger from ignition by fire pi-e- 
 vailed in the minds of all naval men, and the problem Avas 
 to be solved. Its solution was demanded by every naval 
 
392 THE STEAM-SUir FUmCEIOX. chap. xxv. 
 
 power on earth. One of the officers of our navy, Captain 
 William Plimter, submitted a plan by which wheels were to 
 be inserted in the bilge of the vessel on each side— s\ib- 
 merged wheels. Ericsson had demonstrated his plan to be 
 feasible, practicable, and just, already. This "was no experi- 
 ment in the Princeton. The experiment had been made at 
 great cost by Captain Ericsson. . . . The Secretary of 
 the Navy, in authorizing the construction of these two ves- 
 sels, directed one to be constructed on Hunter's plan — not 
 model, as the Senator from Michigan says ; the model -was 
 not in cpxestion at all ; our models were as good then as 
 they are now ; but the point Avas the application of steam 
 to naval purposes. One was to be built on Eilcsson's plan, 
 and one on Hunter's plan. Hunter's plan proved a total 
 failure ; Ericsson's plan laid the foundation of the present 
 steam marine. She was the first war propeller ever built on 
 the face of the earth. In the Princtton he brought for\\ ard 
 not only his propeller invented by himself, but a great 
 many appliances appurtenant to steam navigation which 
 have since been used in our service. 
 
 "Now, to show the estimate in which this shiji was held, 
 I will mention that the American Institute, hearing of it, 
 sent a committee to investigate the condition and success of 
 what they deemed an experiment. I will not read the whole 
 irport I if the committee of the American Institute. My j^ur- 
 pose here is to shoAV the value of Captain Ericsson's services 
 to our country. The committee conclude by saying, after 
 summing up the speed of thirteen knots: 
 
CHAP. XXV. TE£ STEAM- SUIl' PRINCETON. 393 
 
 "'In coucliisiou, your committee beg leave to present the 
 Princeton as every way worthy of tlie highest honors of this 
 Institute. She is a sublime conception, most successfully 
 realized, an effort of genius skilfully executed, a grand, 
 unique combination, honorable to the country as creditable 
 to all engaged upon her.' 
 
 "This is signed by ten members of the Institute. Fur- 
 ther to illustrate : Cai^tain Stockton, in 1844, after describing 
 the ship, her guns and her arniameut, says everything human 
 language can to extol her in the eyes of the American 
 people. As I said, the Princeton is the foundation of om* 
 present steam marine. It is the foimdation of the steam 
 marine of the whole world. She created universal surprise 
 w^herever she was seen. She revolutionized naval vessels ; 
 and hereafter, in maritime war, those who send sailing vessels 
 to sea send them but to be captui-ed." 
 
 The following extract from the work on " Naval and Mail 
 Steamers," published 1853 by Charles B. Stuart, Engineer- 
 in-Chief of the United States Navy, furnishes some imijortaut 
 particulars relating to the Princeton : 
 
 "This vessel was designed by, and constructed under the 
 superintendence of. Captain John Ericsson, of New York. 
 
 Length on deck, .... 
 Length between perpendiculars, . 
 Extreme beam on deck. 
 Depth of lower hold to berth-deck. 
 Depth from berth to spar-deck, . 
 
 164 feet 
 
 
 150 
 
 (f 
 
 
 30 
 
 (1 
 
 G inches 
 
 14 
 
 (1 
 
 
 7 
 
 u 
 
 6 " 
 
TON. 
 
 CHAr. XXV. 
 
 21 
 
 feet 6 inches. 
 
 GTo 
 
 tons. 
 
 418 
 
 a 
 
 954 
 
 ii 
 
 1,04G 
 
 11 
 
 gl^t, 
 
 346 square ft. 
 
 
 390 
 
 394 THE aiEAM-SUIF FmyCHTON. 
 
 Total deptli of vessel, . 
 
 Measurement burthen, . 
 
 Launching ■weight of hull, . 
 
 Displacement at IG^ feet draught, 
 
 Displacement at IS feet draught, 
 
 Immersed midship section at IG^ feet draught, 
 
 a a a jg 
 
 Draught of water at deepest load, with 200 
 
 tons of coal on board, . . . .19 feet 4 inches. 
 
 Draught of water with 100 tons of coal ^ ,. ,,.„,., 
 
 ^ . . forward, 14f feet, 
 
 in, after-bunkers and provisions, and ,- 
 
 aft, 18i " 
 
 water for the crew, half out, J 
 
 Mean draught of water with half coal out, 
 
 and all other weights full, . . . 17 feet. 
 "The peculiarity of model consisted in a ver^^ flat floor 
 amidships, with great sharpness forward and excessive lean- 
 ness aft, the I'un being remarkably fine, Avith a great extent 
 of dead-wood terminating in a stern-post of the unusual 
 thickness of twenty-six inches at the centre of the propeller- 
 shaft, but tapering above and below. This dead-\vood and 
 stern-post was pierced by a hole thirteen inches diameter for 
 the passage of the propeller-shaft. The stern, measuring from 
 a perpendicular from the aft end of the spar-deck, overhung 
 the stern-post fifteen and one-half feet, and depending from 
 it was a false stern-post, leaving a space of six feet, fore and 
 aft, between it and the true stern-post. The pi'opeller (four- 
 teen feet in diameter, composed of bronze, with six propeller- 
 blades, pitch thirty-five feet) was placed within this space. 
 
CUAP. XXV. THE STEAMSHIP PRIXGETO^i. 395 
 
 The false stern-post, or rudder-post, was composed of a 
 wrougbt-irou bar, covered witli half-inch-tluck copper plate ; 
 it was attached at top by brass flanges to a strong oak knee, 
 securely bolted tct the counter of the vessel ; the lower part 
 of it was attached by similar flanges to a solid oak timber 
 placed as a continuation of the keel beyond the true stern- 
 post ; this timber was fourteen inches deep, and securely 
 boltt'tl to the keel and dead-wood. The metallic false stern- 
 post was five and three-eighths inches broad athwart ship, 
 and t\vo feet long fore and aft; the forward part was brought 
 to an acute angle to diminish its resistance to the water, 
 wliilf its after-part was square to receive the attachment of 
 the rudder. The rudder was also composite in its construc- 
 tion, being formed of a wrought-iron frame, the interstices of 
 which were filled in with pieces of five-inch-thick pine plank, 
 the Avhole cased over with a copper plate three-sixteenths of 
 an inch thick. The thickness of the rudder athwart ship was 
 the same as that of the false stern-post, viz., five and three- 
 eighths inches. 
 
 ENGHiTES. 
 
 "The semi-cylinder engine of the Princeton is unquestion- 
 ably the most remaikable modification of the steam-engine 
 that has ever been carrii'd into successful practice. A vibrat- 
 ing piston of a rectangular form moving in a semi-cylinder 
 is an old mechanical device. Mr. Watt, in his celebi-ated 
 patent, embraced this plan for transmitting the motive force 
 of steam to maohineiy. Siuce liis time several engineers liave 
 attempted to buihl engines on this plan, but without succes.s. 
 
396 THE STEAMSHIP PBINCETOW. chap. xxv. 
 
 In common witli Mr. Watt, they have adopted the single 
 semi-cylinder with packing against the piston-shaft. Erics- 
 son's plan differs materially from these various attempts, he 
 having introduced double or compound semi-cylinders of dif- 
 ferent diameters, with double pistons placed in opposite direc- 
 tions on the piston-shaft, both being acted upon by the steam 
 at the same time, their differential force being the effective 
 motive power of the engine. The combination of two such 
 double semi-cylinders, arranged so as to transmit their power 
 in directions nearly rectangular to a crank-pin common to 
 both, also contributes to the complete success of this singular 
 engine. 
 
 "By reference to the plate, it will be seen that the upper 
 semi-cylinder, which contains the reacting piston, is twenty 
 inches in diameter, the lower or working semi-cylinder being 
 seventy-two inches diameter, both ninetj^-six inches long in 
 the clear. The radius of the reacting piston, being deducted 
 from that of the working piston, leaving twenty-six inches 
 effective Avidth of piston, with its centre of pressure placed 
 10 + 13 = 23 inches from the centre of the piston shaft. 
 The active piston area will thus be 26 X 96 = 2,516 super- 
 ficial inches, moving at a mean distance of twenty-three inches 
 from the centre through an arc of ninety degrees. 
 
 " On close examination this engine will be found to pos- 
 sess many peculiar properties, some of which merit particular 
 notice. The vibration of the working piston will be found 
 to correspond nearly to the beat of the pendulum, and thus, 
 unlike the ordinaiy engine, the return movement of the piston 
 
CHAP. XXV. THE STEAM-Snir PRINCETON. 397 
 
 at each termiiiatiou of the stroke is materially assisted by 
 the force of gravity. An undue accumulation of condensed 
 water on the piston, so difficult to carry off in the ordinary 
 engine, presents no inconvenience here, as the inclined posi- 
 tion of the piston allows the condensed water to flow gradu- 
 ally down into the steam-passages before the piston reaches 
 the termination of the stroke. The outward tendency of the 
 packings, induced by centrifugal force, assists materially in 
 forming tight joints, the main packing being held out by the 
 force of gravity. The lateral yielding of the piston-shaft, 
 caused by the pressure of the steam on opening the inlet- 
 valve, tends to give additional tightness to the packings, the 
 pistons being forced into reduced radial limits by the yielding 
 alluded to. The crank-levers attached to the piston-shafts 
 being placed nearly in the same position with the main 
 pistons, it will be found that the crank-journals of those 
 shafts are relieved from j^ressurc, on principles analogous 
 to the i-elieving of water-wheel journals, by transmitting tlie 
 power at some point near the centre of pressure. The increase 
 of force imparted to the crank-pin at each half-turn of the 
 main crank, owing to the angular position of the piston-rod 
 ci-anks, and happening, as it does, when the former presents 
 short leverage, is a marked feature of this engine. 
 
 "The small angular movement (ninety degrees) of the 
 main piston .also deserves attention. A greater motion, while 
 it would augment the power of any given-sized cylinder, 
 would cause undue strain on all the principal beai'ings, as 
 the force of the piston obviously increases in the invei'se 
 
398 THE STEAM-SIIIP PBIXGETON. chap. xxv. 
 
 ratio of tlie shie.s of tLe angles of tlie pistou-sluift cninks, 
 with reference to the position of the conneoting-rod. A very 
 moderate increase of diameter makes up tlie h)ss consequent 
 on the sliort arc tliroiigh A\]iich the piston ^'ibrates. Very 
 deep cylinder-covers, giving great strength tti resist the u[)- 
 ward pressure of the steam, may also l)e named as an advan- 
 tage i-esulting from the shoi't viljration. 
 
 "xVs an instrument of producing. In' dii-ed means, the 
 high speed requisite for screw-propellers, this engine com- 
 mends itself to the engineer. In its fitness for screw-vessels 
 it seems to fulfil every condition. The very limited number 
 of woi-kiug parts, and the small amount of matter to be kept 
 in motion, are self-evident advantages, best understood by 
 reference to the plate. It is important to notice that the 
 pistou-shaft journals are supported by bearings placed out- 
 .side of the heads of the semi-eylinders, and that the brasses 
 admit of adjustment in every direction ; the exact position of 
 the centre of the piston-shafts, -with relati(jn to the centre of 
 the semi-cylinders, being indicated by an external index, 
 enabling the engineer at all times to keep the shaft in line. 
 
 "The main packing is rendered accessible by lifting up 
 one of the covers and removing the nearest side-plate form- 
 ing the packing-groove. The upper packing becomes acces- 
 sil)le by lifting up the centre-piece whicli forms the ujiper 
 semi-cylinder. It may be stated with regard to tlie packings, 
 that the Prineetoi), after having served during the war with 
 Mexico, was desj^atched on a ci'uise to the Mediterranean 
 without recpiiring new packings." 
 
cu.vr. XXV. TUB STEAM-SHir I'L'IXCETON. 399 
 
 Mr. J. O. Siuyvnt, iu ii lectui'e on " Steam Navigation and 
 the Arts of Naval Warfare," delivered before the Boston 
 Lyceum in 1844, stated with reference to the motive power 
 of the Princeton : " The next peculiarity to be noticed in 
 tlie Princeton is the absence of the ordinary tall smoke-pipe 
 employed to pi-oduce the draught for keeping up combustion 
 in the furnaces of the boilers. The smoke-pipe has hitherto 
 formed a serious objection to a steamer as a ship-of-w ar ; for 
 the moment it is carried away the efficiency of the engines 
 ceases from want of steam. The draught in the boilers of 
 tlie Princeton is promoted by means of blowois plat-ed in 
 the bottom of the vessel, and is cpiite independent of the 
 lieigJtt of the smoke-pipe, wliich is only carried about five 
 feet above the deck of the ship. If this inconsiderable pro- 
 jection should become partially deranged by a sliot, the 
 (h-aught kept up by tlie Ijlowers will continue as efficient 
 as before. 
 
 " It is not out of place here to observe that Ericsson was 
 the first to apjily to marine engines centrifugal blowers, now 
 so common in this country in all boilers using anthracite 
 coal. In the year IS.'H he ai)plied such a blower, worked 
 by a separa^te small stcaiii-engiiic, to the steani-packct i'orsdir, 
 of one hundi'ed and twenty liorse-powej', plying between 
 Liverpool and Belfast." 
 
CHAPTER XXYI. 
 
 TWELVE-INCH WEOUGHT-IEON GUN AND CARRIAGE. 
 
 (SEE PLATE 41.) 
 
 The question has frequently been asked, When was 
 VTi'ought-iron orcluauce first introduced on board ships-of-war ? 
 When was breeching first disjiensed with and wrought-iron 
 carriages introduced ? The heading on the plate referred to 
 gives an answer to these queries. 
 
 Captain Robert F. Stockton, of the United States Navy, 
 on visiting England — 1839 — for the purpose of witnessing the 
 trial of a small screw-steamer (the first iron vessel to cross 
 the Atlantic) built for him by Messrs. Laird, to my design, 
 consulted me regarding the possibility of constructing naval 
 ordnance of wrouo;ht iron. Beinar an advocate of that mate- 
 rial, I readily met the wishes of Captain Stockton, and at 
 once prepared drawings of a gun of 12-inch calibre. The 
 Mersey Iron-Works, near Liverpool, being willing to enter 
 into a contract with Captain Stockton, received forthwith 
 
cuAi'. XXVI. ]yi!()iaiiT-n;o.\ auxs axd caiuhaoks. 401 
 
 an order from that euterprisiug and spirited officer to build 
 the gun at his expense. Experienced commodores at the 
 time protested loudly against the proposition to mount "the 
 monster gnu " on board a vessel so lightly built as tlit* 
 Princeton, insisting tliat, among other difficulties, the breech- 
 ing Avould tear her upper works to pieces. It was Urged 
 by the opponents of my new system that the handling of 
 such guns at sea would prove impossible, the constructing 
 carriages of sufficient strength being pointed out as imprac- 
 ticable; while the imprudence on the part of the Navy 
 Department of entrusting such matters to mere engineer- 
 ing skill was severely criticised. In spite of remonstrances, 
 however, Captain Stockton's influence with the Govei-nment 
 prevailed. In the meantime the problem of handling the 
 12-incli gun received due attention. Calculations of the 
 dynamic equivalent of the recoil convinced me that a mode- 
 rate resistance, if continuous and uniform, would suffice to 
 bring the piece to rest in less space than that required by 
 breeching. Friction, being the simplest means of obtaining 
 a continuous resistance, was accordingly resorted to. The. 
 method adopted will be readily comprehended by reference 
 to the illustration already referred to, which represents a side 
 elevation, top view, and imuI views of the gun, carriage, and 
 slide, mounted on board the Princeton in 1843. Two pieces 
 of timber of semi-circular section, placed a few inches apart, 
 slightly taper towards the front end of the slide, are secured 
 to the latter in such a manner as to admit of some vertical 
 motion. A broad hoop of plate-iron, attached to the car- 
 
402 WEOUGET-IBOX GUNS AND CABEIAGES. chap. xxvi. 
 
 riage, clasps the two timbers. An axle pi'ovided with a cam 
 in the middle, operated by a lever, is placed across the front 
 end of the slide passing through the space between the fric- 
 tion timbers. Obviously, these timbers will be pushed apart 
 with considerable force if the transverse axle be turned so 
 as to place the cam in a vertical position. The friction be- 
 tween the hoop and the timbers produced by the pressure 
 of the cam, it is scarcely necessary to observe, "will be greatly 
 enhanced during the recoil of the gun, in consequence of the 
 increasing vertical depth of the timbers towards the rear 
 end of the slide. The transverse axle being turned so as to 
 place the cam in a horizontal position, permitting the friction 
 timbers to approach each other, will of course at once relieve 
 the friction. The monitor gun-carriages, constructed, like the 
 Princeton''s, on the plan of checking the recoil by friction, 
 differed as regards the mode of clasping the friction-timbers, 
 a screw being employed to produce the requisite compres- 
 sion in place of the transverse axle and its cam. But in 
 constructing gun-carriages for the celebrated thirty Spanish 
 gunboats, and in all recent carriages, I have returned to 
 the plan of employing a cam for setting up and relieving 
 the friction, whereby, as in the case of the Princeton car- 
 riage, the operation becomes almost instantaneous. 
 
 The 12-in. Avronght-iron gun of the Princeton, as stated, 
 was manufactured at the Mersey Iron-AVorks, of the very 
 best materials, it has been asserted; but, on being tested in 
 this country, it proved too weak. I therefore resorted to 
 the expedient of hooping the breech of the piece up to the 
 
CUAP. XXVI. WBOUGHT-II;OX GUXS AND CARI^IAGES. 403 
 
 truunion-band ; the hoops beiug made of the best quality 
 of American wrought-irou put on in two tiei-s, shrunk one 
 over the other in such a manner as to break joint. This 
 expedient proved entirely successful, the gun having stood 
 all tests to which it has been subjected. It was a solid 
 cast-iron shot from this 12-in. gun which in 1842 pierced a 
 wrought-iron target 4^ ins. thick, up to that period considered 
 proof against naval ordnance. 
 
 The United States Government having been the firat to 
 introduce heavy MTought-iron ordnance for naval purposes, 
 why does it not continue to build guns of that material ? 
 European artillerists repeatedly jnit this question. Probably 
 the answer will be found in the fact that, although having 
 in the meantime successfully constructed rifled wi-ought-iron 
 ordnance of considerable size, the first essay at buildiuo- 
 heavy guns for naval purposes proved most disastrous. Im- 
 mediately after the trial of the gun referred to, manufactured 
 in England, a 12-in. smooth-bore, of much heavier metal, was 
 forged at Hamersley Forge, bored and turned in New York, . 
 and considered at the time to be a remarkable specimen of 
 good workmanship. It was at once mounted on board the 
 Princeton, by the side of its slender companion, upon a similar 
 carriage. Much confidence was placed in the strength of this 
 magnificent gun on account of the supposed superior quality 
 of American iron. It stood the proof-charge and some preli- 
 minaiy tests, but, on being fired on a festive occasion while 
 the ship was at Washington, the admired piece burst, Avith 
 the sad result recorded in the naval annals of the time. 
 
CHAPTER XXVII. 
 
 APPLICATION OF THE SUBMEEGED PEOPELLER FOI! COM- 
 MERCIAL PURPOSES. 
 
 Dttking tlie constniction of the Princeton numerous pro- 
 peller-vessels were built for carrying freight on the rivers 
 and inland waters of the country, the machinery at first being 
 built in New York, Philadelphia, and Oswego. The line of 
 propeller-steamers between Philadelphia and Baltimore, which 
 seriously interfered with the freight traffic of the Philadel- 
 phia and Baltimore Eailroad, merits special mention. 
 
 The annexed table of propeller-vessels built up to Decem- 
 ber, 1843, prepared by Lieutenant Johnson, of the Swedish 
 Navy, in pursuance of orders from his Government, shows 
 the extraordinary rapidity of the adoption of the propeller 
 for commei'cial purposes. 
 
 The Migineer, in laying the table referred to before its 
 readers. May 11, 1866, observes: 
 
 "The fate of mechanical inventions is much like that of 
 
CHAP. XXVII. SCREW PROPULSION AKD COMMERCE. 405 
 
 tlie seed in the parable. Tlie Invention must fall on a pro- 
 per soil, and be nurtured by favorable circumstances of time 
 and place, in order to bloom into success. The application 
 of the steam-engine to navigation was of greater necessity 
 to the large e.xteut of the rivers and lakes of the States than 
 \vith oui-selves ; and Fulton did right to take his marine 
 engine back to his own country. For similar reasons the 
 screw-propeller worked its way into use there much quicker 
 than with ourselves. This fact is very evident from the 
 table furnished to Mr. Woodcroft by Captain Ericsson, pre- 
 pared by Lieutenant Johnson, of the Swedish Navy, in 1843. 
 Early in the following year (1844) a very large addition 
 was made to the steam fleet provided with Ericsson's pro- 
 peller both on the inland waters of America and on the 
 ocean. Of the latter may be mentioned the bark Bdifh and 
 the steamshi^w JlcICim, Marmora, and Mamachusetts. The 
 last vessel was subsequently purchased by the United States 
 Government. It became the flag-ship of General Scott at 
 the landing of Vera Cruz, which lesulted in the conquest 
 of Mexico. It is woi-thy of notice that Ericsson applied his 
 propeller to upwards of sixty vessels in America before any 
 other form of propeller was adopted or a single attempt made 
 at evading his patent. Nor is it less worthy of remark that 
 the adaptation of his pi'opeller proved a great commercial 
 success from the start, many of the original vessels being 
 now, after fifteen years of service, in good working condi- 
 tion." 
 
406 
 
 SCBEW PllOPULSION AND COMMERCE. 
 
 CHAP. XXVII. 
 
 List of Steam Vessels in North America provided with 
 Ericssox's Screw-Propeller up to December, 1843. 
 
 Names of the vessels. 
 
 Destination. 
 
 
 Robert F. Stockton 
 
 Vandalia 
 
 Delaware and Schuylkill 
 
 Oswego to Chicago 
 
 2 
 2 
 1 
 1 
 2 
 
 2 
 2 
 2 
 2 
 
 Clarion 
 
 New York to Havana 
 
 Baron Toranto 
 
 Rideau Canal and St. Lawrence. 
 Rideau Canal and St. Lawrence. 
 Rideau Canal and St. Lawrence. 
 Rideau Canal and St. Lawrence. 
 
 Philadelphia to Albany 
 
 Philadelphia to Albany 
 
 Philadelphia to Hartford 
 
 Pliiladelphia to Hartford 
 
 Erie Canal 
 
 Royal Barge 
 
 Propeller 
 
 Ericsson 
 
 Ironside 
 
 Anthracite 
 
 Black Diamond 
 
 Vnlcan 
 
 Pioneer 
 
 Oswego 
 
 Oswego to Chicago 
 
 Chicago 
 
 Osw^ego to Chicago 
 
 Cumberland 
 
 Philadelphia to Baltimore 
 
 Philadelphia to Baltimore 
 
 New York to Canada 
 
 Ericsson 
 
 Pilot 
 
 Phcenix 
 
 New York to Canada 
 
 Governor McDowell 
 
 Jefferson (Revenue Cutter) 
 Legare 
 
 James River Canal to Virginia. 
 Lake Erie 
 
 Delaware River 
 
 
 
CHAP, xxvii. SCJii:W rEOPULSIOX asd commeece. 
 
 407 
 
 3 
 
 4 
 5 
 6 
 
 7 
 8 
 
 9 
 
 10 
 
 II 
 
 12 
 
 13 
 14 
 15 
 i6 
 
 17 
 i8 
 
 19 
 20 
 21 
 
 List of Steam Vessels in Xortii Ameuica puovided with 
 Ericsson's Screw-Phopelleu up to December, 1843. 
 
 60 
 40 
 60 
 18 
 36 
 18 
 18 
 40 
 40 
 40 
 40 
 18 
 40 
 40 
 40 
 40 
 18 
 18 
 IS 
 150 
 150 
 
 42 
 
 140 
 
 250 
 
 70 
 
 70 
 
 70 
 
 70 
 
 200 
 
 200 
 
 200 
 
 200 
 
 55 
 
 140 
 
 140 
 
 80 
 
 80 
 
 55 
 
 55 
 
 20 
 
 342 
 
 342 
 
 Ft. 
 
 70 
 
 98 
 
 104 
 
 72 
 
 72 
 
 72 
 
 72 
 
 100 
 
 100 
 
 100 
 
 100 
 
 82 
 
 98 
 
 98 
 
 78 
 
 78 
 
 82 
 
 82 
 
 90 
 
 140 
 
 140 
 
 Ft. In. 
 
 10 
 
 21 6 
 24 6 
 16 6 
 16 6 
 16 6 
 16 6 
 
 22 6 
 22 6 
 22 6 
 22 6 
 14 4 
 21 6 
 21 6 
 18 6 
 18 6 
 13 6 
 13 6 
 13 6 
 24 
 24 
 
 Ft. In. 
 
 7 
 
 6 6 
 
 13 
 
 4 6 
 
 4 6 
 
 4 6 
 
 4 6 
 
 6 
 
 6 
 
 6 
 
 6 
 
 3 6 
 
 6 6 
 
 6 6 
 
 6 6 
 
 6 6 
 
 3 6 
 3 
 1 
 7 
 7 
 
 9 
 
 5 9 
 
 5 9 
 
 6 
 6 
 6 
 6 
 5 6 
 
 5 
 
 9 
 
 5 
 
 9 
 
 6 
 
 6 
 
 G 
 
 6 
 
 5 
 
 6 
 
 5 
 
 6 
 
 5 
 
 
 
 9 
 
 6 
 
 9 
 
 6 
 
 1839 
 1841 
 
 1842 
 
 1843 
 
408 
 
 SCREW PliOPULSIOX AND COMMERCE. chap, xxvii. 
 
 List of Steam Vessels in North America provided with 
 Ericsson's Screw-Propeller up to December, 1843. 
 
 Names of the vessels. 
 
 Destination. 
 
 
 Hercules 
 
 Perry sboroiioii 
 
 Cleveland 
 
 Baltimore 
 
 Priuceton 
 
 Lion 
 
 Eagle 
 
 Mohawk 
 
 C. Bristol 
 
 New London 
 
 Uncas 
 
 IVew London 
 
 Enterprise 
 
 New York 
 
 Washington 
 
 Rufus Page, Owner. . . . 
 Riifus Page, Owner . . . . 
 
 Buck, Owner 
 
 Captain Sanford, Owner 
 
 Williams & Barns 
 
 Captain Coit 
 
 Lake Erie 
 
 Lake Erie 
 
 Lake Erie 
 
 PMladelphia to Baltimore 
 
 Philadelpliia Station 
 
 New York to Hartford 
 
 New York to Hartford 
 
 Albany to Hartford 
 
 Chicago and the Great Lakes. . 
 
 Mobile 
 
 Mobile 
 
 Owners, Wdliams & Barns . . . . 
 Lake Ontario and St. Lawrence. 
 
 Oswego to Chicago 
 
 Hudson River 
 
 East Coast of America 
 
 East Coast of America 
 
 East Coast of America 
 
 East Coast of America 
 
 East Coast of America 
 
 East Coast of America 
 
 2 
 
 22 
 
 2 
 
 23 
 
 1 
 
 24 
 
 1 
 
 25 
 
 2 
 
 26 
 
 1 
 
 27 
 
 1 
 
 28 
 
 1 
 
 29 
 
 1 
 
 30 
 
 1 
 
 31 
 
 1 
 
 32 
 
 2 
 
 33 
 
 1 
 
 34 
 
 2 
 
 35 
 
 1 
 
 36 
 
 1 
 
 37 
 
 1 
 
 38 
 
 2 
 
 39 
 
 2 
 
 40 
 
 2 
 
 41 
 
 2 
 
 42 
 
CHAP. XXVII. scnEW rBoFULswy and commerce. 
 
 409 
 
 List of Ste.vm Vessels ix North Amkrica provided with 
 Ericsson's Sckew-Propellek up to December, 1843. 
 
 
 1 
 
 i 
 
 t: 
 
 .a 
 
 i 
 
 2 
 
 
 
 0. 
 
 1 
 
 a 
 .3 
 
 1 
 
 3 
 
 .a 
 
 J 
 
 
 
 
 /K. 
 
 rt. In 
 
 ft. In. 
 
 rt. In. 
 
 
 22 
 
 60 
 
 250 
 
 
 
 8 
 
 6 
 
 1843 
 
 23 
 
 50 
 
 250 
 
 
 
 8 
 
 6 
 
 
 24 
 
 60 
 
 280 
 
 
 
 8 
 
 6 10 
 
 
 25 
 
 50 
 
 127 
 
 78 
 
 19 6 
 
 6 6 
 
 7 2 
 
 
 26 
 
 400 
 
 672 
 
 164 
 
 30 
 
 17 6 
 
 14 
 
 
 27 
 
 60 
 
 172 
 
 115 
 
 23 6 
 
 6 
 
 6 10 
 
 
 28 
 
 50 
 
 172 
 
 115 
 
 23 6 
 
 6 
 
 6 10 
 
 
 29 
 
 50 
 
 172 
 
 125 
 
 22 
 
 6 
 
 7 
 
 " 
 
 30 
 
 60 
 
 300 
 
 ... 
 
 26 
 
 8 
 
 7 4 
 
 
 31 
 
 50 
 
 172 
 
 125 
 
 22 
 
 6 
 
 7 
 
 
 32 
 
 50 
 
 172 
 
 125 
 
 22 
 
 6 
 
 7 
 
 
 33 
 
 50 
 
 172 
 
 125 
 
 22 
 
 6 6 
 
 7 
 
 
 34 
 
 20 
 
 70 
 
 
 
 
 5 9 
 
 
 35 
 
 50 
 
 140 
 
 98 
 
 21 6 
 
 6 6 
 
 5 9 
 
 
 36 
 
 00 
 
 170 
 
 103 
 
 26 
 
 7 6 
 
 7 4 
 
 
 37 
 
 50 
 
 172 
 
 115 
 
 23 6 
 
 6 
 
 6 10 
 
 
 38 
 
 50 
 
 172 
 
 115 
 
 23 6 
 
 6 
 
 6 10 
 
 
 39 
 
 55 
 
 172 
 
 125 
 
 22 
 
 6 
 
 7 
 
 
 40 
 
 55 
 
 172 
 
 125 
 
 22 
 
 6 
 
 7 
 
 
 41 
 
 56 
 
 172 
 
 125 
 
 22 
 
 6 
 
 7 
 
 
 42 
 
 55 
 
 172 
 
 125 
 
 22 
 
 6 
 
 7 
 
 
CHAPTER XXVIII. 
 
 IRON-CLAD STEAM BATTERY, WITH REVOLVING CUPOLA, 
 SUBMITTED TO EMPEROR NAPOLEON III. 
 
 (ILLUSTRATION, SEE PLATE 42.) 
 
 The illustration referred to is a ■ f ac-simile of a drawing 
 forwarded to the Frencli Emperor at Paris, September 26, 
 1854, accompanied by an elaborate description and demon- 
 stration of tlie utility of the battery.* 
 
 The following extracts from the description of the new 
 system of naval attack, forwarded as stated, furnishes a cor- 
 i-ect idea of its nature : 
 
 The present system of long range is abortive. 1st, because 
 large or heavy bodies cannot be projected to a great distance ; 
 
 * The Emperor promptly acknowledged the receipt of these documents through 
 Gen. Fave, whose letter commences with the following flattering sentence : 
 
 " L'Empereur a examine lui-meme aveo le plus grand soin le nouveaii sysdeme 
 d'attaque navale que vous lui avez communiqm'. 
 
 " S. M. me charge d'avoir I'honneur de vous informer qu'elle a trouve vos idees 
 tres-ingenieuses et dignes du nome celebre de leur auteur." 
 
CHAi'. xxviJi. HEVOLVIXG CUPOLA VESSEL. 411 
 
 2dly, because accurate aim at long range becomes absolutely 
 imi>ossible in practice. The recent trial of the Lancaster 
 gun, when subjected to the unavoidable oscillation of a 
 small vessel, may be cited as proof. Short range, " close 
 quarters," will remove both difliculties, as it admits of lai'ge 
 and heavy projectiles being employed, and because it ensures 
 accurate aim. Besides these advantages, a near approach to 
 the enemy renders attack under water practicable. These 
 facts establish the following propositions : 1st, a complete 
 system of naval attach demands a self-moving vessel capable 
 of passing witliiii range of guns of forts, and of moving at 
 pleasure in defiance of the fire of broadsides. 2dly, with a 
 vessel of such properties, a complete offensive system further 
 requires adequate means of throwing projectiles of large size 
 with absolute precision at short ranges, either point-blank 
 or at very great elevation ; the means of projecting shells 
 (movable torpedoes) under water at short distances being 
 also indispensable. 3dly. These conditions being fulHlled, the 
 system yet demands a projectile that will infallibly explode 
 at the instant of contact. 
 
 Accordingly, the writer has directed his e.xperiments and 
 laboi-s to the solution of the following problems: I. A self- 
 moving shot-proof vessel. II. An instrument capable of 
 projecting very large shells at slow velocities, but very accu- 
 rately, in accordance with previously-determined rate. III. 
 A shell not subject to any rotation in the direction of its 
 course, and so contrived as to explode with infallible cer- 
 tainty at the instant of contact. IV. A sliell (toipedo) 
 
412 EEVOLVING CUPOLA VESSEL. chap, xxviii. 
 
 capable of being projected under water, and certain to ex- 
 plode by contact, together witb an instrument for projecting 
 sucli a shell from the vessel at a certain depth below the 
 water-line. 
 
 The nature of the practical solution of the above prob- 
 lems will be readily comprehended by referring to the illus- 
 trations (see Plate 42). A brief extract of the document 
 forwarded to the Emperor will therefore suffice. 
 
 The vessel to be composed entirely of iron. The mid- 
 ship section is triangular, with a broad, hollow keel, loaded 
 with about 200 tons of cast-iron blocks to balance the 
 heavy upper works. The ends of the vessel are moderately 
 sharp. The deck, made of plate iron, is curved both longi- 
 tudinally and transversely, the curvature being 5 feet ; it is 
 made to project 8 feet over the rudder and propeller. The 
 entire deck is covered with a lining of sheet iron 3 inches 
 thick, with an opening in the centre 16 feet diameter. Over 
 this opening is placed a semi-globular turret of plate iron 
 6 Inches thick, revolving on a vertical column by means of 
 steam-po^ver and appropriate gear-"\voi'k. The vessel is pro- 
 pelled by a powerful steam-engine and screw-propeller. Air 
 for the combustion in the boilers and for ventilation within 
 the vessel is supplied by a large self-acting centrifugal 
 blower, the fresh air being drawn in through numerous 
 small holes in the turret. The products of combustion in 
 the boilers and the impure air from the vessel are forced out 
 through conductors leading to a cluster of small holes in 
 the deck and tiUTet. Surrounding objects are viewed through 
 
CHAP. XXVIII. REVOLVING CUPOLA VESSEL. 413 
 
 small perforations at appropriate places. Reflecting telescopes, 
 capable of being pi-otriuled or witbtlrawn at pleasure, also 
 afford a distinct view of surrounding objects. The rudder- 
 stock passes tbrougb a water-tigbt stufliug-box, so as to admit 
 of the helm being worked within the vessel. Shot striking 
 the deck are deflected, whilst shell exploding on it will 
 prove harmless. 
 
 Tube for projecting the shells to be made of cast iron 
 or brass, 20 inches bore, 2 inches thick, and 10 feet long. 
 It is ojien at one end, the other end being closed by a 
 door moving on hinges provided with a cross-bar and set- 
 screw, in order to be quickly opened and afterwards fli-mly 
 secured. The shell is inserted through this door, and pro- 
 jected by the direct action of steam admitted from the boiler 
 of the vessel through a large opening at the breech. The 
 induction-valve is made with a double face of large areas, 
 and moved by mechanism of instantaneous action, susceptible 
 of accurate regulation in regard to opening. One tube of 
 the above description is placed on a level on the platfonn 
 of the revolving tun-et. Two similar tubes are placed in 
 the body of the vessel, at a fixed inclination of 22 deg., 
 revolving on vertical pivots. These tubes are supplied with 
 steam through the centre of their vertical pivots, the admis- 
 sion of steam being regulated as before described. 
 
 The plan of throwing shells of several hundred pounds 
 by the direct power of steam of ordinary pressure, demands 
 special notice. Without reference to the result of actual 
 trial, a brief investigation of the theoiy on which the plan 
 
414 BEVOLVINQ CUPOLA VESSEL. chap, xxviii. 
 
 is based will show that shells of enormous size may be pro- 
 jected with uuerriug precision. 
 
 The Shell, composed of cast iron, is formed as delineated.* 
 A groove is made round the circumference at right angles 
 to the axis, into which an india-rubber ring is inserted to 
 form a steam-tight joint when the shell is put into the tube. 
 In order effectually to prevent rotation in the line of flight, 
 a tail in the form of a cross, composed of thin plate-iron, is 
 attached to the shell. Opposite to this tail a cavity is 
 formed, into which a cylindrical hammer is inserted. A 
 peroussion-wafer is placed under the hammer, which, being 
 always in advance of the shell, is struck at the instant of 
 contact, infallibly causing an explosion. 
 
 The Hydrostatic Javelin (torpedo-carrier), for convey- 
 ing the shell (torpedo) under water, consists of a cylindrical 
 block of light wood, 16 inches diameter, 10 feet long. At 
 one end of this block a 16-inch shell is attached, charged 
 ■with powder, and furnished with a percussion-hammer, as 
 above described. The other end of the block is pointed and 
 loaded at the under-side sufficient to balance the instrument 
 perfectly. The displacement being 1,000 pounds, the weight 
 of the whole is made to correspond accurately, in order to 
 ensure perfect suspension in the water. The javelin (torpedo- 
 carrier), when required, is passed through the vessel's bow 
 or side by means of a short tube, as shown by the drawing, 
 the water from the sea being kept out during the insertion 
 
 * Unfortunately, the copies of this and other delineations referred to have been lost, 
 henc3 cannot be presented in this work. 
 
CHAP, xxviii. BEVOLVIXQ CUPOLA VESSEL 416 
 
 by the obvious means of a slide-valve. The javelin (torpedo- 
 carrier) is projected — pushed out — by means of a rod attached 
 to the piston of a steam-cylinder of 18 inches diameter, 3 feet 
 stroke. A force of 10,000 pounds acting through 3 feet is 
 more than sufficient to propel the javelin 200 feet, at an 
 average velocity of 12 feet per second. The javelin (torpedo- 
 carrier) is readily kept at any particular depth during its 
 progress by a simple application of the hydrostatic pressure 
 on a tail or radder acting in the horizontal plane. The load 
 inserted at the tail end of the javelin (torpedo-conductor) 
 to balance the shell (torpedo) being applied at the bottom, 
 the instrument cannot turn in the water. 
 
 CONCLUDING REMARKS. 
 
 This new system of naval attack will place an entire 
 fleet of sailing vessels, during calms and light winds, at the 
 mercy of a single craft. "Boarding" as a means of defence 
 will be impracticable, since the turret guns, which turn like 
 the spokes in a wheel, commanding every point of the com- 
 pass at once, may keep off and destroy any number of boats 
 by firing slugs and combustibles. The loading at the breech 
 and the dispensing Avith sponging ensures a rapidity in the 
 discharge of missiles quite irresistible in an attempt at 
 boarding. A fleet at anchor might be fired and put in a 
 sinking condition before being able to get under way. 
 
 Of what avail would be the " steam guard-ships " if at- 
 tacked on the new system ? Alas ! for the " wooden walls " 
 that formerly "ruled the waves." The long-range Lancaster 
 gun would scarcely hit the revolving iron turret once in 
 
416 BEVOLYING CUPOLA VESSEL. chap, xxviii. 
 
 six hours, and tlieii, six cliances to one, its shot or shell 
 Avould be deflected hy the varying angles of the face of the 
 impregnable globe. When ultimately struck at I'ight angles, 
 the globe, which weighs upwards of 40 tons, Avill be less 
 affected by the shock than a heavy anvil by the blow of 
 a hammer. Cousequentl}', a cast-iron shot would crumble 
 to pieces, whilst an exploding shell would strew the arched 
 deck with harmless fragments. 
 
 During contest the revolving turret should be kept in 
 motion, the port-holes being turned away from the opponent 
 except at the moment of discharge, which, however, should 
 be made during full rotation, as the lateral aim in close 
 quarters requires Ijut little precision.*'" 
 
 * Captain Coles, of the British Navy, having claimed priority of invention, the 
 following statement was published in various nautical and mechanical journals (1863) : 
 
 " Absurdity of Captain Coles's Claim. — Captain Coles states, in a letter to the 
 Times of April 5, 1863, that his experience in the Baltic and Black Seas, in 1855, 
 suggested to him the idea of buUding impregnable vessels, and that, towards the 
 latter part of that year, he had ' a rough model made by the carpenter of the Strom- 
 boli,' and that he ])roposed to protect the guns by a stationary shield or cupola. 
 Captain Coles, it appears, met with no encouragement from the Admiralty, and there- 
 fore consulted Mr. Brunei, the celebrated engineer, who warmly embraced the plan. 
 ' He did more,' says Captain Coles in his letter to the Times : ' he assisted me in my 
 calculations, and gave me the aid of his draughtsmen.' Captain Coles further states 
 that, notwithstanding oflficial neglect, he persevered, and in March, 1859, produced 
 drawings of a ' shield fitted with turn-tables.' Lastly, in December, 1860, Captain 
 Coles published in Blachvood's Ilagazine drawings of his ' gun-shield and revolving 
 platform,' the platform being turned by manual power only." 
 
CHAPTER XXIX. 
 
 SURFACE-CONDENSER, OPERATED BY INDEPENDENT STEAM- 
 POWER. 
 
 (SEE PLATE 43.) 
 
 The following is an exact copy of the description accom- 
 panpng the patent granted by the United States (1849) for 
 the independent-action condenser illustrated on the plate re- 
 ferred to : 
 
 Fig. 1 is a longitudinal vertical section ; Fig. 2, a cross- 
 section of the condenser taken at the line (X X) of Fi<y. 1; 
 and Fig. 3, a cross-section of the pumping part of the appa- 
 ratus and the auxiliary engine by which it is operated. The 
 same letters indicate like i^arts in all tlie figiu-es. 
 
 The object of my invention is to condense the steam 
 without admixture with the condeusing-water, that the water 
 produced by the condensation may be carried back to the 
 boiler, to prevent the evil consequences arising from the use 
 of water that contains in solution or suspension mineral or 
 
418 SUBFACE CONDENSATION. chap. xxix. 
 
 other solid matter, and to condense the steam which escapes 
 from the safety-valve, and also for the production of fresh 
 water for any other use. 
 
 In my fresh-water apparatus I use a tubular condenser, 
 through the tubes of which the steam passes, and is con- 
 densed by the cooling influence of a current of cold water 
 taken from the outside of the vessel or ship, and made to 
 pass outside of the tubes ; and to this end the first part of 
 my invention consists in combining the condenser of a steam- 
 engine for the propelling of a ship or vessel with a pump 
 which receives the condensing water from outside the ship or 
 other vessel, and causes it to pass through the condenser, the 
 said pump being operated, irrespective of the engine that pro- 
 pels the vessel, by means of an auxiliary engine, whereby the 
 amount of condensation can be regulated independently of the 
 working of the engine that propels the vessel. The second 
 part of my invention consists in connecting the condenser with 
 the boiler or boilers, or any part thereof, in addition to its 
 or their connection with the exhaust of the engine, when the 
 pump which carries the condensing water through the con- 
 denser is operated by an auxiliary engine, by means of which 
 double connection not only is the steam that escapes from 
 the safety-valve condensed to be carried back to the boiler, 
 but the boiler or boilers may be used to distil and produce 
 fresh water for any purpose desired when the engine is 
 not employed for propelling the vessel. And the last part 
 of my invention consists in connecting the tubes of the con- 
 denser with the cylinder or outer case thereof by connecting 
 
CHAP. XXIX. SURFACE CONDEKSArioy. 419 
 
 oue or botli of the diapliragins to wliicb the ends of the 
 tubes are secured with the outer cylinder or case by means 
 of a ring and flange, or the equivalent thereof, so that the 
 said ring or flange may bend to adapt itself to the unequal 
 contraction and expansion of the tubes and cylinder or outer 
 case of the condenser. 
 
 In the drawings on Plate 43 (a) represents a horizon- 
 tal cylinder, within which are arranged a series of small 
 parallel tubes (h). One end of the said tubes is secured, in 
 the usual way or any other desired and appropriate manner, 
 to a diaphragm (c), which has a turned flange through which 
 rivets or bolts ('/) pass to secure it to the cylinder (a), and 
 within such distance of the ' head as to leave a sufl5cient 
 space between it and the head (e) of the cylinder for two 
 chambers (/) and (g), these two chambers being separated 
 by a horizontal diaphragm or partition (h). The other ends 
 of the tubes are in like manner secured to another diaphragm 
 (/) at the other end, which said diaphragm, instead of being 
 bolted directly to the end of the cylinder in the usual way, 
 is bolted to a ring (j) near its outer periphery, the inner 
 peripheiy thereof being provided with a turned flange bolted 
 to the end of the cylinder ; l)ut, instead of this, the end of 
 the cylinder may be made with a flange corresponding in 
 size and form with this ling, and the diaphragm bolted to 
 its outer periphery. The said ring or flange should be 
 slightly conical, or bent, that the diaphragm may be at some 
 distance from the end of the cylinder, that it may move in 
 and out to adapt itself to the unequal contraction and ex- 
 
420 SURFACE CONDENSATION. chap. xxix. 
 
 pansiou of the tubes and cylinder by reason of the passage 
 of the steam through the tubes and the water for the con- 
 densation through the cylinder. A chamber (Jc) is formed 
 at this end of the cylinder by means of a head (/) secured 
 to the diaphragm by means of a double-flanged ring (??;) and 
 screw-bolts, that it may be removed to give access to the 
 tubes. 
 
 The upper chamber (/), at the end of the cylinder first 
 described, communicates, by means of a pipe («-), in any de- 
 sired manner with the exhaust-pipe of the engine, and, by 
 another pipe (/?'), also with the escape-pij^e of the boiler, 
 and these connections shoiild be governed by appropriate 
 cocks or valves, so that either can be opened or closed at 
 pleasure. Either of these connections being opened, the 
 steam passes into the chamber (/), thence tlirough tlie 
 range of tubes above the diaphragm or partition (Jt) to the 
 chamber (Jc) at the other end, and thence back through the 
 lower range of tubes to the lower chamber (g), which com- 
 municates by means of the pipe (o) with the air-pump and 
 supply-pumps of the engine, or, this connection being closed, 
 by means of a pipe (o') with any desired recipient with 
 which the pipe (o') may communicate. The direction of the 
 passage of steam, and the water produced by its condensa- 
 tion, through the tubes, is indicated by the dotted ai'rows. 
 
 The steam, in passing through the tubes, is condensed by 
 the cooling influence of a constant current of cold water 
 which passes outside of the tubes, and which travels in a 
 direction the reverse of the current of steam, as indicated 
 
CHAP. XXIX. SrRFAC£ CONDENSATION. 421 
 
 by the white arrows, so that the steam a.s it parts with its 
 caloric is constantly approaching a cooler medium. 
 
 The water for the conclensatiou is forced into the cylinder 
 (<a), near the diaphragm (c), through a pi[)e {j)), and passes 
 around the lower half of the series of tubes until it strikes 
 the other diaphragm (<) ; thence it passes up around the end 
 of a horizontal position-plate (<2) on the same plate (/c), 
 which plate (q) extends from the diaphragm (c) to within 
 a short distance of the othei' diaphragm (/), and from this 
 the water passes around all the upper half of the tubes to 
 the firat, where it escapes at the top through a ])ipe (/) 
 that discharges through the side of the vessel above the 
 water-line. 
 
 The water from the condensation is impelled through the 
 condenser by a rotating pump, the case (s) of which is pro- 
 vided with a tangential pipe {t) at the lower part connected 
 \nth the part (/)) by the condenser. And this case is also 
 provided with another pipe ('/) which extends from the 
 centre thereof, to and through the side of the vessel, and 
 so far down as to be always below the water-line, that the 
 water may flow through it to the inside of the pump-case. 
 To the centre of this case is adapted the shaft (^•), the 
 Journals of which run in appropriate boxes («' ir) in the 
 case, and pi-ovided with stuffing-boxes to prevent the escape 
 of water; and on this shaft is a hub (x), Avith four arms 
 or vanes (y) accurately fitted to the case, and yet to rotate 
 without touching it. By the rotation of these arms or vanes 
 the water is drawn in near the centre, and by centrifugal 
 
422 SUBFAVE CONDENSATION. chap. xxix. 
 
 force carried out through the tangential pi^^e (;;) to and 
 through the condenser. And the required rotation of the 
 pump is given by an engine («') secured to the casing of 
 the rotary pump as represented in the drawings, and the 
 connecting-rod (h'), which is jointed in the usual manner to 
 the cross-head {c'), takes hold of a crank (J') on the shaft 
 of the pump, the said shaft being, in the usual manner, pro- 
 vided with an eccentric (e') for working the valves of the 
 engine (a'), which are not represented, as they may be on 
 any of the known plans. The water-supply pump, which 
 receives the water from outside the vessel, and which is for 
 that purpose l)elow the water-line, is provided ^vith a valve 
 (/'), the stem (y') of which passes through stuffing-boxes, 
 and has a handle (/i') by means of which the pipe can be 
 closed at pleasure when it becomes necessary to give access 
 to the inside of the pump. 
 
 From the foregoing it will be seen that, by means of the 
 auxiliary engine which operates the pump, a constant current 
 of cold water is carried through the condenser independently 
 of the working of the propelling engine of the vessel, and, 
 as a necessary consequence, the more the propelling engine 
 labors, by reason of head winds or rough watei-, the more 
 perfect will be the condensation and the vacuum thereby 
 produced, thus Increasing the power of the propelling engine 
 when the power is most needed; whereas if the current of 
 cold Avater were dependent on the working of the proj)elling 
 engine, the sum of the mass of water passing thi'ough the 
 condenser would be exactly in j)i'oportion to the motion of 
 
CHAP. XXIX. SURFACE CONDENSATION. 423 
 
 the engine, and therefore the coudeusation and vacuum would 
 be decreased in the ratio of the decreased motion of the 
 propelling engine. 
 
 It will also be seen that, by reason of the working of 
 the pump which impels the water for the condensation by 
 means of an auxiliary engine, and the double connection of 
 the condenser with the waste-pipe of the boiler or boilers, 
 and with the exhaust of the propelling-engine, whenever the 
 safety-valve is opened, the steam issuing therefrom, instead 
 of being wasted, will be carried through the condenser and 
 condensed, to be returned to the boiler, thus avoiding the 
 necessity of a separate supply of water to make up for the 
 waste by the escape of steam from the safety-valve ; and 
 that when the propelling-engine is at rest the condenser 
 can be used for the distillation and production of fresh 
 water for any desired purpose on board ship, for the con- 
 denser is thus, when desired, rendered entirely independent 
 of the propelling-engine. 
 
 By passing the current of steam in a direction the reverse 
 of the current of condensing-water, the greatest amount of 
 caloric is extracted with the least amount of water. 
 
 The condensing-water in its passage through the condenser 
 never reaches the point of evaporation, and therefore mineral 
 and other matter held in solution will not be deposited to 
 encrust the apparatus ; and by ensuring a constant and rapid 
 current of water around the tubes the danger of unequal 
 contraction and expansion is reduced to the smallest amount, 
 and so small as to prevent all injurious effects by the mode 
 
424 8UBFACI! CONDENSATION. chap. xxix. 
 
 above described of connecting one of the diaphragms, to 
 wliich one end of the tubes is attached, with the cylinder 
 by means of the conical or bent ring or flange. 
 
 Although I have described the use of a rotary pump, 
 operated by a reciprocating engine, for impelling the con- 
 densing-water throi;gh the condenser, I do not wish to 
 confine myself to the use of either a rotary pump or a re- 
 ciprocating engine for this piii'pose, as a rotary engine may 
 be substituted for the reciprocating, and a reciprocating 
 pump for the rotary ; but I have described and represented 
 this arrangement as the one which I have successfully es- 
 sayed and deem the best. 
 
CHAPTER XXX. 
 
 THE CALORIC ENGINE. 
 
 A1>PLICATI0N OF HEATED AIR AS A JIOTOK. 
 (SEE PLATES 44 AND 45.) 
 
 Engineers are aware tliat I built a caloric engine in 
 London, 1833, operated by heated atmospheric air; Faraday, 
 Ure, and Lardner taking great interest in the same in con- 
 sequence of its being based on the principle of returning, 
 at each stroke of the working piston, the heat not convei-ted 
 into mechanical work during the previous stroke. After my 
 arrival in this country, 1839, I prosecuted the plan and built 
 several caloric engines in succession, all of which promised 
 ultimate success. At each step the dimensions were enlarg- 
 ed, until I produced an experimental engine, in 1831, having 
 two working cylinders of seventy-two inches diameter, two 
 feet stroke, and two compressing cylinders of fifty-eight inches 
 diameter (see illustrations on Plates 44 and 45). The lead- 
 ing feature of this large caloric engine was that of cii'culating 
 
426 THE CALOBIG ENGINE. chap. xxx. 
 
 the heated air, as it passed oft' from the Avorkiug cylinder, 
 throiio'h a series of Avire discs coutaiiiiuo- au aow'eii'ate of 
 13,520,000 meshes for each woi'king cylindei'. The cold air 
 iu entering the engine was admitted through the meshes of 
 the heated discs, taking up nearly the whole of the heat 
 previously imparted by the exhaust air iu its passage through 
 the meshes, on its way to the atmosphere. 
 
 DESCRIPTION OF THE ILLtTSTEATIONS KEFEBEED TO.* 
 
 Fig. 1 represents a ti'ansverse section, and Fig. 2 a longi- 
 tudinal section of the engine. 
 
 a, aii'-receiver. h h, supply-cylinder, e', self-acting valve 
 for letting air into, and e" self-acting valve for letting air 
 out of, the same, c, supply-piston ; g\ piston-rod of the same, 
 connected to the working-beam of the engine, d d, work- 
 ing-cylinder ; d' d', holes at the junction of the two cylinders, 
 through which the atmospheric air passes in and out freely. 
 e e, working-piston ; d" d", rods connecting the two pistons 
 together, e'", air-tight vessel, suspended below the working- 
 piston, filled with clay and charcoal to prevent transmission 
 of heat from below, f f, regenerator ; f, discs of wire-net, 
 placed vertically in the regenerator-box. g, valve, Avoi'ked 
 by the engine, for admitting air into the regenerator and 
 working-cylinder ; h, valve for letting air out of the same. 
 i i, pipe, open to the atmosphere, for carrying off the air after 
 having passed through the engine ; h, fire-place. 
 
 The operation of the engine is briefly as follows : A slow 
 
 * Copied from Appleion's Magazine of 1853. 
 
cn.vr. XXX. THE CALORIC ENGINE. 427 
 
 fire being kept up at h for alxiut two lioiiis, until the 
 various parts contained within the hriek-woi'k isliall liave 
 become moderately heated, the air-receiver is charged by 
 means of a haud-pnmp. As soon as the internal pressure 
 shall have reached about six pounds to the scpiare inch — 
 invai'iably effected in less than two minutes — tlie liand-pump 
 is stopped, and the valve g opened by a starting levei-, as 
 in steam-engines; the compressed air from the receiver, thus 
 admitted under the valve g, rushes through the partially 
 heated wires /' into the working-cylinder, forcing its piston 
 e upwards, as also the supply-piston c, by means of the con- 
 necting-rods d" d". The atmospheric air contained in the 
 upper part of cylinder h will, by this upward movement of 
 the supply-piston, l)e forced througli the valve e" into the 
 air- receiver. When the working-piston has reached three- 
 fourths of the full stroke, the valve g is closed by the engine ; 
 and when the piston has arrived at the full np-stroke, the 
 valve li is opened. A free coininniiication with the atmo- 
 sphere being thereby established by means of the open pipe 
 i i, the air under the w^orking-piston passes off, and, owung 
 to the removal of pressure under the working-piston, it will 
 instantly begin to descend by its ow'u ^veight. 
 
 The heated air from under the working-piston, in pass- 
 ing off through the ^vires f, gives out its caloric to the 
 same so effectually that, on reaching the thermometer in, the 
 temperature never e.vceeds that of the entering air at 7 b}' ' 
 more than 30°; on the other hand, the cold air from the 
 receiver, in circulating through the meshes of w^ires in its 
 
428 TEE GALOBIC ENGINE. chap. xxx. 
 
 passage to tlie working cylinder, becomes so effectually 
 heated that, on passing n, its temperature is invariably 
 increased to upwards of 450° wlien tlie machine is in full 
 operation. 
 
 It is evident that during the descent of the supply-piston 
 G the outlet valve e" remains closed by the pressure from the 
 receiver, whilst the inlet valve e' is kept open by suction, 
 and hence that a fresh quantity of atmospheric air enters 
 the supply-cylinder at each down-stroke of its piston, and 
 by the up-stroke is forced into the receiver. There being 
 two supply-cylinders of alternating action, a constant supply 
 of fresh air into the receiver is obtained for feeding the 
 working-cylinders. 
 
 It need hardly be stated that the smaller quantity ob- 
 tained by the supply-cylinder suffices to fill the larger 
 capacity of the working-cylinder, in consequence of the in- 
 crease of volume attending the increase of temperature ; nor 
 need it be stated that an equal amount of force is exerted 
 by the up-and-down movement, as there are two pairs of 
 cylinders attached at opposite ends of a common working- 
 beam. 
 
 The foregoing description being deemed sufficient to ex- 
 plain the mechanical operation of the engine, the result of 
 its prolonged trial may now be considered ; but, before doing 
 so, it will be well to state some particulars in relation to 
 the regenerator. The regenerator measures 26 inches in height 
 and width internally ; each disc of Avire contains G76 superfi- 
 cial inches, and the net has 10 meshes to the inch ; each 
 
CHAP. XXX. THE CALOEIG EXGINE. 429 
 
 superficial inch, therefore, contains 100 meshes, which, multi- 
 plied by 676, gives 67,600 meshes in each disc; 200 discs 
 being employed, it follows that each regenerator contains 
 13,520,000 meshes, and, consequently, if we consider that 
 there are as many small spaces hehceen the discs as there 
 are meshes, we shall find that the air within the regene- 
 rator is distributed in 27,000,000 minute cells. Theory 
 clearly indicates that, owing to the small capacity for heat 
 of atmospheric air (that beneficial property which the Great 
 Mechanician gives to it as a fit medium for animated loann 
 beings to live in), and in consequence, also, of the almost infi- 
 nite subdivision among the wires, the temperature of the 
 circulating air, in passing through the regenerator of the 
 caloric engine, must be greatly changed. Practice has fully 
 realized all that theory predicted, for the temperatures at x 
 and z have never varied during the trials less than 350°, 
 when the engine has been in full operation ; indeed, it has 
 been found impossible to obtain a differential tenijjerature of 
 less magnitude, with sufficient fii'es in the furnaces. 
 
 The reason is evident : the cold air from the receiver is 
 half the time playing upon the wire discs at x, whilst the 
 heated air from the working-cylinder is playing during the 
 other half on the wire discs at z ; as no heated air can 
 reach the former without passing througb the regenerator, 
 and as no cold air can reach the discs at z before likewise 
 passing all the ^dres, it follows that the establishing an equi- 
 librium of temperature becomes impossible. The great num- 
 ber of discs, their isolated character, anil the before-named 
 
430 THE CALORIC ENGINE. chap. xxx. 
 
 distribution of tlie air in sucli a vast number of minute cells, 
 readily explain the surprising fall and increase of tempera- 
 ture of tbe opposite currents passing the regeneratoi', and 
 'which constitutes the grand feature of the caloric engine, 
 effecting, as it does, such an extraordinary saving of fuel by 
 rendering the caloric not converted into mechanical Avork 
 active over and over again. 
 
 In further explanation of the wonderful efficiency of the 
 regenerator, it may be stated that each disc contains 1,140 
 feet of wire in length, and each regenerator 228,000 feet, 
 or 41^ miles, of Avire; the superficial measurement of which 
 is 2,014 square feet, Avhich is equal to the entire surface of 
 four steam-boilers forty feet long and four feet diameter; 
 and yet the regenerator displaying that amount of heating 
 surface is only two feet cube, less than rsVir of the bulk of 
 said boilers ! 
 
 In ]'egard to loss of Jieat, the result of ample trial has 
 been that at no time has the temperature of the escaping 
 air at m exceeded that of the entering air at I by more than 
 30°. As this differential temperature exhibits the positive 
 loss of Iieat, it becomes important to ascertain its amount in 
 pounds of coal : the area of the supply-piston is 2,626 square 
 inches, and its stroke two feet ; hence 36to cubic feet of 
 atmospheric air is supplied for each stroke, and therefore 
 at 30 strokes 1,092 cubic feet, and for both cylinders 2,184 
 cubic feet per minute = 131,040 cubic feet per hour. The 
 Aveight of atmospheric air is nearly 13i cubic feet to the 
 pound, and hence it Avill be seen that 9,706 pounds of air 
 
CUAI'. XXX. TUE CALOUIC EiiGlKE. 431 
 
 pass tlirougli the engine every liuiir. We kuo\v tliat one 
 pound of coal will raise tlie temperature of 10 pounds of 
 water 1,100', while the specific heat of water is to that of 
 the air as I'li : 100; hence it will be seen that 38iV pounds 
 of air will be elevated in temperature 1,100° with one pound 
 of coal. Now, the observed loss of heat in the engine being 
 30°, the fact will be established that the loss will amount 
 to one pound of coal for every 1,408 pounds of air passed 
 through the engine, \vhich, on 9,706 pounds, proves the 
 actual loss of heat in both regeneratoi-s to be only 6^u pounds 
 of coal 2)er hour. A pressure of 13 pounds being sustmned 
 in the receiver, exerting GO horse-power with an actual waste 
 of only G.8 pounds per hour, it \vill be found that Iwo 
 ounces of coal per hour per horse-power is the quantity of 
 fuel absolutely wasted in the process of transfer. The actual 
 consumption of the engine is, however, nearly 4:0 pounds 
 per hour, which is thus proved by the foregoing to be chieily 
 carried off by radiation of heat. On a large scale much of 
 that radiation \v\\\ be prevented. As the machine stands, an 
 indicated horse-power is produced by a consumption of less 
 than 11 ounces to the horse-power per hour. 
 
 The following particulars are of considerable practical 
 importance : 
 
 Ist. The valves g and h are not subjected to heat, the calo- 
 ric being taken up by the wires before reaching the valves. 
 
 2d. The temperature of the packing of the working-pistons 
 does not exceed boiling heat at any time, proving the efficacy 
 of the heat-intercepter e'". 
 
432 TEH CALOEIC ENGINE. chap. xxx. 
 
 3d. As only a slow radiating fire is needed, it has been 
 found that common whitewash, applied to the under side of 
 the heater, remains for several weeks, proving conclusively 
 that the effect of the heat is quite harmless. 
 
 4th. A hole of half an inch diameter, kept open for 
 several hours, in the valve-chest, under the inlet-valve g, 
 does not sensibly aflfect the pressure in the receiver a, so 
 abundant is the supply of air. This fact has surprised all 
 practical men who have witnessed the operation of the en- 
 gine. It proves completely that the machine need not be 
 perfectly air-tight, as supposed by many. 
 
 5th. After putting a moderate quantity of fuel into the 
 furnace, it has been found that the engine works with full 
 power for three hours without fresh feed, and, after remov- 
 ing the fires entirely, it has frequently worked for one hour. 
 
 The regularity of action and perfect working of every 
 part of this experimental engine, and, above all, its apparent 
 great economy of fuel, induced some enterprising merchants 
 of New York, in the latter part of 1851, to accept my pro- 
 position to construct a ship for navigating the ocean propelled 
 by paddle-wheels actuated by the caloric engine. This work 
 was commenced forthwith, and pushed with such vigor that 
 within nine months from commencing the construction of the 
 machinery, and within seven months from laying the keel, 
 the j)addle-wheels of the caloric ship Ericsson tui'ued round 
 at the dock ! In view of the fact that the engines consisted 
 of four working-cylinders of 168 inches diameter, 6 feet 
 stroke, and foiir air-compressing cylinders of 137 inches dia- 
 
CHAP. XXX. 
 
 THE CALORIC ENGINE. 
 
 433 
 
 meter, 6 feet stroke, it may be claimed that, in point of mag- 
 nitude and rapidity of construction, the motive machinery of 
 the caloric ship stands uin-Ivalled in the annals of marine 
 engineering. It may be added that the principal eugineei's 
 of New York all expressed the opinion that a better speci- 
 men of ^vorkIllalls^lip than that presented by the huge 
 engines of the caloric ship had not been produced by our 
 artisans up to that time. 
 
 The following data, published in Appletons' Meclicmic^ 
 Magazine, will interest the professional reader: 
 
 DnCENSIONS OF THE ERICSSOM', 
 
 Length on deck, 
 
 260 feet 
 
 Length of keel, ..... 
 
 250 " 
 
 Breadth of beam, 
 
 40 " 
 
 Depth of hold, .... 
 
 27 " 
 
 Draught of water on trial-trip. 
 
 17 " 
 
 Diameter of wheels, ..... 
 
 32 " 
 
 Length of bucket, 
 
 lOi " 
 
 Breadth of bucket, .... 
 
 20 ins. 
 
 Dip of wheel (supposed about), 
 
 2 feet 
 
 ENGINES. 
 
 
 Number of working-cylinders or single-acting 
 
 air-engines, ...... 4 
 
 Diameter, . 168 inches. 
 
 Area of piston, ..... 22167.07 sq. in. 
 Stroke, 6 feet. 
 
434 
 
 THE GALOEIO ENGINE. 
 
 CHAP. XXX. 
 
 Portion of stroke from commencement at 
 
 which air is " cut off " (about), . . t^ 
 
 Cubical contents of each working-cylinder, 1596024 cub. in. 
 Cubical contents of t^ of working-cylinder, 995495 " 
 Number of suppl3'-c}linders or single-acting 
 
 pumps, 4 
 
 Diameter, ...... 
 
 Area of jjiston or plunger, 
 
 Stroke, necessarily, ..... 
 
 Cubical contents of each pump. 
 
 Number of regenerators, 
 
 Number of discs of iron-wire netting in each 
 
 regenerator, 
 
 Height of each disc, .... 
 
 Width, 
 
 Size of wire, ...... 
 
 Eatio of area of openings in the netting to 
 
 total area of disc, ..... 
 Total area of opening of " air- way " through 
 
 n n- 6X4 
 each disc, = 12 sq. ft. 
 
 Greatest or total heat .of air in working- 
 cylinder above atmosphere, . . . 384° F. 
 
 Heat of issuing air above atmosphere, . . 30° F. 
 
 Pressure necessary to move the engine, . i lb. 
 
 Coal consumed in the four furnaces per day, 6 tons. 
 
 Maximum coal possible to consume in the 
 
 four furnaces per day, .... 7 " 
 
 137 
 
 inches 
 
 . 14741 
 
 sq. in. 
 
 •6 feet. 
 
 1061352 
 
 cub. in 
 
 4 
 
 
 1 
 
 50 
 
 
 6 
 
 feet. 
 
 4 
 
 a 
 
 tV 
 
 incli. 
 
 i to 1 
 
CUA1>. XXX. 
 
 TUi: CALORIC EyuLSE. 
 
 435 
 
 Number of smoke-pipes, ..... 2 
 
 Number uf alr-pipet<, ..... 2 
 
 Height oi each smoki- ami air pipe above deck, 1-* feet. 
 
 Diameter " " " " 30 inches. 
 
 Amount of air passing through the four cylin- 
 ders per houi', ...... 50 to 75 tons. 
 
 Depth of ^vorking•piston, or thickness, . . 6 feet. 
 
 Thickness of cylinder-bottom, . . . 1^ inches. 
 
 Distance of grate from bottom of cjdinder, . 5 feet. 
 
 Ordinary pressure of the engine per sq. incli, 12 lbs. 
 
 Actual pressure on the second trial-trip, Jan- 
 uary 11, • . . 8 " 
 
 Number of revolutions under pressure of 
 
 8 lbs. on trial trip, ..... 9 
 
 Number of revolutions expected with 12 lbs. 
 
 pressure, . . . . . . . 12 
 
 Miles per hour obtained on trial-trip, Jan- 
 uary 11, allowing for tide, etc., . . 7 
 
 Miles per hour expected with 12 lbs. pressure, 10 to 12 
 
 Number of meshes in each disc, . . . 500000 
 
 Temperature in working-cylinder, 60° + 384° = 444° 
 
 Common temperature of the atmosphere (usual 
 
 assumption), 60° 
 
 Specific heat of air — water being 1000, . . .2669 
 
 Common pressure of air per square inch ~ 
 
 14.73 lbs., say 15 lbs. 
 
 Weight per cubic foot, common pressure and 
 
 temperature = .0752914 ll>.s., say . . 1*3 lb. 
 
436 THE OALOBIO ENGINE. chap. xxx. 
 
 Density of air, temperature remaining con- 
 stant, is directly as tlie pressure. 
 
 Weight per cubic foot, at 12 lbs. pressure, 
 
 common temperature, .... t¥j 
 
 Expansion of air at 32° for each degree add- 
 ed, according to Rudberg, . . . ri? 
 
 Dalton and Gay-Lussac, .00208 ; Regnault, rh ; 
 
 common estimate, .... rh 
 
 Expansion of air at 60°, for the 384° added, |f| 
 
 Density of air at (60° + 384° =) 444°, com- 
 
 n . n • o 508 
 
 pared with air at 60 , as . . . „„ , to 1 
 
 ^ 508 + 384 
 
 Weight of a cubic foot at 12 lbs. pressure, 
 
 temperature 444° = M X AV = . . tVWA lb. 
 
 Weight of 995495 cubic inches at 12 lbs. pres- 
 sure, temperature 444° = HffF X tWs^o = 46 lbs. 
 
 Therefore, air passed through each cylinder 
 
 each stroke, . . . . . , 46 " 
 
 Weight of 1061352 cubic inches, at common 
 
 temperature and pressure = ^ Wh^i ^ X ^ — 47.153 lbs. 
 
 Therefore, air passed each pump each stroke, 47 " 
 
 Allowance made for clearance, leakage, etc., 
 
 per stroke, 47 — 46 = . . . . 1 lb. 
 
 Units of heat required to raise 47 lbs. air 384° 
 = 47 X 384 X .2669 = . . . . 
 
 Units of heat retained by the 47 lbs. on es- 
 caping = 47 X 30 X .2669 = . . . 
 
 Units of heat transferred each stroke, , 
 
 4817 
 
 units. 
 
 376 
 
 11 
 
 4441 
 
 li 
 
CIIAP. XXX. 
 
 THE CALORIC ENGINE. 
 
 437 
 
 Absolute theoretical consumption of heat per 
 stroke, per cylinder, ..... 
 
 Mean pressure, per square inch, on ^vorkiug■ 
 piston, allowing for continued addition of 
 heat while expanding, initial pressure being 
 13 lbs., about 
 
 Mean force acting upon working-piston, 10.8 X 
 22167 = 
 
 Mean resistance per square inch to supply pis- 
 ton, commencing with and increasing to 
 12 lbs., at which pressure it continues to 
 end of stroke. 
 
 [The mean resistance in compressing an elas- 
 tic fluid may be found by reversing the 
 ordinary calculation on expansive working. 
 
 The hyp. log. of M is .588] 1.588 X 12 X H = 
 
 Mean resistance against supply-piston 10.55 X 
 14741 = 
 
 Balance tending to move the engine, 239403 
 - 155598 = 
 
 Units of power theoretically obtainable per 
 stroke = 83805 X G = 
 
 Units of power theoretically obtainable from 
 each unit of heat, 
 
 37G units. 
 
 10.8 lbs. 
 
 239403.6 " 
 
 10.55 lbs. 
 
 155598 " 
 
 83805 " 
 
 502830 units. 
 
 1337 
 
 The ship after completion made a successful trip from 
 New York to Washington and back during the winter season ; 
 but the average speed at sea proving insufficient for com- 
 
438 THE CALOIUC EXGINE. chap. xxx. 
 
 mercial purposes, the owners, v^'ith regret, acceded to my 
 proposition to remove the costly machinery, although it had 
 proved perfect as a mechanical combination. The resources 
 of modern engineering having been exhausted in producing 
 the motors of the caloric shij?, the important question has 
 for ever been set at rest : Can heated air as a mechanical 
 motor compete on a large scale with steam ? The commercial 
 world is indebted to American enterprise — to New York 
 enterprise — for having settled a question of such vital im- 
 portance. The marine engineer has thus been encouraged 
 to renew his efforts to perfect the steam-engine, without 
 fear of rivalry from a motor depending on the dilatation of 
 atmospheric air by heat. 
 
 The engines of the caloric ship being an exact counter- 
 j)art of the experimental engine of 1851, excepting dimensions, 
 a description has been deemed superfluous. It may be men- 
 tioned, however, that the pair of engines in the caloric ship 
 actuated a single crank in the middle of the paddle-shaft 
 by connecting-rods working at right angles on a common 
 crank-pin, as in all my marine engines. 
 
CHAPTER XXXI. 
 
 CALOKIC ENGINE FOR DOMESTIC PUEPOSES. 
 
 (SEE PLATE 46.) 
 
 Although the caloric engine has proved inapplicable to 
 navigation, it has been found to be of very great utility as 
 a domestic motor, and for all pui-poses demanding a small 
 amount of motive power. 
 
 The following interesting article from the New York Tri- 
 hune of May 5, 1860, shows how rapidly the caloric engine 
 was adopted after its adaptation to domestic purposes : 
 
 ■ THE NEW MOTOR. 
 
 "It is some eighty-six yeai-s since ^Ir. Boulton, at the 
 great steam-engine works of Soho, made use of the memorable 
 expression to Eoswell : ' I sell here, sir, what all the world 
 desires to have — Poweu.' The mechanical world has been 
 occupied from that time to the present \vith this problem 
 of power, and mechanical ingenuity has tasked and exhausted 
 
440 TEE DOMESTIC CALORIC ENGINE. CHAP. XXXI. 
 
 itself witli efforts to construct a macliine that sliould prove 
 an efficient auxiliary or rival of the steam-engine. And it 
 is most extraordinary that, notwithstanding the amount of 
 inventive genius and science that has been expended in this 
 special field of labor, literally nothing had been accomplished 
 of any practical importance till Ericsson produced the caloric 
 engine, in the particular form and with the peculiar devices 
 which distinguish it from all the engines actuated by heat, 
 that have been built at such an enormous expense of time 
 and money. 
 
 " Motive engines of a moderate or even of a small power 
 play a very important part in the economy of human life. 
 The frightful horrors of the slave-trade ; the scarcely less 
 frightful hoi'rors of the traffic in Coolies; nay, the haggard 
 features and jaded limbs that, in our great cities more espe- 
 cially, speak so distinctly of over-wrought human labor, and 
 cry out so emphatically for relief — all these demonstrate that 
 a compact, manageable, safe, and economical motor, adequate 
 to the work of a single slave or Cooly, or overtasked white 
 man or white woman, would do more to mitigate the suffer 
 ing and diminish the drudgery of mankind than any other 
 conceivable invention. After all the enormous accumula 
 tions of steam-power, water-power, wind-power, and horse 
 power, and their vast achievements, by how much the larger 
 amount of power exercised in the world is the aggregate 
 result of individual force applied to the thousands of little 
 things that occupy the human family in the daily routine of 
 living! Combine these forces, and what a stupendous whole 
 
CHAP. XXXI. THE DOMESTIC CALORIC EXdlNE. 441 
 
 tliey exhibit ! Make au availal)le motor that shall be of 
 oue-inan power, and what a result is obtained ! Make a 
 motor pei-fectly safe, easily kept in order, requiring no water, 
 and consuming but little fuel, of the power of a single horse, 
 to what an extent the aggregate result is augmented, and 
 what an importance in human affairs such a machine assumes ! 
 
 " If Ericsson's caloric engine, then, claimed to be nothing 
 but such a motor, it would be a subject well deserving the 
 most earnest and serious investigation ; but the proof is accu- 
 mulated, of a nature that compels belief and defies contra- 
 diction, which demonstrates the existence in this engine of 
 a power entirely sufficient for all but a very few of the 
 thousand uses for which power is requii'ed. 
 
 " It is not material to our pur2:)ose to indulge in any 
 retrospective review of Ericsson's labors. It is well known 
 that this grand invention has occui:)ied thirty years of his 
 life, during which he has built many engines of the largest 
 size and uncounted experimental engines of smaller power. 
 
 "We have seen an official statement in relation to an 
 engine put up about a year since to supply the locomotives at 
 the South Groton Station, on the Fitchburg Railroad. From 
 April, 1859, to April, 1860, this engine pumped 1,600,000 gal- 
 lons of water, at an expense to the company for fuel and oil 
 of $25, and for an ' engineer ' $25, and has not cost one cent 
 for alteration or repairs. 
 
 "A result more irapoitant, in view of the number of 
 engines employed, is exhibited on the New York Central 
 Railroad, on the line of which there are now some twenty 
 
442 THE DOMESTIC CALORIG ENOTNE. chap. xxxi. 
 
 of ttese engines in daily use. Mr. Chaimcey Vibbard, the 
 Superintendent of that road, reports, over his official signa- 
 ture, after several months' experience with a number of these 
 engines, that they peiform an ' incredible ' amount of labor 
 'for the small quantity of fuel consumed.' One of them, he 
 says, for t\V of a cent per hour, does the woi'lv formerly 
 done by foxir men at an expense of $25 each per month. 
 Another, of the same size, at the Savannah Station, at an ex- 
 pense of eleven cents a day, does the work of five men ^vho 
 received $125 a month. Other engines have been erected 
 on several other I'ailroads for pumping purposes with the 
 same favorable result. 
 
 "The second application of the caloric engine was to 
 the driving of printing-presses. The first trial of the engine 
 for this purpose was made in the office of the Hartford 
 Itnies ; the first that was entirely successful was made in 
 the office of T. W. Strong, No. 98 Nassau Street, in this 
 city. The next engine built was set up in the office of 
 Messrs. French & Wheat, No. 18 Ann Street, and the third 
 in the office of Mr. C. C. Shelley, a job-printer in Barclay 
 Street. The result has been the adoption of the engine in 
 numerous job-offices in every part of the country. There are 
 now no less tlian forty daily papers in the United States 
 printed by Ericsson's engines, most of them of 24-inch, but 
 three or four of 12 and 18 inch cylinders. One of the most 
 recent testimonials to its value is fi-om the proprietor of the 
 Savannah Evening Express, who states that lie regards it as 
 the most perfect and economical motive power ever applied." 
 
CUAP. XXXI. TIIK DOMESTIC CALORIC EXGIXE. 443 
 
 Several tLousand caloric engines were subsequently con- 
 structed in this country and in Europe ; but steam-engineers, 
 finding by the extraordinary demand for caloric engines that 
 very moderate power A\as a great desideratum, have pei-fected 
 the steam-motor until it almost rivals the caloric engine iu 
 sjxfety and adaptability ; consequently, the demand for caloric 
 engines has been greatly diminished of late. Yet this motor 
 can never be superseded by the steam-engine, since it requires 
 no water, besides being absolutely safe fi'om explosion. There 
 are innumerable localities in which an adequate quantity of 
 water cannot be obtained, but \vliere the necessities of civi- 
 lized life call for mechanical motors ; hence the caloric engine 
 may be regarded as an institution inseparable from civiliza- 
 tion. It should be stated that the caloric engine has been 
 found to furnish the only rt'liable motive power for operating 
 the fog-signals on our coasts. The following statement, pre- 
 sented to the Light-House Board, sets forth very clearly the 
 advantages of the air motor for the purpose mentioned : 
 
 With reference to the important question whether steam 
 is a proper motive power for actuating the mechanism con- 
 nected with fog-signals, I beg to express the opinion that 
 unless a safer motor can be found than the steam-engine — 
 more particularly a high-pressure steam-engine — the great 
 practical benefit which you expect from the contemplated 
 system of fog-signals will never be realized. My reason for 
 expressing this opinion will be found in the following brief 
 summary: 1. A high-pressure steam-boiler, even when sup- 
 plied w ith pure, fresh water, is an apparatus which demands 
 
444 THE DOMESTIC CALOIilC ENGINE. CHAP, xxxi, 
 
 the constant attention of an experienced person. Considering 
 that any neglect in keeping up the feed in the boiler will 
 inevitably result in an explosion, it is highly imprudent and 
 scarcelj- humane to put such an instrument in the hands of 
 a light-house keeper. Let us reflect on the well-known fact 
 that when a boiler foams even the practised engineer is 
 sometimes at a loss to determine the height of water within. 
 2. Apart from the difficulty and danger thus alluded to, 
 another circumstance presents itself connected with the em- 
 ployment of steam, which is practically insuperable, viz., 
 that brackish or salt ^vater must be resorted to in most 
 localities. Accordingly, unless a certain quantity of the salt 
 water is regularly drawn off and replaced by water less 
 impregnated with saline matter, the boiler will, at best, be 
 rendered useless by the dej)osit formed. In most cases the 
 first warning to the unskilful light-house keeper will pro- 
 bably be the explosion of the boiler. 
 
 In connection with this most important matter, I cannot 
 omit adverting to the fact that the employment of salt 
 water for land engines attended by skilful engineers has 
 been found so impracticable that means of procuring fresh 
 water — in many cases at very great cost — have been deemed 
 indispensable. It is only in the steam-ship, where the most 
 competent engineers are employed, provided with salinometers 
 and other instruments, that it has been found practicable to 
 employ salt water. But even in the steam-ship salt feed has 
 been dispensed with by employing the surface-coudeuser as 
 the only certain means of saving the boilers from incrustation. 
 
CHAP. XXXI. THE DOMESTIC CALOEIC £yGIXE. 445 
 
 3. During cold weather, another serious difficulty will 
 be encountered if you employ steam, which calls for the 
 application of costly and complicated conti'ivances. Unless 
 the boiler is constantly under steam, it must be kept in some 
 place adequately heated by stoves to prevent pumps, pipes, 
 and cocks from freezing. Not oiil}- this, the cistern Itself, 
 which is to supply the boiler, will, on our iuclenient coast, 
 freeze unless warmed by some means. I need scarcely remind 
 you that, in many localities, the entire supply of water will 
 ^vholly fail during continued cold, dry weather. In fine, 
 the disadvantages of steam for any general system of fog- 
 signals are so numerous and formidable as to render the very 
 proposition to employ that agent an absurdity. 
 
 Having thus briefly disposed of the question of employ- 
 ing steam as the motive power for actuating the machinery 
 of your pi'oposed fog-signals, I have now to state that long 
 practice has shown that the expansive force of heated atmo- 
 spheric air furnishes a reliable dr// motor wholly independent 
 of atmospheric temperature. The advantages of such a raotoi-, 
 more especially as it requires no particular kind of fuel, 
 are so obvious that I will not detain you by enumerating 
 the same. Suffice it to say that it enables you to locate 
 your fog-signal on the dry, barren rock as well as on the 
 moist, sandy beach ; and that its efficiency is not affected 
 by the most intense cold, and that, so far from demanding 
 a heating apparatus during the inclement season, the light- 
 house keeper will find it a very desirable accessory in 
 warming his quarters. Above all, while it thus adds to his 
 
446 THE DOMESTIC CALOBIC ENGINE. CHAP. xxxi. 
 
 comfort, it carries no danger with it. The worst that can 
 ha2:)pen is that the machine Avill stop for Avant of fuel, or 
 that its speed will slacken for want of oil being applied 
 to the bearings. The caloric engine is now so well known 
 that I need not enter on a description of its construction. 
 It will be necessary, however, to advert to the fact that the 
 caloric engine is more bulky and of greater weight than the 
 steam-engine, and that its cost is some 50 per cent, greater. 
 These disadvantages, however, as regards the application to 
 fog-signals, become trifling, in view of the before-named ad- 
 vantages. Indeed, in many places the cost of procuring a 
 suital)le supplj^ of water will be far greater than the differ- 
 ence of price of engine — leaving out of sight the impossi- 
 bility of procuring suitable water in many cases. 
 
 The leading features of the domestic caloric engine will 
 be seen by reference to Plate 46, i-epresenting a longitudinal 
 section through the central vertical plane. Professor Bar- 
 nard having very thoroughly examined one of these engines 
 at the Paris Exhibition, 1867, I propose to present a copy 
 of his re})ort : 
 
 " In its present form the Ericsson engine fails to present 
 to the observer a combination at first view easilj^ intelligible. 
 It even seems to be characterized by a certain amount of 
 complication, which might suggest greater liability to derange- 
 ment than ought to belong to a prime mover. A closer 
 examination, nevertheless, will show that the mechanism 
 itself is in fact very simple, and that it is only the rather 
 puzzling consecution of movements which confuses. 
 
CUAP. XXXI. THE DOMESTIC CALOlilC EXGINE. 447 
 
 "Before referring to the figure of this engine, which is 
 given in the illustnition on the phite mentioned, tlie following 
 general explanation of the mechanical principles of its con- 
 struction will be understood. Let it be supposed that a 
 piston moves air-tight in a cylinder which is closed at both 
 ends. Call one end of the cylinder A, and the other B. 
 Call the piston also C. In the end A let there be a valve 
 opening inward, and in the end B a second valve opening 
 outward. These two valves open, then, in absolute direction, 
 the same \vay. Let the piston C, furthermore, have a valve 
 opening in this common direction. Then, if the piston C 
 move toward B, its own valve will naturally close, and that 
 of B will open, because the movement tends to compress 
 the air between B and C. Also the valve A will open at 
 the same time, because the movement tends to rarefy the 
 air between A and B. Thus, in this movement, continued 
 to the end of the cylinder, all the air on the side toward B 
 may be expelled; but at the same time the cylinder ■will 
 be filled on the other side toward A, l)y the influx of air 
 from without. If the piston C now revei-se its motion, both 
 the valves A and B will be closed, because the movement 
 will tend to rarefy the air on the side of B, and to con- 
 dense it on the side of A. But its own valve will be 
 opened by the joint effect of these causes, so that the air 
 will pass freely thiough the piston, and, if the motion con- 
 tinues, will ultimately be all transferred to the side of B. 
 This operation may go on iudefinitely. 
 
 " Now, if, on the side of A, the cylinder is closed by a 
 
448 THE nOMESTTC CALORTC ENGINE. chap. xxxi. 
 
 second piston (wliicli we may still call A), and not by a 
 fixed cap, both pistons being movable, tlie same succession 
 of occurrences will take place, only modified by the move- 
 ments which may be given to A. If C and A both move 
 in the direction of A, both their valves will open, and air 
 from the exterior of the cylinder will pass throiigh both 
 into the space between B and C. If they both move toward 
 B, but C faster than A, then air will enter on the side of 
 A, and flow out on the side of B, the valve C only remain- 
 ing closed. If both move toward A, but A faster than C, 
 air will still enter the space between C and A, while, in 
 less quantity, it is passing through C into the space between 
 C and B. 
 
 " Let now the piston A be supposed to occupy a posi- 
 tion, say, one-third advanced down the cylinder, the piston 
 C being further advanced still, and let the valve of B be 
 secured by a strong sjDring pressing upon it, so that it can- 
 not be opened without the application of some considerable 
 force ; and in these circumstances let the cylinder, and 
 consequently the air contained in it, be heated. The elas- 
 ticity of the confined air, being increased by heat, will close 
 the valve in A, and that piston will be moved in the direc- 
 tion of A, until, by the enlargement of volume, the elasticity 
 shall be reduced to equality with that of the external air. 
 If the heat be uniform throughout all the mass of confined 
 air, the valve in C will be equally pressed on both sides. 
 Under these circumstances, the piston C could be moved 
 toward A, if there were any means of acting upon it, the 
 
CUAP. XXXI. THE DOMESTIC CALOBIC EXaiXE. 449 
 
 air passing tlirougli the valve toward B. Bnt if au attemjit 
 were made to move the pistou itself toward B, it would 
 encounter resistance, because its own valve would be closed 
 by the movement, and the valve of B is supposed to be 
 forcibly held down. Since now the external piston must 
 move in the direction A, it is only necessary that it should 
 be propei-ly connected with a machine, in order that the 
 force exerted by the heated and expanding air may be turned 
 to some practical account. 
 
 " If, ao-aiu, at the end of the movement the air could be 
 immediately cooled without being discharged, the heat could 
 be again applied and the effort repeated. But this not 
 being practicable, the heated air may be allowed to escape 
 by relieving the valve B of the pressure of the spring which 
 confines it, and by causing the piston C to descend to the 
 extremity B of the cylinder. This movement of C not only 
 drives out the hot air, but it draws in through A a fresh 
 supply of cold air ; and if A descends simultaneously to the 
 position originally supposed — i.e., one-third advanced toward 
 B — there will be a body of air filling the other two-thirds 
 of the cylinder at the common temperature, ready to be 
 acted on anew by heat. 
 
 " In this statement is embraced the general principle of the 
 Ericsson engine. What remains is to explain the mechanical 
 contrivances by which the movements of the pistons are 
 governed, and to describe the heating apparatus which is 
 employed to effect the prompt dilatation of the aii-. Inas- 
 much as the piston which we have called C is shut up in 
 
450 THE DOMESTIC CALOBIC ENGINE. chap. XXXI. 
 
 the cylinder beliliul A, it is necessary that the rods which 
 give it motion should pass through A. They do so, being- 
 packed by means of stuffing-boxes to prevent leakage; and 
 are connected at their external extremities with oscillating 
 levers turning on a fixed centre of motion at their extremi- 
 ties, and kept in motion by the engine. The rod of the 
 external piston A, which is the driving-piston, is also con- 
 nected with an upright oscillating lever, turning on an axis 
 of motion at its lower extremity, and carrying at its upper a 
 horizontal connecting-rod, which acts on the crank of the main 
 shaft of the engine. It would be simpler to connect the piston 
 directly with this crank ; biit if that mode of connection 
 were adopted, the stroke of the piston would have to take 
 place in both directions, forward and back, in equal times. 
 This condition is not favorable to the action of the machine ; 
 and inequality in this respect is still more important in the 
 case of the supply-piston. The peculiar ingenuity of this 
 machine is in fact manifested most signally at this point. 
 By means of the systems of levers interposed between the 
 pistons and the main shaft, provision is made for the per- 
 fect uniformity of the revolution of the shaft, while the 
 pistons, on the other hand, are accelerated and retarded in 
 such a manner as to fulfil the condition that the aspiration 
 of the charge of air should occupy the minimum of time. 
 The oscillating levers which connect Avith the piston-rods of 
 the stipply-piston are kept in oscillation by crank-motion 
 from the main shaft, and in their oscillations they displace 
 the inner piston, encountering no resistance but friction. In 
 
CHAP. xxxr. THE DOMESTIC CAUWIV EXGIXE. 461 
 
 consequence of tlie un-unifoiin ;uul unequal velocities of the 
 two i;)istons, and their intentional adjustment, so that they 
 do not begin and end tlieir course together, the distance 
 between them varies in a manner which is quite important: 
 first, to the aspiration of the charge ; and secondly, to the 
 effectual exposure of the aspired air to the action of the 
 furnace. 
 
 " It is of coui-se of the highest importance that the posi- 
 tions of tlie cranks on the main shaft, and those of the 
 axes of motion of the oscillating levers, should be so related 
 to each other as to produce a rapid separation of the two 
 pistons at the beginning of the negative stroke; because this 
 is the time when the aspiration of the charge must take 
 place. During this time, the inner piston, gaining on the 
 outer, will not only draw in the fresh charge, but it will 
 expel the exhausted one; the escape-valve being lifted foi- 
 the purpose and kept raised during all the period of aspira- 
 ti(in by means of a cam. When the pistons are at the 
 maximum distance from each otlier. the as])iration is ended. 
 From this time until tlie half revolution is complete, the 
 confined air undergoes compression, and the movement is 
 maintained by the fly-wlieel. In the second half revolution 
 the driving-piston is urged by the elasticity of the air 
 whicli is exalted both by conqiression and by heat. 
 
 " The heating is accomplished as follows : The furnace is 
 within the cylinder, at the end which we have called B, 
 where the cylinder is prolonged to receive it. It is of iron, 
 and is cylindrical also, a small annular space only inter- 
 
452 THE DOMESTIC CALOBIO ENGINE. chap. xxxi. 
 
 vening between its Avails and those of the cylinder. This 
 space is open to the interior, but is closed at the extreme 
 end ; so that it forms, in fact, a portion of the proper air- 
 chamber. To the supply-piston C is attached by its crown a 
 sheet-iron cylindrical bell, which enters the annular space Just 
 spoken of without touching the walls of the furnace or those 
 of the surrounding cylinder. The valve in C opens above 
 the crown of this bell ; but any air which comes through 
 the valve from the side of A can only reach the interior by 
 passing down the annular space between tlie bell and the 
 cylinder Avail, and returning up the annular space between 
 the bell and the wall of the furnace. In making this pas- 
 sage, it Avill be exposed in a very thin sheet to the action 
 of the furnace heat, a very large proportion of the molecules 
 being brought into direct contact with the heated iron. 
 
 " That we may understand how this movement of the air 
 is made forcibly necessary, we need only consider the rela- 
 tive movements of the pistons during the period of a com- 
 plete revolution. At the beginning of the negative stroke, 
 or of the movement of A in the direction of B, the supply- 
 piston takes the lead, air enters through the valve of A, 
 and the asioiration is soon complete. The distance between 
 the two pistons, which determines the amount of aspiration, 
 is now of course at its maximum. A next begins to gain 
 on C, but both movements have still for a short time the 
 same (negative) direction. The space occujDied by the air 
 is gradually reduced; or, in other words, the air undergoes 
 compression. The piston C reaches the limit of its course 
 
CHAP. XXXI. THE DOMEHTIC CALOBIC LXGIXE. 453 
 
 sooner than A. It begins to move in the positive direction, 
 while the motion of A is still negative. The valve in C is 
 opened by the pressure, the air passes through, and, having 
 no other channel, descends the aiuiular space outside of the 
 bell, and returns by the annular space inside the bell, becom- 
 ing heated, as above described, in its progress. Presently 
 after this displacement coniniences, the piston A also reaches 
 its limit of movement, and the direction of its motion becomes 
 positive. But C moves the faster of the two, so that the 
 displacement continues throughout the greater part of the 
 positive stroke. A little before the eiul, the distance between 
 the two pistons becomes minimum, and they are then nearly 
 in contact. ^Tien the revolution is quite complete, this dis- 
 tance is slightly increased. Just before this time C will 
 have recommenced its negative movement, Avhile A continues 
 still to be moving in the positive direction. 
 
 " The relative movements here described will be more ad- 
 vantageously compared by presenting them in tabular form, 
 which we are enabled to do by the help of the determinations 
 made by Mr. Mastaing, of Paris, upon the Ericsson engine, 
 which was made the subject of experiment in 1861 at the 
 Conservatoire des Arts et Metiers, by Mr. Tresca, sub-director 
 of that institution. In the first column of this table are 
 placed the angular positions of the driving-crank on the 
 main shaft at different periods of the revolution; putting 
 zero to rejiresent the position of the craidv when the piston 
 A is about to commence its negative stroke. The second 
 column gives the direction of motion of the driving-piston, 
 
454 
 
 TEE DOMESTIC CALOEIC ENGINE. 
 
 CHAP. XXXI. 
 
 and its motion relative to that of tlie other ; and the third 
 column gives the same particulars in regard to the supply- 
 piston. The last column gives the variation of distance taking 
 place bet^veen the two pistons at the several points indicated 
 in the table. 
 
 Angular posi- 
 tion of the 
 cranli. 
 
 Relative motion of the pistons. 
 
 Distance 
 
 between the 
 
 pistons. 
 
 Driving-piston. 
 
 Supply-piston. 
 
 Degs. 
 
 Oto 70 
 
 Negative, losing . . . 
 
 Negative, gaining . . 
 
 Increasing. 
 
 70 
 
 Negative, equal. . . 
 
 Negative, equal . . . 
 
 Maximum. 
 
 70 to 120 
 
 Negative, gaining. . 
 
 Negative, lo.sing. . . 
 
 Decreasing. 
 
 120 
 
 Negative, gaining. . 
 
 Limit of course .... 
 
 Decreasing. 
 
 120 to 170 
 
 Negative, contrary. 
 
 Positive, contrary. 
 
 Decreasing. 
 
 170 
 
 Limit of course .... 
 
 Positive, gaining. . 
 
 Decreasing. 
 
 170 to 310 
 
 Positive, losing. . . . 
 
 Positive, gaining.. 
 
 Decreasing. 
 
 310 
 
 Positive, losing. . . . 
 
 Positive, gaining. . 
 
 Minimum. 
 
 310 to 340 
 
 Positive, gaining . . 
 
 Positive, losing. . . . 
 
 Increasing. 
 
 340 
 
 Positive, gaining . . 
 
 Limit of course .... 
 
 Increasing. 
 
 340 to 360 
 
 Positive, contrary . 
 
 Negative, contrary. 
 
 Increasing. 
 
 " It will be seen that the negative stroke is completed in 
 less than half a revolution foi' either piston, while the posi- 
 tive stroke requires more ; also, that this inequality is con- 
 siderably greater for the supply-piston than for the driving- 
 piston. In the case of the di'iving-piston the inequality is 
 as 170 to 190 deg. ; in that of the supply-piston, as IGO to 
 200 deg. These inequalities, which could not exist if the con- 
 nection between the main shaft and the pistons were made 
 directly, as in the steam-engine, are the effect of the interme- 
 
CHAP. XXXI. THE DOMESTIC CALoh'JC EXGiyE. 455 
 
 diate system of levers, aud arc iutentionally produced. The 
 increase of distance betAveeii the pistons from 310 deg. to 
 the end of the revolution is not an advantage, but it is not 
 a great increase, the total distance amounting finally only to 
 about the one-sixth part of the maximum sepai-atiou, and re- 
 ceiving the principal accession to its amount between 350 and 
 3G0 deg. As, after the second reversal of the niovemeut of 
 the supply-piston, the effective power of the engine is neces- 
 sarily paialyzed, the escape valve is opened at 344 deg. by the 
 action of the cam above spoken of, and the aspiration com- 
 mences before the revolution is quite complete. The valve 
 is closed again at 09 deg., just as the aspiration is becoming 
 maximum. 
 
 "Inasmuch as the effective pt)wer of this engine is negative 
 or zero from 344 deg. onward to 170 deg., or through a little 
 more than half a revolution, it is necessary that the machine 
 should be provided witli a heavy fl}-wheel to maintain the 
 movement during these intervals. Tlie tl3•-^vheel is made to 
 act also as a sort of counterweight, as well as by means of 
 its moment of rotation, the side of the wheel which is de- 
 scending during the peiiod of paralysis being made consider- 
 ably heavier than the other. A companion engine, to act 
 positively during tlie inaction of the first, would render such 
 an exjtedient unnecessary ; but, unfortunately, the bulk is 
 considerable relatively to the power, and it would, in gen- 
 eral, be a disadvantage to double it. 
 
 " The engines of Ericsson are largely in use in the United 
 States, but as yet they have not been constructed of any 
 
456 THE DOMESTIC CALORIC ENGINE. chap. xxxi. 
 
 considerable power. As a general rule, tliey fall within 
 three or four liorse-power as an outer limit, tliouyli it is be- 
 lieved that there have been made some exceeding this limit. 
 On account of their safety and convenience they have been 
 regarded with favor ; and it has been claimed for them as an 
 additional recommendation that they are economical. Such 
 did not appear to be the fact in the case of the particular 
 engine which was the subject of the experiments of Mr. 
 Tresca above referred to. In this machine, which was of 
 two horse-power, the result of very careful ti'ial showed a 
 consumption of 4.13 kilograms (about nine pounds) of coal 
 per horse-power per hour. ■•' In comparison with steam, this 
 cannot be called a large economy. The consumption of a 
 good steam-engine ought not to exceed, per horse-power per 
 hour, two kilograms at the outside. One and a half ought 
 to suffice. 
 
 "It may be observed, in conclusion, that Ericsson makes 
 no attempt to carry the temperature in this engine to a very 
 high point. The mean maximum temperature in the experi- 
 ments at the Conservatoire did not exceed 270° Fahrenheit, 
 though doubtless portions of the air received a greater de- 
 gree of heat than this. The expansion of volume was further 
 determined to be but as 1 : 1.48 — that is to say, about fifty 
 per cent, of the original bulk. 
 
 * There is not a single instance on record in the United States in which the con- 
 sumption has exceeded four pounds of anthracite coal an hour per horse-power. It 
 should be observed, however, that by forcing the combustion by an excess of fuel, put 
 into the furnace under an imperfect draught, the consumption may be more than 
 doubled. 
 
CUAI'. XXXI. THE DOMESTIC CALORIC ENGINE. 457 
 
 " The general description here given will be made more 
 intelligible by reference to the figures of the engine given 
 in Plate 46. 
 
 " Of the two pistons shown at A and F, the first, A, is 
 the driving-piston, and the second the supply-piston, which 
 in the foregoing explanation we have called C. In A is 
 seen a valve marked a. 
 
 " At B is an axis of motion, the office of which is to com- 
 municate movement to the piston A, by means of a crank o, 
 a connecting-rod p, a second crank q, and another rod /■. 
 
 " In the piston F the valve of communication is sliown at 
 f. The solid portion F' is filled with plaster, or other badly- 
 conducting substance, while F" marks the bell-shaped pro- 
 longation which extends into the annular space surrounding 
 the furnace. A\"lien, l^y the approach of the piston F to the 
 piston A, the space between these two pistons is reduced, 
 there is no escape for the air between them but that which 
 is afforded by the annular cavities between this bell and the 
 external wall of the machine /', on the one IkuuI, and the 
 wall of the fuinace itself on the other. The air passes first 
 along the outer space to the mouth of the bell, and returns 
 thi'ough the inner, forming a thin stratum in immediate con- 
 tact witli the hot wall of the furnace. 
 
 " Another axis of motion is shown at C, of which it is 
 the office to communicate movement to the sujiply-piston F, 
 through the crank o, the connecting-roti s, and the cranks i 
 and ", which last two are fixed to the arbor C, at a fixed 
 anirle to each other of seven detirees. 
 
458 THE DOMESTIC CALOBIC ENGINE. chap. xxxi. 
 
 " The escape-valve is placed at D, and kept iu position 
 l)y the spring d. A cam D', acting on this valve through the 
 lever D", opens it Just before the driving-piston commences 
 its descent at the end of the positive stroke. 
 
 " The furnace is enclosed in the iron box G, the grate- 
 bars being shown at g. G' indicates plates of iron designed 
 to protect the walls of the fni'nace. 
 
 " In order to bring the two j)istons into a favorable posi- 
 tion for starting, the fly-wheel is turned on its axis ; and, 
 for the purpose of facilitating this operation, the ai'bor K 
 is introduced, which enables the attendant to act on the fly 
 by means of the clicks marked Ic, and the notches h'. 
 
 " The furnace-door I is made double to reduce loss by radi- 
 ation. The walls of the furnace are similarly protected by 
 means of a double envelope. 
 
 " The products of combustion escape from the furnace 
 through the flues li, protected by fire-brick, and are carried 
 oft' by the chimney H." 
 
 The caloric engine thus described was patented 1858, the 
 Rumford medal being awarded in 1862 for its successful 
 practical application. 
 
 The following address by Professor Horsford, on present- 
 ing the medal, cannot properly be omitted in this ^vork : 
 
 " At the time the vote of the American Academy of 
 Arts and Sciences conferring upon you the Eumford Pre- 
 mium was passed, I had the honor to be Chairman of the 
 Rumford Committee, and, you will remember, signified my 
 wish to relieve myself of the trust imposed upon me ; but, 
 
CHAP. XXXI. THE DOMESTIC CALORIC EXGINE. 459 
 
 as this formal act aud tlie simple cerumouy appropriate to 
 it have been postponed in consequence of the pressure of 
 the war, in which you, sii-, have borne so conspicuous a part, 
 the custody of the vote and medal has been continued w ith 
 me to the present time. 
 
 "1 have the honor now to place in your hands a certi- 
 fied copy of the vote passed by the Academy at its annual 
 meeting, June 10, 1862. It is jis follows: 
 
 " ' Voted, That the Rumford Premium be awarded to 
 John Ericsson, for his improvements in the management of 
 heat, particularly as shown in his caloric eugine of 1858.' 
 
 " In now handing to you the gold and silver medals 
 which have been prepared in accordance with the statutes 
 of the Academy, I beg to congratulate you upon the honors 
 you have ^von through a life of research and experiment, de- 
 voted to the promotion of the prosperity and well-being of 
 mankind, in the field contemplated by the illustrious founder 
 of the Rumford Premium."* 
 
 * A patent was granted June 15, 1809, to a German engineer, tor a machine 
 actuated by heated air, identical in principle and mechanical combination with my 
 caloric engine of 1858, the only difference being that, like the solar engine delineated 
 on Plate 07, aud other air-engines constructed by me in the United States at various 
 times — as far back as IS-tJ — the patented engine uses the same air over and over. 
 The patent refen-ed to also embraces a supposed novel plan of applying a water- 
 chamber round the open end of the cylinder for cooling the same. Now, the leading 
 feature of a large caloric engine built by me at the Delaraater Iron-Works, in Xew 
 York, 185f>, was that of cooling the open end of the cylinder by such a wnter- 
 chambcr. The patented engine, therefore, is a flagrant plagiarism on devices already 
 carried into practice. 
 
CHAPTER XXXIL 
 
 THE MONITOR SYSTEM OP IRON-CLADS. 
 
 (SEE PLATES 47 AND 48.) 
 
 The monitor system is thus noticed in J. Scott Russell's 
 great work on Naval Architecture : 
 
 "It is a creation altogether original, peculiarly American, 
 admirably adapted to the special purpose which gave it 
 birth. Like most American inventions, use has been al- 
 lowed to dictate terms of constriiction, and purpose, not 
 prejudice, has been allowed to rule invention. 
 
 "The I'uling conditions of construction for the inventor 
 of the American fleet were these : the vessels must ' be per- 
 fectly shot-proof, thej' must fight in shallow water, they 
 must be able to endure a heavy sea, and pass through it, 
 if not fight in it. 
 
 "The American iron-clad navy is a child of these con- 
 ditions. Minimum draught of water means minimum extent 
 of sui-face, protected by armor ; perfect protection means 
 
 460 
 
CHAP. XXXII. TEE MONITOR SYSTEM OF lEOX-CLADS. 461 
 
 thickness to resist the heaviest shot, and protection for the 
 \\hole length of the ship; it also means pei-fect protection 
 to guns and gunners. Had they added \vhat our legislators 
 exact — that the ports shall lie in the ship's side, nine feet 
 above the water — the problem might at once have l)ecome 
 impossible and absurd ; but they wanted the work done as 
 it could be done, and allowed tlie conditions of success to 
 rule the methods of construction. 
 
 "The conditions of success in the given circumstances 
 wei-e these: that you slu)uld not require the sides of the 
 ship to rise much above the water's edge ; that you should 
 not require more protection to the guns than would contain 
 guns and gunners ; that you should be content with as many 
 guns as the ship could carry, and no more. 
 
 "To do the work, thereft)re, the full thickness of armor 
 required to keep out the enemy's shot was taken, but the 
 ship was made to rise a feAv inches above watei-, and no 
 more ; and so a narro\v stiip of thick armor, all along the 
 upper edge of the ship's side, gave her complete protection. 
 Thus the least quantity of thickest armor did most work 
 in protecting the ship, engines, boilers, and magazine. Next, 
 to protect the guns, a small circular fortress, shield, or tower 
 ciK'ircled a couple of guns, and, if four guns were to be car- 
 ried, two such turrets carried the armament and contained 
 the gunners. Thus, again, weight of annor was spared to 
 the utmost, and so both ship and armament were completely 
 protected. 
 
 " But the consequences of these conditions are such as 
 
462 THE MONITOR STSTEM OF IROK-CLABS. chap, xxxii. 
 
 we, at least for sea-going ships, would reluctantly accept. 
 The low ship's side will, in a sea-way, allow the sea to 
 sweep over the ship, and the waves, not the sailors, will 
 have possession of the deck. The American accepts the 
 conditions, removes the sailors from the deck, allows the 
 sea to have its way, and drives his vessel through, not over, 
 the sea to her fighting destination, by steam, abandoning 
 sails. The American also cheei-fully accepts the small round 
 turret as protection for guns and men ; and pivots them on 
 a central turn-table in the middle of his ship, raising his 
 port high enough to be out of the water, and then fighting 
 his gun through an aperture little larger than its muzzle. 
 
 "By thus frankly accepting the conditions he could not 
 control, the American did his work and built his fleet. It 
 is beyond doubt that the American Monitor class, with two 
 turrets in each ship, and two guns in each turret, is a kind 
 of vessel that can be made fast, shot-proof and sea-proof. It 
 may be uncomfoiiable, but it can be made secure. The sea 
 may possess its deck, but in the air, above the sea, the 
 American raises a platform on the level of the top of his 
 turrets, which he calls his hurricane deck, whence he can 
 look down with indifference at the waves fraitlessly foam- 
 ino- and breaking themselves on the abandoned deck below. 
 His vessel, too, has the advantage, as he thinks it, of not 
 rolling with the waves ; so that he can take his aim steadily 
 and throw his shot surely. Thus, if he abandons much that 
 we value, he secures what he values more. 
 
 '' I think I have reason to know that the American turret 
 
cJiAr. xxxu. TRE MOMTOIi UlSrEM OF lEOS-CLADS. 463 
 
 ships, of the larger class, with two turrets and four guns, 
 are successful vessels — successful beyond the measure of our 
 English estimate of their success. Like so luauy American 
 inventions, they are severely subject to the conditions of 
 use, and successful by the rigidity and precision with which 
 they fit the end and fulfil the purpose which was their aim. 
 
 "Plate 47 contains side elevation, deck plans, and cross- 
 section of the original American Monitor of Caj)tain Ericsson 
 — the first turret ship that distinguished herself in action, 
 having to engage with her single turret and pair of guns a 
 large broadside ship of much lieavier tonnage and armament, 
 which she thoroughly defeated. 
 
 '' Captain Ericsson, the builder of the Monitor, has long 
 been distinguished equally in England and America. He 
 was known as the builder and designer of one of the most 
 remarkable engines, in the original competition, preliminary 
 to the opening of the Liverpool and Manchester Railroad; 
 he was afterwards distinguished in the introduction of the 
 screw-propeller in steam navigation ; and he has crowned his 
 career by the successful construction of the class of turret 
 ships, which appear to have been taken up with avidity, and 
 prosecuted with energy, by the American Government ; and 
 during the coui-se of their sad civil w.ir the 'monitors' appear 
 to have rendered to the Federal side very important services. 
 The design of these vessels has about it all the character- 
 istics of American audacity. Every conventionality of the 
 ship has been despised and di.scarded ; in the sailor's sense 
 of the word, there is nothing 'sliijKshaiie ' about this original 
 
4C4 THE MONITOE SYSTEM OF IBON-CLABS. ouap. xxxii. 
 
 Ilonitor ; e\'ery thing is unusual. She has neither keel, nor 
 bilges, nor bulwarks. She is very nearly a London bridge, 
 covered by a great horizontal platform of timber, projecting 
 Ijeyoud her deck, and descending below the \vater-line. This 
 great upper platform in no way conforms to the shape of 
 the under-ship which carries it ; it is obviously meant to 
 shelter the rudder and the stern from every attempt to 
 damage them by collision. At the bow the entire hull is 
 equally protected by the overhanging platform of the deck, 
 and the whole upper works of the ship are covered with 
 thick iron armor on both sides, and the wooden deck is pro- 
 tected by iron plates. The rudder is a balanced rudder, and 
 the ship is propelled by a single screw ; the boilers are the 
 double-tier boilers, of the ordinary construction, with four 
 sets of flues. It will be noticed that the arrangements of 
 the turret are very different from Ca2)tain Coles's arrange- 
 ments. The whole turret is on the upper deck, exposed to 
 shot ; it is not carried on a revolving set of rollers, but is 
 pivoted on the centre, which seems to carry most of its 
 weight by means of an iron trussing, from ^vllich it is, as it 
 were, suspended, and it slides on a smooth metal plate lying 
 on the deck. The turret is worked by a small pair of donkey 
 engines, working on tooth gear, and the ports are covered 
 by hanging blocks. Like our turret,* the Monitor shield has 
 two guns worked parallel to each other on slides. The man- 
 
 * The English, in abandoning the cupola of Coles, and copying the monitor 
 turret, also adopted the term turret. For some time, however, the English naval 
 architects adhered to the word cupola ; but in a short while the phrase cupola was 
 dropped, hence "turret ship'' in place of "cupola ship." 
 
CUAP. XXXII. THE MO.MTOR SiHTEM OF IL-OX-CLADS. 4(55 
 
 ner in wliich these turrets were afterwards improved and 
 matured by experience is shown in Plate 49, and it is certain 
 that Captain Ericsson rendered great service to liis country 
 by inventing at once, and successfully introducing, a class 
 of vessels peculiarly suited to action in their inland waters 
 and sluillo\v navigations ; and when we consider the extreme 
 rapidity which attended the execution of the project, we 
 must say that the original Monitor was a remarkable success, 
 :ind that she was a type of an entirely new class of war- 
 shi[i." 
 
 The origin of the name " monitor " calls for an explana- 
 tion in this place. The Navy Department at "Washington 
 having, shortly before the launch, requested me to suggest 
 an appropriate name for the impregnable turreted steam- 
 battery, I addressed a letter to the Assistant Secretary of 
 the Navy, sajing : " The impregnable and aggressive char- 
 acter of this structure will admonish the leaders of the 
 Southern Rebellion that the batteries on the Ijaiiks of their 
 rivers will no longer pi'esent barriers to the entrance v{ the 
 Union forces. 
 
 "The iron-clad intruder will thus prove a severe monitor 
 to those leaders. But tliere are other leaders who Avill also 
 be startU'd and admonished by the booming of the guns from 
 the impregnable iron turret. 'Downing Stieet ' will hardly 
 view with indifference this last ' Yankee notion,' this monitor. 
 To the Lords of the Admiralty the new craft will be a 
 monitor, suggesting doubts as to the propriety of completing 
 those four steel ships at three and a half millions apiece. 
 
466 TKE MONITOB SYSTEM OF IliON-CLABS. CHAP. XXXil. 
 
 " On these and many similar grounds I propose to name 
 tlie new battery Monitor.'''' 
 
 It Avill be recollected that this letter was regarded in 
 England as possessing political significance, several membei'S 
 of Parliament having called for its reading in the House of 
 Commons when the news of the result of the battle between 
 the Monitor and the Merrimack appeared in the Times. 
 Uufpiestionably, the advent of the Monitor materially coun- 
 teracted the pressure which the French Emperor bi'ought to 
 bear on the British Ministry at the time, in favor of the 
 Southern States. 
 
 John Bourne, the greatest authority on naval engineering 
 of our time, in a critical examination of the monitor system 
 published in London, 1866, observes : 
 
 " The confidence of the Americans in the shot-proof quali- 
 ties of their monitors is manifested by many of the incidents 
 of the late Eebellion, one of Avhich is that Captain Worden, 
 of the monitor Montcmlc, attacked and destroyed the Con- 
 federate vessel Nashville, w^hen lying under the guns of Fort 
 McAllister, in Georgia ; and, although the fort Avas all the 
 time pouring a fire upon the monitor from its heaviest guns, 
 the monitor took no notice of it, but proceeded without 
 interruption to the destruction of her antagonist. Another 
 neAV feature in naval war is that, in the attack on Fort 
 Fisher, the fire of the rest of the fleet was directed against 
 the fort over the monitors ; and although shot falling short 
 and shells prematurely exploding could not be prevented in 
 such an engagement, the monitors, it was felt, M-ere able to 
 
CHAP. XXXII. THE MOMTOi; Sl'STFil OF IROS-CLADS. 
 
 407 
 
 encounter such risks with impunity. The monitors, during 
 two years of active service, in all weathers, on a hostile and 
 stomiy coast — sometimes Avatching for blockade-runners in 
 Cuba, sometimes engaged in the Gulf of Mexico, and often 
 at sea in heavy gales — were, on an average, each twenty-five 
 times in action : being a larger amount of service than that 
 of any vessels recorded in history. Shot could not damage 
 them ; storms could not swamp them ; and at the end of 
 the war they wei-e as effective as at the beginning. The 
 following extract from a report of Admiral Dahlgren will 
 show something of the kind of service in whicli some of 
 the monitors were employed during three months in the 
 summer of 1863, and the number of shots they fired and 
 witli impunity received : 
 
 Nftme of Monitor. 
 
 No. of shots fired. 
 
 Hits. 
 
 
 Hits at 
 Ogeechec. 
 
 Total hits 
 
 received 
 
 from 
 
 the enemy. 
 
 15-inch. 
 
 llinch. 
 
 Catsldll 
 
 Montauk 
 
 Lehigh 
 
 Passaic 
 
 Naliant 
 
 Patapsco . . . 
 "NYeehawken 
 Nantucket . . 
 
 IBS 
 301 
 
 41 
 119 
 170 
 178 
 264 
 
 44 
 
 425 
 478 
 28 
 107 
 276 
 230 
 633 
 155 
 
 86 
 154 
 36 
 90 
 09 
 90 
 134 
 53 
 
 20 
 14 
 
 35 
 36 
 47 
 53 
 51 
 
 40 
 9 
 
 1 
 
 106 
 214 
 30 
 134 
 105 
 144 
 187 
 104 
 
 1,255 
 
 2,332 
 
 718 
 
 256 
 
 56 
 
 1,030 
 
 Mr. Bourne also presents the following extracts from the 
 
468 THE MONITOR SYSTEM OF IBON-GLADS. chap, xxxir. 
 
 reports of Captain Jolm Rodgers, of the monitor Weehatvkeii, 
 to the Secretary of the American Navy : 
 
 (1) June 20, 1863. 
 
 " The opinion fc^rmed then confirmed my anticipations, 
 that a hnll rising l)nt little above the surface of the water 
 (in this case only 16 inches), and having a central elevation, 
 as in the monitors, is the sfcape to form a good sea-boat ; 
 and I am convinced that on this idea all successful iron-clads 
 must be built. This form reduces the surface to be plated 
 to a minimum, and puts -the part having the necessary eleva- 
 tion above the sea for fighting guns where it can be carried 
 without inconvenience, and in the Wechaivke/i is easily carried. 
 With us, I think, safety is solely a question of strength. 
 
 " I had relied upon former experience to correct any faulty 
 motion which I might discover in a sea-way, by shifting or 
 reducing weights. I abandoned, however, the idea of improve- 
 ment. As I watched the action of the vessel it was perfect." 
 
 (2) July 22, 1863. 
 
 "On Thui'sday night, when off Chincoteague Shoals, ^ve 
 had a severe gale from east-northeast, with a ver)- heavy sea, 
 made confused and dangerous by the proximity of the land. 
 The waves I measured after the storm abated. I found them 
 23 feet high. They were certainly 7 feet higher in the 
 midst of the storm. 
 
 "During the heaviest of the gale I stood upon the turret 
 and admired the behavior of the vessel. She rose and fell 
 to the waves, and I concluded then that the monitor form 
 had great sea-going qualities. If leaks were |)rc\-ented, no 
 
ciiAi". xxxii. TUI-: MoMToi: iii:;sTJJM of muy-CLAUS. 409 
 
 Imnioaue could injure her. 1 presume in two days we shall 
 be ready for aii}- service, as we ueed uo repaiis, aud ouly 
 some little fittings." 
 
 " It may be added," says Mr. Bourne, "that, on the occasion 
 of the heavy gale which occurred just before the attack on 
 Fort Fisher, the nionitois were the oidy vessels of the fleet 
 which were able to ride it out without di'agging their anchors; 
 aud on the occasion of a common steamer having been sent 
 to escort a monitor, before confidence had yet been established 
 in the seawoi-thiness of tliat class of vessel, the steamer, hav- 
 ing liroken down in a heavy sea, was taken in tow by the 
 monitor, and was carried by her safely into port." 
 
 Mr. Bourne, in proof of the conifcirt and healthiness of 
 the monitors, likewise presents the following extract from 
 A repoi-t of the Secretary of the United States Navy to 
 Congress : 
 
 "It is gratifying to know that an examination of the sick 
 reports, covering a period of over thiity months, shows that 
 so far from being unhealthy, there was less sickness on board 
 the ' monitor vessels than in the same number of wooden 
 .ships with an e(pial niiniber of men, and in similarly exposed 
 positions. The exemjition from siekness in the iron-dads is 
 in some instances i'emarkal>le. Thei'e were on boaid the 
 SaiKjus, from NovemI)er 25, 1864, to April 1, 1805. a period 
 of over four months, but four eases of sickness (excluding 
 accidental injuries), and of these two were diseases from 
 which the patients had suflVred for years. In the M(>iit<tril; 
 for a period of <^ne hundred and sixty-five days prior to May 
 
470 THE MONITOR SYSTEM OF IR0N-CLAD8. chap, xxxii. 
 
 29, 1865, there was but one case of disease on board. Other 
 vessels exhibit equally remarkable results, and the conclu- 
 sion is reached that no wooden vessels in any squadron 
 throughout the world can show an equal immunity from 
 disease. The facts and tables presented are worthy of care- 
 ful study." 
 
 The following vote of thanks was passed by the Thirty- 
 seventh Congress, March 28, 18G2 : 
 
 " Resolved, by the Senate and House of Kepresentatives 
 of the United States of America in Congress assembled, That 
 it is fit and proper that a public acknowledgment be made 
 to Captain John Ericsson for the enterprise, skill, energy, and 
 forecast displayed by him in the consti-uction of his iron-clad 
 boat, the Monitor, which, under gallant and able management, 
 came so opportunely to the rescue of our fleet in Hampton 
 Roads and, perchance, of all our coast defences near, and 
 arrested the work of destruction then being successfully 
 prosecuted by the enemy with their iron-clad steamer, seem- 
 ingly irresistible by any other power at our command ; and 
 that the thanks of Congress are hereby presented to him 
 for the great service which he has thus rendered to the 
 country." 
 
 It will be proper to mention, also, that several iron-ship 
 builders, in conjunction \vith the proprietors of some of the 
 most important marine-engine establishments on the Atlantic 
 coast, in token of their appreciation, honored me by present- 
 ing a model of the Monitor, weighing upwards of fourteen 
 pounds, manufactured of pure gold. 
 
CHAPTER XXXIII. 
 
 THE MONITOR TURRET AND THE CENTENNIAL 
 EXHIBITION. 
 
 (SEE PLATE 49.) 
 
 The imperfect character of the monitor turret exhibited 
 in Fairmount Park has been noticed with surprise by pro- 
 fessional visitors. In view of the important results attained 
 by the adoption of this structure during the war, it cannot 
 be denied that a painted wooden re^u'esentative is unworthy 
 of the occasion. Nor need it be urged that some turret 
 bearing the marks of actual conflict ought to have been 
 transferred to the Exhiliition. Experts are aware tliat, owing 
 to the laminated character of these turrets, they may readil}' 
 be taken down, and the plates transported and put up at 
 any required distance from the vessel. Besides, it may be 
 shown that this process would have been less expensive than 
 erecting a complete representative turret and mechanism. 
 Regarding the armament applied within the wooden turret, 
 
472 THE MOXITOll TUliBET. chap, xxxill. 
 
 naval artillerists from aljroad, who expected to have had 
 an opportunity of examining the detail of the friction-gear 
 peculiar to the mcinitor armament, have been greatly dis- 
 appointed to find that, instead of the carriages on which 
 the guns wavQ mounted during the war, an experimental 
 steam-carriage devised by an engineer from St. Louis, after 
 the war, has been placed in the representative turret. In 
 consequence of this misleading procedure on the 2)art of the 
 authorities, of excluding the carriages which had been used, 
 the majority of visitors have naturally imagined that the 
 monitor guns, during the war, were mounted on the experi- 
 mental carriage referred to." It should be mentioned that 
 the plan of checking the recoil of our guns by friction has 
 not been superseded in the navy, and that friction gun- 
 carriages are at present being constructed for the Navy 
 Department under my patents. 
 
 THE PILOT-TIOUSE. 
 
 The pilot-house — wheel-house — of the monitors, and the 
 steering-gear which it contains (see Plate 49), forming the 
 most important features of the system, their exclusion from 
 the turret at the Centennial Exhibition must be regarded as 
 an untoward circiimstance. It cannot be supposed that the 
 otfieer charged with the duty of putting up the i-epresenta- 
 tive turret would incur the responsibility of excluding with- 
 out authority that part of the system which most interests 
 
 * My oari'iage, operated by a circular compressor, exhibited in the wooden tuiTet, 
 was applied in the Dunderherg, but not in the monitors during the war. 
 
CHAP, xxxill. TIIK MOMTOR TURRET. 473 
 
 the professional visitor. The country is aware that Adniiial 
 Porter's report to the Seci'etary of the Navy, pultlished 
 whiU- the prei)ai'ations for the Centennial Kxhiliitioii were 
 in progress, contained a reconiniendatiou to abolish the pilot- 
 house of the monitor turrets. Possildy this recommendation, 
 condenniatory of my invention, led to the exclusion of the 
 steering machinery and the massive [li lot-house from the 
 representative turret in Fairmount Park. T do not propose 
 to investigate the cause which has led to the exhibition of 
 my invention in a mutilated state, but I feel called upon 
 to show that the I'emoval of the pilot-house from the turrets 
 recommended by Admiral Porter is highly improper and 
 incompatible with the monitor system. 
 
 Well-informed naval officers are aware that Worden failed 
 to sink the MerrimacTc at IIami)ton Roads because he could 
 not personally control the firing and at the same time direct 
 the steering of his vessel from a point enabling him to ob- 
 serve properly the movement of his antagonist. This fact 
 was well understood by the authorities at Washington, the 
 Assistant Secretaiy of the Navy having liimself witnessed 
 the battle and the ineffectual firing. I was accordingly re- 
 quested by the Navy Department, shortly after the conflict, 
 to devise some means of steering from the turret. The de- 
 spatch conveying this request contained several important 
 suggestions, and closed with the following sentence : " The 
 placing the wheel-house cm the turi'et woidd double the 
 formidable character of the vessel." Considering what hap- 
 pened at Ilanqiton Roads, more might liave been said ; for 
 
474 THE MONITOB TURRET. chap, xxxiii. 
 
 had the Monitor been provided with a wLeel-liouse on tlie 
 top of the turret, Wordeu, instead of discontinuing the ac- 
 tion almost blinded, would have forced the MerrwiacJc to 
 surrender as readily as Eodgers compelled the commander 
 of the boasted impregnable Atlanta to haul down the Con- 
 federate flag by being enabled pei'sonally to direct the steer- 
 ing and the firing while watching from the elevated turret 
 wheel-house of the monitor Weehaivhen the movements of 
 his opponent. Again, it was from the turi-et wheel-house of 
 the monitor Montauh that the hero of Hampton Roads 
 detected the Kasliville, and by a feAv fifteen-inch shells, fired 
 under his own supervision, burnt the Confederate vessel and 
 cargo. It was reserved for the Admiral of the Navy to 
 discover that the arrangement which enables the commander 
 of a mouitoi' to direct the firing and the steering from an 
 elevated position above the turret, affording an all-around 
 view, is a great mistake, although it has proved so eflicacious 
 in actual conflict. It was reserved for him also to find out 
 " that the placing of the pilot-house on the top of the monitor 
 turret shows a lack of ingenuity " — a discovery apparently 
 resulting from reflections connected with the fact mentioned 
 in his report, "that this is the most exposed point in the 
 vessel, and is liable to be swept away by the first heavy shot 
 that strikes." The Admiral appears to have forgotten that 
 the fleet of monitors at Charleston, commanded by Dahlgren, 
 engaged the Confederate batteries some twenty times, at 
 easy range, and yet the pilot-houses were not " swept away." 
 The section of the turret and wheel-house (pilot-house) 
 
CHAP. XXXIII. THE MONITOR TURRET. 475 
 
 of a monitor of the Passaic class (shown on Plate 49) repre- 
 sents the structure precisely as built, excepting that the 
 turret wall, in order to protect the base of the wheel-house 
 in accordance with my original plan, should be carried two 
 feet above the turret roof. As the wheel-house and steering 
 gear must remain stationary while the turret revolves, it 
 will be perceived that tin; plan [iresents a mechanical prob- 
 lem of no ordinary character. This is understood by persons 
 possessing correct mechanical knowledge, who have studied 
 the arrangement, and know that it successfully passed the 
 severe ordeal to which it ^vas sul)jected during the war. 
 
 With reference to the original plan of extending the 
 turret wall above the roof, it will be proper to mention 
 that the managers of the plate-mills employed to manu- 
 facture the turret plating during the war refused to furnish 
 plates of more than nine feet in length. Not only did they 
 positively decline to roll plates of an additional length of 
 six inches, but they limited the thickness to fifteen-sixteenths 
 of an inch, owing to their inabilitj- to manufacture plating 
 above a given weight. Those who criticise the strength of 
 the armor of the monitors will do well to bear this in mind. 
 A careful inspection of our illustration, representing a sec- 
 tion of the turret of the Passaic class of monitors, at once 
 disposes of the erroneous assertion that the pilot-house, the 
 internal diameter of which is only six feet, may be "swept 
 away." No target practice has yet shown that a cylinder 
 of such small diameter, composed of solid ii'on eighteen 
 inches thick (of course it might be made thicker), can be 
 
476 THE MOXITOB TUHIiET. chap, xxxill. 
 
 penetrated. At the same time, the inertia of such a cylinder, 
 owing to its weight, is so great that a base-ring of vei-y 
 moderate section attached to the turret roof will effectually 
 jii'event dislodgment under the impact of shot. 
 
 In view of the foregoing, it "would be waste of time to 
 discuss tlie merits of the Admiral's I'ecommenclation to the 
 Secretary of the Navj^, " to place the steering apparatus 
 beloAv, with a small portion of the deck above it raised and 
 heavily plated, with apertures to look through." But his 
 reference to lack of strength calls for some notice. In reply, 
 I mil simply observe that by substituting solid for laminated 
 armor, the original If on if oi; "if in existence to-day," would 
 be the most formidable iron-clad of her tonnage possessed 
 by any naval power. Of course it must be attributed to 
 inadvertency that Admiral Porter has not, in his report, 
 reminded the Secretary of the Navy that the laminated 
 armor of the turrets of the entire monitor fleet ought to be 
 at once substituted by solid plating. 
 
 Naval constractors and engineers will find by examining 
 our illustrations that, excepting the omission to place the 
 pilot-house on the top of the turret, the original Monitor 
 was a perfect fighting machine; and that not a single essen- 
 tial improvement has been added in building the subsequent 
 monitor iron-clads. It will also be found that tlie pi'opeller 
 was better protected in the original Monitor, and that the 
 anchor was handled with greater facility and more perfect 
 protection to the crew, tlian in the recent turret vessels. 
 Moreover, it is susceptible of positive demonstration that a 
 
CHAP, xxxiii. THE MOMTOn TrEJiET. 477 
 
 vessel built precisely like the first Monitor, provided with 
 a turret wheel-house and armor of adequate thickness, would 
 present the most perfect vessel for harbor defence hitherto 
 produced — whether for carrying heavy ordnance i>r fur 
 handling movable torpedoes em[)loyed against an attacking 
 fleet. 
 
 It will be observed by those who have studied the matter 
 that Adiiiiial Porter's statement, at the commencement of 
 his report, is materially modified by a subsequent paragi-a))h, 
 strongly recommending the ovcrJuing of the Monitor, "which," 
 the report states, "prevented the hull l>eing penetrated if 
 the vessel was struck by a ram." The Admiral fiii-ther 
 observes : " The value of this contrivance was sIkjwu in the 
 contest at Hampton Eoads, where the Mrrrimacl: rammed the 
 Monitor, merely turning the latter half round, and doing no 
 damage whatever." It will be seen, by the illustration (PI. 
 47) representing a ti-ansverse section of the original Monitor, 
 that a collision like that between the Iron DuTce and the 
 VangiiarJ, which sent the latter to the l>ottom, would not 
 be productive of greater danger than that caused by the 
 3IerrimacFs ramming, since, owing to the overhang and in- 
 clined sides, the Ir07i Duhe^s spur could not reach the 
 Monitor's hull. 
 
CHAPTER XXXIV. 
 
 THE MONITOR ENGINE. 
 
 (SEE PLATES oO AND 51, REPRESENTING A TOP VIEW AND SIDE 
 ELEVATION.) 
 
 The Engineer of April 20, 1866, contains the following 
 discussion relating to tlie engines employed in the monitor 
 iron-clads, accompanied by a brief description of the me- 
 chanical combination of these motors : 
 
 "The common feature of all Ericsson's screw-propeller 
 engines, however otherwise different in arrangement and 
 principle, consists in what he himself has described as 
 ' bringing the power of two engines to bear at right angles 
 on a common crank-pin ' — a feature already noticeable in 
 the engines built by him in 1839, at Liverpool, for the 
 Robert F. Stockton. This very vessel, tried on the Thames 
 so many years ago, is stated to be, even now, not only the 
 most poAverful tug of her class on the river Dela^\are, but 
 also the fastest. Her engines consist of two steam-cylinders, 
 
CHAP. XX. XIV. THE MOKITOR ENGINE. 479 
 
 placeil tliagoiially, with cioss-lieads aud side-rods connected 
 to a common crank-pin on the propeller-sLaft. In the Edith 
 and Massachusetts Ericsson modified his diagonal form of 
 engine by laying the cylinders against tiie ship's side, nnder 
 the deck, bottom up, with the piston and connecting-rods 
 working downwards, but still connected to a common crank- 
 pin on the propeller-shaft. A couutiymaii of Ericsson, Cap- 
 tain Carlsund, copied this arrangement, applied it in a num- 
 ber of Swedish vessels, exhibited it at the Paris Exhibition, 
 and, with the usual perspicuity of Univei-sal Exhibition juries, 
 was rewarded for this exhibit by the great gold medal. 
 
 "While still adhering to the feature of bringing the power 
 of two cylinders on to a common crank-pin, the present form 
 of marine engine adopted by Captain Ericsson may be looked 
 upon as an outgrowth of the peculiar engine, with semi- 
 cylinders, which he first applied to the United States steam- 
 frigate Princeton. Excellent illustrations of this engine 
 appeared some years ago in a work called 'Im2)erial Cyclo- 
 paedia of Machinery ' ; but as Captain Ericsson has himself 
 observed, in a letter to Mr. Woodcroft, the description began 
 by erroneously stating that 'Watt has described a similai" 
 ari-angement in one of his earlier patents.' This is so far a 
 mistake, as AVatt — according to a practice he adopted, under 
 the then state of the law, of inserting as many ideas as 
 possible into his patents — merely described the bare idea 
 of a piston vibrating within a semi-cylinder. xVs Captain 
 Ericsson states, 'the Princeton^ s engine consists of compound 
 or double semi-cylindei- engines, of different diameters, with 
 
480 THE MONITOli EXGIKE. chap, xxxiv. 
 
 pistons attacliecl to tlie common axle in ojiposite directions, 
 l»otli pistons being acted upon by tlie steam at the same 
 timt', tlK'ir differential force constituting the effective motive 
 power.' The wi'iter of -the description of the engines of the 
 Princeton in the ' Cyclopaedia,' while allowing that ' this 
 species of engine is very compact,' and that it ' admits of 
 being placed entirely below the water-line,' as also that, 
 ' although very many other arrangements have been since 
 brought out in this country, it is still a pre-eminently suc- 
 cessful engine,' yet observes that 'the friction is, of course, 
 more than in many others, inasmuch as it is found practi- 
 cally impossible to obtain the power of steam with so little 
 friction in any form of chamber as a true cylinder.' To 
 these objections Captain Ericsson replies: 'In the first place, 
 the absence of pressure on the main journal of the piston- 
 shaft is not understood by those who are not cognizant of 
 the fact that a straight line drawn from the crank-pin to 
 the opposite journal of the shaft passes through the centre 
 of gravity of the piston.' Then, again, ' the weight of the 
 piston, instead of scraping the bottom of the cylinder, is 
 suspended in the journals, and there produces but a very 
 small amount of friction.' 
 
 " ' Nor have critics recognized the fact that during the 
 passage of the crank-pin of the propeller-shaft through the 
 lower part of the arc of vibration, it is neai'ly relieved from 
 pressure by the opposing action of the connecting-rod — one 
 pushing while the other is pulling. Lastly, the absence of 
 the great friction produced by the diagonal thrust of the 
 
CHAP. XXXIV. THU MONITOR ENGINE. 481 
 
 short coDnecting-roJ against the guides of ordiuary propel- 
 ler engines also forms an important item of saving peculiar 
 to the semi-cylinder engine.' 
 
 " The weak point in this very ingenious engine is un- 
 doubtedly the piston. The ends, for instance, must wear 
 unequally and at a rate increasing with the radius of any 
 given point from the centre of vibration.* Ericsson is too 
 good a mechanic to shut his eyes to this fact ; and accord- 
 ingly, when, in 1859, the United States Navy Department 
 submitted the problem of the best screw-propeller engine 
 for solution by the engineers of America, he presented a 
 plan of an engine similar to that of the Princeton in all 
 essential features, with the exception of the introduction of 
 full cylinders instead of semi-cylinders. Since that pei-iod 
 the greatest success in America has accompanied this last 
 form of steam-engine ; it has been almost universally applied 
 to the later vessels of war of the States, and also in the 
 mercantile navy of that country. There is not a single 
 American monitor without an engine of this kind, and all 
 the Swedish monitors are engined on the same plan. It 
 has been patented by the inventor, and the accompanying 
 plans and descriptions are prepared from information sent 
 by Captain Ericsson himself to Mr. Woodcroft (see Plates 
 50 and 51). He has also sent a beautiful model of it, which 
 
 • In refutation of this objection, I have to state that the steamship Princeton, 
 after serving in tho Golf during the Mexican War, was sent to the JlcJiterranean 
 without repairing her pistons, tho etid-pachinga on examination proving to be in 
 perfect order. 
 
482 THE MONITOR ENGINE. CHAP, xxxiv. 
 
 may be seen in the Patent Office Museum at South Ken- 
 sington. 
 
 "'The several direct-acting screw-propeller engines hither- 
 to constructed,' says Caj^tain Ericsson, ' are all more or less 
 objectionable in the following particulars, viz. : the horizon- 
 tal engines occupy too much space transversely in the vessel 
 to admit of being placed in the run ; the vertical engines 
 pass through decks, and project so far above the water-line 
 as to be useless for war purposes ; and all approved double- 
 cylinder engines oj)erate on cranks placed at right angles to 
 each other, which involves a series of bearings, much fric- 
 tion, and liability to derangement from the shafts getting 
 out of line. In addition to these imperfections, the extreme 
 shortness of the cranks, with the attendant great friction on 
 the crank-pins and journals, to say nothing of the heavy 
 diagonal thrust of the connecting-rods, are serious defects in 
 the direct-acting screw propeller engines in common use.' 
 In Captain Ericsson's present form of screw-engine the two 
 cylinders of a double engine are arranged in such a manner 
 that their base or bottom ranges with a plane passing 
 through the axis of the propeller-shaft, or nearly so, in 
 combination with a certain arrangement of rock-shafts, crank- 
 arms, and connecting-rods, for imparting motion from the 
 pistons to the shaft, whereby he is enabled, first, to bring 
 the cylinders nearer to the propeller-shaft, and hence to 
 economize space and construct the frame of the engine of 
 great strength and compactness ; secondlj^, to avoid the 
 diagonal thrust and friction of the slides, unavoidable -when 
 
CHAP. XXXIV. THE MONITOR EXQINE. 483 
 
 the connecting-rod is attaclied directly to the cross-bead ; 
 thirdly, to oi)erate the connecting-rods nearly at right angles 
 to each other, ^vhich admits of tlie production of a continu- 
 ous motion A\ath a single ci-ank on the propeller-shaft, and 
 with a single crank-pin common to both engines ; fourthly, 
 to employ a crank on the propeller-shaft much longer than 
 half the length of stroke of the piston, thereby diminishing 
 the heavy pressure on crank-pins and on journals, which has 
 hitherto caused so much trouble by the overheating of the 
 bearings, and at the same time diminishing the strain on the 
 engine-fi'ame." 
 
 DESCRIPTIO^J" OF ILLUSTRATIONS OX PLATES 50 AND 51. 
 
 The general character of the Monitor engine will be 
 readily comprehended by a reference to PI. 50, represent- 
 ing a ground-plan, and PI. 51, the side elevation, viewed 
 from the bo^v of the vessel. The crank on the propeller- 
 shaft and the main connecting-rods, being hidden by the 
 steam-cylindei-s, are shown by dotted lines in the side ele- 
 vation. The two cylinders are placed end to end trans- 
 versely in the vessel, trunks or hollow piston-rods being 
 cast on the pistons, as shown by the sectional -^Iaw on 
 page 487, projecting outwards towards the side of the ves- 
 sel. These trunks are sufficiently large to permit the vibra- 
 tion of links connecting the pistons and short vibrating 
 level's attached to the for\vard end of the horizontal rock- 
 shafts. Referring to the top view of the engine, it Avill be 
 seen that vibrating levers of greater length are attached to 
 
484 TffH MONITOB, ENGINE. chap, xxxiv. 
 
 the aft end of tlie rock-shafts. These levers are coupled to 
 the common crank-pin on the propeller-shaft, the connecting- 
 rods acting nearly at I'ight angles to each other. By this 
 arrangement the throw of the crank may be made much 
 longer than in ordinary direct-acting engines ; consequently, 
 the strain on the crank-journal of the propeller-shaft will be 
 correspondingly reduced. A prolongation of the crank-shaft 
 forward — of small diameter — carries the eccentrics ^vhich ac- 
 tuate the steam-valves, while a prolongation of one of the 
 rock-shafts towards the stern operates an air-pump common 
 to both steam-cylinders. The bottom of the cylinders, or 
 the division between them, is formed as shown by the sec- 
 tional plan of the cylinders before referred to. 
 
 The Chief of the Bureau of Steam Engineering at Wash- 
 ington, Mr. B. F. Isherwood, having criticised the principle 
 of the Monitor engine, I published the following reply in 
 the Engineer of June 8, 1866 : 
 
 " deferring to Mr. Isherwood's ' Experimental Researches 
 on Steam-Engineering,' I find it stated, at page 340, that 
 the cost of the horse-power in the engines of the Monitor^ 
 when cutting off at 0.425 of the stroke of the piston from 
 the commencement, is 27.7 per cent, more than in the U. S. 
 paddle-wheel steamer MicMgan. ' Great as this excess ap- 
 pears,' says Mr. Isherwood, ' it is no more than what the 
 conditions fully warrant us to expect, and should be de- 
 cisive against the use of such a type of engine.' Mr. Isher- 
 wood accounts for this great loss of power in the following 
 manner ; ' From the description of the Monitor engine, it 
 
CHAP. XXXIV. THE MONITOR ENGINE. 486 
 
 will be perceived that two cylinders occupy the same barrel, 
 the separation being made ])y a simple partition of cast iron 
 in the centre. Further, that during a large portion of tlie 
 time the boiler-steam occupies one end of the cylinder, while 
 the adjacent end of the other cylinder is open to the con- 
 denser. There is, consequently, one end of one cylinder 
 maintained at the temperature of the boiler-steam, \vhile 
 the adjacent end of the other cylinder, separated only by a 
 cast-iron partition, is exposed to the temperature of the con- 
 denser. This arrangement, immaterial as it appeal's — and is 
 in a mechanical point of view — powerfully affects the econ- 
 omical result by its great influence on the cylinder-conden- 
 sation. To appreciate it, it is only necessary to imagine the 
 piston of the starboard engine, for example, to be near the 
 outboard end of its stroke, in Avhich case nearly the whole 
 of the cylinder of that engine will be filled with steam. At 
 this moment the piston of the port engine is near the centre 
 of its stroke, and about one-half of the port cylinder adja- 
 cent to the starboard cylinder will be open to the condenser 
 and exposed to its refrigerating influence; consequently, the 
 boiler-steam in the starboard cylinder has been exposed for 
 about one-half of the stroke of its piston to this refrige- 
 rating influence from the port cylinder, transmitted through 
 the iron partition of the two cylinders, which, as their dia- 
 meter is great in proportion to the stroke of their piston, 
 forms a large proportion of the surface in contact with the 
 steam. Nor does the evil end here ; for, as the sides of 
 both cylinders are the same piece of iron — those of the one 
 
486 THE MONITOB ENGINE. chap, xxxiv. 
 
 being merely an extension of tliose of the other — the con- 
 duction of heat is very rapid from one cylinder to the other, 
 and the heat imparted by the steam to the sides of the 
 starl)oard cylinder cpiickly passes along by conduction to the 
 sides of the port cylinder, whose interior is in communica- 
 tion w'lik the condenser, and -whose exterior is exposed to 
 the atmosphere. The inevitable result, it is manifest, must 
 be a largely-increased steam-condensation in cylinders of 
 this type of engine over that in tlie cylinders of the usual 
 type — ho^v much larger, is a question which experiment alone 
 can answer. There is still to be added to the already-de- 
 scribed peculiar causes of steam-condensation in cylinders of 
 the Monitor type of engine that of the half-trunk, the effect 
 of \vhich is, for a given capacity of cylinder, to increase 
 both the interior and exterior cylinder surfaces ; while the 
 thin, uo2>roteeted metal of the half-trunk — one side of which 
 is ahvays in contact with the atmosjihere, while the other 
 side is, too, for half the time, and not only in contact with, 
 but in rapid movement through, it — makes it a regenerator 
 of maximum power.' 
 
 " Before analyzing this extraordinary reasoning, let us 
 examine closely the section of the cylinder in the annexed 
 diagram. 
 
 " It will be seen that, although the two cylinders are 
 combined in one casting, each has a separate bottom, with 
 a considerable space between the two ; also that the heat to 
 be transmitted through the metal of the cylinder, as Mr. 
 Isherwood states, must travel a distance of 6 ins. from a to 
 
CUAP. XXXIV. 
 
 THE MOMTOi: ENGINE. 
 
 487 
 
 b, or from b to a, in less than half a second, in order to 
 produce the baneful effect pointed out by the author of 
 'Experimental Researches.' It will not be necessary to de- 
 monstrate that heat cannot be transmitted through 6 ins. 
 of metal in half a second, and it would be an insult to the 
 intelligence of your readers to detain them by disproving 
 Mr. Isherwood's assertion that a considerable auiouut of the 
 motive force is lost by thus transmitting heat back and for- 
 wards thiough the substance of the cylinder. 
 
 " ^Vith regard to the supposed rapid transmission of heat 
 through the 'iron partition of the two cylinders,' you will 
 find on referring to the section that no transmission of lieat 
 can take place, since the two Ijottoms are separated by a 
 stationary body of air or vapoi". In the ordinary cylinder- 
 bottom, the outside of the metal acquires, during regular 
 working of the engine, a ]H>rmanent temperature, attended 
 by a constant loss of heat radiated into, .ind continually ab- 
 
488 TEE MONITOB ENGINE. chap, xxxiv. 
 
 sorbed by, tlie atmosphere. In the case of tlie bottom-plates 
 of tlie Monitor''s cylinder, a permanent temperature is also 
 acquired, but there is no loss of heat by radiation after the 
 intervening small body of air or vapor has attained maximum 
 temperature. It may be truly said that the Monitor engine, 
 with its cylinders combined in one casting, furnishes the only 
 instance in which no heat is lost by radiation thi'ough the 
 cylinder bottom. Having thus disposed of the absurd notion 
 that a vast quantity of heat is transmitted from cylinder to 
 cylinder, we now come to the question of increased internal 
 cylinder sm-face consequent on the application of the trunk. 
 Mr. Isherwood ti'eats this question as one of such great 
 importance that I have taken the trouble to ascertain the 
 exact amount of increase. Area of 40-inch cylinder, 1,256 
 square inches; area of IS^-inch trunk, 143 square inches'. 
 Deducting from this 20 square inches for a 5-in. piston-rod, 
 which the ordinary engine would require, we have a differ- 
 ence of 123 square inches occupied by the trunk. But the 
 trunk only affects the outboard end of the cylinder, and 
 hence the mean area taken up is only 61^ square inches. 
 To make up for this loss of area, the diameter of the cylinder, 
 it will be found by calculation, has only to be increased 
 H of an inch. I should be trifling with the patience of 
 your readers were I to enter on a calculation to show the 
 amount of loss attending such small increase of the diameter 
 of the cylinder. Only one more point, urged by Mr. Isher- 
 wood against the Monitor engine in explanation of the 
 asserted 27 per cent, loss of motive force, remains to be 
 
CUAP. XXXIV. THE MONITOR ENGINE. 489 
 
 consideretl, viz., the effect of the trunk, which he calls a 
 'refrigerator of maximum power.' Mr. Isher\vood devotes 
 so much time to the theoretical consideration of the steam- 
 engine that I can well understand that he lias no time left 
 for practice; otherwise I should feel surprised at his igno- 
 rance of the fact that the great difficulty with trunk engines 
 is tliat of keeping the packing steam-tight without causing 
 overheating. Experience shows that the best that can be 
 done in practice is to prevent the trunk fi-om exceeding the 
 initial temperature of the steam; and hence the trunk, in 
 place of being a ' refrigerator of maximum power,' is actually 
 a super-heater. 
 
 "It will be asked by what process did Mr. Isherwood 
 ascertain the amount of the assumed loss of power for which 
 he accounts by this extraordinary reasoning? He placed, 
 according to his statement in 'Experimental Researches,' a 
 tank on a wharf, to which the Monitor was made fast. From 
 this tank he supplied the boilers of the vessel by means of 
 temporary feed-pipes. Steam having been raised, the engines 
 were started and kept in motion for seventy-two houi-s in 
 succession. Indicator-cards were taken every hour, by means 
 of which the mean indicated horse-power exerted by the 
 engines was ascertained. This w:is compared witli the quan- 
 tity of water measured into the tank and supposed to have 
 been converted into steam. The utmost precision was prac- 
 tised; the water run into the tank was ascertained to the 
 tliii-(l decimal of a pound ; the barometer and direction of 
 the wind carefully noted. The difference between the iiidi- 
 
490 THE MONITOR ENGINE. chap, xxxiv. 
 
 cated power and tLat -winch ought to have been produced 
 according to the quantity of water used, supposed to have 
 been converted into steam, was set down as ' loss occasioned 
 by cylinder-condensation.' I observe that Mr. Isherwood's 
 tables and calculations make no allowance for condensation 
 in the steam-pipes and valve-chests. Loss by leaks through 
 the pistons, valves, and glands ; leaks and waste of water and 
 steam in the boilers — all these sources of loss appear to be 
 ignored by the experimentalist, and every pound of water 
 measured into the tanks is debited to the engine. A balance 
 is then struck by deducting the indicated horse-power, and 
 the difference put down as 'loss caused by cylinder-conden- 
 sation.' I abstain from criticising this rough and unsatisfac- 
 tory mode of deciding the nice jDoint of cylinder-condensation ; 
 but I cannot omit adverting to the fact that, while the pres- 
 sui'e in the boilers, during the trial of seventy-two hours, was 
 kept at 17 lbs. above the atmosphere, the pressure admitted 
 into the cylinders was only If lbs. above the atmosphere. The 
 tables show that, in order to maintain this low working-pres- 
 sure in the engines, the throttle-valve was set permanently 
 at an opening of four square inches. In view of the magni- 
 tude of the engine — two double-acting cylinders of 40 ins. 
 diameter — the admitting the steam through a single opening 
 of such small area was certainly a most singular expedient. 
 Had the object of the trial been to exhibit a minimum indi- 
 cated horse-power, a more effective expedient could not have 
 been devised. As the mean area of the piston, according to 
 the tables, was 1,185 square inches— about three hundred 
 
CHAP. XXXIV. THE MOXTTOR EXOINE. 491 
 
 times greater than the area of the throttle-valve — the speed 
 of the pistons being at the same time upwards of 2.5 feet 
 pt'i- second, some idea may be formed of the amount of dy- 
 namic force exerted and heat extinguished in passing the 
 steam from the boilers to the engines. Yet this waste of 
 force was ignored by the experimentalist, and the indicated 
 piston-power alone was set against the water measured into 
 the tank. 
 
 " Apai't from these facts, the most critical examination 
 of the sectional plan of the cylinders of the Monitor before 
 refei'red to (see page 487) fails to discover any error of con- 
 struction productive of cylinder-condensation to a greater 
 extent than in the screw-engines of the most celebrated 
 makers." 
 
CHAPTER XXXV. 
 
 THE MONITOR DICTATOR. 
 
 (SEE PLATES 52, 53, 54, AND 55.) 
 
 The Dictatoi' is the most powerful and efficient fighting- 
 ship possessed by the United States. Like the Monadnock 
 class of monitors, she is also a good cruising vessel. The 
 hull, engines, turret, and gun-carriages of this ship were de- 
 signed and furnished by the writer, under a contract with 
 the United States Navy Department, 1862. Her length on 
 deck is 314 ft., the overhang aft being 31 ft., and the for- 
 ward overhang 13 ft., leaving a length of 270 ft. between 
 perpendiculars (see side elevation, PI. 52). Her breadth 
 over the sides is 41 ft. 8 ins., and her total beam 50 ft., 
 whilst her draught is 20 ft., the sides midships projecting 
 only 1 ft. G ins. above the water-line. The sides are pro- 
 tected at and near the water-line by 11 ins. of wrought iron, 
 this thickness consisting of inner bars 5 ins. thick, and six 
 plates each 1 in. thick (see transverse section, PI. 58). The 
 plates are placed upon massive timber backing, as shown 
 
CHAP. XXXV. THE MOXITOIi DICTATOR. 493 
 
 by the illustration on the plate referred to. Her tonnage 
 is 3,000 tons. 
 
 The engines of the Dictator (see illustrations on Plates 
 53 and 54) are 4,500 indicated horse-power, consisting of a 
 pair of vertical cylinders 100 ins. in diameter, with a stroke 
 of 4 ft. The pistons have small trunks on their upper 
 side, and are connected by links with the ends of curved 
 horizontal amis attached to the forward end of rock- 
 shafts placed outside the cylinders slightly above the latter. 
 Straight arms nearly vertical are attached to the after-ends 
 of the rock-shafts. These arms are coupled by means of 
 diagonal connecting-rods to a common crank-pin fixed near 
 the circumference of a fly-wheel upon the screw-shaft. The 
 engines drive a propeller 21 ft. G ins. in diameter, provided 
 with four blades set at a pitch of 34 ft. The propeller 
 weighs 39,000 lbs., and its shaft 36 tons. The engines are 
 supplied with steam by six boilers having altogether 50 
 furnaces; the total grate-surface is 1,128 square feet, and 
 the heating surface over 32,000 square feet. The coal-bunkers 
 accommodate 600 tons of coal. 
 
 The Dictator has a single revolving turret, essentially the 
 same as that delineated on Plate 49; it is 24 ft. in diameter 
 inside, by 9 ft. high, and contains two fifteen-inch guns. 
 The sides of the turret are built up of two separate concen- 
 tric cylindei-s, composed of plates 1 inch thick firmly riveted 
 together, a space of five inches being formed between the 
 said cylinders. This space is filled with segmental wi-ought- 
 iron slabs 5 ins. thick and 12 ins. broad; hence the total 
 
494 THE MONTTOR DICTATOB. CHAP. XXXT. 
 
 thickness of tlie turret wall is 1 ft. ?> ins. The turret has a 
 bell-mouthed top formed of iron plates i in. thick, curved 
 outwards, as shown on Plate 52, for the pui-pose of throw- 
 ing oft' the water which, during heavy weather, is dashed 
 up the sides of the turret. Around the bell-mouthed top 
 mentioned is carried a wooden grating provided with a 
 hand-rail, and appropriate stanchions for supporting an awn- 
 ing. The grating referred to forms a convenient balcony 
 for promenade. Over the centre of the turret is placeil the 
 pilot-house, which is 8 ft. in diameter in the inside, and 
 7 ft. high, its sides being formed of twelve thicknesses of 
 1-in. plates. This structure is supported by a strong cross-' 
 beam which rests upon a collar formed on a strong central 
 ■\vrouglit-iron shaft passing down through the turret ; this 
 shaft being stationary, the turret revolving round it. Except- 
 ing the bell-mouthed plate-iron extension at the top, the 
 Dictator turret is constructed precisely as the turrets of the 
 Passaic class of monitors delineated ou Plate 49. Referring 
 to this delineation, it will be seen that the weight of the 
 turret wall is suspended by diagonal rods in such a manner 
 that the entire weight of the turret and armament, as well 
 as the pilot-house, is sustained by the stationary central shaft. 
 The latter rests on a casting bolted to the transverse bulk- 
 heads of the ship, as shown in Plate 4:7. The pilot-house 
 is provided with sight-holes placed at a convenient height, 
 and, like the centi;al shaft, is stationary. The diagonal rods 
 before referred to, it should be observed,, are furnished with 
 screw-euds and nuts, in order to admit of their being tightened 
 
CUAP. xxxv. TJIE MOMTOJi DICTATOB. 495 
 
 in case the turret wall sliuuld sjig. The cross-beam before 
 mentioned, which extends across the top of the turret, sup- 
 l)ort8 rafters carrying iron bars 4 ins. deep by 3 ins. wide, 
 and 2^ ins. apart, these bars being covered with peii'oiatetl 
 plates 1 iu. thick. The base of the turret wall rests on a 
 flat ring composed of bronze, the under-side of which is accu- 
 rately faced. It is supported by anotliei- Hat ring faced on 
 the top and secured to the deck. By this means a water- 
 tight joint is formed between the base of the turret wall 
 and the deck. It is obvious that considerable power would 
 be required to cause the turret to revolve if its weight 
 rested on the I'iugs described. Accordingly, a taper key 
 is inserted under the stationary central shaft, by which the 
 weight of the turret and appendages may be raised so as 
 to rest wholly on the shaft. During the war this key was 
 invariably tightened before going into action, hence the tur- 
 rets revolved with perfect freedom while the gunners pointed 
 the pieces. A circular cliunuel is formed iu the deck near 
 the inside of the turret, in order to carry off water that may 
 leak under the base. It is conducted to the bilge by small 
 scuiiper-pipes. The machinery for turning the turret con- 
 sists of a pair of donkey-engines, which work gearing con- 
 nected with a large cog-wheel secured by strong lugs to 
 the under-side of the gun-slides, which are in their turn 
 tirmly attached to the sides of the turret. It should be 
 observed that all the turning gear is below the deck-line. 
 The floor of the pilot-house consists of a wooden grating 
 fitted with hinged hatches, through which the captain and 
 
496 THE MONITOR DICTATOR. chap. xxxv. 
 
 steersman enter from tlie turret. The steering-wheel is con- 
 tained in the pilot-house, and its motion is transferred, by 
 gearing and by a rack sliding in a groove formed in the 
 central shaft, to a pinion fixed upon the axle of the steering 
 barrel belo^v. From this barrel chains extend to the rudder, 
 which is of the balanced kind. The vessel is strengthened 
 beneath the turret by transverse and longitudinal bulkheads. 
 
 The gun-carriages are run out upon their slides by means 
 of winch-handles moving wheel-work geared into racks, the 
 friction-gear being tightened as soon as the pieces are full 
 out, and of course relieved immediately after the recoil. Each 
 gun has a radial bar placed above it, upon which runs a wheel 
 supporting a block and tackle, provided with a small dished 
 platform upon which the shot is placed. By this contrivance 
 the shot can be rapidly raised from the shot-locker to the 
 muzzle of the gun. 
 
 In the Dictator the air required for ventilation and for 
 supplying the boiler-furnaces is drawn in by several large 
 fan-blowers, partly through the top of the turret and partly 
 thr(.)ugh shot-proof trunks carried high above the decks, as 
 shown in the side-elevation, Plate 52. The vessel is provided 
 with a spacious platform or promenade deck, placed at nearly 
 the same height as the top of the turret (see plate referred 
 to). This deck is supported on vertical iron stanchions, the 
 ship's boats being suspended below the same. The cabins 
 are lighted by means of bull's-eyes fitted into brass frames 
 let into the deck, .these frames being replaced by solid 
 wrought-ii'on covers when the ship is going into action. 
 
CHAP. XXXV. T£I£ MONITOR DICTATOE. 497 
 
 The Engineer, in au article published 1866, states with 
 reference to the Dictator : 
 
 " It may be laid down as a general axiom that when the 
 introduction of fresh elements into any mechanical problem 
 has effected a revolution in its first conditions, originality is 
 at great advantage over hesitating and long-pondering judg- 
 ment. Though without the advantages which our great 
 command over iron in large masses has given in the modern 
 substitution of iron for wood in naval warfare, the Ameri- 
 cans have certainly shown much originality and boldness in 
 their designs for war-shi^js. A good deal of this is due to 
 Captain Ericsson — an original thinker and constructor, \vhose 
 very originality would have led him to be distrusted in 
 this conservative countiy. Broadly stating the matter, it 
 may be said that in France and in England we have pretty 
 much confined ourselves to bolting massive iron slabs, with 
 an intermediate packing, to the skins of our iron vessels of 
 war ; but Ericsson has taken a much more comprehensive 
 view of the capabilities of engineering in its application to 
 the eternal war-problem of doing as much damage as pos- 
 sible to your adversary with as little as possible harm to 
 yourself." 
 
CHAPTER XXXVI. 
 
 THE MONITOR TURRET AND THE CASEMATE. 
 (SEE PLATE 56.) 
 
 An opportunity of instituting a direct comparison between 
 the monitor turret and the fixed casemate was furnished 
 by the completion of the Turkish armor-chxd vessel Moijini 
 Zaffer, launched on the Thames in June, 1869. The build- 
 ing and arming of this iron-clad being the result of the 
 Joint efforts of Sir William Armstrong, Samuda, and Raven- 
 hill, we have a guarantee that whatever merits the fixed 
 casemate system possesses have been fairly developed in this 
 attemj)t to supersede the monitor. 
 
 It cannot fail to be noticed, on careful examination of 
 the illustrations on the plate referred to, that the planning 
 of the casemate of the Moyini Zaffer shows much thought 
 and elaboration ; also that the complication whicli character- 
 izes its form is evidence that the planner was dealing with 
 a difficult subject. Nor can the attentive observer fail to 
 
 498 
 
CUAP. XXXVI. THE MOSITOE TURUET AND TUE CASEMATE. 499 
 
 see at a glauce how imperfectly the dij^aJvantages atteutling 
 the elougatiuu and immobility of the battery — viz., the limited 
 horizontal range of the guns — have been overcome by the 
 combination of curvature and angles resorted to by the con- 
 structor of this substitute for the monitor turret. 
 
 Our illustrations, besides representing a top view of the 
 Moyini Zaffer, accurately drawn to scale, also represent a top 
 view of a monitor provided with two turrets of the same 
 diameter as those of the Pasmic cLiss — viz., 21 feet internally. 
 The length of the Turkish vessel is 230 feet, with 35 feet 
 G inches beam. The monitor, for the sake of exact com- 
 parison, has the same dimensions ; but the thickness of its 
 armor is greater than that of the former, and so proportioned 
 that the %oeujlit of armor of both vessels is alike. The free- 
 board of the Moyini Zaffer, as in all iron-clads built by 
 English engineei-s, is several times higher than that of the 
 monitor, and consequently deeper armor below water must 
 be applied to afford protection, increased rolling being the 
 inevitable result of high freeboard. Referring to the bat- 
 teries, it ^^^ll be seen that the circumference of the fixed 
 liattery is greater than that of the two turrets in the ratio 
 of 25 to 15. 
 
 The Engli-sh mechanical journals, in describing the Moyini 
 Zdffii; point with apjiarent satisfaction to the circumstance 
 that this casemate ship, which is intended for the defence 
 of the Bosporus, has armor-plates "generally six inches in 
 thickness, the whole of the battery (backed with wood) 
 being cased with 5-inch plates." The battery, though pierced 
 
500 THE MONITOB TUFRET AND THE CASEMATE, chap, xxxvi. 
 
 for eight guns, will only carry four of Armstrong's 12-ton 
 riiles. Tlie intention being to transfer the pieces from one 
 side of tlie battery to the other during action, it is evident 
 that Sir William has reached the limit of weight. The 
 difficulty of changing sides with the rapidity called for dur- 
 ing contest with screw-propelled assailants needs no explana- 
 tion. But the constructor of the monitor turret, which, as 
 our illustration shows, commands 340 deg. of the horizon, is 
 not hampered by considerations of weight of metal, a 24-ton 
 gun, or even one weighing forty-eight tons, being pointed as 
 readilj^ by turning the turret as the lightest field-piece. 
 Accordingly, the monitor which our illustration represents 
 is mounted with four 24-ton guns. 
 
 Making proper allowance for the greater area of side- 
 armor and battery-plating of the Moyini Zaffer, it will be 
 found that our double-turreted monitor will, on the same 
 draught of water, support 10-inch thick side-armor, 15-inch 
 thick turret-plating, and carry four 24-ton guns. The greater 
 security — we might say the impregnability — thus attained 
 by the monitor form is, however, only a part of the ad^'antage 
 of this system over that which is represented by the Turkish 
 iron-clad — the latest endeavor of some naval constrictors to 
 demonstrate that the conflict at Hampton Roads was not, 
 after all, so decisive as supposed. 
 
 Impregnability and calibre, although very important, by 
 no means decide the superiority of armored vessels ; hori- 
 zontal range is in many cases of equal importance. A monitor 
 hull provided with a fixed battery may be made as impreg- 
 
CHAP, xxxvr. THE MOXITOn TUFnET AND THE CASEMATE. 501 
 
 nable as a complete monitor, but at least two-tliii-ds of the 
 guns of such a vessel will be ineffective in battle. Samuda, 
 evi(lentl)', was fully aware of tlie impotency of his artilleiy, 
 owing to limited horizontal iviiige, when he adopted the 
 complicated form of the battery of the Moijini Zaffer. 
 
 Let us now consider in detail this question of horizontal 
 range, and inspect closely the extent of ranges marked on 
 oiii' illiistraticin foi' each gun separately. The ranges olitained 
 by the fixed battery of Samuda's construction first claim 
 our attention. To avoid confusion, the ports have been let- 
 tei'ed a, />, c, and '/, the first letter denoting the forward poi't 
 of the battery, and also the muzzle of the piece belonging 
 thereto. Beginning with the first-mentioned port, it will be 
 seen that each gun respectively ranges over a field of 96, 98, 
 98, and 02 deg. Referring to the monitor, it will be seen 
 that each of the four guns sweeps a field of 170 deg. It 
 should be observed that the ranges marked on the illustra- 
 tion have reference only to the starboard side of the line of 
 keel. 
 
 It will be proper, before assigning a numerical value to 
 the eflSciency of each of the systems under consideration, to 
 remember that the real power of naval artilleiy is determined 
 by niiiUiplying the weight of shot by "the horizontal ranfife, 
 the position of the vessel remaining constant. ^Modern taro^et 
 practice having demonstrated that a 24-ton gun is capable 
 of throwing a projectile o"f 600 pounds with adequate force, 
 and that a 12-ton gun is about the projier size for .'JOO-jKmnd 
 projectiles, we are enabled, by applying the rule befoi-e men- 
 
502 THE MOKITOE TUBBET AND THE CASEMATE, chap, xxxvi. 
 
 tioned, to determine with exactness the rehative efficiency of 
 the monitor turret and the fixed battery or casemate. The 
 power of the forward gun a of the casemate will accordingly 
 ])e represented by 300 X 96 = 28,800. In like manner, by 
 multiplying the -weight of the projectiles of the remaiuino- 
 three guns by their respective ranges in degrees, we obtain 
 a sum total of 115,200. Applying the same mode of com- 
 putation to the monitor — viz., multiplying 600 X 170 X 4 — 
 we establish the important fact that the actual efficiency of 
 the monitor is to that of the casemate vessel as 408 to 115, 
 Apart from this superiority as regards the artillery of the 
 monitor over that of the Turkish iron-clad, the armor of 
 both battery and hull of the latter is wholly insufficient to 
 compete with the former. The inference, therefore, is obvious 
 and irresistible that the monitor represented by our illus- 
 ti'ation could readily destroy Samuda's casemate vessel. But 
 it is not my intention to prove the worthlessness of the 
 Moyini Zaffer as a war vessel, the object of discussing the 
 subject being simply that of instituting a comparison between 
 the two systems represented by the illustrations on Plate 5G. 
 It merits special attention that, apart from the limited 
 horizontal range of all the guns of the Moyini Zaffe>\ only 
 one of the four — viz., a — can be j^ointed forward parallel w^ith 
 the ship's course; and that c, the only other gun capable of 
 firing ahead, cannot point nearer than 11 deg. of the line of 
 keel. At a distance of a mile ahead, there is, consequently, 
 a field of 1,200 feet which an assailant may occupy, ex2:)osed 
 to only one 12-ton gun. Chased by an enemy, the Turkish 
 
CUA1-. xxxvi. lUE MoyiTOU TLliliET AM) TUE CASEMATE. 5013 
 
 \var-shii), with the Saimuhi-Aiiiiiitioug battery, will be eiiually 
 impotent; the gun marked d being her only defensive wea- 
 l)on. It will be founil, on inspection, that the piece marked 
 b, like that marked c, cannot be pointed nearer than 1 1 de"'. 
 of the line of keel. 
 
 Let us now turn to the monitor. It will be seen that 
 four 24-ton guns, two forward and two aft, fire in a direct line 
 with the keel; there being no safe position, as in the case 
 of the fixed battery, for the enemy's vessel to occupy. The 
 entire field, viewed from stem to stern, as the plan shows, 
 is swept by all the guns of the monitor. Bearing in mind 
 that these powerful guns are protected by 15-iuch thickness 
 of iron, which, if applied in two thicknesses, is proof against 
 any artillery yet produced, wliile the 1 2-ton guns of the 
 Samuda-Armstrong battery are protected by armor which a 
 7-inch rifle will pierce through and through, the argument 
 in favor of the monitor turi-et becomes overwhelming. 
 
 It will be asked, in view of these incontrovertible facts, 
 why do constructors advocate the fixed battery? I know of 
 no other reason than the assumption that the joint between 
 the rotating turret and the deck cannot be made secure. 
 English engineers, relying on the accounts of the perform- 
 ances of the monitors published by the enemies of the Union 
 during the war, apparently do not take the trouble to investi- 
 gate the matter; while American experts who have written 
 about turrets appear to be ignorant of the leading facts 
 connected with the turret system. 
 
 For instance, Mr. Eads, in a report to the Navy Depart- 
 
504 THE MOXlTOli TUltBET AND THE CASEMATE, chap, xxxvi. 
 
 meut, informs the Secretary that " the baud rouud the base 
 of the tui-ret on the Dictator weighs over 20,000 pounds," 
 aud points out how much better this great weight of irou 
 might be applied for other purposes. Now, this turret has 
 no baud rouud its base, nor was it ever intended to have 
 one, Mr. E. also tells the Secretary that auy down^vard 
 swelling of the plating, produced by the impact of projec- 
 tiles striking low, will stop the rotation of the turret by 
 friction under its base. This assertion proves ignorance of 
 the fact that the Diotatoi' turret rests wholly on the four 
 inner courses of plating (which cannot be swelled), and that 
 the intermediate wrought slabs and outer plating (together 
 11 inches in thickness) do not reach the deck., aud therefore 
 can, by no possibility, cause the predicted stoppage. Again, 
 the apprehensions expressed in several reports, with reference 
 to the base of the pilot-house in connection with the rotation 
 of the turret, prove that another very important circumstance 
 has been overlooked — viz., that the turret projects consider- 
 ably above said base, thereby effectually protecting it. 
 
CHAPTER XXXYII. 
 
 CARRIAGES FOR HEAVY ORDNANCE. 
 (SEE PLATE 57.) 
 
 The Engineer, in discussing the subject of gun-carriao'es 
 (in 1868), says: "Americans mount their big guns in turrets, 
 and France has no peculiarly big guns to mount. In the 
 matter of carriages, as in almost everything else connected 
 with recent improvements in ordnance, England must be 
 content to act as schoolmistress to the rest of the world." 
 This assertion is preposterous, in view of the fact that Eng- 
 land had not mounted a single heavy gun on shipboard at 
 the time when we had a large fleet of iron-clad.s armed with 
 11 and 15 inch guns. Not only that: we iiad effectually 
 used those guns in numerous engagements, and fully estab- 
 lished the reliable character of our system of mounting the 
 same, before English artillerists believed it possible to dis- 
 pense with breeching. But our contest was watched by 
 attentive eyes, and hence our success did not long remain 
 
506 OABBIAOES FOB HEAVY OBDNANOE. chap, xxxvii. 
 
 a secret. An enterprising English captain sjjeedily pro- 
 cured drawings of tlie Monitor gun-carriages and tlieir fric- 
 tion-gear. Ho^v faitlifully lie copied our system the reader 
 will see by comjjaring tlie several devices for producing 
 friction represented on Plate 57. Sir William Armstrong, 
 too, becoming convinced that the Monitor friction-gear was 
 the best for checking the recoil of naval ordnance of heavy 
 calibre, also follo\ved our lead. An anuising contest arose 
 between Sir AVilliam and the enter2:)rising naval officer al- 
 luded to, whose indignation knew uo bounds on finding 
 that the great gunmaker had adopted the same plan as him- 
 self for checking the recoil. But Mr. Scott Russell having 
 in the meantime published accurate drawings of the Monitor 
 gun-carriages and friction-gear. Sir William was in a posi- 
 tion to silence the complaints of his rival by simply point- 
 ing to Plate No. 139 of Scott Eussell's great work on naval 
 architecture. 
 
 By referring to the illustration mentioned, the reader 
 will see at a glance that Captain Scott's friction-gear is iden- 
 tical with that applied to the gun-carriages of the American 
 iron-clad fleet. The principle is very peculiar, and involves 
 the apparent paradox of obtaining increased friction to any 
 desirable extent without adding to the force employed. A 
 brief explanation will show how this singular result is ef- 
 fected. A series of vertical plates are secured to the lower 
 part of the gun-carriage in such a manner as to admit of a 
 slight transverse movement. These plates slide freely be- 
 tween longitudinal friction-timbers, or })lanks composed of 
 
CHAP. XXXVII. CAIil'IAGLS I'OI! HEAVY OEDyAXCK 607 
 
 hard wood, which in broadside vessels are attached to tlie 
 "slides," and in the monitoi-s secured to the base of the 
 turrets. It will he readily understood that, l)y applying 
 lateral force to the two outside vertical plates from \\itli- 
 out, friction will be established between all the plates and 
 the intervening planks ; and it will be evident that the 
 amount of friction between the surfaces in contact will de- 
 pend on the force thus applied, wholly independent of their 
 7iumhei: Thus, by merely doubling the number of plates 
 and idanks, the friction will be doubled without callino' for 
 the application of any additional force. It rarely hajipens 
 in mechanical contrivances that the effect to be produced is 
 so completely independent of the force applied as in this 
 instance. The practical advantage of obtaining requisite 
 friction without employing great manual power is obvious ; 
 and that it is fully appreciated may be infeired from the 
 alacrity with which the system has been copied in Euroj^e. 
 The difficulty of handling the modem monster guns on 
 board ship in l)ad weather, ^vllich at one time was deemed 
 impracticable by experienced sailors, vanished with the in- 
 ti'oduction of my multiple.x friction apparatus thus briefly 
 described. The reader will observe how closely even the 
 detail of the original has been followed by the plagiarists ; 
 the mode of pi'oducing the lateral pressure, for instance, has 
 been carefully copied by Captain Scott. lie employs the 
 transvei-se screw and vertical levers by which the outside 
 friction-plates are forced inwards, precisely as in the Monitor 
 carriages. Sir William Armstrong also emplojs the trans- 
 
508 GAliBIAGES FOB HEAVY OBDNANGE. chap, xxxvii. 
 
 verse screw and vertical levers, but lie divides the screw in 
 the middle — an ill-considered modification, as it calls for the 
 application of force on both sides of the carriage, increases 
 tlie friction, and tends to pull the sides together. Sir Wil- 
 liam Armstrong also introduces the modification of employ- 
 ing iron bars in place of the wooden friction-planks — a most 
 objectionable expedient, as the needed friction is greatly 
 diminished by presenting metal against metal. Moreover, 
 the friction becomes so irregular as to bafHe any attempt at 
 systematic tightening with reference to the charge of pow- 
 der employed, rendering accidents inevitable. It is evident 
 that if the metallic plates are kept dr}", abrasion follows, and 
 that their surfaces are liable to cut and stick. If oiled, the 
 least excess of lubrication will reduce the friction to such 
 an extent as to permit the gun to recoil without check, as 
 experience during experimental practice has shown. Apai't 
 from these objections, the want of that indispensable elas- 
 ticity which the wooden friction-plank affords is fatal to Sir 
 William Armstrong's substitution 6f metal for wood. 
 
 With such facts before it, the Engineer tells its readers 
 that, in the matter of carriages and other improvements 
 connected with naval ordnance, England must be content to 
 act as " schoolmistress " to the rest of the world. 
 
 Alas for the schoolmistress ! She has been endeavoring 
 to teach the \vorld for a long time that our system of naval 
 defence was all wrong, until at last she has discovered that 
 her boasted broadside iron-clads, on which millions have 
 been spent, are hopelessly \nilnerable. 
 
CHAP. XXXTII. CAEFTAGES FCW ]TEArT ORDNAKCK 509 
 
 The leading journal of England, Februaiy 12, 1SG8, frankly 
 admits that "the final blow" has been given to the "already 
 tottering theory of broadside iron-clads," and adds: "Why 
 do we obstinately refuse to build small iron-clad, sino-le- 
 tuiret vessels, with low freeboard, and one or two guns of 
 the heaviest calibre? The American and Russian officei-s 
 who have actually tried them report with enthusiasm of 
 their sea-going properties." It would have been well for the 
 "schoolmistress" if she had not listened to the advice which 
 prejudiced naval constructors have persistently tendered ; it 
 might have spai-ed the naval administration of England the 
 severe censure called forth at the time for having neglected 
 to adopt the monitor system. " It seems to us," says the 
 Times of the date before mentioned, " that the Admiralty 
 have in nothing so neglected their duty as in failing to 
 provide us with a large supply of these formidable little 
 vessels." 
 
 Can the Enghieei\ moreover, point to a single invention 
 connected with our turret iron-clads, naval ordnance, or gun- 
 carriages which has originated in England ? Those best 
 acqutiinted with the matter know that every mechanical de- 
 vice relating to the system Avhich so successfully vindicated 
 itself dui-ing the late wai- was contrived on this side of the 
 Atlantic — a success the more remarkable since the exigency 
 of the time did not admit of previous expeiinients, everv- 
 thing being despatched directly from the foundr}- and woi-k- 
 shop to the scene of conflict. 
 
CHAPTER XXXYIII. 
 
 PIVOT-CAKEIAGES OF THE THIRTY SPANISH GUNBOATS. 
 (SEE PLATE 58.) 
 
 The gim-carriages .and slides constioicted for the SpanisL 
 gunboats present two important features wliich distinguisli 
 the same from other pivot systems — viz., the slide is made 
 to rotate round a permanent central fighting-bolt secured in 
 the middle of the deck near the bow ; consequently, as the 
 bulwarks of the Sj^anish gunboats are low enough to admit 
 of firing en barbette, a horizontal range of 240 deg. is ob- 
 tained. 
 
 The other important feature of the new system is that 
 of enabling the gunner to apply and relieve the compressor 
 instantaneously. 
 
 Naval artillerists are well aware of the advantage of ro- 
 tating slides, but, owing to the circumstance that such an 
 arrangement unavoidably carries the fighting-bolt in the rear 
 of the trunnion when the gun is run out, such slides have 
 been deemed impracticalde. Evidently, if the fighting-bolt 
 
CHAP. XXXVIII. SJ'AXl:sn GUNBOAT riVOT-CAElilAGES. 511 
 
 1h' placed tar iu the rear uf the truiaiioii, the slide will be 
 lifted upwards with great viuleiiee at the iustaut of discliarge. 
 This apparently insuperable difficulty is completely overcome 
 in the arrangement now under consideration, by the expe- 
 dient of raising the circular ring on which the slide turns 
 about one inch above the deck. By this expedient an effi- 
 cient abutment will be obtained for restraiuins the loujri- 
 tudiiial movement of the slide in all jjosltions. A plate at- 
 tached to the front transom of the slide, as represented by 
 the illustration on Plate 58, extending down as far as the 
 bottom of the ring, thus takes the [)lace of the ordinary 
 fighting-bolt. The central pivot, round which the slide re- 
 volves, fits so loosely in the socket of the cross-plate that 
 the whole force of the recoil is received by the descending 
 trausom-plate and the edge of the deck-ring. The latter is 
 sustained by a circular platform of boards 1 in. thick, se- 
 cured to the deck, and flush with the top of the ring. The 
 front transom and outside circumference of the deck-ring 
 being in advance of the centre of the trunnion of the gun 
 when run out, the force of the recoil, iu place of lifting, 
 will evidently tend to depress, the slide. Ample experience 
 in working the slides of the Spanish gunboats has fully 
 demonstrated this fact, and established the superiority of 
 the rotating slide in point of easy handling as well as ex- 
 tensive lateral range. 
 
 It will be evident on reflection that a very slight modi- 
 fication will adapt the rotating slide thus described to broad- 
 side firing. Such a modification was made in December, 1869, 
 
512 SPANISH GUKBOAT PlVOT-VAlllilAGES. chap, xxxviil. 
 
 iiud tliu slide tLii« modified, together with its carriage, was 
 presented to the Ordnance Bureau for trial. A 100-pounder 
 Purrott rifle-guu having been mounted on the carriage, Com- 
 mander E. Simpson was ordered by the Chief of the Bureau 
 to conduct the trial on board the U. S. steamer Tallapoosa, 
 during a run from New York to Washington. Commander 
 Simpson's report of this trial and his description of the 
 new arrangement are so lucid that I adopt the same in 
 preference to any description I could pen : 
 
 " The carriage consists of a slide and top-carriage, con- 
 structed of wrought iron. The slide is composed of two 
 rails, with four bolts connecting them at intervals of two 
 feet. The heurters consist each of two plates of half-inch 
 iron^ between which are placed the rollers for lateral train. 
 They are each strengthened by two castings placed between 
 the plates near the rails, those at the rear end being con- 
 tinued up nine inches above the rails, to Avhich are secured 
 the huffers of india-rubber, designed to receive the recoil 
 when the carriage is permitted to recoil the whole length of 
 the slide. 
 
 "The middle of the slide rests on a rail on the deck 
 designed to support it at that point, and on which it slides 
 \vhen training. 
 
 " The slide has one transom half way of its length riveted 
 to the inner sides of the rails, and on a plane six inches be- 
 low their upper surface. 
 
 "To an angle-iron turned up from the rear of the tran- 
 som, and rising to the level of the rails, is bolted the rear 
 
CHAP. XXXVIII. SrAMSn GUXBOAT PIVOT-CARIirAGES. 513 
 
 vud (if the frictiu)i-lHii; six inclics wide aud inie aiul a (quarter 
 iiidu's tliick, wliicli is coutiuued horizontally to the forward 
 lu'urter, -where it is bolted to a castiiiLr, after jiassing whicli 
 it inclines downwards gradually to the i)iv(>t, where it is 
 secured to the pivut-l)()lt thn)UgIi a liole in its end. 
 
 "A composition rack is bolted on the inside of the riglit 
 rail, the teeth extending above the level of the rails. 
 
 "The carriage rests on four rollers front and rear, the 
 former of 18 inches diameter, the latter of 7 inches. To 
 the inner face of the right forward roller is bolted a work- 
 ing-wheel of composition, with its cogs gearing below into 
 the rack on the slide, while above it gears into a [)inion 
 on a shaft which lias its bearings in the brackets of tlie 
 carriage. A crank is attached to the end of this shaft on 
 the right side of the carriage, and by it the carriage is run 
 in and out on the slide. This shaft has a longitudinal mo- 
 tion, which allows the pinion to be geared or ungeart'il at 
 pleasure. It is always desirable to ungear before iirinii\ in 
 order to prevent motion of the crank, which might i)rove 
 dangerous to the gun's crew. 
 
 " A conveniently-arranged clutch holds the shaft in either 
 position. 
 
 "The carriage has one transom, from the forward i)ai-t 
 of which project two arms, one-third the width of the tran- 
 som apart, extending to a length of 20 inches, and termi- 
 nating in eyes, through which the compression-shaft passes, 
 which has bearings in the lower part of the brackets and 
 well forward of the forward axle of the carriage. 
 
514 SPANISH GUKBOAT PirOT-CAinHAGES. chap, xxxvm. 
 
 "Under the friction-bar is a clamp 17 inches long and 
 6 and 10 inches wide, which binds against the under face 
 of the bar. The compression is produced through the ec- 
 centric motion of a third piece resting on the uj)per clamp, 
 a side-elevation of which represents a half-circle, and which 
 is fitted over the compression-shaft. This eccentric piece is 
 connected with the friction-clamps by two iron straps, with 
 nuts screwed on the lower ends of them. It will be per- 
 ceived that the friction-clamps occupy a position in the 
 centre line of the carriage and between the ends of the 
 two arms projecting from the transom. The friction-clamps 
 are lined with hard wood, which forms the surfaces bind- 
 ing on the friction-bar. The compression-shaft has its bear- 
 ings on the brackets of the carriage, and projects far enough 
 outside the left bracket to receive a long lever which is 
 shipped on its end, and which has a vertical motion, limited 
 by the adjustment of the screw-nuts on the ends of the 
 iron straps which connect the friction-clamj^s. This lever 
 is held in position by a rack on the outside of the left 
 bracket when the required compression is attained. A steel 
 spring at the lower end of the lever binds it against the 
 bracket, and a very convenient eccentric arrangement at the 
 handle of the lever enables this pressure to be overcome 
 when desiring to move the lever." 
 
 Having prefixed this very clear and precise description 
 to his report, Commander Simpson proceeds: "During the 
 firing thus tabulated, the running-oiit gear was but seldom 
 used, the carriage being allowed to move obedient to the 
 
CHAP. XXXVIII. SPAXISir GUXBOAT nrOT-OAEEIAGES. 515 
 
 roll of the vessel, and its motiou was found to be perfectly 
 under the control of one man at the compression-lever, who 
 could check it at any point. The compression being found 
 to work well in deliberate fire, thirty rounds were fired to 
 test the point whether rapid fire would cause the heating 
 of the friction-bar. The thirty rounds consumed nearly 
 thirty minutes in firing, at the end of which time the tem- 
 perature of the bar was slightly raised, but in no way inter- 
 fei-cd \\ith a continuance of firing. Very rajtid firing may 
 be done with this carriage; the time consumed in firing the 
 thirty rounds above mentioned was in consequence of the 
 crew not being accustomed to gun-exercise. 
 
 "The most promiiR'nt advantage — in fact, the essential 
 characteristic — of this carriage is its system of compression, 
 wliich is complete and instantaneous. 
 
 "The compression in use Avith our pivot-guns and with 
 our turret-guns invohes the use of a screw, which requires 
 time to work ; the substitute provided in this carriage is a 
 simple motion in a vertical plane of a lever, which is in- 
 stantaneous in action, and quite as effective in its result. 
 
 "The tardiness of action in the compression of our tuiret- 
 guns may often cause hesitancy in casting them loose in a sea- 
 way, when, with a more speedy means of compression, they 
 might be made of service. The slow and imperfect action 
 of the compressoi-s fitted to our pivot-guns renders necessaiy 
 eccentric rollei-s to the axles, so that the carriage may be let 
 down on the slide, to increase, by the increased surface in 
 contact, the friction that the compressors do not supply. 
 
516 SPANISH GUNBOAT PiVOT-CAEEIAGES. chap, xxxviii. 
 
 "The system of compression doav under consideration ad- 
 mits of keeping tlie carriage always on its rollers, thus 
 simplifying tlie mechanism of the carriage, and dispensing 
 with the levers which are now necessary to bring the rollers 
 in and out of action. The four men now devoted to this 
 duty could be dispensed with. 
 
 " During the experiments here reeoixled the carriage has 
 fulfilled the advantages claimed for it by its inventor, and, 
 unless subsequent experiments or the experience of actual 
 service should develop defects not now apparent, its claim 
 for preference over any carriages now in use in the navy 
 must be allowed. 
 
 " During the firing, the shortest distance at which the 
 recoil was checked was 2 feet 5 inches, which was only 
 half the recoil that would be required in service so as to 
 have the gun in position for loading. If less recoil were 
 required at any time, it can be obtained by a change in the 
 adjustment of the screw-nuts on the strap binding the fric- 
 tion-clamjx" * 
 
 * It will be proper to notice that the new system has proved so successful in 
 practice that the Spanish Government, in addition to the thirty carriages and slides 
 mounted on board of the new gunboats, have recently ordered several sets of similar 
 carriages and slides for other vessels. 
 
CHAPTER XXXIX. 
 
 KOTAKY GUX-CARRIAGE AND TRANSIT rLATFORM. 
 
 AP1>I.IED TO THE SPANISH GUNBOAT TORNADO.* 
 
 (SEE PLATE 59.) 
 
 Tin; illustration on the plate referred to represents a new 
 .system of transferring the battery from side to side, witlK)nt 
 resoiting to the complicated method of pivoting practised 
 in our ve.ssels of war. In addition to the advantage of rapidly 
 transfen-ing the guns from side to side, an all-r<aind lii-e is 
 also secured by this system, as will be seen by the following 
 description : 
 
 The leading feature of the device is that of placing the 
 gun-carriage and its rotary slide on a circular platform, sup- 
 ported on four cylindrical rollei-s (partially shown at //) 
 provided with flanges like those of the wheels of railwjiy 
 
 • It will be remembered that in 1873 this gimboat. Bftcr a spirited chase, cap- 
 tured the American steamer Virginia, having on board a company of " Cuban 
 patriots" and war material intended for the Cuban insurgents. The capture of this 
 vessel, it will also lie remembered, very nearly led to a war between the United States 
 and Spain. 
 
 617 
 
518 BOTABY GUN-CARBIAGE AND PLATFOBM. cn.\p. xxxix. 
 
 carriages. These rollers, the axles of wliicli turn in appro- 
 priate bearings under the platform, move on two flat pai'allel 
 bronze rails, I and m, secured to the deck at right angles to 
 the line of keel. One of these rails, w, is provided with 
 cogs on the outside, thus forming a toothed rack. A small 
 horizontal cog-wheel under the platform is geared into the 
 said rack, and actuated by a set of cog-wheels arranged as 
 in ordinary lifting-jacks. The gear is put in motion by a 
 vertical spindle having a hand-wheel attached to its ujiper 
 end, the lower end being made to fit a square socket, n, 
 formed in the axle of the actuating pinion of the gear. It 
 is hardly necessary to observe that after having transferred 
 the 2^kitform to the desired position, the vertical spindle 
 should be lifted out of its socket and removed, in order not 
 to interfere with the free rotation of the slide. It may be 
 mentioned that in addition to the gear referred to, suitable 
 eyebolts are ajiplied to admit of employing ordinary tackle 
 in transferring the platform from side to side. Having been 
 I'olled into position on the fighting side of the vessel — say 
 starboard — the platform must, of course, be secured by two 
 fighting-bolts. One of these is seen at Tc, the second of the 
 pair being concealed by the slide. On rolling the platform 
 to port, after having removed the fighting-bolts, the latter 
 will be inserted through the bolt-holes of the lugs g on the 
 opposite side of the platform. It should be observed that 
 in housing the gun the fighting-bolts will occupy diagonal 
 positions — plates, as shown at 1c, being inserted in the deck 
 accordingly. 
 
cu.vr. xxxix. BOTAEY CVK-CAKinAGE AND PLATFOUM. 519 
 
 It was pointed out in the previous chapter tliat if the 
 force of the recoil were brought to bear on the central pivot 
 round which the slide revolves, the latter would be lifted 
 uj. violently, or seriously jarred at the instant of discharg- 
 ing the gun,, since the vertical line, passing through the 
 centre of the trunnion, is far in advance of the central pivot 
 when the gun is rolled out. To prevent such lifting or 
 jari-ing, a very effective expedient has been resorted to— viz., 
 that of attaching a l)racket d at the forward end of tlie 
 slide, extending about two inches below its base, and bearing 
 fiimly against the circumference p of the platform. It will 
 be readily seen that if the bracket referred to, which acts 
 as a hook, be placed at a proper distance, while the pivot 
 round which the slide turns fits loosely in its socket, the 
 force of the recoil will be received wholly by the edge 2> 
 of the platform, at the point where the bracket d bears. 
 The professional reader cannot fail to perceive the advan- 
 tage of transferring the strain from the central pivot in rear 
 of the trunnion to a point in advance of the same. Obviously, 
 the practical result will be that of causing the carriage and 
 slide to bear down against the platform, instead of being 
 violently jarred or lifted up, as it would be if the force of 
 the recoil were brought to bear on the central pivot. Re- 
 garding the proper position of the fighting-bolts for securing 
 the platform during firing, it will be evident that if inserted 
 at I; as shown (for the sake of ready explanation) in the 
 illustration, the platform would be lifted or jarred at the 
 moment of discharging the gun; while by inserting the fight- 
 
520 IWTABY GUN-CARRIAGE AND PLATFORM, chap, xxxix. 
 
 iug-l>()lt at <j the teudeucy will be to depress the platform, 
 thereby securing perfect repose throughout the eutire struc- 
 tui-e. 
 
 The gun-carriage itself having been minutely described 
 in Chap. XXXVIII., it will only be necessary to observe 
 that the friction-bar c, which in the carriage referi'ed to was 
 made of iron, has been made of bronze in all the recent 
 carriages constructed under my patents. It was found in 
 practice that the application of oil or grease to this bar, 
 indispensable to prevent its corrosion, considerably diminished 
 the friction — a circumstance of no importance if the dimi- 
 nution had been constant in amount ; but the lubricating 
 medium, varying in quantitj-, obviously causes irregularity in 
 the intensity of the required friction. Bronze being now 
 substituted for iron, renders the application of grease un- 
 necessary ; hence the friction between the wooden lining of 
 the clamp h and the bar c becomes uniform. Accordingly, 
 by pushing the hand-lever a to a given notch of the circular 
 tooth-rack s, it has been found that the length of the recoil 
 may be regulated with remarkable accuracy. 
 
CHAPTER XL. 
 
 GUN-CAKRIAGE FOR COAST DEFENCE. 
 (SEE PLATE 60.) 
 
 Engineering of March 20, 1872, published the following 
 article, headed " Coast Defence " : 
 
 " The problem of how best to defend our coasts has of 
 late attracted so much attention that the illustration which 
 we publish, representing a gun-caniage forming part of a 
 new system of coast defence planned by Captain Ericsson, 
 will possess special interest. The gun is mounted behind 
 an inclined movable shield of solid plate iron, 4 ft. high 
 above the parapet, so arranged that the muzzle is unmasked 
 at the moment of firing, the shield affording protection against 
 projectiles striking above the parapet. It is asserted that this 
 plan is more efficient than that of Moncrieff ; but, as it is not 
 our intention at present to institute a comparison between 
 the rival systems, we proceed at once to describe the illus- 
 trated carriage (see Plate 60). The recoil is checked by 
 
522 
 
 OUN-CARRIAGE FOB COAST DEFENCE. chap. xl. 
 
 friction produced by clamping two longitudinal Lars secured 
 in tlie middle of the slide, tlie compression being effected by 
 a transverse axle, on the principle adopted in the Princeton 
 carriage. The longitudinal bars, however, are composed of 
 bronze in place of wood. The friction-clamp (a longitudinal 
 section of which, through the vertical plane, will be found 
 represented in the accompanying delineation is firmly attached 
 to the front of the carriage. The faces of the fiictiou-bars 
 
 being smooth and true, and the wooden linings of the clamp 
 accurately fitted, it will be perceived that a very slight move- 
 ment of the upper and lower parts towards each other, after 
 contact with the bars, will at once cause friction, provided 
 that the set-nuts of the clamp-cap have been properly tight- 
 ened. In order to comprehend correctly this peculiar system 
 of compression, let us suppose that the clamp, as well as 
 the bolts which hold the same together, are perfectly rigid, 
 and that the transverse axle is flattened so as to make it but 
 slightly oval. Turning the latter through an arc of 90 deg. 
 
CHAP. XL. GUK-CABEIAGE FOJi COAST DEFENCE. 523 
 
 will, uinlt'i- these conditions, obviously produce an enoraious 
 pressure. It Las been objected that, if the axle is but slightly 
 oval, the wear of the wooden linings and the faces of the 
 frictiou-bai-s will soon destroy the efficacy of the compres- 
 sion. To meet this objection the constructor has placed the 
 clamp conveniently in front of the carriage, in order that the 
 set-nuts may readily be screwed down whenever rei^uired. 
 Practice, it appears, has shown that the wear is insignificant 
 with friction-bars composed of bronze. Iron bai-s, on the 
 other hand, owing to corrosion and consequent abrasion, tend 
 to cut the wooden lining unless grease be applied — an expe- 
 dient inadmissible, as it greatly diminishes the desired adhe- 
 sion, besides rendering the same very irregular. It will he 
 seen in the illustration on Plate 6() that the hand-lever by 
 means of which the compi-ession is applied or relieved is held 
 in position by a segment (provided with notches) attached 
 to the side of the carriage; while a spring secured to the 
 end of the transverse axle presses against the outside of the 
 lever, thereby preventing the same from leaving the notches 
 dming the recoil of the gun. 
 
 " Captain Ericsson's ])lan of employing a series of friction- 
 bai"s one above the other, or placed side by side, as in the 
 monitoi-s, appropriately termed the ' multiplex system,' copied 
 by Captain Scott, Sir William Armstrong, and othei-s, involves 
 a paradox ^\•hic•h merits special notice — viz., tliat without 
 employing additional force, any amount of friction may be 
 produced. It \vill be evident on reflection that whatever 
 number of bars be employed, the friction between the face 
 
534 GUN-CAEEIAGi: FOB COAST DEFENCE. chap, xl, 
 
 of eacli and the face of the clamp will depend solely on 
 the pressure applied. Hence, if four bars, presenting eight 
 surfaces, be employed, the retarding force opposing the recoil 
 of the gun will be quadruple that of one bar presenting 
 two surfaces. With reference to running out the gun, the 
 following explanation will suffice : The forward tracks, situ- 
 ated somewhat in advance of the trunnion, are keyed to an 
 axle supj)orted by bearings in the side frames of the carriage. 
 A cog-wheel is attached to the said axle, into which a small 
 pinion, operated by an ordinaiy crank-handle, is geared. 
 
 " Before adverting to the mechanism of the slide, it will 
 be well to observe that light gear of small multiplying power 
 Avill answer for training, since the movement is always in 
 the horizontal plane. It will be noticed that the exterior 
 segment is provided with a projecting rack in the middle 
 — an expedient which overcomes the difficulty so frequently 
 experienced on the old plan of employing separate racks, 
 that settlement of the segment on which the rollers run 
 occasions binding between the cogs of the driving-pinion 
 and those of the rack. It is scarcely necessary to remark 
 that the rollers applied on the opposite sides of the toothed 
 projection of the exterior segment turn freely on the axle 
 to which the driving-pinion and the vertical conical wheel 
 are attached. The action of the small conical pinion geared 
 into the wheel mentioned, operated by means of the hori- 
 zontal hand-wheel, requires no explanation." 
 
CHAPTER XLI. 
 
 THE THIRTY SPANISH GUNBOATS AND THEIR ENGINES. 
 (SEE PLATK 61.) 
 
 The material aid which the Cuban insurrection derived 
 from outside sources compelled the Spanish Government in 
 1809 to build a fleet of swift gunboats to form a cordon 
 round the island jis the only means of preventing blockade- 
 runners from conveying men and warlike stores to the insur- 
 gents. Admiral Malcampo, naval commander in Cuba, was 
 accordingly instructed by his Government to procure the 
 needed vessels in the United States. The Admiral succeeded 
 in canying out his instructions with extraordinary prompt- 
 ness, as will be seen from the following concise statement, 
 copied from the Army and Navy Journal of Isovember 20, 
 1869: 
 
 " The annals of naval construction probably furnish no 
 instance of greater diligence than that displayed in the pro- 
 duction of the thiity Spanish gunboats now floating on the 
 
52G SPAXISE GUNBOATS AND THEIR EN6TXES. chap. xli. 
 
 Hudson. Tlie planning Laving been entrusted to Captain 
 Ericsson, the contract for building the fleet was entered into 
 with the Delamater Iron- Works, in this city, on the 3d of 
 May, 1869. On the 19th of May the fii-st keel was laid, 
 and on the 23d of June the first vessel was launched from 
 Pouillon's ship-yard — thirty-four working days after laying 
 the keel. September 3 — just four months from the signing 
 of the contract, and three months and sixteen days after lay- 
 ing the first keel — the last vessel of this fleet was launched, 
 at which time fifteen of the vessels previously launched had 
 engines and boilers on board ! 
 
 " The Spanish gunboats are sea-going twin-screw vessels, 
 107 feet long on the water-line, 22 feet 6 inches extreme 
 beam, 8 feet depth of hold, and draw 4 feet 11 inches when 
 fully equipped for service, with coal, stores, and ammunition 
 for 100 rounds on board. The lines at the bow are some- 
 what full, in order to sustain a heavy bow gun, the breadth 
 of the deck being carried well forward for the purpose of 
 facilitating the manipulation of this gun, of which we will 
 speak presently. The run is very clean, the lines being 
 deemed faultless for a t\vin-screw vessel. The construction 
 of the hull presents two novelties ^vorthy of special mention. 
 The apparently insoluble character of the problem — a gun- 
 boat of this class drawing only 59 inches of water when fully 
 equipped for service — compelled the designer to dispense with 
 the keel. Shipbuilders, it appears, at first objected to this 
 innovation, but now admit that these gunboats may take 
 ground with far less risk of straining and leaking than 
 
CUAP. XLI. SPANISU GUyBOATH AM) THElIi EyOINES. 527 
 
 oriliiiary liglit-ilrauglit vessels witli their weak keels. The 
 other novelty iilliuled to is the cutting down the rail and 
 substituting a low, heavy tindjer bulwark at the Ijow, pro- 
 vided with substantial \vater-ways and lined with sheet-iron 
 to admit of firing the gun e/i barbette. 
 
 "In addition to their ample steam-power, the Spanish 
 gunboats carry full amount of canvas, being schooner-rigged, 
 with yard and square-sail on the foremast. "Wire-rigging 
 having been adopted, and the masts and smoke-pipe raked 
 more than usual, the appearance of these twin-screw vessels 
 is peculiarly light and saucy. Considering their great num- 
 ber, swiftness, light draught, and the long range of their 
 guns, it is evident that the Spaniards will l,e enal)led iov 
 the future to prevent ett'ectually incursions un the Cuban 
 coast. 
 
 "As might be expected, the .steam ntachiiiery of these 
 novel war-vessels presents features of special interest (see 
 illustration on Plate Gl). It has frequently been urged as 
 an ol)jection against the twin-screw system that the double 
 set of engines, four steam-cylinders, with duplicates of all 
 their working parts, called for in this system, render the 
 whole too complicated and heavy for small vessels ; prevent- 
 ing at the same time the application of ■surface-comhnsation. 
 The designer has overcome these objections by inti-oduciug 
 a surface-condenser which, while it pei-forms the function of 
 condensing the steam to be returned to the boiler in the 
 form of fresh water, serves as the principal support "of the 
 engines, disjjensing entirely with the usual frame-work. Be- 
 
528 SPANISH GUNBOATS AND THEIB ENGINES. chap. xli. 
 
 sides this expedient, eacli pair of cylinders have their slide 
 frames for guiding the movement of the piston-rods cast in 
 one piece. Altogether, the combination is such that the total 
 weight and the space occupied by these novel tvsdn-screw 
 engines do not exceed the ordinary single-screw engines of 
 equal power. Several improvements connected with the 
 Avorking-gear have also been introduced. The outer bear- 
 ings of the propeller-shafts, always difficult to regulate and 
 keep in order on the twin-screw system, are self-adjusting, 
 and accommodate themselves to every change of the direc- 
 tion of the shafts. This is effected by their being spherical 
 externally, and resting in corresponding cavities in the stern 
 braces or hangers. The 'spring-bearings,' for supporting the 
 middle of the shafts, are also arranged on a similai' self- 
 adjusting principle. The thrust-bearing, which receives the 
 pressure of the propeller, is a peculiar construction, the 
 arrangement being such that the bearing surfaces remain in 
 perfect contact, however much the shaft may be out of line. 
 The reversing-gear, likewise, is quite peculiar, insuring com- 
 plete control over the movement of the two propellers under 
 all circumstances. It is claimed that these engines are the 
 lightest and most compact yet constructed for twin-screw 
 vessels. 
 
 " The internal arrangements and fittings show thorough 
 knowledge and experience on the part of the superintending 
 officer. Our friends on the Baltic, who pride themselves on 
 knowing moi-e about gunboats than other nations, will be 
 astonished when they learn how the Spaniards fit out such 
 
CHAP. XLI. SPANISH GUNBOATS AND IHEIB ENGINES. 529 
 
 vessels. Indeed, tlio e«iuipmeiit more resembles that of a 
 yacht than that needed for a plain gunboat. We cannot 
 afford space for a specification, and therefore proceed to 
 notice only that which is essential. The coal-bunkers are 
 placed on each side of the boiler, extending equally forw aid 
 and aft of the centre of displacement of the vessel, in order 
 to preserve pei-fect trim, whether the bunkers are full or 
 empty. The magazine, located in the centre of the vessel 
 between the engine-room and the officers' quartei-s aft, is 
 lined with lead on the inside, with the unusual precaution 
 of having the outside protected by sheet-iron. There are 
 three distinct modes of flooding this magazine — viz., from 
 the sea, by a powerful hand-pump, and by the donkey-engine 
 pump. In addition to the ordinary water-tanks, a 'fresh- 
 water maker' of ample capacity is provided, in which the 
 condensation of the steam is effected by the current of sea- 
 water which passes through the surface-condenser; the fresh 
 water being drawn off through a bent -pipe on deck. A 
 combined capstan and mndlass of novel construction, suffi- 
 ciently low to fire over, is bolted to the deck over the 
 chain-locker at the bow, the combination being such that 
 the capstan may be used alone, or one or both anchoi-s 
 raised at the same time. 
 
 "Respecting the armament, the following brief notice 
 must suffice for the present: It consists of a 100-pound rifle- 
 gun placed at the bow— a Pari-ott rifle— but a very different 
 weapon from that represented by the photographed frag- 
 ments which embellish so many pages of General Gillmore's 
 
530 SPANISH GUNBOATS AND TEEIB ENGINES. chap. XLl. 
 
 famous book. Briefly, it is an improved Panott 100-j)oim(l 
 rifle, witli wrouglit-iron hoops round the chamber, cariied to 
 within three inches of the trunnion, the chase being increased 
 to correspond with the increased strength attained by the 
 extension of the re-enforce. The severe ordeal through which 
 the improved gun has passed during recent trials at Cold 
 Spring, conducted by the Spanish officers, promises so well 
 that no doubt this improved Parrott gun has a future. 
 
 " Of Captain Ericsson's new gun-carriage — on which the 
 improved gun is mounted — one of the leading features of 
 the Spanish gunboats, we have space for only a cursory no- 
 tice. It will be inferred from what has already been stated 
 that the intention is to fire over the bow and in line with 
 the keel. For this purpose, and in order to command a wide 
 horizontal range, a circular platform of wood, surrounded by 
 a brass ring of 12 feet 6 inches diameter, is bolted to the 
 deck at the bow. The gun-slide, composed of wrought iron, 
 provided with friction-rollers at both ends, rotates round a 
 pivot secured to the deck, in the centre of the said brass 
 ring. The carriage is made of light wrought-iron plates and 
 angle-iron, riveted together in such a manner as to ensure 
 great strength longitudinally as well as transversely." (See 
 illustration of gun-carriage on Plate 59.) 
 
 Having described the internal arrangements of the Span- 
 ish gunboats somewhat minutely, the Army and Navy 
 Journal concludes its notice of these vessels by saying : 
 
 " A fleet of thirty war-vessels precisely alike being by 
 no means an ordinary sight, a visit to Delamater's works on 
 
CHAP. XLI. SPAA'ISII GUNBOATS AND TUEIE ENGINES. 531 
 
 the Hudson where the saucy-looking craft are now stationed, 
 ten abreast, cannot fail to be very interesting to naval men. 
 It is a significant fact that this groat display of offensive and 
 defensive force is the result of the efforts of a single estab- 
 lishment, directed Ijy individual skill. Evidence more con- 
 clusive could not be furnished that the progress of the 
 country and its resoui-ces are equal to any future emer- 
 gency.'' 
 
CHAPTER XLII. 
 
 A NEW SYSTEM OF NAVAL ATTACK. 
 (SEE PLATE 62.) 
 
 A HEAVY body of regular form, whose density is greater 
 tlian that of atmosplieric air, moving latei'ally tlirough the 
 atmosphere, is inexorably under the influence of the earth's 
 attraction, and therefore describes a foreshortened parabolic 
 curve during its flight ; while a submerged body, the weight 
 of which is equal to the weight of the water it displaces, is 
 not affected by the earth's attraction ; and that consequently, 
 if 2)ut in motion under the surface of a quiescent fluid of unli- 
 mited extent, such a body will continue to move in a straight 
 line until the motive energy which propels it becomes less 
 than the resisting force of the surroundins; medium. 
 
 In virtue of the first part of this general proposition, a 
 heavy body may be projected in such a manner that the 
 termination of its trajectory shall make any desirable angle, 
 less than 45 deg., with the horizontal line, independently of 
 the length of the chord of the trajectory. In other words, 
 
 633 
 
CHAP. XLII. A XEW SYSTEM OF XAVAL ATTACK. 533 
 
 the body may he projected at variable distances over water, 
 and yet strike its surface at any desirable angle. This 
 important result is effected simply by varying the relative 
 proportion between elevation and strength of charge. The 
 second part of the stated general prop(^sition is of equal 
 importance. It points to the fact that the ti-ajectory may 
 be extended in a straight line under water, to any desirable 
 distance, irrespective of the speed of the submerged pro- 
 jectile. Accordingly, a shot may be projected from one 
 vessel towards another within moderate ranges in such a 
 manner that it shall dip into the water at a considerable 
 distance from, or close to, the vessel assailed, independently 
 of the distance between the two vessels. Also, that the 
 shot may be projected at such an angle that the pi-olono-a- 
 tion of its trajectory in a straight line, after contact with 
 the water, shall strike the hull of the vessel assailed at any 
 desirable depth below the surface. 
 
 That a certain relation between charge and elevation 
 enables us to project a spherical shot in such a manner as 
 to strike the water at any desirable distance from an oppo- 
 nent's vessel, at angles within 45 deg., needs no further 
 demonstration. Hence, if the trajectory be such that its 
 e.xtension in a straight line from the point of contact with 
 the water leads to the hull of the vessel assailed, the latter 
 will be hit — on condition, however, that the shot is not di- 
 verted from its course on entering the water, and provided 
 its vis viva be sufficient to overcome the resistance encoim- 
 teied during its passage through the water. These indis- 
 
534 A NEW SYSTEM OF NAVAL ATTACK. chap. XLII. 
 
 pensable conditions, especially the first-named, wliicli appa- 
 rently cannot be complied with, show the ditficulty of hitting 
 a vessel below the Avater-line. And if we suppose that the 
 projectile is not spherical, another serious difficulty presents 
 itself. An elongated body will not bend to the curvature 
 of the trajectory during the flight through the air, but re- 
 tain during its course the same inclination as the gun from 
 which it has been projected ; hence it will fall nearly flat 
 on the surface of the water when striking. 
 
 Agreeably to our general proposition, a regular body, 
 weighing as much as the water it displaces, is independent 
 of the earth's attraction ; but there is another force which, 
 notwithstanding the absence of any gravitating tendency, 
 will cause a body of regular form moving under water to 
 deviate from a straight line and rise to the surface. A 
 cone moving in the direction of its apex and in the line 
 of its axis horizontally, or on an incline, will, owing to the 
 inertia and the nearly incompressible nature of water, more 
 readily displace the column which rests upon and depresses 
 its upper half than the column from below with its lifting 
 tendency. Consequently, the course of the conical body 
 will be diverted from the straight line upwards, describing 
 a curve nearly elliptical, and quite sudden, if the speed be 
 great. A cylinder with serai-spherical ends will, from the 
 same cause, ascend to the surface if moved in the line of 
 its axis ; while a cylinder with flat ends will take a down- 
 ward course, gradually increasing its inclination, until at last 
 the axis assumes a vertical position. Obviously, the lower 
 
CHAP. XLII. A JV^ir SYSTEM OF NAVAL ATTACK. 535 
 
 part of the forward flat end encountei"S a greater resistance 
 than the upper part ; hence tlie lower half of the transverse 
 section of the cylinder suffers an excess of retardation, which 
 occasions the downward course described. 
 
 The question whether the apparently insuperable diffi- 
 culties thus pointed out can be overcome by mechanical 
 expedients, has occupied my attention for a long time ; and 
 numerous experiments have been made to test the efficacy 
 of certain forms suggested by theoretical considerations. 
 These forms being correct, the direction of the projectile 
 during its flight through the air w^ll be parallel with the 
 trajectory, and on entering the water it will not be diverted, 
 but continue to move under the surface with the same incli- 
 nation it had on coming in contact with the dense medium. 
 
 The illustrations on Plate 62 present the main features 
 of the system under consideration so distinctly that it will 
 be superfluous to enter on a general explanation of the 
 nature of the scheme. It should be stated, however, that 
 the elongated projectile is charged with dynamite or gun- 
 cotton, and provided with a percussion-lock at the for\vard 
 end, which explodes the charge by contact. It may be men- 
 tioned that numerous plans have been suggested during the 
 last few years for projecting solid shot under water, for the 
 purpose of sinking ships. In several instances these plans 
 have been carried into practice, with the invariable result 
 that the resistance of the water has been found so great, 
 even at short distances, that an ordinary wooden hull has 
 proved to be impenetrable. The plan now uuder conside- 
 
536 A NHW SYSTEM OF NAVAL ATTACK. chap. xlii. 
 
 ration bears uo resemblance to these projects, since tlie force 
 of the projectile on reaching its destination need only be 
 sufficient to actuate the trigger which causes the ignition of 
 the explosive charge. 
 
 Apart from the theoretical considerations relating to the 
 course of elongated projectiles under water, the practical 
 question of motive poiver to propel the same claims our 
 attention. It is hardly necessary to state that the force 
 relied upon is the vis viva possessed by the projectile on 
 coming in contact ^vlth the water. Before estimating this 
 force, it will be proper to call attention to the fact that 
 the new system, to be successful, does not call for attack 
 at a great distance, provided the vessel from which the 
 missile is projected has greater speed than the opponent, and 
 at the same time adequate protection against his artillery. 
 Hence the destruction of the vessel assailed would be as 
 certain if the distance of 500 ft. were the limit as if a range 
 of 5,000 ft. better suited the new system. It will be inferred 
 from this explanation that, although there is uo special limit 
 Avithin ordinary ranges, the plan is to attack at distances not 
 much exceeding 500 feet, unless the sea be very smooth. 
 
 The vis viva of a projectile 15 in. in diameter, of such 
 a length that it displaces 500 lbs. of water, may be readily 
 estimated if we suppose the charge of powder in the gun 
 to be so regulated that the speed on entering the water will 
 be 400 ft. per second, necessary to furnish sufficient motive 
 
 power; thus -— = 2500 X 500 = 1,250,000 ft.dbs. A cylin- 
 
CHAP. XLii. A M:W system of XAVAL attack. 537 
 
 (Iric'ul body, 15 in. in diameter, with semi-splieiical cuds, 
 moving at a rate of 100 ft. per second under water, requires 
 a constant motive force of somewhat less than 1,500 lbs. 
 Assuming, then, that the projectile passes through 150 ft. 
 of water — the mean distance represented liy tlie diagram — 
 we have a resistance of 150 X 1500 = 225,000 ft.-lbs. to 
 overcome. The motive force, it will thus be seen, is more 
 than five times greater than the resistance; consequently, 
 no doubt can be raised as to the adequacy of the motive 
 power furnished by the vis viva of the projectile. It should 
 be observed that the resistauce is very great at first, and 
 that the speed diminishes in a very rapid ratio ; but it would 
 be futile to present a formula expressing the ratio of speed 
 and resistance, since the form of the body (withheld for 
 obvious reasons) is the chief element in the calculation. 
 Let us bear in mind that, while the resistauce against a 
 blunt body is exceedingly great, one provided with a sharp 
 point readily enters the \vater, even at the rate of 400 ft. 
 per second. 
 
 With reference to the gun, it should be mentioned that 
 the very low speed of the pix)jectile, and the consequent 
 small charge of powder needed, render heavy metal unneces- 
 sary. Besides, slow-burning cake-powder, contained in cellu- 
 lar caitridges, will be emph)yed, in order to check rapid 
 ignition, and in order to sustain a unifonn pressure duiing 
 the discharge. By reference to our illustration, it will be 
 seen that the guns are loaded from below, and for that pur- 
 pose so ari'anged as to admit of lacing depressed GO deg. 
 
538 A Ni:W SYSTEM OF NAVAL ATTACK. chap. xlii. 
 
 Gnu-carriages are dispensed with, the trnuuious being sns- 
 peuded by adjnstable pemlnlum-links secured under the 
 turret-roof. The recoil is checked by buffers attaclied to 
 the turret wall in rear of the breach. 
 
 It is jiroper to state that the method of loading guns 
 from below deck, as shown in our illustration, was planned 
 by me, and drawings of the same exhibited in New York 
 several years before it was claimed by certain American 
 engineers as their invention. 
 
 Respecting the safety of the charge in the shell from 
 ignition during the discharge, it should be observed that 
 recent impi'ovements in torpedo practice effectually prevent 
 such accidents. With reference to the calibre, it is evident 
 that this system of attack calls for dimensions that Avill 
 admit a projectile of sufficient capacity to contain a charge 
 which, by its explosion, will destroy a first-class ship of war 
 built on the cellular plan. Nothing short of 15-iu. calibre 
 will answer for this purpose. The American and Swedish 
 15-in. guns are admirably calculated for the purpose, although 
 they are unnecessarily heavy. 
 
 European savants, especially certain Swedish naval artil- 
 lerists, who have criticised my advocacy of the 15-in. guns, 
 \vill understand, on looking into this matter, \\\\y I have 
 persisted in advising the Scandinavians to cai'ry tliis large 
 calibre in their monitor turrets as the most effective weajjon 
 against their powerful neighbors. Assuredly the Danes will 
 have no cause to fear tlie Prussian Konhj Willielm or 
 FrkdricJi der Grosse, should their ports be defended by 
 
CHAP. xi.n. A XEW SYSTEM OF XAVAL ATTACK. 539 
 
 vessels aniieil witli guns, l)y means of uliirh sfvcral liiindivd 
 pounds of dynamite or gun-t't)tt()U could be exploded under 
 the hulls of the intruders. 
 
 Tlie important question of hitting tlie intended <)l)Ject 
 will be best answered by a eareful examination of the illus- 
 tration, wliieli eannot fail to convince experts that, in mode- 
 rate weather, the proposed projectile may be made to dip 
 at the pro[>er distance fi'om the opponent's vessel. The 
 different [)arabolic curves marked on tlie delineations on 
 Plate 62 clearly show that no great accuracy is called for, 
 and that the projectile may dip at various distances from 
 the vessel assailed, and yet strike the hull. It should be 
 observed that the vertical .scale is different from that of 
 the horizontal, in order not to place the vessels too far 
 apart for the limited size of the plate ; consequently, the 
 trajectory shown is considerably foreshortened. 
 
 The turret represented on the illustration, in which the 
 light 15-in. shell-guns are mounted, is composed of wTought- 
 iron plates of great thickness, the size of the structure being 
 sufficient to accommodate the two pieces, suspended, as already 
 stated, by pendulum-links secured under the I'oof. A massive 
 central shaft of wrought iron supjiorts the turret, on the plan 
 adopted in the monitors. The vessel designed to caii'v the 
 battery is a mere iron hull, eranimed with motive power, in 
 order to ensure high speed. The midship section is tri- 
 angular and tlie bow raking, as shown by the illustration. 
 The overhanging siiles aud deck are heavily armored. 
 
CHAPTER XLIIL 
 
 SUBMARINE WARFARE— THE MOVABLE TORPEDO. 
 (SEE PLATE G3.) 
 
 It was stated as a general proposition, in tte preceding 
 chapter, that a heavy body, of regular form, projected late- 
 rally through the air, commences to fall from the instant 
 of leaving the muzzle of the gun, describing during its pro- 
 gress a parabolic curve considerably foreshortened owing to 
 atmospheric resistance. But a body of regular form, pro- 
 jected under the surface of water or other fluid, in a hori- 
 zontal or inclined direction, will move in a straiglit line, 
 provided its sjiecific gravity be equal to that of the fluid. 
 In other words, a heavy body moving through the atmospliere 
 is under the influence of the gravitating force of the earth ; 
 while a submerged body, the weight of which is equal to 
 its displacement, is not affected by gravitation. If put in 
 motion under the surface of a quiescent fluid of unlimited 
 extent, such a body will continue to move in a straight liue 
 
CHAP. XLIII. THE MOVABLE TOUrKDO. 541 
 
 until tlie motive energy which propels it liei-onies less than 
 the resisting force of the surrounding niediiini. 
 
 Starting with these cardinal propositions, I entered, some 
 thirty years ago, on the task of solving the problem of sul)- 
 marine attack — viz., the propelling or projecting below the 
 surface of the water of an elongated body containing explo- 
 sive substances to be ignited when reaching some point under 
 the bottom or bilge of an opponent's vessel. The best method 
 of carrying out the idea is that of projecting the elongated 
 body by means of a tube or chamber with parallel sides 
 applied near the bottom of the aggressive vessel. Such a 
 method I proposed to the Emperor of France in the month 
 of September, 1854, Jis mentioned in Chap. XXVIII. 
 
 At close quartei-s the stated plan of attack will unques- 
 tionably be found very effective — indeed, infallible ; but, 
 unless the ojiponent's vessel can be approached very near, 
 it will prove abortive. Oljviousl)^, if the projectile be pushed 
 out in any direction not parallel with the line of keel while 
 the aggressive vessel is in motion, a side resistance will be 
 offered by the stationary water of the sea, \vhich will divert 
 the course of the missile the instant it is deprived of the 
 guiding power of the tube from which it is ejected. Cur- 
 rents will, from the same cause, change the intended course. 
 It need scarcely be observed that, in addition to the difii- 
 culty of controlling the direction of the projectile, tlie force 
 imparted to the same, whether steana or compressed air, will 
 be insufficient to propel it to any considerable distance. In 
 order to meet these sertous practical objections — viz., that 
 
542 THE MOVABLE TOEPEDO. chap, xliii. 
 
 the projectile ('.iiiiint l)e propelled far enougli, and that its 
 course cauuot be controlled — I have resorted to a device by 
 which any desirable amount of pi'opulsive force may be im- 
 parted, irresjiective of the distance traversed, and by which 
 the course of the missile is undei' perfect control dui-iiig its 
 progress to the intended point. Persons of a mechanical 
 turn of mind in almost every country have for a long time 
 been engaged in contriving torpedoes to he propelled under 
 water by independent motive power of various kinds, for 
 the purpose of blowing up vessels. The Austrian torpedo, 
 urged through the water by means of compressed air, may 
 be classed as one of this numeroiis tribe, the reported ter- 
 ]'il)le nature of which has from time to time frightened naval 
 constructors, and amazed some uumechanical sailors who have 
 witnessed the trials, and found that the mysterious body 
 actually can move under water. Proper investigation of the 
 subject, however, exposes imperfections of the Austrian tor- 
 pedo which i-ender its final success problematical. 
 
 It should be borne in mind that atmospheric air com- 
 pressed, so as to exert a pressure of 300 lbs. to the S(|. in., 
 weighs nearly 2 ll)s. to the cubic ft. Consecpiently, the 
 amount of motive force which the torpedo is capable of 
 containing will be found wholly insulficient for its effective 
 jiropulsion unless an imjiracticable or, at any rate, dangei'ous 
 pressure be employed, accompanied bj" great weight, seiiousl}'' 
 intei-fering with buoyancy, while the want of means for direct- 
 ing the torjiedo to the desired point presents an insupei'able 
 objection. As before stated, I have contrived a torpedo 
 
iii.w. XLiii. TUE MOVAHLE 'WHFEDO. 64^ 
 
 that limy Ik- [)ropflli'<I witli auy rc.niisite auiomit of force, 
 invspoi-tive of dLstaiu-e, the course of which is uiuler perfect 
 control, iiotwitli.staiuliug currents, and which may be directed 
 with i)erfect certainty to an object in motion. In contra- 
 distinction to the term [)rojcctiK', ai)[.lied to tlie structure of 
 1854, which was i)roi)elled alone by its via viva on leaving 
 the guiding-tube, I propose to apply the term torpedo to tlie 
 contrivance now to be considered. 
 
 It shoulil be observed that some attempts to prtipel 
 bodies under water have been successful as regards main- 
 taining a given depth. The self-evident device of ai)plyiiig 
 a tin or horizontal rudder, operated by a 2)iston or elastic 
 bag actuated I)y hydrostatic i)ressure, has suggested itself 
 to inventors. It will be readily perceived that an increase 
 or diminution of draught, attended as it is with a corre- 
 sponding variation of pressure, may be made subservient in 
 changing the inclination, thereby establishing a tendency of 
 the horizontal rudder either to elevate or depress the torpedo 
 during its forward motion. Thus, by a proper adjustment 
 and application of the hydrostatic pressure, the torpedo may 
 be made to move at auy desirable depth below the surface 
 of the sea. Nor lias any difficulty been experienced as 
 regards the instrument of propulsion in the experiments 
 made since the introduction of the screw pro]>eller. But 
 the difficulty of procuring the requisite amount of motive 
 force for actuating the propeller, and the aljsence of means 
 for directing the torpedo, have in each instance defeated 
 the object in view. 
 
544 THE MO VABLE TOliPEDO. ■ chap, xliii. 
 
 Before proceeding to consider the important question of 
 gniding the torpedo, I will uow briefly describe my method 
 of obtaining the required power for actuating the propellers. 
 A reel, of suitable diameter, revolving on a horizontal axle, 
 is applied near the chamber from which the torpedo is 
 ejected, one end of the axle being supported by a suitable 
 bearing, while the other enters an air-vessel through a 
 stuffing-box. The end thus inserted in the air-vessel is per- 
 foi'ated longitudinally for a short distance, and provided with 
 an ojjening in the side at the point where the perforation 
 terminates. A tubular rope, the bore of which is about one 
 iucli in diameter, composed of hemp and vulcanized i-ubber, 
 is connected with this opening, and then coiled around the 
 reel a certain niimber of times, and, lastly, connected with 
 the rear end of the torpedo. The air-vessel into -which the 
 perforated axle of the reel enters, being charged with com- 
 pressed air (by means of force-pumps worked by steam- 
 power), it \vill be readily understood that the compressed 
 air will pass through the axle, then through the several 
 coils of tubular rope wound round the reel, and ultimately 
 reach the rear end of the torpedo, where the rope is attached 
 to the engine wliich actuates the propellers. Accordingly, 
 the propulsion of the torpedo may be regulated by simply 
 o[)ening or closing the aperture of the perforated shaft within 
 the air vessel. Tlie rotation of the reel, consequent on the 
 onward movement of the torpedo, obviously cannot interrupt 
 the passage of the compi-essed air through the coils of the 
 tubular rope; hence the supply of motive force will continue 
 
CHAP. XLiii. 77/A' MOVMU.E ToniT.VO. 545 
 
 iiMdiiiiinisIied during the onward niovona-nt. Tlir tulmlar 
 rope being about one inch diameter in the bore, it will be 
 found by calculation that a quantity of compressed air, suffi- 
 cient to develop any desirable amount of power, may be 
 transmitted through it during the progress of the torpedo, 
 whether far oif or near the aggressive vessel. The arrange- 
 ment thus described being sufficiently simple to be com- 
 prehended without entering into detail, it will only be 
 necessary to state that the tubular rope, after leaving the 
 reel under the deck, is made to descend through a vertical 
 tube iuto the torpedo chamljer, iu order to prevent an 
 entrance of water at the point where the rope passes out. 
 Also, that tivo propellers are employed, revolving in opposite 
 directions round a common centre — indispensable to prevent 
 the torpedo itself from rotating when subjected to the 
 powerful torsion produced by a siiKjle projieller actuated 
 by the motive force which may be transmitted through a 
 tuljular rope of one-inch bore. 
 
 I will now proceed to describe my method of guiding 
 the torpedo, premising that the external casing which con- 
 tains the mechanism and explosive compouud is heavier at 
 the bottom than at the top, in order to preserve a vertical 
 position, and that, in addition to the horizontal rudders for 
 regulating the immersion, the torpedo is provided with a 
 vei-tical balauce-rui^lder for directing the lateral course. The 
 reel having a mean circumference of 10 ft., it will be seen 
 that the tu1)ular rope need only l)e coiled rouii<l it 100 
 times to admit of attack at a distance of l,0(in ft., probably 
 
546 THE MOVABLE TORPEDO. CHAP. XLIII. 
 
 far enough, since tbe position of tlie aggressive vessel which 
 carries the torpedo may be changed at all times with de- 
 sirable rapidity. 
 
 The apparently absurd proposition to direct and change 
 the course of the torpedo at will, on board of the aggressive 
 vessel, without external aid, is solved by the following simple 
 expedient: A small elastic bag, connecting the tubular rope 
 Avith the induction-pipe of the rotary engine, is attached to 
 the side of the tiller of the torpedo's balance-i'udder. As 
 the compressed air during its passage to the motor must 
 pass through the elastic bag, the latter will expand and 
 contract wdth every change of internal pressure ; and, as 
 such change will depend on the quantity of compressed air 
 admitted into the tubular rope, the expansion and contrac- 
 tion of the bag is evidently under perfect control. No-\v, 
 the power of this bag, or the powder of a loaded piston in 
 an open cylinder, to resist internal pressure, may be so pro- 
 portioned that when maximum pressure is admitted the 
 swelling of the bag, or the motion of the j)iston, will cause 
 the tiller to move about 20 deg. to port; and Avhen the 
 pressure is ' reduced 25 per cent., the accompanying contrac- 
 tion of the bag, or corresponding motion of the loaded pis- 
 ton in the open cylinder, will move the tiller 20 deg. to 
 starboard. Thus, by admitting moi'e or less compressed air 
 into the tubular i-ope, thereby changing the dimensions of 
 the bag or moving the piston referred to, the tiller will as- 
 sume any desirable angle within 20 deg. on either side of 
 the torpedo's centi'e line. 
 
CHAP. XT.iii. THE MOVABLE TOUrEDO. 5i7 
 
 Accordingly, the direction of the torpedo will be as com- 
 pletely under the control of the hand which admits the com- 
 pressed air to the tubular rope as if an intelligent directing 
 power resided within the torpedo itself. Probably no greater 
 mechanical feat than this can be instanced. The position of 
 the torpedo is indicated by a circular disc, four inches in 
 diameter, attached to the upper end of a perpendicular steel 
 wire, or mast, secured to the top of the torpedo. The said 
 disc is painted sea-green on the forward side and white on 
 the opposite side. It need scarcely be observed that the 
 explosion of the torpedo wnll sever the connection wath the 
 tubular rope, w-hich thus may be hauled in by turning the 
 reel. Should the intended object not be reached, the admis- 
 sion of compressed air to the tubular rope Avill be shut off, 
 and the torpedo hauled in, or sent out on a new errand. 
 
 The scope of the device thus described is, of course, more 
 limited than the scope of the method illustrated on PI. 62 ; 
 yet, had the Italians possessed it, the result at Lissa would 
 unquestionably have been revereed. No harbor can be entered 
 which is protected by it ; nor would any amount of vigilance 
 save vessels from destruction if approaching an enemy's coast 
 defended by it. 
 
 With reference to the reel on which the tubular lope is 
 coiled, it will be well to mention that it may be applied 
 within the toi'pedo itself; in which case the tubular rope, 
 instead of being (otved, will be paid out during the onward 
 motion towards the object intended to be struck. 
 
 The illustrations on PI. 63, Figs. 1 and 2, represent top 
 
548 THE MOVABLE TOBPEDO. chap, xliii. 
 
 vie^\" and side elevation of the actuating steam-engine, air- 
 compressing pump, air-vessel, and reel ; while Figs. 3 and 
 4 show the side elevation and top view of the torpedo. It 
 may be briefly mentioned that the torpedo thus represented, 
 after having been tested during a series of ti'ials in open 
 \vater, has been purchased by the Navy Department at Wash- 
 ington. Negotiations are now pending for the purchase of 
 the invention Ijy the United States Government, on conditions 
 of secrecy, as in the case of the Austrian torpedo «; hence 
 detailed information respecting the internal mechanism can- 
 not be presented in this work. 
 
CHAPTER XLTY. 
 
 TRANSMISSJOX OF MIX'IIAXKAL I'uWER BY COM- 
 PRESSED AIU. 
 
 (SEE PLATKS 04 AM) 05.) 
 
 Professor Barijard, in his admirable report of the Paris 
 Universal Exposition, observes " tliat, next in impoitance to 
 the creation of a new motive power, may lie jilaced any 
 material improvement in tlic methods of makiiiir available 
 the powers which we have. Natnre often fnrnishes ns with 
 such powers in abundance in situations whei-e they cannot 
 be conveniently converted to use. The positions of waterfalls 
 are determined by geographical accidents. These do not 
 always conspire with the causes which promote the growth 
 of to\\Tis and development of industries. If it were jiossible 
 to transfer the imnn-nse forces which are thus unprotitablv 
 ex])ending themselves to points where there are hands to 
 direct them, and materials on which to employ them, thev 
 might be productive of incalculal)le wealth, and of immea- 
 surable benefit to mankind." 
 
650 TBANSMIS8I0N OF POWER. chap. xliv. 
 
 Tlie foregoing views, so well expressed, are quite cor- 
 rect ; but there is another power running to waste which 
 the engineer, ere long, will be called upon to utilize — viz., 
 the 230wer of the tides. Already a prominent association 
 has been formed in France for erecting tidal motors on a 
 very large scale. Thus, while engineering skill has nearly 
 exhausted itself in endeavors to imjM'Ove the steam-engine, 
 a new field opens, boundless in extent, which will demand 
 far greater abilities than those called for within the narrow 
 bounds hitherto limiting the energies of the mechanical engi- 
 neer. The grand scheme of utilizing the natural forces now 
 running to waste divides itself into two distinct branches : 
 1st. The requisite mechanism for receiving the force exerted 
 by nature. 2d. The means for transmitting that force to 
 desirable localities. It is the latter branch which I propose 
 to discuss. But, before entering on the subject, it will be 
 proper to point out that it is not the natural forces alone 
 which the engineer is called upon to devise means for trans- 
 mitting. Indeed, with our present abundant supi:)ly of coal, 
 the transmission of force developed by steam will be most fre- 
 quently called for; since the steam-engine, however j^ortable 
 in its character, cannot be applied in all places where power 
 is required. The experience of late years has shown that 
 the substitution of mechanical power for manual labor in 
 driving tunnels and for mining operations has reduced the 
 cost and greatly increased the amount of work done in a 
 given time. But the presence of steam in tunnels and in 
 the galleries of mines is wholly inadmissible; hence small 
 
C11AI-. xi.iv. TlUXiiMli^SIOy OF I'OWEB. 55] 
 
 motive engines, operated hy compressed air, liave been intro- 
 duced for operating tlie luck-drills and. otliei- cutting tools. 
 Not oidy La;s the work by these means beeh. greatly acce- 
 lerated, but the escape of the exliaust air from tlie motors 
 has in a material degree tended to purify the atmosphere 
 within the mines, rendering the work healthful wliich for- 
 merly proved destructive to the miners. 
 
 The first question which presents itself iu treating of the 
 transmission of force by compressed air is the size of the tube 
 necessary to convey a certain amount of energy in a given 
 inuQ—presmre and velocity being the elements which deter- 
 mine the question. Fortunately, we are not without prac- 
 tical data on the subject, the engineers of the :Mont Cenis 
 tunnel having, some time ago, thoroughly investigated it. 
 The result of their labors has been recorded iu the Report 
 of the United States Commissionei-s at the Paris Universal 
 Exposition of 1S67. The Commissioners state that, at the 
 date of the report on the pi-ogress of the work in the tunnel 
 during the year 1863, the operation was carried on at a dis- 
 tance of nearly two thousand metres from the reservoii-s of 
 compressed air, and that nine borers were in operation witli 
 a force of two and a half horse-power each. The tube con- 
 veying the air w:vs very nearly eight inches in diameter, the 
 air being under a pressure of six atmospheres, and its ve- 
 locity in the tube three feet per second. The transmission 
 of the power under these very favorable conditions was at- 
 tended with no sensible loss, the pi-essure not being percep- 
 tibly less at the working extremity of the tube when all 
 
553 TEANSMIlSi<10X OF FOWEB. chap. xliv. 
 
 the perforations were in operation than when the machinery 
 ■was entirely at rest. 
 
 The Keport of the Commissioners furnishes a very full 
 account of the result of the exj^eriments conducted at Cor- 
 sica, in 1837, by order of the Italian Government, on the 
 resistance of tubes to the ilow of air through them. These 
 experiments were made previously to tlie commencement of 
 the work on the tunnel, the employment of compressed atmo- 
 spheric air as a motive power to actuate the boring a2:)paratus 
 being at the time considered a doubtful expedient. The Re- 
 port states that it was the aim of the investigation not only 
 to ascertain the absolute loss of force attending the trans- 
 mission of air through tubes of certain dimensions at certain 
 velocities, but also to determine what are the laws whicli 
 govern the resistance wlien the velocities of the air and the 
 diameter of the tube are varied. The following conclusions 
 Mere deduced from the experiments : 1. The resistance is 
 directly as the length of the tube. 2. It is directly as the 
 square of the velocity of flow. 3. It is inversely as the 
 diameter of the tube. 
 
 The fact before adverted to — that in the actual Avorking 
 of the machines in the tunnel no perceptible loss of power 
 was experienced at a distance of two thousand metres from 
 the reservoirs — must be attributed to the want of delicacy 
 of the manometer or pressure-gauge employed. Although 
 insignificant at moderate distances and low velocities, the 
 experiments at Corsica proved that the loss becomes serious 
 when the velocity and distance are considerably increased 
 
CHAP. XLIV. TBANSMISSION OF FOWER. 653 
 
 since, agreeably to the law before cited, the resistance varies 
 as the square of the velocity. Consequently, when the velo- 
 city is six times greater than the moderate rate of six feet, 
 or thirty-six feet per second, the resistance will be thirty-six 
 times greater, the power developed increasing in the ratio 
 of the volume of air delivered — viz., six times. It will be 
 perceived, therefore, that while the length and diameter of 
 a tube remain unaltered, and while the absolute resistance 
 opposed to the floAV of a current of air through it varies as 
 the square of the velocity, the relative resistance is only as 
 the simple velocity. It follows from the foregoing facts 
 that the power of compressed air varies as the product of 
 its pressure and its volume; hence, when the pressure is 
 constant, as the volume simply. But the volume delivered 
 varies as the velocity multiplied by the square of the dia- 
 meter of the tube. Now, as the resistance is inversely as 
 the diameter, and the volume directly as the square of the 
 diameter when the velocity remains constant, it follows also 
 that under a given pressure and velocity the relative resist- 
 ance (namely, the resistance divided by the power) will vary 
 invei-sely as the cube of the diameter. Obviously, therefore, 
 by enlarging the diameter of the tube, we may increase the 
 power transmitted, and at the same time diminish both the 
 absolute and relative resistance. In conclusion, I strongly 
 recommend engineei-s who may be called upon to transmit 
 mechanical power by compressed air not to aim at economy 
 by employing tubes of small diameter. 
 
 Having thus disposed of the fii-st branch of the subject 
 
554 TBANSMISSION OF POWEB. chap. xliv. 
 
 under consideration, let us now consider the meclianism 
 needed to compress the air to be transmitted. At first 
 sight the solution of the problem appears to be very simple, 
 but due reflection at once suggests to the practical mind 
 numerous difliculties. Considerations of weight, space, and 
 first cost, of course, demand the adoption of a double-acting 
 compressiug-cylinder ; hence the practicability of employing 
 double action is the very first question that pi'esents itself. 
 Now, in double-acting cylinders both ends must be closed, 
 consequently lubrication of the compressing piston must be 
 effected from without. Supposing that means for effecting 
 such lubrication have been devised (by no means easy), will 
 the packing of the piston be preserved and abrasion pre- 
 vented? In answering this question, we must bear in mind 
 that even at moderate pressure the compression of the air 
 generates a degree of heat which precludes the employment 
 of oil, as it quickly dries up and ultimately burns. Water, 
 if continually replenished, so as to make good the loss caused 
 by the formation of steam, may answer for a short time. The 
 dust drawn into the cylinder from the surrounding atmo- 
 sphere will, however, mix ^\-ith the ^vater, and soon form a 
 paste, resembling mud, on the piston, productive of friction 
 and abrasion of the cylinder incompatible with the functions 
 of a piston. The objectionable plan of compressing air by 
 rising and falling columns of water I do not propose to 
 discuss in this place. 
 
 The illustrations on Plate 64 represent a perspective view, 
 while Plate 65 shows a longitudinal section of a machine 
 
CHAP. XLIT. TIiAN8MTS8I0X OF POWER. 555 
 
 for compressing air, in which the difficulties before referred 
 to have been effectually overcome; the leading features beino- 
 that the compressing cylindei-s, open at the top, are immersed 
 in a cistern through wliich a continuous circulation is kept 
 up by a current of water Avliicli (lows over the corapressino' 
 pistons before entering the cistern. A glance at the sec- 
 tional di-awing on Plate 65 will give a clear idea of the 
 nature of the device and the mode of operation, which may 
 be thus briefly described : A small pipe communicating with 
 a reservoir or other supply of water is applied behind the 
 machine, proWded with a branch for each compressing 
 cylinder. These branch-pipes are bent downwards vertically 
 in such a manner that a stream of water flowing throufli 
 each will fall on the top of the comj)ressing-piston, near 
 its circumference. The conipressing-cylinders, as already 
 stated, are suspended within a water-cistern, and supported 
 by their upper flanges, which rest on the toji of the cistern. 
 Referring to the perspective view of the machine, shown 
 on Plate 64, it will be seen that the water-cistern forms a 
 pedestal supporting the side frames on which the pillow- 
 blocks of the crank-journal rest. It will also be seen that 
 the side frames form slides which guide the cross-head of 
 the piston-rods. A band-wheel, provided with a very heavy 
 rim, to be driven by steam or other motive power, 
 is attached to the crank-shaft between the pillow-l)]ock3 
 formed at the top of the side frames. It scarcely needs ex- 
 planation that the object of making the rim of the band- 
 wheel veiy heavy is tliat of ecpializiug the irregular resistance 
 
556 TRANSMISSION OF POWER. chap. xliv. 
 
 offered by the compressing-pistons. The inlet-valves which 
 supply the atmospheric air to be compressed are inserted in 
 the pistons, while the outlet-valves are placed at the bottom 
 of the cylinder, the valve-chambers of the latter communi- 
 cating directly with an air-conductor which leads to an ordi- 
 nary air-reservoir. Referring again to the sectional repre- 
 sentation of the machine, it will be seen that the sides of 
 the compressing-cylinder are perforated near the top, the 
 position of these perforations being such that when the pis- 
 ton reaches the full up-stroke its upper face will not quite 
 reach the under side of the perforations. It will be readily 
 understood that b)^ this arrangement a certain body of water 
 will always remain on the top of the piston, while at the 
 same time the perforations effectually prevent an overflow 
 within the cylinder. The connecting-rod is very short com- 
 pared with the length of throw of the crank ; hence the 
 piston will remain for a considerable interval of time near 
 the top of the cylinder, during which time the necessary 
 discharge of the water lodged on the top of the piston takes 
 place. To prevent undue accumulation of water in the cis- 
 tern, an overflow-pipe is introduced at the side, as shown in 
 the sectional illustration. It should be particularly noticed 
 that the air, \vhile undergoing compression in the cylinder, 
 is completely surrounded by metallic surfaces cooled by the 
 circulating water. But this is not all. During the recipro- 
 cating action of the piston, the body of water lodged on its 
 top washes the inside of the cylinder both during the upward 
 and downward movement. Now, the speed of the piston 
 
CHAP. XLIV. TRANSMISSIOX OF POWER. 557 
 
 is fully one luimlred and fifty feet per minute; hence an 
 internal refrigeration is established far more efficient than 
 the external circulation. The metal composing the cylinder, 
 it will thus be seen, is actually cooled on both sides, a very 
 remarkable and almost paradoxical achievement. Again, it 
 will be perceived that the circulating cold water continually 
 washes the top of the piston before entering the cistern. 
 Accordingly, the entire quantity of water required for cooling 
 during the compression passes over the piston at the initial 
 low temperature, thereby subjecting the part of the machine 
 that most needs cooling to the greatest amount of refrigera- 
 tion. As regards lubrication, it is self-evident that no con- 
 ceivable plan can be more efficient than that of actually 
 washing the inside of the cylinder with the lubricating 
 medium, both during the up and down movement of the 
 piston. 
 
 Regarding the utility of cooling the compressed air, it 
 needs no demonstration to show that refrigeration after the 
 air has left the compressing-cylinder, recommended by some 
 engineer, is not only useless, but tends to reduce the effi- 
 ciency of the compressed air as a motive agent. Obviously, 
 if the air during its transmission from the compressor to 
 the motor intended to be actuated loses in temperature, it 
 also loses in bulk. On the other hand, refrigeration witltin 
 (lie cylinder during the down-stroke is useful, as it tends to 
 check the swelling of the volume of air under the piston 
 caused by the heat generated by compression, consequently 
 diminishing the necessary motive power. 
 
CHAPTER XLY. 
 
 SUN POWER— THE SOLAR ENGINE. 
 (SEE PLATES GO AND 67.) 
 
 The illustration on Plate 66 derives its cliief interest 
 from the fact that it represents the first motor actuated by 
 the direct agency of the sun's radiant heat. It was con- 
 structed at New York, IS 70, and intended as a present to 
 the French Academy of Sciences. Apart from being a motor, 
 this engine was designed to operate as a meter for register- 
 ing the volume of steam generated by the concentrated heat 
 of a pencil of solar rays of a given section. Regarded as 
 a steam-meter, it proved important, as it verified the results 
 of previous experiments and previous calculations, based on 
 the number of thermal units developed by the evaporation 
 of a certain weight of ^vater in a given time. Engineers will 
 not fail to notice the unusual proportions of the working 
 parts, nor will they fail to appreciate the object in view, 
 that of reducing the friction to a minimum — an iudispens- 
 
 658 
 
CHAP. XLV. SUX POWER— TUE SOLAR E.XGIXE. 559 
 
 able coiulitiou in a meter. Tlie entire nicobiinisni being 
 shown with perfect distinctness in the pei-spective view of 
 the engine on the plate referred to, it is only necessuiy to 
 mention that the square pedestal which suppoi-ts the steam- 
 cylinder (4J ins. in diameter), the beam-centre, and the 
 crank-shaft, conceals a surface-condenser. 
 
 Under a clear sun the engine performed its functions 
 with pei-fect uniformity, at a velocity of 240 revolutions per 
 minute. It consumed, at the stated rate, only part of the 
 steam furnished by a solar steam-generator, temporarily em- 
 ployed, belonging to an engine of greater dimensions then 
 in the course of construction. With reference to ascertaining 
 correctly the amount of mechanical po\ver developed by the 
 concentrated radiant heat applied to thi.s engine, experts 
 need scarcely be reminded that, by dispensing with a vacuum, 
 the atmospheric resistance and back pressure exerted against 
 the piston furnish elements for measuring, with critical 
 nicety, the dynamic foi-ce transmitted by pencils of solar 
 rays of definite sections. 
 
 Drawings and descriptions of the mechanism by which 
 the sun's radiant heat has been concentrated in my experi- 
 mental engines will not be presented in this work, nor wnll 
 the form of the steam-generator which receives the concen- 
 trated heat be delineated or described. Experienced pro- 
 fessional men will appreciate the motive — viz., that of pre- 
 venting enterprising persons from procuring patents for 
 mod if cation. 9. With reference to the course thus adopted, 
 it will be pioper to mention that I have in several instances, 
 
560 SUN PO WEB— THE SOLAR ENGINE. chap. xlv. 
 
 notably in the case of tlie screw-propeller and tlie caloric 
 engine, been prevented from perfecting my invention in con- 
 seqnence of conflicting privileges having in the meantime 
 been granted to otliers. 
 
 Regarding the solar engine, it may be well to state that 
 I shall not apply for any patent rights, excepting for the 
 purpose of protecting the community, and that it is my 
 intention to devote sufficient time and means to ensure its 
 completion. Hence my anxiety to guard against legal ob- 
 structions being interposed before perfection of detail shall 
 have been measurably attained. In the meantime, let us 
 hope that no exclusive privilege will be granted tending to 
 throw obstacles in the way of an unrestricted manufacture 
 and introduction of the solar engine in countries where a 
 continuously clear sky warrants its adoption, especially in 
 Upper Egypt and on the coast of Peru. 
 
 The experiments instituted show that the mechanism 
 ■\vhich I have adopted for concentrating the sun's radiant 
 heat abstracts, on an average, during nine hours a day, for 
 all latitudes between the equator and 45 deg., fully 3.5 units 
 of heat per minute for each square foot of area presented 
 perpendicularly to the sun's rays. A unit of heat being 
 equivalent to 772 foot-pounds, it will be perceived that, 
 theoretically, a dynamic energy of 2,702 foot-pounds is 
 transmitted by the radiant heat, per minute, for each square 
 foot; hence, 270,200 foot-pounds for an area of 10 feet 
 square. If we divide this sum by the adopted standard, 
 33,000, we ascertain that 100 square feet of surface ex- 
 
CHAP. XLV. sry POWEli~TUE HOLAU'ESGINE. otil 
 
 lH)sed to the solar rays develop coiitiuuously 8.2 horse- 
 power during nine houi-s a day, within the limits of lati- 
 tude before iiientioued. But eugiueei-s are well aware that 
 the whole dynamic energy of heat cannot be utilized in 
 piactioe l)y any engine or mechanical combination what- 
 ever, nor at all approached ; hence I have assumed, in order 
 not to oven-ate the capability of the new system, that a 
 solar engine of one horse-power demands the concentration of 
 solar heat from an area of 10 feet square. On this basis, 
 I will show presently that those regions of the earth which 
 suifer from an excess of solar heat will ultimately derive 
 benefits resulting from an unlimited command of motive 
 power which will, to a great extent, compensate for disad- 
 vantages hitherto supposed not to be counterbalanced Ijy 
 any good. But before estimating the magnitude of me- 
 chanical poAver wliich we may produce by availing our- 
 selves of the fuel contained in that great storehouse from 
 whence it may be obtained fi-ee of cost and transportation, 
 let us consider the leading feature of the device resorted 
 to, especially that by which I have succeeded in augment- 
 ing the comparatively low temperature developed by direct 
 solar radiation sufliciently for the production of useful work. 
 The solar engine, when steam is employed as the medium 
 for transmitting the radiant energy, is comjiosed of tlii-ee 
 distinct parts — the engine, the steam-generatoi-, and the me- 
 chanism by means of which the inadequate energy of the 
 sun's rays adverted to is increased to such a degree that 
 the resulting temperature will exceed that corresponding with 
 
563 SUN FOWEll-THE SOLAS ENGINE. chap. xlv. 
 
 the steam-pressure necessary in an efficient engine. The 
 motor itself, wLeu tlie acting medium under consideration is 
 employed, resembles in all essential points a modern steam- 
 engine, utilizing to tlie fullest extent tlie mechanical energy 
 of the steam admitted to the working-c}linder. But when 
 atmospheric air is employed as the medium for transmitting 
 the solar energy to the motor, an entirely different conihiua- 
 tion of mechanism is called for, as will l^e seen hereafter. 
 Regarding the steam-generator, it will he superfluous to point 
 out the advantages resulting from its not being exposed to 
 the action of fire or soot ; hence that it can only suffer from 
 the slow action of ordinary oxidation. As the motor itself 
 resembles a steam-engine, we have of course merely to con- 
 sider the nature of the mechanism by means of which the 
 solar heat is concentrated and the temperature raised above 
 that of the water in the steam-generator. Regarding this 
 mechanism — viz., the concentration apparcUus — it has been 
 asked, Is it costly ? Is it heavy and bulky, so as to render 
 transportation difficult? And, finally, the question has fre- 
 c^uently been put, Is it liable to derangement and expensive 
 to keep in order? The cost is moderate. The weight is 
 small ; indeed, lightness is the most notable peculiarity of 
 the concentration apparatus. As to bulk, it may be observed 
 that this apjjaratus is composed of small parts readily put 
 together. With reference to durability, the fact need only 
 be pointed out that certain metals, however thin, if kept 
 dry, may be exposed to the sun's rays during an indefinite 
 length of time without appreciable deterioration ; hence, 
 
OIIAI-. XLV. SUX rOWER—TllE SOLAE EXGTXE. 503 
 
 unlike the funi;u"L's of steam-boilers, which soon Ijecouie 
 uuserviceable, the couceutratioii apparatus, as it consists of 
 thiu metallic plates, composed of durable materials, cannot 
 be damaged by the mere action of the sun's rays. Another 
 question has been asked, Whether the solar engine will 
 answer as well on a large as it does on a small scale i The 
 following reply disposes of this pregnant query : It is not 
 necessary, nor intended, to enlarge considerably the size of 
 the apparatus by means of which the comparatively feeble 
 intensity of the sun's rays has been successfully concentrated, 
 and the temperature sufficiently elevated to generate steam 
 for actuating the solar engine. The maximum size adopted 
 has been adequate to utilize the radiant heat of a pencil of 
 rays (^sunbeam) of 35 square feet section. The employment 
 of an increased number of such structures will, therefore, in 
 most cases be resorted to when greater power is wanted, as 
 we increase the lumiber of hands when we desire to perform 
 an additional amount of work. The motor itself — viz., the 
 steam-cylinder and the working parts — will obviously be pro- 
 portioned, as at present, in accordance with the pressure of 
 steam employed and the work to be done. 
 
 It should be clearly understood that I do not recommend 
 the erection of solar engines in places where there is not 
 steady sunshine, until proper means shall have been devised 
 for storing up the radiant energy in such a manner that 
 regular power may be obtained from irregular solar radia- 
 tion. E.xperienced engineers need not be toltl that formidal^le 
 difficulties present themselves in stoi-ing up mechanical energy 
 
564 SUN POWEE-THE SOLAS ENGINE. CHAP. XLV. 
 
 of auy kind ; yet when coal can no longer be obtained, 
 necessity, ingenuity, and increased experience will find means 
 of overcoming obstacles wliicli now appear insurmountable. 
 
 Before considering further the nature and capabilities of 
 my solar engine, it will be proper to notice the result of 
 the labors of Professor Mouchot of Tours, formerly of the 
 Lycee of Alengon, who claims to have anticipated me in 
 employing solar heat for the production of motive power. 
 Mouchot bases his claim on some experiments, made in 
 1866, intended to show that by the accumulation of heat 
 which takes place when a blackened surface is surrounded 
 by glass bells, steam may be generated for actuating 
 machinery. Sir John Herschel, it is well known, ela- 
 borated the old idea of concentrating solar radiation, and 
 conducted a series of experiments at Cape Town, in 1838, 
 showing that not only was it possible to produce boiling 
 heat by accumulating solar heat as described, but he suc- 
 ceeded in elevating the temperature sufficiently foi' roasting 
 meat. Some time previous to 1870, Mouchot made a small 
 model engine, a mere toy, actuated by steam generated on 
 the plan of accumulation by glass bells ; but finding the 
 heat insufficient, he added a polished metallic reflector. 
 The increase of temperature resulting from this expedient 
 rendered his steam-generator more effective, and it was 
 found that under favorable circumstances sufficient steam 
 could be produced to actuate his small model. The Con- 
 seil-Gen^ral of Indre-et-Loire having subsequently provided 
 Professor Mouchot with necessary means, he put u^t a steam- 
 
SUN POWEK-THE SOLAR ENGLSE. 
 
 505 
 
 geuerator at Tours in 1S72, wliich he deems a pei-fect inaebiiie, 
 its action beiug based on the results of his previous experi- 
 ments. The accompanying diagram, Fig. 1, represents a ver- 
 tical section of the said steam-generator, thus described by 
 ]\r. L. Simonin in lievue des Deux Mondes : 
 
 "The traveller who visits the libi-ary of Tours sees in 
 the court-yard in front a strange-looking a})paratus. lma"-in«' 
 
 an immense truncated cone, a mammoth lamp-shade, with 
 its concavity directed skyward. This apparatus is of copper, 
 coated on the inside \\\W\ veiy thin silver-leaf. On the 
 small ba.se of the truncated cone rests a copper cylinder, 
 blackened on the outside, its vertical axis being identical 
 with that of the cone. This cylindei', surrounded as it were 
 by a great collar, terminates above in a hemispherical cap, 
 so that it looks like an enormous thimble, and is covered 
 with a bell-glass of the same shape. 
 
 "This curious apparatus is nothing else but a solar receiver, 
 
5G6 SUN rOWEB—TUE SOLAll ENGINE. chap. xlv. 
 
 or, ill <,)tliei' words, a boiler, in wliicli water is made to boil 
 by the beat-rays of tlie suo. This steam-generator is designed 
 to raise water to the boiling point and beyond, by means 
 of the solar rays, ^vhich are thi'o^vn upon the cylinder by 
 the silvered inner surface of the conical reflector. The boiler 
 receives water up to two-thirds of its capacity through a 
 feed-pipe. A glass tube and a steam-gauge communicating 
 with the inside of the generator, and attached to the outside 
 of the reflector, indicate both the level of the water and 
 the pressure of the steam. Finally, there is a safety-valve 
 to let off the steam when the pressure is greater than desired. 
 Thus the engine offers all desirable safety, and may be pro- 
 vided with all the accessories of a steam-l)oiler. 
 
 " The reflector, which is the main portion of the generator, 
 has a diameter of 2.60 metres at its large, and one metre 
 at its small, base, and is eighty centimetres in height, giving 
 four square metres of reflecting surface, or of insolation. The 
 interior walls are lined with burnished silver, because that 
 metal is the best I'eflector of the heat-rays ; still, brass with 
 a light coating of silver would also serve the purpose. The 
 inclination of the walls of the apparatus to its axis measures 
 45 deg. Even the ancients were aware that this is the best 
 form for this kind of metallic mirrors with linear focus, 
 inasmuch as the incident rays parallel to the axis are reflected 
 perpendicularly to the same, and thus give a focus of maxi- 
 mum intensity. 
 
 "The boiler is of copper, which of all the common metals 
 is the best conductor of heat ; it is blackened on the out- 
 
CUA1>. XLV. iiUy I'OWEK-TUE SOLAR EXGIXE. 5C7 
 
 side, liecaiise l)lack possesses tlie property of absorbing all 
 the heat-rays, just as white reflects them; and it is enclosed 
 in a glass envelope, ghvss being ilie most diathermanous of 
 all bodies— that is to say, the most permeable by the rays of 
 liiniinous heat. Glass further possesses the property of resist- 
 ing the exit of these same rays after they have been trans- 
 formed into dark rays on the blackened surface of the 
 boiler. None of the.se applications of physical laws present 
 any novelty ; people induced them to practice instinctively, 
 as it were, before men of science could assign the reasons. 
 Here the arts of cookery and of gardening, and the pro- 
 cesses for warming our rooms, did uot wait for the experi- 
 ments of the physicist. Saussure himself started from these 
 data in his researches ; but the inventor needed the discoveries 
 of modern physics in order to give to these applications a 
 ligorous formula. 
 
 "The boiler proper of the Tours solar engine consists of 
 two concentric bells of copper, the larger one, which alone 
 is visible, having the same height as the mirror — i.e., eighty 
 centimetres— and the smaller or inner one fifty centimetres; 
 their re,spective diameters are twenty-eight and twenty-two 
 centimetres. The thickness of the metal is only three mil- 
 limetres. The feed-water lies between the two envelopes, 
 forming an annular envelope three centimetres in thickness. 
 Thus the volume -of li.piid is twenty litres, and the steam- 
 chamber hits a capacity of ten litres. The inner envelope 
 is empty. Into it pass the steam-pipe and the feed-pipe 
 of the boiler. To the sleamjiipe are attache<l the gauge and 
 
568 SUN POWER— THE SOLAB ENGINE. chap. xlv. 
 
 the safetj"- valve. The bell-glass covering the boiler is eighty- 
 five centimetres higli, forty centimetres in diameter, and five 
 millimetres in thickness. There is everywhere a space of 
 five centimetres between its walls and those of the boiler, 
 and this space is filled with a layer of very hot air. 
 
 "The earth, owing to its diurnal and annual revolution, 
 does not occupy the same position with regard to the sun 
 at all hours of the day, or in all seasons of the year. This 
 being the case, the generator is so contrived as to revolve 
 15 deg., or one twenty-fourth of its circumference, hourly 
 around an axis parallel to the earth's axis — i.e., so as to 
 follow the ajjparent diurnal motion of the sun, and to incline 
 gradually on its axis in proportion to the solar declination. 
 Hence the intensity of the utilized heat is always nearly 
 the same, whatever the hour of the day or the season of 
 the year, inasmuch as the apparatus is always so ari-anged 
 as to reflect with the least possible loss all the rays emitted 
 by the sun. This double motion of the generator is effected 
 by a very simple contrivance." 
 
 The foregoing description of the Solar Steam-Generator 
 of Mouchot is so lucid that it requires no explanation. Mr. 
 Simonin, however, erroneously supposes that the power de- 
 veloped by the appai-atus is neaidy the same at all hours of 
 the day, the fact being that the energy developed by the 
 concentrated solar heat varies with the depth of atmosphere 
 penetrated by the rays. The latter evidently depends on 
 the sun's zenith distance ; hence at Paris, where the maxi- 
 mum solar intensity during the summer solstice is 65°.0 
 
cuAP. xLv. su.\ rowEn—TUJi: solaj: EyaixE. 669 
 
 Fall, at iiouu (see diagram on Plate 9), it .scarcely reaches 
 52°.0 F. at five o'clock in the afternoon, owing to the in- 
 creased zenith distance, and consequent increase of the depth 
 of atmosphere to be penetrated by the sun's rays. Obviously, 
 the efficiency of the solar generator will be diminished in 
 the same ratio as the stated intensities. Mr. Simonin states 
 that on some occjisions, when the sun has been exceptionally 
 clear, the solar generator at Tours has evaporated five litres 
 of water per hour, which he assumes equivalent to half a 
 horse-powei'. The reflector producing this result — a trun- 
 cated cone — being 2.6 metres (8 feet 6 inches) in diameter, 
 it will be found that in order to double the reflective area 
 necessary to generate steam for an engine of one horse-power, 
 a truncated cone of 3.(? metres (11 feet 9 inches) aperture 
 will be requii-ed. Practical engineei-s are aware that an 
 inverted conical body whose base is nearly 12 feet in dia- 
 meter, swinging round an inclined axle at least 60 deg. on 
 each side of the vertical line, presents a structure so for- 
 midal)le, even if counterpoised, that it would not be pradent 
 to increase its size. Accordingly, one hundred of Mouchot's 
 solar generatoi*s w-ould be needed to furnish steam for an 
 engine of 100 horse — a very moderate power, if employed 
 for manufacturing or other industrial purposes. Referring to 
 the diagram on page ~^^^D, Fig. 2, representing a bird's-eye 
 view of the aperture of the conical reflector at Tours in 
 three different positions — viz., a in the morning, h at mid- 
 day, and c during the afternoon — it will be seen that each 
 instrument, owing to the necessary change of position, de- 
 
570 SUN PqWER-THE SOLAB ENGINE. CHAP. XLV. 
 
 mjinds a front sptice of nearly twenty feet. If placed side 
 by side, the conical solar generators required for an engine 
 of 100 horse-power would therefore occupy a front of 2,000 
 feet from east to west. If arranged in four lines, with suf- 
 ficient s^iace north and south to prevent interference, a dis- 
 tance of 500 feet by 200 feet would be required. 
 
 Now, let us consider that the scheme calls for 100 sepa- 
 rate boilers, to be continually fed with water, the height of 
 which can only be known by the indication of outside gauges, 
 while the steam from the scattered boilers must be conveyed 
 by a series of flexible tubes to the motive engine. The hun- 
 dred glass bells can, no doubt, be dusted and kept clean with 
 moderate exertion ; but the hundred silver-plated reflectors, 
 which Mr. Simouin says must be exposed to the vicissitudes 
 of the atmosphere, cannot ,be kept bright without herculean 
 labor, since silver tarnishes in a few hours. In view of the 
 foregoing statement, which embraces only the chief difliculties 
 attending Mouchot's system, the most sanguine might well de- 
 spair of rendering sun-power available for practical pui-poses. 
 
 The Professor of the Lycee of Alenyon, in claiming to 
 have anticipated me, has done so ignorant of the fact that 
 sun-power has been the study of my whole professional life 
 — a life the early part of Avhich was chiefly devoted to the 
 production of a cheaper motive power than steam. The in- 
 dustrious scientist, if he had been correctly informed on the 
 subject, would no doubt have perceived the advantages re- 
 sulting from such antecedents, with reference to a successful 
 practical solution of the problem of utilizing solar heat. 
 
CHAP. xLv. srx rowEn-rnE solar exoine. 571 
 
 On grounds already fully explained, minute plans of my 
 new system of rendering sun-power available for mecbanioal 
 purposes will not be presented in this work. The occasion, 
 however, demands that I should present an outline of the cm- 
 centration apparatus befoi'e referred to. It consists of a series 
 of polished parabolic troughs, in combiinitioii with a system 
 of metallic tubes charged with water under pre.ssure, e.vposed 
 to the influence of converging solar rays, the augmented mole- 
 cular action produced by the concentration being transferred 
 to a central receiver, from which the accunuilated energy is 
 communicated to a single motor. Thus the mechanical power 
 developed by concentrated solar heat is imparted to the solar 
 steam-engine without the intervention of a multitude of boilere, 
 glass bells, gauges, feeders, etc. Moreover, the concentration 
 apparatus, unlike the instrument of Mouchot, requires no 
 parallactic motion, nor does its management call for any 
 knowledge of the sun's declination from day to day. Its 
 jwsition is regulated by simply turning a handle, until a 
 cei-tain index coincides with a certain bright line produced 
 by the reflection of the sun's rays. 
 
 Plate 07 represents a perspective view of a solar engine, 
 in which the concentrated energy of the sun's rays is com- 
 municated to the motor by means of heated atmospheric air, 
 instead of being communicated l)y water heated under pres- 
 .sure and expanded into steam. A glance at the illustration 
 shows that the upper end of the working-cylin<ler is heated 
 by the sun's rays reflected by a curved mirror. It will be 
 seen by carefid examination that the solar rays converge at 
 
672 SUN PO WEE-TEE SOLAR ENGINE. CHAP. XLV. 
 
 a point beyond the axis of the reflector; heuce tliat the 
 form of the latter is not parabolic, but composed of an 
 irregular curve. The object is that of spreading the con- 
 verging rays over a greater length of the cylinder than 
 possible with the divergence which would result from em- 
 ploying a reflector of true parabolic curvature. It will be 
 
 perceived on insi^ection that the upper end of the cylinder 
 will be subjected to a concentration of heat many times 
 greater than the concentration at the lower end. Referring 
 to the accompanying diagram representing a vertical section, 
 of the machine, it will be seen that the working-cylinder, 
 open at the lower end, contains two pistons, a working- 
 piston a and an exchange-piston b. The working-piston is 
 
CHAP. XLV. SUX rOWEU—TUE SOLAR ENGINE. 573 
 
 CDiuiected with tlio crank-shaft d by the beam c and the 
 connecting-rod </. The exchange-piston Z> is connected witli 
 the crank-shaft by the bell-ci-aiik // and connecting-rod //. 
 An annular space is formed round tlie excliange-piston, 
 admitting of a free passage of the air from end to end of 
 the cylinder during the motion of this piston. It will l;e 
 readily understood that during the downward motion of the 
 exchange-piston the cold air from the lower end of the 
 cylinder will be transferred to the upper end, heated by 
 the concentrated solar rays ; lience internal pressure will be 
 produced tending to force the working-piston down. By a 
 careful examination of the combination of the several work- 
 ing parts, it will be easily comprehended how the working- 
 piston is actuated by the confined air, heated and cooled 
 alternately by the peculiar motion of the exchange-piston. 
 It will be evident that the large surface presented by the 
 outside of the exchange-piston, and inside of the cylinder, 
 •will cause a rapid change of temperature of the air while 
 circulating from end to end of the latter. The upper end 
 of the cylinder being heated by the concentrated solar 
 rays, the cold air from the lower end \n\\, during its trans- 
 fer to the upper end caused by the downward motion of 
 the exchange-piston, become heated and expanded; while 
 during the upward motion of the .said piston the air, in 
 being transferred to the lower end of the cylinder, becomes 
 cooled and contracted. It will be found on due considera- 
 tion that the exchange-piston thus performs the office of a 
 regetierator. The engine, therefore, is capable of ojieiating 
 
574 SUN rOWEB-THE SOLAR ENGINE. chap. xlv. 
 
 for a considerable time by exposing tlie upper end of the 
 cylinder to the reflected solar heat during a few minutes at 
 starting. By continuous exposure to the concentrated solar 
 rays, the engine performs fully 400 turns per minute. It 
 should be observed that concentrated solar radiation supplies 
 heat with such extraordinary rapidity that the apparently 
 insufficient amount of heating surface presented by the 
 cylinder has proved adequate, notwithstanding the great 
 speed of the engine. It only remains to be stated that the 
 body fii m represents a radiator carrying off the heat which 
 is not taken up by the circulating air during the motion of 
 the exchange-piston. Of course, the amount of heat carried 
 off by the radiator furnishes a nearly correct measure of the 
 solar energy not converted into mechanical work. Engineers 
 need not be reminded that the form of the solar engine thus 
 described is applicable only for purposes requiring moderate 
 power. In the largest class of solar engines actuated by 
 atmospheric air, in which the radiator is incapable of ab- 
 stracting the sujierfluous heat, I employ valves, and take in 
 fresh air at each stroke of the machine, precisely as in the 
 caloric engine delineated on Plate 46. 
 
 Having thus cursorily examined the construction of the 
 solar engine actuated by the intervention of atmospheric 
 air, and briefly adverted to the steam solar engine and the 
 mode adopted in concentrating the molecular motion im- 
 parted by solar radiation, and also pointed out the nature 
 of the expedient resorted to in transferring the said con- 
 centrated molecular motion to mechanical motors, let us now 
 
CHAP. XLV. SUN roWEli-TUE SOLAIi AWGIXE. 575 
 
 consider the stupfiulous amount of the energy at our com- 
 mand. 
 
 It has already been stated that the result of repeate.l 
 e.\i)eriments with the concentration a^iparatus shows that it 
 abstracts on an average, during nine hours a da}-, for all 
 latitudes between the equator and 45 deg., fully :].'» units 
 of heat per minute for each sipiare foot of area presented 
 perpendicularly to the sun's rays. Theoretically, this indi- 
 cates the development of an energy etpial to 8.2 horse- 
 power for an area of 100 square feet. On grounds before 
 explained, our calculations of the capabilities of sun power 
 to actuate machinery will, however, be based on one horse- 
 power developed for loo s(puire feet e.\-i)osed to solar radia- 
 tion. The isolated districts of the earth's surface suffering 
 from an excess of solar heat being very numerous, our space 
 only admits of a glance at the sunbui-nt continents. 
 
 There is a rainless region extending from the northwest 
 coast of Afi-ica to Mongolia, 9,000 miles in length and nearly 
 1,000 miles wide. Besides the North Afiican deserts, this re- 
 gion includes the southern coast of the Mediten-anean east of 
 the Gulf of Cabes, Upper Egypt, the eastern and i)art of the 
 western coast of the Eed Sea, part of Sp-ia, the eastern part 
 of the countries watered l)y the Euphrates and Tigris, Eastern 
 Arabia, the greater pait of Persia, the extreme western part of 
 China, Tibet, and, lastly, Mongolia. In the western hemisphei-e. 
 Lower California, the table-land of Mexico and Guatemala, 
 and the west coast of South America, for a distance of more 
 than 2,000 miles, suffer from continuous intense radiant heat. 
 
5rG Sm' rOWFAt—THE SOLAR ENGINE. chap. XLV. 
 
 Computations of tlie solar energy wasted on the vast 
 areas tlius specified would present an inconceiva1)ly great 
 amount of dynamic force. Let us, therefore, merely esti- 
 mate the mechanical power that would result from utilizing 
 the solar heat on a strip of land a single mile in width, along 
 the rainless western coast of America ; the southern coast of 
 the Mediterranean before alluded to ; both sides of the allu- 
 vial plain of the Nile in Upper Egypt; both sides of the 
 Euphrates and Tigris for a distance of 400 miles above the 
 Persian Gulf ; and, finally, a strip one mile wide along the 
 rainless portions of the shores of the Red Sea, before pointed 
 out. The aggregate length of these strips of land, selected 
 on account of being accessible by water communication, far 
 exceeds 8,000 miles. Adopting the stated length and a 
 ■width of one mile as a basis for computation, it will be seen 
 that this very narrow belt covers 223,000 millions of square 
 feet. Dividing the latter amount by the area of 100 square 
 feet necessary to produce one horse-power, we learn that 
 22,300,000 solar engines, each of 100 horse-power, could be 
 kept in constant operation, nine hours a day, by utilizing 
 only that heat which is now wasted on the assumed small 
 fraction of land extendiuof alonsf some of the water-fronts 
 
 o o 
 
 of the sunburnt regions of the earth. Due consideration 
 cannot fail to convince i:s that the rapid exhaustion of the 
 European coal-fields will soon cause great changes with 
 reference to international relations, in favor of those coun- 
 tries which are in possession of continuous sun-power. 
 Upper Egypt, for instance, will, in the course of a few cen- 
 
CHAP. XLV. SUN POWEU-THE SOLAU ENGINE. 577 
 
 turies, derive signal advantage and attain a liigli i.ditiial 
 position on account of her iierpetual suiisliinc and the eon- 
 sequent command of unlimited motive force. The time will 
 come when Europe nuist stop her mills for want of coal. 
 Upper Egypt, then, with her never-ceasing sun-power, will 
 invite the European manufacturer to remove his machiueiy 
 and erect his mills on the firm ground along the sides of 
 the alluvial plain of the Nile, where an amount of motive 
 power may be ol)tained many times greatei' than that now 
 employed by all the manufactories of Europe. 
 
V 
 
If 
 
 
 n 
 
 
 a 
 
 
 01 
 
 
 •*) 
 
 
 o 
 
 
 H 
 
 
 S 
 
 
 « 
 
 Q 
 
 en 
 
 01 
 
 
 H 
 
 •?! 
 
 ^ 
 
 o 
 
 o 
 
 W 
 
 W 
 
 > 
 
 rj 
 
 o 
 
 
 
 p-i 
 
 > 
 
 
 H 
 
 
 W 
 
 ^ 
 
 > 
 
 K! 
 
 H 
 
 o 
 
 
 
 
 
 
 
 
 § 
 
 
 
 M 
 
 
 C 
 
 
 cs 
 
 
 K 
 
 
 «-( 
 
 
 O 
 
 
 n 
 
 
 !< 
 
 » 
 
 Flate 1. Hee (Juaf. I. 
 
Instkument fok measuring the Ixtexsity of Radiation from 
 
 Enclosed Concave IlAniATous. 
 
 Designed hy .Toiix Ekicsson. Manufactured at New York, 1873. 
 
 Plate 2. See Chap. I. 
 
DiAouAMS snoTviNo THE Propaoatiox of Radiant 
 Heat tiiikiuoii Space. 
 
 Plate 3. Se£ Chap. I. 
 
Instrument suowing the Rate ob^ Cooling of a Ueated Body 
 
 wiTiiix AN Exhausted Cold Enclosuue. 
 
 Desiuxed hy John Ericsson. Manufactured at ^i'ew York, i8?a. 
 
 ym. I 
 
 Plats 4. See Chap. 11. 
 
IXSTKI'MKNT SHOWIXU TIIK HaTE <iK IIlOATI.NCi OF A CoLl) linUV 
 WmilX AN KXIIAISTKI) IIeATKU EXCLOSUlUi. 
 
 Dksicnkd 1!Y John Euicssox. Manifactiuku at Ni;\v Youk, ist-l 
 
 FLATK 5. i)££ CUAi: II. 
 
IXSTKFMENT SnOWINO TITE RaTE OF COOLINO OF AN IXCAXDESCEXT 
 
 Sphere within an Exhausted Cold Encloscue. 
 "Desioned i?y Johx Eimosson. Manufactured at New Yoi:k, 1874. 
 
 ^}K^ 
 
 Plate G. See Chap. II. 
 
I 3 
 
 5. > 
 
 en 
 
 o w 
 
 ^ i 
 
 53 S 
 
 2 B 
 
 Plate 7. Skf. Ciiav. II. 
 
ACTINOMETEU, FOR MEASURING THE INTENSITY OF SoLAU RaDIATIOX. 
 
 Designed by John Ericsson. MANriACTrnnn at Xkw Yoi:k, i870. 
 
 Pr.ATk 8. See Vuap. HI. 
 
Plate 9. See Cuav. III. 
 
Solar Calokimetkr, i-ou MEAsrniNc; tiik Meciia.vical Enekoy ok 
 
 Solar IlAniATiox. Designed hy Joii.v Ericssox. 
 
 Manufactured at New York, i870. 
 
 Plate 10. See Chap. V. 
 
PoKTAiii.K SoLAK Cai.()i:imktki:. kok MKAsui:i.\(i Tin; Mechanical Enki!oy 
 
 <>I' SoLAi: ItADIAlloV. l)i:si(;XKI> ItY .lollX ElMCSSON. 
 MaM1A( TIIMM) AT XkW Y(I1;K, 1N74. 
 
 PI.ATE 11. See Chap. V. 
 
l)iA(;i:.vMs siiowiNd tiik JIaihatiox fkom Dikkkkknt Parts 
 (IK TKE Soi.Aij Disc. 
 
 Plate 12. See Chap. VI. 
 
PAl^ALLACTIc Mechanism fou mkasukixg tiik Lvtensity ok RadiXtio.v 
 
 FROAI DiFFERKNT PaKTS OF THE SoLAR Dl-Si'. 
 
 Designed uv Joux Euicssux. Coxstkuctkd at New Yokk, 18T5. 
 
 Plats 13. See Cbap. VI. 
 
DiAdu.vM siiowiNd Tin: Attuactiox witiiix tiii-; Solai: Mass 
 AT I)ii'i-ki:i:n r Distaxcks viu>m its Ckntuk. 
 
 Plate U- See Chap. VII. 
 
iNSTTltTMENT FOU MKASURIXC, TIIK RaDIAXT PoWER OF TIIK SoLAR 
 
 Atmosphere. Designed by Jonx Ericssox. 
 
 MANTFArTCKED AT NkW YoRK, 1873. 
 
 Plate 15. Skk Cn.ir. VIII. 
 
DiAiiUAMs siiowixd Tin; ]{ai>ia.nt Puwkk 111- Till-; 
 SoLAU Atmospiikkk. 
 
 Plate 16. See Cuai>. VI 11. 
 
LnSTKUMENT Full MEASUKI.NG TilK AClLAJ. INTENSITY OF TllK Sux's RaY: 
 
 Designed uv John Ekicsson. Maxufactueed at New Yokk. isn. 
 
 Flate 17. i>£i: CuAP. IX. 
 
Instrhment for showin'g the Fkebleness of Solai: Radiation-. 
 Desic.xkp itY Joiix Ericssox. Coxstructed at Xkw York, isn. 
 
 Plate la. .Sav; ( iiai: 1.\. 
 
SOLiVIi PVKOMETKU, FOU ASCERTAIXIXO THK TKMPICnATrKI': OF TIIF, 
 SOLAU SCKFACE. DESIGNED BY JoiIX ElilCSSON. 
 
 Constructed at New York, isto. 
 
 Pr.ATK 10. See ('iiaf. X. 
 
Appaijatus fok MKAsiuiixG THE IIauiant Intensity ok Flames. 
 Designed by John Eiucsson. Constkuoted at New York, is71. 
 
 FIG. 9 
 
 Platb 20. Hee ViiAP. X 
 
INSTKUMKNT FOU MKA.SUKIXG THE UaDIATIOX l-llU.M IXCAXDESCEXT 
 
 Planes. Designed by John Ericsson. 
 Manufactuued at New Youk, 18?^. 
 
 Plate 21. !See C'jjaj: XL 
 
DlAfiUA.MS SlIOWIXO TIIK RaDIATIOX AT DIFFERENT INCLINATIONS 
 
 OF Incandescent Planes. 
 
 Plate 22. See Cum: XL 
 
IXSTUrAfKXT KOU MKASI'IUXC. TUr. UvniATIOX VUOM DUKKKKXT ZOXKS 
 
 OF Incandkscent Spheres. Dksioxkd hy Joiix Ekicssox. 
 Manufactured at New York, i8?3. 
 
 Pr,ATK 2.J. See C'//.tr. XI I. 
 
DiAtSUAMS SHOWINti THK RaDIATIOX KUoM DlKKKKKNT ZoNKS 
 
 OF Incandksoent Spiieues. 
 
 Plate 24. See Cuai: A'II. 
 
OaLOKIMETEK, KOK MEASL'lUNti THE ENKKliV DKVELUPED »Y R.VDIATIOX 
 
 OK FusKD luoN. Desujneu hy Joux Ekicsson. 
 
 CONSTRUOTEU AT NeW YultK, 1B?J. 
 
 Ov 
 
 /'.>/ 
 
 Plate ^o. See Vuaj: All I. 
 
AppAUATrs Koi: mkasiimm; IIakiant IIkat iiy mkans uk tiik Tiikkmo- 
 Elkctrio PiLK. Dksiuxk.i) hy John' Kuicssox. 
 
 CONSTIMTTKU AT XkW Y<iKK, lrtT4. 
 
 Pl.iti-: .'>J. ShK ( IIAI-. .\ I \ 
 
B.VlIoMr.TKIC ACTIVOMKTKI:, von MKASintINt! TIIK IXTKN'SITV OK 
 
 SoLAK 1{ai)[aiion. l)i:si(i\Ki) HY .loiix Ericsson. 
 MANtrKArrruKn at Nkw Vokk, 1874. 
 
 PrMTK 27. Skf r/fAP. ATI. 
 
!25 
 
 M 
 
 > 
 
 R 
 
 Q 
 
 r> 
 
 W 
 
 VJ 
 
 w 
 
 CO 
 
 H 
 
 o 
 
 > 
 
 i< 
 
 
 Flate ^S. ,SA'i' C'//.ii'. A' 1 7/. 
 

 Q 
 
 
 >■ 
 
 
 ^ 
 
 
 CC 
 
 1-H 
 
 3 
 
 
 s 
 
 a 
 
 ^ 
 
 «) 
 
 U 
 
 O 
 
 w 
 
 ^ 
 
 
 o 
 
 H 
 
 
 2 
 
 § 
 
 Q 
 
 ss 
 
 QD 
 
 
 cc 
 
 H 
 
 o 
 
 H 
 
 s! 
 
 CD 
 
 
 g 
 
 
 
 
 5 
 
 O 
 
 O 
 
 o 
 
 
 t? 
 
 H 
 
 '^S 
 
 M 
 
 H 
 
 H 
 
 w 
 
 
 a 
 
 >- 
 
 o 
 
 Q 
 
 H 
 
 O 
 
 H 
 
 d 
 
 o 
 
 w 
 
 
 > 
 
 H 
 
 
 fe5 
 
 O 
 
 w 
 
 >n 
 
 ^ 
 
 1 
 
 ^ 
 
 o 
 
 
 w 
 
 OT 
 
 ?: 
 
 O 
 
 
 tr" 
 
 ^ 
 
 > 
 
 
 53 
 
 !^ 
 
 'T3 
 
 
 «1 
 
 
 a 
 
 
 o 
 
 
 K 
 
 Pijrs ;?5. S£E CuAi: X VIII. 
 
Instrument for measuring the Reflective Power of Silver and 
 
 OTHER Metals. Designed by John Ericsson. 
 
 Manufactured at New York, 1874. 
 
 FIG. I 
 
 FIG. 2. 
 
 Vi. 
 
 Plate 30. See Ciiap. XIX. 
 
Rapid-In'dkwtiox Actinometer, for ME.vsrRixo THE Intensity of 
 
 Solar Radiation. Designed by John Ericsson. 
 
 Manufactured at New York, 18T3. 
 
 Plate 31. See Chap. XX. 
 
ApPAKATUS I'OK ASCKKTAINING TlIK DlATUKlUIANCV OF Fl^JIKS. 
 
 Designed by Joun Ericsson. Constructed at New York, 1872. 
 
 Plate 32. Sek Cuap. XXI. 
 
DiACiKAM KKIMIESENTING A SECTION (»I' TllK EaKTH AXD 
 CERTAIN RiVEU BaSINS. 
 
 Plate 33. :^£E Cu.u: A' A 11. 
 
Byxamic Rkcistkh I'on mkasi-kixc tiik Uki.ativk Pdwku oi' CruitKXT? 
 
 OF NVatku and Vai'oi:. Dksioned ]?y .Tohx Eutcssox. 
 
 Manukactithkh at Nkw York, 1871. 
 
 I'LATE 04. Hee Cj/aj: XXII. 
 
T)i\(;n.\:Nt snowixr, the Result of Experiments with tiik 
 Dynamic Rec.i^jtkr. 
 
 Plate J'j. Ske Ciiai: A' A J I. 
 
DiSTANTK Insi i;r.MK.\T Kui; mi;asiui.\(; DisTANcKs AT Sea. 
 DissiGNEi) i!v ilt'iiN Ki;icss().\. Manufactl'iu:u at Nkw YoUK, l»4l. 
 
 Plate jo. Hes Cuai: A'AJJI. 
 
Steam Fire-Engine. Designed by Jouji Ericsson', 1841. 
 
 Plate 87. See Chap. A'A'IV. 
 

 H 
 
 
 5? 
 
 
 
 
 
 
 
 55 
 
 
 S 
 
 
 
 
 
 •*! 
 
 
 H 
 
 
 B 
 
 
 K 
 
 
 ^ 
 
 
 ^ 
 
 
 :3 
 
 H 
 
 
 P3 
 
 2! 
 
 > 
 
 
 
 •^ 
 
 3 
 
 CD 
 
 H 
 
 <1 
 
 
 
 m 
 
 i? 
 
 
 * J 
 
 fn 
 
 -• 
 
 (S 
 
 
 7: 
 
 -1 
 
 w 
 
 
 
 
 
 
 
 
 
 
 
 i^ 
 
 i< 
 
 T* 
 
 
 
 r 
 
 hp 
 
 _^ 
 
 
 
 W 
 
 M- 
 
 W 
 
 
 y 
 
 ^ 
 
 
 "-' 
 
 
 ^ 
 
 r 
 
 M 
 
 X 
 
 ^ 
 
 c 
 
 
 
 K 
 
 Cfl 
 
 f5 
 
 •Ji 
 
 
 
 
 ^ 
 
 2! 
 
 
 
 
 ^^ 
 
 w 
 
 ij; 
 
 r^ 
 
 w 
 
 
 H 
 
 r 
 
 
 
 -1 
 
 5^ 
 
 b> 
 
 Plate 3S. See Ciiaf. XXV 
 
5< c 
 
 Plate 39, See Chap. XXV. 
 
< 
 
 c S 
 
 K O 
 CQ 
 
 H O 
 
 K 2! 
 
 T'A.iT-i; ^0. 6'£A- (7/.I/'. XXK. 
 
pLATt: J,l. iSEE CuAi: AAV I. 
 
./. ^_^ 
 
 Vlate \2. St:i: Cum: .WVIII. 
 
SlTRFACK-CoNnENSKU. ] )KSI<;NKr) AND PaTKXTKI) IIV .TolIN EuiCSSON, 1849. 
 lUILT AT NKW Vi>i:K. 
 
 Pi.ATK Jt3. See Cum-. XXIX. 
 
EXI'KKIMKNTAL CaLuKIC ExUINK. DksIUXKU ItV .lollN ElMCSSON. 
 
 BlILT AT JsKW YuUK, 18-jI. 
 
 TUAXSVKUSK SKCTIUX. 
 
 Pl.itic 44- ''•^i^-'^' tUAf. XXX. 
 
rt 
 
 M 
 
 5 Z 
 
 2 '-^ 
 
 c B 
 
 K 
 
 Plate Jto. See Cuap. AJ:'X 
 
Oai.oiik- Km;ini'. pdi: Domkstic Pi'im'osks. DiwiuNKn nv John Eimcssox. 
 Biii.T IN Amkkica and Eiiiori', mui.Mi a Skuiks of Vkaks. 
 
 LOXCinriJlNAI. SKClldN. 
 
 Pi^ATK 40. ShK Cum: .\'.\'.\7. 
 
TiiK '• MiiMToi:." T)i:si(;\i:i) isv .loiix Ei;ic»suN. Biilt at ]S'i;\v Y(ii:k, isui. 
 
 niXK PLAN. 
 
 TnAx=;vER'!K '^T.rTiny nv Tiri.L and rrnnET. 
 
 PhATF: 1,7. Ske Cum: XXA'll. 
 
s 
 
 GD 
 
 Q 
 
 o 
 
 H 
 !z) 
 
 o 
 
 w 
 PI 
 
 Plate A8. Sek Chap. XXXII. 
 
MoNiTou OK Tinc "Passaic" Class. Designkd hy John Eijicsson. 
 Ten Monitoks of this Class hiilt at Nk.w Youk and otmku Places. 
 
 SIDE ELEVATION. 
 
 TKANSVEKSE SECTION OF TUUKET AND PILOT-HOUSE. 
 
 Plate 49. See Chap. XXX 111. 
 
y 
 
 'y. 
 
 n 
 
 7: 
 
 a 
 
 y 
 
 M 
 
 
 
 
 
 
 o 
 
 
 C3 
 
 
 t/j 
 
 
 M 
 
 
 :/j 
 
 
 ►^ 
 
 
 Plate hi). See Cuav. XXXIV. 
 
td 
 
 tj "^ 
 
 w 
 o 
 
 ^ S ^^ 
 
 S S 15 
 ^ d 
 
 o 
 
 Pi.ATK ■'>!. Si:k fiiAi: XXXIW 
 
J 
 
MoxiTou •• Dr-iat(ii:." Dksiuxkd uv John Eiucsson. 
 EuiLT AX New York, 1862. 
 
 SIDE ELEVATION. 
 
 DECK PLAN. 
 
 LENGTH ON DECK, 312 FEET. BEAM, 50 FEET. DEPTH, 21 FEET 6 INCHES. 
 
 STEAM-CYLINDERS, 100 INCHES DIAMETER, 4 FEET STROKE. 
 
 PROPELLER, 31 FEET 6 INCHES DIAMETER. 
 
 I'LATt: J J. SEt: LuAP. XXXV. 
 

 S 
 
 
 O 
 
 
 S^ 
 
 
 
 
 H 
 
 
 O 
 
 
 w 
 
 
 ^ 
 
 
 
 
 U 
 
 
 M 
 
 
 Q 
 
 
 1-3 ■ 
 
 
 >► 
 
 
 H 
 
 
 O 
 
 H 
 
 
 > 
 
 ::; 
 
 '^ 
 
 
 <1 
 
 b 
 
 
 
 
 O 
 
 t?d 
 
 ii 
 
 r/3 
 
 M 
 
 M 
 
 O 
 
 o 
 
 
 H 
 
 td 
 
 O 
 
 ><! 
 
 !2! 
 
 =-( 
 
 
 O 
 
 o 
 
 
 ^ 
 
 H 
 
 g 
 
 i 
 
 M 
 
 m 
 
 !yj 
 
 O 
 
 (> 
 
 2! 
 
 !2l 
 
 
 O 
 
 td 
 
 •/. 
 
 a 
 
 ffi 
 
 t-( 
 
 
 tr; 
 
 1^ 
 
 H 
 
 
 t> 
 
 
 H3 
 
 
 tz! 
 
 
 M 
 
 
 ^ 
 
 
 K| 
 
 
 o 
 
 
 w 
 
 
 w 
 
 ili;i.i,., .lii 
 
 Plate 53. Sjce Cuai'. XXXV. 
 
2 " 
 
 Plate 5^. See Cuai: XXXV 
 
Pr.ATE '>',. Skk CiiAi: A' A' AT 
 
w ^ 
 
 3 ^ 
 
 3 / 
 
 ,-£] 
 
 Zl 
 
 m 
 
 ^ 
 
 M S 
 
 3 1-3 
 
 J'latk jo. Sfn L'UAF. A'XA'17. 
 
Cakuiages fou Hkavy Ordxaxck. Designkd by John Ericsson, i8Gi. 
 
 section showing the kkictiox-geak ai'l'lied to the gun-cakkiages 
 oe the united states uton-clau fleet. 
 
 SECTION SHOWING CAPTAIN' SCOTT S I'LACil AltlSM. 
 
 SECTION SHOWING SIi: WILLIAM AKMSTKONG S PLAGIARISM. 
 
 PLATt: o7. St:t: VuAi: XXWll. 
 
w 
 
 Plate .IS. Sek Ciiai; XXXVIJL 
 

 w 
 
 i-i 
 
 o 
 
 •=9 
 
 CO 
 
 O 
 
 ID 
 
 w 
 
 !z! 
 
 s 
 
 
 > 
 
 
 •^ 
 
 trl 
 
 ►t) 
 
 G 
 
 C-i 
 
 
 
 > 
 
 § 
 
 ^ 
 
 H 
 
 W 
 
 K 
 
 2 2 
 
 Pi.ATh: .Vi. Skk Cii.ii: .VA'A'JS. 
 
O 
 
 a 
 
 C>1 
 
 pLATi; (Jij. Hke CuAi: XL, 
 
TllK SrANISII GUNliOAT ExcilNKS. DESUiXKK HY .lollX ElUCSSON. 
 
 Built at Nkw Yoiik, 18G9. 
 
 PhATK 01. .s>.>; CiiAi: XLI. 
 
illSHIlllli 
 
 'I 
 
 iw 
 
 Mh 
 
 Nl 
 
 I 
 
 K^Wi 
 
 Platk OJ. Si:k iii.u: Xl.ll. 
 
^^^)VATiI.K Torpedo. Bestgnkd isv John Eiircssox. Bt'ilt at Nkw Yokk, 187:!. 
 
 I'l.ATH 0-1. Skk Cum: XI.lll, 
 
AlU-CoMPRESSOlt, FOll TIIK TuAXSMISSIOX OK MkCMANICAL PoWKU 
 
 Designed by Joiix Ericsson. Built at New Youk, I8;a. 
 
 PERSPECTIVE VIKW. 
 
 Plate 0'4. tii:^ Ciiai: ALU'. 
 
^ 
 
AiR-CoMPincssoi!, Kou TiiK Tkaxsmissiox of Mechanical Puwkk 
 Designed by John Ericsson. Built at New York, 1873. 
 transverse section. 
 
 Platk 65. See Ouai'. XLIV. 
 
^>f^ 
 
 a ^ 
 
 o 
 
 Pi^r^ C6. 6'£'£ C'i/.4i'. XLV 
 
Solar Exgine, operated by the Intervention of Atmospiierto Am 
 Designed hy John Ericsson. Built at New York, 1872. 
 
 Platk 67. See Chap. XLV. 
 
//v^/^^^i /*-■::>! h 
 
 
 /'.--(' 
 
 r"^f y, 
 
 
 **^\!'^c«s,7 A.^'-; 
 
 n:^':?«;/"%\;I,:;- 
 
 
 
 
 ->La:^^ 
 
 
 ^^^^■;■J'^c"^r^.r^^ 
 
 \,^,^ 
 
 
 \/^' 'W'w N'/r^^M /fcA/r ' " 
 
 ^:aaa^^'/a;»^^^VR?^^/>'^^^ ^a?^^' 
 
^v^ryuA L \f^.