PQ 
 
 w 
 
 O) 
 
 W 
 W 
 IX 
 
 O 
 
 U 
 
 fa 
 O 
 
 w 
 
PETROLEUM: 
 
 PRODUCTION AND USE. 
 
 BOVERTON REDWOOD, F.I.C., F.C.S. 
 
 Abridged from the Cantor Lectures before the Society of Arts, 
 London. 
 
 NEW YOKE: 
 D. VAN NOSTRAND, PUBLISHER, 
 
 23 MURRAY AND 27 WARREN STREETS. 
 
 1887. 
 
O 
 

 PREFACE. 
 
 The following pages have been re- 
 printed from the Journal of the Soci- 
 ety of Arts, London, with the omissions 
 of such portions as would seem to be 
 of little or no interest to American read- 
 ers. The petroleum industry occupies a 
 wide field in this country, and the amount 
 of capital invested in it is very great, and 
 while, therefore, all that was stated 
 by Mr. Redwood in his lectures was ad- 
 mirable, and doubtless of much interest 
 to his English hearers, very much, on the 
 other hand, would prove to be of no 
 material value to the American reader. 
 We feel on the whole, however, that the 
 subject matter of these lectures, with the 
 exceptions stated, are worth preservation, 
 and believe that all that is here given will 
 be found to be of interest to readers here. 
 It has been deemed better to change the 
 matter from the lecture form in which it 
 appeared in the columns of the Journal 
 of the Society of Arts. W. H. F. 
 
PETROLEUM AND ITS PRODUCTS. 
 
 CHAPTEK I. 
 
 In the United States petroleum exists, 
 saturating strata of sand-rock. When 
 first wells were drilled in America noth- 
 ing was known about these strata; but 
 ultimately, in the valley of Oil Creek, the 
 existence of three well-defined oil sands 
 was ascertained. These oil sands in the 
 Oil Creek district are of considerable 
 regularity as regards their thickness and 
 the intervening distance between them, 
 the first sand being 40 feet thick, with an 
 interval of 105 feet between it and the 
 second sand, which is 25 feet thick ; an 
 interval of 110 feet occurring between 
 the second sand and the third, the thick- 
 ness of the latter being 35 feefc. In some 
 localities, however, the second sand is 
 split into two well-defined sands, with 
 from 15 feet to 30 feet of slates or shales 
 intervening, and this has given rise to 
 
the definition of a fourth sand. In drilling 
 on high ground in the Oil Creek district, 
 several upper sand rocks were also per- 
 forated ; these were termed " mountain 
 sands." In addition to the three regular 
 sands, there was often found, about 15 
 feet to 20 feet above the regular third, a 
 fine-grained muddy gray sand, from 12 
 feet to 25 feet thick ; this was termed the 
 " stray third." 
 
 The Venango oil sand group is de- 
 scribed by Mr. Carll* as a group in the 
 strictest sense of the term, having a well- 
 defined top and bottom, and consisting 
 of a mass of sandstone deposits from 300 
 to 380 feet thick, with layers of pebbles 
 and many local partings of slate and 
 shale. These figures may, he states, be 
 varied somewhat, but it will be found, as 
 a general rule, that a thickness of 350 
 feet will, in nearly every case, embrace 
 all the sands belonging to the Venango 
 group, even the fourth, fifth, and sixth 
 sands, as the lower members of the group 
 in some localities have been called. The 
 
 * See Penna. Geological Report Second Report. 
 
oil is principally obtained in the third 
 sand. 
 
 As the area of the oil-producing terri- 
 tory was extended, it was found that the 
 underlying geological formation varied 
 considerably, the sand-rock disappearing, 
 and its place being taken by shales at the 
 same geological horizon. 
 
 The several groups of oil-producing 
 rocks are, as Mr. Carll remarks, locally 
 well-defined under certain areas ; but 
 they have their geographical as well as 
 their geological limits, and as far as is at 
 present known, the geographical limit of 
 one group never overlaps that of another. 
 Hence, it must not be presumed that 
 each particular sandstone or its oil will 
 be found in every locality where its 
 horizon can be pierced with the drill, or 
 that a measured section of the rocks in 
 one place can be precisely duplicated in 
 detail in another. Therefore the most 
 skillful oil-producer, the most expert 
 geologist, cannot tell how many other oil 
 horizons may exist at intermediate depths 
 beneath the surface (i. e., in the scale of 
 
8 
 
 the formation), but which have, as yet, 
 escaped the oil-miner's drill. 
 
 In Western Pennsylvania the sand- 
 rock varies in character from a coarse- 
 grained, uncemented sandstone, to a 
 pebble conglomerate, composed of coarse 
 pebbles, of white, or slightly colored 
 opaque quartz, overlaid by marls and 
 slates, often highly silicated, forming 
 very hard and impervious crusts. This 
 pebble conglomerate consists of two 
 varieties, occupying separate horizons, in 
 one of which the pebbles are nearly 
 spherical, and in the other flattened. 
 Between these beds of sandstone or con- 
 glomerate that contain the oil are beds of 
 shale, with which are thin beds of sand 
 and "shells." These shells are described 
 by Professor Leslie as hard crusts of 
 white flint. 
 
 Mr. Peckharn* remarks that petroleum 
 is found in the principal producing terri- 
 tories in the United States and Canada, 
 saturating porous strata, and overlying 
 superficial gravels ; it occurs in Canada 
 
 *Tenth Census U. S. 
 
and West Virginia beneath the crowns of 
 anticlinals ; while in Pennsylvania it 
 occurs saturating the porous portions of 
 formations that lie far beneath the influ- 
 ence of the superficial erosion, like sand- 
 bars in a flowing stream or detritus on a 
 beach. These formations or deposits, 
 taken as whole members of the geological 
 series, lie conformably with the inclosing 
 rocks, and slope gently toward the south- 
 west. The Bradford field in particular, 
 says Mr. Peckham, resembles a sheet of 
 coarse-grained sandstone, 100 square 
 miles in extent, by from 20 to 80 feet 
 deep, lying with its southwestern edge 
 deepest, and submerged in salt water, 
 and its northeastern edge highest, and 
 filled with gas under an extremely high 
 pressure. 
 
 In relation to the geology of natural 
 gas, Mr. Ashburner points out that the oil 
 and gas regions of Pennsylvania are 
 shown by the strata drilled by the gas 
 wells in the neighborhood of Pittsburg 
 to be one in a geological sense. The 
 general conditions upon which the occur- 
 
10 
 
 rence of natural gas seems to depend, 
 from a consideration of the facts at 
 present at our command, are (a) the 
 porosity and homogeneousness of the 
 sandstone which serves as a reservoir to 
 hold the gas ; ( b) the extent to which 
 the strata above or below the gas- sand 
 are cracked ; (c) the dip of the gas-sand, 
 and the position of the anticlines and 
 synclines ; (d) the relative proportions of 
 water, oil, and gas contained in the gas- 
 sand ; and (e) the pressure under which 
 the gas exists before being tapped. 
 Other conditions may, however, as Mr. 
 Ashburner remarks, be still discovered, 
 which will have as important a bearing 
 upon the problem as those which have 
 been stated. 
 
 Vasilieff, who published an article on 
 the " Oil Wells of Baku,'' in the Russian 
 Mining Journal, for September last, an 
 abstract of which, by William Anderson, 
 has recently appeared in the Minutes of 
 the Proceedings of the Institution of 
 Civil Engineers, states that the petroleum 
 bearing strata of the Caucasus belong to 
 
11 
 
 the lower miocene series of the tertiary 
 epoch, the deposits extending in a N.E. 
 to S.W. direction, and the dip ranging 
 apparently between 20 and 40. The 
 petroleum-bearing beds are composed of 
 sand, calcareous clays, marls, and in 
 places, compact sandstone, often of great 
 thickness, penetrated by bands of pyrites. 
 The number of oil-bearing strata is un- 
 known, but three oil- sands have, up to 
 the present time, been defined. 
 
 Petroleum is generally considered to ^ 
 have resulted from the slow decomposi- 
 tion of vegetable or animal matter, either 
 in the rocks in which it is found, or in 
 underlying strata. Berthelot, however, 
 in 1866, propounded the theory that pe- 
 troleum was formed by the action of 
 carbonic acid and steam on the alkali 
 metals ; and in the following year Men- 
 delejeff, in a celebrated essay, read before 
 the Chemical Society of St. Petersburg, 
 and subsequently published in the Revue 
 Scientifique, gave in detail his reasons for 
 believing petroleum to be the product of 
 the action of water upon iron or other 
 
12 
 
 metal, and carbon, at a high temperature 
 and under great pressure. 
 
 The view that petroleum is indigenous 
 to the rocks in which it is found has been 
 strongly supported by Dr. Hunt and 
 Professor Leslie. Mr. Peckham, after a 
 prefatory remark that his studies, extend- 
 ing over twenty years, have led him to 
 the conclusion that, as yet, very little is 
 known regarding the chemical geology of 
 petroleum, expresses the opinion that all 
 bitumens, from the solid to the gaseous, 
 have, in their present condition, originally 
 been derived from animal or vegetable 
 remains, but that the manner of their 
 derivation has not been uniform. Re- 
 ferring to the hypothesis that petroleum 
 is indigenous in the rocks in which it is 
 found, and to that which regards all 
 bitumens as distillates, he remarks that 
 there remains the modifying fact that 
 there are four kinds of bitumen : 
 
 1. Those bitumens that form asphaltum 
 and do not contain paraffine. 
 
 2. Those bitumens that do not form 
 asphaltum and contain paraffine. 
 
13 
 
 3. Those bitumens that form asphaltum 
 and contain paraffine. 
 
 4. Solid bitumens that were originally 
 solid when cold or at ordinary tempera- 
 tures. 
 
 The first class includes the bitumens 
 of California and Texas, doubtless indi- 
 genous in the States in which they are 
 found; and probably also some of the 
 bitumens of Asia. 
 
 Too little is known, Mr. Peckham adds, 
 about petroleum at the present time to 
 enable any one to explain all the phe- 
 nomena on any hypothesis, but it seems 
 to him that the varieties of petroleum 
 found in New York, Pennsylvania, Ohio, 
 and West Virginia are distinctly of vege- 
 table origin, and are the product of frac- 
 tional distillation, as shown by the large 
 amount of paraffine in the Bradford oil, 
 under the enormous pressure to which 
 it is subjected ; while the Kentucky 
 and Californian oils are evidently of 
 animal origin, and have not been sub- 
 jected to distillation. It is not, he con- 
 siders, the effects of heat as represented 
 
14 
 
 by volcanic action that have produced the 
 petroleum, but rather the effects of slow 
 and gentle changes at low temperatures, 
 due to metamorphic action upon strata 
 buried at immense depth. Regarding 
 the nature of the metamorphic action, he 
 states that it is sufficient for our purpose 
 to know that from the upper silurian to 
 the close of the carboniferous periods, 
 the currents of the primeval ocean were 
 transporting sediment from northeast to 
 southwest, sorting them into gravel, sand, 
 and clay, forming gravel bars and great 
 sand beds beneath the riffles and clay 
 banks, in still water, burying vast accu- 
 mulations of seaweeds and sea animals 
 far beneath the surface. The alteration 
 due to the combined action of heat, 
 steam, and pressure, that resulted in the 
 formation of the Appalachian system, 
 from Point Gaspe, in Canada, to Lookout 
 Mountain in Tennessee, involving the 
 carboniferous and earlier strata, distort- 
 ing and folding them, and converting the 
 coal into anthracite, and the clays into 
 crystalline schists along their eastern 
 
15 
 
 border, could not have ceased to act 
 westward along an arbitrary line, but 
 must have gradually died farther and 
 farther from the surface. The great 
 beds of slate and limestone containing 
 fucoids, animal remains, and even indi- 
 genous petroleum, must have been in- 
 vaded by this heat action to a greater 
 or less degree, and thus, in accordance 
 with the theory of Professor Leslie, a 
 chronic evaporation or distillation of the 
 whole mass of oil in the crust of the 
 earth (within reasonable reach of the sur- 
 face) has been going on, converting the 
 animal and plant remains into oils, the 
 light oils into heavy oils, the heavy oils 
 into asphalte and albertite ; the process 
 being accompanied with the liberation of 
 gas. Dealing, in connection with this 
 theory, with the objection of those who, 
 supporting Berthelot's and MendelejefFs 
 views, point out that there is no evidence 
 of the action of heat upon the rocks hold- 
 ing the oil, and no residues of fixed car- 
 bon, Mr. Peckham replies that we must 
 seek the evidences of heat action at a 
 
16 
 
 depth far below the unaltered rocks in 
 which the petroleum is now stored. If 
 petroleum, he adds, is the product of a 
 "purely chemical process," we should 
 not expect to find paleozoic petroleums 
 of a character corresponding with the 
 simple animal and vegetable organisms 
 that flourished at that period, and tertiary 
 petroleums containing nitrogen, unstable 
 and corresponding with the decomposi- 
 tion products of more highly organized 
 beings, but we should expect to find a 
 general uniformity in the character of the 
 substance, wherever found, all over the 
 earth. 
 
 The subject, while one of speculation, 
 is one that obviously is of practical im- 
 portance, as affecting the sources and 
 duration of supplies of petroleum, its 
 profitable development, and commercial 
 permanence. 
 
 Petroleum occurs, as we have seen, in 
 all forms, from the gaseous to the solid. 
 In its liquid form it varies greatly in 
 physical properties, as well as in chemical- 
 composition, and in regard to the propor- 
 
17 
 
 tion of the different commercial products 
 yielded on distillation. 
 
 Dr. Kriimer gives *780 and -970 as the 
 extreme limits of variation in the specific 
 gravity of liquid petroleum ; and in illus- 
 tration of the well-known fact that con- 
 tiguous wells often yield oils of very 
 different quality, instances the case of two 
 wells in the Oelheim district, of the same 
 depth, and within two meters of each 
 other, one of which yields an oil of sp. gr. 
 880, and the other an oil of sp. gr. -905. 
 
 The material known as "ozokerit," 
 large quantities of which are exported 
 from Galicia, may be described as native 
 paraffine wax. Immense deposits of a 
 similar material are, as I have said, stated 
 to exist in the island of Tcheleken, on the 
 Trans-Caspian coast. In Boryslaw and 
 Stanislaw (Galicia) the ozokerit occurs 
 partly in beds and partly in pockets in 
 the miocene formation, and is obtained 
 in small pieces or in masses of several 
 hundred pounds weight. 
 
 Crude petroleum, in the liquid form, 
 consists almost entirely of carbon and 
 
18 
 
 hydrogen, usually in the proportion of 
 about 85 per cent, of the former to 15 
 per cent, of the latter, but there are also 
 sometimes present in small quantity 
 oxygen, nitrogen and sulphur. 
 
 Reichenbach examined paraffine in 
 1824, and rock oil ten years later. Early 
 attempts to determine the composition of 
 petroleum were also made by Laurent. 
 In 1857, De La Eue and Miiller described 
 the products they had obtained from 
 Rangoon petroleum. In 1863, Schorlem- 
 mer isolated some of the constituents of 
 American petroleum, and about the same 
 time Pelouze and Cahours succeeded in 
 separating from this oil twelve distinct 
 hydrocarbons, which were found to be 
 homologues of marsh gas (CH 4 ), and of 
 which the general formula is C u H 2H + 2 . 
 
 Of the more volatile hydrocarbons, 
 viz., those boiling between C. and 
 130 C. there have been shown to be two 
 series present, those of the first series 
 which have the higher boiling points 
 being normal, while those of the second 
 agree for the most part in boiling point 
 
19 
 
 with the corresponding synthetically pre- 
 pared iso-paraffines. There are grounds 
 for belief in the occurrence of a third 
 series of paraffines in the fraction re- 
 ferred to. From the crude oil, as it 
 issues from the earth, methane, ethane, 
 and propane are given off in gaseous 
 form, so that from American petroleum 
 the paraffines referred to have been sepa- 
 rated. 
 
 Methane is a colorless inodorous gas, 
 burning with a yellow flame of little 
 luminosity. Ethane is also a colorless, 
 odorless gas. Propane liqeufies at 20 
 C. Normal butane condenses at C. to 
 a liquid, boiling at 1 C., which consti- 
 tutes the greater part of the petroleum 
 product known as " cymogene." Normal 
 pentane occurs with iso-pentane in the 
 most volatile portion of petroleum spirit. 
 Hexane, heptane, octane, and certain of 
 their isomers constitute the greater part 
 of the liquid known as " benzoline." 
 
 The less volatile portions of American 
 crude petroleum, boiling above 260 C., 
 contain paraffines of still higher order, 
 
20 
 
 those containing 20 per cent, of carbon 
 atoms, or more, being crystalline solids. 
 There are also present hydrocarbons of 
 the O w H an series, which, however, accord- 
 ing to Markownikow, who terms them 
 naphthenes, differ from the olefmes. Dr. 
 Kramer, who has exhaustively examined 
 three varieties of German petroleum, in- 
 clines to the belief that the so-called 
 naphthenes are mixtures of paraffinoid 
 and benzenoid hydrocarbons. 
 
 From the least volatile portion of 
 American crude petroleum a peculiar 
 solid crystalline hydrocarbon was sepa- 
 rated by Morton in 1873. To this hydro- 
 carbon, which in its reactions resembled 
 impure anthracene, the name of "thallene" 
 was given. Morton subsequently found 
 that the spectrum of " thallene " differed 
 from that of impure anthracene. This 
 product was subsequently examined by 
 Prunier ; and more recentty by Dr. 
 Divers and Mr. Nakamura, who have iso- 
 lated a body of the formula C 4 H 3 ) n boil- 
 ing between 280 and 285 C. 
 
 In 1875, Hell and Meidinger obtained 
 
f/Y 
 
 HTJNIVE 
 21 N^ 
 
 from Wallachian petroleum an acid, form- 
 ing alkali salts resembling soft soap. 
 The analysis of the acid, the ether, and 
 the silver salt, agreed best with the for- 
 mula C n H ao O 2 . These chemists ex- 
 pressed the opinion that heavy Walla- 
 chian petroleum contains a series of 
 probably homologous acids. Markowni- 
 kow has found in a fraction of Russian 
 petroleum, boiling between 220 and 230 
 C. as much as 5.25 per cent, of oxygen. 
 This chemist, continuing the research of 
 Hell and Meidinger, ascertained the oc- 
 currence also of bodies of a phenoloid 
 character. Dr. Kramer has recently ex- 
 amined the oxygenated bodies present in 
 German petroleum. 
 
 Caucasian petroleum has, during the 
 past few years, been the subject of much 
 research at the hands of several chemists, 
 among whom may be mentioned Mar- 
 kownikow and Oglobine, and it has been 
 shown that this material is altogether dif- 
 ferent in composition from American 
 petroleum, and that it consists for the 
 most part of hydrocarbons of the CJEE an 
 
22 
 
 series,|isomeric both with the defines, or 
 true homologues of ethylene, and with 
 the hexhydrides of the benzines. The 
 hydrocarbons in question exhibit the 
 closest resemblance to the paraffines, but 
 are of higher density than their iso- 
 logues. They are attacked by chlorine 
 yielding chlorinated derivatives ; but on 
 oxidation are completely destroyed, with- 
 out furnishing characteristic products. 
 The following is a list of the hydrocar- 
 bons of this group, which have been sep- 
 arated from Caucasian petroleum : 
 
 Boiling point. 
 
 C 8 H 16 119C. A. 
 
 C 9 H 18 136 
 
 C 10 H 20 101 
 
 C^H^ 180 
 
 C 12 H 24 196 
 
 C 14 H 28 240 
 
 ; C 16 H 80 247 J 
 
 Ethylene (C 2 H 4 ), which appears to be 
 the lowest member of the series of ole- 
 fines capable of existing in the separate 
 state, is at ordinary temperatures and 
 pressures a colorless gas, burning with 
 a luminous white flame. 
 
 Hydrocarbons of the aromatic (C n H 2W 
 6) series are also found in petroleum. 
 
Pawlewski reports that a volatile pro- 
 duct of the distillation of Galician petro- 
 leum, examined by him, contained as 
 much as 4.9 per cent, of hydrocarbons of 
 this group. 
 
 The burning oil obtained from Ameri- 
 can petroleum contains a considerable 
 proportion of olefines produced from the 
 paraffines. 
 
 The following tabular statement is 
 based upon the analyses of natural gas 
 made by Mr. Carnegie at his works near 
 Pittsburgh, and quoted by Professor De- 
 war: 
 
 COMPOSITION OF Six SAMPLES OF NATURAL 
 GAS. 
 
 
 1. 
 
 2. 
 
 3. 
 
 \ 
 4. 5. 
 
 6. 
 
 
 
 
 
 
 
 
 Per 
 
 Per 
 
 Per 
 
 Per Per 
 
 Per 
 
 cent. 
 
 cent . 
 
 cent. cent. cent. 
 
 cent. 
 
 Marsh Gas 72 1865.2560.70j49.5857.85 i 75.16 
 
 Ethylic hydride 
 
 3 6 
 
 5 5 
 
 7 92 
 
 12 30 
 
 5 20 
 
 4 g 
 
 Oletiant Gas 
 Oxygen . ... 
 
 .7 
 1 i 
 
 .8 
 8 
 
 .98 
 78 
 
 
 
 .8 
 2 1 
 
 .6 
 1 2 
 
 Nitrogen 
 
 Till 
 
 nil 
 
 nil 
 
 nil 
 
 23 41 
 
 2 8 ( 
 
 Carbonic Acid 
 
 8 
 
 6 
 
 nil 
 
 4 
 
 nil 
 
 3 
 
 Carbonic Oxide 
 
 1.0 
 
 .8 
 
 .58 
 
 !4 
 
 1.0 
 
 .6 
 
The co-efficient of expansion of crude 
 petroleum varies according to the pro- 
 portion of the more volatile hydrocarbons 
 present,' as is shown in the following 
 table : 
 
 Sp. Gr. at 15 C. for 1 G ; . 
 
 Under .700 00090 
 
 .700 to .750 00085 
 
 .750 to .800 00080 
 
 .800 to .815 00070 
 
 . Over .815 00065 
 
 The rate of expansion has also been 
 found to vary according to the tempera- 
 ture. 
 
 In practice it is customary to add to or 
 subtract from the observed specific grav- 
 ity .004 for every 10 F. above or below 
 60 F., and this is found to afford a suffi- 
 ciently close approximation to the truth 
 for all commercial purposes in the case 
 of all the ordinary petroleum products. 
 
 Tables for calculating the alterations 
 in volume of crude petroleum with accu- 
 racy, are in use in America. These tables 
 were constructed on Gay Lussac's for- 
 mula : 
 
25 
 
 \ + kt_~P p 
 1+ar" P 
 
 where P= weight of fluid before heating 
 
 it; 
 
 p= weight of the fluid after heating, 
 and after the apparent expan- 
 sion has been removed ; 
 t= change of temperature ; 
 k= co- efficient of expansion of the 
 
 a = co-efficient of expansion of the 
 fluid. 
 
 CHAPTER II. 
 
 Every oil well is naturally divisible into 
 three sections, viz., (1) surface clays and 
 gravels, (2) stratified rocks containing 
 more or less water, (3) stratified rocks, 
 seldom water-bearing, including the oil 
 sands. The first division requires a 
 conductor ; and the second division 
 requires casing to shut off the water 
 from the third section. The earlier 
 method of excluding the water, by 
 
26 
 
 placing a seed-bag round the tubing 
 was found unsatisfactory, as the tubing 
 could not be removed for repairs without 
 disturbing the seed bag, and letting 
 water into the well. In 1868 cast-iron 
 drive pipe was adopted as a substitute 
 for the wooden conductor used in the 
 earlier wells. The most important alter- 
 ation made in 1868 was, however, the in- 
 troduction of 3J-inch casing as a perma- 
 nent fixture. This casing extended to 
 the bottom of the water-bearing rocks, 
 and was furnished either with the seed 
 bag or with a leather cup, which was 
 forced open against the sides of the well 
 by a pressure of the water. The tubing of 
 2f -inch external diameter and extending 
 nearly to the bottom of the well was then 
 placed inside and suspended from the 
 casing. To obtain a supply of water for 
 the boiler a small pipe was often inserted 
 between the tubing and the casing into 
 the water chamber above the seed-bag. 
 Although the 1868 well was a great im- 
 provement on the earlier wells, it pos- 
 sessed defects. Thus the casing being 
 
27 
 
 S^-inches internal diameter, while the un- 
 cased part below it was 5 inches, fishing 
 tools could not be easily introduced, and 
 if it became necessary to deepen the well 
 only 3^-inch bits could be used. The 
 improvements which followed are best 
 exemplified by describing one of the wells 
 of 1878. This well has an 8-inch wrought 
 iron drive pipe, armed at the bottom with 
 a steel shoe. The pipe is driven down to 
 the bed rock, and an 8 -inch, or strictly 
 speaking, 7J-inch hole is drilled to the 
 base of the water-bearing strata. At 
 this point, the bore is gradually 
 reduced to 5 inches and there a bevel 
 shoulder is made ; 5^-inch casing, pro- 
 vided at the lower end with a collar to fit 
 the beveled shoulder, is then inserted 
 and a sufficiently water-tight joint is thus 
 made. Drilling with 5 J -inch bits is then 
 continued until the required depth has 
 been reached. When gas is obtained in 
 sufficient quantity to furnish fuel for the 
 boiler it is conveyed through a 2-inch 
 pipe connected with the casing beneath 
 the derrick floor,and passing into the door 
 
28 
 
 of the furnace. A J-inch steam pipe, fit- 
 ted with an elbow, and ^-inch jet is in- 
 serted in the gas pipe close to the fire 
 box, and a blast of steam is thus caused 
 to issue with the gas. The apparatus acts 
 as an exhauster, drawing the gas from 
 the well, and preventing the flame from 
 running back. The cost of a well in the 
 Bradford field in 1871 was about $3,000. 
 The " water- packer," introduced in 
 1875, is a device to prevent water that 
 may pass into a well below the casing 
 from gaming access to the oil- sand, and 
 to stop the ascent of gas on the outside 
 of the tubing. It is applied round the 
 tubing at any desired point, and its effect 
 is to shut off all communication between 
 the annular space outside the tubing 
 above it and the oil chamber below. The 
 oil and gas are thus confined in the well 
 chamber, and many wells are thus caused 
 to flow that would otherwise require 
 pumping. Under these circumstances 
 the flow is intermittent, taking place 
 when sufficient gas-pressure has accumu- 
 lated. There are many forms of water. 
 
29 
 
 packer, but one of the simplest consists 
 of a band of india-rubber, which, on com- 
 pression, is forced against the walls of 
 the well. If the well does not flow, the 
 oil requires to be raised to the surface by 
 a pump. The working barrel of the 
 pump is placed at the bottom of the well 
 on the end of the tubing, a perforated 
 piece of casing of proper length, termed 
 the " anchor,'' being attached to the lower 
 end of the working barrel. To the sucker 
 of the pump the required number of 
 wooden sucker rods, screwed together, 
 are attached, the upper end of the string 
 of rods being connected with the walking 
 beam. There is, of course, a valve at the 
 bottom of the working barrel, and in the 
 sucker. The sucker is provided with a 
 series of three or four leather cups, which 
 are pressed against the working barrel 
 by the weight of the column of oil. The 
 sucker rods are of ash, 1 inches in 
 diameter by 24 feet to 28 feet in length. 
 When a number of contiguous wells are 
 to be pumped, an arrangement termed a 
 " grasshopper " apparatus is employed. 
 
30 
 
 By this means several wells can be 
 pumped by the action of a single walking 
 beam. 
 
 Most petroleum wells in the United 
 States are "torpedoed" on the completion 
 of the drilling, in order to increase the 
 flow of oil. The torpedo is a charge of 
 nitro-glycerine in a suitable shell, which 
 is lowered to the oil-bearing rock, and 
 there exploded, with the effect of open- 
 ing fissures into the surrounding rock. 
 The shells, which are of tin plate, are of 
 two kinds. One form is lowered to the 
 bottom of the well by a string that can 
 be easily detached and rests on what is 
 termed an " anchor," which is simply a 
 cylindrical tin tube of such length as will 
 bring the torpedo to the required posi- 
 tion. To the upper end of the shell is 
 fitted a "firing head" consisting of a 
 circular plate of iron, only slightly smaller 
 than the bore of the well, having pro- 
 jecting vertically downwards from its 
 lower surface a rod on which a percus- 
 sion cap is placed. Beneath the cap is 
 an anvil. The lowering cord having been 
 
31 
 
 detached and drawn up, a cast-iron weight, 
 termed a " go-devil," is dropped into the 
 well, and this weight striking the disc ex- 
 plodes the percussion cap and fires the 
 torpedo. The other form of shell is sus- 
 pended by a cord, which serves as a 
 guide for a perforated weight running 
 on it. The usual size of the former 
 description of shell is 3^ inches diameter 
 by 10 feet in length, a shell of these di- 
 mensions holding twenty quarts of nitro- 
 glycerine. Frequently as large a charge 
 as eighty quarts is used, and it is then 
 usual to employ four shells of the dimen- 
 sions given, the lower end of one fitting 
 into the upper end of another, and only 
 the top shell of the series having the 
 firing head. Shells of the other descrip- 
 tion are commonly termed squibs. They 
 are of much smaller dimensions, holding 
 only about a quart of the explosive 
 liquid, and are now generally used to bring 
 about the explosion of the large torpedo. 
 I extract the following from the Petro- 
 leum Age for last August : 
 
 "There are nine glycerine firms at 
 
32 
 
 work in the Bradford field, and all their 
 men are kept busy from morning till 
 night. The size of the shot used is 
 rarely less than eighty quarts. The con- 
 stant enlargement of the cavity in the 
 oil-bearing rocks necessitates the use of 
 dynamite squibs for exploding the shells, 
 and the old method of dropping a * go- 
 devil ' on the firing-head of the torpedo 
 has been almost entirely superseded. 
 The cans in which the nitro-glycerine is 
 transported about the field have been en- 
 larged from six to eight quarts capacity, 
 and each shooter's wagon carries ten 
 cans, or eighty quarts of the powerful 
 explosive. It is estimated that over 
 eight tons of glycerine were used in the 
 Bradford field during the month of 
 July." 
 
 The torpedo is usually exploded 
 under about 50 feet of water. Little 
 or no sound is heard, but a slight 
 quiver of the ground is often perceptible. 
 A few moments after the explosion, how- 
 ever, the fluid in the well is shot into the 
 air with great violence, forming a mag- 
 
33 
 
 nificent fountain, and small pieces of 
 rock are also thrown out. The torpedo 
 and exploding weight are blown into 
 small fragments. A few minutes later 
 the well begins to flow, but there is 
 usually a sufficient interval to admit of 
 the casing being connected with a tank 
 in which the oil is collected. The tor- 
 pedo was invented by Colonel Roberts, 
 and patented by him in 1864. 
 
 Some authorities are of opinion that 
 the use of the torpedo is of little value, 
 its effect being simply to clear the 
 pores of the rock of obstructions, and the 
 apparent increase in the yield of oil being 
 simply due to reaction from the immense 
 gas pressure produced by the explosion. 
 Many wells, however, that produced no 
 oil on the completion of the drilling 
 (technically termed " dry-holes ") have, 
 through the use of the torpedo, been 
 caused to yield abundantly. In Russia 
 the torpedo is never used. 
 
 A modification of the rope system 
 of drilling, known as the rod system, is 
 adopted in Russia and in Galicia. The 
 
34 
 
 rod system consists in the substitution of 
 rods of 40 to 60 feet in length, screwed 
 together, for the portion of the drill ing- 
 cable which passes from the end of the 
 walking beam to the string of tools. 
 Iron rods are used in Russia, and wooden 
 rods in Galicia. In the latter country, 
 where the character of the strata is such 
 that the drilling is difficult, and the hole 
 very liable to depart from the vertical 
 line, in which case the well is rendered 
 useless, the rods, in some cases, work in 
 guides. The rod system is stated by 
 some authorities to be preferable to the 
 rope system for use in Eussia, and also 
 in Canada, but it is a less expeditious 
 method of drilling, the time occupied in 
 disconnecting the rods, when the tools 
 are drawn up for the removal of the pul- 
 verized rock, and the sharpening of the 
 bit, being considerable, especially when 
 the well has become deep. In Russia it 
 is usual to commence drilling with a bit 
 as much as 15 inches or 16 inches, or 
 even more, in diameter, but it is generally 
 found necessary to gradually diminish 
 
35 
 
 the size of the bit as the drilling pro- 
 ceeds. 
 
 The depth of the petroleum wells in 
 the United States increased from 436 
 feet in 1861. to 1,606 feet or more in 
 1878. There has been a further increase 
 in depth since the latter year, especially 
 in certain localities. Thus the compara- 
 tively recently drilled Gordon well in 
 Washington county has a depth of 2,400 
 feet/ The cost of this well is stated to 
 have been $7,500 to $8,000. 
 
 The average depth of the petroleum 
 wells in the Caucasus has also been pro- 
 gressive, having increased from 154 feet 
 in 1873, to 450 feet in 1884. Certain of 
 the wells are, however much deeper, one 
 having a depth of 721 feet. It has 
 been estimated that the average level of 
 the oil in the Baku (Balakhani-Saboont- 
 chi) oil field is lowered to the extent of 56 
 feet for every 500,000,000 gallons ex- 
 tracted. 
 
 The irregular character of the strata 
 renders the operation of well-drilling, as 
 a rule, more costly in Southern Kussia 
 
36 
 
 than in the United States, the expense of 
 a well in the Baku district being stated 
 at from 1,000 to 3,000, according to 
 circumstances. The depth of the wells 
 in Galicia range from 250 to 350 meters. 
 The yield of petroleum wells varies 
 greatly. Of the producing wells in the 
 "United States, numbering 20,000, or 
 more, the great majority furnish only a 
 few barrels of oil per day, but some are 
 stated to have yielded, for a short time, 
 as much as 260,000 gallons per 24 hours. 
 This splendid yield is, however, com- 
 pletely eclipsed by that of some of the 
 wells in the Baku district. In well-drill- 
 ing in the latter locality, it is usual to 
 affix to the top of the casing a strong 
 iron cap, provided with a sliding valve, 
 as soon as the oil is " struck." The pe- 
 troleum is thus bottled up, and is only 
 drawn off as required. In Baku, in the 
 autumn of 1884, one of these capped 
 wells was opened. On drawing the 
 slide a mighty column of crude petro- 
 leum, more than a foot in diameter, 
 immediately shot up to a height of 
 
37 
 
 over 100 feet, with a roar, and this 
 magnificent fountain continued to play 
 as long as the valve remained open, form- 
 ing a lake of oil in the neighborhood of 
 the derrick. This well yields at the rate 
 of 1,125,000 gallons per 24 hours when- 
 ever opened. The celebrated Droojba 
 well, and one of Nobel's wells, both 
 yielded, however, for some time, about 
 double that quantity. In the case of the 
 Droojba well, the flow commenced before 
 the cap could be fixed, and the well was 
 for four months quite uncontrollable, 
 sending up a fountain to a height of 200 
 to 300 feet, and deluging the surround- 
 ing land with the oil. The sand thrown 
 up with the oil did considerable damage to 
 neighboring property, engine-houses and 
 workshops being partially buried in it. 
 The use of the caps is not free from diffi- 
 culty, for the oil contains so much sand, 
 that in its rapid flow under the great 
 pressure prevailing, a considerable thick- 
 ness of iron is quickly cut through. The 
 principle of capping wells is not adopted 
 in the United States, the object there 
 
38 
 
 being to drill the territory acquired as 
 quickly as possible, and take out the oil. 
 It has been demonstrated in the United 
 States that there is a lateral flow of oil 
 through the oil- sand (in one instance red 
 paint put into a well was pumped out of 
 another about half-a-mile distant), and it 
 is therefore impossible for an owner or 
 lessee of oil territory to preserve the oil 
 beneath the surface. The oil must be 
 raised, or it would be drained away by 
 wells on neighboring property. 
 ^ The average length of time during 
 (which an oil well in the United States 
 will yield oil in remunerative quantity has 
 been estimated at five years, but, from 
 what has been stated, it will be apparent 
 that the period must necessarily vary 
 within very wide limits. 
 
 The pressure of the oil in the Baku 
 capped wells is frequently as much as 
 200 Ibs., or even in some cases probably 
 f 300 Ibs. per square inch, and although 
 the upper part of the casing is anchored 
 to the ground, there is some danger of 
 the fittings being blown off when the 
 
39 
 
 valve is closed after drawing off a supply 
 of oil. There are about 400 wells in the 
 Baku district, only some 100 of which 
 are productive,* and of the latter not 
 more than 20 were flowing wells. 
 These 20 wells would, however, 
 for a time, yield more than enough 
 crude oil for the manufacture of 
 the 1,800,000, to 2,000,000 gallons of 
 burning oil which the world daily con- 
 sumes. Wells in the Baku district 
 which do not flow cannot be pumped in 
 the ordinary way, in consequence of the 
 large quantity of sand present (sometimes 
 as much as 30 to 40 per cent.), and the 
 oil is raised to the surface in cylinders 
 resembling the sand-pump. 
 
 The cylinder used commonly holds 
 45 gallons, and it is stated that 
 from 18,000 to 20,000 gallons can thus 
 be raised from each well in a working 
 day of 10 hours. 
 
 The gas-wells in the United States are 
 similar to the oil-wells, the casing heads 
 
 * Vasilieff gives the average yield of these 100 wells 
 as 32 tons per well per day. 
 
40 
 
 of some in the neighborhood of Clarks- 
 ville being firmly secured to the ground 
 by chains. 
 
 According to Mr. Carnegie, the largest 
 gas-well known yields about 30,000,000 
 cubic feet of gas in 24 hours, but half 
 this quantity may be considered as the 
 product of a good gas-well. The pressure 
 of the gas at the mouth of one of these 
 wells was shown by the gauge to be 
 187 Ibs. on the square inch, and at the 
 works where the gas was used, nine miles 
 form the well, the pressure was 75 Ibs. 
 per inch. 
 
 When wells have ceased to yield oil in 
 remunerative quantity in the United 
 States, it is usual to draw out the iron 
 casing for use in other wells ; but as this 
 operation allows surface water to gain 
 access to the oil -sand, and as it has been 
 found that the yield of adjacent wells is 
 prejudicially affected by this " flooding," 
 as it is termed, the Pennsylvania Legis- 
 lature enacted that abandoned wells 
 should be " plugged " by filling them 
 with sand. The prejudicial effect of the 
 
41 
 
 flooding of the oil-bearing strata has, it 
 appears, recently been experienced in the 
 Caucasus, the percentage of water in the 
 oil raised in that locality being, accord- 
 ing to Vasilieff, steadily on the increase. 
 When the oil has reached the surface, 
 either by flowing or being pumped, it is 
 conducted into a tank, usually of wood, 
 holding about 250 barrels. In America, 
 quantities of crude petroleum are 
 always stated in "barrels" of 42 
 American gallons (5 American gallons 
 = 4 Imperial gallons). In the early 
 days of the industry in the United 
 States, the only method of transporting 
 the oil from the well was to place it in 
 oak barrels holding 40 or 50 gallons, and 
 to convey these brarels by road to Oil 
 Creek, where their contents were emptied 
 into bulk barges holding about 2,000 
 barrels. As 'Oil Creek was not navigable, 
 arrangements were made with mill own- 
 ers for the use of the surplus water 
 stored in the dams, and at intervals the 
 barges were floated down from dam to 
 dam until they reached the mouth of the 
 
42 
 
 Creek, at its junction with the Alleghany 
 River, from which point there was good 
 flat-boat navigation to Pittsburgh. This 
 method of transportation was not only 
 very costly, but was also attended with 
 frequent loss of oil through the barges 
 coming into collision while being floated 
 down. On one occasion from 20,000 to 
 30,000 barrels of oil were thus lost. 
 Added to this, the roads over which the 
 barrels had to be drawn to the water 
 were little better than paths through the 
 woods. Nevertheless, the system, for 
 want of a better, was for some time 
 largely adopted, over 1 000 boats, 40 
 steamers, and 4,000 men being engaged 
 in the traffic. 
 
 In the latter part of 1862, a branch of 
 the Atlantic and Great Western Railway 
 was carried into the oil regions, and at a 
 later date the Alleghany Valley Railway 
 was opened up from Oil City, at the 
 mouth of Oil Creek, to Pittsburgh, and a 
 number of narrow-gauge railways were 
 constructed as feeders. 
 
 Crude oil was at first conveyed by rail, 
 
43 
 
 in barrels coated internally with glue, 
 but the small quantity of water present 
 in the oil was found to dissolve the glue, 
 and cause the barrels to leak. To re- 
 move this difficulty, and to reduce the 
 cost of handling the oil, tank wagons 
 were adopted in 1865 or 1866. These 
 wagons at first consisted of an ordinary 
 truck, on which were placed two circular 
 wooden tanks, or tubs, holding from 
 2,000 to 4,000 gallons. In 1871 the 
 tank car now employed was introduced.; 
 This consists of a cylinder of boiler-plate, 
 lying upon a four-wheeled truck, and 
 provided with a dome similar to that 
 which a horizontal steam boiler has. 
 The tank is furnished with means of 
 filling at the top, and with a valve be- 
 neath by which it can be emptied. The 
 tank is now usually about 24 feet 6 inches 
 in length by 66 inches in diameter, and 
 holds from 4,500 to 5,000 gallons. 
 
 Tank barges, 130 feet by 22 feet by 16 
 feet, divided into eight compartments 
 with water-tight bulkheads, and holding 
 2,200 barrels, are also at present used 
 
44 
 
 for the conveyance of crude oil on the 
 Alleghany River. 
 
 In 1862 a bill was introduced into 
 the Pennsylvania Legislature for a pipe- 
 line from Oil Creek to Kittanning, but 
 this and a subsequent scheme for 
 laying a pipe-line down the Alleghany 
 River to Pittsburgh, were strongly op- 
 posed and came to nothing. According 
 to Mr. C. L. Wheeler, the credit of 
 having first suggested the laying of a 
 pipe-line belongs to General Karns, while 
 a Mr. Hutchinson was the first to carry 
 out the idea. Hutchinson's pipe, which, 
 was only about three miles in length, 
 was, however, defectively constructed, 
 and leaked so much that little if any of 
 the oil run in at one end reached the 
 other. Mr. Peckham states that the 
 first successful pipe was laid by Van 
 Syckle, of TitusviUe, in 1865. This line, 
 which was four miles in length, and 
 another, were afterwards worked by 
 the Alleghany Transportation Company, 
 though not at first without considerable 
 opposition from the teamsters, who more 
 
45 
 
 than once maliciously severed the pipes. 
 However, by the employment of armed 
 patrols, the lines were preserved from 
 destruction, and after a time the opposi- 
 tion ceased. Gradually a system of 
 pipe-lines, running from the wells to 
 central stations and thence to loading 
 stages on the railway lines, was con- 
 structed, and in 1876 there were eight 
 or nine companies owning pipe-lines in 
 the oil regions, and issuing negotiable 
 certificates for the oil which they col- 
 lected. 
 
 *l At the present time there is in the oil 
 regions of the United States a complete 
 network of 2-inch piping connecting the 
 various wells with storage tanks and 
 trunk lines. These pipes run across 
 country and through streets ; it is impos- 
 sible to get any accurate statistics of 
 their collective length, but it is safe to 
 say that there are thousands of miles of 
 this 2-inch piping thus employed in the 
 collection of the crude oil. 
 
 In 1875, the first trunk line was laid. 
 This extended from the lower oil country 
 
46 
 
 to Pittsburgh, a distance of sixty miles, 
 and was 4 inches in diameter. Like the 
 first pipe-lines from the wells, it had for 
 a time to be protected by armed men. 
 
 As the refining trade developed it be- 
 came concentrated on the seaboard and 
 on the shore of Lake Erie, and the trans- 
 portation of the crude material to the 
 refineries became a business of very great 
 importance. From 1878 to 1881-2, the 
 construction of great trunk lines was 
 continuous. Consolidation of the trans- 
 porting companies also took place, and at 
 the present time the pipe-lines, with one 
 exception, are under the control of a very 
 wealthy corporation, known as the Na- 
 tional Transit Company, which is said to 
 have $15,000,000 invested in oil trans- 
 porting plant. This company owns the 
 following lines : 
 
 Miles. 
 From Olean, N. Y., to New York, Bay- 
 
 onne and Brooklyn (length) 300 
 
 From Colegrove, Pa., to Phila . " 280 
 
 From Millway, Pa., to Baltimore. . " 70 
 
 From Hilliards, Pa., to Cleveland. " 100 
 From Four Mile, Cattaraugus Co., N. Y., 
 
 to Buffalo (length) 70 
 
 From Carbon Centre, Butler Co., Pa., to 
 
 Pittsburgh (length) 60 
 
47 
 
 A total, including the duplicate lines, of 
 about 1,330 miles. The New York line 
 consists of two 6 -inch tubes for the en- 
 tire distance, with a third 6-inch tube for 
 a portion of the way, and is provided 
 with eleven pumping stations about 28 
 miles apart. The transporting capacity 
 of this line is about 28,000 barrels per 
 day. The greatest elevation of the pipe 
 between stations above tide-water is 
 2,490 feet. The Philadelphia pipe has a 
 diameter of 6 inches with six stations ; 
 the Baltimore pipe is 5 inches in diame- 
 ter without a break ; the Cleveland pipe 
 5 inches with four stations ; and the 
 Buffalo and Pittsburgh pipes 4 inches 
 with two stations. 
 
 The pipe is made specially, and is of 
 wrought iron, lap-welded. It is tested 
 to a pressure of 1,500 Ibs. per square 
 inch, the working pressure being 900 to 
 1,200 or even sometimes 1,500 Ibs. The 
 pipe is in lengths of 18 feet, provided 
 at each end with coarse and sharp taper 
 threads, nine to the inch, and the lengths 
 are connected with long sleeve couplings, 
 
48 
 
 also screwed taper. The line is usually 
 laid two or three feet below the surface 
 of the ground, though in some places it 
 is exposed, and at intervals bends are 
 provided to allow for contraction and ex- 
 pansion. At the different pumping sta- 
 tions there are storage tanks of light 
 boiler plate, usually 90 feet in diameter 
 by 30 feet in height, the oil being 
 pumped from the tanks at one station to 
 those at the next, though sometimes 
 loops are laid round the stations, and oil 
 has thus been pumped a distance of 110 
 miles with one engine. The pumping 
 engines chiefly employed are the Worth- 
 ington engines, constructed at the Worth- 
 ington Works in New York, and at each 
 station there is usually a duplicate set. 
 The characteristics of these pumps are, 
 according to the Engineering News, in- 
 dependent plungers with exterior pack- 
 ing, valve boxes subdivided into small 
 chambers, and leather-lined metallic valves 
 with low lift and large surfaces. The 
 engines vary in size from 200 to 800 
 horse-power. The pumps are so con- 
 
49 
 
 structed that before one plunger has 
 completed its stroke another has taken 
 up the work. The column of oil is thus 
 kept continuously in motion, and the 
 violent concussions which occur when the 
 oil column is allowed to come to rest be- 
 tween the strokes are avoided. 
 
 The tanks usually hold about 30,000 
 barrels. They are of boiler plate, roofed 
 with wood, covered with sheet iron, the 
 roof being usually slightly conical. 
 
 The system of issuing certificates for the 
 crude oil stored, adopted by the National 
 Transit Company, is as follows : When a 
 producer has filled the tank at his well, he 
 summons an officer of the company, who, 
 in association with the well owner, gauges 
 the quantity of oil, issues a voucher for 
 the amount, less 3 per cent, to cover loss 
 in transit, and runs the oil into the com- 
 pany's pipes. The oil thus received by 
 the company is treated like a deposit in 
 a bank, and is transferable by written 
 order. Such order, when accepted by 
 the company, is known as a certificate, 
 but as dealings on the Oil Exchange are 
 
50 
 
 usually on the basis of 1,000 barrels, cer- 
 tificates are, as far as possible, made out 
 only for this quantity. The oil is held 
 rent free for 30 days, and at the expira- 
 tion of this time a charge is made for 
 storage. Only a limited amount of clas- 
 sification of the crude oil in the pipe line 
 system is possible, and obviously the oil 
 from a particular well loses its identity 
 as soon as it passes into the company's 
 pipes. The heavy oil from the Franklin 
 and other districts, and also some of the 
 lighter crude oils, are, therefore, trans- 
 ported in barrels. 
 
 The trunk line is owned by the Tide- 
 water Pipe Company. This line, which 
 consists of one pipe 6 inches in diameter, 
 extends from Kixford, in the Bradford 
 field (about eight miles, as the crow flies, 
 southeast of the town of Bradford), in a 
 general southeasterly direction, to Tam- 
 anend, in Schuylkill county ; there the 
 oil is transferred to tank cars, and con- 
 veyed by the Beading Bailroad to Ches- 
 ter, a town fifteen miles from Philadel- 
 phia, or to Bayonne, New Jersey. From 
 
51 
 
 Bixford to Tamanend is a distance of 
 about 170 miles, and in this distance 
 there are five pumping stations. Instead 
 of the stations being placed, as they are 
 on the National Transit Company's lines, 
 at pretty regular distances of 25 or 30 
 miles apart, they are separated by inter- 
 vals corresponding in some measure with 
 the incline, the greatest distance being 
 55 miles, and the shortest 24 miles. By 
 the use of loop lines round the stations, 
 the oil is, however, frequently pumped, 
 in hot weather, when it is most fluid, a 
 distance of eighty miles. The working 
 pressure is 1,000 Ibs. per square inch, 
 and the capacity of the line 10,000 bar- 
 rels per twenty-four hours. At this high 
 pressure leaks occasionally occur, and 
 workmen have had their hands cut to 
 the bone by the fine stream of oil 
 issuing from some minute orifice when 
 engaged in stopping the leaks. 
 
 A very interesting feature of the pipe- 
 line system of transportation is the ar- 
 rangement adopted for cleaning the 
 pipes, and removing obstructions caused 
 
52 
 
 by sediment. The apparatus used is 
 termed a " go devil," a name which, as 
 we have seen, is also applied to the iron 
 weight which serves to explode the tor- 
 pedo. The pipe-cleaning " go-devil" con- 
 sists in many cases of a brush of steel 
 wire of conical form, fitted, at the base or 
 rear end of the cone, with a leather valve 
 in four sections, strengthened with brass 
 plates and also furnished with long steel 
 wire guides. This instrument is impelled 
 by the stream of oil, and travels at the 
 rate of about three miles an hour. Its 
 progress can be traced by the scraping 
 sound which it makes, and it is followed 
 from one pumping station to another by 
 relays of men on foot. It must never be 
 allowed to get out of hearing, otherwise, 
 in the event of its progress being arrested 
 by an obstruction, it may be necessary to 
 take up a considerable length of piping 
 to ascertain its position. 
 
 We have now to consider the method 
 of transportation of crude petroleum in 
 the Caucasus. Up to the year 1876, the 
 transport of oil in this locality took 
 
53 
 
 place in large barrels, which were con- 
 veyed from the wells to the refineries in 
 primitive two-wheeled Persian carts, 
 termed " arbas," one barrel being placed 
 in the body of the vehicle, and another 
 slung between the lofty wheels. Thou- 
 sands of these carts were at one time in 
 use, and it is stated that as much as 
 100,000 per annum has been paid to 
 the carters for this method of transport- 
 ation. Messrs. Nobel Brothers were 
 the first to substitute a pipe line for the 
 *' arba " system of conveyance, and their 
 example being soon followed, there are 
 now seven or more pipes connecting the 
 Balakhani-Saboontchi oil field with the 
 Baku refineries some eight or nine miles 
 distant. Of these lines, Messrs, Nobel own 
 the two largest, of the respective diame- 
 ters of 6 inches and 5 inches. Messrs. 
 Nobel's first line is stated to have cost 
 10,000, and the average cost of a 6 inch 
 line is stated by Mr. Marvin to be about 
 8,000 roubles per verst. Messrs. Nobel's 
 experiment was attended by the same 
 difficulties that were experienced in laying 
 
54 
 
 the first pipe lines in the United States, 
 the native carters strenuously opposing 
 the interference with their business, and 
 the greater lawlessness prevailing in the 
 Caucasus, and the ferocity of the oppo- 
 nents, rendering the task of protecting 
 the line one of no small difficulty and 
 danger. It was, in fact, found necessary 
 to erect a series of watch-houses along 
 the route, in which armed men were 
 stationed. The capacity of the seven 
 pipe lines is estimated at more k than 
 700.000,000 gallons per annum. A con- 
 siderable quantity of crude oil is also 
 transported in tank wagons by rail. 
 
 The ozokerite of Galicia is obtained by 
 sinking shafts from 130 feet to 260 feet 
 in depth, to the beds or pockets in which 
 the material occurs, and then driving 
 tunnels. The shafts generally pass 
 through about 25 feet to 30 feet of gravel 
 or boulders, and then through blue 
 loam and plastic clay. In this clay, 
 usually at a depth of 140 feet to 150 feet 
 from the surface of the ground, the ozo- 
 kerite, or earth wax, is found in layers of 
 
55 
 
 from 1 foot to 3 feet in thickness, the 
 purest being of a honey-yellow color, 
 and of the hardness of beeswax. Much 
 of the ozokerite is, however, in small 
 pieces, and is obtained in admixture with 
 earthy matter, from which it is separated 
 by fusion. 
 
 CHAPTER III. 
 
 The third division of our subject relates 
 to the processes adopted for the man- 
 ufacture and distribution of the various 
 commercial products, as well as to the 
 methods employed for ascertaining the 
 quality of these products and their suit- 
 ability for the purposes to which they are 
 to be applied. 
 
 Petroleum, as we have seen, consists 
 of a mixture of hydrocarbons of varying 
 volatility, and the first operation per- 
 formed upon the crude oil consists in the 
 partial separation of these hydrocarbons 
 by a process of fractional distillation. 
 Before proceeding to describe the appa- 
 
56 
 
 ratus and process, however, it may be well 
 to mention that some of the heavier crude 
 oils are used for lubricating purposes, 
 either in the state in which they are 
 obtained from the wells, or after their 
 density has been increased by the evapo- 
 ration, at a comparatively low tempera- 
 ture, of a portion of the more easily vola- 
 tilized hydrocarbons present. Such oils, 
 called "natural oils" or "reduced oils," 
 are found to be of greater lubricating 
 value than distilled oils of similar density. 
 They are often purified by nitration 
 through animal charcoal. 
 
 In the manufacture by this process of 
 the best oils for the lubrication of steam 
 engine cylinders, a product of less densi- 
 ty is obtained, which, after being filtered 
 through animal charcoal, and subsequent- 
 ly deprived of fluorescence (or " de- 
 bloomed," as it is technically termed) by 
 exposure to the sun, is known as "neutral 
 oil." The product is employed for oiling 
 wool, and sometimes as a spindle oil in 
 silk mills, where a specially fine oil is re- 
 quired. The term " amber oil " is com- 
 
57 
 
 monly applied to an oil made similarly to 
 cylinder oil, but of less density. This oil, 
 which is red rather than amber in color, 
 is purified simply by filtration through 
 animal charcoal, and therefore without 
 the use of acid. It is to some extent 
 used as an engine oil in the United States, 
 and as a lubricant for printing presses. 
 The first attempts to refine petroleum 
 commercially in the United States were 
 probably made in the year 1854, when a 
 still having a capacity of five barrels was 
 erected in Pittsburgh, for refining the 
 small quantity of crude oil obtained in 
 the neighborhood ; but the scarcity of the 
 raw material prevented for some time any 
 important development of the industry. 
 Up to the year 1862, the stills commonly 
 employed had a cylindrical cast-iron body 
 with boiler-plate bottom and cast-iron 
 dome with goose neck bolted on. The 
 capacity of these stills was usually about 
 twenty-five barrels, and .the charge was 
 distilled to dryness. At the present time 
 the process of distillation is divided into 
 two distinct parts. In the first part of 
 
58 
 
 the process stills of two forms are usually 
 employed. These forms are respectively 
 the plain cylindrical still, and what is 
 known as the " cheese-box " still. The for- 
 mer consists of a cylinder of boiler plate,* 
 30 feet in length by 12 feet 6 inches in 
 diameter, furnished with a dome 3 feet 
 in diameter from which passes a vapor 
 pipe 15 inches in diameter. This still is 
 set horizontally in a furnace of brick-work, 
 usually so constructed that the upper 
 part of the still is exposed to the air. The 
 '" cheese-box " still has a body and dome- 
 shaped top of boiler-plate, and a double 
 curved bottom of steel plate. It is 30 
 feet in diameter and 9 feet in height, 
 and is set on a series of brick arches. 
 The vapor is passed from the still 
 through three pipes into a vapor chest, 
 and thence through forty 3 -inch pipes. 
 Wet steam is usually introduced into the 
 exit pipes of these stills so that it may 
 mingle with the vapor. 
 
 The working charge of the cylindrical 
 stills is about 600 barrels and of the cheese 
 
 The lower hilf of ths cylinder is generally of steel. 
 
(UNIVEESIT' 
 59 
 
 box stills about double that quantity. 
 Connected with the stills are the condens- 
 ers. These were formerly copper worms, 
 but now consist of iron pipes, usually 
 straight, passing through tanks of cold 
 water. A modern condensing arrange- 
 ment may be described as consisting of a 
 series of forty separate 3-inch pipes, the 
 vapor passing in at one end, and the con- 
 densed oil flowing out at the other end 
 of each pipe. In some cases, methods of 
 fractional condensation, more or less 
 complicated, designed to effect the more 
 complete separation of the commercial 
 products, are adopted. 
 
 In the second part of the process of 
 distillation, cylindrical stills, which are 
 commonly of steel, and hold about 260 
 barrels, are employed. 
 
 The crude petroleum on reaching the 
 works is placed in storage tanks, and is 
 thence pumped into the stills. A fire having 
 been lighted in the furnace beneath the 
 still the temperature of the oil is gradually 
 raised, and the more volatile constituents 
 distilled off. The crude naphtha thus 
 
60 
 
 obtained is sometimes collected in two 
 fractions. Sufficient of the lighter hy- 
 drocarbons having been separated to 
 give the next commercial product (which 
 is the burning oil of commerce) the re- 
 quired flashing point or fire test, the dis- 
 tillation is continued until a point is 
 reached at which the operation must be 
 arrested, otherwise the burning oil would 
 have too high a specific gravity. The 
 product of burning oil is frequently 
 divided into two or more fractions. In 
 practice, the termination of the collection 
 of the naphtha and the burning oil re- 
 spectively takes place when the distillate 
 issuing from the condensers reaches a 
 specified density. The percentage of 
 these two commercial products varies ac- 
 cording to the character of the crude 
 oil, and the method of distillation ; the 
 average yield may, however, be stated to 
 be about 12 per cent, of naphtha, and 75 
 per cent of burning oil (110 fire test). 
 Of the higher class of burning oil known 
 as " water-white oil,'' only from 12 to 20 
 per cent is obtained. The fluid residue 
 
61 
 
 in the still, known as residuum, and 
 amounting to about 6 per cent., having 
 been run off, there remains a quantity of 
 coke, which represents from 1 per cent. 
 to 1^- per cent, of the original charge. 
 The time occupied in working a charge 
 is from three to four days. This separa- 
 tion of the naphtha and burning oil con- 
 stitutes the first part of the process of 
 distilling the crude petroleum. The sec- 
 ond part of the process, which is usually 
 conducted at other works, consists in dis- 
 tilling the residuum to dry ness, and obtain- 
 ing lubricating oils and solid paraffine. 
 The residuum, which has an average speci- 
 fic gravity of about 19 B. (equal to .942) 
 is, when practicable, Conveyed from one 
 refinery to the other in bulk barges. The 
 time occupied in the working of the charge 
 is about thirty hours. The distilled pro- 
 ducts are collected in several fluid frac- 
 tions, holding the solid hydrocarbons in 
 solution. The coke which remains in the 
 still at the end of the process amounts to 
 about 12 per cent. 
 
 Superheated steam is often passed into 
 
62 
 
 the stills during the distillation. In the 
 production of some of the special grades 
 of u reduced oils, 7 ' vacuum stills are em- 
 ployed. The crude naphtha is redistilled 
 by steam heat in cylindrical stills hold- 
 ing 500 barrels, and is sometimes sepa- 
 rated into the following commercial pro- 
 ducts, the more volatile of which are 
 colorless : 
 
 No. Density. Name. Use. 
 
 1. 90B .. Rhigolene or cymogene. 
 
 For surgical purposes . 
 
 2. 88/86B .. Gasoline. 
 
 For air-gas machines. 
 
 3. 76B . . Boulevard gas fluid. 
 
 For street naphtha lamps. 
 
 4. 73/68B .. Prime city naphtha (benzoline.) 
 
 For " Sponge Lamps," &c. 
 5 62B .. Benzine. 
 
 For oil-cloth and varnish making. 
 
 The time occupied in working the 
 charge is about 48 hours. The percent- 
 age of these products varies, but as a rule 
 amounts to about 25 per cent, of the first 
 three collectively, rather more than 25 
 per cent, of the 4th and about 40 per 
 cent, of the 5th. 
 
 Such is an outline of the method of 
 distilling crude petroleum in the United 
 
63 
 
 States. There are, however, many detail 
 variations, the precise conditions under 
 which the operation is conducted being 
 by no means uniform. In all cases, how- 
 ever, the principal object in view is to 
 obtain the largest yield of burning oil, 
 lubricating oil and paraffine, consistent 
 with these products being of satisfactory 
 quality. 
 
 The next step in the process of refin- 
 ing consists in a treatment of the distil- 
 late with sulphuric acid. This operation 
 is conducted in tall cylindrical wrought 
 iron tanks, 40 feet or more in height by 
 20 feet or more in diameter, sometimes 
 lead-lined, holding from 1,200 to 1,800 
 barrels, and termed agitators. The oil 
 having been pumped into the tank, a 
 blast of air under a pressure of 5 to 7 
 pounds per square inch is introduced 
 through a pipe at the base of the tank, 
 and the oil thus brought into a condition 
 of active agitation. Oil of vitriol is then 
 forced to the top of the tank by air press- 
 ure, and is gradually " showered " into 
 the petroleum through a perforated 
 
64 
 
 leaden pipe, in the proportion of about 
 six pounds of acid to one barrel of oil. 
 After the agitation has been continued 
 for a sufficient length of time to bring the 
 acid thoroughly into contact with the oil, 
 the air blast is shut off, and the tarry 
 acid (termed acid tar or sludge acid) 
 having been allowed to settle to the bot- 
 tom of the tank, is drawn off. The acid 
 is occasionally recovered and concen- 
 trated for further use. The oil is then 
 agitated with water, next with a solution 
 of caustic soda, to complete the removal 
 of the acid, and finally with water again, 
 to which sometimes a little ammonia is 
 added. The oil is then run into a shal 
 low rectangular iron tank, usually pro- 
 vided with a steam coil for raising the 
 temperature in cold weather, where the 
 water which it contains settles out, 
 and the oil becomes bright. In connec- 
 tion with this tank there is an arrange- 
 ment for " spraying " the oil. This opera- 
 tion consists in running the oil in fine 
 streams through small orifices in pipes 
 placed about the tank, and its effect is to 
 
65 
 
 remove some of the more volatile hydro- 
 carbons present, and thus bring the "test" 
 of the oil up to the required point. From 
 the settling tanks the oil passes to the 
 barreling or canning tanks. 
 
 The treatment of the less volatile of 
 the naphtha distillates with acid is simi- 
 lar to that which is applied to the oil, but 
 mechanical agitation, by means of a ver- 
 tical revolving shaft fitted with arms, is 
 usually substituted for the air blast, as 
 the latter would cause the evaporation of 
 the more volatile constituents. A simi- 
 lar method of agitation was formerly 
 adopted in treating the burning oil. The 
 acid treatment is applied to the naphtha 
 with the object of deodorizing, but in the 
 case of the oil an improvement of color 
 as well as of odor results. 
 
 The lubricating oil distillates are also 
 treated with acid, but before being sub- 
 jected to this process they are cooled in 
 order to separate the solid hydrocar- 
 bons. The cooling was formerly effected 
 somewhat rapidly, but it has now become 
 the practice to operate upon a consider- 
 
66 
 
 able bulk of oil, and to reduce the tem- 
 perature slowly, as it is found that the 
 paraffine is thus obtained in a more crys- 
 talline form. The cooling or chilling 
 tanks, employed at some of the most 
 modern works,have a capacity of 3,000 gal- 
 lons, and the period occupied in the cool- 
 ing is as much as twenty- six hours. The 
 reduction of temperature is effected 
 through the medium of a solution of 
 magnesium chloride, brought to the re- 
 quired temperature by means of an am- 
 monia refrigerating apparatus. This 
 liquid circulates through a number of 
 cells alternating with similar cells con- 
 taining the oil. At the expiration of the 
 period named, the semi-solid mass is re- 
 moved from the cooling tank, placed in 
 bags, and subjected to hydraulic press- 
 ure at a temperature of 40R To effect 
 a further separation of the oil, the press 
 cakes are then broken up, or melted and 
 recrystallized, and subjected to a second 
 pressure of about 200 pounds per square 
 inch at a temperature of 70. The resid- 
 uum yields about 9 per cent, of paraffine 
 
67 
 
 by this treatment. For the production of 
 hard and colorless paraffine wnx, or of 
 wax suitable for candle making, the par- 
 affine scale obtained as described is puri- 
 fied by crystallization from petroleum 
 spirit, or by exposing it to a temperature 
 just sufficient to cause the fusion and 
 draining out of the hydrocarbons of low- 
 est melting point, and by subsequent 
 nitration through animal charcoal. The 
 expressed oil, which varies in specific grav- 
 ity, according to the manner in which the 
 various fractions have been separated in 
 the process of distillation, is then subject- 
 ed to an acid treatment similar to that 
 which the burning oil undergoes, and is 
 thus converted into finished lubricating 
 oil. 
 
 Some lubricating oils are finished by 
 evaporation and by filtration through ani- 
 mal charcoal, the best oils for the lubri- 
 cation of engine cylinders passing through 
 the latter process. Petroleum lubricat- 
 ing oils consist chiefly of olefines (C n H 2n ). 
 The most volatile of the fractions obtained 
 in the distillation of the petroleum resid- 
 
68 
 
 uum, having a specific gravity of .820 
 (42 B.) and low flashing point, contains 
 but little paraffine, and is sold for gas- 
 making in the condition in which it issues 
 from the condensers. 
 
 The semi-solid mixture of uncrystalline 
 hydrocarbons known as vaseline, is 
 obtained by the Chesebrough Manufact- 
 uring Company, under their patent, by 
 evaporating off the more volatile portion 
 of a suitable kind of crude petroleum and 
 purifying the residue by filtration through 
 animal charcoal. The product thus 
 obtained is a colorless or pale yellow 
 translucent semi-solid, possessing slight 
 fluorescence. It is freely soluble in 
 petroleum or shale spirit, benzol, ether, 
 chloroform, carbon disulphide and tur- 
 pentine. Vaseline is a mixture of hydro- 
 carbons of which the chemical composi- 
 tion chiefly ranges from C 16 H 34 to C 20 H 42 . 
 There is reason to believe that olefines 
 are present, but the substance consists 
 mainly of paraffines. 
 
 The yield of burning oil from a crude 
 petroleum of given quality has been large- 
 
69 
 
 ly increased during recent years. This 
 has been accomplished by the adoption 
 of the process known as " cracking." We 
 have seen that the crude petroleum pro- 
 duced in the States of Pennsylvania and 
 New York consists principally of the hy- 
 drocarbons known to chemists as paraf- 
 fines. The researches of Thorpe and 
 Young have demonstrated that the par- 
 affines (C n H 2n + 2 ) when heated to temper- 
 atures above their boiling points, are 
 converted into olefines (C n H 2n ), carbon 
 being deposited, and gaseous products 
 evolved. It is this operation of disso- 
 ciation which is termed "cracking," and 
 its employment enables the refiner to 
 break up the hydrocarbons which are too 
 heavy to be burned in ordinary lamps, 
 and too fluid for use as lubricants, 
 and convert them into hydrocarbons 
 which may be allowed to pass into the 
 burning oil distillate without unduly in- 
 creasing its density. The process of 
 cracking is carried out by conducting the 
 operation of distilling the burning oil so 
 slowly that the less volatile hydrocarbons 
 
70 
 
 become condensed in the upper part of 
 the still, and fall back into the heated oil, 
 where they are heated to temperatures 
 above their boiling points, and become 
 " cracked." A far larger yield of burning 
 oil is thus obtained. The gaseous pro- 
 ducts are conducted into the still fur- 
 nace, and serve as fuel. This process 
 has been subjected to a good deal of ad- 
 verse criticism, based upon the view that 
 the chemical composition of the olefines 
 renders them inferior to the paraffine for 
 illuminating purposes, and that their 
 capacity of forming substitution com- 
 pounds with the acid used in refining 
 leads to the contamination of the burning 
 oil with sulphur products. There is no 
 doubt good theoretical ground for this 
 contention, and it is frequently found 
 that the American petroleum oil of 
 commerce, though showing no trace of 
 the presence of acid when shaken with 
 barium chloride solution, is blackened 
 on heating, and during distillation gives 
 off sulphur dioxide. It is not, however, 
 certain that in the practical use of the 
 
71 
 
 oil any distinct disadvantage can clearly 
 be traced to the action of the opera- 
 tion of cracking, and it is obviously to 
 the interest of the poorer classes, who so 
 largely use petroleum oil as a source of 
 light and heat, that a system of manu- 
 facture which considerably increases the 
 yield, and therefore diminishes the cost 
 of the product, should not be hastily con- 
 demned. Moreover, it should be borne 
 in mind that the proportion of cracked 
 oil in the product is not usually very 
 large, and therefore, that objections 
 which would attach to the use of the 
 cracked oil alone may not apply to the 
 product as a whole. By some it is claimed 
 that distillation of the burning oil over 
 caustic alkali has the effect of removing 
 the substitution products formed in the 
 acid treatment. The burning oils chiefly 
 manufactured in the United States are of 
 the following grades: 110 fire test, 70 
 Abel test ; 120 fire test, 73 Abel test ; 
 150 fire test. 
 
 The color of the first four grades 
 usually ranges from " Prime White " to 
 
72 
 
 " Standard White " (straw color to pale 
 yellow), while the fifth grade is " Water 
 White " (colorless). The rules of the 
 New York Produce Exchange provide 
 that refined petroleum for contract pur- 
 poses shall be standard white or better, 
 with a burning test of 110 Fahrenheit, 
 or upwards, and a specific gravity not be- 
 low 44 Beaume, United States Dispen- 
 satory Standard (not above .811 specific 
 gravity). Formerly the bulk of the oil 
 intended for export was refined to 110 
 fire test, the United Kingdom, however, 
 taking 120 fire test oil. Since the intro- 
 duction of the Abel system of testing, 
 however, a large proportion of the oil 
 has been refined to a test of 70 Abel for 
 shipment to the Continent of Europe, and 
 of 73 Abel for shipment to England. 
 The oil of 150 fire test is considerably 
 lower in specific gravity than the other 
 grades (say .786 to .793). Oils are also 
 specially refined, that is, made to stand 
 special tests, where so required. In 
 addition to the products enumerated, 
 an oil of high specific gravity, and high 
 
73 
 
 flashing point, intermediate in character, 
 in fact, between the ordinary burning 
 oils and the lubricating oils, is made. 
 This oil was originally produced by the 
 Downer Kerosene Oil Company under 
 the name of mineral sperm oil, for use in 
 lighthouses, and its use is now compul- 
 sory on some of the American railroads. 
 It is also largely used on board ship. Of 
 this oil, as now manufactured, the crude 
 petroleum yields about 10 per cent., the 
 product having a specific gravity of about 
 .829, and a fire test of about 300 F. 
 (Flashing point, open vessel, about 270 
 F.; closed vessel, about 240F.). This 
 product is frequently termed mineral 
 colza oil, or mineral seal oil. 
 
 The number of grades of lubricating 
 oil manufactured are so great, and their 
 qualities so various, that it is difficult to 
 give any satisfactory enumeration of 
 them. The oil varies in color from yellow 
 to very dark red, and is strongly fluores- 
 cent. 
 
 The solid paraffine (using the term in 
 its commercial sense) is, like the oil, a 
 
74 
 
 mixture of various hydrocarbons of dif- 
 ferent specific gravities, boiling points, 
 and melting points. The paraffine scale 
 separated from the oil of .905 specific 
 gravity has an average melting point of 
 about 125F. (American test), and that 
 from the oil of .885 specific gravity a 
 melting point of about 117 F. 
 
 " Paraffine wax " is a white or bluish- 
 white translucent waxy solid of lamino- 
 crystalline structure, devoid of taste and 
 smell. The hydrocarbons of which it is 
 composed are characterized by chemical 
 indifference; hence the name (parum 
 affinis.) The specific gravity, melting 
 point, and boiling point of the material 
 vary with its composition. At a temper- 
 ature below its melting point paraffine 
 wax becomes plastic, a characteristic 
 which is of such practical inconvenience 
 when the substance is used as candles, 
 that it is the practice in candle making 
 to add a small quantity of stearic acid, 
 which to some extent diminishes the ten- 
 dency of the candle to bend in a warm 
 room. 
 
75 
 
 When two pieces of paraffine wax are 
 struck together, a sharp metallic sound 
 is emitted, especially if the melting point 
 of the specimen be high. Exposed for 
 some time under a slight pressure to a 
 temperature below its melting point, 
 paraffine wax undergoes a molecular 
 change and becomes transparent ; but 
 upon a change of temperature, or upon 
 being struck, the original translucent ap- 
 pearance returns. 
 
 Paraffine wax is freely soluble in petro- 
 leum or shale oil and spirit, in ether, in 
 benzol, and in essential oils. It is spar- 
 ingly soluble in hot absolute alcohol, but 
 separates on cooling. It is insoluble in 
 rectified spirit and in water. 
 
 When boiled with concentrated nitric 
 acid, paraffine is oxidized with the form- 
 ation of succinic acid, and a small quan- 
 tity of butyric acid. Paraffine is also 
 oxidized when heated with potassium per- 
 manganate and sulphuric acid. At a 
 high temperature it is slowly attacked 
 by concentrated sulphuric acid, and chlo- 
 rine passed into melted paraffine attacks 
 
76 
 
 it slowly. When heated with sulphur, 
 paraffine is decomposed, sulphuretted 
 hydrogen being evolved and carbon de- 
 posited. 
 
 A considerable quantity of crude petro- 
 leum is exported from the United States 
 to France, to be refined there, the French 
 duty on the refined oil encouraging the 
 home industry. The rules of the New York 
 Produce Exchange provide that crude 
 petroleum shall be understood to be pure 
 natural oil, neither steamed nor treated, 
 free from water, sediment or any adulter- 
 ation, and of the gravity of 43 to 48 
 Beaume (.816 to .794 sp. gr.). 
 
 Crude naphtha and residuum are also 
 exported in comparatively small quanti- 
 ties. The naphtha is required to be water- 
 white and sweet, and of gravity from 68 
 to 73 Beaume, while the residuum is to 
 be understood to be the refuse from the 
 distillation of crude petroleum, free from 
 coke and water, and from any foreign im- 
 purities, and of gravity from 16 to 21 
 Beaume (.96 to .93 sp. gr.). 
 
 Natural and reduced lubricating oils 
 
77 
 
 (West Virginia, &c.) are sold under con- 
 tract rules, fixing the limit of setting 
 point, the latter oils being divided into 
 two grades termed respectively summer 
 reduced oil and winter reduced oil. 
 
 Some of the refineries in the United 
 States are of great size. Thus the works 
 at Bayonne, New Jersey, owned by the 
 Standard Oil Company, in whose hands 
 the greater part of the refining trade is 
 concentrated, covers about 67 acres, and 
 is capable of turning but from 10,000 to 
 12,000 barrels of refined oil per day. At 
 some of the most modern works the plant 
 is for the most part uncovered by roof- 
 ing. Precautions are, however, taken 
 against the spread of fire, water sprinklers 
 and steam jets being fitted to the storage 
 tanks, and hydrants being provided. At 
 all large refineries there is a well- drilled 
 fire staff. At the Bayonne works the 
 fire brigade is drilled twice a month* 
 and four fire hoses can be brought 
 into action upon a tank in 30 sec- 
 onds, 1 minute 10 seconds, 1 minute 
 twenty seconds and 1 minute 30 seconds 
 
78 
 
 respectively from the blpwing of the 
 alarm whistle, three of the hoses being 
 brought some distance. At the works of 
 the Pratt Manufacturing Company there 
 is a fire wall which prevented the spread 
 of fire from a burning tank some time 
 ago, although, according to the evidence, 
 the fire burned with great fierceness un- 
 til all of the oil in the tank had been 
 consumed, and the wind was blowing 
 towards the wall. It is considered that 
 a properly protective fire wall should be 
 five feet higher than the tank and about 
 ten feet distant from it. 
 
 The export of the oil has, with the 
 exception of one shipment, hitherto 
 taken place exclusively in "barrels of 
 the size mentioned, or in tin cans of 
 rectangular form, holding five American 
 (or four Imperial) gallons, which are 
 packed in twos in wooden cases. The 
 manufacture of the barrels, cans and 
 cases is, as may be supposed, a very im- 
 portant business. The barrel works of 
 the Standard Oil Company at Bayonne 
 are capable of turning out 10,000 to 12,- 
 
79 
 
 000 barrels per day, and the operation of 
 barrel making at these works is an inter- 
 esting one to witness. The oak staves 
 are purchased ready jointed and sea- 
 soned, in Michigan, and the barrel 
 heads are brought to the Bayonne 
 works ready glued up. The first 
 operation in barrel making at these 
 works consists in fitting the necessary 
 number of staves together in a thick 
 wrought-iron ring encircling their lower 
 ends. This is an operation requiring 
 some experience and judgment. The 
 embryo barrel is then placed in an iron 
 cylinder and steamed, whereby the wood 
 is softened. The staves are next encir- 
 cled by a wire rope connected with an 
 engine, and are thus bent into shape and 
 drawn together, a second strong iron 
 hoop being slipped over their upper ends 
 to hold them in position. The barrel is 
 then " fired " by burning some readily 
 combustible material in the interior, and 
 the curvature of the staves thus rendered 
 permanent. A number of extra tem- 
 porary iron hoops of great thickness are 
 
80 
 
 next slipped on, and drawn toward the 
 bulge of the barrel by means of an ingen- 
 ious arrangement of iron hooks or claws 
 actuated by steam power. The final 
 operation performed upon the staves con- 
 sists in placing the barrel in a lathe, par- 
 ing off the rough ends, and cutting the 
 grooves for the heads. The barrel is then 
 ready to receive the heads, and to be 
 hooped. The hoops weigh collectively 
 about 12 pounds, and the total length of 
 iron required for a set is 443 \ inches, so 
 that putting the out-turn of finished 
 barrels from this one factory at 10,000 
 per day, we have a length of about 70 
 miles of hoop iron (weighing about 55^- 
 tons) used daily. 
 
 In order to render the barrels capable 
 of holding their fluid contents without 
 leakage, they are coated internally with 
 glue, about one pound of glue to 
 3 barrels being required. The glue 
 solution is poured into the barrels ot, 
 the barrel bunged up, and rotated so that 
 the solution coats the entire surface, the 
 surplus being afterwards drained out. 
 
81 
 
 There is some pressure of steam in the 
 barrel during this operation, and a leak 
 is thus at once shown. The barrels 
 finally receive a coating of the well-known 
 blue paint on the staves, and white paint 
 on the heads. Old barrels returned to 
 be refilled are often cleaned externally 
 by an arrangement of rapidly revolving 
 wire brushes, are steamed out, reglued 
 and repainted. 
 
 Before the barrels are filled, the hoops 
 require " driving v to take up the shrink- 
 age of the wood. This was formerly 
 done exclusively by hand, but Mr. Hop- 
 per has invented a successful machine for 
 doing this by steam power. In this ap- 
 paratus the barrel stands on a platform 
 arranged like an inverted steam hammer, 
 and on turning on the steam it is brought, 
 with a succession of blows, against a num- 
 ber of hinged stops, which closely en- 
 circle the barrel and on which the hoops 
 strike. With one such machine the hoops 
 of 2,000 barrels can be driven in ten hours 
 by one man and two boys an amount of 
 
82 
 
 work which formerly entailed the hand 
 labor of ten men. 
 
 The barrels are filled from a rack pro- 
 vided with a series of pipes connected 
 with a barreling tank. Each pipe has 
 at its exit end a float connected with a 
 valve, which shuts off the oil when the 
 barrel has been filled to within one gal- 
 lon of its contents. The shives with 
 which the barrels are closed are of wood 
 and are put in with glue. A package 
 which remains perfectly tight and free 
 from leakage as long as it is handled 
 carefully, and the continuous skin of glue 
 remains intact, is thus produced. 
 
 The can of petroleum is a package 
 standing a good deal of rough usage. 
 The empty can is so strong, not- 
 withstanding that the tin plate from 
 which it is made is thin, that a full- 
 sized man may stand upon it without 
 crushing it. Cans are used exclusively 
 in the shipment of oil to warm climates,, 
 as the barrel, though less expensive per 
 gallon of oil, is liable to become leaky 
 when subjected to a high temperature. 
 
83 
 
 Within the past few months two ves- 
 sels, the Crusader and the Andromeda, 
 have been fitted with tanks for the con- 
 veyance of the refined burning oil in 
 bulk. The former vessel is provided 
 with an arrangement of tanks patented 
 by Mr. L. V. Sone, of New York. The 
 patent specification contains twelve 
 claims, but the most important feature 
 of the system appears to be the provis- 
 ion of auxiliary tanks above the level of the 
 storage tanks and in communication with 
 them. The storage tanks can thus 
 always be kept quite full of oil, the auxil- 
 iary tanks serving to hold the surplus 
 when the contents of the storage tanks 
 become expanded by heat, and supplying 
 the deficiency when contraction takes 
 place. The tanks with which the Crusader 
 is fitted are of boiler plate, cylindrical in 
 form, of various sizes, but of the average 
 capacity of 125 barrels. There are 45 
 tanks, and the total carrying capacity of 
 the vessel is, therefore, about 5,500 bar- 
 rels. All the tanks are connected with a 
 steam pump on deck, by means of which 
 
84 
 
 they can be charged or discharged in 
 about ten hours. 
 
 About a month after the Crusader ar- 
 rived in London, the Andromeda reached 
 Bremen with a cargo of 684,641 Ameri-. 
 can gallons of petroleum oil in bulk. 
 This vessel contains 72 tanks, which, in- 
 stead of being cylindrical, are principally 
 rectangular. Expansion tanks similar to 
 the auxiliary tanks of the Crusader are 
 provided. 
 
 The only special feature in the process 
 of refining petroleum in Canada is the 
 treatment of the burning oil distillate 
 with a solution of litharge and caustic 
 soda for the purpose of removing sulphur. 
 Notwithstanding the adoption of this 
 process, the refined petroleum, or burn- 
 ing oil, has not, however, always been 
 free from this impurity. In 1876 a 
 sample of Canadian petroleum oil pro- 
 duced a deposit of a bluish -black color 
 on the chimney of the lamp in which it 
 was burned. On examining the deposit 
 small beads of liquid were found to be 
 deposited where the glass was very hot, 
 
85 
 
 and a piece of litmus paper applied to 
 these beads was reddened and then 
 charred. On washing out the chimney 
 with water, and adding to the liquid a 
 solution of barium chloride, a copious 
 precipitate was produced. It was evi- 
 dent that the liquid condensed on the 
 glass was sulphuric acid, and on burning a 
 weighed quantity of the oil, and collecting 
 the products of combustion, sulphur was 
 found to be present in the oil to the ex- 
 tent of 116 grains per gallon. Another 
 sample of Canadian oil, used in a green- 
 house to the detriment of the plants, 
 contained 119 grains of sulphur to the 
 gallon. 
 
 We now pass to the consideration of 
 the method -oLjcefining and transporting 
 petroleum oil adopted in Russia. Most 
 of the Caucasian refineries are situated 
 in the Tchorni Gorod or Black Town of 
 Baku, on the shores of the Caspian. 
 There may be seen refineries of all sizes, 
 the smallest being constructed and man- 
 aged in a primitive fashion. The largest 
 works are those of the Nobel Company, 
 
86 
 
 an organization occupying in relation to 
 the refining of Russian petroleum, a 
 position analogous to that which the 
 Standard Oil company holds in regard to 
 the American refining trade. The works 
 of the Nobel Company cover from 78 to 
 80 acres, and the dwellings of the em- 
 ployes another 40 acres. In regard to 
 the substantial character of the buildings 
 and completeness of the arrangements, 
 the refinery is equal to any in the 
 United States. The only essential 
 respect in which the plant differs 
 from that described, in treating of the 
 American refineries, is in the adoption 
 of a process of continuous distilla- 
 tion, the stills, each of the capacity of 
 4,400 gallons, being arranged in groups 
 or series of not more than twenty-five, 
 and a stream of oil continuously flowing 
 through the entire number. The crude 
 oil entering the first still parts with its 
 most volatile constituents, passing into 
 the next still, has rather less volatile 
 hydrocarbons distilled from it, and finally 
 flows from the last still in the condition 
 
87 
 
 of residuum, which in Russia is termed 
 astatJd or masut. The several stills are 
 maintained at temperatures correspond- 
 ing with the boiling points of the pro- 
 ducts to be volatilized. Attempts have 
 been made to carry on a system of con- 
 tinuous distillation of petroleum in 
 America, and an apparatus for the pur- 
 pose was patented in the United States 
 by Stombs and Brace, of Newport, Ken- 
 tucky, April 10, 1860 (Patent No. 27,842), 
 but the process has never been practically 
 employed in that country. Mr. Ludwig 
 Nobel has said that the method of 
 continuous distillation adopted in his 
 Tfrorks was specially suited to Baku petro- 
 leum, the quantity of burning oil sepa- 
 rated being comparatively small, and the 
 residuum, therefore, not very much less 
 fluid than the crude oil. The fuel used 
 under all the stills in Baku is petroleum 
 residuum or astatki. At many of the 
 smaller works, the liquid fuel is simply 
 allowed to flow on the hearth of the fur- 
 nace, and in thus using it a very large 
 amount of dense black smoke is evolved ; 
 
88 
 
 but in the larger and better conducted 
 refineries, arrangements are adopted for 
 burning the fuel with a proper admix- 
 ture of air, and smokeless combustion is 
 thus obtained. At the Nobel refinery the 
 two most volatile distilled products are of 
 specific gravity .754 and .787 respectively, 
 while the kerosene (burning oil) has a 
 specific gravity of .820 to .822. The 
 more volatile products are largely burned 
 as fuel, as there is little demand for them,, 
 and in fact a considerable quantity is 
 burned to waste in order to get rid of it, 
 At the Nobel works the kerosene dis- 
 tillate is pumped into lead lined agitators- 
 of the capacity of 57,000 gallons, where r 
 in the manner already described, it is 
 treated with about 1 per cent, of sul- 
 phuric acid, washed with sea water (fresh 
 water being very scarce in Baku), treated 
 with caustic soda solution, and washed 
 again. The oil then passes to the set- 
 tling tanks, and is afterward stored in 
 large tanks, one of which holds 1,500,000 
 gallons. The whole process of treatment 
 occupies 15 to 16 hours. The sulphuric 
 
89 
 
 acid is made in a neighboring factory from 
 sulphur imported from Daghestan. The 
 oil is almost superfine white in color, and 
 has an average flashing point of about 
 32C. (Abel test.) Of such product the 
 crude oil yields, on an average, only 
 about 27 per cent., and of oil quite free 
 from color, or water- white, and having a 
 flashing point of 50 C., only about 22 per 
 cent, can be obtained. At Baku, the Cas- 
 pian Company make three grades of burn- 
 ing oil of the respective specific gravities 
 and flashing points of (1) -81530 C.; 
 (2) -820-25 C., (3) -821/-822-22 0. 
 Of the first quality, the crude oil yielded 
 20 per cent., of the second 33 per cent., 
 and of the third 38 per cent. The yield 
 of burning oil from Baku petroleum i& 
 therefore comparatively very small, and 
 to this the fact that the oil, although of so 
 high a specific gravity, burns well is no 
 doubt partly due, since the product is- 
 necessarily very homogeneous, the most 
 volatile hydrocarbons present not differ- 
 ing very greatly in boiling point from 
 the least volatile present. Besides this, 
 
90 
 
 however, the hydrocarbons of which the 
 Baku oil is composed possess apparently 
 greater power of ascending the wick of 
 the lamp by capillary attraction than is 
 found in ordinary American oils. Tak- 
 ing ordinary petroleum lamp wicks of 
 two different qualities, and using them as 
 siphons, in a given time the following 
 quantities of the oils enumerated were 
 removed by capillary attraction from a 
 vessel into which the wicks dipped, the 
 conditions of the test being the same in 
 each case ; 
 
 Best Inferior 
 
 Wick. Wick. 
 
 grs. grs. 
 
 American oil (" water-white "). 205.0 .. 104.2 
 
 Russian oil 202.6 .. 94.2 
 
 American (ordinary) 146.0 . . 69.7 
 
 Eussian oil does not burn equally well 
 in all the forms of lamps which have been 
 constructed for use with the American 
 product, but as the result of a large am ount 
 of comparative photometrical testing of 
 the two classes of oils burning in lamps 
 of various construction during the past 
 two or three years, taking the lamps now 
 
91 
 
 in common use as a whole, the Russian oil, 
 though giving less light than good ordi- 
 nary American oil at the commencement 
 of the burning, when the lamp is full of 
 oil and freshly trimmed, affords a flame 
 of somewhat greater permanence, the 
 light emitted in the burning of the 
 American oil diminishing to a greater 
 extent as the level of the oil in the 
 reservoir becomes depressed, the dif- 
 ference being, doubtless, due to the 
 greater power which the Russian oil has 
 of ascending the wick. In the use of 
 Russian oil, however, it is essential that 
 the air passages in the burner should be 
 quite free from obstruction by dirt, other- 
 wise the flame will receive an insufficient 
 supply of oxygen, will be of comparative- 
 ly feeble luminosity, and will be liable to 
 
 smoke. As a rule the Russian oil burns 
 
 
 
 better in a flat- wick burner than in a 
 round-wick burner, but burners of the 
 latter class are now being constructed in 
 which it yields a good flame. 
 
 The astatki, or residuum, as it flows from 
 the kerosene still, has a specific gravity 
 
92 
 
 of about .903, and is chiefly used as fuel. 
 A portion of it is, however, converted 
 into lubricating oil, and a smaller portion 
 into an oil or oils similar to the " mineral 
 sperm oil" obtained from American 
 petroleum. Of the lubricating oil the 
 residuum yields about 30 per cent., and 
 of a product similar to the American 
 " Mineral Colza " oil, about 12 per cent. 
 Messrs. Ragosine were the first to intro- 
 duce an oil of this character made from 
 Russian petroleum. This oil, which is 
 frequently termed " pyronaphtha," as 
 now imported into England has a 
 specific gravity of about .865, a fire test 
 of 265 F., and a flashing point of about 
 230 F. open test, or 205 F. closed test. 
 This oil, like the similar American pro- 
 duct, has not hitherto been much used 
 for domestic purposes, as the lamps in- 
 tended to burn it did not give a satis- 
 factory light. Lamps are now, however, 
 being introduced which burn it well. 
 
 The lubricating oil distillates obtained 
 from Baku petroleum differ from the 
 American products in containing little 
 
93 
 
 or no solid hydrocarbons, the greatest 
 quantity obtainable amounting only to 
 a quarter of one per cent, of the crude 
 oil. In the island of Tcheleken there is, 
 however, petroleum to be obtained which 
 yields as much as 6 per cent, of paraffine. 
 
 This absence of solid hydrocarbons de- 
 prives the Baku refiner of a source of 
 profit possessed by the American refiner, 
 but on the other hand enables him to 
 make cheaply a lubricating oil which 
 bears exposure to a very low temperature 
 without becoming solidified or even de- 
 positing any paraffine. Professor Dewar 
 has suggested that the products of Baku 
 petroleum may be distinguished from 
 those of American petroleum by exposing 
 them to a temperature sufficiently low to- 
 cause the separation of the solid hydro- 
 carbons which are held in solution at 
 common temperatures. 
 
 The lubricating oils manufactured by 
 Messrs. Kagosine include the following: 
 
 Extra heavy cylinder oil, sp. gr. .920. 
 
 Cylinder and valve oil, sp. gr. .912/.915. 
 
 Engine machinery oil, sp. gr. .905/.907. 
 
94 
 
 Dark cylinder oil, sp. gr. .918/.920. 
 
 Russian lubricating oils are character- 
 ized by the possession of great viscosity 
 in relation to specific gravity, but the vis- 
 cosity of the Russian oils is more affected 
 by a rise in temperature, a feature which 
 is undoubtedly a disadvantage, but in the 
 case of both classes of oils the greater 
 the viscosity the greater the diminution 
 in viscosity as the temperature is raised. 
 
 When paraffines are heated to tempera- 
 tures above their boiling points dissocia- 
 tion occurs and olefines are produced : 
 
 C 7 H 16 + C 13 H, C , &c. 
 
 At a very high temperature hydrocarbons 
 containing still less hydrogen are pro- 
 duced. This well known fact has been 
 taken advantage of by Mr. Nobel to man- 
 ufacture from Russian petroleum resid- 
 uum, or astatki, benzine, naphthalene 
 and anthracene. The process employed 
 consists in breaking up the astatki on the 
 highly heated floor of a cupola regenera- 
 tive furnace. The first destructive dis- 
 
95 
 
 VISCOSITIES OF RUSSIAN AND AMERICAN OILS. 
 
 Temper- 
 ature. 
 Fahr. 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 50 
 
 712J 
 
 145 
 
 425 
 
 1030 2040 
 
 2520 
 
 
 60 
 
 540 2 
 
 105 ! 295 
 
 680 
 
 1235 1 19801 
 
 
 
 70 
 
 405 
 
 90 J225 
 
 485 
 
 820 
 
 1320 
 
 
 
 80 
 
 326 
 
 73 171 
 
 375 
 
 580 
 
 900 
 
 
 
 90 
 
 260 
 
 63* 
 
 136 
 
 262 
 
 426 
 
 640 
 
 
 
 100 
 
 213* 
 
 54" 
 
 111 
 
 200 
 
 315 
 
 440 
 
 1015 
 
 110 
 
 169 
 
 50 
 
 89* 
 
 153 
 
 226 
 
 335 
 
 739^ 
 
 120 
 
 147 
 
 47 
 
 78' 
 
 126 
 
 174 
 
 245 
 
 531 
 
 130 
 
 123J 
 
 44} 
 
 63i 
 
 101 
 
 135J 
 
 185 
 
 398J 
 
 140 
 
 105i 
 
 41 
 
 58 
 
 82 
 
 116 
 
 145 
 
 317 
 
 150 
 
 95* 
 
 37J 
 
 52 
 
 70* 
 
 05 
 
 115 
 
 2Q 
 
 160 
 
 85" 
 
 
 
 46 
 
 63* 
 
 
 93* 
 
 200 
 
 170 
 
 76 
 
 
 
 
 
 58" 
 
 70i 
 
 77} 
 
 161 
 
 180 
 
 69 
 
 
 
 
 
 62J 
 
 Wf 
 
 67* 
 
 134* 
 
 190 
 
 64^ 
 
 
 
 
 
 47 
 
 56^ 
 
 61" 
 
 115| 
 
 200 
 
 58| 
 
 
 
 
 
 42 
 
 48* 
 
 54 
 
 99^ 
 
 210 
 
 54 
 
 
 
 
 
 40 
 
 " 
 
 
 
 85 
 
 220 
 
 50 
 
 
 
 
 
 38 
 
 
 
 
 77 
 
 230 
 
 4.7.1 
 
 
 
 
 
 
 
 
 
 
 
 70 
 
 240 
 
 45* 
 
 
 
 
 
 
 
 
 
 
 
 G4 
 
 250 
 
 43! 
 
 
 
 
 
 
 
 
 
 
 
 59 
 
 260 
 
 
 
 
 
 
 
 
 
 
 
 
 
 54 
 
 270 
 
 
 
 
 
 
 
 
 
 
 
 
 
 48 
 
 280 
 
 1 
 
 
 
 
 
 
 
 
 
 46^- 
 
 290 
 
 
 
 
 
 
 
 : 
 
 
 
 44^ 
 
 300 
 
 
 
 
 
 
 
 
 
 
 
 42| 
 
 1. Refined rape oil. 
 2. American mineral oil, sp. r. .855 
 
 3. " .913 
 
 4. ' .923 
 
 5. Russian .909 
 
 6. 4 .915 
 
 7. " ' .884* 
 
 " Semi-solid at common temperatures 
 
96 
 
 tillation thus effected is stated to yield 
 from 30 to 40 per cent, of tar containing 
 from 15 to 17 per cent, of 50 per cent, 
 benzol. By a second destructive distilla- 
 tion of the heavy oils remaining after the 
 separation of the benzol, 70 per cent, of 
 tar is obtained, containing from 7 to 10 
 per cent, of 5 per cent, benzol, 16 per 
 cent, of naphthalene, 2 to 3 per cent, of 
 " dry green grease " (or 30 per cent, an- 
 thracene), and 24 per cent of pitch. There 
 is also obtained in the process 75 to 100 
 cubic feet, per c. f. of astatki, of gas 
 having an illuminating power described 
 as five times that of coal-gas. 
 
 We have already seen that American 
 petroleum oil is principally transported 
 in barrels and cases. As, however, suit- 
 able timber for barrel making is scarce 
 in Russia, Mr. Nobel found it necessary 
 to devise some other n:ethod of distribut- 
 ing the oil, and great credit attaches to 
 him for the admirable system of bulk 
 transportation he has organized. The 
 oil is pumped from the refineries at Baku 
 into tank steamers on the Caspian sea. 
 
97 
 
 These steamers, which were constructed 
 at Motala, in Sweden, are about 250 feet 
 in length by 28 feet beam, and have a 
 draft of water when loaded of 10 to 12 
 feet. The whole of the forward portion 
 of the vessel forms one great tank, the 
 engines and boilers are amidships, and two 
 cylindrical tanks are placed aft. The os- 
 cillation of the oil in the forward tank is 
 prevented by bulkheads. The tanks 
 hold collectively 225,000 gallons (5,500 
 barrels), and can be filled in 4 hours. 
 The fuel used for steam generation is 
 petroleum residuum, 29 tons being con- 
 sumed on the voyage from Baku to the 
 mouth of the Volga, a distance of about 
 460 miles. The engines are of 120 horse- 
 power nominal, and the mean speed of 
 the vessels 10 knots per hour. Several 
 tank steamers have recently been con- 
 structed for the shipment of petroleum 
 oil in bulk from Batoum to ports on the 
 Black sea, the Baltic and elsewhere. 
 These vessels are built on the principle 
 of the boilers and engines being placed 
 in the stern, and the oil storage tanks 
 
98 
 
 forward. Two bulkheads with water be- 
 tween prevent the possibility of any oil 
 finding its way into the boiler space. 
 The steamers are constructed through- 
 out of Bessemer steel, and have been 
 favorably classed at Lloyds. 
 
 In the tank steamers described, the oil 
 is conveyed to mouth of the Volga, 
 where, in consequence of the water being 
 shallow, it is transferred to tank barges, 
 in which it is carried to Tsaritsin, an im- 
 portant storage center on the Volga, 364 
 miles from the point at which the steam- 
 ers discharge. At Tsaritsin the oil is 
 pumped into tank wagons (similar to 
 those used in America) on the Tsaritsin- 
 Griazi railway, and thus finds its way to 
 a large number of storage centers 
 throughout Russia. From these centers 
 the oil is distributed in barrels which can 
 easily be returned to be refilled. 
 
 A considerable quantity of the oil is 
 also conveyed from Baku to Batoum, on 
 the Black Sea, over the Trans- Caucasian 
 railway, in tank wagons, a distance of 
 560 miles, but the carrying capacity of 
 
99 
 
 this line is only about one million and a 
 half of barrels per annum, in consequence 
 of its being carried over a pass (the 
 Suram Pass) 3,000 feet above sea level, 
 where the gradient is in places as much 
 as 1 in 22. It is contemplated to largely 
 increase the carrying capacity of the line 
 by tunneling the pass, and it has also 
 been proposed to lay a pipe line from 
 Baku to Batoum. In the meantime, 
 however, arrangements are being com- 
 pleted to fully utilize the existing facili- 
 ties by shipping the oil from Batoum in 
 tank steamers, and, as the Volga naviga- 
 tion is closed in winter, there is no doubt 
 that the Trans Caucasian route will be 
 largely made use of. 
 
 In the following table the varied yield 
 of commercial products is shown : 
 
o : 
 
 5 --( 
 
 rz: oo oo oq oo oo oo oq oq co QO 
 
 o 
 
 bC 
 g 
 
 OH ^ O, < G a 
 
 J3 
 
 00 ^> *O TH CO GQ ^ O iO "^ CQ ' 
 OB 
 
 I 
 
 I 
 
 t>OOOOOqoOOC5O:;QOCtOOOOCiOOt-Oi<; 
 
 . . ......; o : 
 
 ^ '05 bp i * i i ' os ' ' *^ 
 r-T52 .'O !** 
 
 tZ 53 .S J "02 
 
 **'*.' ? * S 
 
 ' g ** *5 
 
101 
 
 We now enter upon the consideration 
 of the process of manufacturing the 
 various commercial products from shale. 
 True shales, when heated to redness in 
 a closed vessel, do not cake, so that the 
 soft and black residue, after all the 
 volatile matter has been driven off, re- 
 tains the original form of the fragments. 
 The proportion of mineral matter in the 
 shale is usually 73 per cent., but is occa- 
 sionally as much as 80 percent. 
 
 In the earlier days of the industry two 
 methods of distillation, one intermittent 
 and the other continuous, were employed. 
 In the former system, which was that 
 first adopted, the shale, previously 
 broken into fragments of suitable size, 
 was heated in cast-iron retorts, similar 
 to those employed in coal-gas manu- 
 facture, the retorts being discharged and 
 recharged when the whole of the vola- 
 tilizable matters had been driven off. 
 The latter system was conducted in a 
 cylindrical or oval retort of cast iron, 
 about 2 feet in diameter by 8 feet or 10 
 feet in length, set vertically in a furnace. 
 
102 
 
 In this vessel the shale, previously 
 broken by machinery, was exposed to a 
 dull red heat for twelve to twenty hours ; 
 a jet of steain being introduced at the 
 bottom of the retort, and the products 
 conducted from the top. The steam was 
 employed to sweep away the products as 
 formed, and prevent their dissociation by 
 over-heating. The furnace was so con- 
 structed that the lower part of the retort 
 received the greatest heat, and from time 
 to time a portion of the exhausted shale 
 was removed from the bottom of the re- 
 tort, which was sealed with water, and a 
 corresponding quantity of fresh shale 
 was introduced at the top ; the process 
 being thus continuous. By this process 
 about 33 gallons of crude oil and 80 
 gallons of ammoniacal liquor per ton of 
 shale were obtained, the latter being 
 large in volume and of low strength, in 
 consequence of the condensation of the 
 steam employed. 
 
 The method of distillation devised by 
 Mr. Henderson differs principally from 
 the intermittent system already described, 
 
103 
 
 in the retorts being vertical, and in the 
 provision of an arrangement for employ- 
 ing as fuel the spent shale, which some- 
 times contains as much as 12 to 14 per 
 cent, of carbon. The retorts are, in this 
 system, placed in an oven above the 
 furnace. They are about 15 feet in 
 length, and hold about 18 cwt. of shale. 
 The bottom of the retort consists of a 
 door, which can be opened so as to allow 
 the spent shale to drop into a combus- 
 tion chamber beneath the oven. The 
 material, as it is discharged from the 
 retort, is black, but it soon becomes in- 
 candescent, and aided by the incondensa- 
 ble gaseous products of distillation (oc- 
 casionally also by the addition of a small 
 quantity of coal), it constitutes a valuable 
 fuel. In this system of distillation the 
 upper part of the retort is the most 
 highly heated, and the products are re 
 moved from the bottom, a stream of 
 superheated steam being passed in at the 
 top. There are four retorts in each 
 oven, and each retort is charged once in 
 sixteen hours ; but the charging takes 
 
104 
 
 place in rotation at intervals of four or 
 five hours, so that a regular supply of 
 fuel is furnished to the combustion 
 chamber below. 
 
 Although, in the meantime, the older 
 method of intermittent distillation had 
 been improved, it is generally admitted 
 that the introduction of the Henderson 
 retort resulted in an increased yield of 
 paraffine. 
 
 The retort invented by Messrs. William 
 Young and George Beilby, the Pentland 
 pattern of which is now also largely in 
 use, is composite, the upper part being 
 of iron, and the lower of fire brick. In 
 this apparatus the shale is exposed to a 
 comparatively low temperature while in 
 the iron portion the oil products being 
 thus driven off without risk of their being 
 overheated and then passing down- 
 wards into the more highly heated fire 
 brick portion, in consequence of the 
 withdrawal of a portion of the spent 
 material, is subjected to a heat suitable for 
 the formation of the ammonia. Steam is 
 passed into the retort at the base. The 
 
105 
 
 retorts are set in groups of four, aiul 
 usually are heated through the medium 
 of a gas producer instead of by an ordi- 
 nary open furnace. By this method of 
 distillation the yield of paraffine is in- 
 creased, while the gain of sulphate of 
 ammonia is stated to be no less than 14 
 Ibs. per ton of shale. 
 
 The oil vapors are condensed by being 
 passed through a series of 70 to 100 ver- 
 tical 4 inch pipes. 
 
 Crude shale oil is of a dark green color, 
 and has a specific gravity of .865 and up- 
 wards. The first step in the process of 
 refining consists in the distillation of the 
 oil to dryness in cast-iron pot- shaped 
 stills, into which steam, often super- 
 heated, is passed, the product being 
 condensed in the manner already de- 
 scribed when treating of the distillation 
 of petroleum. The oil having been 
 treated with sulphuric acid and caustic 
 soda, these processes being termed 
 "washing," is next subjected to frac- 
 tional distillation in cylindrical boiler- 
 plate stills of a capacity of 4,000 to 5,000 
 
106 
 
 -gallons. Steam is introduced into the 
 stills after the more volatile products 
 have passed off. The first product ob- 
 tained is crude naphtha, and at a higher 
 temperature the burning oil distillate 
 issues from the condensers and is col- 
 lected separately. The heavy oil which 
 remains is distilled in cast-iron vessels. 
 The burning oil distillate thus produced 
 is subjected to acid and alkali treatment, 
 and undergoes a second fractional dis- 
 tillation, an additional quantity of naph- 
 tha being separated, and some heavy oil 
 remaining in the still. The burning oil 
 distillate then passes through a third 
 chemical treatment and distillation, and 
 is, in some cases, finally agitated with 
 acid and alkali, and thoroughly washed. 
 The crude naphtha receives the usual 
 chemical treatment, and is separated by 
 fractional distillation into commercial 
 products. From the heavy oil, lubrica- 
 ting oils are manufactured, and parafime 
 separated and purified substantially as in 
 the case of the similar petroleum pro- 
 ducts. It was formerly the practice to 
 
107 
 
 crystallize the paraffine rapidly, by bring- 
 ing the heavy oil into contact with the 
 surface of a revolving drum through 
 which cooled calcium chloride solution 
 was made to circulate, the paraffine ad- 
 hering to the cool surface being scraped 
 off and removed. Mr. Henderson, and 
 subsequently Mr. Beilby, however, pat- 
 ented systems of slow cooling in con- 
 siderable bulk, as adopted in petro- 
 leum refineries in the United States. 
 Various other systems have been adopted 
 at Addiewell and Bathgate for the slow 
 cooling of the oil containing paraffine, it 
 being now generally recognized that the 
 production of large crystals of paraffine 
 resulting from this method of working 
 facilitates the expression of the oil, and 
 effects a saving in press cloths. From 
 the gaseous products passing away 
 through the condensers crude gasoline is 
 obtained, by a process of scrubbing with 
 heavy oil in a coke tower, or by subject- 
 ing the gas to pressure. The crude 
 gasoline passes through a similar process 
 
108 
 
 of purification to that which the crude 
 naphtha undergoes. 
 
 From the intermediate oils, burning* 
 oils of high specific gravity and flashing 
 points are obtained. 
 
 Without going minutely into details of 
 manufacture, which are by no means the 
 same at all works, an intelligible out- 
 line description of the comparatively 
 lengthy and complicated process by 
 which the various shale products are 
 obtained is here given. In practice the 
 various similar distillates obtained in the 
 intermediate operations are mixed to- 
 gether, the process, as a whole, having for 
 its object to classify the various hydrocar- 
 bons by successive fractional distillations. 
 It should be stated that the crude shale 
 oil yield, in addition to the products enu- 
 merated, acid and basic bodies. 
 
 At the works of Young's Paraffine 
 Light and Mineral Oil Company, the fol- 
 lowing is the average yield of the various 
 commercial products from the crude 
 shale oil : 
 
109 
 
 Per cent. 
 
 Gasoline 0.25 
 
 Naphtha sp. gr. .700 to .760 5.75 
 
 Burning oils : ~] 
 
 No. 1, sp. gr. .802 to .804, F.P. 
 
 110 (Abel test) 
 
 No. 2, sp. gr. .810 to .812, F.P. 
 
 100 (Abel test) f 
 
 Crystal (No. 1, chem. treated). . . I 
 Lighthouse oil, sp. gr. .810 to .820 I 
 
 F.P. 140 (Abel test) j 
 
 Lubricating oils of various specific 
 
 gravities 14.50 
 
 Paraffine (solid) 11.00 
 
 Loss 30.50 
 
 100.00 
 
 The percentages given are only ap- 
 proximate, and are often purposely varied 
 by alterations of the processes to suit 
 the requirements of the markets. The 
 loss is no doubt frequently considerably 
 smaller than the proportion stated. 
 
 At the Broxburn works the average 
 yield is as follows : 
 
110 
 
 Per cent. 
 
 Naphtha sp. gr. .730. , , 5 00 
 
 Burning oils: 
 
 Petroline sp. gr. .800/.802 "1 
 
 No. 1, oil-sp. gr. .808/.810 ^ 37.28 
 
 Lighthouse oil sp. gr. .810 j 
 
 Lubricating oils 17.40 
 
 Solid paraffine 12.52 
 
 Loss.. 27.80 
 
 100.00 
 
 Mr. Alfred H. Allen, in his "Commer- 
 cial Organic Analysis," asserts that shale- 
 burning oil contains about 36 per cent., 
 by measure, of paraffines, and 64 per 
 cent, of olefines and other hydrocarbons 
 acted on by fuming nitric acid ; while in 
 American refined petroleum (burning oil) 
 these proportions are reversed ; and he 
 gives the following comparative state- 
 ment, based upon the relative percent- 
 ages, by measure, of hydrocarbons present 
 in the various commercial products which 
 withstand a consecutive treatment with 
 nitric acid, 1.45 specific gravity, concen- 
 trated sulphuric acid, fuming sulphuric 
 acic, and caustic soda ; remarking, how- 
 
Ill 
 
 ever, that the quantitative composition is 
 liable to considerable variation, and hence 
 must not be interpreted too literally:* 
 
 NAPHTHA. 
 
 From Shale. At least 60 to 70 per 
 cent, of heptylene, C 7 H 14 , and other hy- 
 drocarbons of the olefine series, C n H 2n . 
 The remainder paraffines, C n H 2n+2 . No 
 trace of benzine or its homologues. 
 
 From Petroleum. At least 70 per 
 cent of heptane, C 7 H 16 , and other hydro- 
 carbons of the paraffine series, C n H 2n+2 . 
 The remainder apparently olefines^ with 
 distinct traces of benzine, C C H 6 , and its 
 homologues. 
 
 BURNING OIL. 
 
 From /Shale. 50 to 80 per cent, or 
 more, of the higher members of the 
 olefine series, C n H 2n . The remainder 
 paraffines, C n H 2n+a . 
 
 * In regard to.the burning oil made from American 
 petroleum, the proportion of olefines present depends 
 upon the extent to which the process of "cracking " 
 has been adopted, as already explained. 
 
112 
 
 From Petroleum. 50 to 80^ per cent, 
 of the higher members of the paraffin e 
 series, C n H 2n 4_ 2 . The remainder chiefly 
 olefines, C n H 2ll . 
 
 LUBRICATING OIL. 
 
 From Shale. Chiefly olefines, C n H 2n? 
 with some polymerized members of acety- 
 lene series, C n H 2n _ 2 . 
 
 From Petroleum. A large proportion 
 of higher olefines, C n H 2n ; but less than in 
 corresponding shale product. 
 
 VASELINE. 
 
 From Shale. No such product. 
 From Petroleum. Chiefly higher par- 
 affines of low melting point. 
 
 PARAFFINE WAX. 
 
 From Shale. Solid paraffines, 
 
 C D H, n+a . 
 From Petroleum. Solid paraffines, 
 
 UnH an + a . 
 
 Instead of acting on the olefines by 
 nitric and sulphuric acids as described, 
 
113 
 
 bromine may be used to effect their 
 separation from the paraffines, and a pro- 
 cess of analysis based upon this fact has 
 been devised by Mr. Allen. 
 
 Mr. J. J. Coleman found that the 
 liquid obtained on subjecting the gase- 
 ous products of shale distillation to cold 
 and pressure, consisted chiefly of ole- 
 fines: Butylene, C 4 H 8 ; amylene, C 5 H IO ; 
 and hexylene, C 6 H 12 . 
 
 The following results, obtained by Mr. 
 Galletly with paraffine from Boghead 
 coal, show that the specific gravity of 
 this product increases with its melting 
 point : 
 
 Specific gravity. Melting point. 
 
 .8236 89.6 F. 
 
 .8480 1022 
 
 .8520 104.9 
 
 .9090 128.0^ 
 
 .9110 128.0 
 
 .9243 136-4 
 
 .9248 138.2 
 
 .9400 176.0 
 
 The ozokerite, separated by melting 
 from the earthy matter with which it is 
 associated, and purified by treatment with 
 
114 
 
 Nordhausen oil of vitriol, has a deep 
 yellow color, and in that condition is 
 stated to be used as an adulterant of 
 beeswax. To obtain a material suitable 
 for candle making, it has, however, 
 hitherto been necessary to subject the 
 ozokerite to distillation in a current of 
 super-heated steam, the products being 
 about 5 per cent, of gaseous hydrocar- 
 bons, 3 per cent, of naphtha, 6 per cent. 
 of semi-solid "ozokerine," 12 per cent, of 
 soft paraffine (melting point 44^ to 46 
 C.), distilled ozokerite (melting point 
 61 C.), and a black waxy residue. The 
 grayish colored distilled ozokerite thus 
 obtained is refined by a process similar 
 to that which paraffine from petroleum 
 or shale undergoes. The refining of 
 ozokerite is carried out on a large scale 
 by Messrs. J. C. and J. Field. 
 
 About the time when Mr. James Young 
 laid the foundation of the present shale oil 
 industry, Mr. Eeece Keece and Sir Robert 
 Kane were working upon Irish peat, and 
 Sir Frederick Abel has stated that in 
 1851 or 1852 he saw some small candles 
 
((UNIVEKSIT1 
 115 
 
 made from peat paraffine. According to 
 Dr. Edmund J. Mills, peat furnishes from 
 3 to 6 per cent, of tar, the specific gravity 
 of which is about .954, and Vohl states 
 that peat tar yields, on the average, about 
 20 per cent, of burning oil (specific 
 gravity .82), and about 22 per cent, of 
 lubricating oil. The percentage of par- 
 affine appears to vary considerably, as it 
 is placed by different writers at from 0.1 
 per cent, to 3.4 per cent. 
 
 It will be interesting at this point to 
 consider, briefly, the extent of the world's 
 consumption of mineral illuminating oils. 
 We find that the estimated total out-turn 
 by all the refineries in the United States, 
 for the year ended December 31, 1885, 
 was 732,650,628 American gallons, or 
 14,365,698 barrels of 51 American gallons 
 (equal to about 40 imperial gallons). The 
 approximate home consumption in the 
 United States amounted to about 253,- 
 665,075 American gallons, or 4,973,825 
 barrels; and the quantity exported was 
 5,381,099 barrels, and 17,254.611 cases, 
 collectively equal to 446,982,159 Ameri- 
 
116 
 
 can gallons. The detailed shipping sta- 
 tistics give the respective quantities for 
 the year 1885 as 6,985,637 barrels, and 
 16,528,844 cases, a discrepancy which 
 may be due to differences in dates. On 
 the latter basis, the total gross weight of 
 the barrels would be 1,222,486 tons. If 
 piled six high, as they commonly are 
 when stored, the barrels would cover a 
 space of about half a square mile, and if 
 placed end to end would extend for a 
 distance of 3,638 miles. The tin-plate 
 used in the manufacture of the cases 
 amounts to nearly 38,502 tons, and would 
 cover more than five square miles and 
 431 acres. 
 
 The burning oil manufactured in Baku 
 during the year 1885, amounted to 27,- 
 000,000 poods, equal to 118,800,000 im- 
 perial gallons. Of this, about 17,000,000 
 poods were consumed in Russia, and 
 about 5,000,000 poocls, equal to 22,000,- 
 000 gallons, were exported, leaving a 
 balance of 5,000,000 poods in stock at 
 the end of the year. We have thus a 
 total of over 700,000,000 gallons of burn- 
 
117 
 
 ing oil manufactured per annum in the 
 United States and Kussia ; to this the 
 similar products manufactured at the re- 
 fineries on the continent of Europe and 
 in Scotland have to be added. 
 
 We have seen that petroleum consists 
 of a mixture of hydrocarbons, varying in 
 volatility, and that the line of demarka- 
 tion between petroleum spirit and petro- 
 leum oil is a purely arbitrary one, there 
 being but little difference between the 
 inflammability of the least volatile hydro- 
 carbon present in the spirit and the most 
 volatile in the oil on the one hand, and 
 between the most volatile hydrocarbon 
 present in the still heavier product and 
 the least volatile in the oil on the other 
 hand. The presence of too large a pro- 
 portion of the denser products prevents 
 the oil from burning freely ; while the 
 presence of an undue proportion of the 
 hydrocarbons of lower boiling point ren- 
 ders the oil unsafe for use in ordinary 
 mineral oil lamps. At an early period in 
 the development of the mineral oil in- 
 dustry, attention was accordingly directed 
 
118 
 
 to the necessity of drawing such lines of 
 demarkatioii as would insure the supply 
 to the consumer of an oil of satisfactory 
 character. The specific gravity of the 
 product was found to afford a sufficient 
 indication of the amount of the denser 
 hydrocarbons present, but not of the in- 
 flammability of the product, since a per- 
 centage of the more volatile hydrocar- 
 bons, too small to materially alter the 
 specific gravity of the oil, was found to 
 be sufficient to produce an inflammable 
 atmosphere in the oil reservoir of the 
 lamp. It was, therefore, obviously neces- 
 sary to apply a special test to determine 
 the inflammability of the liquid. The 
 earliest attempts in this direction took 
 the form of pouring the oil on to water 
 heated to a given temperature, passing a 
 light over the surface, and noting wheth- 
 er the oil evolved inflammable vapor or 
 itself ignited, the temperature at which 
 the oil first gave off enough vapor to 
 ignite being termed its " flashing point," 
 and that at which it caught fire its " fire 
 test." 
 
119 
 
 After some years of experience, it was 
 found that the testing of the oil in an 
 open cup was attended with certain dis- 
 advantages, and various forms of closed 
 cup were introduced. 
 
 The greater number of the petroleum 
 testing instruments employed at the pres- 
 ent time, and all those whose use is pre- 
 scribed by law may be divided into the 
 two classes referred to, viz., those which 
 have an open cup and those in which the 
 oil cup is provided with a cover, and the 
 use of instruments of the latter class is 
 largely on the increase. It has also been 
 proposed, however, to determine the in- 
 flammability of the oil by noting the ten- 
 sion of its vapor at a given temperature, 
 and various forms of apparatus have 
 been devised with this object, but since 
 there is no definite, or at any rate no 
 simple relation between vapor tension 
 and inflammability, this method of test- 
 ing has not found favor. 
 
 Of the forms of open cup tester, that 
 which is known as Tagliabue's has been 
 very largely used for many years. The 
 
120 
 
 apparatus consists of a glass cup contain- 
 ing the oil, placed in a water bath heated 
 by means of a small spirit lamp. A ther- 
 mometer is suspended in the oil, and the 
 temperature noted at which on passing 
 a burning splinter of wood across the 
 surface of the oil either a flash of ignited 
 vapor is obtained, or the oil itself takes 
 fire. The ignition of the oil is always pre- 
 ceded by a flash, but the number of 
 degrees of temperature through which 
 the oil must be raised between its flash- 
 ing and igniting varies according to the 
 character of the oil. 
 
 The English Petroleum Act, passed on 
 the 29th July, 1862, provided that " Pe- 
 troleum for the purposes of this Act shall 
 include any product thereof that gives off 
 an inflammable vapor at a temperature 
 of less than 100 of Fahrenheit's ther- 
 mometer," but as the method of testing 
 was not described, the Act was practically 
 inoperative. On the 13th July, 1868, an 
 amending Act was passed, defining " pe- 
 troleum " for the purposes of the Acts as 
 including "all such rock oil, Ban goon 
 
121 
 
 oil, Burmah oil, any product of them, 
 and any oil made from petroleum, coal, 
 schist shale, peat, or other bituminous 
 substance, and any product of them, as 
 gives off an inflammable vapor at a tem- 
 perature of less than 100 of Fahrenheit's 
 thermometer." Appended to the Act was 
 a schedule prescribing the form of appa- 
 ratus and method of testing to be adopted. 
 This apparatus consists of a slightly conical 
 oil cup of thin sheet iron, provided with a 
 flat rim, and a raised edge, -inch high. 
 Across the cup, and fixed to the edge, is 
 a wire, which is thus J-inch above the flat 
 rim. The oil cup is supported by the 
 rim in a tin water bath. The water 
 bath having been filled "with cold, or 
 nearly cold," water, the oil cup, support- 
 ed as described, was filled with the oil to 
 be tested, care being taken that the 
 liquid did not cover the flat rim. A 
 thermometer with a round bulb, and so 
 graduated that every 10F. occupied not 
 less than J inch on the scale, was then 
 placed in the oil so that the bulb was 
 immersed about 1 inches beneath the 
 
122 
 
 surface. A screen of pasteboard or wood 
 of specified dimensions was then placed 
 round the apparatus, and a "small flame" 
 applied to the bottom of the water bath. 
 When the temperature reached 90 F., a 
 " very small flame " was passed across 
 the surface of the oil on a level with the 
 wire, this application of the test flame 
 being repeated for every rise of " two or 
 three degrees " in temperature until a 
 " pale blue flicker or flash " was produced. 
 The temperature was then noted, and the 
 experiment repeated with a fresh sample 
 of the oil, withdrawing the source of heat 
 when the temperature approached that 
 noted in the first experiment, and apply- 
 ing the test flame at every rise of two 
 degrees. Various modifications of the 
 open vessel tester have been devised, 
 especially in the United States. 
 
 Arnaboldi's apparatus is similar to 
 Tagliabue's, but is of larger size, and in 
 one form has an adjustable mechanical 
 arrangement for applying the test flame 
 at a prescribed distance from the surface. 
 Messrs. Lockwood Brothers and Holly's 
 
123 
 
 apparatus is provided with an independ- 
 ent oil lamp, by means of which a test 
 flame can be moved across the testing 
 cnp at any required height above tbe oil 
 surface. In Mr. George M. Saybolt's 
 testing apparatus, which was a few years 
 ago adopted by the New York Produce 
 Exchange, the ignition of tbe vapor is 
 effected by means of an electric spark. 
 
 Tbe greater part of the earlier petro- 
 leum legislation in the United States was 
 based upon fire test and not upon flash- 
 ing point ; and the present rules of the 
 New York Produce Exchange recognize 
 no other test as the basis of commercial 
 transactions in petroleum oil, but in 
 many of the States the petroleum laws 
 now prescribe a test of flashing point. 
 
 One of the earliest forms of closed ves- 
 sel testers is that of Tagiiabue. This is 
 provided with a brass oil cup with a 
 cover, attached to which is a spring valve 
 and dwarf chimney. Tbe opening of this 
 valve and the simultaneous passing of a 
 small flame into the chimney through a 
 lateral orifice determines a current of air 
 
124 
 
 through the upper part of the oil cup, 
 which sweeps out the inflammable vapor 
 and brings it into contact with the flame. 
 In Michigan and Wisconsin the tester 
 has a copper oil cup with a copper- 
 cover, provided with a small orifice to 
 which the test flame is applied. The 
 present New York State tester is precise- 
 ly similar to that last described, except 
 that the oil cup is of large size and has a 
 glass instead of a copper cover to the oil 
 cup. The closed vessel tester employed 
 in Austria is similar in principle to Tag- 
 liabue's. Parrish's naphthometer used 
 in Holland is provided with a stationary 
 flame, fed by the oil in the testing cup. 
 The Foster tester is similar in principle. 
 In Millspaugh's tester the oil cup is of 
 glass, and is immersed only to the extent 
 of one- tenth of its depth in the water 
 bath, the object, apparently, being to 
 prevent the overheating of the surface of 
 the oil. Mann's tester represents an at- 
 tempt to reproduce in the testing appa- 
 ratus the conditions which prevail in an 
 ordinary petroleum lamp, the burner of 
 
125 
 
 the lamp being replaced by a tube, the 
 stopper of which is blown out, when, 
 upon the introduction of a flame through 
 a lateral opening, ignition of the vapor 
 occurs. In Pease's closed tester the 
 vapor is ignited by an electric spark. 
 
 Professor Arthur H. Elliott, of New 
 York, has recently made a large number 
 of comparative experiments with the 
 various instruments described, and has 
 embodied the results in a report to the 
 New York State Board of Health. 
 
 About the year 1870, a closed tester 
 with electric spark arrangement was in 
 use by the late Dr. Letheby in his* 
 laboratory at the London Hospital 
 The oil cup was of glass, and was pro- 
 vided with a hinged metal cover, which 
 was blown open when the vapor was ig- 
 nited by the spark. 
 
 Dr. Attfield has recommended that the 
 flashing point should be taken by warm- 
 ing the oil in a test-tube, and inserting a 
 flame into the mouth of the tube. 
 
 A Petroleum Bill, which sought to 
 amend the law in several very desir- 
 
126 
 
 able respects, was introduced into Par- 
 liament in 1871, but in consequence 
 mainly of the test standard being fixed 
 at a point (85 F.) which was higher 
 than the equivalent of the existing test 
 standard, the bill was opposed by the 
 petroleum trade, and the proposal to 
 change the method of testing withdrawn. 
 The bill passed on the llth of August in 
 that year, repealing the two previous acts 
 but prescribing the open test. In the 
 following year the subject of testing was 
 investigated by a select committee of the 
 House of Lords, and a great deal of evi- 
 dence was taken, but no satisfactory con- 
 clusion was arrived at. 
 
 At this period the position of the petro- 
 leum-testing question in England was 
 by no means promising. It had been 
 found that the existing legal directions 
 for testing were not sufficiently precise ; 
 the results obtained differing greatly, ac- 
 cording to the interpretation of the ex- 
 pression " small flame," as applied to the 
 source of heat ; and " very small flame," 
 as applied to the test flame ; the officials 
 
127 
 
 employed by the local authorities fre- 
 quently condemned oil that had been 
 passed by independent and unbiased ex- 
 perts acting on behalf of the traders ; 
 and retailers had thus no means of pro- 
 tecting themselves from the risk of being 
 fined for selling oil flashing below the 
 legal standard. Accordingly, with the 
 concurrence and approval of the Metro- 
 politan Board of Works, and of the 
 Petroleum Association, Sir Frederick Abel 
 was requested by the Government to un- 
 dertake the investigation of the subject 
 of petroleum testing, with the object of 
 devising a satisfactory test. The out- 
 come of Sir Frederick Abel's long and 
 painstaking experimental inquiry, in 
 which Dr. Kellner rendered valuable as- 
 sistance, was the adoption by Parliament, 
 on the llth of August, 1879, of what is now 
 so well known as the Abel test. The in- 
 strument and its use are thus described 
 in the Schedule of the 1879 Petroleum 
 Act: 
 
128 
 
 SPECIFICATION OF THE TEST APPARATUS. 
 
 The following is a description of the 
 details of the apparatus : The oil cup 
 consists of a cylindrical vessel 2 inches 
 in diameter, 2^ inches high (internal), 
 with outward projecting rim T 5 inch 
 wide, f inch from the top, and 1^ inches 
 from the bottom of the cup. It is made 
 of gun metal or brass (17 B. W. G.) 
 tinned inside. A bracket, consisting of a 
 short, stout piece of wire bent upwards, 
 and terminating in a point, is fixed to the 
 inside of the cup to serve as a gauge. 
 The distance of the point from the bot- 
 tom of the cup is 1^ inches. The cup is 
 provided with a close-fitting overlapping 
 cover made of brass (22 B. W. G.), which 
 carries the thermometer and test lamp. 
 The latter is suspended from two sup- 
 ports from the side, by means of trun- 
 nions upon which it may be made to os- 
 cillate ; it is provided with a spout, the 
 mouth of which is one-sixteenth of an 
 inch in diameter. The socket which is 
 to hold the thermometer is fixed at such 
 
129 
 
 an angle, and its length is so adjusted 
 that the bulb of the thermometer when 
 inserted to full depth shall be 1 inches 
 below the center of the lid. 
 
 The cover is provided with three 
 square holes, one in the center -fa inch 
 by -fa inch, and two smaller ones, fa 
 inch by -fa inch, close to the sides, 
 and opposite each other. These three 
 holes may be closed and uncovered by 
 means of a slide moving in grooves, and 
 having perforations corresponding to 
 those on the lid. 
 
 In moving the slide so as to uncover 
 the holes, the oscillating lamp is caught 
 by a pin fixed in the slide, and tilted in 
 such a way as to bring the end of the 
 spout just below the surface of the lid. 
 Upon the slide being pushed back so as 
 to cover the holes, the lamp returns to 
 its original position. 
 
 Upon the cover, in front of and in line 
 with the mouth of the lamp, is fixed a 
 white bead, the dimensions of which rep- 
 resent the size of the test flame to be 
 used. 
 
130 
 
 The bath or heated vessel consists of 
 two flat-bottomed copper cylinders . (24 
 B. W. G.), an inner one of 3 inches diame- 
 ter and 2J inches high, and an outer one 
 of 5^ inches diameter and 5f inches 
 high ; they are soldered to a circular 
 copper plate (20 B. W. G.) perforated in 
 the center, which forms the top of the 
 bath, in such a manner as to enclose the 
 space between the two cylinders, but 
 leaving access to the inner cylinder. The 
 top of the bath projects both outwards 
 and inwards about f of an inch, that is, 
 its diameter is about -| inch greater 
 than that of the body of the bath, 
 while the diameter of the circular open- 
 ing in the center is about the same 
 amount less than that of the inner copper 
 cylinder. To the inner projection of the 
 top is fastened, by six small screws, a flat 
 ring of ebonite, the screws being sunk 
 below the surface of the ebonite, to avoid 
 metallic contact between the bath and the 
 oil cup. The exact distance between the 
 sides and bottom of the bath and of the 
 oil ' cup is one half an inch. A split 
 
131 
 
 socket similar to that on the cover of the 
 oil cup, but set at a right angle, allows a 
 thermometer to be inserted into the 
 space between the two cylinders. The 
 bath is further provided with a funnel, an 
 overflow pipe and two loop handles. 
 
 The bath rests upon a cast-iron tripod 
 stand, to the ring of which is attached a 
 copper cylinder or jacket (24 B. W. G.) 
 flanged at the top and of such dimensions 
 that the bath, while firmly resting on 
 the iron ring, just touches with its pro- 
 jecting top the inward turned flange. 
 The diameter of this outer jacket is 6 
 inches. One of the three legs of the 
 stand serves as a support for the spirit 
 lamp attached to it by means of a small 
 swing bracket. The distance of the wick 
 holder from the bottom of the bath is 
 1 inch. 
 
 Two thermometers are provided with 
 the apparatus, the one for ascertaining the 
 temperature of the bath, the other for 
 determining the flashing point. The 
 thermometer for ascertaining the tem- 
 perature of the water has a long bulb 
 
132 
 
 and a|space at the top. Its range is from 
 about 90 to 190 Fahrenheit. The 
 scale (in degrees of Fahrenheit) is 
 marked *on an ivory back fastened to the 
 tube in the usual way. It is fitted with 
 a metal collar fitting the socket, and the 
 part of the tube below the scale should 
 have a length of about 3^ inches meas- 
 ured from the lower end of the scale 
 to the bulb. The thermometer for ascer- 
 taining the temperature of the oil is fitted 
 with collar and ivory scale in a similar 
 manner to the one described. It has a 
 round bulb, a space at the top, and 
 ranges from about 55 Fahr. to 150 
 Fahr. ; it measures from end of ivory 
 back to bulb 2 inches. 
 
 DIRECTIONS FOR APPLYING THE FLASHING 
 TEST. 
 
 1. The test apparatus is to be placed for 
 use in a position where it is not exposed 
 to currents of air or draughts. 
 
 2. The heating vessel or water bath is 
 filled by pouring water into the funnel 
 
133 
 
 until it begins to flow out at the spout of 
 the vessel. The temperature of the 
 water at the commencement of the test 
 is to be 130 Fahrenheit, and this is 
 attained in the first instance either by 
 mixing hot and cold water in the bath, or 
 in a vessel from which the bath is filled, 
 until the thermometer which is provided 
 for testing the temperature of the water 
 gives the proper indication ; or by heat- 
 ing the water with the spirit lamp (which 
 is attached to the stand of the apparatus) 
 until the required temperature is indi- 
 cated. 
 
 If the water has been heated too highly, 
 it is easily reduced to 130 by pouring in 
 cold water little by little (to replace a 
 portion of the warm water) until the 
 thermometer gives the proper reading. 
 
 When a test has been completed, this 
 water bath is again raised to 130 by 
 placing the lamp underneath, and the re- 
 sult is readily obtained while the petro- 
 leum cup is being emptied, cooled, and 
 refilled with a fresh sample to be tested. 
 The lamp is then turned on its swivel 
 
134 
 
 from under the apparatus, and the next 
 test is proceeded with. 
 
 3. The test lamp is prepared for use by 
 fitting- it with a piece of flat plaited 
 candle wick, and filling it with colza or 
 rape oil up to the lower edge of the 
 opening of the spout or wick tube. The 
 lamp is trimmed, so that when lighted it 
 gives a flame of about 0.15 of an inch 
 diameter, and this size of flame, which is 
 represented by the projecting white bead 
 on the cover of the oil cup, is readily 
 maintained by simple manipulation from 
 time to time with a small wire trimmer. 
 
 When gas is available, it may be con- 
 veniently used in place of the little oil 
 lamp, and for this purpose a test-flame 
 arrangement for use with gas may be 
 substituted for the lamp. 
 
 4. The bath having been raised to the 
 proper temperature, the oil to be tested 
 is introduced into the petroleum cup, 
 being poured in slowly until the level of 
 the liquid just reaches the point of the 
 gauge which is fixed in the cup. In 
 warm weather the temperature of the 
 
135 
 
 room in which the samples to be tested 
 have been kept should be observed in the 
 first instance, and if it exceeds 65, the 
 samples to be tested should be cooled 
 down (to about 60) by immersing the 
 bottles containing them in cold water, or 
 by any other convenient method. The 
 lid of the cup, with the slide closed, is 
 then put on, and the cup is placed into 
 the bath or heating vessel. The ther- 
 mometer in the lid of the cup has been 
 adjusted so as to have its bulb just im- 
 mersed in the liquid, and its position is 
 not under any circumstances to be altered. 
 When the cup has been placed in the 
 proper position, the scale of the ther- 
 mometer faces the operator. 
 
 5. The test lamp is then placed in 
 position upon the lid of the cup, the lead 
 line or pendulum,* which has been fixed 
 in a convenient position in front of the 
 operator, is set in motion, and the rise of 
 the thermometer in the petroleum cup is 
 
 * The pendulum should be 24 inches in length from 
 the point of suspension to the center of gravity of the 
 weight. 
 
136 
 
 watched. When the temperature has 
 reached about 66, the operation of test- 
 ing is to be commenced, the test flame 
 being applied once for every rise of one 
 degree, in the following manner : 
 
 The slide is slowly drawn open while 
 the pendulum performs three oscillations, 
 and is closed during the fourth oscilla- 
 tion. 
 
 NOTE. If it is desired to employ the 
 test apparatus to determine the flashing 
 points of oils of very low voLitibilifcy, the 
 mode of proceeding is to be modified as 
 follows : The air-chamber which sur- 
 rounds the cup is filled with cold water 
 to a depth of 1^- inches, and the heating 
 vessel or water bath is filled as usual, but 
 also with cold water. The lamp is then 
 placed under the apparatus and kept 
 there during the entire operation. If a 
 very heavy oil is being dealt with, the 
 operation may be commenced with water 
 previously heated to 120, instead of with 
 cold water. 
 
 By no means the easiest part of Sir 
 Frederick Abel's task was the determina- 
 
137 
 
 tion of the equivalent test standard, since 
 the Abel instrument furnishes no excep- 
 tion to the rule that the flashing point of 
 a given sample of mineral oil is far lower 
 in a closed than in an open vessel, and it 
 was therefore necessary to deal with the 
 conflicting views already referred to, as 
 to the proper mode of conducting the 
 test with the open cup instrument, in 
 order to determine the equivalent stand- 
 ard. Eventually, as the outcome of joint 
 experiments, it was ascertained that the 
 difference between the results afforded 
 by the open-cup instrument and the Abel 
 tester ranged from 25 to to 29 Fahr. 
 Taking the mean difference of 27, the 
 new standard was accordingly fixed at 
 73 Fahr. 
 
 There are yet some other systems of test- 
 ing to be described. In 1882, Braun, of 
 Berlin, patented a magnetic pendulum ar- 
 rangement for applying the test flame in 
 the Abel apparatus. In 1881, Engler and 
 Haas made a number of experiments 
 with the Abel apparatus and other testing 
 instruments, and expressed the opinion 
 
138 
 
 that the addition of an arrangement for 
 agitating the oil was desirable.* They 
 based their opinion apparently upon the 
 facts that in all the closed testers there 
 is during the operation a layer of vapor 
 of gradually increasing thickness formed 
 upon the surface of the oil ; that conse- 
 quently no two apparatus will give con- 
 cordant results unless the size and shape 
 of the oil cup, the height to which the 
 cup is filled, the rate of heating, the dis- 
 tance from the surface of the oil at which 
 the test-flame is applied, the size of the 
 test-flame, the dimensions of the orifices 
 in the cover, and other conditions are the 
 same ; and, further, that since the tem- 
 perature of the oil is not uniform through- 
 out, the position of the thermometer bulb 
 must be precisely defined. There is no 
 doubt that with the use of a mechanical 
 arrangement for agitating the oil and the 
 air in the cup, results less dependent 
 upon the conditions enumerated are ob- 
 tained, but since all the standard ineasure- 
 
 * Victor Meyer is stated to have been the first to 
 propose the addition of a stirrer. 
 
139 
 
 ments are most carefully adhered to, 
 and the position of the thermometer 
 bulb is prescribed, the objections to the 
 apparatus are not valid, and it is quite 
 possible that the use of a stirrer might in 
 practice be found to introduce some ele- 
 ment of error. Liebermann has recom- 
 mended the blowing of air through the oil 
 during the process of testing, and Beil- 
 stein devised an apparatus, based upon 
 this principle, which consisted of a glass 
 cylinder to hold the oil, with a tube pass- 
 ing to the bottom, and furnished with a 
 rose jet. At intervals, air was forced 
 through the jet at a rate sufficient to 
 raise a foam of a prescribed depth upon 
 the surface of the oil, and the test-flame 
 was at the same time applied. Stoddard 
 subsequently suggested a modification of 
 Beilstein's apparatus, in which the rose 
 jet was replaced by a glass tube drawn 
 out to a small orifice. 
 
 Bernstein's tester is constructed on the 
 principle of gradually heating the oil 
 until a temperature is reached at which, 
 on raising the level of the oil, vapor is 
 
140 
 
 forced out of the testing chamber, and 
 ignited at a stationary flame. Ehrenberg 
 has proposed to use a syringe to expel 
 the vapor from the closed testing cup, 
 and thus bring it into contact with a 
 flame. 
 
 Of apparatus in which the flashing 
 point of the oil is only determined in- 
 ferentially by noting the vapor tension, 
 the best known is the Salleron-Urban 
 instrument, which is used to some extent 
 in France. 
 
 In a table published in 1866, Salleron 
 and Urban give the following figures of 
 specific gravity and vapor tension of 
 petroleum products at 15 C. : 
 
 Density at 15 C. Tension in mm. of water. 
 
 .812 
 
 .797 5 
 
 .788 15 
 
 .772 40 
 
 .762 85 
 
 .756 125 
 
 .735 410 
 
 .695 930 
 
 .680 1,185 
 
 .650 2,110 
 
141 
 
 In the United States there is found'con- 
 siderable diversity of opinion among au- 
 thorities as to the method of testing and 
 the standard to be selected, and almost 
 all the States have adopted different 
 legislation in these principles. When 
 petroleum was first introduced in that 
 country as an illuminant, it was often 
 used carelessly and improperly, and many 
 accidents occurred. An exaggerated es- 
 timate of the risk involved in the use of 
 the oil was thus formed, and laws were 
 then passed, which, in some instances, 
 were so stringent that they could not be 
 enforced. Moreover, recent experiments 
 by Sir F. Abel and others, in Eng- 
 land, and by the officials of the Normal 
 Eichungs Kommission in Berlin, have 
 shown that an oil of high flashing point 
 may, in some cases, be actually less safe 
 than one of lower flashing point. The 
 experience of several years has demon- 
 strated that the proper use of an oil of 
 comparatively low flashing point (say 
 70 Abel test) is free from danger, and 
 there is reason to believe that authorities 
 
142 
 
 in the United States are beginning to 
 realize that their petroleum laws would 
 t>e rendered far more effective for the 
 protection of the public by the substitu- 
 tion of more moderate test require- 
 ments. 
 
 Petroleum legislation in this country 
 has hitherto been confined to those prod- 
 ucts which give off an inflammable vapor 
 below the legal limit. Exception has, 
 however, been taken to this by the 
 Metropolitan Board of Works, who con- 
 tend that between oil which flashes at 
 72, and oil which flashes at 73, there 
 can be no practical difference in regard 
 to inflammability. In advancing this 
 view, however, it appears to have been 
 forgotten that since the liquids which 
 flash below the standard can only be 
 stored and sold in comparatively small 
 quantities, and under troublesome re- 
 strictions, the practical effect of such 
 legislation is that elaborate precautions 
 are adopted to insure that all the burning 
 oil imported has a flashing point not 
 below the limit; the only liquid dealt 
 
143 
 
 with in accordance with the provisions of 
 the Acts being petroleum spirit (benzo- 
 line and gasoline). Placing of the sale of 
 burning oil (flashing not below 73 F.) 
 under legislative restrictions would be to 
 obliterate, to some extent, the distinction 
 which the present laws have created. 
 Moreover, there are probably no grounds 
 for supposing that petroleum oil consti- 
 tutes a dangerous article in the ordinary 
 stock of an oilman. On the contrary, there 
 is a good deal of evidence in support 
 of the opposite view. Thus, to take one 
 instance only, in the case of one of the 
 most recent fires at an oilman's shop 
 almost the only portion of the con- 
 tents of the shop which had escaped de- 
 struction were three barrels of petroleum 
 oil, constituting, according to the evi- 
 dence given, the whole stock of this 
 material on the premises at the time of 
 the fire. These barrels were a good deal 
 charred, but still held a considerable 
 quantity of oil. Petroleum is, in fact, a 
 far less dangerous liquid than is com- 
 monly supposed, as was pointed out 
 
144 
 
 some years ago, and again last year, by 
 Sir Frederick Abel in lectures at the 
 Koyal Institution. Statistics show that 
 the destruction of petroleum-laden ships 
 by fire is very rare, and at least one case 
 is on record where a vessel carrying 
 petroleum having been set on fire by 
 lightning, the fire was extinguished and 
 the cargo brought safely into port. 
 Many barrels discharged from the ves- 
 sel in question bore evidence of the 
 heat to which they had been subjected, 
 being in some cases so much charred that 
 a penknife blade could be driven through 
 the staves, and yet these barrels still held 
 the oil intact. 
 
 Inconsequence, partly, of the expressed 
 views of the Metropolitan. Board of 
 Works, a Petroleum Bill was introduced 
 by the Government in 1883 to place the 
 storage and sale of petroleum oil under 
 legal restrictions ; but the measure was 
 practically condemned by a Select Com- 
 mittee of the House of Lords, and was 
 withdrawn. Subsequently Colonel Ma- 
 j en die, her Majesty's Chief Inspector of 
 
145 
 
 Explosives, made a visit to Germany, Aus- 
 tria, France, Belgium, and Holland, for 
 the collection of information in regard to 
 petroleum legislation in those countries. 
 In a memorandum issued in 1884, Colo- 
 nel Majendie set forth the result of his 
 observations and inquiries, the informa- 
 tion he collected being supplemented by 
 particulars obtained from consular re- 
 ports and other sources. The result of 
 the inquiry showed that in all the princi- 
 pal countries of Europe the traffic in 
 mineral oil was to a greater or less extent 
 under the control of State or municipal 
 authorities, but in many respects the re- 
 strictions were far less stringent than 
 those proposed by the Bill of 1883. 
 There is no doubb much to be said in 
 favor of the contention that the storage 
 of mineral oils in large quantities should 
 take place only under specified condi- 
 tions, so that in the event of a fire occur- 
 ring, outflow of burning oil from the 
 premises could not occur, but it is diffi 
 cult to see how legislation to give effect 
 to this view could fairly and justifiably 
 
146 
 
 be advocated unless it took the shape 
 of a comprehensive measure embracing 
 other inflammable liquids. 
 
 We have now dealt with that which is 
 the most important of all the tests ap- 
 plied to petroleum products, since, as we 
 have seen, legislative enactments framed 
 in the interest of public safety are based 
 upon it. There are, however, many 
 commercial tests worthy of attention. 
 
 The color of petroleum oil is deter- 
 mined in this country (as regards oil 
 for export), in England, in Germany, and 
 in Russia (in the case of oil for export), 
 by the use of a chromometer patented by 
 Mr. B. P. Wilson, a member of the com- 
 mittee of the Petroleum Association of 
 London, in ]870. This instrument is 
 fitted with two parallel tubes furnished 
 with glass caps, and at the lower end of 
 the tubes is a small mirror, by means of 
 which light can be reflected upwards 
 through the tubes into an eye- piece. 
 One of the tubes is completely filled with 
 the oil to be tested, and beneath the 
 other tube, which remains empty, is 
 
147 
 
 placed a disk of stained glass of standard 
 color. On adjusting the mirror and 
 looking into the eye-piece, the circular 
 field is seen to be divided down the 
 center, each half being colored to an ex- 
 tent corresponding with the tint of the 
 oil and of the glass standard respectively. 
 An accurate comparison of the two colors 
 can thus be made. The glass disks, 
 which for the English trade are of five 
 shades of color, termed good merchant- 
 able, standard white, prime white, super- 
 fine white, and water white, are issued 
 by the Petroleum Association of London, 
 and the instruments are all precisely sim- 
 ilar in construction ; thus the testing of 
 color, wherever these chromometers are 
 used, is placed upon a uniform basis. 
 
 A German modification of the instru- 
 ment (devised by Stammer), is provided 
 with an arrangement for shortening the 
 column of oil, and thus obtaining an 
 exact match with the standard. The ex- 
 tent of shortening being indicated on the 
 scale, the color of the oil can be recorded 
 
148 
 
 in terms of the standard with considera- 
 ble precision. 
 
 Both fire test and flashing point of 
 lubricating oils are determined by testing 
 the oil in an open cup. There is no 
 generally, or at any rate universally, 
 accepted method of applying these tests, 
 and important discrepancies between the 
 results of different operators frequently 
 occur. These discrepancies chiefly arise 
 from variations in the rate of heatiDg the 
 oil. Pen sky, of Berlin, constructs a 
 closed cup apparatus for testing lubrica- 
 ting oils, which in principle resembles the 
 Abel instrument. As, however, the trade 
 are accustomed to judge of the volatility of 
 lubricating oils by their open vessel flash- 
 ing points or their fire tests, the closed cup 
 cannot usually be employed. The method 
 of heating the oil adopted by Pensky, which 
 consists in placing the oil cup in an air 
 chamber in a cast-iron vessel which can 
 be strongly heated, is a convenient one, 
 and on using the Pensky apparatus 
 without the oil-cup cover, and adjust- 
 ing the Bunsen burner flame, which is 
 
149 
 
 used as a source of beat, so that the 
 temperature of the oil is raised at the 
 rate of about 10 per minute, there are 
 obtained concordant flashing points 
 and fire tests, which agree with 
 those furnished by the method of 
 testing usually adopted in America. 
 A gas flame not more than a quarter 
 of an inch in diameter, produced by the 
 use of a small jet, is a good test flame 
 to use. 
 
 The so-called "cold test "of lubrica- 
 ting oils is the temperature at which the 
 oils either become cloudy or cease to flow 
 through crystallization of parafifine. The 
 test is usually applied by slowly cooling 
 a sample of the oil in a tube about If 
 inches in diameter, and noting the tem- 
 perature at which on inclining the tube 
 the oil no longer flows, or that at which 
 separation of paraffine commences. 
 
 The viscosity of mineral lubricating 
 oils is a feature which is intimately asso- 
 ciated with their lubricating properties. 
 The subject has been dealt with somewhat 
 fully in a paper read at a meeting of the 
 
150 
 
 Society of Chemical Industry by the 
 author,* and the apparatus commonly em- 
 ployed in testing consists of an arrange- 
 ment for noting the length of time occu- 
 pied by a given quantity of the oil in 
 flowing through a small orifice of given 
 dimensions and form at a given temper- 
 ature. 
 
 Paraffine scale, or partially refined 
 paraffine wax, usually contains oil, and 
 sometimes water. The percentage of oil 
 is determined by subjecting a weighed 
 quantity of the material to a given 
 pressure at a given temperature for 
 a specified time, and noting the loss 
 in weight. The author published in the 
 Journal of the Society of Chemical In- 
 dustry, in August, 1884, the particulars 
 of the method of testing which he 
 employs, and gave comparative results 
 obtained under various pressures, with 
 various lengths of exposure to the press- 
 ure at various temperatures, and with 
 various quantities of material. The experi- 
 
 * Published in the Journal of the Society of Chemical 
 Industry, on the 29th of March, 1886. 
 
151 
 
 ments showed that differences of tempera- 
 ture affect the results far more thai] differ- 
 ences in pressure and in the quantity of 
 material do. This is not surprising when 
 we consider that there is no actual line 
 of demarkation between the solid and 
 liquid hydrocarbons obtained from petro- 
 leum. It is, however, important that the 
 same pressure should always be employ- 
 ed, and accordingly, some years ago 
 a press was constructed which indicates 
 the amount of pressure by the extent of 
 deflection (magnified by levers) of the 
 steel cross head. The test is made at 
 60 F., the temperature of the press 
 plates being indicated by thermometers 
 placed in mercury cups ; the quantity of 
 material employed is 500 grains, the 
 pressure is 9 tons over the whole surface 
 of the circular press cake, 5f inches in 
 ' diameter, and this pressure is maintained 
 for five minutes, the oil expressed being 
 absorbed by blotting paper. The author 
 recently had a press constructed in the 
 United States for use in testing, which is 
 furnished with a long, heavily- weighted 
 
152 
 
 lever in place of the spring cross-head. 
 With this press similarly concordant 
 results were obtained, and the ob- 
 jection to a spring, that it may become 
 weaker through use, does not, of course, 
 attach to the weighted lever method of 
 indicating pressure. Mr. McCutchon, of 
 Young's Paraffine Oil Company, con- 
 structed a testing press in which coiled 
 steel springs were substituted for the 
 elastic cross-head, and this press has been 
 adopted by the Scottish Mineral Oil 
 Association. Mr. William Walls, of 
 Glasgow, recommends the use of a 
 hydraulic press, and Messrs. Clarkson 
 &Beckitt, of Glasgow, a few months ago, 
 constructed a well-made press of this 
 nature. A hydraulic press is portable, 
 occupies less space than a screw press, 
 and admits of a specific pressure being 
 applied with ease and certainty. This 
 press acted promptly even after being 
 out of use many clays. Under a great 
 pressure the scale is liable in some cases 
 to be squeezed out of the cloth, and the 
 quantity of oil expressed is not mate- 
 
153 
 
 rially increased. Thus, in some compara- 
 tive experiments conducted at various 
 pressures, all the other conditions being 
 alike, the author obtained the following 
 results : 
 
 PRESSURE ON WHOLE CAKE. 
 
 9 tons 13.3 tons 20 tons 
 
 Sample. oil. oil. oil. 
 
 A. . . 2.9 per cent. . 3 1 per cent, .not taken. 
 B...14.8 " ..14.4* " ..not taken. 
 C... 2.5 " ..not taken. . .2.9 per cent. 
 D. .. 5.5 " ..not taken. . .6.1 
 
 Water in paraffine scale is usually de- 
 termined by heating a weighed sample to 
 a temperature somewhat above the boil- 
 ing point of water, and maintaining it at 
 that temperature until it ceases to lose 
 weight. 
 
 The so-called <fc melting point " of 
 paraffine is, in the case of the recognized 
 American and English methods of mak- 
 ing the test, the temperature at which 
 the sample, after having been melted, 
 and while in the process of cooling, 
 
 * Temperature slightly lower. 
 
154 
 
 begins to solidify. The American test is 
 conducted by melting sufficient of the 
 samples to three parts to fill a hemispheri- 
 cal dish 3J in. in diameter. A thermome- 
 ter with a round bulb is suspended in 
 the fluid so that the bulb is only three- 
 fourths immersed, and the material being 
 allowed to cool slowly, the temperature 
 is noted at which the first indication of 
 filming extending from the sides of the 
 vessel to the thermometer bulb occurs. 
 The English test is made by melting the 
 sample in a test tube, about three-quar- 
 ters of an inch in diameter, and stirring 
 it with a thermometer as it cools, until a 
 temperature is reached at which the 
 crystallization of the material produces 
 enough heat to arrest the cooling, and 
 the mercury remains stationary for a 
 short time. The results afforded by this 
 test are usually from 2 to 3 F. lower 
 than those furnished by the American 
 test. The melting point is also some- 
 times determined by observing the tem- 
 perature at which a minute quantity of 
 the sample, previously fused into a capil- 
 
155 
 
 lary tube and allowed to set, becomes 
 transparent when the tube is slowly 
 warmed in a beaker of water. 
 
 CHAPTER IV. 
 
 By far the most important of the uses 
 to which the products of petroleum are 
 as yet applied is that of illumination, 
 and it is to this use that we should ac- 
 cordingly direct our attention. In re- 
 gard to the employment of the solid pro- 
 ducts of petroleum in the form of par- 
 affine candles not much need be said. 
 Liebig in 1851 expressed the opinion that 
 "it would certainly be esteemed one of 
 the greatest discoveries of the age if any 
 one could succeed in condensing coal-gas 
 into a white, dry, solid, odorless sub- 
 stance, portable, and capable of bting 
 placed upon a candlestick or burned in a 
 lamp," and it may fairly be said that in the 
 paraffine candle we have a source of light 
 possessing all the characteristics which 
 
156 
 
 Liebig considered so desirable. Price's 
 Patent Candle Company and Messrs. J. C. 
 and J. Field are well known to have been 
 conspicuously successful in their efforts 
 to improve the manufacture of paraffine 
 candles, the introduction of the " self- 
 fitting " end, and the adoption of various 
 devices to render the candle more orna- 
 mental. 
 
 The principal defect possessed by par- 
 affine candles is a liability to bend when 
 exposed to the air of a warm room. This 
 tendency is to some extent overcome by 
 mixing the paraffine with a small percent- 
 age of palmitic or stearic acid, and it is 
 found that candles made of this mixture 
 are also less liable to smoke than those 
 made from pure paraffine. Ozokerite 
 candles have a higher melting point than 
 those which are made of petroleum or 
 shale paraffine, and are in this respect 
 better suited for use in warm climates. 
 
 Young's Paraffine Light and Mineral 
 Oil Company have made an ingenious use 
 of soft paraffine in the handy little miners' 
 lamp. 
 
157 
 
 There probably is no exhaustive ac- 
 count of the various improvements which, 
 from time to time, have been effected in 
 appliances for burning mineral oils with 
 a view to the production of light ; and 
 in the attempt to compile a reccord hav- 
 ing any pretensions to completeness, great 
 difficulties may be anticipated. 
 
 We know, from the accounts given by 
 historians who wrote prior to the com- 
 mencement of the Christian era, that 
 petroleum was used as an illuminant in 
 remote ages. The appliances for burn- 
 ing the oil, were, without doubt, of a 
 very primitive description, consisting 
 simply of an oil vessel furnished with 
 a wick of some fibrous material. Lamps 
 of this character are still used for burn- 
 ing crude petroleum in various localities. 
 The earthen- ware petroleum lamp of some- 
 what artistic form, but very primitive con- 
 struction, is found in use in provincial 
 Russia, arid little glass and tin lamps are 
 met with in the bazaars in India. 
 
 For the lighting of the derrick during 
 the operation of well-drilling in the 
 
158 
 
 United States, the large iron two-spout- 
 ed kettle-shaped lamp, fed with crude 
 petroleum, is still commonly employed. 
 With none of these lamps is a chimney 
 used, and the flame produced is dull and 
 smoky. 
 
 Much interesting and valuable infor- 
 mation in reference to the subject of 
 mineral oil lamps was given by Sir Fred- 
 erick Abel in a lecture delivered at the 
 Royal Institution last year. The lec- 
 turer expressed the belief that min- 
 eral oil lamps were first constructed 
 in Germany, about the years 1852-3, 
 and somewhat extensive inquiries made 
 since quite confirm this view. Miny 
 years previously, however, viz., in 1820, 
 coal-naphtha, produced under Lord Dun- 
 donald's patent of 1781, was introduced 
 as an illuminant by Mr. Astley, for use in 
 the so-called founder's blast lamp, in 
 which the combustion was aided by a 
 current of air artificially produced, Read 
 Holliday was also among the first to sug- 
 gest the employment of coal naphtha for 
 illuminating purposes, and the vapor 
 
159 
 
 lamp which he patented is still largely 
 employed. 
 
 Other inventors designed lamps for 
 vaporizing coal-naphtha, and burning the 
 vapor at a jet. Various ingenious 
 arrangements for maintaining a uniform 
 oil level were also devised during the lat- 
 ter part of the 18th century. These 
 arrangements, many of which were des- 
 cribed by Mr. Leopold Field in his Can- 
 tor lectures on " Solid and Liquid Illumi- 
 nating Agents," delivered in 1883, were 
 applied to lamps burning fixed oils ; but 
 some are of special interest to us as being 
 applicable to mineral oil lamps. Thus 
 the fc * bird-fountain '' system of maintain- 
 ing the oil level, which is stated to have 
 been first adopted by Miles in 1781, is 
 applied in one class of petroleum lamps 
 now in use. This arrangement consists 
 of an oil reservoir closed at the top, and 
 communicating by a tube with the wick 
 case of the burner, which is at a lower 
 level than the top of the reservoir. As 
 the oil in the wick case is drawn up to 
 the flame and consumed, a few bubbles 
 
160 
 
 of air pass into the reservoir, and a cor- 
 responding quantity of oil flows out. The 
 principle of pumping the oil np to the 
 flame, which was devised by Carcel in 
 1708, is also at present in use in burning 
 mineral oils, chiefly for lighthouse illu- 
 mination, but also, to a small extent, for 
 domestic lighting. 
 
 The first lamps in which mineral oils 
 were burned in this country with any 
 degree of success were those which had 
 been designed some years previously for 
 use with rectified oil of turpentine, 
 known as "camphine." These lamps 
 were constructed upon the principles de- 
 scribed by Ami Argand in the specifica- 
 tion of the patent granted to him in 1784, 
 and, in view of its great value, it wiy. be 
 interesting to consider the nature of Ar- 
 gand's invention. The specification 
 states that the patent is for " a lamp that 
 is so constructed as to produce neither 
 smoak nor smell, and to give consider- 
 ably more light than any lamp hitherto 
 known ; " and the method by which these 
 results are obtained is thus described : 
 
161 
 
 " My method of giving light by lamps 
 or any other illuminating instruments or 
 things, consists in wholly converting into 
 flame and light any inflammable and com- 
 bustible matter whatever that may be 
 used as the pabulum, fuel, or subject of 
 such light, which in the common mode of 
 producing light is only in part burnt, 
 consumed, and converted into flame, the 
 rest going off in smoak or soot. This 
 is effected in its various modes of appli- 
 cation, first, by causing a current of air to 
 pass through the inside of the flame ; sec- 
 ondly, by increasing that current of air 
 and producing another current of air on 
 the outside of the flame by means of a 
 chimney, dome, funnel, tube, pipe, or 
 other contrivance, covering, surrounding 
 or enclosing the flame, or the vehicle, or 
 the receptacle or vessel containing it, so 
 that the fresh air which continually rushes 
 into the said chimney, dome, funnel, tube, 
 or pipe, or passes through it to supply 
 the place of that which is decomposed 
 and rarefied by fire or flame, and which, 
 growing lighter, escapes continually 
 
162 
 
 through the top of the said chimney, 
 dome, funnel, tube or pipe, shall have free 
 access and circulation ; or thirdly, by 
 applying such chimney, dome, funnel, 
 tube, or pipe to common wicks, where 
 the full effects of the principle are not re- 
 quired.* < This effect of increasing the 
 light, by converting the smoakinto flame, 
 is obtained by the means above mentioned, 
 which may be -used either separately, or, 
 where a greater advantage is required, the 
 methods above described may be com- 
 bined. The vehicles which contain the 
 light are made of various forms and 
 various materials.'' 
 
 The Argand lamp, to quote a descrip- 
 tion of it published many years ago, 
 " embraces so many improvements upon 
 the common lamp, and has become so 
 general throughout Europe, that it may be 
 justly ranked amongst the greatest dis- 
 coveries of the age. As a substitute for the 
 candle, it has the advantage of great 
 economy and convenience, with much 
 
 * It is stated that the use of a chimney had been 
 previously suggested by Quinquet. 
 
163 
 
 greater brilliancy ; and for the purpose 
 of producing heat it is an important in- 
 strument in the hands of a chemist. We 
 may, with some propriety, compare the 
 common lamp and the candle to fire 
 made in the open air, without any forced 
 method of supplying it with oxygen ; 
 while the Argand lamp may be compared 
 to a fire in a furnace, in which a rapid 
 supply of oxygen is furnished by the 
 velocity of the ascending current. This, 
 however, is not the only advantage of this 
 valuable invention. It is obvious that .if 
 the combustible vapor occupies a consid- 
 erable area, the oxygen of the atmos- 
 phere cannot combine with the vapor in 
 the middle part of the ascending 
 column. The outside, therefore, is 
 the only part which enters into 
 combustion ; the middle constituting 
 smoke. This evil is obviated in the Ar- 
 gand lamp, by directing a current of 
 atmospheric air through the flame, which, 
 instead of being raised from a solid wick, 
 is produced by a circular one, which sur- 
 rounds the tube through which the air 
 
164 
 
 ascends." Argand's original lamp had an 
 iron chimney, but a cylinder of glass was 
 soon substituted. Smethurst, or Lange, 
 afterwards proposed the employment of 
 a chimney with a shoulder or constriction 
 in place of the simple cylinder. 
 
 In ]818 Sir Thomas Cochrane obtained 
 a patent " for lamps in streets, which ef- 
 fectuate and regulate the combustion of 
 a certain purified essential oil or spirit 
 obtained from different ligneous, carbon- 
 aceous, or bituminous substances usually 
 called spirit of tar, or oil of tar." The 
 principal feature of the invention seems 
 to have been the construction of the 
 " burner or wick- holder of a length pro- 
 portioned to the volatility of the said 
 essential oil or spirit compared to that of 
 the viscous oil of whales, or of other 
 gross and sluggish oils now in use." In 
 1822 A. and D. Gordon patented the 
 construction of lamps with wicks made 
 of platinum, gold, silver, or copper drawn 
 into fine threads, or of glass drawn into 
 capillary tubes, bound into a compact 
 bundle ; also the use with such wicks of 
 
165 
 
 a burning fluid composed of spirit of 
 wine or wood spirit mixed with essential 
 oils. In 1839 Goldsworthy Gurney and 
 Frederick Rixon obtained a patent for 
 the 4 *oleo-oxygen or Bude light, produced 
 by administering to the flame of oil or gas 
 lamps or burners a jet or stream of pure 
 oxygen." In 1840 George Halpin, Jr., 
 patented the application of a mechani- 
 cally produced air blast to an Argand 
 burner, and described the use of " an 
 additional tube placed within that usually 
 forming the inner casing of an Arg.tnd 
 lamp." The drawings accompanying the 
 specification also show the construction 
 of a burner with three concentric wicks. 
 In the same year Thomas Young, of 
 Queen Street, London, obtained protec- 
 tion for improved methods of supplying 
 the oil to the burner, and for the use of 
 a perforated plate " at a position above 
 the point of combustion of the wick of 
 lamps, and thereby obtaining a more 
 favorable application of air to the flame 
 of lamps." A year later a patent was 
 granted to William Newton for " lamps, 
 
166 
 
 burners, and apparatus for the produc- 
 tion of light and heal for domestic and 
 other useful and ornamental purposes, 
 partly from certain substances which 
 have not yet been brought into general 
 use for such purposes and by such means, 
 some of the said substances being the 
 hydrocarbons usually called coal-tar, oils 
 or liquids, naphtha, vegetable tar, oils or 
 liquids, and the liquids obtained by dis- 
 tillation and rectification of the resins, 
 schistus, petroleum, maltha or mineral 
 pitch, mineral naphtha, asphaltum, bitu- 
 men, caoutchouc, animal oil and the 
 various hydrocarbonaceous substances 
 that may be extracted from vegetables, 
 grain, plants, and trees, arid most of the 
 inflammable oleaginous or resinous sub- 
 stances that are both to be vaporized 
 at certain temperatures, according as 
 they may vary in nature and specific 
 gravity." The burners in respect 
 of which the use of this remarkably 
 comprehensive list of burning fluids 
 was claimed, were provided with 
 various complicated arrangements for 
 
167 
 
 supplying the fluid to the wick, and in 
 some cases for causing the vaporization 
 of the fluid before its ignition. In the 
 same year (1841) William Young filed a 
 lengthy specification of improvements in 
 lamps, part of which related to arrange- 
 ments for burning " naphtha or turpen- 
 tine spirit." William Young's lamp, 
 which was of the Argand type, was at first 
 used with rectified oil of turpentine, 
 known as camphine, but on the intro- 
 duction of coal oil as an illuminating 
 agent, Young's " Vesta " lamps were 
 used to burn it. In one form of this 
 lamp two, three, or more flat wicks were 
 so held in curved wick holders as to pro- 
 duce a tubular flame, and air was admit- 
 ted to the inner surface of the flame be- 
 tween the wick tubes. A collar of wood 
 was inserted between the burner and the 
 oil reservoir, to prevent the conduction of 
 heat to the latter. 
 
 In 1842 George Eoberts, a miner, was 
 granted a patent, part of which related 
 to improvements in lamps for burning 
 " naphtha, turpentine, or such spirits as 
 
168 
 
 are usually burned 'n lamps." The lamp 
 described in the specification as suitable 
 for use with naphtha or other spirits was 
 of the Argand form, the essential feature 
 of the invention b^ing the use of as many 
 as four perforated disk air deflectors, so 
 arranged as to " deflect air on to the wick, 
 and also on to and all around the flame 
 after it is formed, such deflection of the air 
 being at intervals apart, consequently the 
 air will be deflected on to the flame at 
 different heights, the air being prevented 
 passing into the upper part of the chim- 
 ney except with the flame.'' The patentee 
 states that " by combining the use of the 
 four disks, as above described, a lamp, such 
 as above described, may be made suitable 
 for burning naphtha and other spirits, 
 without smoke, producing a very power- 
 ful light.'' This early recognition of the 
 important principle of distributing the air 
 supply is of considerable interest. Three 
 years later, viz., in 1845, the same invent- 
 or obtained protection for, among other 
 improvements in the construction of 
 lamps, the use, in the center of an Ar- 
 
169 
 
 gand wick tube, of " a flat disk of metal, 
 to the outer edge of which a curved piece 
 of perforated metal is attached, thus 
 forming a cap through which the air will 
 pass in small streams to the flame ;" also 
 for the similar use of a trumpet-mouthed 
 deflector receiving air through a hollow 
 rod, the top of the deflector being closed 
 with a flat disk, and the sides perforated. 
 Air diffusers of similar construction form 
 important features of mineral oil lamps 
 patented later. Mr. Roberts also specified 
 the use of wicks of cane or porous wood, 
 or of asbestos. Samuel King's patent of 
 1856 relates chiefly to arrangements for 
 providing in a suitable manner a supply 
 of air to the flame of an Argand burner. 
 An additional patent for further improve- 
 ments in the arrangements for the distri- 
 bution of the air was taken out by King in 
 1859. A patent for a chimney less burn- 
 er (flat wick) was granted to George 
 Young in 1867. 
 
 N. J. Fenner is authority for the state- 
 ment that a distiller of the name of Baker, 
 carrying on business on Bow Common, 
 
170 
 
 was probably the first to introduce recti- 
 fied oil of turpentine as an illuminating 
 agent under the name of camphine. This 
 was somewhere about the year 1835. The 
 manufacture of camphine was subsequent- 
 ly, according to Mr. Fenner, carried on 
 by Messrs. Jupp, C. Price and Co., P. 
 Murphy, Flockton, Garton, and Fenner, 
 all of whom were turpentine and resin 
 distillers, the wet steam process of recti- 
 fication being generally employed. In 
 consequence of the high price of colza 
 and fish oils, camphine had at this time 
 a large sale as an illuminant, notwith- 
 standing its liability to fill the room with 
 " blacks " and its many other defects. In 
 1856 Mr. Chappel, a solicitor, consulted 
 Messrs. Fenner as to commercially utiliz- 
 ing an asphaltum imported from Cuba, 
 and experiments having shown that the 
 material yielded 75 per cent, of crude oil 
 from which good burning and lubricating 
 oils could be made, a company was formed 
 and the process carried on successfully 
 for some time, the oil selling at 3s. 7d. to 
 3s. 8d. per gallon. Before long, how- 
 
171 
 
 ever, the introduction of the cheaper 
 American petroleum rendered the indus- 
 try unprofitable. 
 
 The products at first manufactured by 
 James Young were not well adapted for 
 illuminating purposes, attention having 
 been concentrated upon the manufacture 
 of lubricating oils ; but in 1853 the in- 
 creasing consumption of the burning oil, 
 which, for three or four years previously 
 had been manufactured in Hamburg, by 
 Noblee, led Mr. Young to make inquiries 
 which resulted in his finding that suit- 
 able burners were being constructed by 
 Mr. C. H. Stobwasser, of Berlin. Mr. 
 Young then obtained some of the burners 
 in question, and entrusted to Messrs. R. 
 Laidlaw and Son, of Edinburgh, the man- 
 ufacture of similar appliances. The 
 earliest of these is on the Argand princi- 
 ple, with an annular wick case, and is 
 called a "solar-oil lamp," while those made 
 subsequently for burning petroleum have 
 flat wick tubes. 
 
 Stobwasser & Co. first made the solar-oil 
 (Argand) burners, and the flat wick burn- 
 
172 
 
 ers about 1852 or 1853. The solar-oil burn- 
 er was constructed for use with the liquid 
 termed "pbotogene/' made by Noblee 
 from coal, while the flat wick burner was 
 used with the oil distilled from lignite. 
 The solar-oil at first manufactured would 
 not burn in the flat wick burners, as 
 originally constructed, but a slight alter- 
 ation of the burner, and an improvement 
 in the quality of the oil, overcame the 
 difficulty. 
 
 Gesner states that the first lamps suit- 
 able for burning kerosene that were em- 
 ployed in America, were imported from 
 Vienna by Mr. J. H. Austen, who, about 
 the year 1854, was engaged in the coal 
 oil industry. According to Dr. Schweit- 
 zer, Professor of Chemistry in the Uni- 
 versity of Missouri, the coal oil manu- 
 factured about this time was of such com- 
 paratively high density that it could 
 be burned in ordinary lamps. The sub- 
 sequent development of the petroleum in- 
 dustry, however, rendered it very neces- 
 sary to devise special lamps. The first 
 patent taken out for a petroleum lamp, 
 
173 
 
 so-called, in the United States, was dated 
 1859, and in that year forty applications 
 for patents for petroleum lamps, burners 
 and appliances in general were granted. 
 In the succeeding year the number of 
 similar grants was 71, in the next year 
 53, and in the one after 101. From that 
 time to 1878 the number ranged from 63 
 to 186 ; and judging by the number of 
 new kinds of lamps which are offered for 
 sale every year in America, the energy of 
 the American inventor shows no signs of 
 diminution. 
 
 Meanwhile, in England, the improve- 
 ment of appliances for burning mineral 
 oils was steadily progressing. Messrs. 
 Joseph Hinks & Son, of Birmingham, 
 who had commenced business as lamp 
 manufacturers about the year 1860, 
 patented October 28, 1865, the well- 
 known duplex burner,* the introduction 
 of which has done so much to popularize 
 the use of mineral oils for illuminating 
 purposes. The patent which was granted 
 
 * Dr. Angus Smith had previously constructed a 
 two wick burner. 
 
174 
 
 to James and Joseph Hinks was described 
 as being for " the improvements herein- 
 after described in the burners of lamps 
 for burning paraffine oil and other vola- 
 tile liquid hydrocarbons, whereby two or 
 more flat flames, or one circular, or nearly 
 circular flame, may be produced by the use 
 of two or more single flat wicks." The 
 general construction of the burner was 
 thus described : ' fc Our invention consists 
 in the employment in the same burner of 
 two or more flat or curved wick cases or 
 holders, in which said cases or holders 
 single flat wicks are placed. Each of the 
 said wick cases is provided with an axis 
 and pinions for raising and lowering the 
 wick therein contained. The wick cases 
 or holders are either straight or slightly 
 curved, or of the figure of semi-ellipses, 
 or semi-circular, so as to produce, when 
 arranged in the burner, flat, or elliptical, 
 or circular flames. The cone or deflector 
 has two or more straight or curved open- 
 ings in it, through which the wicks may 
 pass. That portion of the top of 
 the cone between the curved open- 
 
175 
 
 ings (when curved openings and curved 
 wick holders are employed) serves as a 
 substitute for the ordinary button used 
 with circular wicks ; or a circular hole 
 may be used in the cone, and the ordi- 
 nary button employed. When flat wick 
 cases are employed, straight openings are 
 made in the cone." In the original form 
 of the flat-wick burner the two wicks 
 were brought together into one wick 
 tube in the lower part of the burner. 
 Subsequent improvements effected by 
 this firm have resulted in the production 
 of a burner, the good features of which 
 are too well known to need enumeration ; 
 and their conspicuous merit, from an 
 artistic as well as mechanical point of view, 
 is apparent. In 1871, Holver Holver- 
 stone, of Cambridge, Mass., patented a 
 dual burner, having two flat wicks in 
 a cone furnished with a single slot. Lit- 
 igation between this inventor and Messrs. 
 Hinks resulted in an amicable arrange- 
 ment. 
 
 On January 22d, 1886, Messrs. Thomas 
 Rowatt & Sons patented the "anucapnic" 
 
176 
 
 burner, which differs from those already 
 referred to in being provided with a 
 double cone. The use of this appliance 
 admits of the ordinary chimney being re- 
 placed by a globe, without any diminu- 
 tion in the brilliancy of the light. The 
 provisional specification, filed by Thomas 
 Kowatt, the younger, states that his 
 "improvements in lamps for burning 
 paraffine, petroleum, belmontine, and 
 other hydrocarbon oils without the use 
 of a chimney," consists in obtaining "the 
 necessary current of air by carrying down 
 a tube of metal or any other suitable 
 material from the dome to within a short 
 distance of the bottom of the burner, ex- 
 tending its lower rim laterally, so that 
 the opening remaining between the edge 
 of this rim and the lower body of the 
 burner becomes an air channel with a 
 much greater superficial area than the 
 area of the bottom neck of the tube, con- 
 sequently the tube becomes an air- sluice 
 when the lamp is lighted/' The specifi- 
 cation further states that in order that 
 complete combustion might be effected, 
 
177 
 
 it was found that a second current of air 
 must be projected on to the flame above 
 the dome referred to. The arrangement 
 is therefore described in the complete 
 specification as consisting of " two round- 
 topped hollow cones or domes, one ex- 
 terior and the other interior, both being 
 so arranged as to produce two separate 
 and distinct currents of air. J ' Messrs. 
 Thomas Bo watt & Sons patented an im- 
 proved anucapnic burner in 1878, and in 
 1882 obtained protection for a duplex 
 chimneyless burner with single-slotted 
 double dome, which they term the 
 "Lome" burner. In the latter burner 
 the wick tubes are inclined towards each 
 other at the top and bottom, so that the 
 flames coalesce, and the wicks beneath 
 the burner are in contact. Increased 
 luminosity and increased capillary attrac- 
 tion result. This firm was the first to 
 construct burners with a screw to attach 
 them to the collar of the lamp, the burn- 
 ers previously made in this country being 
 provided simply with a conical tube fit- 
 ting into the collar. 
 
178 
 
 Young's Paraffin e Light and Mineral 
 Oil Company, some years ago, com- 
 menced the manufacture of lamps on a 
 large scale, and have introduced several 
 burners of novel and valuable character. 
 Among these are the "Champion" burn- 
 er, with circular wick, the "Regulator" 
 burner, and the " Triplex " burner, which 
 is furnished with three flat wicks arranged 
 triangularly. The " Champion " burner 
 was patented by A. E. Ragg, in 1876 and 
 1880. The claims in the specification of 
 the former patent are for the use of a 
 ceniral perforated cone or tubular cham- 
 ber within the wick-tube circle, for sup- 
 plying the upper portion of the flame 
 with a series of small separate injected 
 currents ; for the use of a series of rays 
 or divisions projecting from or approach- 
 ing close to the wick-tube or tubes, " so 
 as to separate the impinging current of 
 air into a series of currents separated by 
 quiescent intervals, thus causing serra- 
 tions in the flame ;" for the mode of "pre- 
 venting smoke in lamps by adjusting the 
 cap to that height abo -j the surface of 
 
179 
 
 the wick, that on the cap being placed in 
 position the flame changes from yellow to 
 white ;" and for the use of two flat wicks 
 in an Argand burner, in such a manner 
 that they are brought to a cylindrical 
 contour at the top. In the specification 
 of the subsequent patent the claims are 
 for the use of a disk or button (either 
 alone or in conjunction with a perforated 
 tube), of such size as to extend over the 
 entire area embraced by the inner wick 
 tube, and the greater portion or whole of 
 the wick, " so as to deflect the flame out- 
 wards to a much greater extent than 
 heretofore, and at the same time to cause 
 the central current of air to be deflected 
 into the flame above the wick with a cor- 
 respondingly greater force." Under this 
 patent, improvements in the construction 
 and fitting of the air chamber and gal- 
 lery, and an arrangement for regulating 
 the level of the oil, are also claimed as 
 novelties. The patent for Ragg's " regu- 
 lator" burner, dated 24th October, 1878, 
 is based upon his other patent of the 
 same year (drted 3d June) already re- 
 
180 
 
 f erred to. A considerable portion of the 
 specification relates to improvements in 
 the method of constructing the wick 
 tubes so as to admit of the use of two 
 flat wicks as described. One claim is, 
 however, for the " regulator," which con- 
 sists of a cap attached to the cone, and 
 so formed that it covers up a portion of 
 the wick, and excludes it from the action 
 of the flame, "It is preferred that the 
 portion of the wick covered up shall be 
 the extreme ends." The inventor states 
 that this addition to the burner regulates 
 the height to which the wick may be 
 turned up, and thus prevents it from 
 being turned up high enough to cause 
 the throwing off of smoke or unconsumed 
 carbon, and insures a uniform and per- 
 fect trimming of the wick. The uncon- 
 sumed portion of the wick protected by 
 the " regulator " is also said to act as a 
 feeder to the flame, so that in the use of 
 the lighter oils the flame is maintained of 
 full size for a greater length of time, 
 while oils too heavy for use in lamps of 
 the usual construction are caused to burn 
 
181 
 
 satisfactorily. An additional claim is for 
 a chimneyless flat-wick burner, provided 
 with a shallow trough placed above the 
 wick-tube, 'and with the bottom cut away 
 over the wick. This trough becomes 
 heated, and is stated to deflect the air 
 currents so as to produce a broad flame. 
 The " Triplex " burner was patented by 
 Doty, in 1876, the claims being for the 
 arrangement of three flat wicks triangu- 
 larly ; for the " flaring " or outward in- 
 clination of the wick- tubes ; and for the 
 construction of the cap with three inter- 
 secting hemispheres. 
 
 Mr. A. M. Silber is well known to have 
 devoted a large amount of time and at- 
 tention to the improvement of mineral 
 oil burners, and has published the results 
 of his efforts in an instructive paper read 
 before the Society of Arts in 1870. The ex- 
 cellent results obtained with the Silber 
 burner, which is of the round, or Argand 
 form, are due largely to the use of an 
 inner tube so constructed and placed as 
 to direct a current of air on to the inner 
 surface of the flame at a point some dis- 
 
182 
 
 tance above its base. Professor Barff 
 has stated that the prismatic spectrum of 
 the light emitted by this burner is nearly 
 identical with that of the solar ray ; it 
 should, therefore, allow nearly all shades 
 of color to be distinguished as in day- 
 light.* The Silber burner with central 
 tube was patented in April, 1873, and the 
 claim specified is for "the improved 
 burner hereinbefore described, in which 
 there are one or more air-tubes receiving 
 atmospheric air through apertures in 
 their sides or peripheries at the lower 
 part thereof, substantially as described." 
 Two years previously Mr. Silber patented 
 a lamp provided with an annular air- 
 space passing vertically through the body 
 of the lamp between the wick chamber 
 and the oil reservoir. Oil was supplied 
 to the wick through several small tubes 
 connecting the base of the oil reservoir 
 with the wick case, and thus the main 
 body of the oil was kept cool, while the 
 
 * This remark would, no doubt, be equally applica- 
 ble to the light given by other mineral oil lamps of 
 modern construction. 
 
183 
 
 small quantity supplied to the wick be- 
 came considerably heated. 
 
 The lamp which has been recently' in- 
 troduced by Messrs. J. Defries & Sons, 
 has several valuable characteristic fea- 
 tures. It is on the Argand principle, 
 and is the invention of Mr. Sepul- 
 chre. The air passes to the interior 
 of the flame through the base of the 
 lamp, and the wick is contained in an an- 
 nular space formed by the air- tube (which 
 passes completely through the oil reser- 
 voir) and an outer tube attached to the 
 burner collar. The latter tube reaches 
 nearly to the bottom of the oil reservoir, 
 and, being slightly longer than the wick, 
 is always sealed by the oil. Thus the 
 passage of flame from the burner to the 
 air space of the oil reservoir is prevented, 
 and the oil cannot flow out in the event 
 of the lamp being overturned. 
 
 The "Lampe Beige" has a similarly 
 placed central air-tube, but is not pro- 
 vided with the long outer wick- tube. 
 The " Excelsior " lamp bas an Argand 
 burner with two flat wicks, which, meet- 
 
184 
 
 ing in the wick case, furnish a tubular 
 flame. The wicks are fixed in wick-hold- 
 ers sliding in the wick-tube, and are 
 raised or lowered by a rack and pinion. 
 
 The " Waterbury " lamp has an Argand 
 burner provided with an ingenious ar- 
 rangement for facilitating the insertion of 
 the wick, and with an equally ingeniously 
 contrived extinguisher, to which I must 
 again refer. The " Star " lamp is also on 
 the Argand principle, and has the annu- 
 lar air-space, surrounding the wick tube 
 and passing through the oil reservoir, 
 already described as ' having been pat- 
 ented by Mr. Silber several years ago. 
 It is an objection to lamps so con- 
 structed that the oil creeps over the 
 upper edge of the wick-tube, and it has 
 accordingly been found necessary to pro- 
 vide a drip- cup in the base of the lamp. 
 A later form of the "Star" lamp manu- 
 factured by Holmes, Booth, & Hayden 
 Co., New York, has a good extinguisher. 
 One of the Argand burners (patented by 
 J. Funck in 1882) has a long inner wick- 
 tube, extending almost to the bottom of 
 
185 
 
 the cylinder, which serves as an outer 
 wick-case. The wick is held in another 
 long tube of thin metal, to which is at- 
 tached a stud working in a spiral groove 
 made in the fourth tube extending down- 
 wards from the collar which supports 
 the cone. The tube which holds the 
 wick is revolved by a pinion working in 
 teeth cut in the lower edge of the cone ; 
 and through the action of the spiral 
 groove, the wick is thus raised or low- 
 ered. Of the various flat-wick burners 
 the " Sun hinge " is so made that, without 
 removing the chimney, the cone can be 
 raised like the hinged lid of a box, for 
 trimming the wick or lighting. Another, 
 the "Manhattan," an exceedingly good 
 burner, giving a tall, well- shaped flame, is 
 provided with two cones. The principal 
 general features of the American flat- wick 
 burners are the comparative thinness of 
 the metal of which they are constructed, 
 and the shortness of the wick tubes. 
 
 Both in this country and in England 
 flat-wick burners have hitherto been far 
 more largely used than Argand or round 
 
186 
 
 burners. On the European continent, on 
 the other hand, flat-wick burners are 
 rarely employed. The very commonest 
 form of German round burner is that 
 which is known as the "Cosmos." 
 This burner has neither the additional 
 inner tube nor the button, and accord- 
 ingly it is necessary to use with it a 
 chimney much constricted a short dis- 
 tance from the base. Better results, 
 " especially with the larger sizes, are ob- 
 tained with the improved burners sup- 
 plied by Messrs. C. H. Stobwasser & Co., 
 Messrs. AVild & Wessel, and Messrs. 
 Schuster & Baer, under the names of the 
 " Victoria," the " Phoenix," the "Moon," 
 the " Helios," and the u Solar-oil" burn- 
 ers. These improved burners differ from 
 the old form of " Cosmos " burner prin- 
 cipally in being provided with an air tube 
 within the inner wick tube, or with a 
 button to deflect the inner air current. 
 
 The Eussian burners differ only in 
 details of construction from those al- 
 ready described. One of the Argand 
 burners and the flat -wick burner are in- 
 
187 
 
 tended for use with the heavier oils. The 
 latter burner has a short but capacious 
 wick tube, and takes a wick of unusual 
 thickness. The former burner has a 
 slightly conical chimney held in position 
 by a metal collar fitting over its base. 
 In 1884, the Committee of the Russian 
 Chemical and Physical Society awarded 
 to Kumberg the premium which had 
 been offered for a lamp suited to Cau- 
 casian oils of specific gravity .865 to 
 .875. 
 
 In the construction of all mineral oil 
 lamps, the principal points aimed at are 
 the production of a current of air passing 
 into the burner with suitable velocity ; 
 and the direction of this air current on to 
 the flame in a manner best adapted to 
 secure the proper combustion of the oil. 
 The heated chimney or other arrange- 
 ment induces the required flow of air, but 
 if in the case of an Argand lamp the 
 chimney be cylindrical and the burner 
 merely two tubes to hold the wick, the 
 air would not do what is required of it, 
 and the lamp would burn with a flame of 
 
188 
 
 comparatively little luminosity. In the 
 simplest forms of flat wick burners the 
 metallic dome which surrounds the top 
 of the wick tube serves to cause the air 
 current to impinge upon the flame, the 
 base of the burner being fitted with a 
 disk of perforated metal to moderate the 
 velocity of the current. In the best 
 round burners the result is attained by 
 the use of a metallic disk or other contri- 
 vance, sometimes in conjunction with a 
 cone surrounding the wick tube ; or as in 
 the Silber lamp, and other lamps, by the 
 use, in addition to the cone, of an inner 
 air tube of suitable size and form. The 
 " air diffuser " of the Defries lamp 
 (Sepulchre's patent) may be described as 
 a combination of the button and a per- 
 forated tube closed at the top, while the 
 air diffuser of the " Rochester " lamp and 
 of Young's " Champion" burner consists 
 of a cylinder closed at the top and with 
 perforated walls. The action of the 
 button or tube is aided by the employ- 
 ment of a chimney formed with a 
 shoulder, and if the chimney be con- 
 
189 
 
 stricted just above the top of the burner, 
 the use of the button or tube may in fact 
 be dispensed with. The upper part of 
 the Bayle chimney is in the form of a 
 truncated cone, the chimney practically 
 consisting of two cones united at their 
 smaller ends. It is claimed that the em- 
 ployment of this form of chimney causes 
 the outer air current, entering between 
 the glass and the wick tube of an Argand 
 burner, to pass upwards with the same 
 velocity as that of the inner current which 
 supplies the center of the flame with air. 
 The use of the double cone of the anu- 
 capnic lamp as a substitute for the chim- 
 ney, has been referred to. Lamps are 
 made in wLnbh the air current is supplied 
 by a revolving fan, driven by clockwork 
 in the base of the lamp ; this method of 
 construction being illustrated in the lamp 
 made by Messrs. Gardners. Martin 
 Rae, in 1861, patented the use of a small 
 lamp in the base of the lamp proper to 
 create a current of air, the use of a flame 
 "of the size of a common pea" being, he 
 states in his specification, " sufficient to 
 
190 
 
 rarefy the air of a lamp with a flame of 
 two inches square.'' 
 
 The ''blast lamp/' patented by Robert 
 Lavender in 1875, and introduced by 
 Young's Paraffine Oil Company for use 
 in illuminating large spaces in the open 
 air, was provided with an arrangement 
 for introducing a jet of steam into the 
 chimney, on the principle of the "blast" 
 employed in loc jmotives, and thus pro- 
 ducing a strong upward current of air, 
 suitable for the combustion of the heavier 
 mineral oils. 
 
 In the case of Argand oil burners, it 
 may be said that there are four principal 
 points at which the air currents impinge 
 upon the flame. Thus as regards the 
 interior surface of the flame, the current 
 of the air first comes into contact with 
 the base of the flame, while a portion is, 
 by the use of a deflector or central tube, 
 or other contrivance, caused to strike the 
 flame at a higher point. Similarly as re- 
 gards the outer surface of the flame a 
 portion of the air current meets the flame 
 at its base, and another portion comes 
 
191 
 
 into contact with the flame surface higher 
 up. As we shall presently see, when 
 considering the subject of lighthouse 
 illumination, Sir James Douglass's min- 
 eral oil burner has a cone provided with 
 three openings for air, at gradually in- 
 creasing heights from the base, and the 
 air is thus supplied to the exterior of the 
 flame in comparatively small quantities at 
 different elevations. 
 
 Scarcely less important than the produc- 
 tion and direction of the air current is the 
 maintenance of a proper supply of oil to the 
 flame. With most mineral oil lamps this 
 is effected through the unaided capillary 
 attraction of the wick, and it is obvious 
 that the quality of the wick is a point 
 of great importance. In the student or 
 reading lamp the action of the wick is 
 aided by maintaining a constant level of 
 oil on the principle of the bird fountain 
 already mentioned. The principle of the 
 " Moderator " lamp is also sometimes ap- 
 plied to lamps for domestic use, as in the 
 Peigniet-Changeur lamp shown at the 
 International Health Exhibition, the oil 
 
192 
 
 pumped up being, however, allowed to 
 overflow at a lower level in the wick tube 
 than in the case of colza oil. Mr. Silber 
 has proposed to lay on a supply of oil to 
 the various burners throughout a dwell- 
 ing house from a reservoir on each floor, 
 kept filled from a storage tank, the sup- 
 ply being regulated by the use of ball 
 valves. Peter Brash and William Young 
 patented a similar arrangement in 1867, 
 and Mr. D. C. Defries has recently 
 patented a method of arriving at the 
 same result. Messrs. Hinks some years 
 ago manufactured a lamp in which the 
 oil container from which the wick drew 
 its supply could be caused to descend 
 into a well or reservoir forming the base 
 of the Limp, and could thus be refilled. 
 In some lamps the action of the wick 
 proper is aided by the provision of a 
 "wick feeder," consisting of a thick and 
 loosely woven wick attached to the under 
 surface of the burner, in contact with 
 the wick proper, and dipping into the 
 oil. 
 
 The wick is usually raised or lowered in 
 
193 
 
 the wick case by the action of toothed 
 wheels pressing lightly against it, the 
 revolution of the wheels being effected 
 by turning a button on the end of a 
 spindle which carries them. In some in- 
 stances, however, the wick is held -in a 
 tube, or frame, which is raised or lowered 
 by a rack and pinion, or by a worm cut 
 on the burner tube, and actuated by re- 
 volving the burner, or, as in the "Roches- 
 ter" lamp, by means of a vertical rod 
 attached to the wick frame. The flame 
 has always hitherto been produced from 
 the extremity of the wick, but an arrange- 
 ment for presenting a fold of a continu- 
 ous flat wick as the burning surface was 
 shown in the International Inventions 
 Exhibition. The burner thus fitted was 
 provided with a device for causing the 
 wick to travel through the burner so that 
 a fresh portion of wick could be exposed 
 when desired. 
 
 The wicks at first employed in lamps 
 for burning fixed oils were of the nature 
 of a loosely woven cord or solid cylinder. 
 In 1773, Leger used a flat wick, and a 
 
194 
 
 few years later Argand adopted a tubular 
 wick. Ditmar and others have used two 
 wicks ; one to bring the oil up to the 
 burner, and the other to burn it. With 
 a large number of the round burners of 
 the present day, a flat wick of such 
 breadth that the edges meet in the annu- 
 lar wick space is employed, and in some 
 instances two flat wicks are similarly 
 used. The "Mitrailleuse" burner is, 
 however, furnished with a number of 
 solid cylindrical wicks, arranged in a 
 circle, and held in a frame, which is 
 raised or lowered by a rack and pinion. 
 Messrs. Browne & Co., and Mr. Ret- 
 tich, manufacture this form of burner. 
 The Rettich "Mitrailleuse" burner has 
 an air deflector of improved form. 
 The " Sirius '' burner, patented by Mor- 
 rison & Smith, and one form of the 
 " Mirtin " burner, are provided with two 
 concentric wicks ; but compound Argand 
 burners are, for the most part, used only 
 for lighthouse illumination. 
 
 Considerable attention has for some 
 years been paid to the construction of 
 
195 
 
 wicks in the United States. The various 
 operations, in the order in which the 
 cotton is subjected to them, are in the 
 United States termed: (1) opening; 
 (2) lapping ; (3) carding ; (4) railway 
 drawing ; (5) doubling and first drawing ; 
 (6) doubling and second drawing ; (7) 
 slubbing and roving; (8) speedy roving; 
 (9) spinning; (10) twisting, which makes 
 the yarns ready for the looms. 
 
 It is important to dry the wick before 
 it is used, and wicks are frequently found 
 to have absorbed from 4 to 6 per cent, of 
 their weight of moisture, and to the ex- 
 tent to which this moisture is present, 
 the capillary attraction of the wick when 
 used in the lamp is impaired. The fol- 
 lowing results of experiments which have 
 been made with wicks of various qualities 
 in common use, clearly indicate the im- 
 portance of employing a wick of good 
 quality. The figures indicate the relative 
 quantities of a given oil of good quality 
 drawn through wicks of the same width, 
 by capillary attraction in a given time : 
 
196 
 
 Wick of best quality 198 
 
 Wick of medium quality 100 
 
 Wick of inferior quality 76 
 
 The extent to which the behavior of 
 the lamp is affected by the quality of the 
 wick, especially with oils that do not flow 
 very freely under the influence of capilla- 
 ry attraction, is strikingly shown by the 
 following results which the author ob- 
 tained under otherwise similar condi- 
 tions : 
 
 Wick of 
 
 Wick of best ordinary 
 quality. quality. 
 
 Maximum illuminating "1 ^ 
 
 power 10.43 9.99 
 
 Minimum illuminating | ^ 
 
 po WIT after six hours* 
 
 burning ....:... 9.63 7.64 } 
 
 Average illuminating " 
 
 power during six 
 
 hours' burning 10.14 8.99 
 
 Diminution in illumi- . J 
 
 n at ing power, per 
 
 cent 7.6 23.5 
 
 Oil consumed per hour. 529 500 ) g 
 Oil consumed PLT candle 
 
 light per hour 52.17 55.61) 5 
 
 Many of the complaints of unsatisfac- 
 tory burning quality of oil have undoubt- 
 edly arisen from the use of inferior wicks. 
 
197 
 
 As an illustration of the ignorance which 
 frequently prevails as to the importance 
 of using a good wick, it may be men- 
 tioned that an American, engaged in the 
 petroleum trade, being unable to account 
 for the reiterated statements of a custom- 
 er that the oil supplied would not burn, 
 caused an examination to be made of the 
 lamp in which the oil was used, and found 
 that the wick having become short the 
 complainant had ingeniously lengthened 
 it by attaching with two pins a strip torn 
 from an old flannel garment. This ar- 
 rangement appeared to have answered 
 the purpose fairly well until one of the 
 pins fell out unperceived, when the sur- 
 faces of contact of the flannel and wick 
 were no longer sufficiently large, and the 
 lamp having ceased to burn, the oil was 
 hastily assumed to be in fault. 
 
 In certain cases, however, it is found 
 that a wick gives satisfactory results for 
 several days and then appears to become 
 choked. Even in such cases, however, 
 the use of a sufficiently good wick over- 
 comes the difficulty, as is shown in the 
 
198 
 
 following tabular statement of the results 
 of some experiments made two years ago. 
 For convenience of comparison the illu- 
 minating power is expressed at the com- 
 mencement of the experiment as 100 in 
 each case : 
 
 Wick of 
 
 Wick of bet ordinary 
 quality. quality. 
 
 Maximum illuminating power 
 
 when the wick was new 100 . . 100 
 
 Maximum illuminating power 
 
 after the wick had been used 
 
 for a total of 21 hours on 3 
 
 successive days . . 58.5 
 
 Maximum illuminating power 
 
 after the wick had been used 
 
 for a total of 32^ hours on 5 
 
 successive days . . 17.8 
 
 Maximum illuminating power 
 
 after the wick had been used 
 
 for a total of 50 hours on 7 
 
 successive days 93.8 . . 
 
 Mr. Nakamura formed the opinion that 
 this choking results from the deposition 
 of water in the wick, since the wick re- 
 covers its capillary power on drying. 
 Among those who have experimentally 
 investigated the subject of the diminution 
 in illuminating power in petroleum lamps 
 are Colonel Junker, director of the Test 
 
199 
 
 Bureau of the Bremen Petroleum Borse, 
 Mr. L. Schmelk, chemist to the Norwe- 
 gian Sea-lighting establishment, and Dr. 
 J. Biel, of St. Petersburg. 
 
 While on the subject of wick manu- 
 facture, it may be pointed out that it is 
 much to be desired that lamp manufactur- 
 ers would adopt standard gauges for the 
 wick tubes. At the present time an un- 
 necessarily large number of wicks, vary- 
 ing in thickness and in width by sixteenths 
 of an inch, have to be made and kept in 
 stock, and since the nominal widths of the 
 wick tubes sometimes differ to a con- 
 siderable extent from the actual widths, 
 it frequently results that the wick sup- 
 plied does not fit the tube properly. The 
 wicks in use in England with miner- 
 al oils range in width from J of an 
 inch to 5 inches, the greatest difference 
 in width between any two of the interme- 
 diate sizes being ^ of an inch, and the 
 difference being in many cases as little as 
 one-sixteenth of an inch. There is also 
 considerable variation in different burn- 
 ers in regard to the thickness of the wick 
 
200 
 
 the holder will properly take. If the 
 wick is too narrow, the tamp will not 
 burn well, and its use may even be 
 dangerous. If the wick be too thin, it is 
 deficient in capillary attraction, and if 
 too thick it will not move freely. 
 
 Several kinds of incombustible wicks 
 have at various times been introduced. 
 The wick patented in 1876, and in an 
 improved form in 1877, by Hein- 
 richs, consists of a lower portion of 
 felt, an intermediate portion of miner- 
 al wool, and an asbestos top or 
 ring. In the International Inventions 
 Exhibition last year, a lamp with an 
 asbestos wick patented by Messrs. Flatau 
 and Turner was shown. One of the 
 claims in the patent specification relating 
 to this lamp is for the division of the 
 wick horizontally into two portions, one 
 of which can be moved up or down so as 
 to be put into or out of contact with the 
 other portion. The inventors appear to 
 prefer that the upper part of the asbestos 
 wick should be a fixture in the wick tube 
 and that the lower part should be moved 
 
201 
 
 downwards out of contact with the upper 
 part when it is desired to extinguish the 
 lamp. 
 
 It is now usual to fit the larger burn- 
 ers with some form of extinguishing ap- 
 paratus. The earliest attempt to fit an 
 extinguisher was probably that made 
 in the case of the improved "Brigh- 
 ton " burner, patented in 18G2. This 
 burner has an air deflector or but- 
 ton, the stem of which rests upon a 
 pin passing horizontally through the 
 burner. On withdrawing the pin, the 
 button drops on the wick and extinguish- 
 es the flame. In the " Wateubury " burn- 
 er, the dropping of the button is also 
 effected by drawing out a pin, but the 
 action compresses a spring and the but- 
 ton resumes it normal position on the pin 
 being released. The button extinguisher 
 of the new " Star " lamp is brought into 
 action by depressing a thumb plate, and 
 in this case also the button returns to 
 its original position when the pressure is 
 removed. The u Duplex " burners of Mr. 
 James Hinks & Son, Messrs. Wright & 
 
202 
 
 Butler, and others, are fitted with ingen- 
 iously contrived extinguishing apparatus, 
 which, on depressing a lever, bring" a pair 
 of metallic plates into contact over the 
 top of each wick tube. In one of the 
 "Duplex" lamps made by Messrs. 
 Wright & Butler, the extinguisher is 
 automatically brought into action as soon 
 as a weighted rod suspended beneath the 
 burner passes to any material extent out 
 of a line perpendicular to the base of the 
 lamp. It is therefore impossible for the 
 lamp to become tilted when falling with- 
 out the flames being extinguished. An 
 automatic extinguisher has also been 
 patented by Messrs. King & Godfrey. 
 
 Two of the " Duplex " burners and 
 Kettich's " Mitrailleuse " burner are pro- 
 vided with mechanical arrangements for 
 raising the gallery carrying the chimney 
 and globe, so that the wicks may be con- 
 veniently lighted. In Wright & Butler's 
 " Duplex " the gallery is supported on 
 levers, which, on turning a key, not only 
 raise the chimney and globe, but also 
 move them to some extent horizontally. 
 
203 
 
 In Hinks' "Duplex '* the movement of 
 the gallery, effected by turning the key, is 
 vertical only. In Rettich's '^Mitrailleuse" 
 there is no key action,but the gallery slides 
 up to an extent sufficient to admit of in- 
 serting a lighted taper. 
 
 Messrs. C. H. Stobwasser & Co., of 
 Berlin, have recently adopted a process 
 for preventing the oil from "creeping " 
 over the edge of the burner collar and 
 soiling the exterior of the reservoir. 
 The process consists in placing in the 
 collar, between the edge of the reservoir 
 and the plaster of Paris which is used to 
 attach the collar, a layer of some com- 
 pound which looks like a mixture of gel- 
 atine and glycerine. The oil will pass 
 through the plaster, but is arrested 
 by the compound referred to. The upper 
 surface of the collar of the lamp is also 
 made slightly conical, and at its inner 
 and lower edge, at the junction between 
 the collar and the burner, a small hole 
 communicating with the oil reservoir is 
 made. Any oil, therefore, which drops 
 from the burner returns to the reservoir. 
 
204 
 
 On the pin of the wick winder is soldered 
 a small star shaped wheel, the points of 
 which are over the conical collar. Oil 
 passing along the pin is stopped by the 
 star wheel, and drops from its lowest 
 point on to the collar. 
 
 We have now to consider the principles 
 of construction of mineral oil lamps in 
 relation to the question of safety. It 
 is well known that accidents in the use 
 of mineral oil lamps are, unfortunately, 
 of by no means rare occurrence, though 
 the number bears a very small proportion 
 to that of the lamps in use. The atten- 
 tion of authorities and experts in the 
 United States has long since been direct- 
 ed to the comparative frequency of such 
 accidents, and Dr. Chandler, of New 
 York, as long ago as 1871, published a 
 report of a lengthy series of experiments 
 which he had undertaken with the object 
 of ascertaining the conditions under 
 which the accidents occurred. It seems, 
 however, to have been somewhat hastily 
 assumed that the accidents were the re- 
 sults of explosions of the mixture of 
 
205 
 
 petroleum vapor and air formed in 
 the upper part of the oil container, 
 and the experiments were therefore 
 chiefly directed to ascertaining the rela- 
 tion between the flashing point of the oil 
 and the temperature to which the oil was 
 raised when burning in various forms of 
 lamps. Commenting on these experi- 
 ments, Mr. Peckham very properly points 
 out in his census report that, although 
 explosions undoubtedly sometimes break 
 lamps, the danger arises principally from 
 the risk of overturning and breaking the 
 lamp. 
 
 In a lecture, delivered at the Royal 
 Institution eleven years ago, Sir Fred- 
 erick Abel stated that a large propor- 
 tion of the accidents arising out of the 
 use of mineral oil lamps were not actually 
 due to the occurrence of explosions; 
 and in a subsequent lecture he added, 
 that instances might be quoted in which 
 the breaking out of a fire, or the de- 
 struction of or injury to life, which had 
 evidently been caused by upsetting or 
 allowing to fall a mineral oil lamp, had 
 
206 
 
 been erroneously ascribed to an explo- 
 sion. There are, however, as Sir F. 
 Abel said, numerous cases of accidents 
 which have been caused by explosions 
 in lamps, followed by the ignition of the oil. 
 The experiments which have been made 
 by Sir Frederick Abel and the author, 
 with the valuable aid of Dr. W. Kellner, 
 Assistant Chemist of the War Depart- 
 ment, have enabled them to arrive at 
 several definite conclusions with respect 
 to the immediate causes of lamp explo- 
 sions, and to certain circumstances which 
 may tend to favor the production of such 
 explosions. These conclusions were so 
 clearly set forth by Sir Frederick Abel, 
 which is here quoted verbatim: 
 
 "If the lamp of which the reservoir is 
 only partly full of oil, be carried or 
 rapidly moved from one place to another, 
 so as to ngitate the liquid, a mixture of 
 vapor and air may make its escape from 
 the lamp in close vicinity to the flame, 
 and, by becoming ignited, determine the 
 explosion of the mixture existing in the 
 reservoir. This escape may occur through 
 
207 
 
 the burner itself, if the wick does not fit 
 the holder properly, or through openings 
 which exist in some lamps in the metal 
 work, close to the burner, of sufficient 
 size to allow the flame to pass them 
 readily. A sudden cooling of the lamp, 
 by its exposure to a draught, or by its 
 being blown upon, may give rise to an 
 inrush of air, thereby increasing the ex- 
 plosive properties of the mixture of 
 vapor with a little air contained in the 
 reservoir, and the flame of the lamp may 
 at the same time be drawn or forced into 
 the air-space filled with that mixture, es- 
 pecially if the flame has been turned 
 down, as the latter is thereby brought 
 nearer to the reservoir. The sudden 
 cooling of the glass, if it had become 
 heated by the burning of the lamp, may 
 also cause it to crack if it is not well an- 
 nealed, and this cracking, or fracture, 
 which may allow the oil to escape, may 
 convey the idea that an explosion has 
 taken place. If the evidently common 
 practice is resorted to of blowing down 
 the chimney with a view to extinguish 
 
208 
 
 the lamp, the effects above indicated as 
 produceable by a sudden cooling may be 
 combined with the sudden forcing of the 
 flame into the air space, and an explo- 
 sion is thus pretty certain to ensue, es- 
 pecially if that air space is considerable. 
 If the flashing point of the oil used be 
 below the minimum (73 Abel) fixed by 
 law, and even if it be about that point or 
 a little above it, vapor will be given off 
 comparatively freely if the oil in the lamp 
 be agitated, by carrying the latter, or 
 moving it carelessly ; the escape of a 
 mixture of vapor with a little air from the 
 lamp, and its ignition, will take place more 
 readily, but, on the other hand, it will 
 probably be feebly explosive, because the 
 air will have been expelled in great meas- 
 ure by the generation of petroleum vapor. 
 If the flashing point of the oil be high, 
 the vapor will be less readily or copiously 
 produced, under the conditions above 
 indicated, but, as a natural consequence, 
 the mixture of vapor and air existing in 
 the lamp may be more violently explosive, 
 because the proportion of the former to 
 
209 
 
 the latter is likely to be lower and nearer 
 that demanded for the production of a 
 powerfully explosive mixture. If the 
 quantity of oil in the lamp reservoir be 
 but small, and the air-space consequently 
 large, the ignition of an explosive mixture 
 produced within the lamp will obviously 
 exert more violent effects than if there 
 be only space for a small quantity of 
 vapor and air, because of the lamp being 
 comparatively full. If the wick be low- 
 ered very much, or if for some other 
 reason the flame becomes very low, so 
 that it is burning beneath the metal work 
 which surrounds and projects over the 
 wick holder, the lamp will become much 
 heated at those parts, and the tendency 
 to the production of an explosive mixture 
 within the space of the lamp will be in- 
 creased, while, at the same time, heat 
 will be transmitted to the glass, and it 
 will be correspondingly more susceptible 
 to the effects described as being exerted 
 by its sudden exposure to a draught. 
 Experiments have demonstrated that a 
 lamp containing an oil of high flashing 
 
210 
 
 point is more liable to become heated 
 than if it contained a comparatively light 
 and volatile oil, in consequence of the 
 much higher temperature developed by 
 the combustion, and of the comparative 
 slowness with which the heavy oil is con- 
 veyed by ih& wick to the flame. It 
 therefore follows that safety in the use 
 of mineral oil lamps is not to be secured 
 simply by the employment of oils of very 
 high flashing point (or low volatility), and 
 that the use of very heavy oils may even 
 give rise to dangers which are small, if not 
 entirely absent, with oils of comparatively 
 low flashing points." 
 
THIS BOOK IS DUE ON THE LAST DATE 
 STAMPED BELOW 
 
 AN INITIAL FINE OF 25 CENTS 
 
 WILL BE ASSESSED FOR FAILURE TO RETURN 
 THIS BOOK ON THE DATE DUE. THE PENALTY 
 WILL INCREASE TO SO CENTS ON THE FOURTH 
 DAY AND TO $I.OO ON THE SEVENTH DAY 
 OVERDUE. 
 
 '..." 
 
 
 ' .- I3|] 
 
 
 HFf; $ 154,| 
 
 
 FEB 28 1944 
 
 
 
 
 
 
 
 
 
 
 : 
 
 
 ' 
 
 
 
 
 ?MvU^:,v 
 
 
 
 
 
 
 
 LD 21-100m-7,'40 (6936s) 
 
YA 03414 
 
 
 
 ,