UC-NRLF 
 
 SbM 333 
 
ffttturmHjr 
 
 JVo 
 
 Division 
 ^ Range 
 
 Shelf. 
 
 Received 
 
 / 
 
 187/. 
 
 / 
 
FRICTIONAL ELECTRICITY. 
 
A TREATISE 
 
 FRICTIONAL ELECTRICITY, 
 
 |n Cjjeori) aitb Iratfhe. 
 
 BY 
 
 SIR WILLIAM SNOW HARRIS, F.R.S. 
 
 \ 
 
 EDITED, WITH A MEMOIR OF THE AUTHOR, 
 
 BY CHARLES TOMLINSON, F.R.S. 
 
 LI B It \ i. V 
 
 j| UNlVKilM i V OF 
 
 CALIFORNIA. 
 
 LONDON: 
 
 VIRTUE AND CO., 26, IVY LANE, 
 
 PATERNOSTER ROW. 
 1867. 
 
LONDON ! 
 
 PRINTED BY VIRTUE AND CO., 
 CITY ROAD. 
 
PREFACE BY THE EDITOR. 
 
 THE great work on Frictional Electricity with which Sir William 
 Snow Harris proposed to close his scientific career he was not 
 permitted to complete. The whole of the First Part up to page 
 205 of the present volume, had the benefit of his final revision. 
 After his death, his papers were placed in my hands by Lady 
 Harris and Mr. Harris, her son. I found them to be in a confused 
 and unsatisfactory state, but I have done my best to complete the 
 work in as methodical a manner as the materials under command 
 would admit of. All that is given in the Second Part is in the 
 actual language of the author, and in passing it through the 
 press I have had the benefit of Mr. Harris's revision, which 
 has in many cases been of value to the work, while it has given 
 additional confidence to my own labours. Among Sir William's 
 papers I found some notes intended to form the basis of a Preface 
 to this work. I have retained them as they were written. It is 
 probable that had the author revised his Preface he would have 
 exercised his caution in mitigating its severity, or have withheld 
 it altogether ; but when a distinguished man is dead, we care 
 more about knowing his opinions than his caution when alive in 
 withholding them. 
 
 KING'S COLLEGE, LONDON, 
 1st October, 1867. 
 
LIBRARY 
 
 UNIVERSITY OF 
 
 CALIFORNIA. 
 
 BIOGRAPHICAL NOTICE OF THE 
 AUTHOR. 
 
 BY THE EDITOR. 
 
 THE life of a scientific man seldom calls for much exertion on the 
 part of his biographer. Such a life may be even less eventful 
 than that of a literary man ; for not only has the latter a wider 
 and more varied audience, but his work has more individuality 
 and more character ; it is more a reflection of the author's mind, 
 more rounded and complete, than a scientific work : the one 
 belongs to the man, the other to nature ; the one remains for all 
 time as an exponent of the author's mind, the other is soon 
 absorbed into the general body of science, and its author more or 
 less forgotten. This need not be regretted, since the object of 
 science is the discovery of truth, not individual celebrity ; and 
 our love and gratitude are not the less due to the men who have 
 laboured in the cause of truth, even though their work be quickly 
 absorbed, and their books and memoirs soon cease to be read. 
 
 The author of the volume now before the reader completed the 
 allotted term of man's life. He made some important discoveries 
 in electrical science, laboured long and successfully to introduce 
 into the Eoyal Navy his system of lightning conductors, was an 
 excellent sailor, an accomplished musician, a lively intellectual 
 companion, a steady friend, and he died honoured and beloved. 
 
 Such was William Snow Harris, and it will scarcely add to our 
 interest to be told that he was born on the 1st of April, 1791, the only 
 son of Thomas Harris, Esq., of Plymouth, whose family had settled 
 in that town as solicitors as early as the year 1600. He was educated 
 first at the Plymouth Grammar School, then for the medical profes- 
 sion, and he completed his studies at Edinburgh. He appears to 
 
Vlll EARLY PAPERS. 
 
 have been struck with Dr. Hope's lectures on chemistry, and has 
 related to the editor several amusing anecdotes of that distin- 
 guished teacher. Snow Harris began the practice of his profession 
 as a militia surgeon, and afterwards as a general practitioner in 
 Plymouth. But his great love of science, especially of electricity, 
 interfered with his medical practice ; and after his marriage, in 
 1824, with the eldest daughter of Richard Thorne, Esq., of Pilton, 
 North Devon, he devoted himself chiefly to the cultivation of his 
 favourite science. He had already, in 1820, invented a new method 
 of arranging lightning conductors for ships, with a view to the 
 defence of the Royal Navy from the destructive effects of light- 
 ning. In the electrical and mechanical arrangements of this 
 system, the metal was permanently fixed in the masts, and 
 extended throughout the hull, so as to afford the required security 
 at all times and under all the variable circumstances in which the 
 ship might be placed, 
 
 Harris showed his affection for Edinburgh by laying before its 
 Royal Society a series of papers, the first of which was entitled, 
 " Experimental Inquiries concerning the Laws of Magnetic Forces/' 
 This paper appeared in the " Transactions " in 1829, but it is dated 
 " Plymouth, July 1st, 1827." It contains an account of the Hydro- 
 static Magnetometer. The second paper, " On a New Electrometer, 
 and the Heat excited in Metallic Bodies by Yoltaic Electricity," 
 is dated May 5th, 1831, and was read 19th December of that year. 
 The third paper, dated April 5th, 1833, " On the Investigation of 
 Magnetic Intensity by the Oscillations of the Horizontal Needle," 
 was read 6th January, 1834. By this time the author adopts the 
 title, "F.R.SS. Lond, and Ed." 
 
 Harris's introduction to the Royal Society arose from his skill 
 in the improvement and construction of instruments and apparatus, 
 of which there is abundant evidence in his early papers. While 
 Sir Humphry Davy was President he had been attracted by the 
 simplicity and ingenuity of the thermo- electrometer, and invited 
 Harris to lay an account of it before the Society. This was 
 accordingly done, and the paper " On the Relative Powers of various 
 Metallic Substances as Conductors of Electricity, by Mr. William 
 Snow Harris, of Plymouth, Surgeon," was read before the Society, 
 14th December, 1826, and appeared in the "Transactions" for 
 1827. 
 
MANIPULATIVE SKILL. IX 
 
 Harris was justly proud of the favour with which the President 
 and also the Vice- President, Dr. Wollaston, regarded his inventions 
 and discoveries. This was, perhaps, natural in the case of Wol- 
 laston, who delighted in manipulative skill ; but it was remarkable 
 in the case of Davy, who valued the end of scientific research, and 
 rather despised than cherished the means. 
 
 Harris was elected into the Royal Society in 1831. His papers 
 contributed to the Society in 1834, 1836, and 1839, on the 
 elementary laws of electricity, contain his best work, and display 
 in a striking manner the author's ingenuity and delicate mani- 
 pulative skill. He was not satisfied with attaining his end by 
 any means, but the means themselves were all-important, and 
 were the subject of long and anxious thought and repeated trials, 
 until at length the best means possible, under the circumstances, 
 had been hit upon. This care in the selection and improvement 
 of apparatus might seem to an ordinary observer to be often 
 superfluous, but it led to success and to the thorough understanding 
 of the conditions of success, so that the failure of an experiment 
 in Harris's hands became next to impossible. The chapter in the 
 present volume on Electrical Manipulation will give some idea of 
 the author's scrupulous care in the management of apparatus. 
 He left nothing to others ; even the making of paste for attaching 
 the tinfoil to his jars was a matter of study until the best recipe 
 had been found. But with all this love of apparatus, and of 
 its minute details, Harris had none of the spirit of a mere instru- 
 ment maker. He knew that the best instrument does the best 
 work under the guidance of the best mind ; for, as Liebig has it, 
 Das Instrument macht ja das Werk nicht, sondern der menschliche 
 Geist. Nor was his ingenuity confined to his apparatus. There 
 was not a room in his house that did not bear marks of the pre- 
 sence of an original mind. He converted the ceiling of his children's 
 nursery into a planetarium, and the floor into a compass-card. He 
 taught them book-binding and printing ; he was ready to repair 
 a roller-blind, or to keep in order his ingenious kitchen-range. 
 
 In 1835 the Copley Medal, " the olive crown " of the Royal 
 Society, as Davy loved to call it, was bestowed on Harris, in recog- 
 nition of the value of his papers on the laws of electricity of high 
 tension. In 1839 his " Inquiries concerning the Elementary Laws 
 of Electricity " formed the subject of the Bakerian Lecture, and 
 
X LIGHTNING CONDUCTORS. 
 
 was rewarded by the Society agreeably to the bequest of Henry 
 Baker, Esq., F.R.S., made in 1774. 
 
 But, in the midst of these highly successful researches, Harris 
 had never ceased to exert himself in behalf of his favourite system 
 of lightning conductors for ships. So little was the subject known 
 at the time, that the most erroneous views were held respecting 
 the phenomena of a thunder-storm, and the action of conductors. 
 It was considered by many highly-educated persons, especially 
 among naval men, that conductors did more harm than good that 
 they were even positively mischievous by attracting lightning to 
 the structure they were intended to protect. To set these doubts 
 and difficulties at rest, a mixed commission of naval and scientific 
 men was appointed by the Government to investigate and report 
 on lightning conductors for ships. The Committee met several 
 times in one of the rooms of the Royal Society at Somerset House, 
 when Harris gave his evidence, exhibited his experiments, and 
 answered the inquiries of the Commissioners. Dr. Wollaston took 
 great interest in this inquiry. On one occasion it was proposed to 
 send the charge of a large battery through a mixed system of con- 
 ductors in a curved form. It was asked what the effect would be 
 if a thin wire was extended along the chord of the curve. Harris 
 replied that the discharge would be along the chord. Wollaston 
 then took out of his pocket some of his fine platinum wire, and 
 joined with it the two ends of the curve. Some difficulty 
 arose as to how to support the curve. Harris at once said, " I 
 will hold it." The Committee begged him not to run so much 
 risk. He replied, "If I were not certain of my principles I 
 should be ashamed to appear before you. There is no danger." 
 The discharge went through the platinum wire, which, of course, 
 disappeared in vapour. It is needless to say that the conclusion 
 arrived at by the Commissioners was, " that the fixed conductors 
 of Mr. Harris are superior to all others, and they earnestly recom- 
 mend their general adoption into the Royal Navy." 
 
 Still, however, the official mind was not satisfied. Ten ships 
 had been ordered to be fitted with the conductors, trials of their 
 value had been made in various parts of the world, and, although 
 experience fully justified their use, an order was given to 
 have the copper conductors stripped off every ship as it came 
 into dock. About this time some granite chimneys had been 
 
WORK ON THUNDER-STORMS. XI 
 
 erected in the Victualling Department at Devonport, and the 
 engineer, Sir John Hennie, had fitted one of them with lightning 
 conductors on Harris's plan, and was about to protect the other in 
 a similar manner, when an order was given to remove the con- 
 ductor from the protected chimney, and to leave the other chimney 
 as it was. On the receipt of this order the engineer waited on 
 Harris in a state of great excitement, and, after talking the 
 matter over, Harris said, " Well, never mind, Nature will avenge 
 us." Never was prophecy more truly uttered, for before the order 
 could be obeyed a thunder-storm passed over Plymouth, the light- 
 ning struck the unprotected chimney, and made a great rent in 
 its side, while the protected chimney was of course uninjured. It 
 happened, fortunately for the cause of science and good sense, that 
 the British Association held its meeting in Plymouth soon after 
 this event, and it was a common joke during the week to taunt 
 Harris by pointing to the damaged chimney. Harris, however, 
 smiled, and said nothing, for he knew how fruitful the accident 
 had been to him. His conductors not only got into favour with 
 the Admiralty, but his scientific discoveries were now recognised 
 in the same quarter, and he was recommended to the Government 
 as worthy of an annuity of 300 " in consideration of services in 
 the cultivation of science." 
 
 But the conductors were not yet admitted into the Royal 
 Navy. To remove the prejudices which still existed on the 
 subject of protection against lightning, in the minds of naval 
 men and others, Harris published, in 1843, his well-known 
 work on Thunder-Storms. The author requested the editor 
 to see the work through the press, and this led to a friend- 
 ship which was broken only by death. In the preparation 
 of this work, and of several pamphlets on protection from light- 
 ning, &c., subsequently published, the editor was in frequent 
 communication with Harris, either personally or by letter. It 
 was Faraday's theory of induction by means of contiguous 
 particles that first made the phenomena of the lightning - 
 stroke clear, and Harris was the first to take advantage of the 
 theory in his book. In order to be quite sure that he repre- 
 sented the theory fairly, the early sheets of the book were sent 
 to Faraday, who returned them with approval and without a 
 single alteration. 
 
Xll WORK ON THUNDER-STORMS. 
 
 Allusion has already been made to the uncertain opinions held 
 even by scientific men on the subject of thunder-storms. The 
 following extract from a letter from Harris to the editor will 
 show that he at least had clear ideas, and knew how to express 
 them. Referring to his work on Thunder- Storms, then going 
 through the press, he says : 
 
 " I think you will find the last section very interesting, as bearing forcibly 
 upon the results of our experience of the protecting effect of conductors ; and 
 I think I have succeeded in treating satisfactorily all the cases in which 
 buildings armed with conductors are said to have been damaged. 
 
 " I read in Arago's Notices Scientifiques, at pp. 597, 598, in the Annuaire 
 for 1838, an account of the damage done to an angle of the magasin at 
 Bayonne having a pointed conductor. This ia a capital case for me. I dis- 
 agree with Arago and all of them about it, and think I can show that it 
 proves, beyond a question, the wonderful effect of the conductor in that 
 instance in saving the building. Arago says that the Colonel of Artillery at 
 Bayonne reported on it, and that M. Gay-Lussac subsequently drew up a 
 memoir for a section of the Academy of Sciences. The accident occurred in 
 February, 1829, so that I suppose the account must be found in the volumes 
 of the Academy. As I have the vanity to think I can put the Academy 
 right about this, I am extremely anxious to ascertain at what distance the 
 south-west angle (damaged by the lightning) was from the conductor, and 
 if they have given any diagrams or descriptions sufficiently clear to put the 
 representation on paper. By Arago's account I take it that the arrange- 
 ment was something of this 
 kind, in which is the 
 conductor, and A the point 
 damaged. I mean to write 
 to Arago about it. If, how- 
 ever, in the meantime you 
 should happen to be at the 
 Museum, and could just look 
 into the memoirs of the Aca- 
 demy for 1829, you would 
 soon see if such a report is 
 there, and if there are any 
 drawings. I only want two 
 things : 
 
 " First. The distance of 
 conductor from the damaged 
 angle. 
 
 " Second. A general notion of the position of the conductor on the 
 building, something in the way of the sketch I have just given. 
 
 " This building was evidently struck by the bifurcated discharge in two 
 places, viz., on the angle connected with the conductor, and likewise on the 
 
WORK ON THUNDER-STORMS. Xlll 
 
 conductor itself, and really that the conductor carried off the whole without 
 damage to the building. The point of the angle on which the shock fell 
 would naturally enough be ruffled." 
 
 Soon after the date of the above lie writes (6th April, 1843) : 
 
 " I have found the notice of Gay-Lussac's Eeport in the Annales de Chimie, 
 tome xl., p. 386 398 ; but it is, to a certain extent, very vague and un- 
 satisfactory, and does not give me the information I want by any means. 
 It is a clumsy affair altogether. I have written to Arago about it. It was 
 of the utmost importance to have given the relative positions of the conductor 
 and the point of the building struck, and the distance. 
 
 " I have also found the date of the explosion at Brescia, but I do not know 
 where to find the original accounts. 
 
 " I must now make the best of what I have. I think I see the elements of 
 a more extensive and complete work on Lightning than has hitherto appeared. 
 I have obtained an amazing amount of cases from the Logs of the Navy." 
 
 The work on Thunder- Storms did not attract much attention, 
 and, indeed, can scarcely be said to have paid its expenses. One 
 of Harris's most intimate friends, Professor Daniell, of King's 
 College,* expressed to the editor his regret that Harris should 
 have treated the subject in a separate volume. " It would/' said 
 he, "have made a splendid chapter in Harris's large work on 
 Electricity;" for even at this period Harris had such a work in 
 contemplation. But his object was not so much the sale of the 
 book as the facility it afforded him of distributing information 
 respecting his system in certain influential quarters, as will appear 
 by the following letter, dated August 10th, 1843, which also shows 
 that he was not indifferent to the circulation of information among 
 the general public ; the first part of the letter referring to a notice 
 of his work which the editor had been requested to write for a 
 popular magazine. 
 
 "Nothing can be better than your condensed view of my work intended for 
 the ***** Magazine. The only fault I find with it is the rather partial 
 reference to myself. However, I am quite content to put up with that, espe- 
 cially at your hands. I think you should be cautious, lest it appear that 
 the object of the article was to uphold my system of conductors as applied to 
 the Navy ; but I think the rapid and clear condensation of the matter of my 
 book is very admirable. 
 
 * Professor Daniell included Frictional Electricity in his annual course at King's 
 College, and it was his earnest wish that Harris should accept a Professorship, and 
 deliver these lectures once a year. He felt that it would reflect honour on the 
 College, and that the Council would readily make the appointment ; but Harris 
 declined the offer, fearing that it would restrict his movements. 
 
XIV HARRIS AS A LECTURER. 
 
 1 ' With respect to the idea I have advanced relative to the suffocating 1 
 odour attendant on electrical discharges, and that the electrical explosion is 
 accompanied by ignited master dragged into the track of the discharge, it 
 appears to me worth enlarging on, and made subject for your original think- 
 ing. It is stated in the Nautical Magazine, p. 439, vol. i., that in the case of 
 the smack London, struck by lightning off St. Abb's Head, 'the hold 
 was filled with sulphureous smoke, and a coating of black bituminous matter 
 left on a large extent of surface.' This suffocating smell of sulphureous 
 matter is the most marked on shipboard. You say it is not always compared 
 to that of burning sulphur ; but in all the cases in the logs of our ships 
 there is not an instance to the contrary. The sailors always describe the 
 odour as an intolerable stench of sulphur and brimstone. See the last three 
 numbers of the Nautical Magazine for my publishing cases of damage by 
 lightning in the Navy. 
 
 ' ' And now, having despatched your affair, I proceed to thank you for the 
 plumbago, &c., and to lay myself under fresh obligations to you. First, 
 will you be so good as to inquire of Mr. Parker (to whom. I beg kindly to be 
 remembered) whether the book is bound fit to be presented to Prince Albert ? 
 If so, to let me know. Second, I wish to send copies of my work to Quetelet, 
 of Brussels, and to Arago ; for, although the latter has not treated me very 
 civilly, yet it will be well to send him the book, I think, as I understand 
 that some inquiries on the subject are going on in France. Will you, then, 
 be so good as to get two copies, to be placed to my account ? Enclose the note 
 herewith sent, and the book, in a cover, directed to Professor Quetelet, 
 Brussels, and leave it with !M>. Eobertson, the Assistant Secretary of the 
 Eoyal Society. I think, also, it would be well to secure with a little paste 
 to the first blank leaf the strip of paper with the address on it herewith sent. 
 
 " I think Robertson could also send the work to Mons. Arago, and I send 
 notes to be served in the same way for this great scientific aristocrat, who 
 has never had the good manners to reply to any one communication I ever 
 sent him. Quetelet is an intimate friend and a most excellent fellow, so I 
 must treat him as such ; the other chap I treat on the noli me tangerc prin- 
 ciple." 
 
 Another example of the confidence Harris had in his own know- 
 ledge of his favourite science is afforded at the meeting of the 
 British Association at Plymouth already referred to. He had to 
 give one of the Evening Lectures illustrated by experiments, before 
 the assembled Association, on his own electrical discoveries, and the 
 day was very wet. Several of the leading members condoled with 
 him on the unfavourable state of the weather, and even advised him 
 to put off the lecture. He accepted their condolence, but declined to 
 act on their advice. He knew his own resources, and relied on them. 
 By means of his peculiar heating-irons he surrounded his appa- 
 ratus with a warm dry atmosphere, and the result was a brilliant 
 
EXPERIMENTS ON LIGHTNING CONDUCTORS. XV 
 
 success. He lias described to the editor how anxious he was for 
 success ; for he had, in the physical section, been opposing some 
 of Coulombe's views, and had tried in vain to shake the sectional 
 President's (Dr. Whewell) belief in the French theory. He has 
 told us how quickly his heart beat as he measured off different 
 quantities of electricity by means of his unit jar, lest the disc, 
 counterbalanced by different weights, should not go down at the 
 exact times required to prove his law, that with a constant surface 
 and at a constant distance, the attraction increases as the square of 
 the quantity of electricity. 
 
 Few persons are aware of the long? continued struggle Harris 
 had to undergo to impress upon the public mind the importance 
 of adopting his system of lightning conductors for the ships of the 
 Royal Navy, and few are aware of the varied means used for the 
 purpose. He contributed a number of papers to the Nautical 
 Magazine illustrative of damage by lightning ; he was always on 
 the watch for the slightest scent of a good case ; and he never gave 
 it up until he had tracked it to the ship's log deposited in Somerset 
 House, or obtained an account from the captain or one of the 
 officers of the ship that had been struck. He embodied these 
 cases in letters and pamphlets, which he circulated among Mem- 
 bers of Parliament and various persons in authority, including 
 the foreign ambassadors ; and it may be mentioned to the 
 honour of the Emperor of Russia, that Harris's system was 
 adopted in the Russian navy before it was fully admitted into our 
 own. In 1845 the Emperor presented Harris with a valuable 
 ring and a superb vase, in acknowledgment of the merits of his 
 system. Harris also instituted a series of experiments, on a large 
 scale, in Plymouth Sound, showing that he could direct the dis- 
 charge to any point of the vessel, or to the sea, at pleasure ; and he 
 made the points of discharge evident by firing gunpowder. These 
 experiments attracted public notice, and crowds assembled on the 
 Hoe to witness them. This led to a ludicrous circumstance, which 
 Harris related to the editor with great glee. One evening an 
 old woman was passing along the Hoe, when a man called her 
 attention to summer lightning that was flashing in the horizon. 
 " Don't talk to me about summer lightning," remonstrated the 
 incredulous dame ; " it is that Dr. Harris playing some of his 
 tricks. If he doesn't take care, he will play them once too often." 
 
XVI HARRIS IS KNIGHTED. 
 
 Harris not only interested himself in protecting ships from 
 lightning, but he also endeavoured to get his system applied to 
 public buildings. He drew up a long list of buildings that had 
 been damaged, not nearly so full and complete as his list of ships,* 
 but still a formidable indictment against folly and prejudice. He 
 even addressed a memorial to a Church Building Society, pointing 
 out the necessity of protecting every new church that was built. 
 He invited the editor to accompany him to hear the verbal reply, 
 which was to this effect, that the cost of fitting conductors to a 
 church viz., from 60 to 100 was a fatal objection ; for in many 
 cases this additional charge in the estimates would most likely 
 turn the scale against the church being built at all. 
 
 At length all difficulties in the way of his long-cherished object 
 were removed or overcome. All the various objections to his con- 
 ductors had been met : persons in authority had declared that 
 letting the copper bands into the masts weakened them ; but 
 Harris proved experimentally that their powers of resistance to 
 flexure were increased. The flagstaff of a ship, placed on the top 
 of the mast above the point where the conductors began, had 
 been struck by lightning and shivered to pieces. This and similar 
 slight accidents, which really proved the efficiency of the system, 
 at length were so clearly understood, that no further doubt 
 remained. It was felt that some public recognition was due to 
 the man who had made our ships safe from the attacks of a 
 destructive foe which had formerly deprived the country of the 
 full services of its navy, killed or crippled its sailors, and wasted 
 many thousands annually of the public money. In 1847 the 
 honour of knighthood was conferred on Snow Harris at the express 
 command of her Majesty the Queen, in consideration of his " very 
 useful inventions," to use the words of Earl Russell. So little was 
 this honour expected, that when Earl Russell's letter arrived at 
 Plymouth, Harris thought it was a hoax ; for he, in common 
 with all men of genius, had a strong sense of humour (without 
 which, indeed, genius seems to be scarcely complete, unless power 
 take its place, as in the case of Milton and Dante), and this 
 humour was so often let loose upon his friends in good-natured 
 jokes (often practical ones), that no wonder if he were sometimes 
 repaid in his own coin. He took the letter to a gentleman in 
 * Ordered to be printed by the House of Commons, 1854. 
 
SUCCESS OF LIGHTNING CONDUCTORS. XYil 
 
 Plymouth, and said, " Have you not a collection of autographs, 
 including that of Lord John Russell?" The autograph in ques- 
 tion was produced. He examined it carefully, and said, " No, it is 
 no hoax ; the writing in my note is identical with that in yours." 
 But even then he consulted with his friends as to whether he had 
 not better ask leave to decline the honour ; but he was nervously 
 anxious not to appear in the slightest degree to oppose himself to 
 her Majesty's gracious wish. He accordingly came up to town 
 and received the well-merited honour. The editor called on him 
 next day to congratulate him, when he gave an amusing de- 
 scription of his own feelings on finding himself for the first time 
 in a court-dress. He expressed himself extremely gratified at 
 the gracious manner of the Queen, and the smile of recognition 
 which he received from the Prince Consort. Indeed, the confidence 
 of her Majesty and the Prince in the perfect safety of Harris's 
 conductors was shown in a request that he would fit up conductors 
 at Buckingham Palace and at Osborne, and also similarly protect 
 her Majesty's yacht. Harris was also employed some years later 
 to design a complete system of conductors for the palace at West- 
 minster. His written instructions, which are full of interest, were 
 ordered by the House of Commons to be printed, and they will be 
 found under the head of " Estimates, &c., Civil Services, for the 
 year ending March 31, 1856." 
 
 There was now no difficulty in the way of admitting Harris's 
 conductors into the Royal Navy, and the Government, in con- 
 sideration of his great -services, proposed a vote of 5,000 to the 
 inventor. Sir James Graham, in moving the vote, said that he 
 never voted away money with more pleasure. 
 
 In 1850 Sir William was elected an honorary member of the 
 Naval Club at Plymouth, when a number of eminent officers 
 warmly congratulated him on the great service he had rendered 
 to the Navy. In 1854 he was elected an honorary member of the 
 Royal Yacht Squadron at Cowes, by the consent of all the mem- 
 bers, in acknowledgment of his public services. Nothing could be 
 more congenial to Sir William's tastes, for he was always more of 
 a sailor than a landsman. His very walk reminded you of the 
 deck of a ship, and the warmth and simplicity of his character of 
 a sailor. It was quite a treat to be with Harris near or on 
 
 b 
 
XV111 CHARACTER AS A SAILOR. 
 
 the sea. There was not a craft that he was ignorant of, and 
 he was never tired of pointing out the merits and defects of 
 the vessels around him. He loved to be on the sea, in what- 
 ever craft. The editor once accompanied him to the Eddystone 
 in the lighthouse tender, and on another occasion to Cornwall in a 
 limestone barge. He had invented a new form of ship's compass 
 in which he took great interest, and was proud to see it in use. 
 Being told that a yacht had arrived in the Sound with one of his 
 compasses on board, he asked the editor to go with him to inquire 
 how the yachtsman liked it. As soon as he got on board and sent 
 in his message, the proprietor came up and said, " Oh ! I don't 
 like your compass at all ; but I have one here by a man named 
 Harris that is a great favourite of mine." " I'm Harris," was the 
 bursting, eager reply, whereupon apologies and warm congratu- 
 lations ensued. 
 
 Sir William had long had a yacht of his own, in the manage- 
 ment and sailing of which he took the greatest delight. He was 
 proud of his nautical skill, and was pleased when some one told 
 him that a naval man once observing a sailing-boat tacking about 
 in the Sound, exclaimed, " Egad, the fellow in that boat well 
 knows what he's about ! " It need hardly be said that the fellow 
 was Harris himself. He had many other characteristics of a sailor, 
 but there was one point in which he did not resemble Jack Tar : 
 that is, in his indifference to dancing, although, as we have 
 said, he was an accomplished musician. Sir William gratified 
 his love of inusic in various ways. He had been his children's 
 tutor in music as in other matters, and the concerts got up in 
 his drawing-room were almost professional in their finish.* 
 He was often in London, where, as may be supposed, he had frequent 
 opportunities of hearing good music ; but, in common with the 
 best minds, he preferred cheap to costly pleasures. He delighted 
 to go to the pit of a minor theatre when operas were performed. 
 He was fond of a good play well acted, and reckoned among his 
 intimate friends Mr. and Mrs. Charles Kean. 
 
 Sir William was fond of conversation, and excelled in it. His 
 sympathies were wide, and he was ready to talk or to listen on 
 any subject. Captain Lockyer once said of him, " Harris is like 
 
 * See note page xxi. 
 
CHARACTER AS A WRITER. XIX 
 
 a barrel organ ; you may set him to any tune." He was also a 
 man of tender feeling, and many will cherish his memory as of 
 one more than a friend, for he was their doctor and nurse. 
 
 In the midst of all his exertions in the cause of lightning con- 
 ductors, Sir "William did not neglect science. When Mr. Weale, 
 the publisher, in the year 1847 was about to start his Rudi- 
 mentary Series of Scientific Manuals, he obtained the assistance 
 of Sir William, who wrote several treatises on electricity, 
 magnetism, and galvanism, which met with great success and 
 passed through several editions. Sir William's style was clear, 
 because he thought clearly ;. but it was modelled on that of a 
 former age, when one scientific man seldom named another 
 without giving him all his titles, with "justly celebrated" in 
 addition. He was very fastidious as to the way in which 
 his thoughts were stated ;, he would write and re- write a page 
 many times, and when at last it reached him in proof he would 
 so twist it and re-arrange it, that the printer's charge for cor- 
 rections was sometimes higher than that for composing. Sir 
 William would even make organic changes in wood engravings, 
 so that if adopted it would have been necessary to have new 
 blocks. But in all such cases, whether of writing or engraving, 
 he was willing to listen to remonstrance ; and indeed sometimes 
 felt it to be a relief to have some one to decide for him. The 
 editor frequently hesitated at this responsibility, and was care- 
 ful how he advised, lest his advice should be too readily 
 adopted and afterwards repented of. Harris was so grateful for 
 any little services of this kind, that it was a pleasant though a 
 responsible duty to serve him. And then he was so trusting and 
 confiding, so child-like and affectionate to any one whom he 
 loved, that he concealed nothing, not even his own weaknesses, 
 for he had his weaknesses, and had he not he would not have 
 been so lovable a character. He was sensitive as to what 
 others thought or said of him, and sometimes fancied they did 
 not do him justice, or that their discoveries had borrowed too 
 much from his. And when he resumed his labours in the field 
 of original research, and found that his papers were not so 
 welcome in the Royal Society as they had been thirty years 
 before, he could not understand the reasons for their rejection. 
 
XX SCIENTIFIC CHARACTER. 
 
 He could not admit the fact, still less receive it with calmness 
 (few men can), that his working days were nearly over ; and 
 that his later work was but a reflection of the light that once 
 shone so brightly. His sympathies were with the Bennetts, the 
 Cavendishes, the Singers, the Voltas of a past age. Frictional 
 electricity was his forte and the source of his triumphs. He was 
 too good a Tory of the old school of science to recognise the broad 
 and sweeping advance of the new ; and he did not feel that he 
 was behind his age when, in 1861, he presented to the Royal 
 Society an elaborate paper on an improved form of Bennett's 
 discharger, and yet later when, in 1864, he discussed the laws of 
 electrical distribution, and still relied upon the Ley den jar and 
 the unit measure. 
 
 It cannot be said that Harris, with all his ability, had a creative 
 mind, in the sense that Wollaston's or Davy's was creative. He 
 was highly ingenious and inventive, and in his best work one is 
 always struck with the advance he had made beyond previous 
 occupiers of the same ground. He did his work well, and left 
 his mark 011 the science of his day ; and although some of his 
 labours will be forgotten, and others be absorbed and blended 
 with the branch of physical science that he cultivated, still there 
 are many points in Harris's character as a man, and in his habits 
 as a philosopher, that the student may dwell upon with pleasure 
 and profit. 
 
 The few remaining years of Sir William's life were embittered 
 by bodily suffering. In August 1861, after returning from a trip 
 in his yacht, he was seized with an attack of iritis which did not 
 yield to medical treatment during some months. He was confined 
 to the house until the following May. In the autumn of 1862 the 
 attack returned, and Sir William was advised to undergo two 
 painful operations ; but the result, after all, was the loss of one 
 eye and the partial loss of the other. His general health, too, 
 suffered from the medical treatment. 
 
 When he had partially recovered, he was anxious to bring out 
 his complete treatise on Frictional Electricity, materials for 
 which had been accumulating during many years. He intended 
 to add to this work short biographical notices of the leading 
 electricians. He had also prepared a minute account of the history 
 
ILLNESS A:ND DEATH. xxi 
 
 of the Leyden jar. The first few chapters of the work now before 
 the reader were sent to the editor early in 1866. There was con- 
 siderable delay in getting the woodcuts properly executed, so that 
 the work made but slow progress. The first part, however, up to 
 p. 205 of the present volume, was completed under the author's 
 supervision, when he was seized with his last illness, which ended 
 fatally on the evening of the 22nd January, 1867. " He bore his 
 sad calamity during five years and a half with the greatest pa- 
 tience, calmness, and fortitude, and was never heard to murmur," 
 were the words of one who announced to the writer of this notice 
 the loss of her husband and his friend. 
 
 [The following note, referred to at p. xviii., has been contributed by a 
 friend of the family, who took part in the concerts referred to.] 
 
 Sir William was a musician of rare ability and attainments ; indeed, few 
 amateurs possessed such critical acumen or so comprehensive an acquaint- 
 ance with music, scientifically and practically considered, as he did ; for in 
 addition to his extensive study of theory, he was a performer both on the 
 harp and piano, and played with exquisite taste and feeling. With thes3 
 qualifications, Sir William devoted much time to the musical education of 
 his daughters, who became accomplished pianists. 
 
 Provided with two of Collard's grand pianofortes, Sir William was fond of 
 collecting a few " strings," and with them enjoying the luxury of concerted 
 music. These parties, small at first, soon swelled into complete orchestras 
 formed of amateurs, to whose aid in cheerful spirit came professors, civil 
 and military, the latter bringing from their bands the necessary wind ac- 
 companiments. With such a force Sir William would take the field, and 
 fen red not to encounter the most powerful of composers. The names of 
 Beethoven, Weber, Mendelssohn, Mozart, and others, were familiar as 
 household words in his home at Windsor Villas. 
 
 Nor was this all, for our enthusiast added to his instrumental productions 
 all the vocal music of which the singers in his neighbourhood wore capable. 
 He invited all amateurs found to possess the requisite voices to assist him, 
 and the result was a succession of splendid concerts, not often surpassed in 
 private circles. 
 
 To perfect these reunions no efforts were spared ; consecutive portions of 
 operas were performed, and descriptive programmes from the ready pen of 
 Sir William were issued to the company, who were thus made familiar with 
 the subject of the pieces produced ; to this was added a brief sort of lecture 
 or synopsis of the opera from which the music was selected, and, finally, 
 small scenes were used to bring the more striking incidents of the opera 
 vividly before the audience. 
 
 In the year 1851 a wider field seemed to open for the exercise of Sir 
 William's powers of musical organization, and he accordingly took the lead 
 in a committee of management on the formation of a philharmonic society 
 which included subscribers from Devon and Cornwall, who were invited to 
 
XX11 MUSICAL ATTAINMENTS. 
 
 participate in or enjoy the really good music intended to be performed for 
 their gratification. 
 
 In this undertaking Sir William was indefatigable, and for two or three 
 years excellent concerts were produced, aided by eminent singers from 
 London, amateurs in the neighbourhood, and an orchestra and choir num- 
 bering 150 persons. 
 
 The frequent change amongst the inhabitants caused the maintenance of 
 this society to become a work of such labour, that the promoters, sanguine 
 as they were, could not continue their efforts, and Sir William was obliged 
 to fall back on his delightful home concerts. 
 
 In addition to his high appreciation of music, Sir William had a great 
 admiration for the drama. He was proud to reckon Mr. and Mrs. Charles 
 Kean amongst his most intimate friends, and often spoke of his acquaintance 
 v\dth the late Mr. Matthews, whom he professionally attended towards the 
 close of his life. 
 
THE AUTHOR'S PREFACE. 
 
 PERSONS who have been trained more immediately in what may be 
 called the mathematics of electrical physics look generally with 
 much indifference and suspicion at any new experimental processes 
 or tangible investigations which at all run counter to their long- 
 admitted theory. E/eposing with a reverential confidence upon a 
 few early experimental deductions, conjoined with analytical forms 
 of expression, they appear to imagine that the limit of our know- 
 ledge of the laws of electrical action has been reached, and that to 
 doubt in any degree the truth of what has been termed the 
 Coulombian theory of electricity, and the received laws of the 
 necessary distribution of electricity upon bodies of variable figure, 
 is little less than philosophical heresy. In this sense many pro- 
 found writers, distinguished for analytical skill, betray an amount 
 of prejudice not very favourable to the advancement of science : 
 they judge of every theory by the quantum of symbolic expression 
 which enters into it, or by the form under which it appears. 
 Thus M. Biot, in his life of Yolta,* seeks continually to disparage 
 the quality of mind of that truly great man, upon the ground that 
 his researches and discoveries had not a rigorous mathematical 
 basis similar to the method of Coulombe, and is hence led, although 
 constrained to admit the vast genius of Yolta, to refer nearly all 
 Yolta effected to previous principles mathematically deduced by 
 -ZEpinus, although no one ever heard of the condenser or the 
 electrophorus before Yolta. Still, we are informed that the prin- 
 ciples were foreseen, and their theory given twenty years before 
 in the "Tentamen Theoriae Electricitatis, " by .ZEpinus, where 
 most likely they would have remained dormant to this day if somo 
 great genius in experimental physics had not invented the instru- 
 ments. The real fact is, that these beautiful instruments having 
 
 * Hioyraphie Universette, tome xliv. 
 
XXIV THE AUTHOR S PREFACE. 
 
 been invented and perfected, the theory of JEpinus is applied in 
 explanation of them. Many other theories of electricity could be 
 equally well applied, and it would be just as easy to say that 
 Yolta's instruments had been prophesied in those theories as in 
 the theory of JEpinus. If the principles had been so clearly 
 expounded by -ZEpinus twenty years before Yolta invented the 
 instruments, it seems a sad reflection on those who so strongly 
 contend for mathematical rigour not to have applied them. 
 
 The real state of the case is, that the events virtually preceded 
 the prophecy, and having taken place, the prophecy was applied to 
 them. M. Biot, evidently biassed by his veneration for a mathe- 
 matical form of inquiry, and which he calls la rigueur mathematique, 
 treats Yolta's physics of electricity very slightingly, and so far 
 betrays a great want of appreciation of what Yolta advanced, if 
 not a failure of comprehension of Yolta's principles. Thus he 
 says, Yolta's discoveries of the condenser and electrophorus were 
 for him merely combinations of experience ; that he never referred 
 them to their true theory, but to electrical atmospheres, an idea 
 which the most profound geometers could never dissuade him 
 from ; that Yolta's dissertation upon electrical conductors betrays 
 a total absence of all abstract rigour of research ; that he does not 
 fix any of the rigorous elements of the important question of the 
 influence of the general form of conductors upon the conservation 
 or loss of electricity, or upon the energy of their discharges, whilst 
 the rigorous method of Coulombe fixes, and for ever, the exact laws 
 of electrical distribution and equilibrium upon the surface of bodies 
 of different forms. 
 
 The substantiality of all this kind of criticism is very doubtful. 
 It is possible that the genius of Yolta did not admit of his employing 
 an analytical process similar to that pursued by Coulombe in his 
 experimental inquiries. Yet it does not follow, on that account, 
 that it did not equally involve physico-mathematical accuracy 
 under other forms ; and really we have yet to learn the truth of 
 the assertion that his notions of what he calls electrical atmo- 
 spheres, taken in the sense in which he uses this term, are not per- 
 fectly correct, and demonstrable by experiment, or that his views 
 of the variable intensity of a given quantity of electricity, disposed 
 upon conductors cf equal surface, but of different extension, are 
 
THE AUTHOR S PREFACE. XXV 
 
 not quite sound. They may not coincide with a given mathema- 
 tical theory of electricity, but they need not be " vague " on that 
 account. Neither is it so certain that the course pursued by 
 M. Coulombe has really "fixed, and for ever," the exact laws of 
 such variations of electrical force. The experiments upon which 
 Coulombe rests are really very few, and not altogether unexcep- 
 tionable. The theory of the Proof Plane, upon which depends all 
 the assumed laws of electrical distribution, is extremely doubtful, 
 and its operation very precarious, and in many instances absolutely 
 fallacious ; and although a theory of repulsive force, assumed to 
 exist in the particles of a supposed electrical fluid, may be brought 
 to square with a mathematical and analytical application of the 
 observed effects, however fallacious the interpretation, yet that 
 is not evidence of so conclusive a nature as to preclude the 
 reception of other equally general views of electrical force. It 
 would be, p3rhaps, both an invidious and useless course to seek to 
 oppose to each other the relative merits of simple and pure experi- 
 mental researches, and of investigations made through the medium 
 of symbolical and abstract mathematical analysis. It is, after all, a 
 liberal combination of the two methods which gives us the most 
 powerful means of advancing physical knowledge. That rigorous 
 mathematical forms and methods of experiment and deduction are 
 most desirable in physical inquiries, few can doubt. The mathe- 
 matical mind can doubtless see further into futurity than other 
 minds not so constituted > and can thereby. save the great loss of 
 time attendant on mere tentative skill ; whilst the great beauty 
 and elegance of its analytical processes, and its wondrous develop- 
 ments, lend an increased charm to experimental labours. Never- 
 theless, we must never forget that it is the experimental labours 
 upon which all our knowledge of nature mainly reposes. Newton 
 himself would not consent to hazard his grand theory of universal 
 gravitation upon any other basis, however near its probability 
 approach to certainty, and, as is well known, withheld it from the 
 philosophical world until correct measurements of the earth had 
 been obtained. The great and justly- celebrated Galileo, who was 
 much more remarkable for a profound geometry than for any skill 
 in symbolical analysis, never failed to verify every assertion by 
 experiment. We have, in treating questions of this kind, to 
 
XXVI 
 
 distinguish (1) that class of mind adapted to mere mathematics, 
 considered as an instrument of research ; (2) those who pos- 
 sess great skill in the handling and management of symbols and 
 of general symbolical processes, or geometrical analysis, taken 
 abstractedly ; and (3) those who, with less analytical skill, pursue 
 inventive experimental research solely ; or (4) those who, without 
 the aid of any mathematical processes, repose entirely upon 
 experimental investigation, still possessing intuitively, as it were, 
 a peculiar quality of mind involving really mathematical rigour 
 under another form. There may undoubtedly exist great and 
 innate powers of thought and philosophical intellect without any 
 of that quality we are accustomed to designate as mathematical. 
 Bacon, for example, one of the most wonderful and accurate 
 thinkers of any age, was totally ignorant of geometry and mathe- 
 matics. All the great discoveries and advances in Physics have, 
 for the most part, been derived from men essentially experi- 
 mentalists, men who either applied great and original powers of 
 thought in the investigations of nature in their own way, without 
 the aid of mathematical processes, or who allowed mathematics to 
 follow, as it were, in the wake of experiments, to correct and assist 
 them, rather than to develop in advance new and, as yet, undis- 
 covered facts. Franklin, whose researches and discoveries in 
 ordinary electricity were most marked and important, had very 
 little skill in mathematics. Neither Volta, the inventor of impor- 
 tant electrical apparatus, nor Davy, the great discoverer of the 
 bases of the fixed alkalies, can be said to have been led to such 
 discoveries through any mathematical process. The same may be 
 said of a host of others. To judge, therefore, of the value of any 
 physical inquiry solely with reference to its mathematical method, 
 or the amount of analytical symbolic representation it may contain, 
 would be not only to apply a fallacious measure of value, but would 
 be really a philosophical injustice. Mathematical or symbolical 
 "analysis is, after all, but an instrument of research of a peculiar 
 kind, undoubtedly most valuable in its operation, but still an 
 instrument which many minds do not require, and which is often 
 very greatly abused and overworked by persons who, conscious of 
 their knowledge of symbolical arithmetic, are rather content to 
 exemplify processes than to investigate truth. Yery little, if any, 
 
THE AUTHOR S PREFACE. XXV11 
 
 really useful knowledge of nature is found in the elaborate and 
 interminable pages of symbolic analysis in which many modern 
 philosophical papers abound. As specimens of mere analytical 
 skill they are no doubt valuable, but for any practical result they 
 are frequently valueless. It is the show of learning without the 
 reality. Whilst, therefore, we admit the vast importance of 
 mathematical knowledge and address, and the great advantage of 
 mathematical method in its sound and healthy application to 
 experimental physics, we still think that in considering the exhibi- 
 tion of analytical skill as the great object of philosophy, we sacrifice 
 the end to the means, and thereby fail to arrive at any new result. 
 We substitute a sort of paper philosophy based upon data which 
 are rather taken for granted than definitively proved for the 
 philosophy of nature, and really close the door upon further 
 inquiry. Things are taken for granted upon authority much in 
 the way of the followers of Aristotle, and the result is, that it is 
 difficult to obtain anything like candid consideration of facts which 
 do not fall in with what is denominated par excellence THE THEORY. 
 It is well known that, when Galileo asserted that bodies falling to 
 the earth simultaneously from the same altitude would all fall in 
 the same time, whatever their weights might be, the Aristotelians 
 refused to listen to such a heresy, although, in his famous experi- 
 ment from the tower of Pisa, the ring of the falling bodies on the 
 ground was still sounding in their ears. 
 
ANALYSIS OP THE CONTEXTS. 
 
 THE EDITOR'S PREFACE , v 
 
 LIFE OF THE AUTHOR, BY THE EDITOR .-*-.. . . . . vii 
 
 THE AUTHOR'S PREFACE . . , xxiii 
 
 ELECTRICITY IIS" THEORY AND PRACTICE. 
 PART I. 
 
 ELEMENTARY ELECTRICAL PHENOMENA. 
 CHAPTER I. ATTRACTION AND REPULSION. 
 
 ATTRACTION, various kinds of, 1 ; development of electrical force early knowledge 
 of electricity, 2 ; terms Tised in electricity, 3 ; modes of showing electrical 
 action, 3 ; electroscopes trial reed, 4 ; simple experiments on electrical excita- 
 tion, 5 ; list of idio-electrics, 6 ; Gray's discovery, excitation and attraction, 
 insulation and conduction, 7 ; list of electrical conductors, 9 ; insulation, 9 ; 
 carriers and transfer-planes, 10 ; limit of the terms insulation and conduction, 
 10 ; list of electrics or insulators, intermediate substances, and conductors or 
 non-electrics, 11 ; remarks on this table, 11 ; experiments on attraction and 
 repulsion, 12; law of attraction and repulsion, 15 ; vitreous and resinous elec- 
 tricities, 15; illustrative experiments, 16; positive and negative, or plus and 
 minus electricity, 17 ; origin of the two electricities, 18 ; further illustrations of 
 attraction and repulsion, 19 ; effects of varying the surface, 19 ; determination 
 of the kind of electricity, 21 ; single gold leaf electroscope, 21 ; double ditto, 
 22 ; Bennett's gold leaf electrometer, 23 ; Henley's quadrant electrometer, 23 ; 
 Cavendish's electrometer, 23 ; Harris's electroscope electrometer, 24 ; Harris's 
 electroscope electrometer of double repulsion, 25. 
 
 OCCASIONAL MEMORANDA AND EXPLANATORY NOTES TO CHAPTER I. 
 
 Note A. Suspensatory filaments of silk and spider's web, 27. 
 ,, B. Construction of electroscope needle, 27. 
 C. Management of leaf-gold, 27. 
 D. Preparation of amalgam, 27. 
 E. Construction of sliding insulator, 28. 
 ,, F. Preparation of glass and gutta-percha tubes, 28. 
 ,, G. Preparation of gold-leaf electroscope, 28. 
 H. Preparation of a portable diverging electroscope, 29. 
 I. Preparation of Henley's electrometer, 29. Value of the indications afforded 
 
 by this instrument, 30. 
 ,, K. Mode of using Cavendish's electrometer, 30. 
 
XXX ANALYSIS OF THE CONTENTS. 
 
 CHAPTER IT. ON INDUCTION. 
 
 INDUCTION Electric-it e par influence Electricitt* dissimule'e, 31 ; experimental illustra- 
 tions of induction, 31; direct and reflected induction inductive power, 33; 
 preliminary process in electrical attraction and repulsion, 34 ; experimental 
 illustrations, 34 ; induction and attraction apparatus, 35 ; further illustrations, 
 35 ; mode of propagation at a distance, 37 ; Faraday's nomenclature, 38 ; 
 polarisation, 38 ; action of contiguous particles in the dielectric, 39 ; action 
 traced by Cavendish and Faraday from an electrified body to the walls of the 
 room, 39 ; action of a second body near the charged body, 39 ; further illustra- 
 tion, 40 ; the two electricities always present, 40 ; experiment with ball and 
 vase, 40, 41 ; experiment with concentric vases, 42 ; distribution of electricity 
 on insulated surfaces, 44 ; experiments by Coulombe, Saussure, and Cavendish, 
 45 ; experiment with movable hemispheres, 46 ; Biot's notice of such an ex 
 periment, 47 ; Volta's experiment on the tendency of electricity to the surface, 
 48 ; Franklin's " can and chain " experiment, 48 ; Faraday's insulated muslin 
 bag, 49 ; Faraday's insulated cubical electric chamber, 49 ; difference between 
 insulation, conduction, and discharge, 50 ; tension, 51 ; striking distance, 51 ; 
 Franklin on the action of points, 51 ; action of points explained by modern 
 theory, 52 ; conditions of discharge, 53 ; nature of electrical force as explained 
 by induction, 53 ; Faraday's view of electricity as a force, 54 ; conduction and 
 insulation, 54 ; induction, 55 ; intensity and tension, 55 ; mutual action of in- 
 sulated conductors, 55 ; what is electricity ? 57 ; tension and intensity denned, 
 58 ; specific induction, 58 ; experimental illustrations, 59 61. 
 
 EXPLANATORY NOTES TO CHAPTER II. 
 
 Note L. Insulated carrier and gold-leaf electroscopes, 63. 
 M. Sliding insulated platforms, 63. 
 
 N. Construction of induction and attraction apparatus (Fig-. 24 incorrectly 
 referred to in Note as Fig. 22), 63. 
 
 CHAPTER III. ON ELECTRICAL APPARATUS. -INSTRUMENTS OF ACCUMULATION. 
 
 THE electrical machine, 64 ; early forms, 64 ; cylindrical electrical machines, 65 ; the 
 rubber, 66 ; Nairne's electrical machine, 67 ; preparation of the cylinder, 68 ; plate 
 electrical machines, 68 ; Cuthberston's machine, 68 ; defect of, 69 ; Van Marum's 
 machine, 69 ; Harris's improved plate machine, 71; method of mounting plate, 
 &c., 71 ; details of cushion, 72 ; telescopic joint conductor, 73 ; negative conduc- 
 tor, 73 ; details of handle, &c., 74 ; Harris's plate machine for negative electricity, 
 75 ; frame- work, glass plate, 76 ; dimensions of plates, &c., 76 ; effects of a fine 
 plate machine, 76 ; sensation of spider's web, 77 ; Winter's ring conductor, 77 ; 
 ozone, 78 ; sparks, 78 ; equable action of machine, 79 ; experimental illustration , 
 79 ; steam electrical machine, 81 ; the electric column or perpetual electrical 
 machine, 82 ; action of dry pile on electroscopes, 84 ; perpetual or electric 
 chime, 84 ; differential electroscope, 85 ; the electrophorus, 85 ; single electro- 
 phorus, 86 ; compound electrophorus, 87 ; action of electrophorus, 87 ; electro- 
 phorus manipulation, 88 ; the electrophorus on a large scale, 89 ; Phillips's 
 electrophorus, 89 ; electrophorus electrical machine, 89 ; details of, 90 ; the 
 condenser, 91 ; Volta's paradox, 92 ; semi-insulating bodies, 93 ; single and 
 double condenser, 94 ; Cavallo's condenser, 94 ; the electrical jar or Leyden 
 phial, 96 ; the coated pane, 96 ; experimental illustration of, 97 ; free and con- 
 cealed electricity, 98 ; fulminating square, 98 ; the electric shock, 98 ; further 
 experiments with coated pane, 99 ; improved form of Leyden jar, 100 ; spon- 
 taneous discharge, 101 ; circuit, 102 ; discharging- rod, 102 ; the two electricities 
 obtained from a charged jar, 102 ; electrical battery, 103. 
 
ANALYSIS OF THE CONTENTS. XXXI 
 
 NOTES TO CHAPTER III. 
 
 Note 0. Franklin's electrical machine, 104. 
 P. Origin of steam electrical machine, 104. 
 Q. Priestley's electrical battery, 105. 
 
 CHAPTER IV. INSTRUMENTS OF QUANTITATIVE MEASURE MORE ESPECIALLY ADAPTED 
 TO ELECTRO-MATHEMATICAL AND PHYSICAL RESEARCH. 
 
 II ARRIS'S unit jar, 106; arranged for charging a Leyden jar, 107; self-discharging 
 Leyden jar and unit jar, 108 ; uniformity of action in the unit jar, 108 ; Lane's 
 electrometer, 109 ; as improved hy Harris, 111 ; method of using, 112 ; as modi- 
 fied for connecting with Leyden jar, 113; details of screw measure, 114; pre- 
 cautions in \ising Lane's electrometer, 114; Cuthbertson's steel-yard electro- 
 meter, 115; method of using, 116; erroneous views of the older electricians 
 respecting tension and intensity, 117 ; Harris's charging and discharging quan- 
 titative electrometer, 117; Harris's quantity jar and transfer measure, 119; 
 positive and negative quantity jars, 120 ; transfer measure, 120 ; practical ap- 
 plication of, 121 ; quantity of electricity on different surfaces in units of charge, 
 121 ; the globular or circular inch, 122; Harris's hydrostatic electrometer, 122; 
 advantages of, 127; method of using, 128; experiments with, 129; Harris's 
 scale-beam electrometer, 130 ; experiments with, 133 ; electrical attraction of 
 spheres, 133 ; Harris's two-threaded or bifilar balance, 135 ; modes of applica- 
 tion, 136; the balance described, 137; formula for oscillating bodies, 138; ve- 
 rification of results, 138 ; tabular results, showing rate of oscillation, with 
 different lengths and distances of threads, 139 ; table, showing the weight in 
 grains required to resist the reactive force of the threads, 140 ; table, showing 
 the weight required to balance the reactive force of the threads at various angles 
 of deflection, 140 ; Harris's improved form of bifilar balance, 141 ; mode of sus- 
 pending needle, 142; graduated circle, 143; experiments with the bifilar balance, 
 143 ; conclusions from these experiments, 145 ; Coulombe's torsion apparatus, 
 145; Coulombe's balance of torsion, 146; method of using, 146; remarks on, 
 148; critical examination of, 149; Coulombe's experiments with, 149; Harris's 
 thermo-electrometer, 150; fusion of wires by the. early electricians, 150; con- 
 ducting power of wires as determined by the thermo-electrometer, 151 ; method 
 of fixing the wire, 152; experiments with thermo-electrometer, 152; thernio- 
 electrometer with two wires, and experiments, 154. 
 
 NOTES TO CHAPTER IV. 
 
 Note R. Elimination of error from small quantity of charge left on transfer plate, 
 or globe, 154. 
 
 S. Wood used for plates, hollow spheres, &c., used with scale-beam electro- 
 meter, 155. 
 
 ,, T. Professor Lloyd's investigation of the reactive force of torsion with bifilar 
 suspension, 155. 
 
 U. On the torsion force of the suspending wire, 156. 
 
 CHAPTER V. ELECTRICAL MANIPULATION. 
 
 COMMON articles required by the practical electrician, 157; heating irons, 158; 
 various other articles, 158, 159, 160 ; white cedar lath-wood, 161 ; gilder's 
 cushion, &c., 162 ; pulverising bowl, 162 ; glass in various forms, mica, resinous 
 substances, silk, dry and baked wood, animal furs and skins, flannel, straw 
 reeds, quills, &c., 163 ; threads, balls of pith, cork, &c., metallic tubes, rings of 
 lead, materials for amalgams, &c., 164 ; mode of preparing electrical balls, 165 ; 
 
XXX11 ANALYSIS OF THE CONTENTS. 
 
 of metal, wood, cork, &c., 16-5, 166 ; method of piercing holes through corks, 
 166; metallic conducting rods, 167; varnishing, varnishes, and cements, 167; 
 lacquers for metallic, surfaces, 168 ; varnish for wood and French polish, 169 ; 
 mode of applying electrical varnishes, 169; varnish cup and brushes, 170; var- 
 nishing glass rods, tuhes, and glass generally, 170; varnishing electrical jars, 
 171; varnishing glass plates, metallic surfaces, and wooden surfaces, 173; 
 varnishing paper, 175; amber varnish, 176; application of varnish and oil to 
 silk, 177 ; electrical cements, 177 ; preparation of electrical plates and cylinders, 
 brimstone cylinders, cones, &c.,|l79 ; brimstone plates, 180 ; cylinders and plates, 
 of resin, wax, &c., 180 ; electrophorus plate, 181 ; junction of plates of glass by 
 cementing, 183 ; colouring matters for cements, 184 ; value of compound plates 
 185; isinglass, glue, and size, 185; Japanese cement and paste, 186; mode of 
 covering wood with tin leaf, 187 ; electrical suspensions, 187 ; Chinese mode 
 of suspending compass needle, 188 ; suspension filaments, 188 ; management 
 and handling of gold-leaf, 189 ; selection of glass for electrical tubes and elec- 
 trical machines, 190 ; precautions required in constructing electrical machines, 
 191 ; cylindrical machine, 191 ; construction of rubber or cushion, 193 ; plate 
 machine, 194 ; preparation of amalgam, 195 ; various recipes for, 196 ; method 
 of preparing amalgam, 197 ; method of applying amalgam, 198 ; aienon musivum, 
 or mosaic gold, 199 ; excitation of cylindrical electrical machine, 200 ; excitation 
 of plate machine, 201 ; preparation of the cushions of rubbers, 203. 
 
 NOTE TO CHAPTER V . 
 Note V. Various kinds of lac, 201. 
 
 PAET II. 
 
 ON THE LAWS OF ELECTRICAL FORCE. 
 
 CHAFER I. BRIEF ENUMERATION OF FACTS AND PHENOMENA OF ELECTRICAL AGTIOX 
 AS DEDUCIBLE FROM AN INVESTIGATION INTO THE THEORETICAL AND PRACTICAL 
 NATURE, OPERATION, AND LAWS OF ELECTRICITY. 
 
 EARLY notions as to the nature of electricity, 207 ; information obtained only by 
 means of experiment, 207 ; first great step in the science by Du Fay, in dis- 
 tinguishing two kinds of electricity, 208 ; Cavendish's researches, as contained 
 in his MS. papers, 208 ; all the charge on the surface, 208 ; Robison's and Cou- 
 lombe's researches, 208 ; law of electrical force as deduced by Coulombe and 
 others, 209 ; attempt to define the law more clearly, 209 ; speculations thereon, 
 210; how the force varies, 210; physical conditions under which the forces of 
 attraction and repulsion exist, 211 ; objections to the use of the tangent plane, 
 211 ; effect of induced action, 212 ; effect of varying the form of a charged con- 
 ductor, 213 ; experiment with hollow globe and coatings of mercury, 214 ; action 
 on carrier ball, 214 ; influence of change of form on electrical tension, 215 ; ideas 
 of the early experimentalists as to the law of magnetic force, 216 ; influence of 
 reflected induction, 217; limit of the law of electrical or magnetic force, 217 ; 
 influence of opposed spheres, 218 ; objections to the measurement of the force 
 by means of repulsion, 218 ; operation of torsion balance, 219. 
 
 NOTES TO CHAPTER I. 
 
 Note A. On the author's views of the Coulornbian theory, 219. 
 B. Details of experiment on hollow glass globe with coatings of mercury, 221. 
 
ANALYSIS OF THE CONTEXTS. XXX1H 
 
 CHAPTER II. ON THE NATURE OF THE ELECTRIC FORGE. 
 
 NOTIONS of the early electricians on an electrical fluid, 226; objections to the us; 
 of the term fluid, 227 ; want of experiments on electricity considered as a force, 
 228 ; mathematical investigations refer chiefly to the distribution of electricity 
 on surfaces, 228 ; importance of obtaining data with respect to number, weight, 
 and measure, 229 ; the author's instruments contrived with that view, 229 ; 
 leading features of electrical action, modes of interpreting plus and minus, elec- 
 trical action at a distance differs from gravity, induction precedes attraction, 
 230 ; reflected induction or successive influences, 231 ; law of force between two 
 bodies, 232 ; discordant results by different observers, 232 ; reflected magnetic 
 action, 232 ; various cases of this action, 234 ; general conclusion, 235. 
 
 NOTE TO CHAPTER II. 
 
 Note C. Geometrical representation of the distribution of force in the charge anl 
 discharge of the electrical jar, 235. 
 
 CHAPTER III. FURTHER INQUIRIES CONCERNING THE LAWS AND OPERATIONS 
 OF ELECTRICAL FORGE. 
 
 DEFINITIONS of quantity of electricity and intensity with reference to electrical forco, 
 238 ; increased capacity of conductors when the surface is extended in length, 
 239 ; Volta's experiment, 239 ; mutual action of charged spheres at varying dis- 
 tances, 239 ; action between charged and neutral surfaces at various distances, 
 240 ; relation of quantity to surface, 241 ; quantity received by plain rectan- 
 gular surface depending on surface, and perimeter, 241 ; electrical boundary 
 for different surfaces, 242 ; law of variation of charge, as boundary and surface 
 vary, 242 ; electrical intensity of plane rectangular surfaces, 242 ; effect of 
 dividing the surface into equal parts, 243 ; law of variation, 245 ; quantity on 
 a circular plate, 246 ; effect of contact of two circles, two spheres, two rectan- 
 gular plates, two cylinders, &c., 247 ; two spheres of unequal surface, 247 ; 
 charges of spheres, 248 ; electrical boundary of the figure with relation to sur- 
 rounding matter, 248 ; hollow cylinders and globes and other variations of 
 surface, 249 ; charge of a cube, compared with that of a globe, 250 ; intensity 
 of closed and open surfaces, 250 ; formulae used in deducing the laws of statical 
 electrical force, 251 ; calculation of circular or globular measures from the laws 
 of charge for circles and globes, 252 ; instruments used by the author, 252 ; 
 transfer measures, globes and plates, suspension-loop, vulcanite-rod, &c., 253 ; 
 quantity jar, and measure of circular or globular inches, 254 ; law of attraction 
 between plates, 255 ; modes of manipulation, 255 ; experimental illustrations on 
 the attraction of plates, 256 ; electrometer test experiments, 257 ; laws of surface 
 and boundary as regards rectangular plane surfaces, 259 ; charge as determined 
 by intensity, 259 ; charge and intensity of square plates, circular plates, spheres, 
 and closed and open surfaces, 263 ; comparison of plate and sphere, 266 ; of plate 
 and cylinder, 267 ; of cube and sphere, 268 ; conclusion, 268. 
 
XXXIV ANALYSIS OF THE CONTENTS. 
 
 TWO LECTUEES ON ATMOSPHEKIC ELEOTEICITY AND 
 PEOTECTION FEOM LIGHTNING. 
 
 FIRST LECTURE. 
 
 ORIGIN of lightning conductors, 272 ; Franklin's hypothesis, 272 ; erroneous views 
 and practices respecting a thunder-storm, 272 ; rough estimate of damage from 
 lightning to ships, buildings, gunpowder magazines, &c., 273 ; case of the TMsbe, 
 274 ; of the Zowestofe, 275 ; case of St Michael's Church, Black Rock, Cork, 275 ; 
 electricity a force, 275 ; to he studied by its effects like gravitation, 276 ; com- 
 pared with a bent bow, and distinguished as positive and negative force, 276 ; 
 further illustration of modes of force, 277 ; action at a distance, or induction, 
 277 ; the electrical machine and Leyden jar, 278 ; conductors and insulators, 278 ; 
 vitreous and resinous electricity, 278 ; electrical attraction and repulsion, 279 ; 
 general law of electrical action, 280 ; electrical charge and law of attraction, 
 280 ; action of points, 281 ; attractive force equal between disks of different 
 material, 281 ; importance of induction, 281 ; erroneous views respecting attrac- 
 tion, 282 ; experiments on spark discharge, 282 ; effect of points and continuous 
 conductors, 283 ; recapitulation, 284. 
 
 SECOND LECTURE. 
 
 GENERAL conditions of a thunder-storm, 285 ; induction between the clouds and the 
 earth, and condition of the dielectric air, 286 ; polarisation of particles and dis- 
 ruptive discharge, 286 ; experimental illustrations, 286 ; line of least resistance, 
 287 ; position of a building or of a ship with reference to the discharge, 287 ; 
 whirlwinds, waterspouts, and comazants electrical phenomena, 287 ; Frank- 
 lin's experiment with the kite, electrical bells, 288 ; experiment to illustrate 
 whirlwinds, 288 ; lightning and thunder described, 289 ; striking distance, -289 ; 
 law of discharge, 290 ; line of least resistance, 290 ; electrical feelers sent out to 
 mark line of least resistance, 291. , 
 
L I B R A R Y 
 
 UNIVEKSITY OF 
 
 CALIFORNIA.^ 
 
 ELECTRICITY 
 
 PAET I. 
 
 ELEMENTARY ELECTRICAL PHENOMENA. 
 
 CHAPTER I. 
 
 ATTRACTION AND REPULSION. 
 
 1. A multitude of varying natural phenomena, continually 
 present, indicates the operation of certain invisible subtle agencies, 
 or physical forces, by which the material universe is apparently 
 controlled. These physical forces or powers, although differing 
 in their modes and forms of action, are still characterised by one 
 remarkable property common to them all, in a greater or less 
 degree. This property has been designated by the general term 
 attraction. That is, the agencies or forces, whatever they be, 
 cause particles of ordinary matter, both at finite and at indefinite 
 distances, to approach each other, or tend to approach each other. 
 We perceive this in the falling of a stone towards the ground ; in 
 the planetary motions ; in the aggregation, or growing together, 
 as it were, of minute particles into solid and compact masses ; in 
 the approach of ferruginous matter towards a certain ore of iron 
 termed the loadstone or magnet ; as also in the approach of light 
 substances towards certain bodies when subjected to friction ; and 
 in some other remarkable instances. In these phenomena we 
 recognise the operation of one or more subtle natural agencies ; and 
 although careful not to regard such phenomena except as effects, 
 
Z ELECTRICITY AMONG THE ANCIENTS. 
 
 we nevertheless usually designate them by such terms as gravity, 
 magnetic attraction, electrical attraction, and not unfrequently refer 
 to them as so many distinct and operative causes, or independent 
 powers of nature. In treating, however, of one or all of these 
 apparent forces, we do not pretend to deal with occult causes. We 
 concern ourselves only with such agencies considered merely as 
 force, and with the laws and mode of their operation ; and until 
 these are fully developed and comprehended, all speculation as to 
 the nature of the forces themselves will be defective. 
 
 2. The peculiar invisible power, then, which we term Electricity, 
 may be viewed as one of those incomprehensible and mysterious 
 natural agencies known to us only by its effects ; and, although 
 involved in deep obscurity as a source of physical power, it is, in 
 common with other agencies, ever present, often under a form not 
 cognisable by our senses, and is hence termed latent. The slightest 
 change, however, in the attendant circumstances is sufficient to call 
 up this mysterious power, electricity, from its apparently dormant 
 state, and to make us immediately sensible of its presence. Many 
 substances, for example, on being simply brought into close con- 
 tact under pressure, and subsequently separated, not only tend to 
 come together again, but seem actually to exert a species of attrac- 
 tive force, both on each other and on other bodies. A similar 
 result ensues in a great variety of operations, natural and arti- 
 ficial ; such, for example, as in the formation of rain, hail, snow, 
 mist, and other meteorological phenomena ; also in the case of 
 variations in temperature, as in the heating and cooling of certain 
 mineral bodies, such as the tourmaline, the topaz, and some others ; 
 also in changes of form, as in the liquefaction or consolidation of 
 brimstone ; also in chemical action, as in the decomposition of 
 water, the solution of metals in acids, &c. ; also in mechanical 
 operations, such as cleavage, abrasion, friction, and such like; 
 all these are sources of electrical developmentj and may superin- 
 duce on common matter attractive force. 
 
 3. The first written notice of electrical action appears to have 
 originated, 600 years before the Christian era, with a Greek 
 philosopher, Thales, a native of Miletus, founder of the Ionic 
 philosophy. He observed, as a remarkable property of amber, 
 that when subjected to a slight friction it acquired an attractive 
 power, and would draw light bodies towards it. Thales was so 
 much struck with this phenomenon that he considered the amber 
 to be endowed with a species of animation. 
 
 Theophrastus, also a Greek philosopher, subsequently observed, 
 after a lapse of three centuries, a similar property in a stone 
 
ELECTRICAL TERMS. 
 
 termed Lyncurium, supposed since to have been identical with the 
 tourmaline. This, he says, on being subjected to friction, will 
 not only attract light straws and sticks, but also thin pieces of 
 metal. The elder Pliny noticed the attractive property of amber, 
 and a similar property appears to have been discovered about the 
 time of Pliny, in agate and other precious stories. 
 
 4. The attractive power of amber when rubbed may be con- 
 sidered as the basis of our electrical nomenclature ; the Greek 
 word denoting amber being rjXeKrpov (electron), in Latin elect rum. 
 The unknown element which, according to Thales, gave it life, 
 has been termed Electricity. As our knowledge of the operation 
 of this active principle began to advance, and other substances 
 were found to display a similar property to that of amber, such 
 substances were said to be amber -like, or electrical, and were hence 
 termed electrics. Again, any substance having attractive force 
 superinduced upon it, after the manner of amber, was said to be 
 electrified, or electrically excited, whilst its peculiar state at this 
 instant was said to be a state of electrical excitation. These terms 
 are still retained, and others have been introduced on the same 
 basis. Thus, the attractive force displayed under this excitation 
 is termed electrical attraction ; and the particular substance itself 
 is denominated an excited electric ; while any particular contrivance 
 for better observing and detecting the presence of electrical force 
 is termed an electroscope, and when constructed so as to measure 
 its power is termed an electrometer. 
 
 5. Electrical excitation in any given substance may be observed 
 easily ; it is only requisite to present the excited body to some 
 light substance, when its attractive power becomes immediately 
 apparent. The substances best adapted to the exhibition of elec- 
 trical attraction are downy feathers, bog down, small balls of 
 elder pith, or the pith of the Sola plant, fragments of gold or 
 silver leaf, a thread of cotton, and such like. Light substances 
 of this kind, especially if delicately suspended so as to be free to 
 move in any given direction, are convenient and efficient electro- 
 scopes, A thread of fine sewing-cotton two or three feet in length 
 suspended from the ceiling of a room by means of a small piece 
 of sticking-plaster or gummed paper, is very obedient to electrical 
 attractive force, and constitutes a simple and efficient instrument ; 
 any electrified substance, however weakly excited, will, on being 
 presented to the suspended thread, at the distance of some feet, 
 set it in motion towards itself (Note A). 
 
 Gilbert employed a delicately balanced magnetic needle for 
 testing the presence of electrical force, a species of electroscope 
 
 B 2 
 
SIMPLE ELECTROSCOPES. 
 
 frequently resorted to by later electricians, in form, at least, if not 
 in material. A delicate electroscope needle of this form is shown 
 in Fig. 1. It is sensitive and convenient for many purposes (B). 
 
 6. A fine cotton or linen thread suspended from an arm or 
 support, and having a small spherical mass of cotton wool or 
 bog-down attached to it, about the size of a hazel-nut, as at m, 
 Fig. 2, constitutes also a delicate and sensitive electroscope. The 
 
 Fig, 1. Electroscope Needle. Fig. 2. Bob Electroscope. 
 
 terminating pendent mass m is sometimes named the bob, and serves 
 to expose a larger amount of attractive surface. A light sphere 
 of elder pith, or other light vegetable substance, about one-fourth to 
 three-eighths of an inch diameter, may be employed for this purpose. 
 A simple and efficient electroscope is also obtained by means of a 
 light straw reed, such as is employed in straw bonnet making. 
 This reed may be attached to a fine fibre of silk or cotton thread 
 
 Fig. 3. Trial Eeed. 
 
 Fig. 4. G-old-leaf Electroscope. 
 
 by means of a little sealing-wax or gum, and hung from any 
 convenient support, as in Fig. 3. A light pith ball should be 
 placed on its lower extremity. This kind of electroscope may be 
 called a trial reed, and if suspended from a light glass or metal 
 arm, as in Fig. 3, may be held in the hand* Straw reeds are useful 
 in detecting electricity. 
 
ELEMENTARY EXPERIMENTS. 5 
 
 One of the most delicate tests of electrical excitation is a slip of 
 leaf gold, F, Fig. 4, attached to a short paper holder, and suspended 
 under a ^bell glass, in any convenient manner, so as to shield it 
 from currents of air. A strip of gold-leaf arranged in this way is 
 obedient to an extremely small force of attraction (C). A great 
 variety of contrivances for detecting electrical force will be duly 
 noted as we proceed. 
 
 7. The following experiments may be adduced as simple illus- 
 trations of electrical excitation : 
 
 Exp. 1. An ordinary stick of sealing-wax, or a roll of common 
 brimstone, is to be gently rubbed with a dry handkerchief of soft 
 silk, or with a piece of flannel. When the excited body is 
 presented to any one of the electroscopes above described, the 
 part free to move will be drawn towards it, even though it be 
 held at the distance of some feet. 
 
 Exp. 2. Let a tube of glass, about 15 inches long, and three- 
 fourths of an inch in diameter, be gently warmed, and drawn 
 through the folds of a dry silk handkerchief, held in the hand. It 
 will become so powerfully excited that faint luminous flashes may 
 be frequently observed on its surface in a dark room, and also 
 luminous sparks, termed by the older electricians electrical fire. 
 
 The excitation of glass is greatly promoted by the applica- 
 tion of an amalgam of tin, zinc, and mercury CD).- The bisulphide 
 of tin, also termed aurum musivum, a friable substance of a yellow 
 colour, may be substituted for the amalgam. 
 
 A tube of glass about 2 feet in length and 1 inch in diameter, 
 made very dry, and warm inside and out, excited in this way 
 by warm dry silk, throws out in the dark, vivid sparks and 
 luminous flashes, attended by a crackling noise and a peculiar 
 odour, causing a singular sensation on the face or hand, termed 
 by electricians spider's web. 
 
 Exp. 3. Two silk ribbons, the one black and the other white, 
 each about a yard in length, are to be applied fairly to each 
 other, and then drawn, with slight pressure, through a fold 
 of silk velvet, or even through the fingers ; they will be found 
 to adhere, although they do not at the moment affect the 
 electroscope ; but directly they are separated, each ribbon will 
 be attractive of the other, and of surrounding matter, and if 
 presented to the electroscope will powerfully affect it. 
 
 Such are a few elementary illustrations of electrical excitation 
 as dependent on simple friction and pressure. 
 
 8. Every known substance is electrically excitable in a greater 
 or less degree, provided due precautions be observed. But it is 
 
6 
 
 ELECttUCS. 
 
 only under certain conditions, and in a certain class of bodies, 
 that we are able, under ordinary circumstances, to develop this 
 property to any extent. We are unable, for example, to excite 
 a metallic tube under the same circumstances and in the same 
 way as we excite a tube of glass or a roll of brimstone (7). Hence 
 has arisen a classification of substances in relation to their electrical 
 excitability. For example : substances which are amber-like that 
 is, when subjected to friction, display attractive force have been, 
 from an early period, classed as electrics, and considered more espe- 
 cially as electrical bodies, or idio electrics, or substances naturally 
 electric in themselves. Later researches, however, do not permit 
 us to recognise electrical excitability as an exclusive property of 
 any particular substance. There is no physical evidence for such 
 a conclusion, and it can only be admitted as being peculiar to 
 certain substances, under certain conditions. With this limita- 
 tion the classification is not altogether without advantage, and 
 is so far admissible. 
 
 The following table contains a list of idio- electrics, or those 
 bodies which have the property of attracting light substances when 
 subjected to friction in the common way : 
 
 TABLE I. 
 LIST OF IDIO-ELECTBICS. 
 
 Gum lac, shell lac ; gums of all kinds, including camphor. 
 
 Eesinous substances of every kind, including common wax. 
 
 Gutta-percha. 
 
 Bituminous substances, including amber and jet. 
 
 Sulphur, and some of its compounds, including vulcanite and other 
 artificial preparations. 
 
 The diamond ; nearly all precious stones ; agate. 
 
 Mica, and other laminated minerals. 
 
 Glass and all vitreous substances. 
 
 Porcelain. 
 
 All crystalline transparent gems, especially tourmaline. 
 
 Silk of every kind and form, especially silk gut. 
 
 Dried animal skins, and fur ; skins of living animals in their ordinary 
 state. 
 
 Gun-cotton, and wool generally ; hair and feathers. 
 
 Baked dry wood, and dry vegetable substances ; very dry warm paper. 
 
 Oils and fatty fluids, including turpentine. 
 
 All dry gases, including atmospheric air. 
 
 Steam of very high temperature. 
 
 Ashes of animal and vegetable bodies generally. 
 
 Phosphorus ; lime and chalk, deprived of all moisture. 
 
 Dry metallic oxides. 
 
 Lycopodium. 
 
 Dry ice, below the zero of Fahrenheit's thermometer. 
 
ELECTRICAL ATTRACTION. 7 
 
 9. The most available and efficient electrics for ordinary ex- 
 perimental inquiries are gum lac, gutta-percha, brimstone, 
 amber, animal fur, cat skin, hare skin, glass, *ilk, vitreous and 
 resinous bodies generally. There are also many artificial com- 
 pounds of these, which are valuable as electrics, such as the com- 
 pound of caoutchouc and brimstone, termed vulcanite, ordinary 
 sealing-wax, compounded of shell lac and turpentine, to which is 
 added occasionally a little resin. Gun-cotton and collodion, formed 
 from a solution of gun-cotton in ether, are available as electric 
 bodies. 
 
 10. Although metallic substances and many other bodies do not 
 display electrical attractive force when subjected to friction in 
 the usual way, they may nevertheless display this property and 
 appear to have attractive force superinduced upon them by placing 
 them in communication with an electrically excited substance (4). 
 This remarkable discovery was made by Mr. Stephen Gray, of 
 the Charter House, London, and others, so long since as the year 
 1729, and is a fine instance of experimental inductive science. 
 
 The history of this discovery is worthy of notice. After many 
 attempts to render metallic substances electrically attractive by 
 friction, the experimenters were led to notice the phenomenon of 
 an attractive power acquired 
 by a cork with which they 
 had closed one of the open 
 ends of a glass tube, T, Fig. 5, 
 subjected to excitation. This 
 led them to insert a light 
 rod of fir wood, c, into the 
 cork b, so as to pass within 
 the tube, while on its ex- 
 posed extremity was a ball of 
 
 r at i ri - 5> Ex ample of Excitation and Attraction. 
 
 ivory, d. W nen the glass tu be 
 
 was briskly rubbed they noticed to their great astonishment that 
 both the fir rod and its terminating ball had acquired, in common 
 with the glass, an attractive force, and drew toward it light sub- 
 stances, even more freely than the excited tube itself. The length 
 of the fir rod c, and consequent distance of the ball d, from the 
 source of power, appeared in no way to interfere with the result ; 
 so that in suspending a second ball, h, from the rod by means of 
 a line of hemp of greater or less length, they were enabled on 
 exciting the glass tube to attract a light substance, n, from off the 
 pavement, from the summit of a high building. 
 
 On extending the inquiry with a view to discover the limit 
 
INSULATION AND CONDUCTION. 
 
 of distance through which, this attractive power might be trans- 
 mitted, they were led to take a horizontal direction, in which 
 case it became requisite to suspend the line of communication 
 with the glass tube and rod, Fig. 5, from a beam, A, Fig. 6, 
 in a sheltered avenue, in which their experiments were con- 
 ducted. This they effected by means of a hempen cord of 
 suspension, m. Under these conditions no evidence of attrac- 
 
 tion was observable in 
 the ball h. The cord 
 of suspension m being 
 thick, they were led 
 to imagine that the 
 electric virtue was ab- 
 sorbed by it, and car- 
 ried up to the beam 
 to which the cord 
 was appended. This 
 led them to abandon 
 
 Fig. 6. Illustration of Insulation and Conduction. suspensioilj 
 
 and substitute a line of fine silk. The result fully justified 
 their expectations ; the attractive power now appeared in the 
 ball h, as before. In this way they succeeded by means of silk 
 suspensions in transmitting the attractive force through a distance 
 of 765 feet. With a view of still further perfecting their research, 
 they again changed the silk suspension for a still finer suspension 
 of metal wire; but this, so far from improving the electrical 
 transmission, destroyed it altogether, as in the case of the hempen 
 cord. It was hence evident that the silk cord had satisfied 
 some important condition in this interesting experiment, and 
 that the result they arrived at rather depended on the nature of 
 the substance of the cord of suspension m than on its size. In this 
 way a most important discovery was made, namely, the conducting 
 power of certain bodies. Both the hempen cord and the small wire 
 had evidently caught up and transmitted the excited force to the 
 beam A, from whence it had passed away to other bodies ; whilst 
 the silk having no such power, had caused the electricity to be 
 retained on the line of electric communication between the excited 
 tube T and the ball h. 
 
 11. This fine elementary discovery speedily led to a new and dis- 
 tinct class of phenomena dependent on what has been since termed 
 electrical conduction, and to another classification of substances 
 considered as conductors of electricity, in contradistinction to the 
 class enumerated in Table L, termed electrics or non-conductors. 
 
CONDUCTION AND INSULATION. 
 
 The substances which fall more especially under the class of 
 conductors are placed in the following Table : 
 
 TABLE II. 
 ENUMERATION OF ELECTRICAL CONDUCTORS. 
 
 All the metals and metallic bodies. 
 
 Plumbago. 
 
 Well-burned charcoal. 
 
 Concentrated acids and saline fluids generally. 
 
 Water, including aqueous vapour. 
 
 Ice and snow above 10 of Fahrenheit's thermometer. 
 
 Animal solids and fluids. 
 
 Vegetable fluids and most vegetable solids. 
 
 Flaine and smoke. 
 
 Alcoholic and other spirituous vapour. 
 
 Vapour of ether. 
 
 Highly rarefied air. 
 
 Nearly all earths and stones. 
 
 12. Many substances owe their conducting power to the moisture 
 contained in them. Thus, baked wood is an electric, and does not 
 easily transmit electrical force ; but wood in a natural state, con- 
 taining moisture, is a conductor. By exposure to very great heat, 
 the volatile parts of wood are driven off, and, if air be excluded, its 
 indestructible base, charcoal, remains. This is a conducting sub- 
 stance. But charcoal further exposed to heat with an access of air 
 undergoes combustion, and is converted first into a volatile gas 
 known as carbonic acid, or carbonic anhydride, while the solid mineral 
 part of the wood remains in the form of ash. The carbonic acid 
 gas and the charcoal are both electrics. The influence of heat in 
 this process is not very intelligible ; it being difficult to determine 
 the more immediate sources of electric and conducting power. We 
 must, however, conclude as a general fact that all electric sub- 
 stances (Table I.) are non-conductors, and reciprocally all con- 
 ductors (Table II.) are non-electrics. 
 
 13. We infer from the preceding facts (10), that in order to 
 retain electricity on any given conducting body, as on P, Fig. 7, it 
 is requisite to support the body on one of the non-conducting sub- 
 stances enumerated in Table I., as for example on a glass rod, N. 
 By this means we imprison, as it were, the electrical agency, and 
 confine it to the given body, where otherwise it would not remain 
 (10). A conductor of electricity thus circumstanced is said to be 
 insulated, and is termed an insulated conductor ; when electrified it 
 is said to be charged (). 
 
 14. By means of a small conducting sphere, or thin disc, u, 
 Fig. 7, insulated on a long slender rod of varnished glass, or some 
 
10 INSULATION AND CONDUCTION. 
 
 other electric substance, we are enabled to transfer electricity 
 from a charged insulated conductor, p, to another insulated con- 
 ductor, M, not so charged. This may be effected by first touching 
 the charged body P with the small insulated neutral sphere 
 or disc, u, which body now participates in the charge, as if it 
 were a portion of the electrified body p. We then proceed to 
 
 transfer the electricity thus 
 taken up by the transfer 
 sphere or disc u to the given 
 neutral body M, by bring- 
 ing it into contact with 
 M. The insulated body M, 
 in virtue of the contact, shares 
 in its turn the electricity of 
 the transfer ball or disc u, and 
 
 Fig. 7. Insulated Conductors. ^^ ^ repeated contac ts Or 
 
 transfers we are enabled to bear off a great portion of the electricity 
 of the charged body p, and deposit it on a second and distant insu- 
 lated conductor, M. 
 
 The insulated transfer sphere or disc u has been termed a carrier, 
 and is largely employed in electrical research. A small thin 
 circular metallic disc, termed by the French philosophers a tangent 
 plane, or plate, is frequently substituted for the carrier ball, as 
 being better adapted to certain peculiar methods of experiment. 
 
 As insulation (13) is the great desideratum in this experiment 
 particular care is necessary in perfecting it. The insulator should 
 be small, long, and nicely varnished. 
 
 16. Although the march of experiment up to an advanced 
 period of electrical inquiry has been such as to completely 
 separate the phenomena of insulation and conduction as dependent 
 on two distinct and opposite principles, yet according to Fara- 
 day there is really no substance which absolutely arrests, or 
 which perfectly transmits electrical action ; conductors and 
 insulators being so far identical in physical constitution, that 
 the electric or insulating power and the non-electric or con- 
 ducting power differ in degree only. Admitting this as highly 
 probable, if not positively demonstrable, we have still to observe 
 that the great practical value of the division of bodies into 
 electrics and conductors (as first proposed by Desaguliers so 
 long since as the year 1740) is unquestionable. The extremes 
 or limits of the differences are so wide apart as really to admit of 
 substances being considered generally either as conductors or 
 insulators, allowing an intermediate class of substances partaking 
 
ELECTRICS AND CONDUCTORS. 
 
 11 
 
 of either property, and hence considered as imperfect either as 
 conductors or as insulators. 
 
 If we tabulate bodies in respect of their conducting or insulating 
 power, we find metals, charcoal, and concentrated acids to be the 
 best conductors at one extremity of the series ; shell lac, brim- 
 stone, silk, vitreous and resinous bodies generally, the best electrics 
 at the other extremity of the series. In the intermediate or tran- 
 sition class we have bodies such as marble, chalk, earths, and stones. 
 
 In the following Table the substances are systematically 
 arranged in the order of conducting and insulating power. If 
 taken in relation to insulating power, the series commences in the 
 first column with resinous bodies and terminates with metallic 
 substances in the last. If taken in the order of conducting 
 power, the series must begin with the last substances of the third 
 column that is, with metallic bodies and must be read in the 
 reverse order, terminating with the resinous bodies in the first 
 column. 
 
 TABLE III. 
 
 Electrics or Insulators. 
 
 Intermediate Substances. 
 
 Conductors or Non-Electrics. 
 
 
 Lac, and all resinous 
 
 Spermaceti and fatty 
 
 Flame. 
 
 
 bodies. 
 
 substances. 
 
 Saline substances. 
 
 
 Brimstone , gutta-percha . 
 
 Dry metallic oxides. 
 
 Aqueous fluids. 
 
 
 Amber and bituminous 
 
 Dry vegetable bodies. 
 
 Water. 
 
 
 substances. 
 
 Wool and hair. 
 
 Saline fluids. 
 
 
 Camphor and gums. 
 
 Feathers. 
 
 Aqueous vapour. 
 
 
 Gun-cotton and collo- 
 
 Phosphorus. 
 
 Alcoholic vapour. 
 
 
 dion. 
 
 Wood. 
 
 Snowandice at32F. 
 
 
 Mica. 
 
 Paper, parchment. 
 
 Concentrated acids. 
 
 
 Glass, and all vitrifica- 
 
 Leather. 
 
 Plumbago, 
 
 
 tions. 
 
 Marble. 
 
 Charcoal. 
 
 
 Tourmaline and jet. 
 
 Stones and earth. 
 
 All metals. 
 
 
 Diamond and other gems. 
 
 Ice above 10 F. 
 
 
 
 Silk, and silk-worm gut. 
 Animal furs and skins. 
 
 Earefied air. 
 Steam. 
 
 L I B R A R 
 
 Y 
 
 Oils, especially turpen- 
 
 Smoke. 
 
 
 
 tine. 
 
 I 
 
 rNIVERSITY 
 
 OF 
 
 Dry air and gases. 
 
 
 
 
 High-pressure steam. 
 Fish-glue or isinglass. 
 
 I 
 
 CALIFOMNI 
 
 A. 
 
 * 
 
 Dried animal gut. 
 
 ^ 
 
 
 
 
 16. The order of the series in the above Table can only be con- 
 sidered as a useful approximation, since we really cannot, in the 
 present state of electrical knowledge, assign in the scale of conduct- 
 ing power the precise position of every kind of substance, natural 
 and artificial. Besides which, many circumstances arise that tend 
 
12 ATTRACTION AND KEPtJLSION. 
 
 greatly to embarrass the judgment. Thus, in determining the 
 position of charcoal as a conductor, we must look at the process of 
 its manufacture, and ascertain from what kind of wood it was 
 prepared, for there is an amazing difference in the amount of ash 
 in different woods, and this has an influence on the conducting 
 power of the charcoal. " On the other hand, when we examine the 
 insulating or electric power of glass, we have to consider in like 
 manner its manufacture and the kind of glass, whether made with 
 potash or with soda, whether the silica in its composition is free 
 from iron, or contains that or any other metal, and so on. We 
 therefore must not regard these bodies with a very rigid eye ; for 
 many substances owe the greater portion of their conducting power, 
 as already observed, to the water contained in them, whilst many 
 insulators are dependent in great measure on the state of the atmo- 
 sphere. In any order of sequence therefore that may be attempted 
 we necessarily repose on a large generalisation. Nevertheless the 
 position of most bodies, as electrics or as conductors, is pretty well 
 determined. Thus the metals, although differing amongst them- 
 selves in degree of conducting power, are certainly upon the whole 
 the best electrical conductors, and stand first in the conducting 
 series ; on the other hand, resinous and vitreous substances are 
 the best insulators, and come first in the order of electrics, but 
 last in the sequence of conductors. Such bodies as marble, parch- 
 ment, wool, &c., are certainly to be classed as intermediate or 
 transition substances. The best conductors are silver and copper. 
 The best insulators are amber, lac, brimstone, and resinous and 
 vitreous bodies generally. The liability, however, of glass to con- 
 dense aqueous vapour on its surface, by which its insulating power 
 is diminished, renders it necessary to cover the surface with a 
 thin solution of shell lac in naphtha, dried off by heat. Resinous 
 bodies not being so liable to aqueous deposition, this precaution 
 is not so immediately called for. 
 
 17. Attractive force, although the first, and commonly the most 
 evident result of electrical excitation, is not the only exhibition of 
 power consequent on the electrical state of common matter. On 
 a further examination of the attendant phenomena, new facts 
 present themselves. If the electrical excitation be vigorous, 
 and the electroscope, or other attracted substance, be insulated, the 
 attractive force appears to vanish at the instant the two bodies 
 come into contact, and an opposite or apparently repulsive force 
 succeeds, so that the insulated electroscope now recedes from the 
 excited or charged body with considerable violence, and cannot be 
 again attracted until it has been touched by some conducting non- 
 
ATTRACTION AND REPULSION. 
 
 13 
 
 electrified body, of such magnitude as can deprive it of the 
 electricity it has acquired. 
 
 18. The following experiments are strikingly illustrative of this 
 singular and important fact : 
 
 Exp. 4. Suspend the electroscope, Fig. 2 (6), by means of a 
 filament of unspun silk, by which its ball will be insulated. This 
 may be termed the insulated electroscope, in contradistinction 
 to the first electroscope suspension, Fig. 1 ; or if the needle 
 electroscope, Fig. 2, be employed, it is to be sustained on a 
 varnished rod of glass, in which case it will be an insulated 
 needle. Excite the glass tube T, Fig. 8, furnished with a con- 
 ducting rod and ball d. 
 Present its electrified 
 ball to the insulated 
 electroscope m, the 
 electroscope will be 
 attracted and drawn 
 toward the ball d, and 
 if allowed to come into 
 contact with it the elec- 
 troscope ball m will 
 instantly rebound from 
 it as indicated by the 
 dotted line m', and will 
 not again come near 
 the charged ball d of 
 the excited tube T until touched by the finger or some freely con- 
 ducting substance. 
 
 Exp. 5. Let the metallic sphere s, Fig. 8, about 2 inches in 
 diameter, sustained on a conducting rod, w, be placed on one side 
 of the pendent ball m of the insulated electroscope, and present 
 to the ball m on the opposite side the excited tube and ball d. 
 The pendent ball m of the electroscope will be first attracted and 
 then repelled by the charged ball d, and will fly to touch the 
 opposite uninsulated ball s, to which it will communicate its 
 acquired charge (14) ; it will then be in a condition to be again 
 attracted by the charged ball d, and in this way will vibrate for a 
 certain time between the two balls, depending on the force or 
 amount of electricity in operation. 
 
 This experiment may be repeated under a very simple form. 
 It is only requisite to bring the charged ball of the excited tube 
 near any light substance or fragment of conducting matter resting 
 on a deal table, and a series of attractions and repulsions of a similar 
 
 Fig. 8. Attraction and Eepulsion. 
 
14 ATTRACTION AND REPULSION'. 
 
 kind, exerted between the table, the light body, and the charged 
 ball, will be immediately apparent. A light feather, a fragment 
 of paper or of leaf gold, may be employed in this experiment. 
 
 19. The repulsive force thus evinced is better displayed by a 
 charged insulated conductor than by a simply excited electric. If 
 the attractive force be weak, the attracted substance frequently 
 remains, as if adhering to the excited surface, and does not imme- 
 diately leave it. It may probably remain attached to it, but it 
 will be always thrown off after repellent power can operate 
 between the bodies at any very small distance. 
 
 20. On a further investigation of the electrical repellent power 
 we find a most important relation of electrical force to the different 
 kinds of ordinary matter, a relation which, in the early progress 
 of our knowledge of this wonderful agency, could scarcely have 
 been anticipated. We find, for example, that when certain excited 
 electrics repel the electroscope ball, other excited electrics actually 
 attract it, so that when repelled by some excited electrics it is 
 attracted by others, a result evident in the following experimental 
 illustrations. 
 
 Exp. 6. The glass tube T, Fig. 8 (18), being powerfully 
 excited, let it attract and repel the insulated electroscope ball m, 
 the metallic sphere s being removed. Whilst in this repulsed 
 state withdraw the repellent ball and tube T, and present to the 
 electroscope ball m an excited roll of brimstone, gutta-percha, or 
 gum lac. These substances do not continue to repel the electro- 
 scope ball ; on the contrary, the ball flies toward them with 
 remarkable facility. 
 
 Exp. 7. Let a gutta-percha tube of an inch or more in diameter, 
 and about 18 inches in length, be prepared and fitted with a con- 
 ducting rod and ball as in the case of the glass tube T (10). Excite 
 these tubes, the glass one with dry silk, the gutta-percha one 
 with dry woollen cloth. Bring the ball d of the excited glass 
 tube on one side of the insulated electroscope ball m, and the ball 
 of the excited gutta-percha tube on the opposite side, as at s and d, 
 Fig. 8. The electroscope m will for a moment or two vibrate 
 between the two balls in a similar way to that shown in Exp. 
 5 (18). 
 
 21. A simple and useful kind of instrument, in which both kinds 
 of excitation may be produced, is obtained by coating a glass tube 
 with a solution of shell lac through one half its length, leaving 
 the other half free, and then inserting a conducting wire and ball 
 into each end of the tube. This instrument, Fig. 9, supplies the 
 place of the two separate tubes just noticed, according to the 
 
TWO KINDS OF ELECTRICITY. 15 
 
 portion we excite, whether the glass or the lac. It is desirable to 
 guard the centre c, when held in the hand, by rolling a few turns 
 of silk ribbon round it, and covering the ribbon with a short piece 
 of varnished glass tube ; we thus obtain a convenient insu- 
 lating handle ; and this enables the operator to excite 
 either side with more freedom. The exciting tube may be 
 about 2 feet in length ; either half may be excited with 
 dry soft white silk made moderately warm (If). 
 
 22. The foregoing phenomena (18) led the early electri- 
 cians to the following important deduction, viz., excited 
 electrics develop two kinds of electricity, attractive of 
 each other, but repulsive of themselves. We are indebted 
 to M. Du Fay, Intendant of the gardens of the King 
 of France in 1730, and Member of the Royal Academy of 
 Sciences, for this remarkable fact. His valuable memoirs, 
 printed in L'Histoire de I'Academie, from 1733 to 1737, 
 were of great importance to the progress of electrical 
 research at that period of its history. 
 
 Conceiving that the opposite electricities developed 
 by vitreous and resinous bodies (20) were peculiar to 
 these substances, M. Du Fay inferred the existence of two 
 distinct electric principles. " Excited glass," he observes, Tube - 
 "always repels excited glass; whilst, on the contrary, it attracts 
 all bodies the electricity of which is of the nature of amber. 
 Reciprocally, amber, and other substances of the same electricity 
 as amber, repel each other, but attract the electricity of excited 
 glass." M. Du Fay considers that his fourth memoir established 
 two great facts : first, that electrically excited bodies begin by 
 attracting all other bodies, that when they have rendered bodies 
 electrical by communication (18), then they repel them, but not 
 before ; second, that there exist two kinds of electricity differing 
 essentially from each other, one produced by excited glass and other 
 vitreous bodies, the other by excited amber and other resinous 
 bodies ; these he styles the one vitreous the other resinous elec- 
 tricity, as depending on the sources from which he supposes them 
 to be derived, whether from vitreous or resinous bodies. 
 
 23. This notion of vitreous and resinous electricity did not, how- 
 ever, long survive the discovery on which the doctrine of M. 
 Du Fay was based. Indeed, M. Du Fay himself began at last to 
 countenance an hypothesis more or less prevalent at the time, 
 which assumed an identity of the two forces, the observed diffe- 
 rence being merely a difference in degree of power by which the 
 stronger overcame the weaker. 
 
16 TWO KINDS OF ELECTRICITY. 
 
 That the two supposed electricities are not either of them 
 peculiar properties of vitreous and resinous bodies, was soon made 
 apparent by the fact, that both may be obtained in the excitation 
 of either a vitreous or a resinous substance. Moreover both are 
 developed in every case of electrical excitation, and are co- 
 existent. 
 
 Exp. 8. Let the resinous half of the electrical tube (21) be 
 excited by dry soft silk, or woollen cloth : it attracts the electro- 
 scope when charged with vitreous electricity, that is to say, 
 when repelled by the glass half, excited by silk; but will repel 
 the electroscope if charged with resinous electricity, that is to 
 say, when repelled by the glass half, excited with cat skin ; evi- 
 dently showing that both vitreous and resinous electricity have 
 been developed in the excitation of the glass by silk and by cat 
 skin. (29) Table IV. This is further shown by exciting the glass 
 tube T, Fig. 8 (18), with silk, and when it freely repels the electro- 
 scope charged with vitreous electricity, exciting the tube with the 
 fur of cat skin, made warm and dry ; the tube instead of repel- 
 ling, now attracts the electroscope, when charged with vitreous 
 electricity. 
 
 Exp. 9. Excite the gutta-percha tube Exp. 7 (20) with dry, 
 warm silk, it will repel the electroscope charged with resinous 
 electricity that is to say, when attracted by the excited glass 
 tube T, Fig. 8. Excite now the same gutta-percha tube with the 
 rough side of dry oiled silk, smeared over with the metallic 
 amalgam already referred to (7) (D) ; the same gutta-percha tube 
 now attracts the electroscope when attracted by the excited glass ; 
 or in other words, the resinous surface will have developed both 
 the electricities. (29) Table IT. 
 
 Exp. 10. A wide silk ribbon is. to be rolled round a long 
 cylindrical cork, c, Fig. 10, and a varnished glass rod is to be 
 
 passed into the cork to serve as an in- 
 sulating handle. Let now this insulated 
 silk rubber be passed with slight friction 
 a few times in one direction over a dry, 
 warm slip of window glass, A B, retaining 
 the silk rubber by its insulating handle 
 
 Fig. 10. suk Rubber and Glass ^ *& one hand, and the glass by one of its 
 extremities in the other. Both the silk 
 
 and the glass will attract the uninsulated electroscope (6), and each 
 may be caused to attract or repel a charged insulated electroscope 
 (18), but they will not both attract or both repel at the same time. 
 The phenomena will be precisely the same as in Exp. 9. The silk 
 
POSITIVE AND NEGATIVE. 17 
 
 will attract when the glass repels and conversely, and the effects 
 of Du Fay's vitreous and resinous electricities be obtained. 
 
 24. It is quite evident by these experiments that both elec- 
 tricities are produced by the same substance ; that they are both 
 developed in the process of excitation, and consequently co- existent, 
 although not always both apparent. The reason for this is that 
 in ordinary cases the rubber is not insulated, but is in communica- 
 tion with the ground, through the medium of the hand, or of some 
 other conducting body. Hence the electricity of the rubber is con- 
 tinually passing away, or 'being otherwise neutralised. For a 
 similar reason we fail under ordinary circumstances in endeavour- 
 ing to excite conducting bodies such as the metals. In order to 
 obtain the excited electricity of such bodies,, it is requisite to in- 
 sulate them (13). 
 
 "We have further to observe that in this proeess of excitation by 
 friction, the electrical development is by no means perpetual, 
 except under certain conditions. If both the rubber and the 
 electric be insulated, the development after a certain time ceases. 
 It would seem as if the rubber could only develop a certain 
 quantity of the opposite electricity. It is not, therefore, until we 
 have restored these bodies to their normal state by contact with 
 free conductors, that we are enabled to continue the excitation. In 
 the case of the excited tubes (20) the hand, being a conductor, 
 continually preserves the rubber in its normal state, whilst the in- 
 sulated rod and ball, projecting from within the tubes, operates 
 continually in abstracting the excited electricity of the tubes, and 
 in giving it off to other bodies. It is in this way we are enabled to 
 obtain a continuous succession of sparks from the conducting ball. 
 
 25. The terms vitreous and resinous, however, as applied to 
 opposite electricities, displayed in the foregoing experiments, carry 
 with them an hypothetical deduction not warranted by the facts of 
 the case. It has been hence considered more philosophical to cha- 
 racterise the opposite electrical forces by the terms positive and 
 negative, and to represent them by the algebraic signs + and , or 
 plus and minus. By this means we disembarrass the question of 
 an illusory form of expression, and are enabled to reason from the 
 effects of the unknown causes of these phenomena, as if the causes 
 were known ; much in the same way as the astronomer calculates 
 the movements of the bodies in planetary space, without at all 
 understanding the nature of the occult cause of their movements. 
 
 26. A striking and important result of electrical excitation by 
 friction is, that so long as the electric and rubber remain in contact, 
 no electrical development appears to result. It is only when the 
 
18 ORIGIN OF THE TWO ELECTRICITIES. 
 
 two bodies are separated that electrical force becomes apparent in 
 
 either. 
 
 Exp. 11. P w, Fig. 11, is a circular plate of glass about 10 in. 
 
 in diameter, sustained in a slit ball, n, by means of a compressing 
 
 screw. R is a circular rubber in contact with the glass plate. 
 
 R is fixed at the extremity of a 
 long glass rod, a R, movable in the 
 supporting balls c y, so as to be 
 easily turned round against the 
 glass. The glass rod and rubber 
 are secured in position upon a con- 
 venient base and pillar, fixed on an 
 elliptical stand, M r> ; an electro- 
 scope ball, e, suspended by a silk 
 thread from a light insulating rod, /, 
 
 Fig. 11, Glass Plate and Kubber. . , ., ., , , . 
 
 is nung opposite the glass plate p w, 
 
 so as to indicate any electrical force developed in the plate by the 
 movement of the rubber. 
 
 The glass being warm and dry, and the rubber R lightly covered 
 with electrical amalgam (7), turn the rubber round against the 
 glass so as to produce a continuous friction. Not the slightest 
 effect will be observable on the electroscope ball e. Break the 
 contact between the rubber and the glass by withdrawing the 
 glass rod a R through its supporting balls c y t an immediate elec- 
 trical development on the glass plate ensues ; the electroscope ball 
 is forcibly attracted towards the glass plate. 
 
 It has been inferred from these phenomena that the develop- 
 ment of electricity by friction is the result of the separation and 
 renewal of contact between dissimilar bodies, and has been further 
 elucidated by a simple hypothetical exposition. The electrical 
 agency, in whatever it consists, may be supposed to be distributed 
 through bodies in quantities proportionate to their affinities for 
 it. Let now the affinity of glass for electricity be represented 
 by the number 6, and the affinity of the rubber by the number 4, 
 making together 10. When the two bodies are brought into 
 contact these relative affinities are supposed to be changed : the 
 affinity of the glass being increased from 6 to 8, the affinity of the 
 rubber decreased from 4 to 2, the total 10 being still unchanged. 
 When we separate the rubber from the glass or break the con- 
 tact, the original affinities 6 and 4 are restored, whilst the new 
 distribution remains. The quantity of electricity in the glass, 
 therefore, has been increased by 2. Hence the glass remains posi- 
 tively electrified, or plus 2, and the rubber negatively electrified, 
 or minus 2. 
 
ATTRACTION AND REPULSION. 19 
 
 27. The important results so completely established in the course 
 of the preceding experimental investigations, lead directly to the 
 conclusion that substances in opposite electrical states, one plus and 
 the other minus, attract each other ; substances in similar electrical 
 states, both plus or both minus, recede from each other. This great 
 general and characteristic law of electrical force, is concisely 
 expressed in the following formula : Opposite electricities attract, 
 similar electricities repel. This prominent feature of electrical 
 action is ever to be kept in view, since it is found more or less in- 
 volved in every species of electrical research. 
 
 28. An instructive and elegant practical illustration of this 
 elementary law is seen in the following experiment: 
 
 Exp . 12. Excite, as before, the glass tube p, Fig. 12, and 
 having selected a sheet of leaf gold, such as is commonly used by 
 gilders and bookbinders, blow it gently into the air from 
 off any convenient plane surface. The leaf will for the <A 
 
 W'Y/> 
 
 moment float as it were in free space, as indicated in the yjjffl 
 figure. Whilst thus floating, present to the leaf the charged 
 ball p of the excited tube. The leaf will be immediately 
 drawn toward it ; but directly the bodies touch, the leaf 
 recedes with astonishing rapidity from the ball, and may 
 now be pushed through the air, by a distant impulse of the 
 positive electricity of the tube, in any direction. Whilst 
 thus driven away from the excited glass, present to the leaf 
 the ball of the excited gutta-percha tube (20) ; the gold 
 leaf will now fly rapidly toward the ball of the gutta- 
 percha tube, illustrating in a very elegant and striking 
 manner the opposite character of the two electrical forces ; 
 clearly showing that they are attractive of each other but 
 repulsive of themselves. In this experiment the forces rig . 12. 
 may be observed to operate at great distances, which gives Sifand 
 it a very definite character and renders the experiment a Go 
 fine example of electrical action. 
 
 29. The two opposite electrical forces thus called into play, in 
 every case of electrical excitation are mainly dependent on the rela- 
 tive condition or state of the surfaces or bodies in contact, or of the 
 surface of the excited electric, in respect of what is usually termed 
 the rubber. We find, for example, as already observed (24), that 
 opposite electricities may be obtained from the same substance. A 
 polished glass surface, for example, when excited by dry silk, 
 develops positive electricity; but when rubbed with cat skin, 
 negative. If we change the condition of the polished glass surface 
 by abrasion with emery powder, then the glass excited by means 
 
 c2 
 
 ; 
 
20 GENERAL DEDUCTIONS. 
 
 of woollen stuff develops negative electricity. So again, if we 
 give a shell-lac surface to a glass tube, by coating it with a solution 
 of shell-lac in naphtha (as in the positive and negative exciting 
 tube), (21), laid on as in varnishing, we obtain, on exciting this 
 surface with dry silk stuff, negative electricity ; but if we smear the 
 silk with metallic amalgam (7), we then obtain from the same lac 
 surface positive electricity. The method of rubbing produces also, 
 according to Faraday, in some cases very considerable difference, 
 although, as he remarks, it is not easy to say why. If we strike 
 a feather lightly against dry canvas, it becomes electrified nega- 
 tively ; if drawn with pressure between folds of the same canvas, 
 it is electrified positively. Friction applied across the grain of a 
 substance may develop an opposite electricity to that of friction in 
 the direction of the grain. The colour, or probably the colouring 
 matter of a substance, will often determine the kind of electricity 
 developed. Thus black and white silk frequently give opposite 
 electricities, although excited by the same substance and in the 
 same manner. 
 
 30. The following are some general experimental deductions rela- 
 tive to the development of electricity by the process of friction : 
 
 1, Cat skin is positive in all cases of friction with other sub- 
 stances. 
 
 2, Smooth or polished glass is positive with all other substances 
 except cat skin. 
 
 3. Rough glass is positive with dry silk, but negative with 
 woollen stuff. 
 
 4. Resinous bodies are negative with non^metallic substances, 
 but positive in most cases with metals. 
 
 Table IV., originally drawn up by Cavallo, and improved by 
 modern electricians, exhibits a concise and comprehensive view of 
 the results of experiments of this kind. But although this table 
 may be considered as pretty correct, we must remember that many 
 sources of error are inseparable from the experiments by which it 
 has been determined : anomalies are liable to arise in examiiimg 
 the particular electrical state of substances. For example, it is 
 stated by Faraday that one part of a cat skin may, after excita- 
 tion, exhibit positive electricity, and another portion negative 
 electricity. The practical manipulation with delicate electro- 
 meters, and the manner of treating them experimentally, may also 
 give rise occasionally to apparent discrepancies. The general 
 deduction, however, that vitreous matter develops positive elec- 
 tricity when excited by most other substances, and resinous bodies 
 negative electricity, is doubtless fully established. 
 
KIND OF ELECTRICITY DETERMINED. 
 
 21 
 
 TABLE IV. 
 
 SHOWING THE KIND OF ELECTRICITY EVOLVED IN THE FRICTION OF 
 
 VARIOUS SUBSTANCES, ONE WITH THE OTHER, UNDER DIFFERENT 
 CIRCUMSTANCES. 
 
 Substances. 
 
 Cat skin. 
 Smooth glass. 
 
 Eough glass. 
 
 Tourmaline. 
 Baked wood. 
 
 Amber. 
 
 Eesinous bodies. 
 Black silk. 
 
 White silk. 
 
 Positive, by friction with 
 
 Every other kind of matter, 
 Every substance except 
 
 cat skin. 
 Dry oiled silk, sulphur, 
 
 metals. 
 
 Amber. 
 
 Silk, paper, metals. 
 
 Many metals. 
 Sealing-wax, glass. 
 
 Black silk. 
 
 Negative, by friction with 
 
 Cat skin. 
 
 Woollen-cloth, paper, 
 wood, sealing-wax, the 
 hand. 
 
 Diamond, the hand. 
 
 Flannel, hare skin, po- 
 lished glass. 
 
 Silk, and most other 
 substances. 
 
 Substances generally. 
 
 White silk, hare skin, 
 ferret skin. 
 
 Cat skin. 
 
 31. The practical determination of the kind of electricity in 
 operation being of great importance, it may be as well to observe 
 that this can always be ascertained by presenting to an insulated 
 electroscope (18), charged with either of the electricities, the parti- 
 cular substance the electrical state of which is to be determined. 
 If an excited body be of the same kind as that with which the 
 electroscope is charged, repulsion will ensue ; if of an opposite kind, 
 attraction (27). Electrical attraction and repulsion have hence 
 been extensively employed as fundamental principles in the con- 
 struction of electroscopes and electrometers ; and since a correct 
 determination of the force and kind of electricity which an electri- 
 fied body evinces is of paramount importance in every kind of 
 electrical investigation, it may be as well to advert to the con- 
 struction and application of some of the most 
 
 elementary and useful of these instruments. 
 
 32. Gold leaf, as already observed (6), is an 
 extremely sensitive and efficient test of electrical 
 action. Fig. 13 represents a gold leaf electro- 
 scope of a simple and convenient kind. A single 
 gold leaf, R, is suspended within a glass receiver, 
 mounted on a firm foot, E. It has a lateral 
 opening, q, to admit of bringing any substance to 
 operate on the leaf, which will be found ex- 
 tremely obedient to electrical force. The object 
 of this species of electroscope is to ascertain first 
 
 whether a body evinces any electrical sign whatever, or is electrified 
 
 I'ig. 13. Gold Leaf 
 Electroscope. 
 
22 
 
 ACTION OF ELECTROSCOPES. 
 
 in any minute degree. Secondly, to determine the kind or quality 
 of the electricity. In order to ascertain the first, we set a small 
 metallic disc, g, fixed at the extremity of a sliding rod, g h, 
 within a short distance of the suspended leaf R. We then bring 
 the given subject of experiment in contact with a communicating 
 plate or ball, p. If the substance to be tested be in the smallest 
 degree electrically excited, the leaf R is attracted toward the^disc g, 
 which should be in this case uninsulated ; or, conversely, we may 
 make the given substance under examination touch the ball h of 
 the insulated rod g k, so as to communicate with the disc g. So 
 sensitive is this" instrument, that the mere touch of a silk hand- 
 kerchief on the plate or ball p causes the leaf to strike the disc. 
 To determine the kind or quality of the electricity we set the disc g 
 at a convenient distance from the leaf ; and having communicated 
 to the leaf through the ball or plate p a weak charge of either 
 positive or negative electricity, so small as not to be sensible to the 
 disc g at a given distance, we then make the body under exami- 
 nation touch the ball h of the insulated rod g h+ If the electricity 
 be of the same kind as that communicated to the leaf through the 
 plate p, the leaf will be repelled and conversely (27). (Gr.) 
 
 A glass receiver, with one lateral opening, q, as seen in Fig. 13, 
 suffices for ordinary operations. A second lateral opening, on the 
 opposite side, however, as at q', Fig. 14, gives greater experimental 
 facility, since it enables us to oppose the substance itself under 
 examination directly to the suspended leaf through either of the 
 lateral openings q q', or enables us to place the leaf between two 
 
 Fig. 14. Single G-old Leaf 
 Electroscope. 
 
 Fig. 15. Double Gold Leaf 
 Electroscope. 
 
 discs by means of a second slide and insulator, w, similar to the first 
 already described ; the tubular rod w being passed clear through 
 the foot, for the support of the additional slider w. Fig. 14 repre- 
 sents the single leaf electroscope as thus constructed with two 
 lateral openings. 
 
ACTION OF ELECTROMETERS. 23 
 
 33. An extremely sensitive electroscope with two gold leaves is 
 represented in Fig. 15. Two strips of gold leaf, a b, are suspended 
 within a glass receiver, R. Electricity is communicated to the 
 leaves through a light conducting rod and ball, p. When elec- 
 tricity is communicated to the ball p, the leaves a b imme- 
 diately diverge, as indicated by the dotted lines x y. If the 
 electricity of the substance under examination be different from 
 that with which the ball p has been charged, then on making it 
 touch the ball the leaves will immediately close. If, on the 
 contrary, the electricity be the same as that with which the ball 
 p is charged, the leaves will have an increased divergence (27), 
 and thus the kind, and in some degree the force, of the electricity 
 in operation is easily observed. 
 
 The application of leaf gold to the purpose of an electroscope 
 was first suggested by the Rev. Abraham Bennett, F.R.S., Hector 
 of Wirksworth, in Derbyshire, who gives a description of what he 
 terms his Gold~lea/ Electrometer, in a letter to Dr. Priestley, read 
 before the Royal Society, 7th December, 1786, and printed in the 
 seventy-seventh volume of the Philosophical Transactions. Ben- 
 nett's celebrated gold-leaf electrometer, as originally constructed, 
 consisted, as in the present case, of two delicate slips of leaf gold 
 suspended from a conducting rod within a glass receiver (H). 
 
 34. Henley's Quadrant Electrometer. This species of electro- 
 scope was invented by Mr. Henley, F.R.S., so long since as the 
 year 1772. It is represented in Fig. 16. A light reed, cd, termi- 
 nating in a pith ball, d, is mounted in a delicate axis, c, attached to 
 a vertical conducting rod or stem, A B. This axis is in the centre of 
 a graduated quadrant or semicircle, also affixed to the 
 
 same stem. The stem has a ball below at B, against 
 which the terminating pith ball d of the reed reposes, 
 when, in a vertical position. But when electrified by being 9 
 connected with a charged body, the reed rises, and 
 denotes on the graduated arc the angle of divergence 
 by which the comparative amount of charge acting 
 on the reeds may, in some cases, be estimated. The in- 
 strument as commonly employed is not very available as Y 
 an accurate quantitative measure. It is, however, prac- p . 
 tically useful, and may by a little care and management H u e a ^ r e ^ s 
 be occasionally employed as an electrometer with advan- Electrometer. 
 tage (I). 
 
 35. Cavendish's Electrometer. The Hon. Henry Cavendish also 
 employed the principle of diverging reeds as a measure of electrical 
 force. The kind of instrument used by him is represented in Fig. 17, 
 and the following description of it is, for the most part, taken from 
 
THE ELECTROSCOPE ELECTROMETER. 
 
 B0> 
 
 Fig. 17, Cavendish's 
 Diverging Electrometer. 
 
 the manuscripts of this distinguished philosopher, in the possession 
 of the author. Two wheaten straw reeds, a A and b B, each about a 
 foot in length, are set on fine steel pins as axes, movable in notches 
 in an insulated metallic plate, P, or on centres in a ring of support, 
 so as to turn with the axis. The straws are left open below, 
 and carry two light cork balls, or balls of pith, A B, about one-third 
 of an inch in diameter. In order to increase the force with which 
 the straws tend to close when diverging, the lower open ends of 
 the reeds are occasionally loaded with short pieces of wire. Now 
 it is not difficult to find, on ordinary mecha- 
 nical principles, the force with which the reeds, 
 considered as levers, will tend to the perpen- 
 dicular from any given angle of inclination, 
 when light and when loaded. We may thus 
 estimate the relative repulsive power required 
 to maintain the same angle of divergence in 
 these two cases. For example, let the force 
 tending to the perpendicular from a given 
 angle of divergence be four times greater when 
 loaded than when light ; we might infer that if 
 divergent by electrical repulsion, four times the repulsive force 
 must be exerted to sustain the loaded reeds at the same angle as 
 the light reeds. If we suppose the force of gravity in this kind 
 of instrument to be concentrated in the balls themselves, " the force 
 required to separate them will be as their weights directly," so 
 that by a careful manipulation we may, in 
 many cases, with this kind of instrument, 
 measure the repulsive force pretty accu- 
 ^rately (K). 
 
 36. The Electroscope Electrometer. 
 Fig. 18 represents a simple and available 
 instrument invented by the author, and 
 termed by him an Electroscope Electro- 
 meter, as combining, to a certain extent, 
 the object of both instruments. 
 
 An electroscope reed, m n, about a foot 
 in length, terminating in small pith balls, 
 m, n, is mounted on a delicate axis, set on 
 fine points within an elliptical metallic ring, 
 rig. is. Harris's Electroscope T - This electroscopic reed consists of two 
 
 Electrometer. ^^ m ^ n ^ m{i ^ fa^gh the m edi um 
 
 of a central axial-pin at right angles to the axis. The ring, with 
 the electroscope m n, is joined by means of a short projecting 
 
ELECTROSCOPE ELECTROMETER OF DOUBLE REPULSION. 25 
 
 stud and ball to the extremity of an insulating glass arm, d e. 
 The ring and axis allow the electroscope m n free motion in a 
 vertical plane in either direction, which is determined to the 
 vertical by a reed slider on its lower arm ; this slider enables us 
 to adjust the balancing power in such a manner as to measure, 
 with some degree of accuracy, the force in operation. The attrac- 
 tive or repulsive force is developed in opposing a charged body at 
 a given distance, to either ball, m n, of the electroscope, either in 
 a neutral state or charged with one of the electricities. The elec- 
 troscope ball will incline toward the charged ball or recede from 
 it, according to the attractive or repulsive force developed. In 
 either case it will be found more convenient to operate upon the 
 lower ball n rather than on the upper one, m, although either may 
 be employed, according to the circumstances of the experiment. 
 In the case of attractive force the electroscope is neutral. In the 
 case of repulsive force the electroscope is charged either positively 
 or negatively, which may be easily done by communicating posi- 
 tive or negative electricity to the ball e. The amount of attractive 
 or repulsive force is shown by the deviation of the electroscope 
 reed from the perpendicular. In order to estimate the angular 
 deviation of the arm m n, a small quadrantal arc, x y, about 
 4 inches radius, is fixed to the elliptical ring T, immediately 
 behind the central axis. This arc, with its radial arms, is of 
 light mahogany, the arc itself being faced with a similar arc of 
 thin cardboard, divided into 60 degrees on each side of its centre, 
 marked . The Zero line coincides with the index arm of the 
 electroscope, when at rest in a vertical position, The amount of 
 attractive or repulsive force may hence be measured in degrees of 
 the arc included between the Zero line and the line of deviation 
 of the index arm. We thus not only determine the kind of 
 electricity in operation, whether positive or negative, but we 
 have also, in a certain degree, a measure of its amount (35). 
 All we require to know is the distance of the attracting or repelling 
 bodies, the amount of charge in operation, and the relation of the 
 force to the degree of angular deviation, all of which may be cal- 
 culated experimentally (I). It will be sufficient in any case to 
 take the distance between the charged body and the ball of the 
 electroscope. This simple instrument is efficient and convenient in 
 ordinary electrical elementary inquiries. 
 
 37. Electroscope Electrometer of Double Repulsion. This instru- 
 ment, Fig. 19, is constructed on similar principles to the Quadrant 
 Electrometer of Henley (33). A small elliptical metallic ring, c, 
 is set obliquely on a short brass rod, c w, sustained on a sliding in- 
 
26 
 
 THE ELECTROSCOPE ELECTROMETER. 
 
 sulating support, s ; two light metallic wires, c a, c b, proceed in 
 opposite perpendicular directions from the extremities of the long 
 diameter of the ring, and these wires terminate in gilt balls of 
 pith or cork, a b. A delicate axis is set on fine points in the direction 
 of the short diameter, which, by a fine central transverse axis, 
 carries two light reeds, c d, cf, forming together, as in the preced- 
 ing instrument, Fig. 18, one long index, df. This index also 
 terminates in gilt pith balls, df. When unelectrified, the balls 
 df of the index repose against the balls a b of the vertical 
 wires c a, c b ; when electrified, either alone or by 
 connecting the rod c m with a charged conductor, 
 the index is repelled in opposite directions by 
 a double repulsion, above at d and below at f. 
 The amount of divergence is estimated by a 
 graduated quadrant fixed in the centre of the axis 
 of the index behind the elliptical ring c. The 
 tendency of the index to a vertical position is re- 
 gulated by short reed sliders on the opposite 
 arms of the index, and which move on them 
 with friction so as to admit of being placed 
 at various distances from the centre of motion. 
 The electroscope, therefore, admits of a similar 
 measurement of force, as in the case of the electro- 
 scope of Cavendish (35), when loaded and when 
 light ; thus the index diverges with an extremely 
 small force. This kind of electroscope, although 
 convenient and useful in many instances, is, never- 
 theless, of very limited application in others. We 
 are certainly enabled to say that, cceteris paribus, when the 
 angular deviation is the same, the same quantity of electricity 
 is in operation; with a greater divergence there is a greater 
 quantity, with a less divergence, less, but how much greater or 
 less is not easy to determine, since we have to take into the 
 account the diminishing force of repulsion as the distance increases, 
 and the simultaneous increasing force of gravity at different 
 angles, together with all the different and varying distances from 
 the centre to the extremities of the repelling arms, and also the 
 variable oblique action, all tending to complicate any mechanical 
 calculation to which we can subject it, and render it more or less 
 dependent upon empirical experiment. 
 
 Fig. 19. Kama's 
 Electroscope Electro- 
 meter of Double 
 Kepulsion. 
 
OCCASIONAL MEMOEANDA AND EXPLANATOEY NOTES. 
 
 (A) A fine filament of unspun silk from the cocoon of the silk-worm, with 
 a light pith ball at its extremity, is very sensitive of electrical force. A 
 still more delicate test of electrical attraction is a similar suspension by the 
 spider's thread, which may be twisted round some thousand times without 
 exhibiting the least torsion or reactive force. It requires, however, great 
 manipulative skill in the construction of an electroscope of this kind. If we 
 should be fortunate enough to catch a spider in its descent from the ceiling, 
 leaving its thread behind it, there is the electroscope required ; the spider 
 itself being the attached bob, or substitute for the pith ball. By a little 
 care we may cut the spider adrift from its thread, and attach by a little 
 gum a light pith ball to the thread itself. The method of securing light 
 spider's threads for delicate electrical suspensions is very simple. A wood 
 spreader should be prepared after the manner of a pair of compasses, so as 
 to admit of being extended to any required distance, the points of the 
 spreader being touched with a little weak gum, the observed spider's thread 
 is carefully caught between them, and the thread subsequently attached to 
 the electroscopic pendulus body, and to tlj.e given point of suspension. 
 
 (B) This needle is constructed of fine silver wire, about one-twentieth of an 
 inch in diameter, and 10 inches in length. The centre of motion is near one 
 extremity of this wire, dividing the needle into two arms, c a> c &, of unequal 
 length. At the centre of motion c the wire is bent into a small curve, and 
 a fine point soldered to the middle of the curve, projecting downwards, so 
 as to allow the system to turn freely on it, preserving at the same time 
 the point of suspension above the centre of gravity of the mass. The ex- 
 tremity of the long arm of the needle carries a light ball of the pith of elder, 
 about three-tenths of an inch in diameter, fixed on its point ; the extremity 
 of the short arm carries a balance -weight, &, which may consist of sealing- 
 wax neatly run into a large globule of sufficient size to preserve the needle 
 when suspended in a horizontal position : the whole is sustained on a hard 
 metal centre, united to a small vertical support of brass or of glass , d. Any 
 electrified substance presented to the pith ball a, at the extremity of the 
 long arm of the electroscope needle, affects it powerfully from a considerable 
 distance. 
 
 (C) The strip of leaf gold may be about 3 inches long and from a quarter 
 to half an inch wide. It should be prepared, and cut fair upon a gilder's 
 cushion. The most simple and extemporaneous way of suspending such 
 a strip of gold is to attach it to a short slip of moistened gummed paper ; 
 and insert the paper-holder in a cleft stick of what is called skewer-wood, 
 which, being neatly tapered, may be passed into a bung of fine cork ; the 
 bung being pressed upward in the open neck n of a bell glass, the gold 
 leaf downwards. The whole may now be sustained from a short arm of 
 support by means of a slight clip of wood or metal, as shown in the figure, 
 being sustained on an insulated rod and foot, M. 
 
 (D) A good metallic amalgam for promoting the excitation of glass 'is 
 obtained by melting together in a clean iron ladle 1 part tin, 2 parts zinc, 
 and 4 parts mercury, the proportions being 1,2, and 4, or double of each 
 other. The zinc should be just heated, and when nearly at the point of 
 fusion the tin should be added ; the two metals will run freely together. 
 The temperature of the mercury being raised to somewhat short of its boil- 
 ing point, should be poured on the fused metals, stirring the mass at the 
 same time with an iron rod. The resulting amalgam, after being allowed 
 
28 PREPARATION OF APPARATUS. 
 
 to cool somewhat, should be now poured into a covered wooden box, and 
 kept in a state of agitation until cold, when it will be commonly found in 
 the state of a black powder, or otherwise in a soft metallic mass, which after 
 a time will harden, and may be pulverised in an iron mortar. The triturated 
 substance should now be passed through a muslin sieve. When wanted for 
 use a little of it should be rubbed up with some purified lard, and applied 
 to the surface of the rubber with which the glass is to be excited. For 
 a more especial account of the preparation and application of metallic 
 amalgams and other substances in promoting electrical excitation we must 
 refer to the chapter on Electrical Manipulation (85). 
 
 (E) Insulated supports of various kinds and altitudes are frequently called 
 for in Experimental Electricity. Fig. 20 represents an insulating support, 
 
 the altitude of which may be varied to any required extent, 
 without interfering with the insulation. It may be termed 
 a sliding insulator. It is a very efficient and convenient aux- 
 iliary electrical instrument. Its construction is as follows : 
 n o is a solid cylindrical rod of glass or vulcanite, of any con- 
 venient length and thickness. It is movable within a varnished 
 glass tube, t, capped by a small ball of wood, v, through which 
 the rod passes. This insulator is movable within the glass 
 tube t, after the manner of a piston-rod, having a compressible 
 cylinder of fine cork at its lower extremity, o, covered by a layer 
 or two of silk ribbon, which enables it to slide freely in the tube. 
 There is a small stop-screw, v, in the wooden ball through which 
 the insulating-rod passes, by which it may be secured at any 
 given height. The body to be insulated is placed on the upper 
 extremity, n, of the insulated rod, by means of any convenient 
 holder attached to it. For the support of spherical bodies a 
 
 Fig 20 Sliding concave holder such as an inverted watch-glass will be 
 insulator. found convenient. Plane surfaces may also be sustained upon 
 
 inverted spherical segments ; and thus the altitude may be regulated to 
 
 any convenient height. The glass tube t is fitted in a neat foot and holder, 
 
 a (or other basis), furnished, if necessary, with a ring of lead to give it 
 
 stability. 
 
 For general purposes the rods of these sliding insulators may be about a 
 
 foot in length and from one-eighth to one-quarter of an inch in diameter. 
 
 The glass tubes within which they move may be from 5 to 8 inches in 
 
 length and from '3 to '8 of an inch in diameter. 
 
 (F) Two hollow cylindrical tubes, one of glass, the other of gutta-percha 
 (20), are very useful in Practical Electricity, since they furnish a ready 
 means of producing the opposite forces. These tubes should be well 
 selected and carefully prepared ; they may be from 20 inches to 2 feet in 
 length, and a full inch in diameter. They can be powerfully excited, and 
 pretty strong sparks may be obtained from them. The old electricians 
 confined their experimental researches within the limits of the powers of 
 excited glass tubes. These were, in fact, their electrical machines, and they 
 took especial care of them, and enclosed them, when not in use, in paper or 
 woollen envelopes. Glass tubes for electrical excitation should be rather 
 thin than otherwise. When in good condition a gentle excitation with a 
 cloth smeared with metallic amalgam (7) will be sufficient. 
 
 (G) The glass R, Fig. 13, within which the leaf is suspended, may be of the 
 form given in the figure. It has two contracted necks, with open mouths, t u, 
 about 2^- inches in diameter, and a lateral mouth opening, q, about the same 
 diameter. The dimensions of this circular glass screen may be about 6 inches 
 in diameter, and 8 inches in height. It should be mounted on a firm foot, E, 
 
PREPARATION OF ELECTROSCOPES. 29 
 
 so as to admit of a small tubular rod of brass, w, being passed horizontally and 
 partially through it beneath the glass. This rod carries a short slider, w, and 
 insulating support, upon which is sustained a light movable conducting rod, 
 g h, which enters the receiver through the lateral opening q. The rod g h has 
 fixed to its inner extremity the gilt disc g, about half an inch in diameter 
 and one-tenth of an inch thick, and terminates externally in a conducting 
 ball, h ; by means of the slide w the disc g may be set at any convenient dis- 
 tance from the surface of the suspended leaf. The leaf is appended from a wide 
 metallic pincer, at the extremity of a short metallic rod, passing through an 
 insulating glass tube, d, sustained within the receiver through a varnished 
 cork in its upper neck, t. The suspension rod terminates above in a brass 
 ball, p, about three quarters of an inch diameter, or in a small brass plate, 
 about 2 inches diameter, and through which electricity may be commu- 
 nicated to the leaf R. 
 
 (H) The Portable Diverging Electroscope. A simple and useful modification 
 of the Double Gold-Leaf Electroscope of a portable kind is obtained by sub- 
 stituting two light diverging reeds or pith balls 
 for the gold leaves. A small brass rod, m n y 
 about 4 inches in length and about one -tenth of 
 an inch in diameter, is supported through a cen- 
 tral ball, d, insulated at the extremity of a short 
 glass rod, c d ; the extremities of the brass rod 
 are capped by small metallic balls, m n, about 
 one-eighth of an inch in diameter. Two light 
 threads, or very light straw reeds, m o, mp, 
 terminating in small pith balls, op, are sus- 
 pended from the ball m ; these are very deli- 
 cately hung to the ball, either by very fine " ig . 2 i. Portable Diverging 
 wire ring joints or small silk filaments; so that Electroscope, 
 when the ball n at the opposite extremity of the brass rod m n touches an 
 electrified body (the instrument being held by its glass insulator, c d), the reeds 
 or threads mo, mp, diverge, or, if divergent, tend to come together ; thus 
 showing not only the excited state of the body touched, but the kind of 
 electricity with which the given substance is affected. 
 
 (I) Henley's Electrometer is often roughly and ill-constructed, and is 
 hence generally a somewhat rude instrument. The most refined method of 
 construction is as follows: The stem A B, Fig. 16, should be a small 
 silvered brass tube about 6 inches long and one-eighth of an inch in diameter, 
 terminating at its upper extremity in a small silvered or wood ball, A, about 
 half an inch in diameter. A light straw reed, c d, about 5 inches long, is 
 affixed to a delicate axis, in the centre of an elliptical ring, c, attached to 
 the stem A B, about an inch below the terminating ball A. The axis is set 
 on fine watchmaker's points, so that the reed has great freedom of motion ; 
 and has a short pin passing centrally through it. The under portion 
 of this pin carries the electroscope reed c d ; the upper portion is furnished 
 with a small counterpoise ball about one-tenth of an inch in diameter. 
 A light quadrantal arc, q, divided into 90 degrees whose centre is the centre 
 of the axis carrying the reed is fixed behind the elliptical axial ring to 
 the stem A B. This arc should be an open segment about 2^ inches radius 
 and a quarter of an inch wide on its circumference. It may be of ivory or 
 varnished cardboard glued upon a light wood segment. A pith ball, d, of 
 about a quarter of an inch in diameter, is fixed on the extremity of 
 the reed, and reposes, when the reed is in a vertical position, against a 
 small metallic ball on the lower part of the stem. The whole when in 
 use is sustained on a stout metal point, proceeding from the electrified or 
 
30 INDICATIONS OF ELECTROSCOPES. 
 
 charged body. As the charge accumulates the reed rises and marks upon the 
 graduated arc the progress of the charge, as already observed. 
 
 Much elaborate investigation has been expended upon the value of the 
 indications of this instrument, but not with any great advantage. We may, 
 however, put some confidence in the following general principles : It is a 
 well-known mechanical fact that the momentum or moment of the force with 
 which a pendulous body delicately suspended tends to the vertical, is 
 measured by the weight of the body multiplied into the distance of its centre 
 of gravity from point of suspension, multiplied into the sine of the angle of 
 inclination. Having in this way ascertained the value of the mechanical 
 force in operation for any given angular divergence of the reed, we are 
 enabled so far to employ the instrument as an electrometer of measure. The 
 centre of gravity of the reed with its pith ball and axis may be determined, 
 as practised by Cavendish, by balancing the reed across a fine knife edge. 
 If the reed, therefore, be furnished with a small slider set at different 
 distances from the centre of motion, the distance of the centre of gravity 
 from the point of suspension will vary, and thus by a simple graduation of 
 the reed we may estimate the mechanical force at a given angle. This is 
 very simple, but whether it is all we require it is not easy to determine. 
 The inclination, for example, of the repellent reed to the vertical pillar 
 undergoing continual change, the electrical force not only acts at variable 
 distances from the vertical, but also acts more or less obliquely. Hence it 
 does not at all follow that the quantity of Electricity producing a double 
 divergence would be also double. In order, therefore, to arrive at an 
 exact quantative measurement we are in a great degree dependent upon 
 empirical experiments. Cavendish states in his manuscripts that when this 
 electrometer is considerably elevated on a long stem above a charged 
 conductor on which it is usually placed, the indications will be different 
 from those obtained when close upon it. In the first case the electrometer is 
 more sensible at the beginning of the motion ; in the latter case it is less 
 sensible. The difference by experiment with similar quantities of electricity 
 was found to be considerable ; thus, when close upon the conductor the 
 divergence was only 5 degrees, when elevated the divergence amounted to 
 21 degrees. On the other hand, towards the close of the motion the sensibility 
 of increase was found to be less in the elevated than in the low position. The 
 safest method, therefore, perhaps is to determine experimentally for each 
 particular instrument the angular quantity corresponding to a given charge, 
 and to estimate accordingly. 
 
 (K) In applying this electrometer, Fig. 17, we suspend it at a distance 
 of about 6 inches before a strongly-marked pasteboard scale, the eye 
 being situated about 30 inches before the scale, so that by means of an eye- 
 piece the angle of divergence is easily observed. The straws a A, & B, reach 
 nearly to the bottom of the cork or pith balls A B, but not quite ; the lower 
 ends of the small wires, therefore, with which the reeds are occasionally 
 loaded may be just even with the surface of the ball, being retained in their 
 places by a little soft wax. If, instead of loading the reeds with weights, we 
 furnish them with small graduated sliders, it would be easy to accommodate 
 the centre of gravity of the reeds to any given angular divergence. The 
 electrometer as employed by Cavendish, when loaded with the wires, required, 
 according to his calculation, three times the force to maintain the same 
 angle of divergence than when not so loaded ; in which case Cavendish in- 
 ferred, on the principles already noticed (I), that with the same angle of 
 divergence the repulsive forces were in these cases at 3.1. 
 
L I B U A R Y 
 
 UNIVERSITY OF 
 
 CALIFORNIA. 
 
 
 CHAPTER II. 
 
 INDUCTION. 
 
 38. The preceding elementary facts being fully understood, we 
 are prepared to enter on a more advanced class of phenomena, 
 remarkable for their singular interest and importance, namely, 
 electrical action at a distance, or Induction as it is called, or by 
 the French, Electricite par influence, also Electricite dissimulee. 
 Hitherto our attention has been limited to the'development of certain 
 electrical forces of attraction and repulsion, either by simple excita- 
 tion (7), or by immediate communication (10). We now proceed 
 to consider a new species of electrical force, consisting, apparently, 
 in the development of similar powers by the influence of excited 
 bodies on other bodies beyond the limits of contact ; in short, a species 
 of apparent electrical emanation or sympathetic action, operating 
 between bodies at sensible, and even at considerable distances. 
 
 39. This new species of electrical action is well exemplified in 
 the following experiments : 
 
 Exp. 13. Let an insulated neutral conductor, Q, Fig. 22 (which 
 may be a light hollow cylinder of gilded wood, about 5 inches 
 in length, and from 3 to 4 inches in diameter), be directly opposed 
 to a similar insulated conductor, p, charged either positively or 
 negatively (30), suppose positively. Under such circumstances 
 an action, apparently of a sympathetic kind, arises at a distance 
 between these conductors, p and Q. The result is to change the 
 actually existing electrical state of the two bodies P and Q. The 
 previously neutral body Q, without any direct communication 
 with P, evinces a state of excitation so that a delicate neutral 
 electroscope, e (36), placed near the remote surface p' of the body 
 Q, becomes attracted towards Q, as indicated in the figure. On the 
 other hand, a similar electroscope, f y repelled from the remote 
 face q of the body p, falls back towards p in a greater or less 
 degree, thereby showing a decrease in the repellent force of p at 
 its distant extremity, q. 
 
32 ACTION OF INDUCTION. 
 
 The state thus indicated in Q is a peculiar state ; its near face, n, 
 immediately opposed to the charged body P, assumes an opposite 
 electricity to that of p, whilst its remote face p indicates the same 
 electricity. Thus, supposing p to be charged with positive or 
 vitreous electricity, the face n of the body Q will be negative, 
 and the face p positive. This change or disturbance of the 
 neutral condition of Q, however, is not all ; it is further attended 
 by a reciprocal or similar influence reflected back from the 
 
 Fig. 22. Apparatus for showing some effects of Induction. 
 
 body Q, upon the charged body p, shown, as just observed, in 
 the diminished divergence of the electroscope /. This reflected 
 influence, therefore, is such as to increase the force in the charged 
 body P, at its face, p, directly opposed to the face n of the 
 body Q, and diminish the force of its face, q> distant from the 
 face n of the opposed body, so that the immediate result of 
 the induction is a succession of comparative alternate positive and 
 negative states. These states may be tested by means of a 
 carrier (L). 
 
 40. This succession of positive and negative force by inductive 
 influence, may be still further, and perhaps more directly, demon- 
 strated in the following manner : 
 
 Exp. 14. Let two cylindrical conductors, p, Q, Fig. 23, similar 
 to the former, be fitted with movable faces, q, p, n> p, sustained in 
 position on slender rods of varnished glass, a, b, c, d, inserted in 
 small movable feet upon sliding bases. Let the two bodies be 
 opposed to each other at some given distance, as indicated in the 
 figure, the body P being positively charged. When so charged 
 withdraw in succession the faces, p , n, p, q, examine their elec- 
 trical states through the medium of one of the gold leaf electro- 
 
PHENOMENA OF INDUCTION. 33 
 
 scopes already adverted to (32), weakly charged with positive or 
 negative electricity. These faces will evince, as just observed, a 
 succession of positive and negative states ; that is to say, the 
 
 Fig. 23. Induction Apparatus. 
 
 face p will be positive, the face n negative, the face p again 
 positive, the face q less positive (M). 
 
 41. We see by these experiments (39, 40) that when an elec- 
 trified body is directly opposed to a neutral body, two induc- 
 tive actions ensue. First, we have a direct induction of the 
 charged body p upon the neutral body Q, changing its neutral 
 state into a state of excitation ; secondly, the neutral body thus 
 influenced reacts in its turn upon the charged body, and changes 
 the actual condition of the charged state of that body. The first 
 of these actions may be considered as direct induction ; the second, 
 or reactive state, may be termed reflected induction. Both these 
 inductions will be more apparent in some inverse ratio of the dis- 
 tance between the bodies, in some direct ratio of the quantity of 
 electricity in operation, and the superficial extent of the two bodies. 
 Thus, if the body Q opposed to the body p (39), instead of being 
 a small cylinder of equal dimensions, have very great exten- 
 sion, its reflected induction upon the body p will be considerably 
 increased. If Q be put in communication with the ground, so as 
 to give it unlimited extension, or inductive capacity, the influence 
 upon P is the greatest possible. Thus the diminished repulsive 
 force upon the electrometer/ 1 , by the influence of the body Q upon 
 p, will be the greatest possible when Q has unlimited extension ; 
 that is to say, when its distant face p is connected with the earth ; 
 so that the influence of a neutral on a charged body would in great 
 measure depend upon what may be termed its inductive power. 
 Reciprocally, the electrical change which the charged body p 
 undergoes by the influence of the body Q, excited by direct 
 induction, will mainly depend upon the extent of its charged 
 surface. 
 
 42. The ordinary phenomena of electrical attraction are inva- 
 
 D 
 
34 ATTRACTION, REPULSION, AND INDUCTION. 
 
 riably attended by these different inductive actions. When, for 
 example, a charged body (39) exerts an attractive force upon a 
 neutral body, it first changes the neutral state of that body into 
 an excited state. It calls up in the near surface of the neutral 
 body an opposite electricity to that of itself, while the neutral 
 body, thus excited by induction, reacts upon the near surface of 
 the charged body. The two approximated surfaces being hence 
 in opposite electrical states, the bodies if all impediment to motion 
 be removed immediately approach each other (27). 
 
 43. The phenomenon of electrical repulsion is attended by 
 similar inductive changes in the repellent bodies. 
 
 For example, let two bodies (39) be charged with the same 
 electricity suppose positive and be directly opposed to each 
 other at a given distance. Inductive actions immediately ensue 
 between the two bodies similar to those just alluded to. One of 
 the bodies endeavours to reverse the electricity of the other, and 
 to excite in its near surface an opposite electricity to that with 
 which it is already charged. The two bodies resist these changes, 
 both being charged alike ; so that, if the existing forces are suffi- 
 ciently powerful to withstand the new inductive actions brought to 
 bear on them, the two bodies recede from each other, and apparent 
 repulsion ensues, there being no impediment to motion. Thus 
 it is that similar electricities are said to repel each other (27). 
 "We see, therefore, that in the phenomena of electrical attraction 
 and repulsion, bodies are first rendered attractable or repellent of 
 each other by induction, and then attraction or repulsion follows 
 as a subsequent result. Hence we may infer, as an especial charac- 
 teristic of electrical force, that the attracted or repelled bodies are 
 first rendered attractable or repellent before attraction or repulsion 
 ensues. 
 
 44. Electrical attraction and repulsion, together with their 
 attendant inductive phenomena, admit of the following elegant 
 and striking experimental illustrations. 
 
 Exp. 15. N, Fig. 24, represents a light disc of gilt wood about 3 
 or 4 inches in diameter, and one-twentieth of an inch in thickness, 
 insulated on a slender rod of glass or vulcanite, at the extremity 
 of a balanced cylindrical rod, m N, of light wood. This rod is 
 mounted on a delicate vertical axis, #, set in an elliptical metallic 
 ring, so as to be susceptible of circular movement in a horizontal 
 plane. A radial arm, b d, projecting from a movable centre, 
 revolves about a central pivot-foot fixed in a strong base of sup- 
 port, R s. The foot carries a central vertical insulating rod of 
 
REFLECTED INDUCTION. 
 
 35 
 
 Fig. 24. Induction and Attraction Apparatus. 
 
 support, a b. A second equal similar disc, P, insulated on a 
 slender glass rod, at the extremity of the projecting radial arm b d, 
 is opposed to the disc N. Each disc has a strip of gold leaf, n', p , 
 attached to its distant face, after the manner of an electroscope 
 leaf, so as to hang freely and parallel along the faces of the discs, 
 and diverge readily when the discs 
 are electrically excited. The disc 
 p being turned away at a dis- 
 tance from N, and charged either 
 positively or negatively say posi- 
 tively its gold leaf, p, will im- 
 mediately be repelled from the disc, 
 as indicated in the figure. Whilst 
 thus divergent, turn the arm b d, 
 and cause the charged disc p to 
 approach the neutral disc N. As 
 the discs approach each other, the 
 leaf ri of the disc N will evince a state of excitation by induction. 
 At this instant attraction will ensue, and N will move toward p ; 
 and thus the simultaneous operation of the inductive and attractive 
 forces are elegantly and experimentally demonstrated. The re- 
 flected induction (41) of N upon P will be also more or less apparent 
 as the two discs approach each other, so that the gold leaf p will 
 tend to collapse as the leaf n' diverges. This last, or reflected 
 inductive action, however, is not so apparent as the direct induc- 
 tion, owing to the limited extension of the attracted disc N. 
 In order to render both the inductive actions very sensible, the 
 attracted body N should have great extension (41). This it is diffi- 
 cult to effect without inconveniently loading the balanced rod m N. 
 "We may, however, remedy this inconvenience by substituting for 
 the disc N a very light gilt paper cylinder, about 4 inches long and 
 3 inches diameter, in which case its reflected induction upon p 
 will be more sensible. 
 
 The following experiment, demonstrative of the reflected in- 
 duction, will be perhaps equally, if not more satisfactory than the 
 preceding, as being more sensible in its operation. 
 
 Exp. 16. Turn the discs p, N, away from each other, as at first. 
 Let the disc P be connected with the ground. Charge the disc N 
 with one of the electricities, say positive. Let the disc P, thus 
 neutral and free, be caused to approach the charged disc N. The 
 gold leaf, n', will immediately tend to collapse, in consequence of 
 the reflected induction of P upon N, at which instant attraction 
 
 D2 
 
36 EXPERIMENTS ON INDUCTION. 
 
 ensues between the two discs. If, instead of placing the disc P in a 
 free state by connecting it with the ground, we give it considerable 
 extension (41) by connecting its charged face with an insulated 
 cylinder about 4 inches in diameter and 5 or 6 inches long, both 
 ihe direct and reflected inductions may be experimentally observed 
 oy the electroscope gold leaves attached to the respective discs, 
 which will be simultaneously affected. 
 
 Exp. 17. Place the discs P, N, far apart. Charge the one, P, 
 positively, the other, N, negatively. The leaves will then be re- 
 pelled from the discs, one with positive, the other with negative 
 electricity. Let the disc P approach the disc N, as before ; reflected 
 inductions from the discs upon each other will arise, and the 
 leaves tend to collapse by the reciprocal action of the positive and 
 negative electricities, but at a greater distance than in the former 
 experiment, and attraction will again ensue. 
 
 Exp. 18. The two discs being neutral, and at a distance from 
 each other, charge them with the same electricity, either positive 
 or negative suppose positive. The electroscopic leaves will diverge 
 equally. Let the discs P, N, thus charged, approach each other. 
 The electroscope leaves will exhibit an increased divergence ; and at 
 the same instant the disc N will recede from the disc p, the two discs 
 being apparently repulsive of each other. It is, however, a question 
 with some electricians whether this be an instance of pure repul- 
 sive force, or whether the two discs do not recede from each other 
 in consequence of their respective attractions and inductions upon 
 surrounding neutral matter more immediately influencing the 
 distant surfaces of the oppositely charged discs. 
 
 If we substitute for the conducting discs p, N, light discs of mica 
 or other non-conductor little susceptible of electrical induction, the 
 attractive or repulsive force is not so apparent. Hence it follows 
 that electrical attraction is most sensible between bodies having 
 large inductive capacities (41). 
 
 45. Such are a few of the more elementary phenomena of elec- 
 trical action at a distance, as exhibited in opposing to each other a 
 neutral to a charged body, or otherwise two charged bodies, either 
 charged with the same or with opposite electricities. These phe- 
 nomena, although of extensive application, constitute a portion 
 only of the whole inductive process. In order to understand more 
 fully this kind of electrical action, and the importance of its appli- 
 cation in the propagation of electrical force, so as to cause it to 
 assume the appearance of action at a distance, it becomes necessary 
 to take note of what is going on in the insulating medium, such as 
 the air, by which conducting bodies, said to be charged with elec- 
 
NATURE OF INDUCTION. 37 
 
 tricity (13), are necessarily surrounded and separated from each 
 other. 
 
 Exp. 19. Let s, Fig. 25, be a spherical or other conductor, in- 
 sulated in free atmospheric space. Charge this insulated body 
 with either positive or negative elec- 
 tricity suppose positive. We have 
 then a stratum of atmospheric particles, 'fy 
 + + + +> &c. &c., immediately surround- //./ 
 ing its surface, which, if the electricity 
 communicated to the sphere s be positive, ' , ^ 
 
 will also be positive. The stratum of elec- \\\^ 
 trined particles, + + -f +, &c., now ope- 
 rates by- induction upon the next stratum 
 of adjacent particles, causing in them a ( 
 development of the opposite or nega- Fig ' 25 ' Di 
 tive electricity, ... The negative electricity thus called 
 forth excites, in its turn, positive electricity in the next 
 adjacent particles, until the electrical wave, or alternating 
 series of positive and negative particles, reaches distant neutral 
 matter, A B c D, causing electrical force to appear there of an 
 opposite kind to that of the sphere s (40) ; so that, in fact, we 
 have a series of disturbances between the charged conductor 
 and distant neutral matter, in the form of an alternating series 
 of positive and negative forces, and which may be represented 
 as in Fig. 26, in which p N may 
 represent portions of opposed 
 conducting bodies, with an inter- "*" 
 vening insulating medium, such 
 as the air. Let abed, &c., stand 
 for and represent adjacent or con- 
 tiguous particles of the insulating Fig - 26 - &*&** to illustrate induction. 
 air. If, in this case, P be charged positively, then we have some 
 such change as that just alluded to ; that is to say, an alternate 
 series of positive and negative forces, as indicated by the black 
 and white portions of the particles abed, &c. ; these reaching 
 eventually the distant surface N, cause a force, N n, to appear there, 
 opposite to that of P p, from which the inductive action is sup- 
 posed to proceed. 
 
 46. In the charged sphere s, Fig. 25, this kind of action, sup- 
 posing the sphere to be in free space, extends equally all 
 around. 
 
 Exp. 20. Place around and at equal distances from the insulated 
 metallic sphere s, Fig. 27, about 5 inches in diameter, four or more 
 
 -+ -4- i- -4- 4- i- 
 3 c d e -P 
 
 -4- -4 -4- -4- -4- .+ 
 
38 LANGUAGE OF INDUCTION. 
 
 insulated conducting discs, abed, each, about 3 inches in diameter, 
 with a light electroscope reed and pith ball appended to each of 
 them, as shown in the figure. Charge the sphere either positively 
 
 or negatively. Each of the discs will 
 be acted on inductively, and the elec- 
 troscope reeds will all equally diverge, 
 evidently showing a sort of radiation 
 of power from the central sphere s in 
 all directions. 
 
 47. With a view of more clearly 
 identifying these several actions, Fara- 
 Fig. 27. sphere and Electroscopes. day, with his usual precision, has 
 applied to them a comprehensive and appropriate species of no- 
 menclature. First. It being important, when two insulated con- 
 ductors are in inductive relation to each other, to distinguish the 
 charged body, or surface originating and sustaining the induction, 
 from the opposite or neutral body, in which the action terminates, 
 or to which it extends (45), it is agreed of the two bodies, to call 
 the charged body the inductric or inductive body ; and the opposed 
 body in which the action terminates, the inducteous body. Second. 
 The medium separating the bodies is denominated the dielectric 
 medium ; the term dielectric being employed to designate any kind 
 of insulating substance through, or across, which the electric forces 
 operate. Third. The peculiar induced state (45) of this medium is 
 termed a polar state ; and the medium itself is said to be polarised. 
 Electrical polarity, therefore, is such a disposition of force as 
 gives to the same particle opposite electrical powers. Induction, 
 then, consists in a polarised state of the particles of a dielectric 
 medium, caused by the inductric body sustaining the action ; and 
 in which medium positive and negative forces become symmetri- 
 cally arranged. By contiguous particles, we are to understand 
 particles next to each other, whether we imagine them in actual 
 contact, or whether there be something or nothing between 
 them. 
 
 48. Under ordinary conditions, induction may be supposed to 
 consist of a species of electrical force resulting from the action of 
 matter charged with electricity through insulating matter, tending 
 to produce in the surrounding dielectric particles, a series of oppo- 
 site electrical states, and this it does by first polarising the particles 
 next it. These, in their turn, polarise the next particles, and so 
 on, until the action propagates itself from the charged surface to 
 any distant mass, where the contrary force is caused to appear. 
 Hence it is only through, or across, insulators that induction is 
 
ACTION OF CONTIGUOUS PARTICLES. 39 
 
 sustained. We have, according to this view, no such thing really 
 as electrical action at a distance, or any kind of sympathetic force 
 between particles or masses separated from each other by empty 
 space. 
 
 Electrical influence is an actual propagation of force through 
 the medium of intermediate matter. The distinguishing difference 
 in the polarisation of an electric substance, and the polarisation of 
 a conducting substance, consists in this : the first is a polarisation 
 of forces, which do not communicate their forces to each other ; 
 the second is a wholesale polarisation, as it were, of the entire 
 mass (40), so that opposite portions of the body exhibit opposite 
 electricities ; and this arises from the circumstance that conducting 
 particles will not admit of the same constrained state as non- 
 conducting particles. It is to be further observed that induction 
 upon conducting bodies does not require sensible thickness for the 
 development of the positive and negative states above alluded to ; 
 the thinnest leaf gold may by induction become positive on one 
 surface and negative on the other, without any interference what- 
 ever of the opposite electricities with each other ; that is, so long 
 as the source of the induction remains. Conducting bodies are, in 
 fact, in mass, as regards induction, what electric or non-conducting 
 bodies are in particles. 
 
 49. He verting for a moment to a further consideration of the 
 inductive phenomena with opposed conductors, we find them all 
 resolvable into the propagation of electrical force across an insu- 
 lating or dielectric medium (48). If we take a positively charged 
 conductor standing alone in free atmospheric space, then, as in the 
 case of the charged sphere s (45), we have a polarisation of the 
 surrounding atmospheric medium in all directions, which would 
 finally extend, like a wave, to other conducting matter, however 
 distant, in which an opposite electricity to that of the charged 
 body would eventually appear. Cavendish traced this from a small 
 electrified globe placed in the centre of a room sixteen feet square 
 to its surrounding walls. Faraday traced it from a ball, suspended 
 in the middle of a large apartment, to the walls distant twenty- six 
 feet. Such would be the case supposing the charged body to be 
 alone in free atmospheric space. Directly, however, we bring into 
 the field a second body, and place it near the charged body (39), 
 then the induction becomes determined, more especially upon the 
 second or near body ; and the action is so confined to the adjacent 
 surfaces, and the insulating medium immediately between them, 
 that the induction in other directions frequently becomes so small 
 as to admit of being neglected. 
 
40 TWO ELECTRICITIES ALWAYS PRESENT. 
 
 The decreased divergence of the electroscope/ (39) is strikingly 
 illustrative of this exclusive reciprocal action. This diversion, or 
 compensation of force, by the reflected influence of near neutral 
 matter upon a charged body, is characteristic of electrical induc- 
 tion, or electrical action at a distance. 
 
 Exp. 21. Let a neutral body, o, Fig. 28, in a free state, be 
 caused to approach a charged mass, M, whilst attracting a distant 
 
 body, n. The influence of the near 
 body o is such as to weaken or neu- 
 tralise the attractive force between the 
 body M and the distant mass n, so as 
 frequently to supersede the action be- 
 tween M and n altogether. 
 
 50. The simultaneous presence of two 
 opposite electricities in every instance 
 of electrical excitation (24), is a 
 
 Fig. 28. Induction Apparatus. 
 
 striking and very important feature in 
 
 the operation of electrical force, however produced. These two 
 apparently constituent elements of electrical agency (in whatever 
 that agency consists) seem to be inseparable, one having no exist- 
 ence independently of the other. Thus, in the communication of 
 electricity to simple insulated conductors, or in the development 
 of electricity by friction, or in electrical accumulation by any 
 other means, it is impossible so completely to sever the two elec- 
 tricities as to get rid of one of them, and leave the other free to 
 act alone. We may, it is true, electrify a body either positively 
 or negatively, and in this sense may be said to have charged the 
 body with one of the forces only ; but this is not an independent 
 condition. If, for example, we excite a glass tube, we develop 
 positive electricity on the glass. This positive electricity, which 
 commences at the points of friction, becomes at once related to an 
 equal amount of negative electricity developed in the rubber itself 
 (23), and also to distant negative electricity called up by induction 
 in surrounding matter. 
 
 51. This elementary and wonderful fact has been well shown by 
 Faraday, and is apparent in the following experiments : 
 
 Exp. 22. Let a c d, Fig. 29, be a cylindrical vase of thin metal, 
 about 7 inches in diameter and 1 foot deep, suspended from any 
 convenient support by means of fine insulating lines of varnished 
 silk gut, or other insulators, a m, d n. Let p be a light metallic 
 ball about 3 inches in diameter, suspended by a fine insulating- 
 line, p h. Let E be a double leaf electroscope communicating 
 
EXPERIMENTS ON THE TWO ELECTRICITIES. 
 
 41 
 
 Fig. 529. Vase and 
 Electrometer. 
 
 with the outer surface of the vase a c d. Raise the ball P by its 
 insulating line P h out of the vase, and communicate to it a 
 charge of either electricity, suppose positive (21). Then return 
 the ball to its place within the vase. As the ball descends 
 into the vase, the gold leaves of the electroscope E 
 begin to diverge, and continue to diverge, until 
 the ball is about 3 inches below the edge of the 
 vessel, when the divergence remains constant. 
 Now, the exact position of the ball p within the 
 vase is of no moment whatever, whether it be 
 near or distant, from either the sides or the bot- 
 tom, provided it be not so near as to discharge 
 its electricity against the vase. Under this con- 
 dition, place it where you will, the divergence of 
 the leaves remains unchanged. The sum of the 
 forces of induction, therefore, is the same constant 
 quantity. 
 
 Again : raise the charged ball P by its insulating 
 line, P h, clear of the vase. The leaves of the elec- 
 troscope E will now close, but will again diverge on restoring the 
 charged ball to its former position. 
 
 If we now examine the kind of electricity communicated to the 
 electroscope E from the outside of the vase, it will be found to be 
 the same as that of the ball p ; that is to say, positive. The in- 
 ductive influence of the ball p, therefore, has been such as to deve- 
 lop on the outside of the vase the same electricity as that of the 
 ball itself; namely, positive. If we now examine the interior 
 surface of the vase by means of a small insulated carrier (14), 
 raising the electrified ball p without the vase whilst the carrier is 
 in contact with its interior surface ; then, on removing the carrier, 
 we find the interior surface of the vase negative. We see, there- 
 fore, that the two sides of the vase are placed by the inductive 
 influence of the ball p in opposite electrical states ; the surface 
 next the ball being negative, the distant or outer surface posi- 
 tive ; phenomena strictly in accordance with the teaching of (40). 
 That this is a pure result of inductive influence from the charged 
 ball P is evident from the fact that everything becomes restored to 
 its original state on withdrawing the ball P, or discharging its 
 electricity. 
 
 Exp. 23. The apparatus being the same as in the last experi- 
 ment, but in a neutral state, charge the ball p whilst raised clear 
 of the vase, positively, as before ; and when so charged, lower it 
 
42 ELECTRICITY ON THE OUTER SURFACE. 
 
 carefully into the interior of the vessel, so as to remain suspended 
 about its centre. The leaves of the electrometer will again diverge. 
 Whilst thus divergent, cause the charged ball p to touch the in- 
 terior surface, so as to communicate to the cylindrical vase its own 
 proper positive charge. Not the least change will take place in 
 the divergence of the gold leaves. They will still remain divergent 
 to the same extent. If, therefore, the ball p be merely suspended 
 in the interior of the vase a c d, it acts upon the vase by induc- 
 tion, and evolves electricity of its own kind on its distant or outer 
 surface. The electricity, therefore, evolved upon the outer surface 
 a c d, by the inductive action of the ball P, must be exactly equal 
 to that with which the ball P is actually charged, without which 
 the divergence of the leaves of the electroscope E must necessarily 
 change. This is further evident from the fact that, on with- 
 drawing the ball p from the interior of the vase, after touching its 
 inner surface, the ball p comes away uncharged and perfectly 
 neutral, which could not be the case if it had not parted with all 
 its charge to the vase itself. The effect, therefore, of the contact 
 of the charged ball p with the interior surface of the vase, has 
 been to unite the two opposite electrical forces, existing upon the 
 ball and upon the interior surface of the vase ; the charged ball p 
 being positive, and the interior surface of the vase opposed to it, 
 negative ; and these two electricities must be co-existent, and 
 exactly equal in amount, since the ball p, after contact with the 
 interior surface, comes away perfectly neutral. 
 
 As this electrical change, however, produces no effect upon the 
 leaves of the electroscope, we may infer that the electrical charge 
 induced by the ball p, and the electrical charge existing in p, are 
 exactly equal in amount. 
 
 52. These phenomena do not appear to be affected by any number 
 of cylindrical surfaces interposed in series between the charged 
 ball and the surrounding walls of the vase. 
 
 Exp. 24. Let several concentric vases, r, s, t, u, Fig. 30, be 
 suspended, by fine insulating lines, one within the other, so as not 
 to touch in any point. The charged ball p, suspended within this 
 system of concentric cylindrical vases, produces the same effect as in 
 the case of the single vase, so that the intervention of many con- 
 ducting surfaces causes no difference in the amount of the inductive 
 effect. If we connect the two interior cylinders, t and u, by means 
 of a wire suspended from an insulated thread, and insert it between 
 them, the leaves still remain in a state of divergence, as before ; 
 and continue to do so, if the cylinders s and t be connected in a 
 
ELECTRICITY ON THE SURFACE. 
 
 43 
 
 similar manner, and so on. The result, therefore, in this case is 
 
 precisely the same as if the interval between the outside of u, 
 
 and the inner surface of the exterior cylinder r, were a solid 
 
 mass. Supposing, however, the interior cylinders, r, s, t, u, to 
 
 remain insulated from each other, and the 
 
 ball P to be in the centre of the system, 
 
 then we have a succession of positive 
 
 and negative surfaces (40) ; that is to say, 
 
 the inside of u y next P, is negative, and the 
 
 outside positive ; and so on for the successive 
 
 cylinders, until the induction reaches the 
 
 outside of r, where we find a charge by in- 
 duction of the same kind as that of the ball P, 
 
 and of exactly the same amount, as already 
 shown (51). . 
 
 If the charged ball p be now made to 
 
 touch the interior surface of the cylinder u, 
 
 then, as before (51), the opposite equal, 
 
 positive, and negative states neutralise each 
 other ; that is to say, the positive elec- 
 tricity of the ball unites with the negative 
 electricity induced upon the inner surface 
 of u, so that the charge originally on the ball P, and all the in- 
 ductions dependent on it throughout the system, vanish ; and the 
 ball, on being withdrawn from without the system, comes away 
 perfectly neutral, leaving the amount of its charge upon the 
 exterior surface of the cylinder r, as in the preceding experi- 
 ment. 
 
 It appears, therefore, by these experiments, that the divergence 
 of the electroscope leaves E remains the same, whether there be 
 only one cylinder, or whether there be a series of concentric 
 vessels, r, s, t, u, insulated one within the other ; or whether the 
 interval between the ball and the external cylinder be a solid mass, 
 so that a certain amount of electricity upon a ball, acting within 
 the centre of a vessel, exerts precisely the same power externally, 
 whether it operates by induction through the space between the 
 ball and the surrounding wall of the vessel, or whether it be trans- 
 ferred by communication through solid matter to that wall, so as 
 absolutely to annihilate or neutralise the induction within by the 
 union of the opposite electrical states, which we have shown to be 
 exactly of the same amount, and of opposite kinds. We may con- 
 clude, therefore, from these experimental inquiries, that, whether 
 
 Fig. 30. Concentric Vases and 
 Electrometer. 
 
44 ELECTRICITY ON THE SURFACE. 
 
 a conducting substance be a solid mass, or merely a hollow shell, 
 the amount of electricity which can be accumulated upon it is 
 solely dependent upon its external surface. 
 
 53. It appears by the foregoing results that neither induction, 
 nor attraction, nor repulsion, has any relation to solidity or quantity 
 of matter, or even to the kind of matter of which substances con- 
 sist. "Whether a body be metallic or not, hollow or solid, the 
 quantity of electricity it can finally receive and support, together 
 with the subsequent induction, attraction, or repulsion it exerts, is 
 in each case the same ; the only element which appears to enter 
 into the development of the force is time. Inferior conductors, 
 such as wood, require a somewhat longer time for accumulating or 
 abstracting electricity, or for operating inductively, than more 
 perfect conductors, such as the metals, the action of which is appa- 
 rently more instantaneous. The French philosopher Coulombe 
 found that an equal division of electricity took place between sub- 
 stances of similar form and equal surfaces, whatever difference 
 existed in the kind of matter of which they were composed ; or 
 whether hollow as a mere shell, or solid. 
 
 In fact, in communicating electricity to an insulated conductor, 
 as, for example, an insulated metallic sphere (45), we cannot 
 be said to electrify the metal itself bodily. The metallic sur- 
 face merely enables the electricity to expand upon the dielectric 
 medium surrounding it, and in such way as to cause the surface of 
 the metal to become enveloped by dielectric particles adhering so 
 closely to the metal as to admit of being considered an electrical 
 extension of the body itself. It is this envelope of dielectric 
 particles, supported by the metallic surface, which sustains the 
 induction, and through which the action reaches distant matter. 
 If it were practical to remove the conducting sphere without dis- 
 turbing the surrounding medium immediately in contact with it, 
 there would remain a hollow globular vase of dielectric particles. 
 
 It may be hence further inferred that accumulated electricity 
 exists on the surfaces of bodies, and has little or no relation to their 
 mass, so that as much electricity can be accumulated on a hollow 
 metallic sphere as on a solid sphere. In no case does an electrical 
 charge ever penetrate the substance of the charged body ; or, so 
 far as yet ascertained, does it even affect the particles beneath the 
 surface in any very sensible degree. What we require, therefore, 
 in accumulating electricity on insulated conductors, is a large 
 extent of inducteous surface, together with free inductive action 
 upon other matter ; a deduction abundantly verified in the fol- 
 lowing experiments : 
 
ELECTRICITY ON THE SURFACE. 45 
 
 54. Saussure, in the year 1766, observed, as an interesting 
 experimental fact, that an electrical charge tended to the surface 
 of the charged body. 
 
 Exp. 25. Charge an insulated conducting mass, R, Fig. 31, in 
 which there are deep pits, s, t, u. Introduce a small carrier ball, 
 p, p, into these pits, and withdraw it without touching their edges. 
 Then, as found both by Saussure and Coulombe, not a particle of 
 charge will have been taken up by the carrier. Bring the carrier 
 now in contact with the surface of the body ; it will immediately 
 receive an electrical charge. 
 
 55. The Hon. Henry Cavendish directed his attention to this 
 question, and, as we find by his manuscripts, contrived, so long 
 since as the year 1775, the following elegant and conclusive expe- 
 riment, by which he anticipated not only all the great facts which 
 were discovered in 1785 and the following years, as recorded in 
 
 Fig. 31. Saussure's Apparatus. Fig 32. Electrified Spheres. 
 
 the Memoirs of the Hoyal Academy of Sciences in Paris, 
 but also in other Memoirs of learned societies of a more recent 
 date. 
 
 Exp. 26. The intention of this experiment (says Mr. Cavendish) 
 is to find whether, when a hollow globe is electrified, a small 
 globe contained freely within it, and communicating with the 
 outer globe by a wire or other conducting substance, becomes 
 at all over-charged or under-charged (as he terms it) ; that is to 
 say, positive or negative. To this effect, a light, metallic, insulated 
 globe, g, Fig. 32,was enclosed between two insulated, hollow, metallic 
 hemispheres, M, N, leaving a space between the interior and ex- 
 terior shells, and constituting, when joined, an external globe of 
 about 13 inches in diameter. 
 
 The two hemispheres, M, N, were insulated in rectangular frames, 
 which admitted of being turned back so as to expose the interior 
 
46 ELECTRICITY ON THE SURFACE. 
 
 globe g, and leave it quite free of the two hemispherical enve- 
 lopes, an operation indicated in Fig. 32 by means of the insu- 
 lating supports, A c p and B d r, and the sliding bases A B, 
 Thus circumstanced, a temporary conducting communication was 
 established between the inner and the outer globes by a short 
 brass wire, w, attached to an insulating rod, x 9 and by which 
 the wire could be easily removed. Mr. Cavendish then commu- 
 nicated an electrical charge to the enveloping globe M N ; in 
 which case, as is evident, if any part of the charge tended to per- 
 vade the system as a mass, it could freely do so by means of 
 the communicating wire w. "I drew out the wire w," says 
 Cavendish, " connecting the inner with the outer globe, which, as it 
 was drawn away by an insulator, could not discharge the electricity 
 either of the interior globe or its exterior envelope. I then in- 
 stantly withdrew the hemispheres, and applied a pair of small pith 
 balls, suspended by fine linen thread, to the inner globe, to see 
 whether it was at all over or under-charged ; " that is, either positive 
 or negative. This electroscope he further explains as being fixed 
 at the extremity of an insulating glass rod, covered with a little 
 tinfoil in that part intended to touch the globe. " The result was 
 that, though the experiment was repeated several times, I could 
 never," says Cavendish, "perceive the pith balls to separate, or 
 show any signs of electricity." Every precaution seems to have 
 been taken in this experiment. On withdrawing the hemispherical 
 shells, their electricity became speedily discharged, so that no sub- 
 sequent electrical action could possibly arise. 
 
 Cavendish further endeavoured to discover how small a quantity 
 of electricity, not sensible to his electroscope in its ordinary state, 
 might be made apparent ; for which purpose he communicated to 
 the electroscope balls a weak positive or negative charge. In this 
 way he found he could render sensible on the inner globe a 
 quantity of redundant electricity less than the one- sixtieth part of 
 that on the outer ; and hence concludes that in this experiment 
 the redundant electricity, if any, existing on the interior globe, 
 must " certainly have been less than the one-sixtieth part of that 
 on the outer globe ;" but thinks " there is no reason to believe that 
 the inner globe was at all charged." 
 
 The double gold leaf electroscope (33) would be a still more 
 delicate test than the pith balls. 
 
 Although this simple and perfectly conclusive experiment fully 
 proves the tendency of electricity to the surfaces of bodies, yet the 
 following, which may be taken as the converse of Cavendish's 
 process, is by no means unimportant. 
 
EXPERIMENTAL ILLUSTRATIONS. 47 
 
 Exp. 27. Charge the interior globe g, Fig. 32, either positively 
 or negatively, the hemispherical envelopes M N being withdrawn, 
 as in the figure, and the communicating wire w removed ; 
 when so charged, replace the enveloping hemispheres, and insert 
 the communicating wire by means of its insulating handle, so as 
 to make a conducting communication between the inner and the 
 outer globes ; again withdraw the wire, and separate the hemi- 
 spheres, all the electricity will have left the inner and previously 
 charged globe, and be found on the exterior surface of the 
 hemispherical shells, charged, in fact, with the electricity of the 
 interior globe. 
 
 These experiments may be performed with a small sphere of 
 2 inches diameter, and light hemispherical shells of sheet 
 copper. 
 
 56. An experiment somewhat similar to that of Mr. Cavendish 
 has been described by M. Biot in his Traite de Physique, which, if 
 the student should fail to verify, he must not be disappointed. 
 In this experiment a conducting ellipsoid, M, Fig. 33, has 
 closely fitted to its surface two similar superficial ellipsoidal 
 
 a 
 
 Fig. 33. Ellipsoidal Covers. 
 
 envelopes, a b c, a d c, each furnished with a light insulating 
 holder, b n, do, and without leaving any sensible space between. 
 These shells being removed, the interior body M is charged 
 with electricity, after which the envelopes are replaced. On 
 being again withdrawn, all the charge, it is said, will have passed 
 into the superficial shells, and the interior body M will be found 
 perfectly neutral. Now it is quite clear that this experiment 
 cannot possibly succeed unless the envelopes be rapidly and 
 simultaneously withdrawn, so rapidly, indeed, as to exceed the 
 rapidity of the electrical expansion over any small space or open- 
 ing upon the surface of the interior body M, which will necessarily 
 occur between the envelopes at the instant of their separation, and 
 so simultaneously, that one shall not be in any sensible degree 
 after the other, for if it be, the charge will certainly expand from 
 the remaining envelope over the exposed part of the interior 
 ellipsoid M. This experiment, therefore, is much more precarious 
 
48 EXPERIMENTAL ILLUSTRATIONS. 
 
 and inconclusive, and certainly far inferior to that of Cavendish 
 
 just described (55). 
 
 57. Yolta likewise, in 1779, instituted the following conclusive 
 
 experiment, illustrative of the tendency of electricity to the 
 
 surface of a charged body. 
 
 Exp. 28. Let s, Fig. 34, be a thin hollow metallic sphere, about 
 
 5 inches in diameter, and having a circular opening at d y about 
 1| inch across. Place this sphere on a long 
 insulating support, v. Let a be a carrier ball 
 about three quarters of an inch in diameter, and 
 insulated on a slender glass rod. Charge the 
 carrier, and introduce it within the interior of 
 the shell s, but without touching the edge of the 
 opening at d. Having brought this electrified 
 ball in contact with the interior of the shell, again 
 carefully withdraw it clear of the opening d. 
 Every particle of the charge will have left the 
 carrier #, to appear on the outer surface of the 
 sphere s, which will have now become attractive 
 and repellent of a delicate electroscope. * By a 
 few repetitions of the experiment, the exterior 
 
 Fig. 34. Hollow Metallic _ p . 
 
 sphere. charge upon s may be rendered very powerful ; 
 
 for notwithstanding the previous communication of electricity to 
 the inner surface of s, the small ball a will be continually and 
 completely robbed of its charge. 
 
 If the hollow shell s be originally charged, and the insulated 
 ball a in a neutral state introduced within it, then, as in the expe- 
 riment by Saussure (54), it will not, on being removed, exhibit 
 any sign of electricity. 
 
 Exp. 29. Let v, Fig. 35, be a metallic can, sustained on an 
 insulator, i, and having attached to it an electroscope of diver- 
 gence, e. Let a c be a fine metallic chain, reposing at first in 
 the bottom of the can, having a silk cord, p, attached to it, 
 extending outside the can, so that the chain a c may be raised out 
 of the can without interfering with any electrical charge that may 
 be given it. 
 
 The chain reposing at the bottom of the can, charge the can 
 positively, until the electroscope e becomes divergent. Now raise 
 the chain by the insulating silk line p until it is freely sus- 
 pended in the air. The electroscope e will now decline, showing 
 that a portion of the charge had quitted the can to expand upon 
 the elongated chain, thus confirming, in a remarkable way, all the 
 preceding experiments. This is known as Franklin's " Can and 
 
ELECTRICITY ON THE OUTER SURFACE. 
 
 49 
 
 chain" experiment, and was contrived by him so long since as the 
 year 1754. 
 
 Exp. 30. A conical muslin bag, a c b, Fig. 36, rendered suffi- 
 ciently stiff to preserve its figure, but without materially affect- 
 
 Fig. 35. Can and Chain Experiment. 
 
 Fig. 36. Insulated Muslin Bag. 
 
 ing its pliancy, is fixed to a wire hoop, a b, and insulated on a 
 glass rod, m ; a silk line, /, is attached to the vertex c of the 
 conical bag, so as to pull it inside out, or again reverse it. If 
 this have an electrical charge communicated to it, and the interior 
 be then tested by a carrier in the usual way, no electrical indica- 
 tion is apparent, whilst the external surface a c b attracts and 
 repulses the electroscope freely. Reverse the surfaces of the bag 
 by means of the silk line attached to the vertex of the bag, so 
 that the interior may become the exterior surface ; then we have 
 precisely the same result ; the electricity is all found upon the 
 surface externally. If we again reverse the surfaces, we restore 
 the bag to its first condition, in which case the original electrical 
 development is again restored. This elegant experiment is due 
 to Faraday. 
 
 Exp. 31. The most general, and perhaps the finest experiment on 
 electrical induction, was carried out in the lecture-room of the Royal 
 Institution by the same philosopher, in November, 1837. Having 
 constructed a light cubical chamber, c, Fig. 37, of 12 feet cube, 
 the walls of which consisted of slight wire, covered with metallic 
 paper, and insulated by adequate suspending insulators, m, n, o, p, 
 he placed himself within it, used lighted candles, very delicate 
 electroscopes, and other tests of electricity, but although the cube 
 was placed in communication with a powerful electrical apparatus, 
 and charged so highly through an insulating conducting rod, A, 
 
50 
 
 IMPORTANCE OF INDUCTION. 
 
 as to throw off powerful sparks and brushes of light from the 
 exterior surface, yet not the least effect was produced upon the 
 electroscopes or other bodies within. The cube was now put into 
 communication with a perfect discharging train, and the air 
 
 Fig. 37. Cubical Electric Chamber. 
 
 within strongly electrified by means of a conducting wire passing 
 into it from the electrical apparatus ; but every attempt to charge 
 the air bodily and independently with either electricity failed. 
 The conclusion arrived at was, that non-conductors, as well as 
 conductors, have never yet had communicated to them an abso- 
 lute and independent charge of either electricity, and that to all 
 appearance such a state or condition of matter is impossible. 
 
 58. Induction appears to be the essential function both in the 
 first development of electrical force and all the subsequent pheno- 
 mena attendant on it. 
 
 Insulation and common conduction seem to consist in an action 
 of contiguous particles (47), and to depend upon the two opposite 
 forces developed in them. These forces are the particles in a state 
 of polarity (47), in which state they have a greater or less degree 
 of power in communicating their forces one to the other. Bodies 
 are better or worse conductors, or better or worse insulators, ac- 
 cording to the greater or less degree in which the developed forces 
 can re-unite (47). Induction and conduction appear, therefore, 
 
TENSION, CONDUCTION, DISCHARGE. 51 
 
 so far, to be in principle the same, though generally assumed to be 
 totally different (15). If the particles maintain their polarised state 
 (47), insulation is the result in a greater or less degree. If they 
 communicate their forces more or less freely to each other or to 
 other bodies, conduction ensues, and the phenomenon termed elec- 
 trical discharge necessarily follows. Retardation of electrical dis- 
 charge, therefore, is virtually insulation. If the induction, there- 
 fore, across a dielectric remains undiminished, then insulation is the 
 consequence, and the higher the degree of polarisation, that is to 
 say, the greater the amount of disunion of the opposite electrical 
 forces, the greater the polar state, which state or disunion has 
 been usually characterised by the term tension, or recombining 
 power. 
 
 If the contiguous particles do not maintain the polarised state, 
 but communicate their forces in a greater or less degree one to 
 the other, this state of tension falls, and, as just observed, is 
 followed by conduction and discharge. 
 
 The reduction, therefore, of two effects, insulation and conduc- 
 tion, so long held distinct, to an action of successive particles 
 obedient to one common law, is no unimportant result. 
 
 When the polarisation of the intermediate particles is raised to 
 a degree which they cannot support, the forces recombine with a 
 sort of convulsive effort, attended by an evolution of light and 
 heat, the result being the ordinary electrical spark. The dis- 
 tance through which discharge can take place between two 
 conductors (45) is called their striking distance. When we con- 
 tinue to dimmish the extent of surface originating an electrical 
 spark, and finally arrive at a small terminating point, very curious 
 effects are observable. 
 
 59. The influence of a pointed body in promoting the depolari- 
 sation and discharge of a polarised dielectric (47), was first 
 noticed by Franklin in 1750. He showed that when a pointed 
 conductor was presented to an electrically charged body, the 
 point rapidly caught up the electricity of the charged body, even at 
 considerable distances. He charged an iron ball, about 4 inches 
 in diameter, and observed that on presenting to it an uninsulated 
 pointed needle, the attractive force of the ball on a small thread 
 immediately ceased, as if its electricity had vanished. He further 
 observed that this influence of a pointed body was also exerted when 
 the pointed body was projecting from the charged body itself, the 
 charge being rapidly dissipated by the projecting point. At the 
 same time a current of aerial particles set off apparently from the 
 electrified point, so that if the latter were free to move, it would recede 
 
 E 2 
 
52 ACTION OF FOISTS. 
 
 in a direction opposite to that of the serial current, or electrical aura, 
 as it is called. On this principle the little toy known as the elec- 
 trical fly, acts. It consists of a wire with the ends pointed, and bent 
 in opposite directions, and a cap in the middle for supporting it 
 on a vertical wire attached to the electrified body. The aura, 
 or wind, which apparently escapes from the points of the fly, will 
 cause it to rotate in an opposite direction. So also the electrical 
 aura may be made to drive light models, representing water- 
 wheels, &c., for which purpose they are made to move on centres, 
 and are furnished with vanes, against which the stream from the 
 point plays. 
 
 60. It is not difficult to explain, upon the principles of induc- 
 tive action just adverted to, the more immediate effects of a 
 pointed body in dissipating the electricity of a charged body. 
 According to the theory of induction (41), it is assumed that the 
 particles of the dielectric are in a certain state of tension (58), 
 which rises higher and higher in each particle as the induction is 
 raised higher and higher, either by the closer approximation of 
 the inducing surfaces, increase of the charge, or variation of 
 form, &c. (41). 
 
 The spark- striking distance (58) will therefore be dependent on 
 the discharge of a few particles of the dielectric, occupying a very 
 limited space, in consequence of which the induced polarised 
 state of the whole series is lowered, and the molecules return to 
 their previous or normal condition, in the inverse order in which 
 they left it, whilst their powers to propagate or continue the 
 discharging operation from the point where the subversion first 
 occurred, become united. A good mechanical illustration of this 
 may be derived by setting up on end and near each other a series 
 of thin rectangular pieces of wood ; if we overturn one at the 
 end, the next, the next, and so on, must fall, because each in suc- 
 cession becomes pressed upon by the united forces of the preceding, 
 which now complete the downfall of the whole series in a given 
 direction. The few particles originating the discharge are gene- 
 rally next one of the terminating conductors ; in this point of 
 subversion, however, they are not merely pushed aside, but they 
 assume for the time an extreme tension, and the powers dis- 
 charge throughout the series with violence and explosive force : 
 the ultimate effect is the same as if we had put a discharging 
 wire in place of the dielectric particles, and operated by conduc- 
 tion immediately between the limiting conducting surfaces. 
 The tension of the particles of the dielectric next the points in 
 the limiting conductors being greater than those in the middle 
 
CONDITIONS OF DISCHARGE. 53 
 
 of the series, it is hence in these points that the discharge com- 
 mences ; so that when these conductors terminate in mere points 
 or small surfaces, the tension upon the particles of the dielectric 
 in contact with them is excessively increased ; in fact, all the lines 
 of inductive force may be supposed to concentrate upon the point. 
 If a spherical uninsulated conductor be placed some distance from 
 a point projecting from a charged conductor, then the lines of 
 inductive force will concentrate, as it were, upon the point, and 
 the latter will originate an active mechanical force, and pre- 
 serves its predominance over the other portions of the con- 
 ductor, behind it, by a continued discharge of the accumulated 
 electricity. Hence currents of wind arise by the recession of the 
 charged particles of air, and which are, in every way, favoured 
 by the shape and position of the conducting surface in the rear. 
 If the point be more or less central to the walls of a room, with- 
 out any more immediately opposed conductor, or be exposed to the 
 induction of any other substance in its vicinity, still the same 
 result ensues, since there is no distance so great as to limit the 
 operation of this inductive action (45). 
 
 The theory applies, by the converse of this, to an uninsulated 
 point, opposed to a charged body. 
 
 The conditions, therefore, requisite for the production of an elec- 
 tric spark are, two opposed conductors (45), charged, one positively 
 and the other negatively, with an intervening dielectric medium. 
 If the opposite forces be continually increased in strength by 
 exalting the electric state of the two conductors, or by bringing 
 the conductors nearer together, or by diminishing the density of 
 the intervening dielectric, a spark at length appears, discharge 
 ensues, and the two forces combine, and are reduced to a state of 
 neutrality. The opposed conductors may be considered as the 
 termini of the inductive action, all the effects prior to discharge 
 are inductive, and the degree of tension required before the spark 
 passes, is the limit of the influence of the dielectric in resisting 
 discharge, that is to say, of its conservative power, and, there- 
 fore, a measure of the forces in operation. 
 
 61. The phenomena of electrical induction, which we have been 
 hitherto considering, are calculated to throw much further light 
 on the theory and nature of electrical force. According to the 
 hitherto received doctrines, electricity has been referred to an 
 invisible, imponderable, highly elastic fluid, intimately associated 
 with common matter, and pervading the solar system. The dis- 
 covery of Du Fay (22) led to the idea that this assumed fluid 
 consisted of two primary elements, possessing distinct and opposite 
 
54 NATURE OF ELECTRICAL ACTION. 
 
 properties ; these elements, as already observed, have been termed 
 vitreous and resinous, or positive and negative electricities ; they 
 are further assumed to be highly repulsive of themselves, but 
 attractive of each other. When combined they are neutralised, 
 and electrical repose is the result. When disunited, each becomes 
 active, and is the source of electrical phenomena. This has been 
 termed the double fluid or French theory. 
 
 According to Franklin, Watson, Cavendish, and other English 
 philosophers, electrical phenomena depend on the presence of a 
 single agency in a greater or less degree, repulsive of its own 
 particles, but attractive of the particles of common matter. When 
 distributed in bodies in quantities proportionate to their capacity 
 for it, such bodies are said to be in their natural state ; the result 
 is electrical quiescence. When we increase or diminish the quantity 
 in any substance, we disturb this equilibrium, and the bodies, 
 according to Cavendish (55), are said to be under or over- 
 charged, and to be either positively or negatively electrified, in 
 which case a powerful action ensues, arising out of the tendency 
 of the bodies to regain their natural state. This it does by taking 
 away electricity from other bodies if its own quantity be dimi- 
 nished, or by throwing off electricity upon other bodies if the 
 quantity be increased. 
 
 Faraday's researches, however, throw a further light on the laws 
 and nature of electrical action. Without entering upon any assump- 
 tion in explanation of the occult nature of electricity, Faraday 
 regards it simply as force ; much in the same way as Newton con- 
 sidered gravity, without pretending to explain its nature. Accord- 
 ing to this view, then, all electrical phenomena are resolvable into 
 induction (58) . In every case of electrical action two opposite forces 
 are invariably present ; one cannot exist without the other ; we can- 
 not charge a body with one of the forces only, independently of the 
 other, such a condition of matter being impossible (50). It is 
 assumed that the particles of every kind of matter are as wholes 
 conductors ; that not being polarised (47) in their original state, 
 they may become so by the influence of neighbouring charged 
 particles ; that the particles, when polarised (47), are in a tense or 
 forced state, and are constantly endeavouring to return to their 
 normal condition ; that the ready communication of the opposite 
 forces between near particles constitutes conduction, and the diffi- 
 cult communication, insulation; that induction can only take place 
 through or across insulators. The leading feature of the electric 
 power is, that it is limited and exclusive, and that the two forces 
 always present are exactly equal in amount (51), the forces being 
 
INTENSITY AND TENSION. 55 
 
 related to each other in one of two ways, either as in the ordinary 
 state of an insulated uncharged conductor, or in the charged state 
 of an insulated conductor, which last state is a state of induction. 
 
 62. There are two terms, namely, intensity and tension, fre- 
 quently employed in treating of electrical force, and more or 
 less involved in the phenomena of electrical induction, that may 
 require a little further explanation, especially as they are fre- 
 quently employed in a loose and indefinite manner. In order 
 fully to comprehend their real signification and immediate appli- 
 cation, it should first be observed that the term quantity a 
 term sufficiently intelligible designates the actual amount of 
 the unknown agency, whatever it be, constituting electrical 
 force, as referable to some arbitrary standard of measure, in 
 terms of which it may be expressed. The term intensity , in its 
 usual acceptation, has been referred to hypothetical views of 
 the occult nature of electricity itself, often vague and unsatis- 
 factory. Intensity has also been used to signify the peculiar state 
 of an hypothetical elastic fluid, the electrical agency being sup- 
 posed capable of changing its condition, just as we may imagine 
 a spring to have greater or less elastic power. It will be found, 
 however, upon a critical investigation of the facts upon which 
 this assumption rests, that the term intensity applies more especi- 
 ally to electroscope or electrometer indications, and is but another 
 term for the quantity of electricity operating at a given point in 
 a given direction ; as, for example, in the direction of the electro- 
 scope. 
 
 For example, let A, c, s, Fig. 38, be three insulated conducting 
 
 Fig. 38. Mutual action of Insulated Conductors. 
 
 bodies of different forms and magnitude. Let A be a circular 
 plate of small thickness, c a cylinder, and s a sphere, and suppose 
 each furnished with a delicate electroscope of divergence, a, c, s 
 (the three electroscopes being precisely alike). Suppose also these 
 three conductors to be charged with electricity to such an extent 
 as will bring each of the electroscopes to the same angle ; in this 
 case, as is well known, the quantity upon each of these bodies 
 will be very different, notwithstanding the sameness of the elec- 
 troscope indication. Here the electrical intensity of each of the 
 
56 EFFECTS OF EXTENT OF SURFACE. 
 
 bodies is said to be the same, and the electrical charge to have 
 the same density in each. Again, take two insulated conducting 
 spheres, M, N, Fig. 39, of unequal diameter, and suppose the 
 surface of the one, M, to be three times that of the other, N, 
 having delicate electroscopes, m n, attached 
 to them as in the former case ; then, if we 
 charge each of the globes with the same 
 quantity of electricity, the electroscope on 
 the small sphere N will have a much greater 
 angular divergence than the electroscope 
 upon the large sphere M, from which it has 
 been inferred that the electricity upon the 
 small sphere has greater elastic power 
 
 Fig. 39. Insulated Con- * 
 
 ducting spheres. than the electricity on the large sphere ; 
 
 or, in other words, that we have varied the intensity with- 
 out changing the quantity. A further investigation of these 
 phenomena, however, leads to a more definite and simple 
 explanation. When we place different quantities of electricity 
 upon the three bodies A, c, s, Fig. 38, without affecting the 
 angular divergence of the electroscopes, it is the total quantity 
 which varies, and not the quantity in any given point of the 
 surface ; that is to say, the absolute quantity affecting the electro- 
 scope ; the quantity affecting the electroscope is really the same 
 in each case, although the total quantity upon each of the three 
 bodies may greatly differ. It is, in fact, quite evident that since 
 the three bodies vary in extent of surface, the greater that extent 
 the greater the number of points the surface will contain ; the 
 quantity, therefore, in any one point of either of the bodies may 
 still be the same, notwithstanding the quantity in each of the 
 bodies is different. 
 
 The total quantity or charge must be taken to act upon an 
 electroscope projecting from its surface, as it would do supposing 
 the electricity were equally distributed ; this is evident from the 
 fact that, in whatever point of the charged surface the electroscope 
 be placed, its angular divergence remains the same. Intensity, 
 therefore, in this case, when rigidly interpreted, is nothing more 
 or less than the quantity of electricity at a given point acting on 
 the electroscope, and cannot be taken to express any dif- 
 ference in the actual state or condition of the electrical agency 
 itself. 
 
 Let us now take the case of a different angular divergence, the 
 quantity of electricity being the same. Take for example the 
 case of the two unequal spheres M N, Fig. 39, each furnished 
 
WHAT IS ELECTRICITY? 57 
 
 with, a delicate electroscope, and let tlie same quantity be accu- 
 mulated on each. The electroscope of the larger sphere will, 
 in this case, as already observed, have a much less angular 
 divergence than the electroscope of the smaller sphere. Here we 
 perceive that the quantity in any one point of the two spheres is 
 not the same, since the total quantity being alike in each sphere, 
 and the surface over which it is expanded in the larger sphere 
 three times as great as the surface over which it is expanded in 
 the smaller sphere, there would be a greater number of points 
 in the larger sphere for the reception of the charge, taking the 
 surfaces, as supposed, 3 to 1. Hence where there is one particle 
 of force on a point of the larger sphere, there will be three 
 particles of force on a point of the smaller sphere. Here again, 
 intensity, when correctly interpreted, is nothing more than 
 the quantity of force operating at a given point ; but it does not 
 follow that the agency in operation has a higher amount of elastic 
 power in the one case than in the other, or is necessarily more or 
 less dense. If it exhibit a greater degree of energy in any given 
 point of the small sphere than it exhibits in any given point of 
 the large sphere, it may be because there is a greater amount of 
 mere force in operation in that point, Until we nave a clearer 
 conception of the nature of electrical force, we cannot say whether 
 that force is susceptible of change in quality or constitution or not. 
 Any inference therefore of a change in electrical density must be 
 mere assumption. The idea of a difference of density in the two 
 cases of the small sphere and the large sphere, supposing the elec- 
 trical agency to be material, and subject to the laws of ordinary 
 matter, is certainly in accordance with the deduction that where 
 there is one particle of electricity acting in a given point of the 
 large sphere, there are three particles in a given point of the 
 small sphere. So far this corresponds with the hypothesis of 
 density, or as better expressed, perhaps, by the French philo- 
 sophers, with "thickness of stratum," but as already observed, 
 since we have no knowledge whatever of the occult nature of the 
 electrical agency, we can scarcely venture to rely upon any hypo- 
 thesis of this kind, but must be content to consider electricity as 
 mere force, without assigning to it any specific elementary con- 
 dition, much in the same way as we accept gravity as a mere 
 force, without troubling ourselves as to its occult nature. 
 
 It follows from this that there is no such element as intensity 
 independent of quantity ; and that what we are to understand 
 by intensity, is only the greater or less amount of force in a given 
 point operating on the electroscope. 
 
58 SPECIFIC INDUCTION. 
 
 63. We have already remarked (49), that supposing a charged 
 body to be alone in free space, there is a propagation of force 
 in all directions. Directly we place another body near the 
 charged body, then the inductive action becomes shared, more 
 especially with the near body. For example, let the cylinder c, 
 Fig. 38, be charged with a given quantity, so as to give to its 
 electroscope c a given angular divergence. Let now an insulated 
 neutral body, s, furnished also with an attached electroscope, s, be 
 brought near the charged cylinder c ; the angular divergence of 
 its electroscope c becomes sensibly less, whilst the electroscope s of 
 the insulated neutral body s begins to diverge, evidently showing 
 a propagation of force in that direction. The electrical intensity 
 of the cylinder is in this case said to be diminished by the 
 proximity of the neutral body, a phenomenon already noticed 
 (49). It is easy to perceive in this case that the angular diver- 
 gence of the electroscope of the cylinder is not diminished in 
 consequence of any change in the assumed elasticity of the 
 electricity with which it is charged ; but in consequence of the 
 force which at first operated exclusively on the electroscope of 
 the cylinder being shared with another body, and therefore ope- 
 rating in two directions instead of one ; that is to say, in the 
 direction of the neutral body, as well as in the direction of the 
 electroscope. It was on this principle that Yolta succeeded in 
 rendering sensible minute quantities of electricity not appreciable 
 by the electrometer. The term tension is more especially applicable 
 to the polarised state of electric particles, or to the state of the 
 particles of any dielectric medium intermediate between a positive 
 and negative surface, as already explained (58). 
 
 Tension may therefore be taken to represent the recombining 
 power of the disunited electricities in the polarised particles of a 
 dielectric medium, interposed between a positive and a negative 
 surface, whilst intensity is more especially referable to electrometer 
 indication. 
 
 We may further observe that whilst intensity, as measured by 
 direct attractive force between a charged and a neutral body, is 
 as the square of the quantity accumulated, tension, as measured 
 by the striking distance (58) of the electrical spark, is as the 
 quantity itself, phenomena perfectly consistent with each other, 
 and the laws of electrical force. 
 
 64. Specific' Induction. Cavendish (as appears by his valuable 
 MSS. of the year 1771, confided by the Earl of Burlington to the 
 care of the author and the late learned Dean of Ely) observed a 
 great difference in inductive action when transmitted through 
 
SPECIFIC INDUCTION. 
 
 59 
 
 different kinds of glass and some other insulators. Thus, coated 
 crown glass, for example, differed in its charging power from 
 coated window- glass. Coated resin or bees- wax differed from 
 coated brimstone or shell-lac. Faraday, in the course of his valu- 
 able researches in Experimental Electricity, printed in the Philo- 
 sophical Transactions for 1837, was led to treat this subject in a 
 novel and instructive way, under the title of Specific Induction. 
 
 The question whether different dielectrics have any degree of 
 influence in promoting inductive action through them, is a question 
 of singular interest and importance. The ques- 
 tion may be thus stated. Suppose M, Fig. 40, 
 to be an electrified plate of metal, insulated in 
 free air, and x y two exactly similar but unin- 
 sulated plates at equal distances, parallel to and 
 on each side of it. The charged plate M will, as 
 is evident, operate equally by induction on x and 
 y (46). If some other dialectric than air, such 
 as shell-lac, be introduced between the plates, 
 the question is, will the induction remain the 
 same ? Will it depend solely on the distance 
 between the plates, without relation to any in- 
 tervening matter (as was once imagined by 
 Coulombe and other philosophers to be the 
 case), or will it vary ? Now, it is to be here observed that 
 what is termed electrical charge, is, in fact, a polarisation of 
 dielectric particles between opposed conductors. If, therefore, 
 induction be related to the particles of the surrounding dielec- 
 tric, then it is related to all the particles under the influence 
 of the inducing conductors, and not merely to the few particles 
 next the charged body. The nature and properties, therefore, of 
 the dielectric intervening between opposed conductors would be 
 very likely to be attended by what we have termed Specific Induc- 
 tion ; that is to say, the induction, for example, across air might 
 possibly differ from induction across shell-lac ; and such is really 
 found to be the case. Faraday compared together the inductive 
 capacities of different dielectric media, and found shell-lac to evince 
 greater facility in propagating inductive action through its sub- 
 stance than air. Assuming the specific induction of air to be 
 1, shell-lac would be 1-5. The conclusion that shell-lac exhibits a 
 higher specific inductive capacity appeared irresistible. Glass was 
 found to have little or no specific induction, although different 
 kinds of glass, according to Cavendish, exhibited different charging 
 powers. Sulphur, as compared with air, is found to be as 1 : 2 '24, 
 
 Fig. 40. Specific In- 
 duction Apparatus. 
 
60 SPECIFIC INDUCTION. 
 
 being greater than any substance yet tried. An extensive experi- 
 mental inquiry, with a great variety of substances including gases, 
 various fluids, air, rare and dense may be adduced to show that 
 most bodies have a specific inductive capacity of greater or less 
 extent ; but in the case of gases and vapours, the specific inductive 
 capacity seems to be pretty much alike in all. A complete experi- 
 mental investigation, however, of this interesting question is 
 difiicult, and occasionally precarious. Glass and many other insu- 
 lators are found, on examination, totally unfit for the purpose of 
 accurate experimental research. Thus, glass, for example, princi- 
 pally in consequence of the alkali contained in it, though made 
 warm and dry, always has a certain degree of surface conducting 
 power, mainly depending on the state of the atmosphere. This 
 renders it unfit for a delicate test experiment. It is evident that a 
 very feeble degree of surface conduction may tend to produce effects 
 indicating a greater inductive capability than that of a more per- 
 fect insulator. Shell-lac and sulphur, of all other substances, appear 
 to be the most free from this objection. 
 
 Some experimental inquiries by the author, relating to Specific 
 Induction, will be found in the Philosophical Transactions of the 
 Royal Society for 1842, Part ii. 
 
 The given substance to be examined being disposed under the 
 form of a circular plate of 1 foot in diameter, and about four- tenths 
 of an inch in thickness, tinfoil coatings, 6 inches in diameter, were 
 applied to its opposite surfaces, after the manner of coated glass. 
 One of the coatings being connected with the attractive disc of an 
 electrometer, it was easy to determine, by a simple and direct ex- 
 periment, the three following elements requisite for the elucidation 
 of the question of Specific Induction : First. By insulating the 
 system and depositing a given measured quantity of electricity on 
 the coating operating on the electrometer, we determine the intensity 
 of the given quantity (63). Secondly, By connecting the opposite 
 coating with the electrometer, we determine the direct induction 
 between the coatings through the intervening dielectric. Thirdly. 
 By connecting the one coating with the earth and charging the 
 opposed coating, we determine, by means of the electrometer, the 
 sensible portion of the charge. In this way we may examine any 
 dielectric medium, whether solid, fluid, or gaseous, contained 
 between the metallic coatings, and compare their respective 
 influences over the degree of induction which takes place through 
 them. 
 
 The following are some results of the author's inquiries. Call- 
 ing the specific inductive capacity of air 1, we have for resin, 1*77 ; 
 
SPECIFIC INDUCTION. 61 
 
 for pitch, 1-8; for lees-wax, 1-86; for glass, 1-9; for brimstone, 
 1-93; for lac, 1'95. 
 
 The results in the case of lac and air very nearly coincide with 
 those of Faraday, who found (Experimental Eesearches, 1270) the 
 relation of lac to air as 2 : 1, or very nearly. He also found a 
 very high inductive capacity for sulphur, although the specimen 
 employed in the author's experiments did not give a higher 
 capacity than lac, as appears to be the case in the experiments of 
 Faraday. The author has given in this paper many interesting 
 practical observations relative to experiments on specific induction. 
 
 65. The general question of specific inductive capacity may be 
 reduced to an experimental form through the medium of an instru- 
 ment (Fig. 41), which may 
 be termed a Specific Induc- 
 tion Electrometer ; or by Fa- 
 raday, a Differential Inducto- 
 meter. m, n are two electro- 
 scope gold-leaves suspended 
 from two small rounded gilt 
 conductors within a glass 
 receiver, K, as in the gold- 
 leaf electroscope. The leaves 
 are from 3 to 4 inches in 
 length, and about four- tenths 
 of an inch in width. The 
 gilt conductors from which 
 
 Fig. 41. Specific Induction Electrometer. 
 
 the leaves are suspended 
 
 are fixed at the extremities of small stout brass wires, passing 
 through fine corks inserted in holes drilled through the sides 
 of the glass receiver K, by which the distance of the projection 
 of the leaves within the glass may be varied, a d is a small ver- 
 tical metallic rod passing through an insulating cork in the mouth 
 of the receiver, and terminating in a thin gilt disc, d, about half 
 an inch in diameter, immediately between the leaves ; so that on 
 slightly charging the disc through its vertical metallic rod, the 
 leaves may be acted upon by the disc on either side. 
 
 Exp. 32. Connect one of the plates A with one of the gold- 
 leaves m, and the opposite plate B with the other leaf n, 
 and adjust the distance of the plates A and B at 1-J- inch 
 from the middle plate c. Let the gold-leaves m and n be 
 2 inches apart. Slightly charge the centre plate c positively, 
 the plates A and B, with their gold-leaves, being at the same time 
 connected with the ground, that is to say, uninsulated; which 
 
62 SPECIFIC INDUCTION. 
 
 being done, immediately re-insulate them. In this state of things 
 c remains positively charged, and A and B negatively, by induction 
 (39), the same dielectric air remaining in the interval between 
 the plates c A and c B, the gold-leaves m and n hanging parallel 
 to each other in a relatively unelectrified state. Thus circum- 
 stanced, let a plate of shell-lac, three-quarters of an inch thick 
 and 4 inches square, perfectly insulated, be introduced between 
 the plates c and A: the electric relation of the three plates 
 becomes immediately changed ; the gold-leaves m, n y now attract. 
 Remove the shell-lac-plate from between the plates c and A : the 
 attraction ceases. Introduce now the shell-lac between c and B : 
 attraction again takes place, but again ceases on removing it ; 
 evidently showing that induction through air is different from 
 induction through lac. Similar results are obtained with a plate 
 of sulphur. By means of this instrument (Fig. 41) Faraday was 
 enabled to show the difference of the specific inductive capacity 
 between a space partly filled with air and partly occupied by 
 thin shell-lac-plate. 
 
 Thus circumstanced, the gold-leaves m y n, become as delicate a 
 test of specific induction as they would be of ordinary electrical 
 charge in Bennett's Electrometer, 
 
63 
 
 OCCASIONAL MEMOEANDA AND EXPLANATORY NOTES. 
 
 (L) The opposite electrical states of tlie terminating surfaces of the bodies 
 p, Q, Fig. 22, are best examined by the gold leaf electroscopes adverted to 
 (32, 33). The body P being charged positively, apply a well-insulated carrier 
 plate (14), about three-quarters of an inch in diameter and one-tenth of 
 an inch thick, to each of the faces, p' n, p q ; and after each contact, transfer 
 the carrier to the double gold leaf electroscope (33), or other of the elec- 
 troscopes already described (33), weakly charged with positive or negative 
 electricity, suppose positive ; the divergence of the leaves or balls will tend 
 to increase if the carrier plate has been applied to the faces p or p' of the 
 bodies P, Q, and diminish if it has been applied to the face n of the body Q. 
 If applied to the face q of the body P, we have a comparative difference in the 
 actual states of p, q, q will be less positive than p. 
 
 (M) The mechanical arrangement referred to, (39) is well adapted to expe- 
 rimental purposes in electricity. 
 
 The bodies p Q, e /, are insulated and supported on small wooden plat- 
 forms, T T T T, about 5 inches square, movable in an open rectangular frame 
 of wood, A B, about 4 feet in length by 8 inches in width. This rectangular 
 frame is constructed of mahogany battens 2 inches square, stiffened by 
 cross pieces, in order to preserve the sides straight and parallel. The plat- 
 forms T T T T are accurately fitted within the sides of the frame, so as to 
 slide freely between guide pieces throughout its whole length. The bodies 
 p Q, e f, may hence be conveniently placed at measured distances from each 
 other. P Q are light hollow cylinders of gilt wood, each of them having an 
 adjacent electroscope electrometer, e /, to indicate any electrical change 
 which they undergo. 
 
 (N) The rod m N, Fig. 22, carrying the disc N, is about 14 inches long, 
 mounted on a light axis, and counterpoised by a small balance ball, u. The 
 axis is set on fine centres, counterpoised by a small balance ball, u. This 
 disc is set within an elliptical brass ring, x, 6 inches long and 1 inch wide, 
 so as to admit of a very free motion of the rod in a horizontal plane, the rod 
 being counterpoised by a small ball. The whole is supported by means of a 
 short projecting pin, on a small wood ball, a, insulated on a glass rod, h &. 
 The radial arm I d, sustaining the insulated disc P, is attached to a circular 
 base, &, movable about the pivot foot b of the insulating rod a &, supporting 
 the elliptical ring x and movable arm m N. We can thus make the two 
 discs, P N", approach or recede from each other. The whole is supported on a 
 convenient base, E s. The electroscope leaves p 1 n' are applied to the discs P N 
 by means of a small piece of cork, which causes the leaves to hang freely, 
 and at short distances from the surfaces of the discs. 
 
CHAPTER III. 
 
 ELECTRICAL APPARATUS. INSTRUMENTS OF ACCUMULATION. 
 
 66. The Electrical Machine. It having been found essential to 
 the progress of electrical inquiry to produce and accumulate the 
 two opposite electricities (22) in great quantity and in a more easy 
 and expeditious way than could be effected by means of the excited 
 tubes already referred to (20), or other simple methods, electricians 
 were led to a mechanical arrangement, more especially termed an 
 Electrical Machine. The elements of any arrangement of this kind 
 are these 1. An electric to be excited ; 2. A rubber for excita- 
 tion ; 3. An insulated conductor, or conductors, for collecting 
 either the electricity of the rubber, or the electricity of the excited 
 electric (23). The glass and gutta-percha tubes already referred to 
 (note F) may be considered as electrical machines of the most simple 
 form. We have, in the glass or gutta-percha tubes, the electric to 
 be excited ; the silk or other substance held in the hand, the 
 cushion or rubber ; the projecting rod and ball, the insulated con- 
 ductor for collecting the electricity of the tube. We are not 
 enabled to collect the electricity of the rubber, because the rubber, 
 being held in the hand, is in this case not insulated, and, hence, 
 its electricity becomes dissipated. 
 
 67. For the purpose of subjecting electrics to powerful friction, 
 it occurred to the old electricians to mount them upon an axis in a 
 wooden frame, and turn them round, by means of a winch, against 
 a fixed cushion, or even against the hand. Otto Guericke, so long 
 since as the commencement of the seventeenth century, was the 
 first to resort to this method of excitation. He mounted a globe 
 of brimstone upon an axis, and turning it round against his hand, 
 evolved negative electricity in abundance (30). Since his time, 
 other electrics have been employed, including vitreous and resinous 
 substances generally. Glass, however, has been commonly used in 
 the form of hollow globes, cylinders, or plates. 
 
 Electrical machines, as at present constructed, are principally of 
 two kinds, termed the cylindrical and the plate machines ; the 
 electric to be excited being either a hollow cylinder of glass, or a 
 glass plate. 
 
THE CYLINDRICAL MACHINE. 
 
 65 
 
 Fig. 42. Harris's Cylindrical Machine. 
 
 68. Cylindrical Electrical Machines. A machine of this kind of 
 improved construction is shown in Fig. 42, in which A m B is a 
 hollow glass cylinder, the diameter being about two-thirds the length 
 of its side. This cylinder is blown with a short open projecting neck, 
 A B, at each extremity 
 of the axis ; these open 
 necks are closed air-tight 
 with fine cork, the air 
 within the cylinder being 
 previously made per- 
 fectly dry, and the glass 
 within freed from all 
 deposit of moisture. The 
 necks are covered with 
 neat brass caps, or caps 
 of varnished mahogany 
 cemented over them. 
 The caps have short 
 axial projections, A B, of hard wood, on which the whole is 
 mounted on two mahogany or glass pillars, c D, so as to admit 
 of the cylinder being turned round by a winch, iv, or a wheel 
 and band, against a fixed rubber, R R'. The rubbing cushion is 
 freely sustained with pressure against the glass cylinder by an 
 insulated cylindrical conductor, N, termed the negative conductor, 
 insulated on the glass pillar N n. This insulating pillar is based 
 on a sliding piece, t, having a stop-screw to fix the slide piece at 
 any required point. The rubber has a flap of oiled silk, m, 
 attached to it, extending from its upper edge over the cylinder to 
 about one -third of its circumference. The silk is oiled on one 
 side only, the rough side being next the glass. The cushion R R', 
 with its flap m, is attached to a wooden back, and is held in place 
 by means of holes drilled in the conductor and two projecting 
 pins, as shown at F, which represents the detached cushion and its 
 flap. Two small spiral springs, s s', project from the back of the 
 rubber. These springs, under compression, set nearly fair with 
 the surface of the conductor N, and hold the cushion with variable 
 gentle pressure against the glass cylinder. The application of a 
 flap of silk to the electrical machine is very important. "We owe 
 this improvement to Dr. Nooth,* who in this way sought to cut 
 off the advancing portion of the excited glass from all electrical 
 communication with the cushion it had just left, and thus prevent 
 any retrograde movement of the developed electricity. 
 * Phil. Trans. 1774. Vol. Ixiii., pt. 2. 
 
 F 
 
66 THE PKIME CONDUCTOR. 
 
 69. The rubber n' consists of several layers of thick woollen 
 stuff, or some other elastic substance, supported, as just observed, 
 by a wooden back, from which the holding pins project, and is 
 faced up by fine morocco leather. The silk flap consists of very 
 thin silk, termed Persian, oiled on one side, the opposite or rough 
 side being next the glass (68). The construction of the cushion 
 and flap, although of a very simple form, is by no means a matter 
 of indifference. The best material for the stuffing of the cushion 
 is a modern manufacture termed spongeopiline, a combination of 
 sponge and wool ; it is thick, pliable, even, and elastic.* 
 
 70. The positive electricity generated by the movement of the 
 cylinder against the rubber and under the silk flap is received 
 upon a row of metallic points, q q (59), projecting from the 
 insulated or positive conductor p, termed the prime conductor, 
 immediately opposite the termination of the silk flap m. This 
 conductor is called the positive conductor. It is insulated on the 
 glass pillar p p, which is also set upon a sliding-piece, u, similar 
 to that of the insulator of the negative conductor N, and is steadied 
 in position by a stop-screw. The two conductors p N are cylin- 
 drical, and of a T shape. Each conductor consists of two elongated 
 portions, P and q q, well rounded at their extremities ; one portion, 
 q q', of the T being parallel to the side of the cylinder, its length 
 being three times its diameter ; the other, a shorter portion, p, is 
 set at right angles from its centre, and has a rounded extremity, 
 from which projects a brass ball of an inch in diameter set in a 
 sliding brass tube. This ball carries a sliding rod, s s', terminating 
 in light metallic balls. The sliding rods s s', with their terminating 
 balls, can be turned aside to any convenient angle. As it may be 
 occasionally desirable to effect a large electrical accumulation, a 
 cylindrical conductor with full rounded extremities of about twice 
 the length of the glass cylinder and half its diameter is employed. 
 This, when required, is united to the projecting central portion of 
 the positive or negative conductors, according to the kind of 
 electricity to be accumulated. In this way we obtain a powerful 
 and dense electric spark ; the brass tubes and balls projecting 
 from the centres of the conductors being in this case removed to 
 admit of the junction of the new conductor. The electricians of 
 the last century were in the habit of employing prime conductors 
 of very large dimensions. 
 
 When the cylinder is made to revolve freely, under friction of 
 
 * For the most approved method of preparing the cushion and flap, see 
 the Chapter on Electrical Manipulation, Nos. 83 and 84. 
 
LARGE CYLINDER MACHINE. 
 
 67 
 
 the cushion and silk flap, the conductor N being connected with 
 the ground, a surprising stream of positive electricity is obtained 
 from the insulated conductor p, in the form of a current of 
 bright sparks or scintillations. If the conductor p be connected 
 with the ground and the cushion insulated, we have a copious 
 evolution of negative electricity from the conductor N ; thus we 
 see that both electricities are produceable by the same machine 
 (24). If we desire to obtain positive electricity we connect the 
 negative conductor N with the ground, and take sparks from the 
 positive conductor p. If we wish to obtain negative electricity 
 we connect the positive conductor p with the ground, and take 
 sparks from the negative conductor N. [Note 0.] 
 
 71. The most powerful and complete cylindrical machine of past 
 
 Fig. 43. Nairne's Electrical Machine. 
 
 days appears to have been constructed about the year 1773 by Mr. 
 Edward JSairne, F.R.S., of London, mathematical instrument maker. 
 This instrument having been celebrated in the pages of electrical 
 history it may be worth while to describe it here. It is represented 
 in Fig. 43. The glass cylinder G is 1 foot in diameter and 19 inches 
 long between the shoulders. The rubber is 14 inches long and 
 5 inches wide, is insulated on two horizontal glass rods, A B, and 
 is acted on by two wood springs, c D, so as to press the rubber 
 against the glass cylinder. The prime conductor p is of large 
 size, it being 5 feet in length and 12 inches in diameter. It termi- 
 nates in a short brass rod and ball, and is insulated on two solid 
 glass pillars, m n. A receiving ball, K, is opposed to the ball at the 
 end of the conductor, and is fixed on the extremity of a brass tube, 
 
 F 2 
 
68 THE PLATE MACHINE. 
 
 movable in a hole in the insulating stand s, which supports it, 
 and by which it may be caused to communicate with the ground. 
 The cylinder is set in motion by a multiplying wheel and band, w. 
 With this machine electrical sparks of great power, from 12 to 
 14 inches in length, were obtained, Nairne employed two prime 
 conductors ; a large one, P, for dense and a smaller one for lesser 
 accumulations. 
 
 72. It is important in the construction of these instruments that 
 the interior of the glass cylinder be perfectly free from moisture, 
 otherwise little or no result is obtained. Hence it is that many 
 well-formed cylinders fail in exciting power until made warm by 
 means of heated irons. In constructing the instrument the 
 cylinder should not be closed air-tight until the atmosphere be in 
 a perfectly dry state and the barometer at 30 inches. The older 
 electricians were in the habit of coating the interior of their 
 cylinders with melted sealing-wax, or some other resinous sub- 
 stance, which was found to be efficacious in promoting excitation, 
 doubtless in consequence of the resinous coating protecting the 
 interior of the cylinder against a deposition of moisture. 
 
 73. Plate Electrical Machines. Dr. Ingenhousz, about the year 
 1764, proposed to substitute a plate of glass for the hollow globes 
 and cylinders in use up to that time. This kind of machine, 
 which soon became general throughout Europe, consisted of a 
 circular glass plate of a foot or more in diameter, mounted on 
 a horizontal axis, between two strong vertical supports. The 
 plate was made to revolve between two pairs of cushions. Mr. 
 Cuthbertson, a celebrated mathematical instrument maker in 
 Holland, and afterwards of London, greatly improved this appa- 
 ratus, about the year 1770, He constructed plate electrical machines 
 of 2 feet or more in diameter, and of such power that they 
 exceeded any kind of electrical machine previously designed. 
 Since that period plate machines have been constructed from 2 
 to 8 feet in diameter. 
 
 74. An improved plate machine by Cuthbertson is represented 
 Fig. 44, in which c d is the basis upon which the instrument 
 rests, A c the framework within which the plate revolves, x y z 
 the horizontal axis carrying the plate, z the winch handle, q v 
 the cushions set in light spring frames, so as to apply with 
 pressure to the glass plate by means of compressing nuts and 
 screws, s p silk flaps (68) projecting from the cushions, A w b the 
 prime conductor, projecting from the framework in front of the 
 machine, and terminating in two curvilinear branches, armed with 
 points, for collecting the excited electricity of the glass (59) as it 
 
THE PLATE MACHINE. 69 
 
 passes out from under the silk flaps. The conductor A w b is 
 insulated on a stout glass rod, #, attached by a sliding dovetail 
 piece, u, to the front upright of the framework of the machine. 
 The conductor consists of a light hollow centre piece, w, and two 
 tabular curvilinear branches, A w, w b, extending across a hori- 
 zontal diameter of the plate. When the plate is caused to revolve 
 between the cushions, the excited electricity accumulates on the 
 conductor A w b, from whence it 
 is evolved in bright sparks and 
 scintillations. To remedy the 
 inconvenience arising from the 
 absence of the negative conductor 
 of the cylindrical machine (68), 
 the whole apparatus is insulated 
 upon glass legs, m, n, o, p, fixed in 
 a firm base, F H. A conducting 
 rod and ball, s, projects from the 
 base c d of the framework, as 
 a negative conductor. By these 
 means the entire framework of 
 the apparatus is rendered negative *& 44 ' 0****'* Plate Machine. 
 when the machine is in motion (24), the framework being imme- 
 diately in connection with the rubbers. In order to perserve to 
 the framework a sufficiently perfect insulation, the plate is turned 
 by an insulating varnished glass handle, z. Cuthbertson subse- 
 quently constructed his instrument with two equal glass plates 
 supported on the same axis, by which its power was greatly 
 increased. The two plates were rubbed by four pair of cushions, 
 and the excited electricity taken up by collecting points passing 
 between the plates. 
 
 75. The defect of these machines appears to have been their 
 liability to fracture, and the absence of an especial negative conductor 
 immediately connected with the cushion, as in the cylindrical 
 machine. The method of converting the entire framework into a 
 negative conductor, although allowing negative sparks tobe obtained 
 from the operation of the rubbers, did not succeed very well. Pro- 
 fessor Copland, of Aberdeen, found that, although insulated, the 
 winch handle emitted a flash of negative electricity to the pillar of 
 the machine at every revolution of the plate, perfectly visible in 
 the dark, and the effect of which he felt nearly to his shoulders.* 
 
 76. Van Marum, in 1791, remedied this inconvenience by a 
 machine of new construction (Fig. 45). The plate was 2 feet 
 
 * Nicholson's Journal, vol. xxvi., p. 10. 
 
70 
 
 THE DUTCH PLATE MACHINE. 
 
 7 inches in diameter. This machine, as well as Cuthbertson's double 
 plate, was of an ingenious, although of a rather heavy and some- 
 what complicated character. The cushions A B are each separately 
 insulated on stout glass pillars, c D, and are applied nearly in the 
 
 direction of a horizontal 
 diameter of the glass plate. 
 The prime conductor P is 
 a large ball, to which is 
 attached a light semicircu- 
 lar rod, M P N, terminating 
 in two metallic cylinders, 
 M N, each 6 inches long 
 and 2J inches in diameter. 
 The semicircular branch, M 
 p N, may be turned from a 
 vertical to a horizontal 
 position, so as to bring the 
 metallic cylinders M N into 
 contact with the rubbers, in 
 which case the prime con- 
 ductor exhibits negative 
 electricity. In order that 
 the rubbers may communi- 
 cate with the ground, when positive electricity is required, there is 
 a second semicircular branch, v, on the opposite side of the plate, 
 nearly at right angles to the first, and connected with the ground. 
 "When negative electricity is required, the semicircular branch, MPN, 
 is brought into contact with the cushions, and the opposite semi- 
 circular branch v brought into a nearly vertical position, so as to 
 carry off the electricity of the glass. The axis w of the plate is 
 supported on a single column, G, for which purpose a bearing piece, 
 z, is provided, furnished with two brass collar guides. With a 
 view to counteract friction in the collar guides, the axis is balanced 
 by a counterpoise of lead, E. The conductor P is provided with a 
 long curvilinear copper tube and ball, H, movable like a radius about 
 a stem projecting from the conductor, so as to communicate elec- 
 tricity in any required direction. The middle of the conducting 
 part of the axis is a cylinder, x t of baked wood capped with brass. 
 The glass plate is secured to the front face of this central cylinder 
 by means of a hole drilled through the glass, and a projecting 
 screw and nut, being guarded from undue pressure by means of 
 intervening rings of felt. The whole plate thus turns freely with 
 the axis and winch handle. 
 
 Fig 45. Van Marum's Plate Machine. 
 
HARRIS'S PLATE MACHINE. 
 
 71 
 
 77. In the plate machine about to be described, we have all the 
 advantages of the cylindrical machine and the plate machine of 
 
 Fig. 46. Harris's Improved Plate Machine. 
 
 Yan Marum, under a light form and simple construction. P, 
 Fig. 46, is a circular glass plate 3 feet in diameter, mounted on a 
 horizontal brass axle, x /, and sustained in a strong mahogany 
 
 Fig. 47. Method of Mounting Plate, &c. 
 
 frame, m q. The glass plate is held fast upon its axle x I, between 
 two hollow brass flanges, a b, Fig. 47, guarded by intervening collars 
 
72 
 
 DETAILS OF HARRIS S PLATE MACHINE. 
 
 of leather or felt ; one of the flanges, a, is fixed, the other, b, mov 
 able, as a sort of screw nut, so that the plate revolves with the axle 
 and handle. As in turning the plate the screw flange b is liable to 
 press forward upon the glass, and thereby cause it to crack through 
 the centre, a serious objection to the ordinary plate machine, 
 a small stop screw is passed at b through the projecting edge of 
 the flange, into the substance of the axle, which effectually prevents 
 the pressure from increasing beyond a given point requisite for the 
 movement of the plate. For more effectual security against pres- 
 sure two circular glass plates, c d, 6 inches in diameter and one- 
 quarter of an inch thick (drilled through their centres for the 
 passage of the axle) are attached by thin varnish to the centre 
 of each face of the plate. These circular plates serve to thicken 
 
 and strengthen the large plate at 
 its centre and effectually preserve 
 it against fracture. The plate is 
 excited by two pairs of rubbers, 
 R R', Fig. 46 ; each rubber is 12 
 inches long by 2 a- inches wide and 
 about half an inch in thickness, 
 constructed in the manner of the 
 cushion of the cylindrical ma chine 
 (69). The rubbers R R, Fig. 48, 
 are loosely enclosed between light 
 spring pieces of mahogany, z z'. 
 These pieces are fitted by dove- 
 tails k k r , in a wooden block, B, 
 and are compressible against ^the 
 brass plate by means of short brass 
 rods, a b, passing through the spring pieces immediately before the 
 block. The rods have screw terminations and small globular nuts, 
 a b. These press the rubbers gently toward the plate from opposite 
 sides. They are insulated upon stout glass pillars, n m, o q, as in 
 Fig. 42, one at each extremity of a diameter of the plate ; the 
 blocks and spring holders, in which the rubbers are fixed, are 
 secured to the insulated pillars by means of brass caps, with 
 projecting screws and globular nuts. Each rubber, R R', has an 
 oiled silk flap, R s, R u, Fig. 42, attached to it, shaped to the plate, 
 the silk being prepared as directed in the Chapter on Electrical 
 Manipulations, and applied with the rough side next the glass. 
 Each pair of flaps is united over the edge of the plate by narrow 
 silk binding. They are prevented from dragging upon the plate 
 by silk cords,//*, carried round a descending rod of glass,/, on one 
 
 48. Details of Cushion. 
 
DETAILS OF PLATE MACHINE. 73 
 
 side, inserted in a sliding brass tube passing through the block 
 of the rubber, and on the opposite side, round a light perpen- 
 dicular tube of brass, f y, inserted in the ball confining the block 
 to the insulator, as shown in Fig. 46. 
 
 78. The prime conductor a c b, Fig. 46, is placed, in a vertical 
 position in front of the plate, and consists of two curvilinear tubu- 
 lar branches a c, c b, proceeding from a large central hollow ball, 
 c. A cylindrical brass socket, s, Fig. 49, 3 inches long, proceeds 
 from this ball, and receives 
 
 within it a similar cylindrical 
 socket, T, terminating in a 
 second hollow ball, D> the two 
 sockets fitting so * closely as to 
 admit of an easy circular and 
 sliding motion. This last ter* 
 minating ball, D, carries a central Fi g- <19; Telescopic joint conductor. 
 
 sliding-rod, t t, having a metallic ball at each extremity. The rod 
 1 1 may be turned to any angle by means of the movable telescopic 
 socket T. The experimentalist may hence lead off connecting wires 
 from the conductor in any convenient direction. The curvilinear 
 branches a c, c b, of the conductor a c b, extend to within a short dis- 
 tance of the glass plate, and have affixed to them light vertical rods, z z', 
 Fig. 46, about 8 inches long, and three-eighths of an inch in diameter, 
 terminating in balls of varnished wood. These vertical rods are fur- 
 nished with thin knife-edged pieces of brass plate, about half an inch 
 wide, and 6 inches long, carrying small collecting points (59). The 
 points project about one-quarter of an inch beyond the knife-edge, 
 immediately opposite the terminations of the silk flaps n s, R u, 
 attached to the cushions. The vertical conductor in front of the 
 plate is insulated on, and supported by, a glass rod, g, of 1 inch in 
 diameter, and 1 foot in length, furnished with two brass caps, one 
 of which connects it with the brass socket within the hollow of the 
 ball c, so that the ball with the conductor can be turned about upon 
 the brass cap of the insulating rod g, if required ; the other cap is 
 connected with a dovetail plate fitted into a groove attached to the 
 centre of the front cross-bar supporting the axle, as in the figure. 
 
 79. The negative conductor n e h, Fig. 46, passes behind the plate, 
 and consists of two metallic tubular curvilinear branches, e n, e h, 
 three-quarters of an inch in diameter. These connect together the 
 insulated blocks n R' that carry the rubbers. The curvilinear 
 branches en, eh, are united over a piece of flexible metal tube, 
 through a brass ball, e, insulated, on a glass rod, p, 6 inches long 
 and half an inch in diameter. This insulating rod is fixed by a brass 
 
74 DETAILS OF HARRIS'S MACHINE. 
 
 cap upon the centre of the back cross-bar of the framework of the 
 machine, immediately over the axle of the plate. To render this 
 negative conductor perfectly efficient, and further cut off all com- 
 munication with it, the plate is turned by a strong insulating winch- 
 handle, H, of varnished glass, about 1 foot in length and a full inch 
 in diameter. The negative conductor n e h is supported upon the 
 blocks carrying the rubbers by means of spherical sockets, attached 
 to two circular pieces of brass about half an inch thick, which 
 resist the pressure of the balls, securing the blocks of the rubbers 
 to the insulating pillars. The brass ball x, Fig. 46, which fixes the 
 rubber on the side x to the insulator, carries a vertical tube of brass, 
 w x, terminating in a small ebony ball, w. The vertical brass tube 
 proceeding from the ball x, carries a sliding ball, v, upon it, for the 
 
 convenience of attaching con- 
 necting rods to the negative 
 conductor, when required. The 
 brass ball y, uniting the rubber 
 on the side y to the insulator, 
 carries also a light vertical brass 
 tube,/' ^terminating, as in the 
 opposite side, in a ball, w. A 
 light open circle, c, of silvered 
 brass, or varnished cardboard, 
 6 inches in diameter, and 
 divided into eight parts, is fixed 
 centrally round the axle, where 
 
 Fig. 50. Back view of Harris's Mudihie. it paSSCS through the back CrOSS 
 
 bar of the framework of the machine, as shown in Fig. 50, which 
 is a back view of the whole apparatus. By means of an index, i, 
 upon the axle, we are enabled in turning the plate to estimate 
 the number of its revolutions. 
 
 80. When the instrument is so arranged as to convert it into a 
 negative machine, the prime conductor is turned upon its insulator 
 into a horizontal position (Fig. 51), and its collecting points placed 
 in contact with the rubbers. The original negative conductor, 
 passing behind the plate, is in this case entirely removed, and its 
 place supplied by a vertical positive conductor, of the form a R b, 
 attached to a wide cylindrical brass ring it, movable with friction 
 upon a fixed socket surrounding the axle. This conductor consists 
 of two light tubular arms, A R, E, B, about a quarter of an inch in 
 diameter, each inclining at an angle toward the plate, being held 
 fast on two stout brass pins screwed into the movable ring R. 
 These arms carry collecting points, a b, as in the original prime 
 
DETAILS OF PLATE MACHINE. 75 
 
 conductor. The branches a R, R b, fit closely upon the pins of sup- 
 port, and may be removed at pleasure. We are enabled, by the 
 circular motion of the ring upon the brass socket surrounding 
 the axle, to adjust the conductor that carries off the electricity of 
 the glass, in an exact vertical position ; or in again replacing the 
 original prime conductor in its first position, we can turn down 
 the new positive conductor a R b so as to connect the rubbers, as 
 
 ^IBRARY 
 
 IVKKSifY OP I 
 
 AL1FOHNIA. 
 
 Fig. 51. Harris's Plate Machine for Negative Electricity. 
 
 seen in Fig. 50. When acting as a positive conductor, however, 
 it is merely employed to carry off the electricity of the glass, in 
 which case it is connected with the ground ; whilst the original 
 prime conductor, acting horizontally, receives the electricity of the 
 rubbers. The machine arranged in this way as a negative machine, 
 is wonderfully efficient, and sends forth a torrent of negative sparks 
 from the horizontal prime conductor ; so that both positive and 
 negative electricity are thus obtained. 
 
 Such is the construction of the instrument as fitted for positive 
 and negative electricity. It is novel in form, and of convenient 
 and general application. 
 
 The base and framework upon which the machine is supported 
 are of seasoned mahogany. The base consists of two open rectangular 
 frames, each 4 feet 6 inches long, and 2 feet wide, the pieces form- 
 
76 VARIOUS DIMENSIONS OF PLATES. 
 
 ing the frames being 4 inches wide by 3 inches thick, neatly 
 moulded, and rounded at the angles. The lower of these frames, 
 w, Fig. 51, is set upon castors to admit of an easy movement of 
 the whole instrument, and has six levelling screws through it, in 
 order to give it, when in place, a firm bearing upon the table or 
 floor. The upper frame m q is insulated upon the lower frame w 
 by four stout conical glass feet, 51 inches long. These are received 
 below in small shallow cavities, and are screwed into the under 
 part of the upper frame, through the medium of brass caps, nuts, 
 and screws. 
 
 The glass plate of the machine is supported about 28 inches 
 above the upper frame m q, by means of four cylindrical mahogany 
 pillars, from 11 inch diameter above to 21 inches below. They 
 are about 20 inches apart, and are united above by stout cross- 
 pieces, which receive the axle of the plate. The supporting 
 columns and cross-bars are carefully moulded and fashioned, and 
 are securely united to each other, and to the framework, by iron 
 screws and nuts. The centres of the cross-bars, through which 
 the axle passes, are formed ornamentally into cubical blocks 
 2-J inches wide by 3? inches deep : the bars being nicely rounded 
 up to their terminations in other blocks joining the vertical 
 pillars. 
 
 81. The diameter of a plate machine may vary, as already 
 observed (73), from 2 to 8 feet, or more. A 3-foot plate, however, 
 is upon the whole the most convenient size. Plates of 18 inches, 
 2 feet, and 30 inches diameter, are powerful and efficient. The 
 electrical machine constructed by Cuthbertson in the year 1780, 
 for Teyler's Museum at Haarlem, consisted of two circular glass 
 plates, each 65 inches in diameter, fixed in the way already 
 referred to (74), upon the same horizontal axis, parallel to each 
 other, and at a distance apart of 7J inches. The glass plates were 
 excited by eight rubbers, each 15^ inches long, placed in appro- 
 priate frames. The prime conductor was divided into branches 
 supported by three glass pillars, 57 inches long. The branches 
 entered between the plates, and by means of fine points, collected 
 the excited electricity. This machine appears to have been of 
 great power. Two, and occasionally four men were required to 
 work it. When in good action, the following effects were observ- 
 able : A sharp steel point presented to the prime conductor drew 
 forth a luminous electrical stream full half an inch in length. The 
 same point fixed to the conductor, so as to project 3 inches from 
 it, emitted similar streams 6 inches long, when a ball of 3 inches 
 in diameter was presented to it ; and 2 inches long when another 
 
WINTER'S ELECTRICAL MACHINE. 77 
 
 point was presented instead of the ball. The peculiar creeping 
 sensation, commonly called spider's web, was felt on the face of the 
 bystanders often at the distance of 8 feet. A delicately suspended 
 thread, 6 feet in length, was sufficiently attracted by the prime 
 conductor, at the distance of 38 feet, to cause it to deviate sensibly 
 from the perpendicular (5) . The electricity of a pointed wire rod, 
 28 feet distant from the conductor, appeared luminous, A second 
 conductor of similar dimensions being presented, in order to receive 
 the sparks, and a perfect metallic communication being made 
 between the second receiving conductor and the earth, by means 
 of a long brass wire three-eighths of an inch in diameter, it was 
 found that whilst a stream of electricity passed from the prime 
 conductor to the receiving conductor, the brass wire gave small 
 sparks to conducting bodies placed near it. Hence the quantity 
 of electricity evolved from the prime conductor must have been 
 considerable, since a wire one- eighth of an inch in diameter could 
 not perfectly transmit the accumulation to the earth. The sparks 
 between the two conductors were generally 21, but sometimes 24 
 inches long. The electrical stream between the conductors appeared 
 crooked, and sent forth lateral branches of a large size. 
 
 Since that time, plate electrical machines have been constructed 
 from 8 to 10 feet and upwards in diameter, turned by steam 
 machinery, the effects of which have been of a surprising character. 
 
 82. The old electricians, with a view of obtaining dense and 
 powerful sparks, were frequently in the habit of employing prime 
 conductors of enormous dimensions (71). In our own time, that 
 is, about the year 1856, Professor Winter, of Vienna, applied to 
 a plate machine of a peculiar construction (the plate was 19 
 inches in diameter), a circular prime conductor under the 
 form of a large ring, from which he was enabled to obtain 
 unusually long and brilliant sparks, quite equal to the sparks 
 obtained from the machine of Teyler's Museum, at Haarlem. 
 This new form of prime conductor consisted of a ring of polished 
 wood, nicely rounded at its edges, about li inch thick, and 
 27 inches in diameter, the ring being constructed in parts, and 
 having a stout metallic wire passing throughout its substance. 
 The ring is upheld by a stout metal rod, 20 inches long, in- 
 serted in the prime conductor of the machine, insulated on a 
 strong pillar of glass. Fig. 52 represents Winter's ring, c is 
 the ball of the prime conductor, g the insulating pillar, d the 
 metallic rod, inserted in the conductor that carries the ring. The 
 electricity evolved by the machine accumulates on the ball c 
 of the conductor in the usual way, and is transferred to the 
 
78 EFFECTS OF PLATE MACHINE. 
 
 ring R, through its metal support. It is necessary to keep the 
 ring R at least 2 or 3 feet away from the ceiling of the room. 
 With an electrical plate of 3 feet in diameter, Professor Winter 
 is said to have obtained vivid electrical sparks of nearly 2 feet 
 in length. This machine was excited between two circular 
 rubbers, each 6 inches in diameter, furnished 
 with collecting points, and consisted of leather 
 pads stuffed with cotton wool. It appears to 
 have been powerful. 
 
 The advantage of Winter's ring seems to 
 consist in this, that whilst its rounded edges 
 tend to retain the accumulated electricity, the 
 intensity of action is increased by its prominent 
 circular elongation, and thus a brilliant dense 
 spark escapes from it. 
 
 83. Electrical machines demand especial care 
 and attention in preparing them for excitation. 
 If of glass, it is requisite the glass be clean 
 and dry ; the exciting rubbers should be carefully 
 covered with electrical amalgam (7), and this should not be 
 applied beyond the junction of the silk flap with the rubbers, 
 but quite in the rear of it. All the insulators should be varnished 
 with a solution of shelUlac in naphtha, dried off by heat ; and 
 it will be found advantageous tq interpose a layer of clean, dry, 
 warm writing-paper between the rubbers and the glass, previously 
 to turning the plate a few rounds. Indeed the exposure of the 
 plate to friction with clean, warm, dry paper for a short time, or 
 during a few revolutions, has a very beneficial effect. 
 
 84. When a powerful electrical machine is set in motion in a 
 dry atmosphere, the rubbers pressing gently on the glass, the 
 following phenomena will be apparent : 
 
 The conductors being removed, brilliant circular streams of 
 light, or, as the old electricians styled it, electrical fire, attended 
 by a sharp crackling sound, dart round the plate or cylinder, whilst 
 luminous corruscations play about its circumference, producing 
 a very brilliant and beautiful effect. The peculiar creeping sensa- 
 tion, or spider's web, is felt on the face and hands, and a vaporous 
 odour, termed ozone, affects the sense of smell. The conductors 
 being in place, similar phenomena present themselves. Bright 
 sparks occasionally shoot back over the glass not covered with 
 the silk flap, to that part of the conductor carrying the collecting 
 points, and often continue to the rubbers under the flap. If the 
 knuckle be presented to the glass plate or cylinder beyond the 
 
EQUABLE ACTION OF MACHINE, 
 
 79 
 
 rubber, a sharp spark will frequently be experienced. "When 
 one knuckle is presented to the positive, and the other to the 
 negative conductor, both conductors being insulated, a pungent 
 and unbearable succession of electric sparks is experienced. If 
 the two conductors be united by a metallic wire or rod, all these 
 phenomena vanish. If for the metallic rod a series of small shot 
 be substituted, strung upon a silk line about one-thirtieth of an 
 inch, or even less, apart, a luminous and brilliant effect is pro- 
 duced by the electricity passing from one shot to the other, 
 as if an electrical current existed between the two conductors. 
 Either conductor emits the most powerful spark when the oppo- 
 site conductor is connected with the ground (41). A similar 
 effect ensues on linking the two conductors together by an 
 iron jack chain, which may be prolonged in festoons by means of 
 insulators. 
 
 85. Of the different forms of electrical machine the cylinder is 
 the most simple and convenient, and the least troublesome, as 
 there is only one rubber. In this form, and without any com- 
 plicated mechanical arrangement, both electricities are at command, 
 for which purpose we have merely to transfer our operations from 
 one side of the machine to the other. 
 
 86. The electrical machine is, when well and perfectly con- 
 structed, equable and continuous in its action, and may be con- 
 
 rig. 53. Opposed Discharging Balls. 
 
 sidered to produce at every revolution the same quantity of excited 
 electricity. This is an unquestionable experimental fact, although 
 several electricians, especially in Germany, have entertained an 
 adverse opinion. A very little reflection, however, will suffice 
 to put the fact in evidence. 
 
 Exp. 33. Let p N, Fig. 53, be two light, polished, hollow 
 balls of copper, from 3 to 4 inches diameter, mounted on insulating 
 supports inserted in sliding bases, T v, and set at a given distance 
 apart. Let one of the balls, p, be connected with the positive con- 
 ductor, and the opposite ball, N, with the negative conductor. 
 
80 QUANTITY OF ELECTRICITY FROM MACHINE. 
 
 Let the touching points p n of the balls be set at such a distance 
 apart, that on carefully turning the electrical machine a given 
 number of discharges between the balls may take place at each 
 two or three revolutions. By means of a pendulum, or the musical 
 regulator termed a metronome, let the time of the revolutions 
 be accurately noted. It will be found that the same number 
 of revolutions in the same time will evolve the same number of 
 discharges between the balls, consequently the same quantity 
 of excited electricity must have been developed at each revolu- 
 tion of the machine. We suppose, of course, the machine to have 
 been carefully constructed, and in good order. The number of 
 discharges between the balls therefore may stand for and repre- 
 sent the quantity of electricity at each given number of revolu- 
 tions. It has, however, been said by Professor Riess of Berlin, 
 and others, in opposition to this very plain demonstrative fact, 
 that inasmuch as the sparks are commonly observed to decrease 
 in vigour, and also in number, as the charge proceeds, therefore 
 the machine does not produce equal quantities of electricity 
 in equal times, or the same quantity at each revolution. Now it is 
 by no means difficult to perceive the insufficiency of this observa- 
 tion ; it is clear that any diminution of the sparks in frequency 
 from the prime conductor, as the charge of a limited surface of 
 accumulation advances, does not depend upon any decreasing 
 power in the machine to produce the increased quantity, but in 
 the decreasing power of the receiving surface to accumulate the 
 quantity produced. Unless the quantity evolved from the machine 
 can be as rapidly taken up as it is produced, it will be, of course, 
 impossible to maintain an infinite succession of equal quantitative 
 sparks between the balls p N, supposing the capacity of the nega- 
 tive ball N to be limited. Directly, however, we connect the 
 ball N with the negative conductor, or with the earth, and set it 
 free from this limitation, we obtain an infinite series of explosive 
 sparks of precisely the same magnitude. Imagine, therefore, the 
 insulated conductor, or other insulated surface charged from the 
 prime conductor of the machine, to have sufficient electrical 
 capacity or extension ; the revolutions of the machine are uniform 
 and equable up to a given limit of charge, and may be taken as 
 an accurate measure of the quantity of excited electricity up to 
 that limit, or nearly .so. Let, for example, the limit of charge 
 of a given surface charged to saturation be one hundred revolu- 
 tions of the plate, we may conclude, without sensible error, that 
 when charged with forty revolutions the given surface would 
 have received twice the quantity of electricity it would have 
 
ELECTRICITY FROM STEAM. 
 
 received with twenty revolutions, and such is really the case, as 
 is demonstrable by experiments. 
 
 87. Steam Machine. This species of electrical machine is of 
 modern date and construction, and is the result of an accidental 
 observation (P). 
 
 Armstrong's machine consists of a steam boiler, A, Fig. 54, insu- 
 lated on stout pillars of glass. The steam is made to issue from a 
 general steam-pipe through bent iron tubes, a b c, terminating in 
 jets of wood, and of which there are a large number. An insulated 
 projecting conductor, N, is placed in connection with the boiler 
 for the convenience of collecting the excited electricity, and a 
 second conductor, P, formed of a metallic case furnished with 
 several rows of points, is placed im- 
 mediately in front of the jets to 
 receive and carry off the electricity of 
 the steam, and prevent its return upon 
 the boiler, by which the opposite forces 
 would be neutralised. Faraday, who 
 investigated this question with his 
 accustomed tact and penetration, en- 
 deavours to show, by a series of 
 masterly experiments, that the elec- 
 tricity thus produced does not depend 
 upon any chemical or other change 
 which may be supposed to arise from 
 evaporation or condensation, but is the Fi - 64 - steam Electrical Machine, 
 result of the friction of condensed particles of water whilst being 
 driven by the still issuing steam through the jets, so that, in fact, 
 these particles perform the office of the glass of the common 
 electrical machine (68), and give out vitreous or positive electricity ; 
 the wood jets and pipes act as the rubber, and give out resinous 
 or negative electricity ; the friction of the steam in passing through 
 the jets being the source of electrical power. The electricity pro- 
 duced by this apparatus is enormous in quantity. The sparks from 
 the conductor N, upon an insulated metallic ball, are dense and 
 rapid, presenting frequently the appearance of a continuous flame, 
 and will readily set fire to inflammable matter. 
 
 Although the friction theory of the operation of this instrument 
 adopted by Faraday appears consistent with striking facts, it is 
 nevertheless not perfectly clear that the condensation and evapo- 
 ration of water as steam is altogether without influence in the 
 electrical development. 
 
 88. The Electric Column; or, Perpetual Electrical Machine. 
 
82 THE DllY PILE, 
 
 When a series of thin discs of two different metals zinc and silver, 
 for example, or zinc and copper are alternated with similar discs of 
 common writing paper in a dry state, the result is a peculiar electro- 
 motive action. The zinc disc of the series exhibits positive, the 
 copper or silver disc negative, electricity ; and the arrange- 
 ment will act powerfully on the gold-leaf electroscope (32). 
 For example, let A B, Fig. 55, be a clean tube of glass, 
 about three-quarters of an inch in diameter, varnished 
 both internally and externally with a thin clear coating of 
 shell-lac dried off by heat. "Within this tube is a series 
 of circular discs of very thin zinc and silver, or zinc and 
 copper, alternated with similar discs of dry paper, in the 
 succession of silver, zinc, paper, silver, zinc, paper, and 
 so on. The zinc disc at one extremity of the series 
 exhibits positive, and the terminating silver or copper 
 disc negative, electricity. If the tube A B be capped at 
 each extremity with short caps or rings of brass, A B, 
 communicating with the first, and terminating discs, by 
 compressing screw rods, a b, passing through the caps at 
 Dry Pile, each extremity of the column, one of the caps will affect 
 the electroscope positively ; the other negatively. 
 
 The amount of positive and negative action will of course depend 
 on the extent of the series. A series of a thousand alternations 
 will be extremely sensible to the gold-leaf electroscope, and may 
 be employed with advantage to produce a given divergence of 
 the leaves without violence. With a series of from two to three 
 thousand discs, comparatively powerful electrical effects result. 
 
 The most simple way of constructing the electric column is to 
 cut out, by means of a hollow punch, from very thin zinc plates, 
 circular discs, about three-quarters of an inch in diameter. Simi- 
 lar discs should then be prepared in the same way from dry 
 cartridge paper, covered on one side with silver leaf. We then 
 proceed to arrange the discs in succession within the tube, between 
 its capped extremities, in the order of zinc, silvered paper the 
 silvered side next the zinc ; then again zinc, silvered paper the 
 silvered side next the zinc and so on. We have then a series 
 grouped in the order of zinc, silver, paper, zinc, silver, paper, and 
 so on, the zinc being in contact with the silver throughout, and 
 each pair of zinc and silver plates separated by dry paper. One 
 of the caps should be fixed before the discs are introduced into the 
 tube, and each cap should have a compressing screw- rod passing 
 through its centre, in order to press the discs closer together, 
 and effect a good contact with the terminating discs of the 
 
OR ELECTRIC COLUMN. 83 
 
 column. The extremities of the screw rods should terminate in 
 small metallic balls, these being found convenient for conveying 
 weak electrical charges to the gold-leaf electroscope. 
 
 The operation of this species of electrical apparatus appears to 
 depend on a peculiar kind of electro-motion, induced by the contact 
 and separation of dissimilar bodies, and has apparently some rela- 
 tion to the phenomenon of excitation, as previously exemplified 
 (26). We here observe that when the rubber and glass, two dis- 
 similar bodies, are first brought into close contact, and subsequently 
 separated, both evince a state of electrical excitation ; the one 
 being electrified positively, the other negatively : thus leading to 
 the conclusion that electrical excitation is more or less dependent 
 on the contact and separation of dissimilar bodies. 
 
 The celebrated experimental electrician Singer, by bringing 
 various insulated bodies into contact and then separating them, suc- 
 ceeded in developing positive and negative electricity. Thus an in- 
 sulated polished circular plate of zinc, of about 4 inches in diameter, 
 being brought into contact with a well-insulated polished circular 
 plate of copper, of the same diameter, the two plates evince on 
 separation opposite electrical states, which become sensible to the 
 electroscope ; the copper plate being negative, the zinc plate posi- 
 tive. These states are more fully apparent after several contacts 
 of the two plates. Hence it has been inferred that in the opera- 
 tion of the electric column, the associated metals, separated by 
 semi-conducting matter, such as paper, give rise to a propagation 
 of electrical action analogous to electrical excitation; the zinc 
 acquiring a positive electrical state by contact with the copper or 
 silver, and the interposed paper transferring the positive electricity 
 thus developed from the copper to the next zinc, and so on through- 
 out the series, so that at last all the developed positive electricity 
 becomes accumulated on the terminating zinc disc, and all the 
 negative electricity on the terminating silver or copper disc. Thus 
 the column is, as it were, a spontaneous source of electrical excita- 
 tion. It will be seen that in this kind of electrical machine there 
 are three associated substances two metals, and an imperfect con- 
 ductor viz., zinc "and silver, or copper, and paper ; being so far 
 analagous to the voltaic apparatus, which consists of two metals, 
 and an interposed fluid, and which in the hands of Davy effected 
 such marvellous decompositions. There is, however, this remark- 
 able difference between them, that whilst the voltaic apparatus is 
 celebrated for its electro-chemical power, the electric column has no 
 electro- chemical power at all, but is principally remarkable for its 
 pure electrical effects, resembling common electrical excitation (7) . 
 
 c2 
 
84 
 
 THE DRY PILE. 
 
 It was the opinion of Yolta that the fluid interposed between 
 the metals of his apparatus serves only to conduct the electricity 
 developed in one pair of metals to the succeeding pair. In a 
 similar way the discs of paper in the electric column may be con- 
 ceived to break the continuity of the opposite metals, and at the 
 same time transmit the electricity developed by the one pair to 
 the succeeding pair, constituting- a species of electro-motion. 
 
 89. The pure electrical effects of the column show that the 
 arrangement of a series of zinc and silver, or copper discs, with 
 intermediate paper, is a semi- conducting column, which in its in- 
 sulated state is positive at one extremity, negative at the other, 
 and neutral in the centre. 
 
 Exp. 34. Connect a column, A B, Fig. 56, of 500 to 1,000 
 
 Fig. 56. Action of Dry Pile on Electroscopes. 
 
 alternations, with three delicate gold-leaf electroscopes, A, B, c ; one 
 being at one extremity, A ; another at the opposite extremity B ; 
 and a third at the centre c. The electroscope connected with 
 the copper termination will diverge with negative 
 electricity ; that connected with the zinc termina- 
 tion will diverge with positive electricity ; whilst 
 the electroscope at the centre c will appear neutral. 
 If either extremity, A or B, of the column be con- 
 nected with the ground, and thus set free, the 
 leaves of the electroscope connected with that ex- 
 tremity will close ; those of the central electroscope 
 c will immediately diverge with the same electricity, 
 whilst the leaves of the electroscope at the remaining 
 insulated extremity will, in accordance with ordinary 
 inductive action, have their original divergence in- 
 creased (41). This indicates a real electro-motive 
 property in the apparatus by which the zinc 
 extremity accumulates positive electricity, and the 
 copper extremity negative. 
 90. The opposite states of the electric column have given rise 
 to an ingenious instrument termed the Perpetual or Electric Chime, 
 represented in Fig. 57, in which A B, c D, are two active electric 
 columns, containing about 1,000 groups each, set upon two small 
 
 Fig. 57. 
 Electric Chime. 
 
THE ELECmOPHORUS. 
 
 85 
 
 insulated bells, B D, their zinc and copper terminations, A and c, 
 being connected by a small conducting rod, w. A light metallic 
 ball or clapper, q, is suspended between the bells by a fine silk thread, 
 which, if the series composing the column be extensive, will con- 
 tinue to vibrate between the bells (18), keeping up a perpetual chime, 
 and affording some ground for the conclusion that the electro-motive 
 power is perpetual. A series of about 1,500 groups was found by 
 De Luc to sustain a vibration in this way for several years, giving 
 rise to many important meteorological results. If the electric 
 power be really continuous, we have no doubt arrived at perpetual 
 electrical motion. If a sufficiently extensive series be employed, 
 the electrical development appears to be so continuous and of such 
 duration as to go far in support of this deduction. 
 
 91. The electric column has been further employed as a delicate 
 means of determining the positive or negative state of a body, 
 through the medium of the gold-leaf electroscope. 
 
 Fig. 58 represents a sensitive differential arrangement applicable 
 to this purpose. L is a single gold-leaf elec- 
 troscope constructed in the usual way (32). 
 The gold leaf is suspended from an insulated 
 metallic rod, between the hemispherical ter- 
 minations of two conductors, p n, passing 
 through small holes drilled in opposite sides 
 of the glass enclosing the gold leaf, n s is 
 a horizontal insulated electric column, the 
 extremities of which communicate with the 
 rods p n by small conducting wires, and thus 
 place its flattened terminating hemispheres 
 in opposite electrical states (89). Conse- 
 quently, if the electricity affecting the gold 
 leaf through the plate q be positive, the leaf is attracted toward 
 the negative side n ; if negative, it is attracted toward the positive 
 side^? (27) ; thus the electrical state of any body affecting the plate 
 q is immediately made evident. 
 
 92. The Electrophorus. If an ideo-electric plate (8), N, Fig. 59, 
 be insulated on a glass support, s, its upper sur- 
 face being excited by friction, and covered by a 
 
 metallic plate, p, having an insulating handle, 
 
 m, the excited electricity does not leave the 
 
 surface of the ideo-electric plate N to accumulate 
 
 on the metal cover p, but acts upon the metal 
 
 cover by simple induction (38), and in such way 
 
 as to determine the electricity of the metallic plate P upon the 
 
 excited surface of the ideo-electric plate N. 
 
 Fig. 58. Differential 
 Electroscope. 
 
 Fig. 69. Single 
 Electro] ' 
 
 strophorus. 
 
86 THE ELECTROPHORTJS. 
 
 Let, for example, the ideo- electric plate N be a circular resinous 
 disc suppose a disc of coarse sealing wax, about half an inch, in 
 thickness its upper surface being negatively excited by rub- 
 bing it with dry flannel or soft silk (30). Let P be a circular 
 metallic plate, or coyer, of nearly the same dimensions. It may 
 be of wood covered with tinfoil, and about three-eighths of an 
 inch thick, reposing upon the negatively excited surface of the 
 resinous plate. The negative electricity of the excited plate does 
 not leave it, but acts inductively on the metallic cover, 
 determining its positive electricity upon the negatively excited 
 surface of the resinous plate, leaving the metallic cover nega- 
 tively electrified. If, under these circumstances, we present a 
 conducting ball or the knuckle of the hand to the cover, positive 
 electricity will be given off to it from the knuckle under the form 
 of an electric spark, the opposite electrical forces will combine, and 
 the cover will be again rendered apparently neutral. Let the cover 
 be now raised from off the excited plate by its insulating handle ; it 
 then bears off or carries away the positive spark communicated to 
 it, and which satisfies its previously negative state, so that it will 
 now appear positively charged, and will return to the knuckle or 
 a conducting ball the identical spark it had previously received, 
 and again render the cover neutral. By repeating this process of 
 successive contacts of the excited plate and its cover, and the 
 subsequent withdrawal of the cover, we may bear off and obtain a 
 series of electric sparks to a large amount without a new excitation 
 of the resinous plate. An excited electrical plate and cover 
 thus circumstanced constitutes a species of spontaneous electrical 
 machine, termed an Electrophorus, from the Greek words r]\eKTp6v, 
 electricity, and 0opw, to bear or carry off, the metal plate carrying 
 away after each contact an electrical charge. The simple combi- 
 nation of an electric plate having a metallic cover with an insu- 
 lating handle is all that is absolutely requisite to this result. An 
 electric plate, therefore, having an insulating conducting cover, as 
 represented Fig. 59, may be termed a single electrophorm. Any 
 electric substance may be employed for the purpose. A resinous 
 electric plate is, however, generally chosen, resinous bodies being 
 more especially retentive of excited electricity. A resinous plate, 
 consisting of shell-lac, common resin, and Venice turpentine, in 
 equal parts, cast at a temperature of from 230 to 250 degrees, 
 within a metal or wood ring upon a marble slab, or smooth metal 
 surface, is very efficient. [Electrical Manipulation. 94.] A plate 
 of second or inferior sealing-wax is very efficient. A circular 
 plate of glass may be employed with success, especially when con- 
 
THE COMPOUND ELECTROPHORTJS. 87 
 
 sisting of two plates of flat glass joined by means of sealing-wax. 
 [Manipulation. 94.] A plate of this kind is extremely excitable. 
 A plate of brimstone, cast upon glass in a similar way to the 
 preceding, is remarkably efficient. It is, however, liable to crack 
 by changes of temperature.* 
 
 93. The Compound Electrophorus. The ordinary method of 
 preparing an electrophorus is to place the electrophorus disc N, 
 Fig. 59, between two metal discs, P and Q, Fig. 60, of somewhat 
 less diameter, one of them, Q, being in connection with the ground, 
 through any convenient conducting stand of support, T, the upper 
 plate P being placed as in the preceding case. Such an arrange- 
 ment may be termed the Compound ^Electrophorus. The conduct- 
 ing plates may be of light wood, well rounded at the edges and 
 smoothly covered with tinfoil 
 
 Let N, Fig. 60, be an electrophorus plate, resting on the 
 metallic plate Q, supported on a light brass 
 pillar and foot, T, and covered by a metallic 
 plate, P, as before. This arrangement is 
 subject to the general laws of induction (41). 
 We have in fact two metallic conducting 
 plates, P and Q, opposed to each other through 
 an intervening dielectric, N. The upper disc 
 P has been termed the cover, the under disc rig. eo. compound 
 
 .,, .-, ,T .1 7 Eleetrophorus. ' 
 
 Q, in communication with the earth, the sole. 
 This has been found to increase the action of the single electro- 
 phorus (92). How it contributes to this result is not imme- 
 diately apparent, seeing that the evolution of an electrical 
 spark may be obtained without it. 
 
 94. In order to explain the action of the compound electro- 
 phorus many complicated theoretical views have been advanced. 
 The action of the compound electrophorus, however, admits of a 
 simple and practical solution. 
 
 We have seen (41) that an insulated conductor has its capacity 
 for electricity increased by induction when it is opposed to an 
 uninsulated conductor of great superficial extent, or in connection 
 with the ground. In the case of the compound electrophorus, 
 Fig. 60, the cover P is an insulated conductor, the sole Q an 
 uninsulated conductor of unlimited extent, being in connection 
 with the earth, and directly opposed to the cover, through an 
 insulating medium. Hence, by the laws of induction (41), the 
 cover P can receive from the electrophorus plate N, under the 
 
 * The best method of preparing electrophorus plates will be found in 
 our chapter on Electrical Manipulation (94). 
 
88 THE ELECTROPHORUS. 
 
 influence of the sole, a greater charge, under a given intensity, 
 than it could receive without such auxiliary influence. Such 
 appears to be the advantage of the compound electrophorus. An 
 inductive action is set up between the sole and the cover ; in- 
 creasing the power of the cover to bear off electricity from the 
 excited resinous plate. 
 
 95. Electrophorus Manipulation. Having raised the cover from 
 off the electrophorus plate (93), so as fairly to expose its surface, 
 we proceed to excite it. Supposing the electrophorus plate to be 
 resinous, this is best effected by gentle friction with woollen stuff, 
 or by dry soft silk, as in whisking a dry soft silk handkerchief 
 across it. Both the woollen stuff and the silk, together with the 
 electrophorus plate, should be warm, dry, and perfectly free from 
 moisture. The woollen stuff may be a piece of warm dry flannel, 
 loosely formed into a roll about a foot long ; one end of the roll 
 being held in the hand and the other extremity swung round over 
 the surface of the plate in an oblique direction with a quick turn 
 of the wrist, so as to have repeated contacts with the plate in a 
 way varying between a rub and a blow. This will excite the 
 plate to a considerable extent. The cover P, held by its insulating 
 handle, is now placed on the excited surface. If a conducting 
 communication be made between the cover and the sole by means 
 of the thumb and forefinger, a slight electrical shock will be felt, 
 and a small spark will pass, owing to a combination of the opposite 
 electricities. If the knuckle or a ball be presented to the cover, a 
 strong electrical spark will be elicited in the way already explained 
 (92). If the cover be now raised by its insulating handle from 
 off the excited surface, then, as in the single electrophorus, 
 the same identical spark which had been communicated from the 
 knuckle to the coyer will be returned to the knuckle, or any con- 
 ducting substance presented to the cover. If the electrophorus 
 plate be resinous, we obtain in this way a succession of positive 
 sparks. If the electrophorus plate be vitreous, a succession of what 
 may be termed negative sparks will be obtained in a similar way. 
 The whole operation, as is evident, is an affair of induction. 
 
 The electrophorus is a convenient instrument for obtaining 
 sparks of positive or negative electricity, or minute electrical 
 charges, at any required moment. In the case of minute elec- 
 trical charges, we employ extremely small electric discs, arranged 
 as in the single electrophorus (92). The cover may be of 
 gilt wood. The electrophorus disc may be of the best ordinary 
 sealing-wax. Minute electrophori of this kind will always supply 
 positive or negative electricity, to a greater or lesser amount, for 
 
THE ELECrrHOPHORUS. 89 
 
 electrical investigation, more especially in the employment of the 
 gold-leaf electroscopes. Instruments varying from 1 inch to 5 
 inches in diameter, may be advantageously resorted to. An electro- 
 phorus from 10 to 20 inches in diameter, if well constructed, has 
 considerable power. The celebrated Yolta, Professor of Physics at 
 Pavia, was an original inventor of this instrument ; an account of 
 it was published by him in 1775. The Professor of Physics at 
 Stockholm claims, however, to have been the first inventor in 
 August, 1772 ; as may be seen in the memoirs of the Stockholm 
 Academy. Lichtenberg, in 1777, constructed an electrophorus of 
 gigantic dimensions, and of extraordinary power. His electro- 
 phorus plate was 6 Paris feet in diameter. It consisted of com- 
 mon resin, turpentine, and Burgundy pitch, melted together, and 
 poured whilst in a fluid state upon the surface of a sole (93), up to 
 the given thickness. When cold and solid, the surface was 
 smoothly polished by the usual mechanical processes. The cover 
 of Lichtenberg's electrophorus was well rounded at the edges, had 
 a diameter of 6 Paris feet, and was lifted off and applied to the 
 electrophorus plate by means of a suspension pulley. When this 
 electrophorus was freely excited and in action, sparks are said to 
 have been obtained from it 15 inches long, and of such density and 
 power as scarcely to be endured. 
 
 Several gigantic instruments of this kind were also constructed 
 by the Germans ; one, 9 feet in diameter, from which sparks are 
 said to have been obtained 13 inches in length, as thick as the 
 little finger. 
 
 96. We owe to Professor Phillips a convenient modification of 
 the compound electrophorus. His electrical plate is perforated in 
 three or four points, and short pieces of brass wire inserted in the 
 perforations, so as to be level with its opposite surfaces, and form 
 a communication between the cover and the sole, whenever the 
 cover is applied to the excited plate. The object of this construc- 
 tion is to communicate, by means of the knuckle or other con- 
 ductor, a supply spark to the insulating cover every time the cover 
 is raised off the surface. In this arrangement the cover is supplied 
 whilst in place through the conducting communication with the 
 sole. The cover may in this way be charged from 50 to 100 times 
 in a minute, merely setting it down and lifting it up as fast as the 
 operator can work. This instrument was about 20 inches in 
 diameter, the cover about 16 inches. It emitted flashing sparks 
 of electricity full 2 inches long. 
 
 97. Electrophorus Electrical Machine. Herr Holtz, of Berlin, 
 availing himself of the principle of the electrophorus, has applied 
 
90 
 
 it in the construction of an electrical machine acting not in the 
 usual way, by simple friction, but by mere induction, he being 
 under an impression not only that such machines were possible, 
 but that with a moderate expenditure of force they would exceed 
 ordinary machines in quantitative effects. 
 
 Holtz's induction machine consists of two circular glass discs 
 (Figs. 61, 62), one of which, A, Fig. 61, is about 16 inches in 
 diameter, and is mounted on an axis, a, accurately centered, after 
 the manner of an ordinary plate electrical machine. It is made 
 to revolve rapidly by means of a multiplying apparatus, at the 
 distance of about one-eighth of an inch from a stationary plate, B, 
 Fig. 62, of window : glass, about 2 inches larger in diameter. Both 
 
 Fig. 61. Movable Disc, Fig. 62. Fixed Disc. 
 
 discs are coated with shell-lac varnish. The fixed plate B has a large 
 central hole, to allow the axis a of the rotating plate, A, Fig. 61, 
 to pass freely through and travel clear of it, and admit of the 
 two plates being set at one-eighth of an inch distant from each 
 other, or at any other given distance apart. B is also furnished 
 with two openings or windows, o p, near which are attached two 
 surfaces of paper, m n, with a point from each projecting into the 
 plane parallel to the openings o p. 
 
 The principle upon which this machine works pre-supposes, as 
 in the case of the electrophorus, a certain amount of previously 
 existing electrical excitation in the rotating plate A ; which is 
 effected by any ordinary means, as by holding an excited glass 
 tube or a stick of resin near the paper points, o p. On setting 
 the rotating disc in motion, induced electricity is evolved from 
 beneath the paper coatings, m, n. Two insulated metal rods, or 
 conductors, furnished with collecting points, are set before the 
 rotating plate A, parallel to the axle a, Fig. 61. The points are in- 
 tended to take up the induced electricity as rapidly as it is evolved, 
 and give it off under the form of powerful current discharges. 
 
 Holtz's machine is mounted on a strong insulating rectangular 
 frame, with insulating pillars and cross bars (Fig. 63). The strik- 
 ing distance of a spark from this machine is said to vary from 
 
THE CONDENSER. 91 
 
 1 to 4 or 5 inches. From three to four discharges are reported 
 to be obtained in one second.* This machine is of ingenious and 
 scientific construction, but is somewhat precarious and tedious in 
 practice. It is necessary to clean it every time it is used, and 
 
 Fig. 63. Holtz's Electrical Machine. 
 
 to free the rotating disc from the dust which accumulates on it, in 
 consequence of the decomposition of the resinous varnish with 
 which the glass is coated. It is hence questionable whether an 
 ordinary plate machine of Cuthbertson's construction (74) is not 
 more simple and efficient. 
 
 98. The Condenser. Analogous to the electrophorus is the 
 electrical condenser, the object of which is to render sensible to 
 the electroscope minute quantities of electricity, which, without 
 its aid, would be inappreciable. 
 
 It has been already remarked, with reference to the compound 
 electrophorus, that the increased power derived from the sole 
 is referable to the inductive action between an insulated and an 
 uninsulated conductor (94). Such is, in fact, the principle of the 
 electrical condenser, a contrivance originating in the ingenious 
 experiments of Volta, which led to his memoir " On the advan- 
 tages of a kind of imperfect insulation." He remarks that the 
 Marquis Bellisoni, having by chance placed his charged electro- 
 phorus disc on a table covered with animal skin, and after a short 
 time raising it up, was astonished at obtaining a spark from it. 
 This fact he communicated to Yolta, who instituted some further 
 researches bearing on the subject, and eventually proposed the 
 following problems, which he entitled " Electrical Paradoxes." 
 
 * A detailed account of this instrument is given in the Philosophical 
 Magazine for December, 1865, p. 425. 
 
92 
 
 1. To cause any conductor to retain electricity whether insulated 
 perfectly or imperfectly, or not insulated at all. 
 
 2. To accumulate a much greater quantity of electricity on a 
 conductor imperfectly insulated, than can be effected by means of 
 the most perfect insulation. 
 
 3. To cause a metallic conductor of small dimensions to retain a 
 charge to a small extent, although repeatedly touched by the 
 finger or a metallic body for twenty or thirty seconds. 
 
 4. To avoid a total dissipation of an electrical charge imparted 
 to a conducting body, although the conducting body remains in 
 contact with a neutral conductor for a considerable time. 
 
 Yolta's experiments go to show that a conductor may better 
 acquire the power of retaining a charge when in contact with an 
 imperfectly insulated body than when perfectly insulated. Thus 
 a conducting plate, M, Fig. 64, suspended 
 by silk lines, and flatly applied to a marble 
 table, N, does not yield up all its electricity, 
 but will, after being removed from the 
 marble, still give a spark, even although 
 the contact with the marble surface be 
 rig. 64. voita's Paradox, continued for a considerable time. This is 
 what Yolta calls a paradox. The more extensive the contact, and 
 the better the surfaces apply to each other, the more perfect the 
 experiment. It is only when we touch the table with the edge of 
 the plate that the electricity leaves it. If the electrified plate be 
 insulated in free space in the air, all its electricity disappears in a 
 few minutes, whereas if you place it upon a dry piece of wood or 
 marble the electrical charge remains. Again, if the charged 
 plate M be placed upon a cake of brimstone or resin, still its 
 electricity leaves it, although these bodies may be considered as 
 electrical insulators (15) ; and the more perfectly the electrified 
 plate M is insulated, the more rapidly will its electricity disappear. 
 Yolta ma'kes a great difference between this and the action of the 
 electrophorus (92). The electrophorus disc is never in action 
 unless its electrical surface be first excited ; the excited electricity 
 in this case dissipates but slowly, but remains adherent on the 
 surface as if pasted on it. The case of the charged metallic disc M, 
 just alluded to, is different. If it be made to repose upon a free 
 metallic plate, or upon a plate of wet wood or moist marble, its 
 electricity is soon dissipated. 
 
 99. Yolta inferred from these results that the circumstances 
 under which a charged conductor can best preserve its charge is 
 not that of perfect insulation, but, on the contrary, a state of such 
 
THE ELECTRICAL CONDENSER. 93 
 
 imperfect insulation that we can scarcely consider it as being 
 insulated at all. Yolta further observed that imperfect insulation 
 enables a conductor to absorb or take up electricity more readily 
 than perfect insulation. The effect of a semi-insulating plate 
 upon a metallic plate is to restrain any free electricity communi- 
 cated to it, and so, by diminishing its intensity (62), enable it to 
 receive under a given electrometer indication a much greater 
 quantity. On separating the metallic plate from its semi-insu- 
 lating base, the increased charge is immediately apparent (49). 
 
 100. The condenser, therefore, consists simply of a semi-con- 
 ducting disc, M, Fig. 64, reposing on a semi-insulating plate, N, 
 the disc M having an insulating handle, or be suspended by insu- 
 lating silk lines. If now a very weakly electrified body, charged 
 with an infinitesimal quantity of electricity, 'quite inappreciable 
 by the most delicate electroscope, touch the insulated metallic 
 plate M, the imperfect insulating plate N will enable the metallic 
 plate M so to absorb every particle of electricity in the weakly 
 charged body, as to cause the metallic plate to abstract the whole 
 of it, or nearly so ; which otherwise, without such influence, it 
 could not do, some portion must still always remain. The 
 semi-insulating plate, in fact, increases by induction (41) the elec- 
 trical capacity of the metallic plate, by which any small amount 
 of electricity imparted to it is at the instant rendered, as it 
 were, insensible to the electroscope. Directly, however, the con- 
 ducting disc M is raised in an insulated state off the surface of the 
 semi-insulated base on which it rests, the small quantity of charge 
 it had absorbed whilst in contact with the semi-insulating sur- 
 face N becomes (in accordance with what has been already stated, 
 98) immediately sensible. Such is the operation of the electrical 
 condenser. 
 
 101. Yolta gives a list of semi-insulating bodies best adapted to 
 the purpose of a condenser. He prefers marble. But all kinds of 
 marble do not succeed equally well. He thinks Carrara marble 
 the most efficient. Some varieties of alabaster also answer the 
 purpose as semi-insulators. All these substances, however, must 
 be very clean and dry, not only on the surface, but within the 
 substance. Yolta tried ivory and other kinds of bone, but not 
 with the same success. He found nothing better than plates of 
 wood previously steeped in oil and well dried by heat. A table 
 covered with white wax, velvet and silk stuff stretched on a 
 table, were found efficient as semi-insulators. 
 
 102. The electrical condenser, as more recently constructed, 
 dispenses with a semi-insulating substance altogether, and consists 
 
94 
 
 THE CONDENSER. 
 
 of two metallic plates, P Q, Fig. 65, one of them, p, being insulated 
 on a light rod of glass, r, the other, Q, uninsulated, it being fixed 
 on a brass rod, t, supported on a sliding piece, v. The two plates 
 are set quite parallel to each other, and are brought, by means of 
 the slide v, as near together as may be without touching. The un- 
 insulated plate Q supersedes, in this way, the necessity of a marble 
 or other semi-insulator, there being a small stratum of air inter- 
 posed between the two plates. When we desire to detect the 
 presence of an infinitesimal quantity of electricity in a given body, 
 contact is made with the weakly charged body and with the insulated 
 plate P, through a small projecting ball jt? ; the minute quantity of 
 electricity in the weakly charged body becomes immediately 
 absorbed and condensed by the influence of the approximated 
 uninsulated plate Q. The plate Q is now withdrawn by means 
 of the slide v, at a distance from p ; the condensed electricity in p 
 then becomes immediately sensible to a delicate electroscope. The 
 rod supporting the uninsulated plate Q has occasionally a joint 
 and stop, w, at its lower extremity, which admits of the plate Q 
 being turned back from p without moving the slide v. This may 
 be termed the single condenser. If the insulated plate P, Fig. 65, 
 
 Fig. 65. Single Condenser. 
 
 Fig. 66. Double Condenser. 
 
 be placed between two uninsulated plates, the plates being separated 
 by an extremely small interval of air or other dielectric, the 
 condensing power is greatly increased. This may be called the 
 double condenser, and is shown in Fig. 66. 
 
 103. Cavallo extended the condensing power still further by 
 transferring the condensed electricity in a first plate to the 
 insulated plate of a second condenser. The insulated plate of the 
 second condenser is generally attached to the cap of the double 
 gold-leaf electroscope (33). The uninsulated plate opposed to the 
 plate on the electroscope is fixed on a light brass rod, having a 
 joint and stop attached to a slider connected with the foot on which 
 the whole is supported, as seen in Fig. 67. The diameter of the 
 condensing plates in an instrument of large size may vary from 5 
 
THE CONDENSER. 95 
 
 to 12 inches : the auxiliary condenser attached to the cap of 
 the gold-leaf electroscope need not exceed 2 inches. 
 
 104. Many eminent electricians of past days, not content with 
 this increase of condensing power, endeavoured to arrive at still 
 more delicate arrangements, which, although evincing great 
 ingenuity and wonderful sen- 
 sibility, tend to produce spon- 
 taneously the electricity they 
 
 were designed to detect, and 
 are hence liable to give rise to 
 equivocal results. "We may, 
 for example, refer to Bennett's 
 Doubler of electricity, Cavallo's 
 
 -r TI- T TVT'ii > o Fig. 67. Cavallo's Condenser. 
 
 Multiplier, Is icholson s Spin- 
 ning Condenser, Wilson's Double Multiplier, and such like.* 
 
 105. A simple and efficient condenser may be easily constructed 
 by dropping three small spots of sealing-wax on the lower face of 
 an electrophorus cover (93), by which it may be supported at a 
 short distance above the even surface of an ordinary table. We 
 have in this arrangement a metallic plate under the influence of a 
 semi-insulator. If now a weakly- charged body, insensible to a 
 delicate electroscope, be made to touch the metallic plate, and the 
 plate be then raised off the table by its insulating handle, the 
 abstracted electricity immediately becomes sensible directly the 
 plate is applied to the gold-leaf electroscope. Bennett's doubler 
 is said to multiply an infinitesimal quantity of electricity 500,000 
 times. It is evident, however, that no instrument of this descrip- 
 tion can actually multiply the electrical agency. Its power is 
 necessarily limited to rendering sensible to the electroscope a small 
 quantity of electricity actually existing, which, under ordinary 
 circumstances, is inappreciable. 
 
 106. Although the condenser may be considered as an instrument 
 more especially adapted to the rendering very small quantities of 
 electricity sensible to the electroscope otherwise inappreciable, yet 
 the converse of this is not by any means apparent ; it does not 
 follow that the condenser renders quantities of electricity, sensibly 
 affecting the electroscope; insensible. It can only diminish the 
 intensity to a certain point. The condensing or inductive power 
 of the semi-insulated plate has a limit, so that the electroscope in 
 
 * An account of these instruments will be found in the 77th and 78th 
 volumes of the Phil. Trans., for 1787 and 1788; Cavallo's "Treatise on 
 Electricity;" "Nicholson's Journal," 4to., vol. i. p. 16; and in the article 
 ELECTRICITY, by Sir David Brewster, in the eighth edition of the "Encyclo- 
 paedia Britannica." 
 
96 THE COATED PANE. 
 
 communication with an insulated electrified conductor, will always 
 have a certain degree of divergence, the condensing induction of 
 the semi-insulated plate decreases in some inverse ratio of the 
 electricity communicated to the insulated plate, as is, in fact, 
 found by experiment (41). 
 
 107. The Electrical Jar, or Ley den Phial. If the plates of the 
 condenser (102), instead of being of sensible thickness, and brought 
 extremely near each other, separated by a thin stratum of air, 
 were to consist of thin metallic leaf, such as tinfoil, and be 
 separated by some solid dielectric, such as glass, we then arrive 
 at an electric combination productive of marvellous effects. For 
 example, let A B, Fig. 68, be a square of common window glass ; 
 
 c d, e f two squares of tin leaf of 
 less size than the glass, smoothly 
 attached to the opposite surfaces of 
 the glass by thin strong paste, leav- 
 ing a wide insulating margin e df 
 all round in order to prevent the 
 opposite edges of the tin leaf sur- 
 faces from touching each other. 
 
 Fig. 68. Coated Pane. ^ Q ^aye tnen wnat nas been 
 
 termed a coated pane ; the opposed metallic surfaces c d, e f being 
 termed the coatings. With a view of rendering the insulation of 
 the edges of the two opposed coatings c d, 0/more complete, the 
 uncoated interval or margin of glass, e d f, separating them, is 
 covered with a thin varnish of shell-lac dissolved in naphtha and 
 dried off by heat. If this coated pane be placed upon a con- 
 ducting stand, s, one of the coated surfaces, ef, being connected 
 with the ground, we have virtually another form of condenser, 
 the upper plate c d being the insulated or accumulating 
 plate, e /the condensing, or what may be termed the semi-insu- 
 lating, plate. We may, therefore, in a similar way accumu- 
 late or condense upon the insulated surface c d a considerable 
 quantity of electricity not possible to so accumulate without the 
 aid of the lower plate e f. In this arrangement, as is evident, we 
 have an insulated conductor, c d } opposed to a semi-insulated con- 
 ductor, ef. 
 
 108. Thus circumstanced, if we communicate to the upper 
 insulated surface c d a, given quantity of electricity, while at the 
 same time a Henley's electrometer, H, is standing on the insulated 
 plate c d, the ball of the electrometer will be repelled to a given 
 distance by the uncondensed electricity on the charged plate, and 
 the twO plates, c d, e f, will exhibit opposite electrical states or 
 
THE COATED PANE. 97 
 
 forces ; that is to say, if the upper surface or insulated conductor 
 c d be positive, the opposed surface or uninsulated conductor ef 
 will be negative when freed from the influence of the positive 
 coating, by which its negative electricity is held as it were 
 in abeyance, much in the same way as the cover of the excited 
 electrophorus emits an electrical spark, directly it is raised off 
 the surface of the excited resinous plate (92), being then set free 
 from the inductive action of the excited electric surface. 
 
 Exp. 35. Let A B, Fig. 69, represent a coated circular plate of 
 glass of about a foot or more in diameter ; p N the two coatings, 
 the whole being mounted on a convenient supporting foot s, as 
 shown in the figure, p, n are two delicate reed electro- 
 scopes, freely suspended along the opposite coatings p, N. 
 
 If, in this arrangement, we connect one of the coatings 
 N with the ground, by means of a metallic wire, we have, 
 as in the case of the coated pane (107), an insulated con- 
 ductor p, opposed to an uninsulated conductor N ; and all 
 the elements of the condenser (100) realised in a vertical 
 position. Communicate to the insulated coating p a series 
 of electrical sparks, suppose positive ; the electroscope p F> 69 
 of the charged side P will diverge, whilst the electro- coated Pane. 
 scope n, in connection with the uninsulated surface N, will appeal 
 neutral, although the two coatings P, N are really in opposite elec- 
 trical states ; the coating N being actually negative, its negative 
 electricity, however, is, as just explained, held in abeyance by the 
 uncondensed positive electricity on the opposite surface p, operating 
 through the glass ; hence the negative state is not sensible to the 
 electroscope n. Remove now the connection of the coating N with 
 the ground, and we have the whole system insulated. Withdraw 
 from the surface P of accumulation a portion of its charge (which 
 is easily done by taking a spark from it by means of the knuckle), 
 the electroscope p immediately falls back to the surface P, whilst 
 the electroscope n diverges with the opposite or negative elec- 
 tricity. If now in a similar way we withdraw a negative spark 
 from the coating N, we in like manner liberate positive electricity 
 from the surface P, previously held in abeyance by the inductive 
 influence of the negative electricity of the coating N, so that 
 the electroscope n, now divergent, falls back in its turn, the elec- 
 troscope p again rising, and thus, by withdrawing a spark, first 
 from one coating, and then from the opposite coating (which can 
 always be done so long as the system remains insulated), the 
 electroscopes p n may be kept in alternate reciprocal movement. 
 
 Hence, the two coatings p, N are really in opposite electrical 
 
 H 
 
98 THE COATED PANE. 
 
 states, and under the influence of each other. The coated pane 
 A B, thus electrically effected, is said to be charged. 
 
 109. The principles just adverted to involve the theory of free 
 and condensed, or simulated and dissimulated electricity (38), as set 
 forth by the French philosophers, Faraday calls this theory in 
 question as conveying erroneous views, if by the terms free and 
 condensed electricity we mean to imply any difference in the 
 electrical agency itself, the one not being really more free or 
 condensed than the other. But although Faraday's objections 
 are just, and consistent with the nature and operation of electrical 
 force, yet the employment of the terms free and concealed elec- 
 tricity are not only admissible, but convenient, when limited 
 by definition, and their meaning and application clearly appre- 
 hended. We may, for example, in the case of the condenser, 
 or the coated pane just alluded to (108), fairly term that portion 
 of the accumulation not sensible to the electroscope concealed 
 electricity, or electricite dissimulee, and that portion operating 
 more immediately on the electroscope, not so concealed or com- 
 pensated, as it were, free electricity. It is certainly allowable to 
 draw a distinction between that portion of a charge not sensible 
 to the electroscope, and that portion by which the indications of 
 the electroscope are more especially affected. 
 
 110. If the electricity communicated to the accumulating surface 
 (107) (which may be either of the coatings), the opposite coating 
 being uninsulated, be positive, the pane is said to be charged 
 positively; if negative, the pane is said to be charged negatively. 
 If the opposed metallic coatings of the pane, being charged 
 either positively or negatively, be joined by a bent conducting- 
 wire, the opposite electricities unite with a loud brilliant 
 explosion, resembling in a minor degree the phenomenon of 
 thunder and lightning. The pane is now said to be discharged, 
 and has been termed by the French philosophers, in consequence 
 of this explosive effect, a fulminating square. If we bring the 
 finger of one hand in contact with the uninsulated coating, and 
 a finger of the other hand in contact with the insulated coating of 
 accumulation, a painful sensation through the arms and across 
 the breast is experienced, termed the electric shock. 
 
 111. We may conclude, from the operation of the coated pane 
 (108), that in charging or discharging a coated surface, the two 
 opposite electricities or forces are so related, that as much of 
 either force as we add to or subtract from the one coating, we 
 subtract from or add to the opposite coating, an important ele- 
 mentary fact, first announced by Franklin. 
 
THE COATED PANE. 99 
 
 Exp. 36. Let a coated circular pane, Q, Fig. 70, be mounted on con- 
 venient supports, and a stand, s, between two small metallic ballsy n, 
 placed at short and equal distances from each, coating on either side 
 of it. Let the ball p communicate with the positive conductor p of 
 the electrical machine, and the ball n with the ground by means of 
 a wire or chain. Let the machine be now gently set in motion until 
 the ball p sends off a spark to the adjacent coating. The opposite 
 coating will at the same instant send off a spark to the ball n, 
 and this reciprocal action will continue until the pane Q be fully 
 charged (108) ; that is to say, 
 until the positive electricity 
 emitted by the ball p has been 
 held in abeyance by the opposite 
 electricity of the ball n. 
 
 We see by this experiment that 
 in charging a coated electric, as 
 much of the one electricity as we 
 communicate to one of the coat- Fi - 7U - Coated P{ine - 
 
 ings, we abstract from the opposite coating, and reciprocally. 
 Suppose, for example, that twenty electrical sparks had been 
 received upon the one coating from the ball p, then twenty equal 
 and similar sparks would be given off by the opposite coating to the 
 ball n, so that the coating opposed to the ball p will be charged 
 positively with twenty sparks, and the coating opposed to the ball n 
 would be charged negatively with twenty sparks. When the 
 two opposed coatings p, n are united by a bent conducting wire in 
 the way just mentioned (110), the whole twenty measures received 
 from the ball p, rush in a dense state of accumulation through 
 the connecting wire upon the coating n from which twenty 
 measures have been abstracted, giving rise to the fulminating 
 effect and electrical shock already adverted to (110). The coated 
 dielectric pane being but another form of condenser (107), we 
 see the necessity of one coating being uninsulated, or in con- 
 nection with the earth, without which the system cannot be 
 what we have termed charged, or in other words, cannot be 
 brought under the dominion of electrical induction. We have, 
 in fact, seen (41) that in opposing two conductors to each other 
 the inductive effect is the greatest possible when one of the con- 
 ductors is uninsulated or free. In taking alternate sparks from 
 either coating, as in the former experiment, we eventually effect 
 discharge also in successive and alternate portions. 
 
 112. The form and extent of the dielectric glass separating 
 the coatings being of little or no consequence to the experiment, 
 
 H2 
 
100 
 
 THE LEYDEN JAR. 
 
 it has been found convenient, instead of a thin flat plate of glass, 
 to employ an ordinary glass jar, coated on its inner and outer 
 surfaces with tinfoil to within a certain distance of its mouth. Such 
 an arrangement has heen termed the electrical jar or Leyden phial, 
 so called from its marvellous electric shock (110), first accidentally 
 noticed by some Dutch philosophers at Leyden, in Holland, whilst 
 endeavouring to electrify water enclosed in a common phial ; a 
 discovery which caused much astonishment throughout Europe. 
 
 113. The most approved form and dimensions of the electrical 
 jar are represented, Fig. 71, m b represent a jar of clear glass of 
 almost any shape or size ; it is, however, generally a hollow cylinder. 
 Its height from the base b, to the mouth m, should be about 
 20 inches ; diameter, a d, one-half its height 10 inches ; width 
 of mouth, m, one-quarter its height -5 inches ; height to shoulder, 
 s, about a diameter and three-fifths 16 inches; thickness of 
 glass, something less than the one- tenth of an inch. The jar 
 should be rounded and somewhat compressed at the shoulders, 
 rising gradually in a flattened curve to 
 the mouth. It is coated internally and 
 externally with tin-foil up to the shoulder 
 s, the coatings being attached to the 
 glass by strong thin paste, and smoothly 
 rubbed down upon its surface. The jar when 
 coated generally exposes about 4 square 
 feet of coating. The insulating, or, as it 
 is termed, the uncoated interval, q r, 
 separating the edges of the internal and 
 external coatings, is varnished with a 
 filtered solution of shell-lac dissolved in 
 naphtha. 
 
 114. Electricity is communicated to the 
 jar from the electrical machine or other 
 source of electric accumulation, by means 
 of a light metallic tube, E F, termed the charging rod, passing 
 through the mouth of the jar to its internal coating. As it is 
 often desirable, especially in employing single jars, to vary the 
 length and extent of the charging rod, it is convenient to have 
 the rod in two separate portions, E p, p r, which consist of two 
 light drawn metallic tubes, sliding with friction one within the 
 other ; one of these, p F, being securely held in a wooden foot, F, 
 covered with tinfoil, is a fixture within the jar. The foot F is 
 strongly glued to the internal coating at the base of the jar. This 
 fixed portion of the charging rod passes completely through the foot, 
 
 Fig. 71. Leyden Jar. 
 
THE LEYDEN JAR. 101 
 
 so as to touch the tinfoil beneath, and extends from the foot to a 
 little above the height of the internal coating ; where it is further 
 steadied by passing with friction through a similar inverted foot, p. 
 This upper support, p, is furnished with four light cross-arms, 
 t, u, v, w, covered with tin-leaf, inserted in four separate parts, in 
 corresponding grooves, cut for their reception in the upper surface 
 of the inverted foot p, which allows of their being easily placed in 
 position through the mouth of the jar, and extended within it 
 so as to touch the internal coating at opposite points, t, u, v, w, of 
 its circumference. The cross-arms of support are secured by means 
 of long slots cut in them, and small metal screws passing through 
 the slots into the grooves beneath. We are thus enabled to give 
 the cross-arms any required amount of lateral extension, and 
 pressure against the internal coating of the jar. The remaining 
 portion p E of the charging-rod slides freely within the fixed 
 portion, so as to project beyond the mouth of the jar to any 
 required extent, its upper extremity terminating in a charging 
 ball E. Several fine holes are drilled in the projecting portion of 
 the charging-rod for attaching light and finely pointed tubular rods. 
 
 The communicated electricity condenses on the inner coating 
 by the inductive influence of the external coating operating 
 through the intervening glass (108) ; with this view, and in order 
 to satisfy the conditions of the semi-insulated plate of the con- 
 denser, the outer coating must be free, the resulting pheno- 
 mena being precisely the same. In charging a coated jar, as 
 the charge begins to accumulate, it is made evident by means 
 of a Henley's electrometer, h, usually fixed on the ball E at the 
 extremity of the charging-rod. As in the case of the revolutions 
 of the electrical machine (86), there is a limit to the quantity of 
 electricity which the jar can receive. On approaching the maxi- 
 mum or limit of charge, the electrical sparks from the prime 
 conductor begin to subside, and the uncondensed or accumulated 
 electricity (86) flies off in luminous coruscations, and a bright 
 electrical discharge frequently ensues between the coatings over the 
 intervening glass, or uncoated interval. This spark constitutes the 
 phenomenon known as spontaneous discharge. In all experimental 
 inquiries with the electrical jar, therefore, it is requisite, as in the 
 case of the revolutions of the electrical plate, to determine the 
 limit of charge (86) of which the jar is susceptible, and not to 
 press the accumulation beyond that limit, 
 
 115. If the charging-rod communicate with the external coat- 
 ing by means of a bent wire, w, Fig. 71, an explosion takes place, 
 and the jar is said to be discharged. This bent wire w, or other 
 
102 THE LEYDEN JAR. 
 
 conducting communication between the coatings is termed a 
 circuit. When mounted on a glass handle, N, with a joint in the 
 centre, and terminating in small balls, p, n, the circuit is called 
 a discharging rod. The glass handle N shields the hand from any 
 shock incidental to the discharge, whilst the joint at w admits of 
 the circuit being extended to any convenient distance between 
 the discharging balls. 
 
 116. Such is the tendency of the two electricities developed on 
 the opposite coatings to unite, that absolute contact between the 
 balls of the discharging rod and the opposed coatings is not 
 requisite for effecting the discharge of the jar ; it is enough if 
 one ball of the discharging rod, usually the negative, first touch 
 the negative coating, whilst the other ball be brought near the 
 charging rod of the jar. In such case a brilliant explosive spark 
 breaks through the interval separating the points of discharge, 
 and the opposite electricities immediately unite with a brilliant 
 explosion (110). 
 
 The jar being charged, its inner and outer coatings, as in 
 the case of the fulminating square (108), are in opposite 
 electrical states ; that is to say, if one exhibit positive, the other 
 will exhibit negative electricity. Both states may be made evident 
 by placing the jar on an insulating base, in which case we may 
 demonstrate experimentally, as in the case of the coated pane (111), 
 that as much of either force as is accumulated on, or abstracted 
 from the one coating, is abstracted from, or accumulated on the 
 opposite coating. For example, let each coating in a neutral state of 
 the jar be supposed to contain a quantity of electricity represented 
 by the number 20, that is to say, let each coating be supposed to 
 contain a quantity represented by the number 10. When the jar is 
 fully charged, the whole quantity, 20, is accumulated on one of the 
 coatings ; 10 being taken from one coating and added to the other, 
 when the jar discharges through any circuit, the electrical 
 equality between the coatings again obtains, and the quantity 
 10, taken from the one side, rushes through the discharging circuit 
 (115), and restores the original distribution. The circuit which 
 enables this phenomenon to occur may be of unlimited extent, 
 and may consist of any given kind of substance. If there be 
 many such circuits of discharge, the jar in the act of discharging 
 will affect them all. On applying one of the balls of the dis- 
 charging-rod first to the outer coating, suppose negative, and 
 then bringing the opposite ball towards the ball E of the 
 charging-rod, suppose positive, the explosive spark of the dis- 
 charge takes place before the two balls, as already observed (116), 
 
THE ELECTRICAL BATTERY. 103 
 
 actually touch ; a phenomenon arising from the attractive force 
 between the opposite electricities, and their tendency to com- 
 bine (27). Hence if two conducting balls, one connected with 
 the charging-rod, or inner coating of the jar, the other con- 
 nected with the outer coating, be directly opposed to each other 
 at a given distance, an explosive discharge will ensue between 
 the balls directly the electrical accumulation on the positive 
 coating of the jar has reached a given point, and thus a measure 
 of the charge may be obtained. 
 
 117. When a number of electrical jars are joined so as to 
 op3rate as one whole, we have an electrical battery, the construc- 
 tion of which has been usually of a complicated and expensive 
 character, and not always effective. The simplest method of form- 
 ing an electrical battery is to group about a central jar, c, Fig. 72, 
 upon a square or circular base, A B, 
 
 covered with tinfoil, a given number of 
 
 coated jars. All the charging-rods of 
 
 the group are to be united by means of 
 
 light conducting wires radiating from 
 
 the charging-rod of the central jar, so 
 
 that the whole group may be charged 
 
 and discharged from the central rod as 
 
 a single jar. We may in this way, by 
 
 grouping together nine jars, numbered 
 
 from 1 to 9 (Fig. 72), easily obtain a 
 
 battery exposing 36 square feet of coated 
 
 glass, within a circular space the diameter 
 
 of which is 30 inches, each jar being 
 
 similar to the jar already referred to Kg. 72. Electrical Battery. 
 
 (113), and exposing 4 square feet of coated glass. By connecting 
 
 together two or more such groups, a battery of almost any extent 
 
 may be constructed. The great battery in Teyler's Museum at 
 
 Haarlem consisted of 135 jars, each jar exposing something less 
 
 than 1 square foot of coating. The explosion from this battery 
 
 melted an iron wire 25 feet long and -2 1 oth of an inch in diameter. 
 
 118. In constructing a battery in the way just described (117), 
 the conductors connecting the charging-rods of the jars should con- 
 sist of small brass tube, about one-eighth of an inch in diameter, 
 with solid, pointed, sliding wires, projecting from each extremity 
 of the tube, so as to allow of the rod being elongated or contracted 
 to a given distance ; small holes are drilled in the charging-rods 
 for the reception of the pointed connecting- wires. The base A B 
 on which the jars rest is supported on varnished glass columns. 
 
104 
 
 EXPLANATORY NOTES. 
 
 The whole battery, therefore, may be insulated, if required, on 
 withdrawing the conducting communication with the earth. It 
 is evident that by connecting, by means of the discharging-rod 
 (115), the central ball c, in which all the charging-rods terminate, 
 with the tinfoil base on which all the jars rest, we may discharge 
 the whole system as a single jar (Q). 
 
 OCCASIONAL MEMOBANDA AND EXPLANATORY NOTES. 
 
 ( 0) The cylinder machine may be of any given dimensions, according to 
 the power required. A cylinder 10 inches in diameter and 12 inches in 
 length between the shoulders, gives considerable power in a machine. A 
 
 cylinder 1 foot in diameter and 13 
 inches in length between the shoulders, 
 is extremely powerful. A cylinder of 
 7 inches in diameter and 9 inches in 
 length, with a rubber of 8 inches, is 
 very efficient. 
 
 Fig. 73 represents the cylindrical 
 machine as constructed by Franklin, 
 the cylinder P being about 7 inches in 
 diameter and 9 inches in length between 
 the shoulders. Franklin's machine was 
 set up in a firm frame, ff f , and was 
 turned by a wheel and band, as shown 
 in the figure. 
 
 (P) In 1840, an intelligent workman 
 in charge of a steam-engine at Sighill, 
 near Newcastle, noticed a considerable 
 escape of steam, arising from a leak in 
 the cement about the safety-valve. The 
 engine man being about to adjust the 
 weight of the valve, was surprised by 
 the emission of a powerful spark of 
 electricity, which he found to proceed 
 from the metal work connected with 
 the boiler; as also from the boiler, if he attempted to touch it during 
 the escape of steam, especially if one of his hands were immersed in the 
 vapour. Mr. Armstrong, a scientific gentleman at Newcastle (now Sir 
 William Armstrong), having been informed of this result, lost no time in 
 investigating this new phenomenon. 
 
 By means of an insulated brass rod, with a metallic plate at one extremity, 
 and a ball at the other the latter placed near the boiler, and the plate in the 
 issuing vapour he obtained sixty or seventy sparks per minute. After a 
 series of interesting inquiries, he succeeded in constructing a new form of 
 electrical machine (87), depending on the excitation of particles of water 
 driven by steam through small orifices. 
 
 (Q} The largest jars which the glass-blowers of former days could con- 
 veniently make, were 17 inches in height, and were seldom more than 4 
 
 Fig. 73. Franklin's Electrical Machine. 
 
THE ELECTRICAL BATTERY. 
 
 105 
 
 inches in diameter. They were thin and very liable to fracture from 
 spontaneous explosion (114). An electrical battery of the old construction 
 was a hazardous and expensive piece of apparatus. It usually consisted 
 of a large number of jars of variable dimensions and thickness, arranged 
 within a rectangular box, Fig. 74, lined with tinfoil. Each jar had a 
 
 Fig. 74. Electrical Battery. 
 
 varnished, wooden cover, with a brass charging-rod, connected by a short 
 brass chain with the internal coating of the jar* The charging rods 
 were all connected together, as represented in Fig. 74, by cross rods, 
 nicely polished and lacquered. From the great number of connecting 
 rods and balls, &c., &c., employed in the construction of an electrical 
 battery of the olden time, this piece of electrical apparatus was neces- 
 sarily very complicated and costly. The battery employed by Dr. 
 Priestley consisted of sixty-four jars, each jar 10 inches in height, about 
 3 inches in diameter, and coated to within If inch of the top, exposing 72 
 square inches, or half a square foot each of coating. The first battery con- 
 structed by him consisted of forty-one jars, each 17 inches in height and 
 3 inches in diameter, coated to within 2 inches of the mouth, each jar ex- 
 posing a square foot of coating. The whole battery, therefore, exposed 41 
 square feet, an amount of coated surface which could be very well grouped 
 under an arrangement of 10 jars, in the manner described (117). The jars 
 thus arranged, and grouped together in a box, were very liable to fracture in 
 consequence of spontaneous explosion, by which the whole battery was ren- 
 dered useless, and consequently involved a costly reconstruction. The simple 
 method described ( 1 17) is not only efficient but economical, and easily managed, 
 involving no complication whatever. It is evident, as remarked (118), that 
 by connecting all the charging rods of the jars with the tinfoil lining of the 
 box, Fig. 74, by means of the discharging rod, the battery will be at once 
 discharged as a whole through any given circuit. It is clear that in order 
 to discharge any battery we only require to unite the outer and the inner 
 coatings by a metallic communication. 
 
 Let, for example, N, Fig. 74, be a projecting rod or wire passing through 
 the box, so as to touch the inner tinfoil lining upon which each jar rests. 
 If we connect one of the charging balls, by means of the discharging rod, 
 with the projecting wire N proceeding from the tinfoil lining of the battery, 
 the whole system discharges in the manner just observed (118). 
 
 > It A It 
 
 O 
 
 i , 
 
CHAPTER IY. 
 
 INSTRUMENTS OF QUANTITATIVE MEASURE, MORE ESPECIALLY ADAPTED 
 TO ELECTRO-MATHEMATICAL AND PHYSICAL RESEARCH. 
 
 119. The Unit Jar. It was one of Franklin's great facts that 
 in charging the electrical jar, as much of the one electricity 
 as is caused to accumulate on the one coating, the same quantity 
 of the other electricity is developed on the opposite coating (111). 
 
 In accordance with this great principle; the author of this 
 work proposed to interpose between the prime conductor and the 
 battery or jar to be charged, a small insulated coated jar, exposing 
 a few square inches only of surface ; connecting the inner coat- 
 ing of the small jar with the prime conductor of the machine, and 
 the outer coating with the jar or battery to be charged, or recipro- 
 cally, so as to charge the battery with the electricity given off 
 from the opposite coating of the interposed jar. When the small 
 jar becomes charged to a given point or height, an explosion or dis- 
 charge ensues between the two coatings of the interposed jar, by 
 means of two small exploding balls, and marks the reception by the 
 battery of one charge or unit of the small jar. Hence this piece 
 of electrical apparatus has been termed a unit measure, each of its 
 explosive discharges being considered as a unit of charge of a 
 given magnitude, taken in terms of the distance between the 
 two small exploding balls of discharge^ 
 
 120. The unit jar or measure, Fig. 75, is a small electrical 
 jar about 4 inches long, six- tenths of an inch in diameter, and 
 
 one-twelfth of an inch thick, ex- 
 posing about 5 square inches of 
 coated surface, either under the 
 form (113), or straight and open- 
 mouthed. The unit jar is con- 
 Fig. 75. unit Jar. structed of thin-drawn glass 
 tube, hermetically sealed at one end, varnished both internally 
 and externally with a thin solution of shell-lac. This unit 
 measure is carefully prepared ; the coatings are well rubbed 
 down upon the surface of the glass, and silvered over. The 
 
THE UNIT JAR. 
 
 107 
 
 charging-rod c C' is a small silvered metallic tube, one-fifth 
 of an inch in diameter, and consists of two portions, movable 
 one within the other, in the manner already described (114). 
 One of these, c, is a fixture, within the jar, and projects for 
 about an inch beyond its mouth. The other portion c' slides 
 freely within this fixed portion so as to admit of extension 
 or contraction to any required amount. The fixed portion 
 carries upon a sliding piece, s, a small exploding ball, p, about 
 three-tenths of an inch in diameter ; an opposite similar exploding 
 ball, n, is fixed on the extremity of a sliding tube move- 
 able within a tubular piece, t r, secured by light metallic bands, 
 m- y b, to the outer coating. . These exploding balls, p, n, can, 
 by means of the sliders, be set at any given distance apart 
 as determined by accurate measures or gauges, or a delicate 
 regulating screw, affixed to the tubular piece t r, and acting on 
 the ball n. The sliding tube of the ball n, where it emerges from 
 the tube t r, is further graduated into divisions one-twentieth 
 of an inch apart. The base of the jar is capped externally with 
 a brass band, b, terminating in a small ball, a. The whole is 
 sustained upon a ball and pin attached to the metallic band m, 
 so as to admit of turning the jar into any convenient direction. 
 
 121. The unit measure is principally employed for measur- 
 ing the quantity of electricity accumulated on coated surfaces, 
 as in charging the electrical battery, in which case it is 
 either mounted on an inde- 
 pendent insulator, and inter- 
 posed between the machine and 
 
 battery to be charged, Fig. 76, 
 or is directly sustained on the 
 prime conductor. 
 
 In charging a given body, as 
 the charge proceeds, successive 
 explosions between the neutra- 
 lising balls of the unit jar mark 
 with great precision the number 
 of measures in the battery. The 
 precise dimensions of the unit jar will of course depend on the 
 requirements of the experimentalist. 
 
 122. A small jar exposing about 5 square inches of coated 
 surface will, for general purposes, be found sufficient, although 
 larger units may be occasionally required. If employed to denote 
 small quantities of electricity, unit measures of very small dimen- 
 sions may be employed, which may be constructed of thin coated 
 
 Fig. 76. Position of Unit Jar in charging 
 Leyden Jar. 
 
108 
 
 THE UNIT JAR. 
 
 glass tube exposing about half a square inch of surface, or of small 
 thin coated glass plates. 
 
 123. Objections have been made to the unit measure similar 
 to those advanced relative to the uniform action of the electrical 
 machine (86), but they are equally inapplicable. It has been 
 urged, for example, that the unit measure does not communicate 
 the same quantity of electricity at each discharge, in consequence 
 of the accumulation in the battery operating upon its outer coat- 
 ing, which interferes with the freedom of accumulation. Such, 
 however, is not the case in the practical application of the 
 instrument. We have already seen (86) that no circumscribed in- 
 sulated conductor can receive and maintain an unlimited quantity 
 of charge. The limit of accumulative power being determined, 
 every unit discharge within that limit communicates an equal 
 quantity to the jar or battery with which it is connected. This 
 fact is made quite evident by a simple and plain experiment. 
 
 Exp. 37. Let the jar A, Fig. 77, be fitted with two exploding 
 balls, p, N ; the one, p, fixed on the extremity of a light sliding 
 
 Fig. 77. Self-Discharging Leyden Jar and Unit Jar. 
 
 rod, P q, passing through the charging-rod of the jar ; the other, 
 N, fixed at the extremity of a sliding rod, N o, passing through 
 a slide, s, projecting from a ring on the outer coating. Let 
 the balls be placed and adjusted at successive and equal dis- 
 tances, suppose !, -2, -3, &c., of an inch apart. Connect the jar A 
 with an interposed unit measure, u, in the ordinary way. The 
 electrical machine being set in motion, note the number of revolu- 
 tions, and the number of unit explosions required for discharge of 
 the jar at distance !, the number required for explosion at distance 
 2, and so on. The unit explosions, or the revolutions of the 
 machine (86), are found to be as the distances between the discharg- 
 ing balls P, N, which could not possibly be the case if the unit 
 measure had not been uniform in its action up to the distance three- 
 tenths at least. Let the distance of the discharging-balls P, N be 
 now increased up to a point of distance at which discharge between 
 
LANE'S ELECTROMETER. 109 
 
 the balls p, N becomes altogether impeded, or ceases to ensue, in 
 consequence of the jar having reached its limit of accumulation ; 
 this immediately determines how far the unit measure may be relied 
 on as a quantitative accumulator. 
 
 124. Faraday, who examined this question, arrived at the con- 
 clusion that the unit measure is a true indicator of quantity. He 
 remarks : " Suppose the unit jar has one- tenth the capacity 
 of the large jar or battery to be charged ; and that being charged 
 up to its exploding point, it contains ten particles of positive 
 electricity ; these ten particles will have then passed into the 
 large jar or battery as a unit accumulation, and none will remain 
 in the unit jar. At this moment the conductor of the machine, 
 the outside of the unit jar, and the charging- rod .and ball of the 
 large jar or battery, will be positive to a carrier ball (14). Now 
 although on continuing to turn the machine, the positive state 
 of all these surfaces increases, still the mutual relation of the two 
 exploding balls, and of the inner and outer coatings especially 
 of the unit measure, will be quite the same as before, for no 
 external relation can change their mutual relation, though it may 
 affect the outer coatings both of the charging jar or battery and 
 the unit jar, so that whenever a unit spark does pass between the 
 unit balls, the quantity of electricity passing must be the same 
 because the inductive relation of the coatings to each other through 
 the glass, and the inductive relation of the two exploding balls 
 to each other remain absolutely the samp. This is, I think, a 
 rigid consequence of the principles of inductive action." 
 
 125. Lane's Electrometer. -We have seen (116) that in dis- 
 charging the electrical jar by means of a discharging rod the 
 explosive spark takes place befpre the ball of discharge actually 
 touches the positive ball of the jar. A similar result ensues in the 
 case of drawing a spark from the prime conductor of the electrical 
 machine. The celebrated electrician J^ane, availing himself of this 
 fact, constructed an instrument for deterinming the comparative 
 force of an electrical discharge, either under the form of a spark 
 from the electrical machine, or that of an electrical explosion 
 between the opposite coatings of an electrical jar or battery. 
 
 126. Lane's instrument (Fig. 78) appears to have been simple 
 and efficient. In this figure M is a pillar about 8 inches high, and 
 as originally constructed is for the most part of baked wood boiled 
 in linseed oil. The pillar is bored through two-thirds of its length 
 for the reception of a brass rod, R, the upper extremity of which 
 carries a brass ball, B, from 2 to 3 inches in diameter. The rod R 
 is movable within the pillar M, and may be set to any given point 
 
110 
 
 of altitude, and finally secured by a stop screw, M. A brass rod, 
 N H, about a quarter of an inch in diameter, and 6 inches in length, 
 is cut as a screw with the threads one twenty-fourth of an inch 
 >L apart ; this rod passes through the 
 central ball, B, as a nut. One extremity 
 of this rod carries a well-polished metallic 
 ball, N, about 1^ inch in diameter, the 
 other extremity terminates in a milled 
 head, H, for turning the screw N H. A 
 graduated rule or scale, s E, divided into 
 twelve parts, measures the movement of 
 the screw, and passes out horizontally 
 
 Fig. 78. Lane's Electro- j ,-i *il i T_ i i -i_i 
 
 meter as originally constructed. under me milled nead H ; and as tne 
 threads of the screw are one twenty-fourth of an inch apart, 
 each turn of the screw measures the one twenty- fourth of an inch. 
 A circular plate fixed upon, and movable with the screw rod, 
 indicates each turn of the screw on the graduated scale s E below. 
 This plate is divided into twelve parts. The ball N is opposed 
 to a similar ball, p, projecting from the prime conductor c L of the 
 electrical machine. 
 
 127. The principle of this instrument is simple. The quantity 
 of electricity in operation will, as Lane remarks, always be 
 directly as the distance between the balls p, N. If a spark occurs 
 after four turns of the electrical machine when the balls P and N 
 are one turn of the screw N H distant, the spark will again occur 
 after eight turns of the electrical machine, when the balls p and N 
 are two turns of the screw distant, and so on ; that is to say, the 
 number of revolutions of the machine for each explosion increase 
 with the distance between the balls p, N. 
 
 328. In the simple application of Lane's electrometer, the positive 
 or exploding ball p is, as just stated, either fixed to the prime con- 
 ductor, or otherwise insulated and connected with it. By noting the 
 distance between the exploding balls p, H, together with the number 
 of explosions in a given time, Lane compared the power of electrical 
 machines with each other. In applying the instrument to the 
 explosions of the electrical jar, the ball p is placed in connection 
 with the knob or positive side of the jar, and the ball N, with the 
 negative side. The quantity of electricity for each explosion at 
 a given distance between the balls p, N, is, according to Lane, 
 proportionate to the extent of coated glass, and the distance between 
 the exploding balls. Such is Lane's Discharging Electrometer, as 
 originally constructed. 
 
 129. Although the instrument, as at present improved and per- 
 
Ill 
 
 footed, is deserving of consideration, yet it seems to have been ill- 
 appreciated and not well understood as an instrument of electrical 
 research, electricians having taxed it with defects from which it is 
 certainly exempt. They have complained, for example, of its being 
 liable to premature and irregular discharge, in consequence of 
 particles of dust, or other matter floating in the atmosphere, and 
 accumulating between the discharging balls ; also from repeated 
 discharges rendering the surfaces of the discharging balls rough 
 and irregular, and have raised other similar objections to its em- 
 ployment. The author of this work, however, having critically 
 investigated the nature and operation of Lane's instrument, con- 
 siders these objections, if the electrometer be properly constructed 
 and employed, as unsound and of little moment. 
 
 130. The instrument as now revised and perfected by the 
 author, is represented Fig. 79, in which G is a central varnished 
 glass insulating column about 1 inch in diameter, and 14 inches 
 high. This insulating column 
 is fixed in a firm circular foot, 
 F, resting on levelling screws, 
 and steadied by a stout central 
 ring of lead. The column ter- 
 minates in a strong ebony cap, 
 having a plane surface, about 
 2 inches in diameter. This inr 
 sulating column, G, carries a 
 light mahogany stage ? s E, 
 upon which are two light 
 
 Varnished insulating glaSS pilr Fig. 79. Lane's Electrometer as Improved by 
 
 lars of support, p' N', about half 
 
 an inch in diameter, and from 3 to 6 inches long. These are secured 
 in mahogany sliding bases, a, b, movable on the stage s E. Upon 
 the insulator p' is a brass cap, giving support to a stout horizontal 
 brass rod, p q, about one-eighth of an inch in diameter, and 6 inches 
 long. This rod is movable within a small brass tube, fixed in tie 
 upper ball of the cap of the insulator P' ; one extremity of this rod 
 carries a finely-polished copper ball, p, about 1| inch in diameter, 
 easily revolvable within a semicircular, axial, metallic ring. 
 Upon the insulator N is a lesser varnished mahogany stage s' E', 
 about 6 inches long, and 1 inch wide. Upon this stage are secured 
 two firm vertical metal supports, s', E', between which is the 
 regulating screw c D. The vertical supports s', E' terminate in light 
 metallic balls, v, w, through which slides a plain metal rod, w v, 
 about 10 inches long and one-eighth of an inch in diameter. This 
 
112 LANE'S ELECTROMETER IMPROVED. 
 
 rod is acted upon by a central globular nut, m, through which passes 
 a screw-cut rod one-fifth of an inch in diameter, and 5 inches long, 
 the threads of which are one-tenth of an inch apart, and is move- 
 able between the two uprights, s', E'. The movement of the screw is 
 regulated by a milled head, D, fixed on one extremity of the screw- 
 cut rod c D, and by one turn of which the screw is caused to 
 advance or recede one-tenth of an inch. The terminating extremity 
 of the rod w v carries a finely-polished copper ball, N, Ij inch in 
 diameter, similar and directly opposed to the ball p just described; 
 also set in a semicircular axial ring. Balls varying from 1 inch 
 to 5 inches in diameter may be easily and accurately opposed to 
 each other by means of the sliding bases a, b, and caused to present 
 renewed surfaces of explosion by turning them in their axial 
 rings when required. The sliders a, b are steadied in position 
 by tightening screws passing through them. The insulating 
 rod s' carries a graduated circle, divided into ten parts, at the ex- 
 tremity of the screw rod c D, indicating by an index the separation 
 of the balls p, N, to the one-hundredth of an inch. 
 
 Lane's Electrometer, Fig. 79, as thus arranged, may be employed 
 to measure the uniform action of an electrical machine (86), in which 
 case the exploding balls P, N, being insulated, are put in communi- 
 cation with the positive and negative conductors through the 
 medium of their respective rods, p q, and w v. The length and 
 frequency of the sparks may be accurately measured and estimated 
 by means of the opposed balls p, N. In order, however, to obtain a 
 full, round, definite spark for ordinary purposes, the balls should not 
 be less than 4 inches in diameter, nor less than 1 inch apart. 
 "Where very minute explosions are required, the balls may be 
 small and near together, but this is seldom needful for ordinary 
 electrical investigation. Gassiot, in applying the principle to 
 small voltaic discharges, constructed a beautiful instrument, by 
 which the distance between two discharging balls may be adjusted 
 to the one-hundredth of an inch, or even less. 1 
 
 131. When Lane's instrument, as thus constructed, is employed 
 for measuring discharges between two balls, P, N, the balls should 
 not exceed 1^- inch in diameter, and seldom be placed more than 
 four-tenths of an inch apart at the utmost. By a little care and 
 attention to the apparatus the exploding balls may be kept free 
 from dust and extraneous matter, and a nicely polished surface of 
 discharge preserved. WTien employed to measure the explosions 
 of an electrical jar, the ball P is connected with the positive coat- 
 ing through the charging rod, and the ball N with the negative 
 coating (123). 
 
 (1) Phil. Trans., 1840. 
 
ELECTROMETER. 113 
 
 132. Although this form of the instrument is well adapted for 
 general purposes, the electrometer may be constructed in such 
 way as to be more immediately applicable to the electrical jar, and 
 occasionally constitute a portion of the rod of the jar itself. In 
 this improved electrometer, a c b, Fig. 80, is a stout arm of varnished 
 glass rod, about 6 inches in length, and three- 
 tenths of an inch in diameter, bent at c to a right 
 angle, each arm, a <?, c b, being 3 inches in length. 
 The arm, a c, is cemented at a into a metallic 
 socket of support, projecting from a short tubular 
 slider, K. The slider K moves on a tubular rod 
 of brass, d e, and with sufficient friction to remain 
 at any required point. The other arm, c b, 
 is also cemented at b into a socket holder, 
 attached in a similar way to a tubular piece, h, 
 within which is a screw movement acting on a 
 short horizontal rod of brass, f\ this rod carries 
 the polished silver or copper exploding ball N, Fig . 80 . Improved 
 \\ inch in diameter, which is mounted on axial Electrometer. 
 points, as already explained (130). The rod /, and the ball N, 
 are acted on by turning the milled head 0, the degree of motion 
 being measured by an index and graduated circle attached to 
 the tubular piece h ; one turn of the screw moves the ball N 
 through a distance of one-tenth of an inch ; the graduated circular 
 plate being divided into ten parts, we may set the distance to the 
 one-hundredth of an inch. Immediately opposite the ball N 
 is a similar ball, P, set up in like manner on axial points within a 
 semicircular metallic ring, fixed at the extremity of a sliding rod, 
 T P, movable with friction through a short tubular piece on the 
 vertical stem of support d e, and terminating in a covering ball of 
 wood. By these several movements the exploding balls P, N may 
 be nicely regulated and adjusted in respect of each other. By the 
 sliding motion at K, the centre of the balls may be accurately 
 placed in the same right line. The horizontal sliding motion of 
 the rod T p enables us to bring the exploding points P, N into con- 
 tact when the index of the screw motion is at zero. By this 
 motion the exploding points of the balls p, N may be separated by 
 any required interval. Finally; by the axial movement of the 
 balls P, N, we are enabled to renew, as it were, the exploding 
 surfaces, should any abrasion or other defect present itself. The 
 instrument thus constructed is either set upon an independent in- 
 sulator, i, Fig. 80, and connected with the jar or battery for experi- 
 ment through the sliding ball T, or terminating ball d, or it may 
 
 i 
 
114 
 
 DETAILS OF ELECTROMETER. 
 
 Fig. 81. Leyden Jar 
 with Attached 
 Electrometer. 
 
 be placed immediately on the charging rod of the jar itself, Fig. 81. 
 All the junction edges of the caps, or other joints, are neatly 
 covered by small varnished ebony balls. It being 
 of importance to determine the exact distance of 
 the balls p, N, it is convenient, in addition to the 
 measuring screw within the tubular piece h, Fig. 
 80, to be prepared with a set of small ivory or other 
 gauges, varying from one-tenth to four-tenths of 
 an inch in thickness ; these may be simple oblongs, 
 about 2 inches in length, by 1 inch in width, 
 carefully and accurately measured and prepared, 
 and occasionally inserted between the touching 
 points of the exploding balls for given distances, 
 so as to verify the distance given by the turns of 
 the screw, or for estimating the distance of the 
 balls in the absence of a screw measure. 
 
 133. There are several methods of constructing 
 the screw measure. That represented in Fig. 82, after the manner 
 of what opticians term " the rising sight of a sextant/' is neat and 
 compact, the screw part being concealed. We have here three pieces, 
 A, B, c, one within the other ; the outer piece A is a hollow cylinder, 
 and fixed ; the second piece B moves within the outer cylinder, for 
 which purpose it is drilled, and tapped about two-thirds up its length, 
 constituting a hollow nut about the one-eighth of an inch in dia- 
 meter ; the inner piece c is a small cylindrical piece, having the 
 threads of a screw cut on it to match the hollow cylindrical nut. 
 The inner screw terminates in a projecting square and ball, 
 
 H, by which the inner screw 
 c may be moved round within 
 the cylindrical nut B, in doing 
 which the piece B, within 
 Fig. 82. screw Measure. which the nut is cut, is made 
 
 to advance or recede according to the direction in which the screw 
 is moved by the ball H, and it is prevented from turning round in 
 position, within its cylindrical enclosure, by a projecting pin, or 
 by being squared. The movable piece B sends forth a short pro- 
 jection, upon which is fixed an exploding ball, N. By means of 
 a graduated circle and index, s, behind the ball H, the amount of 
 advance or recession of the ball N is accurately measured. 
 
 134. We have already observed (129) -that electricians in the 
 application of Lane's Electrometer have not resorted to sufficient 
 experimental investigation ; they have hence been led to some- 
 what hasty conclusions concerning it. If we critically consider 
 
STEEL- YARD ELECTROMETER. 
 
 115 
 
 the conditions under which the explosive discharges between 
 the balls occur, we immediately detect several sources of error to 
 which it is not unfrequently exposed. It is, for example, quite 
 evident that the exploding balls of the instrument, when in con- 
 nection with the electrical jar, are really coatings to the air 
 between the balls. The exploding balls of the instrument, when 
 connected with a charged jar or battery, are really coatings of a 
 second charged electric medium; the interposed air, operating 
 between the balls in precisely the same way as the interposed glass 
 between the coatings of the jar. We have, therefore, two coated 
 electrics more or less dependent on each other, charging at the 
 same time the air between the balls, and the glass between the 
 coatings of the jar. Now the dielectric air between the balls 
 being a lesser resisting medium than the dielectric glass between 
 the coatings of the jar, will break down or give way under the ac- 
 cumulation, whilst the glass remains, and thus a circuit of discharge 
 is opened for the quantity actually accumulated at the instant in 
 the jar itself; one of the exploding balls being connected with the 
 charged side of the jar, and the other with the opposite side (128), 
 and thus a relative measure of the accumulation is obtained by 
 noting the distance of discharge between the exploding balls, which 
 allows a given accumulation to take place before discharge occurs, 
 the exploding balls being supposed 
 to be always the same, and always 
 affected in the same manner. 
 
 .135. Cuthbertson's Steel Yard 
 Electrometer. This instrument 
 was invented by Cuthbertson about 
 the year 1790. 
 
 D E, Fig. 83, is a stand of seasoned 
 mahogany, about 18 inches long, 6 
 inches wide, and an inch thick. A B 
 is a light metallic tubular rod, about 
 13 inches long, and one quarter of an 
 inch in diameter, terminating in 
 light metallic balls, each about one 
 inch in diameter. The rod A B is delicately balanced on a knife- 
 edged centre, or axis, the centre of gravity being a little above the 
 centre of motion. The beam and its axis are sustained on a 
 strong glass central rod i, through the medium of a hollow 
 metallic ball, c. The ball c is made up of two hemispheres 
 one immediately over the other. The lower hemisphere is 
 fixed on a brass cap terminating the insulating rod of support 
 
 i 2 
 
 Fig. 83. Steel Yard Electrometer. 
 
116 STEEL- YARD ELECTROMETER. 
 
 i. The upper hemisphere is turned with a groove, and shuts 
 closely down upon the lower hemisphere. The fulcrum upon 
 which the knife-edged central axis of the beam A B rests is enclosed 
 between the two hemispheres constituting the ball c. In order to 
 allow of the free ascending or descending motion of the beam A B, 
 the two hemispheres are slit open in opposite directions at each 
 extremity of a diameter of the hollow ball. One arm, e B, of the 
 axial beam has a slider, s, on it, which, when set at different dis- 
 tances from the centre of motion, loads the arm. c B with a pro- 
 portionate weight, the arm c B being graduated from one grain to 
 60. The ball B at the extremity of the graduated arm A B rests 
 upon a similar ball, P, either supported by a light bent insulating 
 arm, T, proceeding from the ball at the extremity of the insulator, 
 i, connected with the central ball c, or sustained, like the ball N, 
 upon an independent insulator. The ball, N, is immediately under 
 and at a short distance from the ball A, at the opposite extremity 
 of the arm. 
 
 136. In the application of this instrument to a jar or battery, 
 the central ball c is put into metallic communication with the 
 positive or charged side, and the ball N with the negative or un- 
 charged side, through a light metallic connecting wire or small 
 chain. A Henley's electrometer, H, is usually fixed on the central 
 ball c to denote the progress of the charge. 
 
 137. It is evident from the construction of this instrument that 
 if the stand D E be horizontal and the ball B at one extremity of 
 the axial beam lightly rest upon the ball P, it will remain in that 
 position without the help of the slider s ; and if the ball B receives 
 u weak electrical charge, the two balls B, p will repel each other, 
 the balls A, N at the same time exerting on each other a slight 
 attraction in consequence of the weak charge accumulating in the 
 ball A, through the medium of the central ball c and axial beam A B. 
 
 The combined operation of these two forces causes the arm c B to 
 ascend, and on account of the centre of gravity being a little above 
 the centre of motion, the ascension will continue till A rest upon N. 
 If the beam be again adjusted horizontally and the slider set to any 
 given weight, it will cause B again to rest upon p with a pressure equal 
 to that weight, so that more electricity must be accumulated before 
 the balls B p again separate, and as the weight on the graduated 
 arm c B is increased or diminished by the slider s, a greater or 
 less degree of repulsion will be required to separate the balls. It 
 is further evident that when the electrical accumulation in the jar 
 or battery is sufficiently strong, the arm c B rises, and the arm c A 
 falls and descends upon the ball N, when discharge immediately 
 
CHARGING AND DISCHARGING ELECTROMETER. 117 
 
 ensues, since the ball A is in communication with the positive side 
 of the battery through the central ball c, and the ball N in com- 
 munication with the negative side through a light metallic wire or 
 chain, n (110). 
 
 138. There seems, however, to have been some little misappre- 
 hension on the part of Cuthbertson and the older electricians in 
 their views of the electrical terms Tension and Intensity (62). 
 They assumed that the quantity of electricity accumulated in 
 a jar or battery will be directly as the force required to separate 
 the balls B, p, and consequently as the position of the slider on the 
 graduated arm. Thus, if the position of the slider s indicated a 
 resistance of 5 grains when the quantity of electricity accumulated 
 was equal to unity or 1, the quantity accumulated would, accord- 
 ing to Cuthbertson, be equal to 2 when the slider was set to 10 
 grains, or double the former resistance. But since it is demon- 
 strable (63) that the intensity of an electrical accumulation is as 
 the square of the quantity, if intensity be taken as a measure of 
 the force, in order to accumulate twice the quantity the slider 
 should be set to 20 grains instead of 10, or four times the resist- 
 ance instead of twice. The older electricians have evidently sub- 
 stituted what has been termed electrical tension (58) of the explod- 
 ing distances of the electrical spark for its attractive force, a very 
 important distinction, as exemplified in Lane's Electrometer. 
 
 Many errors in electrical research have arisen from not being 
 aware of this important law. 
 
 139. The Charging and Discharging Quantitative Electrometer. 
 A convenient and ready means of charging and discharging an 
 electrical jar or battery, or other electrical accumulation, with 
 ease, convenience, and safety to the operator, either in the light 
 of day or amidst the darkness of night, has been deemed a 
 desideratum in practical electricity. 
 
 Fig. 84 represents and describes an instrument calculated to 
 satisfy the conditions of this important mechanical problem. In 
 this figure A B, is a mahogany elliptical stand or base, sustained on 
 levelling screws, its long diameter being about 15 inches, its lesser 
 diameter about 7 inches, and the thickness three-quarters of an inch. 
 
 At each extremity of the long diameter is a strong vertical var- 
 nished glass insulator A N, and B p, each three-quarters of an inch 
 in diameter, and about a foot in length. These insulators are set 
 in varnished mahogany feet, each about 2 inches in diameter, 
 firmly screwed into the elliptical base at each of its extremities. 
 The insulating column B p is capped with a stout wooden cap 
 and ball, p ; a brass rod, p <?, about 6 inches long and 1 quarter 
 
118 
 
 CHARGING AND DISCHARGING ELECTROMETER. 
 
 of an inch in diameter, passes centrally through the ball of the 
 wood cap P, and terminates in a brass ball, p, about an inch 
 in diameter. The wood cap of the insulator sends up a vertical 
 brass tube, /, about 8 inches long and three-eighths of an inch in 
 
 Fig. 84. Charging and discharging quantitative Electrometer. 
 
 diameter, having a sliding ball, I, on it, which may be adjusted to 
 any convenient altitude. 
 
 The lower portion of the tube / is a fixture, and proceeds from 
 the ball P of the wood cap, and directly communicates with the 
 transverse rod p q, passing through the ball. Within the fixed 
 portion of the tube / is a second light tube terminating in a brass 
 ball, b, sustaining a small unit measure,, u (120), communicating by 
 a light tubular connection with the conductor of the electrical 
 machine. The outer coating n of the unit measure u commu- 
 nicates with the jar or battery to be charged through the sliding 
 ball /, which has holes drilled in it for the reception of small con- 
 necting rods. The insulator A N, at the opposite extremity of the 
 elliptical base, is capped with brass, supporting a metallic ball, s, 
 within which is a strong joint movable with friction, and acted 
 upon by a varnished glass rod, a m, about 5 inches long and one- 
 quarter of an inch in diameter, firmly cemented in a short brass 
 socket passing through the flanges of the joint, and terminating in 
 small, varnished mahogany balls, so that by means of the glass 
 rod a m and axial tube, passing through the flanges of the joint, 
 we may cause the joint to turn in either direction, according to 
 the direction of the force applied to the terminating balls at the 
 extremity of the axis. A stout brass rod, s n about 6 inches 
 long and a quarter of an inch in diameter, carrying a brass ball, 
 n, about three-quarters of an inch in diameter is firmly screwed 
 into the centre piece or flange of the joint, acted upon by the axial 
 insulating rod a m above mentioned, so that by turning the axis 
 a m in either direction, the ball n may be raised or depressed to 
 
QUANTITY JAR AND TRANSFER MEASURE. 
 
 119 
 
 any required extent. The ball N on the brass cap on the insulator 
 A N carries a short stout brass rod, r d, in which is inserted the ter- 
 mination of the circuit leading to the outside of the jar or battery. 
 
 140. It is evident by this arrangement that since the ball p is 
 in direct communication with the positive coating of the battery, 
 through the connection of the sliding ball and tube /, with the 
 outer coating of the unit measure u, and the ball n, in connection 
 with the circuit leading to the outside of the battery, directly we 
 cause the ball n to descend upon the ball p, we discharge the 
 whole accumulation by putting the inside and outside of the battery 
 in direct metallic communication with each other, through any 
 given circuit we choose to establish (110), and this may be effected 
 either in the broad light of day, or in the darkness of night, with 
 ease and safety to the operator, by simply turning the insulating 
 axis a m. 
 
 141. The Quantity Jar and Transfer Measure. The object of these 
 instruments is to communicate to insulated conducting, or other 
 bodies, given measured quantities of electricity, for which purpose 
 we require a sort of reservoir of force of given magnitude, from 
 which we may abstract, at any instant, accumulated electricity, 
 and by means of a carrier (14) deposit the quantity abstracted on a 
 given surface. The reservoir of force may be termed a Quantity Jar. 
 
 This jar, as prepared and arranged, is represented in Fig. 85. 
 j is an ordinary electrical jar of the form already described (113). 
 It is one foot high, 6 inches diameter, and '1 
 of an inch in thickness, and exposes about 
 200 square inches, or about Ij square foot, of 
 coating, the height of the coating being about 
 9 J inches. It has an open mouth of about 3 
 inches diameter. The jar is carefully var 
 nished, and fitted with a charging rod, its 
 coatings being well and smoothly rubbed 
 down on the glass. The jar, thus prepared, 
 is placed upon a circular varnished glass salver, 
 s, within a light open ring of varnished wood, 
 fitted to the salver so as to retain the jar in a 
 central position. The jar is finally covered 
 by a light bell-glass screen, E, resting in 
 a concentric groove in the wood-ring, within 
 which the jar is placed. The bell-glass screen 
 has an open mouth of sufficient diameter to 
 allow the charging-rod and ball, p, to project 
 through and beyond it to any required extent ; this bell-glass 
 
 Fig. 85. Quantity Jar. 
 
120 TRANSFER MEASURE. 
 
 envelope E exceeds the diameter of the jar by about 2 inches, and is 
 varnished, inside and out, from its mouth to about an inch below the 
 height of the outer coating of the enclosed jar. A wide band of 
 tinfoil is pasted diametrically across the surface of the glass plate of 
 the salver, upon which the jar rests, and is continued over the rim 
 of the salver, and terminates in the under surface. A similar short 
 band of tinfoil is pasted around the wooden ring at opposite extre- 
 mities of a coincident diameter, so as to have contact with the tinfoil 
 band traversing the surface of the glass beneath. A communication 
 is thus ensured with the outer coating of the jar when required, 
 and for which purpose there is a small wire and ball, N, projecting 
 from the band of foil where it passes over the edge of the wooden 
 ring. The jar thus circumstanced is charged to a given intensity, 
 through a unit measure, in the usual way (121), communication 
 being at the same time made with the outer coating of the jar and 
 the ground by means of the small projecting ball N. 
 
 142. A jar, when charged, insulated, and covered up in this way, 
 retains its charge for a long time, and will communicate a succes- 
 sion of precisely equal charges to an insulated carrier (14) when 
 brought into contact with the metallic ball at the extremity of the 
 charging rod of the jar, communication being at the same time 
 made between the outer coating of the quantity jar and the ground, 
 when applying the carrier. 
 
 It is desirable to employ for experiment two quantity jars thus 
 arranged and prepared, one for positive and the other for negative 
 electricity, or for two jars combined when required. The positive 
 jar may have a small band of red paper, about a quarter of an inch 
 wide, pasted over the edge of the outer coating, in the way recom- 
 mended by Mr. Singer. The negative jar may be furnished with 
 a similar band of blue paper. These bands, whilst they serve to 
 restrain spontaneous explosion, denote at the same time the elec- 
 tricity with which each jar is charged. 
 
 143. The Transfer Measure. This instrument is to be employed, 
 in conjunction with the quantity jar, for the purpose of abstract- 
 ing from it given measured quantities of electricity, and 
 communicating the abstracted quantity to a given insulated sur- 
 face (14). The transfer measure consists of a light, thin, circular 
 metallic disc, D, or globe, G, Fig. 85, having an eye of fine silk gut 
 firmly secured to it, by which the plate or globe may be taken up 
 on a long slender rod of varnished glass or vulcanite, v v. 
 
 If the measure be a plate, it is requisite to secure the suspen- 
 sion eye effectually, for which purpose a small loop is tied cen- 
 
TRANSFER MEASURE. 121 
 
 trally in a fine thread of varnished silk-gut ; two fine holes are 
 drilled obliquely upward, on each side of the transfer plate, from 
 the lower edge of the circumference at each extremity of a 
 diameter, so as to extend from the edge of the circumference to 
 a little within the surface : the extremities of the suspension gut 
 are then passed outward through these holes to the edge of the 
 plate, in order to bring the loop in the centre. The gut is secured 
 by fine wooden pegs, nipped off fair with the circumference. The 
 vulcanite rod or holder is tapered, rounded off, and slit open by 
 a thin saw cut for the reception of and for holding fast the sus- 
 pension loop of the plate. When a light rod of varnished glass is 
 employed, the glass is bent a little upward at one extremity to 
 prevent the gut loop from falling off it. 
 
 The suspension loop is easily secured in a globular transfer by 
 drilling a fine hole in it, inserting the extremities of a simple loop 
 in the hole, and finally securing it by a small wooden peg, nipped 
 off fair with the surface. A complete insulation is essential to the 
 perfect operation of the transfer measure. The transfer measure 
 thus prepared is charged to saturation, or to a given intensity by 
 contact with the charging ball of the quantity jar. 
 
 144. The practical application of the quantity jar and transfer 
 measure is as follows. The jar being placed on its insulating 
 salver, Fig. 85, and covered up (the ball p of its charging-rod 
 projecting clear of the mouth of its bell-glass screen), it is charged 
 to a given intensity, through a unit measure, mounted on a long 
 insulator (121). The jar being charged, is now withdrawn 
 from the unit measure and electrical machine, when it is ready 
 for experiment, and the application of the transfer plate, which, 
 after contact with the charging ball as just mentioned, is trans- 
 ferred to the subject of experiment. 
 
 145. It has been shown in the author's paper in the Philoso- 
 phical Transactions for June, 1864, that the quantity of electricity 
 which circular planes, or globes, can receive up to a given point 
 of intensity, or saturation, is as their diameters. It is easy, there- 
 fore, to calculate a series of circular planes, or globular transfers 
 of definite values, taking the circular or globular inch as unity, 
 calling the circular plate of an inch diameter, after Cavendish, a 
 circular inch of electricity ; and a globe of an inch diameter, a 
 globular inch of electricity ; the plate or globe being, as just 
 observed, charged to saturation, or up to a given intensity. 
 
 In the following table are given the quantities of electricity in 
 particles, or units of charge, contained on circular plates, or globes 
 of given diameters varying from -25 to 2 inches, together with 
 
122 
 
 HYDROSTATIC ELECTROMETER. 
 
 the respective intensities. In this table the circular inch, that 
 is, a circular plate of an inch in diameter, and one-fifth of an inch 
 thick, is taken as unity, and supposed to contain 100 particles, or 
 units of charge. 
 
 Diameters, 
 or Units o 
 Charge. 
 
 CIRCLE. 
 
 GLOBE. 
 
 Particles. 
 
 Intensity. 
 
 Particles. 
 
 Intensity. 
 
 0-25 
 
 25 
 
 0-062 
 
 35 
 
 0-124 
 
 0-50 
 
 50 
 
 0-250 
 
 70 
 
 0-500 
 
 0-75 
 
 75 
 
 0-560 
 
 105 
 
 1-120 
 
 1-00. 
 
 100 
 
 1-000 
 
 140 
 
 2-000 
 
 1-25 
 
 125 
 
 1-560 
 
 175 
 
 3-120 
 
 1-40 
 
 140 
 
 1-960 
 
 196 
 
 3-920 
 
 1-50 
 
 150 
 
 2-250 
 
 210 
 
 4-500 
 
 1-60 
 
 16Q 
 
 2-560 
 
 224 
 
 5-120 
 
 1-75 
 
 175 
 
 3-060 
 
 245 
 
 6-120 
 
 2-00 
 
 200 
 
 4-000 
 
 280 
 
 8-000 
 
 146. Although the foregoing table will be found useful in the 
 course of experimental inquiry, it is seldom requisite for general 
 purposes to employ transfer measures exceeding the globular 
 or circular inch. For estimating small quantities of electricity, 
 however, to be communicated to the gold leaf, or other very 
 sensitive electroscopes, very small transfer plates or globes of *25 
 of an inch diameter, or even less, may be resorted to. 
 
 Perfect insulation of the transfer-measure being necessary, care 
 must be taken in perfecting the insulating power of the vulcanite 
 or glass rod, by which the plate or globe is to be suspended. 
 These insulations should be as slender as possible, nicely varr 
 nished, and be made very dry. 
 
 147. Hydrostatic Electrometer. In this instrument (Fig. 86), 
 the force of attraction between a charged disc, P, and a neutral 
 disc, N, in connection with the earth, is hydrostatically counter- 
 poised by a cylinder of wood, c, accurately weighted, and partially 
 immersed in a vessel of water. The neutral disc N and its balance 
 are freely suspended by two fine lines, I /', in opposition to each 
 other, over the circumference of a light brass balance-wheel, w, 
 2-4 inches in diameter, weighing about 160 grains. The line /, 
 suspending the disc N, is of fine gold thread, by which its free 
 electrical state is always preserved. The thread of suspension 
 passes through a small hole drilled through the circumference of 
 the wheel at the extremity of a vertical diameter, and is held there 
 by a knot tied in it ; from this it passes over the circumference, 
 
DETAILS OF HYDROSTATIC ELECTROMETER. 
 
 123 
 
 and terminates in a small hook, to which the suspension lines of 
 the plate N are appended ; the plate itself is hung by three lines 
 
 Fig. 86. Harris's Hydrostatic Electrometer. 
 Figs. 87, 88, and 89 show some of the separate parts enlarged. 
 
 of gold thread attached to opposite points of its upper surface, and 
 converging to a centre after the manner of a scale pan. The line 
 /, suspending the counterpoise c, is of silk thread, and is fixfed in 
 a similar way through a hole in the circumference of the wheel, 
 close to the former ; from this hole it passes over the circumfer- 
 ence, and also terminates in a hook sustaining the counterpoise. 
 The circumference of the wheel is slightly grooved for the recep- 
 tion of the lines of suspension. The attracting discs, P N, are of 
 
124 HYDROSTATIC ELECTROMETER. 
 
 light wood, smoothly gilt, about li inch in diameter and -1 of an 
 inch thick. These dimensions, however, may be varied. 
 
 The wheel, w, is mounted on a steel axis, the extremities of 
 which are hardened and turned down to fine pivots ; these pivots 
 rest centrally on the circumferences of two light friction wheels 
 (Fig. 87) of 11 inch in diameter, each weighing about 40 grains 
 or less. The axis of the large wheel is prevented from falling to 
 either side by four smaller, or check wheels, each 1 inch in 
 diameter, and weighing 20 grains. These are placed two on each 
 side of the terminating pivots of the axis of the large central 
 wheel. The last-named six wheels are delicately poised, and set 
 on hard steel axes centred in jewels or in hammered brass, similar 
 to the wheel- work of chronometers, so that the large wheel has 
 perfect freedom of motion, and is influenced by the slightest force 
 between the attracting plates P, N. 
 
 "Whatever this force be, it is speedily counterpoised, hydro- 
 statically, within certain limits, by the elevation or depression of 
 the cylindrical float c above or below the water level. The force 
 thus counterpoised is measured by an index, i (Fig. 86), of light 
 straw reed 7 inches long, weighing 11 grain, attached to a 
 radius of fine steel wire, s (Fig. 88), passing through the 
 circumference of the wheel w. The weight of the index i 
 is counterpoised by a small brass ball, B, movable with screw 
 motion on a similar steel wire, continued on from the boss to 
 an opposite point in the circumference, so that the two steel 
 wires constitute a diameter of the wheel. The index moves over 
 a graduated quadrant, Q T (Fig. 86), of about 7 inches radius, 
 accurately divided into 45 degrees on each side of its centre, o ; 
 indicating the movement of the wheel, in degrees, in either 
 direction, o Q, o T, the centre of the arc being zero. Each 5 
 degrees of the arc corresponds with the -A-nd part of the circum- 
 ference of the wheel, so that the plate N ascends or descends about 
 one-tenth of an inch for every 5 degrees of the arc. 
 
 The arc, Q T, is composed of cardboard about three-fourths of an 
 inch wide, and is attached to a light open quadrant of support, con- 
 structed of thin slips of wood fixed immediately behind the wheel, 
 and its index upon a brass plate, M R, carrying the wheelwork : 
 this brass plate is 4 inches long and 1 inch wide, and is, with 
 the wheelwork and quadrant of measure, supported upon two small 
 pillars of brass 1| inch high and one-fourth of an inch in diameter, 
 screwed into a brass stage or platform, s E, 11 inches long, and 
 from li to 2 inches wide, where it receives the small brass pillars. 
 The stage is secured at its narrow extremity upon the cap of a 
 
HYDROSTATIC ELECTROMETER. 125 
 
 vertical column, x z, by means of a nut and projecting screw 
 passing through a slot in the stage, so that the stage may be 
 adjusted to project from the column to any required distance, 
 not greater than 8 inches. The hydrostatic counterpoise c, Fig. 
 89, is about 2 inches in length, and '3 of an inch in diameter ; 
 but may be greater or less than this, if required. A stout 
 brass wire is screwed centrally into its lower extremity, leaving 
 a short projection of about one-eighth of an inch, upon which is 
 the thread of a screw for the reception of a small cylindrical or 
 globular weight, w, sufficient to counterpoise the suspended disc, 
 N, and cause the float to sink freely in the water when not held 
 in equilibrio by an opposite force. The upper extremity has a 
 conical cavity sunk in it for the reception of fine shot, or other 
 small weights, which may be required to occasionally regulate its 
 position in the water. Should the equilibrium, in certain cases, 
 be not readily obtained on the side of the suspended electrometer 
 disc, N, it is effected by placing a thin metallic ring upon its upper 
 surface (Fig. 86). The water vessel carrying the hydrostatic 
 counterpoise c, is about 2J inches in diameter, and 3 inches deep, 
 being an ordinary inverted cupping-glass. It is closed in by a 
 light wood cover, having a round hole through its centre for the 
 passage of the counterpoise float, and is held in a ring of brass at 
 the extremity of an horizontal rod, acted on by a vertical screw 
 motion, contained in a short brass cylinder fixed in the platform 
 carrying the wheelwork. The water vessel c, by this screw motion, 
 may be raised or depressed to a certain extent, so as to cause the 
 index i to move in either direction, o Q, o T. The stage s E, sus- 
 taining the electrometer wheels and other portions of the apparatus, 
 is, as just stated, secured upon the cap of a vertical column, x z, 
 fixed on an elliptical base, Y z, 1 foot long by about 7 inches wide, 
 set on four levelling screws. The lower portion of this column 
 consists of two brass tubes, u H, u z, each about 5 inches long and 
 1 inch in diameter, movable one within the other by a rack and 
 pinion at u. This portion of the column, when fully elongated by 
 the rack and pinion, is about 9 inches in length ; it carries, by 
 means of a tube v H, a wooden column, v x, about 1 inch in dia- 
 meter and 1 foot long, terminating in the cap x ; the lower end of 
 this wooden column is fixed in a brass tube 2| inches long, movable 
 by a second rack and pinion, v, within the tube v H, which is 
 screwed upon the upper tube, u H, of the lower rackwork, so that 
 the wooden part of the supporting pillar can be raised or depressed 
 alone. 
 
 The attracting disc p, immediately opposed to the suspended plate 
 
126 HYDROSTATIC ELECTROMETER. 
 
 N, rests on a small circular plate of brass three-fourths of an inch in 
 diameter, screwed upon the brass cap of an insulated rod of glass, g, 
 10 inches long by one-fourth of an inch in diameter. This insulating 
 rod is movable with friction, through fine cork, within a graduated 
 slider, movable within a tube fixed upon a sliding piece, Y, at the 
 extremity of the elliptical base. The insulating support g, of the 
 attracting plate P, being movable within the graduated slider, 
 may be regulated to any convenient height, giving a clear insu- 
 lation of about 8 inches. A short piece of small brass tube, about 
 1 inch in length and one- tenth of an inch in diameter, is fixed 
 transversely through the cap of the insulator, immediately under 
 the brass plate supporting the attracting disc P, for the occasional 
 reception of connecting wires, or of a transverse movable rod 
 terminating in small brass balls ; and by which an electrical jar, 
 or any other insulated body, may be conveniently connected with 
 the attracting plate P of the electrometer. In order to estimate 
 the elevation or depression of the suspended disc N, by means of 
 the rack u, carrying the column u v, a vertical scale of measure 
 of about 4 inches in length divided into twentieths of an inch, is ap- 
 plied to the upper brass tube, so as to measure the movement, up or 
 down, of the wooden column carrying the electrometer wheels. The 
 zero line of this scale coincides with the edge of the lower tube 
 when the two tubes are coincident, and there is a small tangent 
 screw, t, by which the movable zero line may be made perfectly 
 coincident with the fixed line. One turn of the pinion of the rack 
 raises the wooden column carrying the wheelwork Hths of an inch. 
 In addition to this scale of measure, a light open circle of 4 inches 
 diameter, and divided into 64 parts, surrounds the pinion of the 
 rack. The axle of the pinion carries a light index movable upon 
 it with friction, so that the index may be set to zero of the circle, 
 when the small vertical scale of the rack is accurately adjusted to 
 zero. We may in this way estimate the vertical movement of the 
 rack to the ^th of an inch, or very nearly ; for since the gra- 
 duated circle surrounding the pinion is divided into 64 parts, one 
 division corresponds to the - 6 Vth part of ifths of an inch, equal to 
 0101 ; that is, the T <h>-th of an inch very nearly. The suspended 
 disc N of the electrometer, therefore, will be raised or depressed 
 i th of an inch for each degree of the circle, according to the 
 direction in which the milled head u of the pinion of the rack is 
 turned. 
 
 The levelling screws on which the elliptical base Y z, carrying 
 the whole of the electrometer apparatus, is fixed, rest on the 
 platform of a small railway carriage, movable on rails, R, R', so as 
 
HYDROSTATIC ELECTROMETER. 127 
 
 to transfer the instrument from one point to another on the line 
 of the rail with the least possible disturbance. The platform is 
 10 inches long and 7J inches wide. The rails are of brass, and 
 are set 5 inches apart upon an open mahogany frame about 5 feet 
 long by 7J inches wide. The frame is sustained on levelling 
 screws, supported at its opposite extremities on two circular tables 
 fixed on central columns, which being movable in stout brass 
 tubes fixed on levelling stands, can be adjusted to any convenient 
 height. There are two light travelling platforms, a, b, movable 
 on the rails, 1 foot long by 7 inches wide ; these are for the support 
 of insulated or other bodies to be placed in connection with the 
 electrometer in the most convenient way. Beside these travelling 
 stages there is a light platform stage at the side of the rails 3 feet 
 long and 7 inches wide, resting on two projecting bars attached 
 by clamp screws and nuts to the under-part of the open rect- 
 angular frame upon which the rails are based. This platform is 
 for the purpose of placing insulated or other bodies in convenient 
 positions at the side of the electrometer. 
 
 The bars supporting the platform being movable, may be caused 
 to project on the other side of the rails, so as to transfer the 
 platform to the other side of the electrometer, or admit an addi- 
 tion of a second platform on the opposite side of the rails, for the 
 convenience of supporting on insulators extensive conducting 
 surfaces in any required direction. 
 
 148. It may be observed that unless the wheelwork is delicately 
 mounted, and accurately set in place, a small index error of a 
 degree or a degree and a half may arise upon the measurement of 
 extremely small forces. This, however, is easily detected by a 
 slight oscillation communicated to the index I, by which it will be 
 immediately seen whether the pivot of the great central wheel is 
 liable to shift its position either to the right or to the left. There 
 is no difficulty in making a small allowance for this error, should 
 it occur. 
 
 149. This electrometer is peculiarly adapted to the measurement 
 of electrical force, which can be estimated in terms of an admitted 
 standard of weight. If, for example, with a given hydrostatic 
 counterpoise, 1 grain added to either side moves the index 5 
 degrees in either direction, 2 grains should move it 10 degrees,, 
 and so on. We may, therefore, estimate degrees in grains weight. 
 Thus, if we had a registered force between the attracting plates, 
 p, N, of 15 degrees, that would, in this case, be equivalent to a 
 force of 3 grains. 
 
 Another advantage in using this instrument is, that the expe- 
 
128 HYDROSTATIC ELECTROMETER. 
 
 riment requires only a short time for its accomplishment ; so short 
 a time, indeed, as to supersede the necessity of any kind of calcu- 
 lation for electrical dissipation. When the electrometer insulations 
 are perfectly dried off, and the air dry, the index will often remain 
 unmoved for at least ten minutes, and sometimes longer ; whilst 
 any experiment with the instrument may be concluded in less 
 than three minutes, and occasionally in less than one minute. 
 
 150. The mechanical conditions of the electrometer being fully 
 apprehended, the experimental manipulation will be as follows : 
 The index being accurately set at zero of its arc, either by pouring 
 a certain quantity of fluid into the water- vessel, or by a small move- 
 ment of the screw motion that supports it, the accuracy of its indi- 
 cation is determined by placing small weights of a grain each, or 
 less, either upon the suspended plate N, or in the conical cavity 
 of its counterpoise. If 1 grain move the index 5 degrees in 
 either direction, 2 grains should, as just stated (149), move it 
 10 degrees, and so on, which it will not fail to do if the instru- 
 ment be carefully constructed, so that the index error, if any, will 
 be small. The hydrostatic counterpoise c must, however, be previ- 
 ously immersed for a short time in water, so that no part of it 
 remain dry ; for if it were so, it would interfere with its correct 
 position in the fluid. When in perfect order the index may be 
 caused to oscillate by a slight movement of the wheel, in either 
 direction, with perfect freedom, and will invariably return to the 
 zero of its arc when the oscillation ceases. 
 
 151. In the application of the electrometer to experiment, as 
 for example, to the determination of a charged electrical jar, or 
 to that of an insulated circular plate, or any other charged surface, 
 we first determine the distance at which the force between the 
 attracting plates is to operate. This distance is measured in the 
 following way : The index of the arc being carefully adjusted to 
 zero, and the vertical scale and the index of the circle of the lower 
 rackwork also adjusted to zero, the surfaces of the attracting plates 
 p N are brought into contact by the elevation or depression of 
 the slide supporting the lower disc, or by a more accurate adjust- 
 ment, by means of the upper rack- work v operating on the wooden 
 column supporting the wheel-work of the apparatus. By means 
 of this rack the suspended plate may be lowered or raised to the 
 last degree of precision. The plates P N being thus in contact, 
 and the index of the arc, together with that of the scale of 
 measure of the rack, adjusted to zero, we now proceed to raise the 
 column supporting the wheel-work and the suspended plate N 
 through a space equal to the required distance between the plates 
 
EXPERIMENTS THEREWITH. 129 
 
 p N, suppose, for example, half an inch. We know, therefore, 
 that when the index of the arc is at zero, the two plates P N are 
 5 of an inch apart. 
 
 152. The following experiments may serve as illustrations of 
 the operation of this instrument. 
 
 Exp. 38. Let it be required, for example, to investigate the 
 law of electrical attractive force, between the circular planes P, N, 
 as regards quantity. Connect the electrometer with a hollow 
 spherical conductor of 5 inches diameter, having an open project- 
 ing circular mouth of 3 inches diameter, so as to freely admit a 
 transfer measure of 1 circular or globular inch (145). Proceed 
 now to transfer to the interior of this sphere a given quantity 
 of electricity, suppose 5 globular inches ; the discs of the 
 electrometer will attract each other, and the index will move 
 forward in the direction o Q, Fig. 86. (To prevent the suspended 
 disc from descending beyond a certain limit, a small stop-pin is in- 
 serted in one of the radial arms of the arc on the side o Q of the index.) 
 The water vessel is now lowered by the screw motion, until the 
 index is again brought back to zero of the arc. We are now 
 assured that whatever be the force between the attracting plates, it 
 is operating at the given distance of half an inch. To determine 
 the amount of this force, we discharge the electrical accumulation. 
 The hydrostatic counterpoise now descends in the fluid, and the 
 index recedes in the direction o T, until an equilibrium again 
 ensues between the counterpoise and the suspended disc, showing 
 the amount of force previously in operation. Let the force thus 
 shown be, for example, 5 degrees, being supposed to be equivalent to 
 aforceof 1 grain (149). Charge the electrometer, through the hollow 
 sphere, with 10 globular measures, or twice the former quantity ; the 
 hydrostatic counterpoise will again be raised, and the index move 
 forward in the direction o Q, until impeded by the stop-pin, against 
 which it rests. We now further lower the water vessel, until the 
 index again rests at zero ; the force between the plates, whatever 
 it be in this case, is still operating at distance '5. We determine 
 the amount of this force as before, by discharging the electrical 
 accumulation. A hydrostatic balance again ensues between the 
 suspended disc and its counterpoise float, so that the index will 
 have again receded in the direction o T ; showing the amount of 
 force in degrees due to 10 circular inches, or twice the former 
 quantity, and which, in this case, will be 20 degrees, or four 
 times the former force. Thus, whilst the quantities of electricity 
 are as 1 : 2, the forces are as 1 : 4, or as the squares of the quan- 
 tities. The electrometer, therefore, together with the method of 
 
 K 
 
130 SCALE-BEAM ELECTROMETER. 
 
 experiment, is in strict accordance with an admitted and well- 
 known law of statical electricity, and is so far entitled to our 
 confidence (R). 
 
 Exp. 39. Let it be now required to investigate the law of 
 attractive force as regards distance. Raise the vertical pillar 
 another half inch, which will double the distance between the 
 attracting plates when the index is at zero of the arc. Repeat 
 the charge of 10 globular inches, which by the last experiment 
 shows a force of 20 degrees at distance '5. The index will 
 again advance in the direction o Q. Bring the index again to 
 zero, which is effected by operating on the water vessel as in the 
 former case. We now, as before, discharge the accumulation. 
 
 The hydrostatic counterpoise will again descend in the fluid, and 
 the index again recede in the direction o T, indicating the force 
 between the plates at 1 inch, or double the former distance, which 
 will be now 5 degrees, or only one-fourth the former force. In 
 this case we see that whilst the distances are as 1 : 2, the forces 
 are as 4:1, the quantity of electricity being constant ; that is 
 to say, the force of attraction between the electrometer discs is 
 in an inverse ratio of the squares of the distances. The electro- 
 meter, therefore, together with the method of experiment, is in 
 strict accordance with well-established laws of electrical action. 
 The same laws may be observed with increased quantities at in- 
 creased distances. It is requisite, however, to observe that to get 
 a greater range of distance or force, we must take such distances 
 or forces as come within the grasp of the instrument. We could 
 scarcely, for example, obtain a staple equilibrium of 45 degrees at 
 the short distance of '5 of an inch only between the plates, which 
 would be the force due to 15 globular inches measured as before. 
 To exemplify the law of quantity, therefore, for several terms of 
 the series, it is requisite either to take an increased distance or a 
 less force. 
 
 153. The Scale-beam Electrometer. In this instrument the 
 common scale-beam is adapted to the purpose of measuring the 
 attractive or repulsive forces between bodies of various kinds 
 and forms, charged with given quantities of electricity, and 
 placed at given distances from each other. The Scale-beam Elec- 
 trometer is represented in Fig. 90. L M is a delicate scale- 
 beam, 10 inches in length, weighing about 300 grains, and will 
 readily turn with one-tenth of a grain when loaded with 100 
 grains. It is hung in the usual way from a convenient support, 
 w. From one arm of the beam is suspended, by three fine gold 
 threads, a light circular gilt plate, N, or other body, accurately 
 
SCALE-BEAM ELECTROMETER. 
 
 131 
 
 balanced by an equivalent counterpoise, in a scale-pan, s, sus- 
 pended from the opposite arm M. Being accurately balanced, 
 the attractive force between the suspended plate N and a 
 plate, P, placed immediately under it, is estimated by grain 
 weights placed in a second scale pan, T, hung by three silk 
 lines from the opposite arm M of the beam L M. The scale- 
 pan s is attached beneath the scale-pan T by its three lines of sus- 
 pension, which are continued on for the purpose. 
 
 L I B R A R Y 
 
 UNIVERSITY Of 
 
 CALIFORNIA. 
 
 Fig. 90. Scale-beam Electrometer. 
 
 The attracting plate P, immediately under the suspended disc N, 
 is supported on a varnished insulating rod, I, 6 inches long; 
 and three-quarters of an inch in diameter. * In order to insure 
 the parallelism of the two plates P N, the attracting body P rests 
 on four small levelling screws, passing through a brass plate, 
 supported on the insulator i, by means of a tube, sliding with 
 friction through another tube, proceeding from a ball, c, at the 
 extremity of the insulator, and by which the altitude of the plate 
 p may be adjusted, and electrical charge communicated to it. 
 
 The insulator I, with its attracting plate P, is fixed in a sliding 
 foot, A, forming part of an elliptical base, A B, 16 inches long, 
 its width across being 9 inches, and its thickness 1 inch. By 
 
 K2 
 
132 SCALE-BEAM ELECTROMETER. 
 
 means of the movable foot A, the insulating support I carry- 
 ing the attracting plate p may be adjusted to any given point. 
 The opposite extremity B of the elliptical base carries a compound 
 column of support, B E, variable in length from 20 to 30 inches. 
 This column carries a light rectangular projecting brass plate of 
 support w, about 10 inches long, half an inch wide, and one-eighth 
 of an inch thick. This plate supports the delicate scale-beam L M, 
 and its appendages. 
 
 In order to steady the beam and give it occasional support, there 
 is a small open rectangle of brass, attached to a rod, made to 
 slide within a tube screwed into the plate of support w. The 
 beam passes through this rectangle, so that by means of a cross- 
 pin, and small holes in the sides of the rectangle, the beam may 
 be steadied in any position, and its motion, by a due adjustment of 
 the sliding rod, restrained. 
 
 The compound column B E supporting the scale-beam M N, and 
 its suspended bodies, N, s, consists of different parts, E, F, G, H, 
 and B. 
 
 The upper part E is a cylinder of extremely light wood 
 (Chap. V., 35), about a foot in length, and a full inch in diameter. 
 
 The cylinder is inserted and fixed in a brass socket, F, three 
 inches in length, at the side of which is a delicate rack, acted on 
 by a pinion fixed at the upper extremity of a brass tube, G, within 
 which the socket carrying the pillar slides with friction. By 
 means of this rack and pinion the portion E may be adjusted 
 to any given altitude. 
 
 The part G is screwed on the upper extremity of a following 
 stout brass tube, H, at the side of which is a rack similar to that 
 of the tube F, acted on by a pinion fixed at the upper extremity 
 of a final brass tube, B, screwed into the elliptical base A B at the 
 extremity B. By means of this rack and pinion the total height 
 of the compound column B E may be varied. 
 
 In order to estimate any change which may be effected in the 
 altitude of the column B E by means of the lower rack, the 
 pinion moving the rack is furnished with a light open graduated 
 circle, divided into 64 parts, having a delicate index movable 
 with friction about the milled head. One revolution of the pinion, 
 as measured by the index on the graduated circle, raises the 
 pillar B E iS ths of an inch ; and since the circle is divided 
 into 64 parts, one division raises the pillar the one-hundreth of an 
 inch, or very nearly. Besides this circular measure there is a 
 small movable scale, graduated into twentieths of an inch, attached 
 to the lower tube H of the column, and acted upon by a tan- 
 
ELECTRICAL ATTRACTION OF SPHERES. 133 
 
 gent screw, t, so as to accurately adjust the zero-point of the scale 
 to the upper edge of the lowest tube B of the column. In order* 
 to steady the scale-pan s with its counterpoise, there is a light 
 tube, sliding with friction on the tube G, carrying a small project- 
 ing circular plate on which the scale-pan rests. 
 
 154. It is easy to comprehend by the following instructive 
 experiments, the nature and action of this instrument. 
 
 Exp. 40. Let it be required, for example, to determine under 
 given conditions the attractive force between two equal spheres, 
 p, N, Fig. 91, of a given diameter, suppose 2 inches, each weighing 
 
 Fig. 91. Electrical attraction of spheres. 
 
 150 grains. The sphere N is delicately suspended from the beam 
 immediately over the sphere P by a fine, double, gold thread ; 
 the attracting sphere p resting on the conducting tube q, at the 
 extremity of the insulating rod i. The sphere N being accurately 
 balanced in the way already described (153), the attracting sphere 
 p immediately opposed to it is then put into communication with 
 an electrical jar of given dimensions, exposing suppose about 
 three square feet of coated surface. The jar is then charged 
 from the prime conductor of the electrical machine through 
 a unit measure (121). When the accumulation amounts to 
 a given number of units or measures of charge, tlie two spheres 
 
134 ATTRACTION OF SPHERES. 
 
 attract, and the sphere N descends upon the sphere P. The force 
 of this attraction is estimated by grain weights placed in the 
 scale-pan T. Thus the electrical attractive force between the 
 spheres with a given quantity of electricity is estimated in 
 a known standard of weight. Suppose, for example, at the in- 
 stant 10 units have been accumulated, the sphere N has descended 
 on the sphere p, and raised a weight of 12 grains placed in the 
 scale-pan T. In this case the attraction between the spheres p, N, 
 at a given distance from each other, and measured between their 
 near points, may be considered as amounting to 12 grains. It 
 is however necessary to accurately determine this distance, which 
 is arrived at by first bringing the touching, or near, points of the 
 two spheres into contact, and causing the spheres to rest freely 
 on each other, by means of the upper rack and pinion. They are 
 then separated by a given distance, as measured by the vertical 
 scale or the graduated circle of the lower rack. When the distance 
 of the surfaces is for example, -5 of an inch, the quantity of 
 charge being 10 units, suppose the attractive force between the 
 two spheres to be equivalent to a weight of 12 grains in the 
 scale pan T. 
 
 Exp. 41. Let the diameters of the spheres remain the same as 
 at first, that is to say, 2 inches, the quantity of electricity (that 
 is to say, 10 units), and the distance being also the same. Then 
 as we have seen, (Exp. 40), the force of attraction is 12 grains 
 placed in the scale-pan T. 
 
 Let the quantity of electricity be now doubled, that is to say 
 20 units instead of 10 ; the attractive force, as measured by the 
 number of grains placed in the scale-pan T, will in this case be four 
 times as great as with half the quantity ; that is to say, the weight 
 in the scale pan T will now be 48 grains instead of 12, as in the 
 former case. 
 
 Thus whilst the quantities are as 2 : 1 the forces are as 4:1, or 
 as the squares of the quantities. 
 
 Exp. 42. Increase the distance between the plates to 1 inch, 
 or double the former distance, the quantity of electricity remaining 
 as in Exp. 40 that is to say, 10 units, which is equivalent to a 
 weight of 12 grains ; the attractive force decreases to 3 grains, or 
 one-fourth the former weight. We have in this case (as deter- 
 mined by this instrument,) the distances as 1 : 2, and the attractive 
 forces as 4 : 1, in which case the attractive force varies in an inverse 
 ratio of the square of the distance a long sought and important 
 problem. 
 
 Exp. 43. Let the diameters of the two spheres p, N, be now 
 
Bl FILAR BALANCE. 135 
 
 doubled, that is to say, 4 inches instead of 2, as at first. The areas 
 of the attracting surfaces opposed to each other will, in this case, 
 be four times as great, the surfaces of spheres being as the squares 
 of their diameters. Let the suspended sphere N of 4 inches 
 diameter, and which weighs about 350 grains, be balanced by an 
 equivalent weight placed in the scale-pan s. Let the number 
 of unit accumulations remain as in the preceding experiment (42), 
 that is to say, 10 units, the distance between the near points 
 of the hemispheres being adjusted as before (Exp. 40) ; that is to 
 say, to -5 of an inch. In this case the attractive forces between 
 the spheres P, N, of double diameter, will now be only one- 
 fourth as great as between those of half the diameter; that is 
 to say, the attractive forces will be inversely as the areas of 
 the opposed spheres, or as the squares of their diameters. Hence 
 the attractive force between the spheres of double diameter will 
 now only raise 3 grains, placed in the scale-pan T, instead of 12. 
 In this way the attractive forces between spheres, plates, and 
 other bodies opposed to each other under certain conditions, may 
 be accurately weighed, and estimated (S). 
 
 155. Harris's Two-threaded, or Bifilar Balance. There are no 
 departments of science which call more for the perfection of 
 quantitative measurement, and a clear perception of what we 
 really measure, than those of Electricity and Magnetism. If 
 we except the valuable researches of Professor Robison, of the 
 French physicist, Coulombe, the recent investigations of Dr. Fara- 
 day, and the experiments of the electrician, Brook, of Norwich, 
 we can scarcely be said to possess, in common electricity at least, 
 any connected series of experiments with electrical repulsive 
 forces, carrying with them a rigid numerical value. In the 
 various inquiries which philosophers have instituted into the 
 elementary laws of electricity, it has been their endeavour to 
 perfect our methods of electrical research, whether relating to 
 the quantity of electricity in action, intensity, inductive power, 
 or other elements requiring an exact numerical value, and by 
 operating with large statical forces, both attractive and repulsive, 
 to avoid many sources of error inseparable from the employment 
 of minute and almost insensible agencies. Coulombe' s balance of 
 torsion, and other delicate instruments of repulsion, have been 
 employed with this view. The Bifilar, or Double-threaded Balance 
 (invented by the author in the year 1831, and described in the 
 Transactions of the Royal Society for the year 1836) has been 
 resorted to with much advantage as a measure of repulsive force 
 in electrical and magnetic researches. 
 
136 
 
 HARRIS S TWO-THREADED 
 
 156. This instrument is constructed on the following principles. 
 If a delicate electrical needle, M N (Fig. 92), of moderate weight, 
 and about ten inches in length, be suspended from two fixed 
 points a a' by two similar filaments of unspun silk a b, a' b', 
 without torsion or twist, placed parallel to each other at equal 
 distances from the axial line c c, it is evident that its position of 
 rest will be horizontal, and in a plane passing through the 
 needle. Whenever, therefore, we turn the suspended needle 
 about the imaginary axis c c, the lines of suspension a b, a' b' 
 tend to cross each other, or become deflected from the vertical, 
 so that the distance c c' will be shortened. We have hence a 
 reactive force derived from the weight of the suspended needle, 
 which becomes imparted, as it were, to the threads of suspension 
 
 I*' 
 
 Fig. 
 
 !. Bifilar Suspension 
 of Cylinder. 
 
 Fig. 92. Bifilar Suspension of Needle. 
 
 a b, a' b' } since the centre of gravity of the mass will again tend to 
 rest in its previous position, and the suspended needle will be simi- 
 larly circumstanced to that of a body falling down a very small 
 circular arc. If, therefore, the needle be freely abandoned to this 
 reactive force, a vibratory oscillation of the needle ensues, by 
 observing which we may determine, by the formulae for oscil- 
 lating bodies, the reactive force of the bifilar threads producing 
 the oscillations. 
 
 157. With the view of determining this reactive force, a solid 
 cylinder of wood, p (Fig. 93), 2 inches high, and 2 inches diameter, 
 is suspended by two parallel filaments of unspun silk, a b, a b', in- 
 serted in the points b b' in a diameter of the upper surface of the 
 cylinder p at equal distances from its centre, c, and kept apart by 
 light, small, intermediate cross stays of cork, s. An index of light 
 reed projects from the cylinder for the purpose of allowing observa- 
 tions to be made on a graduated circular card, c D, as to the dura- 
 tion and extent of the oscillations produced by the crossing of 
 
OR Bl FILAR BALANCE. 
 
 137 
 
 the threads on each other. The threads a b, d b' (Fig. 93), are 
 sustained in a convenient light mahogany frame (Fig. 94), 
 mounted on a solid base, A A'. The suspension threads a b t d b' 
 pass through fine holes in a movable cross-bar, f r, Fig. 94, and 
 are joined above to a strong 
 piece of silk thread continued 
 through holes in a second bar, 
 d e, and are finally wound round 
 regulating cylindrical winders 
 at d and e. By these means 
 the respective lengths and dis^ 
 tances of the threads of oscilla- 
 tion a b, a b', are regulated, and 
 the cylindrical weight p caused 
 to hang parallel with the plane 
 of the graduated card c D. By 
 varying the situation of the 
 bar r r', the length of the 
 threads of oscillation may be 
 changed, and by a succession 
 of fine holes in the cylindrical 
 weight P, corresponding to 
 similar holes through the bar 
 r r', we may vary their dis- 
 tances apart. 
 
 158. By these mechanical arrangements, and the oscillations of 
 the cylinder, it is shown that the reactive force imparted to the 
 bifilar threads of suspension in turning the cylinder upon its 
 vertical axis, and allowing it to oscillate, is directly as the dis- 
 tances between the threads, inversely as their lengths, directly 
 proportionate to the weight of the suspended body, and as the 
 angle of twist or torsion of the bifilar threads upon each other ; 
 results similar to those arrived at by the celebrated Coulombe 
 in determining the reactive force of the fine wire employed by 
 him in his beautiful instrument the Balance of Torsion (166). 
 
 159. Having obtained a given length and distance between the 
 threads, the centre of the cylinder p (Fig. 94) is caused to hang 
 immediately over the centre of the graduated circle c D, as shown 
 by contact with the finely-pointed extremity of a short cylindrical 
 rod, passing from beneath upwards through friction corks in the 
 central block B B'. Small stays s, s', s'', s'", of light reed or 
 cork, are inserted between the threads at given distances, in order 
 to prevent them from closing upon each other. 
 
 Fig. 94. Bifilar Balance. 
 
138 THE TWO-THREADED 
 
 The index I turned aside to an angle of 60, being the sus- 
 pended body P allowed to oscillate freely. In order to preserve 
 a very free oscillation, the fine central point is depressed from 
 beneath the base and block B B'. 
 
 By carefully noting the rate of oscillation, the following results 
 are immediately arrived at : 
 
 The time of an oscillation is as the square root of the length of 
 the threads of suspension, divided by their distance apart, and 
 is altogether independent of the weight of the oscillating body. 
 
 The oscillations are isochronous at all angles. 
 
 From these results we may, by the general formula for oscillating 
 
 bodies n = %- employed by Coulombe, in his experiments 
 
 tu Cf T 
 
 on the torsion of wires, easily deduce the laws of the reactive 
 force imparted to the threads. 
 
 In this formula n is the force, in terms of a unit of weight 
 = 1 grain, which applied perpendicularly at the extremity of 
 a lever of a unit of length = 1 inch will resist the reactive force 
 imparted to the threads, when the suspended body has been turned 
 about its axis through an arc of 60, whose chord is equal to the 
 unit of length, = radius, = 1 inch. P is the weight of the 
 cylinder in terms of the unit of weight, = 1 grain ; TT the ratio 
 of the circumference to the diameter of the circle ; a the radius of 
 the cylinder or circle = 1 inch ; g the force of gravity, = 386 
 inches, equal to the distance through which a body falls in free 
 space in a second of time ; T the time of an oscillation in terms 
 of a unit of time = 1 second. 
 
 In applying this formula, it is easy to perceive that the value of 
 n will vary with the squares of the distances between the threads 
 of oscillation divided by their lengths, and that, contrary to the 
 law of torsion, as deduced by Coulombe, is as the weight of the 
 
 p d 2 
 cylinder p, hence we have n cc - ; and since the oscillations 
 
 of the cylinder are, as just stated, isochronous at all angles, 
 we may conclude that n is also proportional to the angle of 
 deflection of the threads. 
 
 160. These results are verified mechanically and experimentally 
 in the following manner : - 
 
 A weight placed in a light paper scale-pan, / (Fig. 94), and 
 carefully counterpoised, is caused to act tangentially to the 
 circumference of the cylinder p, by means of a slender fila- 
 ment of unspun silk, f p P, passed round the cylinder and 
 led over an extremely delicate watch-pulley p. This pulley is 
 
OR B1F1LAR BALANCE. 
 
 139 
 
 attached to a slide and socket, u, fixed to a circular rim, u D, 
 movable about the interior block carrying the graduated card 
 c D, hence the line / p p may be exactly set at right angles 
 to the radius of deflection in all positions of the index I, and 
 the precise weight determined requisite to balance the reactive 
 force of the threads at any given angle, or otherwise, by turning 
 the whole frame of suspension d s'" y in a circular socket formed 
 in the transverse piece A A' through any number of degrees, 
 as measured by a graduated card and index n x, and preserving 
 always the index i of the oscillating cylinder p at zero, we arrive at 
 the reactive force of the threads of suspension at any required 
 angle. 
 
 The results of a series of experiments conducted in this way, 
 completely verified the above deductions (158), the weight requisite 
 to maintain the index at an angle of 60 being as the weight of the 
 cylinder P multiplied into the squares of the distances between the 
 filaments of the suspension threads, divided by their length. It 
 was also found to vary with the angle of repulsion, as determined 
 by Coulombe. 
 
 161. The following tables, abridged from a greater number of 
 experiments than it is desirable to mention here, afford a sufficiently 
 practical evidence of the truth of these results. In these tables 
 the unit of weight is 1 grain, the unit of length 1 inch, the unit 
 of time 1 second. 
 
 TABLE I. 
 
 Showing the rate of oscillation with different lengths and dis- 
 tances of the threads. 
 
 Weight of Suspended Body, 960 grains. Angle of Oscillation 45. 
 
 Length. 
 
 Distance. 
 
 Oscillations 
 in 60". 
 
 Time of ten 
 Oscillations by 
 observation. 
 
 Time of 
 one 
 Oscillation. 
 
 6 
 
 24 
 
 | 0.25 ( 
 
 28-50 
 14-25 
 
 21 
 
 42 
 
 2-1 
 4-2 
 
 6 
 
 \ ( 
 
 46 
 
 13-1 
 
 1-31 
 
 12 
 
 0-4 
 
 32-50 
 
 18-5 
 
 1-85 
 
 24 
 
 ) ( 
 
 23 
 
 26-2 
 
 2-62 
 
 24 
 
 0-8 
 
 46 
 
 13-1 
 
 1-31 
 
 Similar results were obtained when the angle of oscillation was 
 increased to 180 and upwards, as also when the weight of the 
 cylinder p was varied from 960 to 480, and 240 grains respectively, 
 the radius being in each constant. 
 
140 
 
 HARRIS'S TWO-THREADED 
 
 The rate of oscillation was taken with a valuable chronometer, 
 by which portions of time so little as the one-sixtieth part of a 
 second could be well estimated. 
 
 TABLE II. 
 
 Showing experimentally the weight in grains requisite to resist 
 the reactive force of the threads at an angle of 60, their length 
 and distance apart being varied, as also the altitude of the 
 cylinder p. 
 
 Length. 
 
 Distance. 
 
 Weight in grains on a lever of 1 inch. 
 
 = 960 grains- 
 
 P=480 grains. 
 
 P=240 grains- 
 
 6 
 24 
 
 } ' 25 { 
 
 2-675 
 0-67 + 
 
 1-325 
 0-325 
 
 0-66 
 0-15 
 
 6 
 12 
 24 
 
 ],, | 
 
 ?' + 
 3-55 
 1-75 
 
 3-525 
 1-775 
 
 0-875 
 
 1-75 
 0-885 
 0-425 
 
 24 
 
 0-8 
 
 6-85 
 
 3-425 
 
 1-750 
 
 The smaller weights employed in these experiments could not 
 be considered as mathematically exact; they were, however, 
 sufficiently accurate for the purposes required. They consisted of 
 lOths of grains, 20ths, 40ths, and lOOths. The numerical 
 values in the above table are those which resulted from the 
 positition of the index, so far as these small weights could deter- 
 mine ; and it will be seen that the approximation to the values 
 deducible from the preceding Table I., by means of the formula 
 
 2 2 
 
 , are as near as could be expected from such an 
 
 n = 
 
 2 gl 
 experiment. 
 
 TABLE III. 
 
 Showing the weight in grains by calculation and experiment, 
 required to balance the reactive force of the threads at various 
 angles of deflection, from to 300, the threads being 24 inches 
 in length, and -25 apart, and prevented from collapsing by small 
 stays of cork (159), inserted between them at equal distances. 
 
 Weight of Cylinder P=960 grains. 
 
 Angle of Deflection 
 
 10 
 
 20 
 
 30 
 
 60 
 
 90 
 
 100 
 
 120 
 
 150 
 
 180 
 
 200 
 
 240 
 
 270 
 
 300 
 
 Force by Formula 
 
 115 
 
 23 
 
 34 
 
 69 
 
 67+ 
 
 1-03 
 
 1-15 
 
 1-38 
 
 1-72 
 
 2-07 
 
 2-3 
 
 2-76 
 
 3-1 
 
 3-45 
 
 Force by Experiment 
 
 11+ 
 
 22 
 
 34- 
 
 1- 
 
 1-1 
 
 1-35 
 
 1-7 
 
 2- 
 
 2-25 
 
 2-725 
 
 3'+ 
 
 3-425 
 
 The pulley p (Fig. 94) employed in these experiments is ex 
 
OR BIFILAR BALANCE. 
 
 141 
 
 tremely delicate. The small scale / in which the weights were 
 placed, weighed the *1 of a grain, and was suspended by a filament 
 of a thread of the silkworm ; that part of it passing over the pulley 
 being particularly slender and flexible. The -01 of a grain was 
 in this way rendered very sensible on the index. The approxi- 
 mations in this table are sufficiently close to show that the reactive 
 force of the threads is as the angle of repulsion or deflection of 
 the needle ; the threads being prevented from collapsing by inter- 
 mediate stays. 
 
 162. Fig. 95 represents the bifilar balance in its most perfect 
 form. A B is a flat, polished, mahogany base, about 16 inches 
 square and an inch thick. A, B, c, D are four light grooved maho- 
 gany pillars, each 10 inches 
 high and half an inch in 
 diameter, screwed into the 
 angles of the base. The 
 grooved pillars A, B, c, D re- 
 ceive four glass panes, constitut- 
 ing, together with the pillars, 
 a sort of light glass cage, 
 within which is suspended by 
 bifilar threads, a b, a b', (157), 
 a light insulating needle M N 
 (156). Each extremity of the 
 needle carries a small circular 
 gilt disc, M N, four-tenths of 
 an inch in diameter, attached 
 to the needle by an insulating 
 thread of shell-lac. A stout 
 
 vertical metallic axis, H, pass- 
 
 ing with friction through a " r T 
 
 central cork, firmly fixed in 
 
 the base A B, carries a circular disc or movable base of light 
 wood, o x, 10 inches in diameter, and a quarter of an inch 
 thick, firmly united, centrally, to the extremity of the axis 
 H, so that by turning a milled head, h, beneath at the lower 
 extremity of the axis, the circular base o a? can be moved round. 
 To facilitate this, movement upon its square base of support, 
 through which the central axis of motion passes, very fine rollers 
 are delicately inserted in the under surface of the movable base o x. 
 Two small gilt discs, p Q, similar to those at the extremity of 
 the needle, are fixed on delicate sliding insulators (Chap. I., 
 Note E), sustained in small feet inserted in opposite points 
 
142 
 
 of a diameter of the movable base o x, .immediately opposite 
 and nearly to the discs M, N of the needle, so as to admit of 
 one of the discs M or N, at the extremity of the needle, coming 
 very nearly to and parallel with one of the discs p or Q at the 
 extremity of the sliding insulator. The bifilar threads a b and 
 a' b' are sustained in a convenient, light, open, metallic frame, e t', 
 such as already described (157). This light frame- work is con- 
 structed of two small, parallel, vertical, metallic tubes a t, a' t', 
 14 inches long, fixed in a firm circular metallic base, movable, 
 centrally, in a socket within a light elevated cross-piece, y x. 
 Two light metallic bars e f, g h, attached to sliders movable on 
 the vertical tubes a t, w i, unite the vertical side-pieces of the 
 frame e t' with each other. The bifilar threads a b, a b', pass 
 through fine holes in thin metallic plates, connected with the 
 bar g h. Thus, by means of the fine holes in the cross-bar g h, 
 and the sliders on the side-pieces a t and a' t, the lengths of the 
 bifilar threads of suspension and their distances apart may be regu- 
 lated to any required extent, in the way already explained (157). 
 The suspension threads a b, a' b', are attached to short lengths of 
 stout silk thread, passing round cylindrical winders in the bar ef 
 above, and terminate in the needle suspension below. The needle 
 is suspended by means of very fine wire hooks b b' (Fig. 96), 
 
 united to short sliding- pieces 
 of small silvered brass tube. 
 These sliders are movable 
 upon a short length of brass 
 tube carrying the needle, and 
 by which it is sustained. A 
 descending vertical tubular 
 rod, u v (Fig. 95), 6 inches 
 long, and one-eighth of an 
 Fig. 96. Mode of Suspending Needle. inch in diameter, is attached 
 
 to the small tube b b' carrying the needle, and extends to the 
 centre of the under surface of the movable circular base o x y where 
 it is steadied by a fine needle-point, in which it terminates, and 
 corresponds with a fine hole drilled in the termination of an 
 interior axis h, passing through the axis H of the movable base, 
 in such a way as to admit of the two axes moving one within the 
 other, after the manner of the hands of a watch. 
 
 Thus, whilst the needle is accurately centred, it is prevented 
 from acquiring any undue lateral motion. The axis h carries a 
 forked lever for arresting and regulating the position of the 
 needle through a light index, i, of reed at the lower extremity of 
 
OR BIFILAR BALANCE. 143 
 
 the vertical tubular rod u v proceeding from the central tube b b' of 
 the needle. 
 
 A fixed graduated circle surrounds the movable base o x y 
 within which the circular base moves. This circle is divided into 
 360 degrees. The long index i at the lower extremity of the 
 descending tubular rod u v attached to the centre of the needle, 
 moves over the graduated circle surrounding the movable base 
 o x, and indicates in degrees the angular deflection of the needle 
 by the repulsive force between the discs M or N and p or Q when 
 brought near each other and electrically charged with the least 
 possible quantity. The instrument indicates an extremely small 
 charge communicated to the opposed discs. 
 
 By turning the circular basis on which the supporting frame e t' 
 of the bifilar threads rest, we are enabled to deflect the needle 
 from its position of rest any number of degrees. Hence the 
 number of degrees required to reverse the torsion is a measure of 
 the repulsion that is to say, a measure of the turn of the bifilar 
 threads upon each other. To effect this with precision, a gra- 
 duated circle, divided into 360 degrees, is fixed on the upper sur- 
 face of the circular base t t', together with a fixed line or index 
 to denote the zero point. By this arrangement the instrument 
 with the needle is under perfect command, and the index I easily 
 adjusted to the zero points of the graduated circle o x. 
 
 As it is often convenient to pass the fixed insulated discs p, Q 
 downward through the cover c D of the cage, the insulators, carry- 
 ing the discs, slide through fine cork, closely fitted in capa- 
 cious holes in one of the bars of the cover movable under the 
 bridge y z, so as to admit of the fixed discs descending within 
 the cage ; and thus one of the fixed discs p or Q, and one of the 
 discs M or N of the needle may be placed parallel and very nearly 
 to each other by turning the movable circle y z, or movable base 
 o x. We may in this way operate upon the repelling discs either 
 above, through the cover of the cage, or beneath, by means of the 
 movable circular base o x, the distance between the repelling 
 discs being measured in degrees of a circle of 1 foot in diameter 
 divided into 360 parts. The insulators carrying the fixed discs 
 p, Q, consist of strong vulcanite or other insulating rods, about 
 6 inches long, and the sixteenth of an inch in diameter. 
 
 163. One or two simple experimental illustrations will be suffi- 
 cient in explanation of the operation of this instrument the Two- 
 threaded or Bifilar Balance. 
 
 Exp. 44. The required lengths and distances of the bifilar 
 threads being duly adjusted, together with the horizontal position 
 
144 EXPERIMENTAL ILLUSTRATIONS. 
 
 of the needle, by means of the arrangements just described (162), 
 bring one of the insulated discs p or Q parallel and very nearly to 
 one of the discs M or N of the needle, which may be done either 
 by the movable circle y z of the cage above, with which the fixed 
 discs P and Q are very frequently connected, or by means of the 
 movable circular base below, in which the fixed discs p and Q are 
 commonly inserted. Charge either of the fixed discs p or Q with 
 a given quantity of electricity by means of a transfer plate and 
 quantity jar (144), suppose 1 circular inch (145). The disc M or 
 N of the needle will be immediately repelled by the charged disc p or 
 Q to a given distance (27), suppose 10 degrees, as indicated by the 
 index i, previously set to zero of the circle o x. 
 
 In this case the indicated arc of 10 degrees is very little dif- 
 ferent from its chord, so that the distance of the fixed disc P or Q 
 and the disc of the needle M or N may be correctly represented by 
 10 degrees. Hence we have distance of repulsion 10 degrees, and 
 repulsion 10 degrees, quantity of electricity being 1 circular inch. 
 
 JExp. 45. The quantity of electricity remaining the same that 
 is, 1 circular inch bring the two repulsive discs M or N and p or Q 
 within 5 degrees of each other, or half the former distance. The 
 repulsion, as indicated by the index i, will rapidly increase as the 
 distance between the discs decreases ; so that the index i when 
 the discs are one-half the previous distance apart, that is to say, 
 5 degrees instead of 10 will indicate a repulsion of 40 degrees, 
 or four times the former. 
 
 Thus, whilst the distances are as 1 : ^, the repulsions, as measured 
 in degrees, are as 1^ : 2^, that is, as 1:4, clearly showing, as in 
 previous instances of attraction (154), that the repulsion varies in 
 an inverse duplicate ratio of the distances between the repelling 
 surfaces. We arrive at the actual reactive force of the threads upon 
 the needle by some such means as those referred to (162),' as for 
 example, by turning the movable basis (on which the frame sup- 
 porting the bifilar threads rests) in a direction contrary to the 
 twist of the threads, and observing how many degrees we 
 require to turn the circle in a reverse direction to bring the 
 needle back to its primitive position. 
 
 Exp. 46. The fixed disc P or Q, and one of the discs M or N 
 of the needle being previously set parallel and very nearly to 
 each other, as in the former experiment indicating a force of 
 10 degrees, charge the fixed disc p or Q with 2 circular inches, 
 or twice the former quantity, by means of the transfer plate and 
 quantity jar (144). The index i will now indicate 20 degrees, or 
 twice the former repulsion of 10 degrees. 
 
BALANCE OF TORSION. 145 
 
 Thus the distances of repulsion are in this case directly as the 
 quantities of electricity in action. 
 
 164. "We find by these experiments : 1st. That electrical repul- 
 sion varies in an inverse duplicate ratio of the distances between 
 the repelling surfaces. 2nd. That with a double quantity the 
 repulsion extends to a double distance, and is directly as the 
 quantity. 
 
 165. It may perhaps be as well to observe that in Exp. 45, 
 whilst the quantity remains constant, the distances vary; and 
 in Exp. 46, whilst the distance remains constant, the quantities 
 vary (T). 
 
 166. The Balance of Torsion. If a cylindrical weight, w 
 (Fig. 97), be suspended by a fine, elastic, metallic wire, or silk line, 
 w e, from a fixed point, w, so that the axis of 
 
 suspension be immediately under the fixed 
 point, and the cylindrical weight and wire be 
 turned round a little upon its point of suspen- 
 sion, the wire will be somewhat twisted, and will 
 have acquired an elastic force, which causes it, 
 together with the suspended weight, to tend to 
 regain its previous position, and if set free, the 
 weight together with the wire will assume a 
 state of vibration or oscillation, 
 
 The elastic force set up in the wire w e has rig. 97. couiombe's Tor- 
 
 , 1 .7 / // / , T> sion Apparatus. 
 
 been termed the jorce oj torsion, or twist. Joy 
 carefully observing the duration of a certain number of oscilla- 
 tions of the suspended weight and wire, we may determine, by 
 means of the formula for oscillating bodies, given by Coulombe 
 in his celebrated memoir on the force of torsion, in the Proceedings 
 of the Royal Academy of Sciences of Paris, the reactive force pro- 
 ducing the oscillations. 
 
 The simple apparatus employed by Coulombe is represented in 
 Fig. 97. The weight w, suspended by the metallic wire or silk 
 line w e, carries an index, i, which marks on a graduated card, i, 
 the duration and extent of the oscillations. By varying the weight 
 of the suspended body w, the length of the wire of suspension, its 
 tension and thickness, and the kind of metal of which it consists, 
 we may determine the laws of the reactive force of the torsion of 
 the wire, in reference to the elements upon which the laws of 
 torsion depend. 
 
 167. To determine the reactive force of a wire thus subjected 
 to torsion, different weights are suspended from a fixed point by 
 metallic wires of various kinds, lengths, and thicknesses. 
 
 L 
 
146 
 
 BALANCE OF TORSION. 
 
 The following are the general results of Coulombe's physical 
 investigations : 
 
 The reactive force of the twist or torsion of a wire is as its length, 
 thickness, and tension. 
 
 The force of torsion of a wire is as the angle of torsion that is 
 to say, the angle at which the wire is twisted or turned upon 
 itself. 
 
 Upon these elements of force Coulombe constructed his beauti- 
 ful Balance of Torsion for estimating small physical forces, more 
 especially those of electricity and magnetism. 
 
 168. The balance of torsion (Fig. 98) consists of a light 
 needle, m n, 6 inches long, suspended 
 by a fine wire, w e, within a cylindrical 
 cage, A B, 12 inches in diameter, and 12 
 inches in height ; it is covered by a circu- 
 lar plate of glass, A c, 13 inches in dia- 
 meter. The centre of the glass plate 
 is pierced by a hole, .H, 2 inches in dia- 
 meter, to receive a vertical glass tube, 
 T H, through which passes, centrally, the 
 fine metallic wire, w e. The needle 
 consists of three separate parts, m, e, n 
 a central part, e, terminating in two stiff 
 threads of gum- lac, m, n. 
 
 At one of the gum-lac extremities, m y 
 is a light pith ball, about half an inch in 
 diameter, weighing about a grain and a 
 half. The opposite gum-lac extremity, n, 
 carries a circular gilt paper disc, about 
 1| inch in diameter, saturated with 
 turpentine, which, whilst it tends to 
 Fig. 98. Balance of Torsion. counterpoise the pith-ball m, restrains 
 the oscillation of the needle. 
 
 Directly an electrified body approaches the pith-ball m at the 
 gum-lac extremity of the needle m n similarly electrified, the 
 needle is repelled, and turns the wire suspending the needle so 
 as to impress upon it, from its fixed point, torsion or twist. It is 
 the reactive force of this torsion which was employed by Coulombe 
 to measure the force of the repulsion of the needle. 
 
 The torsion wire w e, carrying the needle m n, is subjected to 
 the strain or tension of a weight appended to it, just less than 
 sufficient to break it. The circular glass plate A c closing the 
 cage, A B, has another hole, li y through it, about an inch in 
 
BALANCE OF TORSION. 147 
 
 diameter for the convenience of passing any particular substance 
 into the cage. A second pith-ball, v, at the extremity of a stiff 
 thread of gum-lac, similarly electrified to the ball m of the needle, 
 is passed through the hole h into the cage A B so as to act upon 
 and be exactly opposite the ball m of the needle. 
 
 The upper extremity of the glass tube T H receives, after the 
 manner of a telescopic joint, a short, polished, metallic tube, about 
 2 inches in length, on which is fixed a circular silvered plate, i. 
 
 The edge of this circular plate is divided and graduated into 
 360 degrees, having a zero point, from which the divisions of 
 the circle are estimated. 
 
 The torsion wire w e is suspended through the medium of delicate 
 pincers, fixed in the under surface of the plate i, closing the upper 
 extremity of the tube T H. A graduated circle, E F, also divided into 
 360 degrees, surrounds the cage A B at the level of the needle. 
 The position of the graduated circle is such that its zero commences 
 from the point opposite the fixed pith-ball v. Since the tube T, 
 terminating the tube T H and carrying the plate i, turns with a 
 little friction within the tube T H, the zero of the graduated plate i 
 on the tube may be brought into any required position relatively 
 to the divisions of the graduated circle E F, surrounding the cage 
 A B below ; and as the pincers holding the torsion wire w e pro- 
 ceed from the under surface of the plate i, the pincers holding 
 the wire 10 e and the wire itself, together with the suspended needle, 
 must also turn with it. Thus, the wire we becomes subjected to 
 torsion to a given extent. 
 
 169. If the centre of the pith-ball m of the needle be supposed to 
 correspond precisely with the zero of the circle E F surrounding the 
 cage before the introduction of the pith-ball v ; and if the centre 
 of the fixed pith-ball v, when it is introduced, be made to take the 
 place of the centre of the pith-ball m of the needle, it will displace 
 the ball m, and therefore turn it aside through a space equal to 
 the sum of the radii of the two balls. If either of the two balls 
 thus in contact be electrified, the electricity will be shared between 
 them, and they will repel each other (27), but as the ball v 
 is fixed, the repulsion will only take effect upon the ball m, which 
 will separate from v, and cause the needle m n to move from v ; 
 m will continue thus to depart from v, twisting the wire as it turns, 
 until the reaction of the torsion of the wire balances the repulsive 
 force, m will then remain at rest at a distance from v, indicated 
 by the division of the circle E F, surrounding the cage, oppo- 
 site to which it stands. The number of that division may be 
 taken to represent the angle of torsion of the suspending wire, 
 
 L2 
 
148 BALANCE OF TORSION. 
 
 and this angle is always proportional to the reactive force of 
 the torsion, or the force with which the torsion wire w e endea- 
 vours to recover its position of rest. 
 
 In this manner the force of repulsion at different distances may 
 be measured and observed. 
 
 170. Such was the elasticity of the wire employed by Coulombe 
 that after having been subjected to the torsion of many circles it 
 accurately returned to its first position. 
 
 By turning the plate I, suspending the torsion wire to e and the 
 needle m n, the ball m of the needle may be brought opposite the 
 zero of the graduated circle E r, when it will be immediately under 
 the hole h, through which the pith-ball v and its gum-lac insu- 
 lator is passed into the cage A B, so as to be directly opposed to 
 the pith-ball m of the needle. 
 
 171. The extreme degree of sensibility of Coulombe's Balance 
 of Torsion, and the infinitesimal quantities it is capable of measure- 
 ing may be collected from the consideration of the dimensions of the 
 torsion wire employed. In the experiments of .Coulombe (174), by 
 which the laws of electrical attraction and repulsion were investi- 
 gated, the suspending wire was 28 inches in length, and was so 
 fine that one foot of its length weighed only the one- sixteenth part 
 of a grain. This, however, was found too fragile for actual experi- 
 ment. The radius of a circle described by the ball m of the needle 
 was four inches. The elasticity of the wire employed by Coulombe 
 was so perfect that when turned through one entire revolution, or 
 360, its reactive force amounted to no more than the 340th part 
 of a grain, and since Coulombe found that the reactive force 
 of torsion is proportional to the angle of torsion (166), the force 
 corresponding to the motion of the ball m through one degree of 
 the circle was only the 122,400th part of a grain ; thus dividing 
 a single grain into 122,400 parts, and rendering each part dis- 
 tinctly visible. Coulombe, however, as we have already remarked, 
 found it convenient to employ wires of less sensibility. The sensi- 
 bility of a torsion wire was found by Coulombe to increase with the 
 length of the wire, all other things remaining the same, and to in- 
 crease inversely as the fourth power of its diameter, the length being 
 constant ; so that the sensibility of the balance of torsion is in the 
 direct proportion of the length of the torsion wire, and in the 
 inverse proportion of the fourth power of its diameter. If, for 
 example, the length of the wire be doubled, the sensibility of the 
 instrument will be doubled ; and if the diameter of the wire be 
 reduced in the proportion of two to one, its length being the same, 
 the sensibility of the instrument will be sixteen times as great ; 
 
EXPERIMENTAL ILLUSTRATIONS. 149 
 
 that is to say, the reactive force will be one-sixteenth greater with 
 a wire of one-half the diameter. 
 
 172. To communicate electricity to the fixed ball v of the 
 balance we employ a transfer of given dimensions (145), according 
 to the quantity of electricity we wish to impart, the transfer being 
 introduced into the cage through the hole h in the cover. 
 
 Such is the sensibility of the torsion balance, that if, after having 
 excited a stick of sealing-wax and weakly electrified the ball m of 
 the needle, negatively, the excited wax be held at a distance of 3 
 feet from the electrified ball m, the ball will be repelled 90 degrees. 
 
 173. In the forces which are manifest between small particles 
 of matter, the first circumstance to be noticed is the fact that the 
 energy of the particles on each other is augmented in some propor- 
 tion as the distance between them is decreased. The analogies 
 suggested by physical forces generally, the intensities of which 
 increase with the diminution of distance, and more especially the 
 law of gravitation, by which the energy of that force increases 
 inversely as the square of the distance between the particles, 
 naturally leads to the question : according to what law does the 
 force of electrical attraction or repulsion increase or decrease as the 
 distance between the attracting or repelling bodies decreases or 
 increases ? 
 
 If the nature of electricity were perfectly known, this law could 
 be deduced by general reasoning, so that the manner in which 
 electrified bodies would comport themselves in any position in 
 which they might be respectively placed could be certainly fore- 
 told. But the physical principles from which electricity arises not 
 being known, the investigation of its laws and the establishment 
 of a just theory respecting its nature depends upon our knowledge 
 of observed phenomena. To determine, therefore, the law of 
 electrical action, it is necessary to submit electrified bodies to their 
 mutual attractions and repulsions at different distances, to measure 
 the actual amount of that attraction and repulsion at those dis- 
 tances, and, by comparing the results of such measurement with 
 the distances themselves, to discover the dependence of one upon 
 the other. This was effected by Coulombe, by the aid of his 
 Balance of Torsion. 
 
 174. The following are elegant experimental illustrations 
 originating with Coulombe in his early investigations in 
 electricity of the application of his torsion balance to physical 
 research. 
 
 Exp. 47. The centre of the pith-ball m, Fig. 98, corresponding 
 precisely with the zero of the circle E r, surrounding the cage, and 
 
150 HARRIS'S THERMO-ELECTROMETER. 
 
 the ball v being passed into the cage through the hole h, so as to be 
 in contact with the ball m when at zero of the circle E F, charge 
 the balls m, v, by means of a transfer plate of given dimensions 
 (145), with a quantity sufficient to repel the needle 36. 
 
 In this case the repulsion will be 36 and the torsion force 36. 
 
 Turn the plate i, closing the tube T H, 126 back against the 
 direction of the already existing torsion ; the effect of this will be 
 to impress an additional torsion of 126 on the wire of suspension 
 w e and to bring the balls m y v, within 18 of each other, or half the 
 former distance of 36. 
 
 We now have for the two torsions, 36 and 144. In this case 
 the distances of repulsion are as 36 : 18, that is as 2 : 1 ; whilst 
 the forces of repulsion corresponding to those distances are as 
 36 : 144, that is as 1 : 4 ; or, in other words, the electrical forces 
 are in an inverse ratio of the squares of the distances. 
 
 Exp. 48. Turn the plate i farther back against the repulsion 441, 
 this will impress on the torsion wire w e an additional torsion of 441; 
 This will bring the balls m y V, within 8^ or 9 of each other, or half 
 the former distance. 
 
 Comparing these three experiments, we have for the distances of 
 repulsion 36, 18, and 9, and the forces corresponding to those dis- 
 tances 36, 144, and 576. The forces, therefore, still continue to 
 vary in the inverse ratio of the squares of the distances, being in 
 accordance with the great general law of electrical action (U). 
 
 175. The Thermo- Electrometer. It was long since observed by 
 Mr. Children, that the heat evolved by metallic bodies in trans- 
 mitting an electrical charge is in some inverse ratio of their con- 
 ducting power, a principle generally admitted, not only as a 
 reasonable deduction, but also as being established by a great 
 variety of facts. 
 
 176. The effect of the electrical discharge on metallic bodies is 
 to raise their temperature to a greater or less degree, and in many 
 instances to render such bodies red-hot, and to dissipate them in 
 melted globules. 
 
 The fusion of wire has been hence frequently resorted to by the 
 older electricians (Van Marum, Cuthbertson, Singer, and others) as a 
 measure of electrical discharge. This method, however, has often 
 been found tedious and uncertain in its results, and, in many 
 instances, quite inapplicable to refined inquiry. Thus Cuthbertson, 
 in comparing the results of some experiments made by himself 
 on the fusion of wires by electricity with those of similar experi- 
 ments by Dr. Yan Marum, observes : 
 
 " The doctor might, perhaps, have been been led into a mistake 
 
THERMOELECTROMETER. 
 
 151 
 
 in the following manner. He might not have been aware of the 
 different degrees of ignition caused by electric discharges, but only 
 judged of the force, by the wires being converted into balls, by 
 which great mistakes will frequently arise ; for if a wire be taken 
 18 inches long, and of such a diameter that when a jar or battery is 
 charged to such a height as just to cause it to run into balls, much 
 shorter lengths of that same sort of wire may be subjected to the 
 same force, and still only be converted into balls. If only 7 inches 
 were taken, nothing but balls will appear ; the only difference will 
 be that the balls will be smaller and dispersed to a greater distance, 
 which might be easily overlooked. If 6 inches of the same sort of 
 wire be taken it will be converted into balls and flocculi, or brown 
 oxide of iron ; so that, to be accurate in this point, the lowest 
 degree of ignition must be had, which is known when the charge 
 has passed ; the wire will be red-hot the whole length, and after- 
 wards fall into balls "* a rather precarious condition. 
 
 177. The Thermo-Electrometer, whilst it avoids all destruction 
 of the metal, indicates, at the same time, the comparative heating 
 effect of the electrical discharge, and admits of an accurate estimate 
 of its force. 
 
 178. This instrument (Fig. 99) is of very simple construction, 
 being little more than an air-thermometer, having a fine metallic 
 wire, or wires, passed air-tight through 
 
 its ball. A glass tube, a b c, the interior 
 diameter of which is regular and some- 
 what less than one-tenth of an inch, 
 has one of its extremities, b c, bent up- 
 wards and outwards for about 2 inches, 
 and is united by welding to an elliptical 
 glass cup or reservoir, R, about 2J inches 
 in length, 2 inches in width and 
 depth; this reservoir contains a small 
 quantity of coloured spirit, such as is em- 
 ployed in the spirit thermometer. The 
 vertical part of the tube is sustained by a 
 graduated scale, a b, 2 feet in length, di- 
 vided into 150 degrees, each degree being 
 one-tenth of an inch. This scale is fixed upon a convenient 
 base, 1) c, and the point on it at the level of the coloured spirit 
 in the tube is marked zero. Upon the reservoir R is screwed a 
 glass ball of 5 inches diameter, having a metallic wire, WE, or 
 wires, passed air-tight across its centre. 
 
 * Cuthbertson's " Practical Electricity," pp. 185186. 
 
 Fig. 99. Thermo-Electrometer. 
 
152 CONSTRUCTION AND USE OF 
 
 The method of fixing the wire is easy. Two flanches of 
 brass, /, /' (Fig. 100), with shoulders and projecting screws, are 
 cemented in and over the holes drilled through the glass ball for 
 the passage of the wire ; the wire is passed directly through the 
 ball by means of corresponding small holes drilled through the 
 flanches, and, being gently put on the stretch, is secured by fine 
 
 wooden or metallic pegs, by which 
 the wire is slightly compressed in 
 the hole. When metallic pegs are 
 used, a small groove should be cut in 
 them for the passage of the wire. Both 
 the pegs and extremities of the wire 
 project a little for the convenience 
 of removal, and thus wires of various 
 kinds and of different diameters may 
 be easily substituted. The whole is 
 
 Fig. 100. Mode of Fixing Wire. P ,-, , , . . . , . , 
 
 finally rendered air-tignt by means 
 
 of small brass balls, B, B', flattened on one surface, and screwed 
 on the flanches against leather collars, so as to render the ball 
 air-tight. Beside these flanches, the ball is also furnished with 
 a valvular projection, v, attached in a similar way to its upper 
 part, which being rendered air-tight by a screw and leather can 
 be occasionally opened, so as to form a communication with the 
 external air, and thus the level of the coloured spirit in the reser- 
 voir R may at all times be adjusted to zero of its scale. 
 
 179. For the convenience of passing more than one wire through 
 the centre of the glass ball, there are additional holes, furnished with 
 brass flanches cemented to the glass, which are supplied with covering 
 balls in the way above described. We may thus employ several wires 
 of different metals, and cause an electrical charge to pass through 
 either of them, without interfering with the wires already in the 
 ball ; and thus obtain a Thermo-Electrometer of different degrees 
 of sensibility, according to the metallic wire employed. Under 
 these circumstances, when an. electrical charge of sufficient force 
 is passed through the wire or wires in the ball, the relative degree 
 of heat evolved is made evident by the ascent of the fluid along 
 the graduated scale a b, Fig. 99. 
 
 180. The application of the Thermo-Electrometer in promoting 
 electrical investigation is easily seen by a few, simple, direct, and 
 efficient experiments. 
 
 Exp. 49. Pass successively through the ball B of the Thermo- 
 Electrometer (Fig. 101) fine wires of different metals ; firmly fix 
 them in position, as already described (178). The wires should 
 
THE THERMO- ELECTROMETER. 
 
 153 
 
 be drawn down through successive holes in a hardened steel plate 
 to about the one-eightieth of an inch in diameter, and consist of 
 silver, copper, gold, zinc, platinum, iron, tin, and lead. 
 
 Charge an electrical jar, A (Fig. 101), exposing about 4 square 
 feet of coated surface (113), with a given number of units (119), 
 through a unit measure, u, and having completed the required 
 metallic communications, discharge the given accumulation by means 
 of a charging and discharging quantitative electrometer E (139), 
 
 Fig. 101. Mode of Using the Thermo-Electrometer. 
 
 through the particular wire under examination. Observe the heating 
 effect of each metal, by the ascent of the fluid along the scale. 
 We find for copper, 6 ; for silver, 6 Q ; for gold, 9 ; for zinc, 18 ; 
 for platinum, 30 ; for iron, 30 ; for tin, 36 ; for lead, 72 ; from 
 which it appears that silver and copper are the best conductors, as 
 being the least heated by a given electrical discharge ; lead is the 
 worst conductor, as being the most heated by the same electrical 
 discharge, whilst iron and platinum are between the two. 
 
 181. The susceptibility of fine wires, of different metals and of 
 given diameters, to become heated to a greater or less extent by 
 the electrical discharge enables us to convert the Thermo-Electro- 
 meter into an ordinary electrometer of different degrees of 
 sensibility. Thus (180), whilst a given electrical charge passed 
 through a fine wire of copper of given diameter in the ball, causes 
 the fluid to ascend along its scale only 6, the same charge will, 
 when passed through a platinum wire of the same diameter en- 
 closed in the ball, raise the fluid to five times the height, or 30, 
 
154 THERMO-ELECTROMETER. 
 
 and when passed through a wire of lead of similar dimensions will 
 raise the fluid to no less than twelve times the height, or 
 72 ; thus affording a wide range of observation between copper, 
 platinum, and lead. 
 
 The metal best adapted to the purposes of an ordinary electro- 
 meter is platinum ; it is easily acted on, and not liable to oxidisation. 
 The wire in the ball may vary in diameter from the one-fiftieth to 
 the one-hundred- and- fiftieth of an inch, according to the require- 
 ments of the experiment. This method of estimating electrometer 
 effects, by means of the electrical discharge through metallic wires, 
 is convenient, and susceptible of much greater accuracy than the 
 ordinary means by fusion (176). 
 
 182. Fig. 99 represents a Thermo-Electrometer with two wires, 
 silver and platinum, each one-hundredth of an inch in diameter, 
 passed air-tight through its ball, by which electrometer effects are 
 readily determined. 
 
 Exp. 50. Charge the electrical jar A (Fig. 101), exposing about 
 4 square feet of coated surface, with 10 units through the medium 
 of the unit measure u. Having completed the requisite metallic 
 communications, through the charging and discharging quanti- 
 tative electrometer, discharge the accumulation through the plati- 
 num wire. Observe the ascent of the fluid along the scale 
 suppose 10 degrees. 
 
 Exp. 51. Accumulate 20 units, or double the former quantity of 
 electricity, on the given surface of 4 square feet, and discharge the 
 accumulation through the same wire. The fluid will now be 
 observed to ascend along the scale of the instrument four 
 times the former height, or 40 degrees ; by which we perceive 
 that the heat evolved by a given electrical discharge is as the 
 square of the quantity of electricity. 
 
 OCCASIONAL MEMOEANDA AND EXPLANATOEY NOTES. 
 
 (E) The error arising from the small quantity of charge left upon the 
 transfer-plate or globe after each deposit may in most cases be neglected, 
 or otherwise calculated or avoided altogether. Suppose the relative 
 diameters of a transfer-plate or globe, and a given circular plate or globe 
 to be as 1 : 12 ; in that case the charge would be twelve times as great 
 upon the larger plate or globe as upon the smaller plate or globe. To put 
 in evidence how small a quantity would in this case be' left upon the trans- 
 fer after contact with the given surface, we may observe that the charge of 
 which the larger plate is susceptible is at least twelve times that of the smaller 
 plate, and therefore after the contact the charge of the smaller disc will be 
 shared between the two, the larger one consequently taking eleven-twelfths of 
 
EXPLANATORY NOTES TO CHAP. IV. 
 
 155 
 
 the whole charge, leaving only a residual of one-twelfth part of the transfer 
 charge on the circular inch, which (145) would be considerably less than 
 could be sustained by a disc of one-tenth of an inch in diameter. There 
 will of course be a slight increase after each transfer^ but for the few 
 transfers generally employed the error arising therefrom would be of no 
 moment. It is easy to estimate the amount of the residual on the transfer 
 by placing a hollow sphere, having an open mouth (57), in connection with 
 a delicate electrometer, and after each contact of the transfer measure with 
 the given insulated conducting plate, deposit the residual within the hollow 
 sphere. We know in this case that all the residual will be taken up by the 
 hollow sphere ; the electrometer will, consequently, indicate the total sen- 
 sible quantity of electricity which the small transfer plate has carried away 
 after any given number of transfers, the distance between the electrometer 
 attracting discs remaining constant. 
 
 This kind of experiment will be quite sufficient for estimating the greatest 
 error which could possibly arise from the small quantity carried off by 
 the transfer measure after a given number of deposits. 
 
 (S) Light circular plates, hollow spheres, and other light bodies for ex- 
 perimental investigation by means of the Scale-beam Electrometer should 
 be constructed of a very light wood, termed white cedar lath wood (Chap. 
 V., 35), and should be constructed in the way referred to Chap. III., 53. A 
 circular plate of cedar lath wood, 7 inches in diameter and three-tenths 
 of an inch thick, when carefully prepared in the lathe, and lightly var- 
 nished, was found to weigh only 700 grains. 
 
 A hollow sphere of 5 inches diameter and three-twentieths of an inch 
 thick, constructed of this wood, was found to weigh less than 24 drachms, 
 or 3 ounces ; and, consequently, well adapted to the delicate Scale-beam 
 Electrometer. 
 
 (T) The Eev. Dr. Lloyd, Professor of Mathematics in the Dublin Uni- 
 versity, appears to have arrived at a different result from that of Coulombe. 
 Instead of the reactive force of torsion being, 
 according to Coulombe, as the angle of tor- 
 sion, it appears, by Dr. Lloyd's magnetic 
 investigations, the reactive force is as the 
 sine of the angle, and not as the angle itself. 
 This difference, however, with Coulombe's 
 result has probably arisen from the circum^ 
 stance that Dr. Lloyd's experimental inves- 
 tigations depended upon bifilar threads, not 
 sustained at given distances apart by cross- 
 stays (159), but by bifilar threads which in 
 turning were allowed to collapse upon each 
 other, and which would greatly influence 
 the result of the mathematical analysis of 
 the force. 
 
 Let, for example, 2 r = distance between 
 the threads = m m' = n n' ; P P' = length 
 of suspended needle = 2 a ; angle of incli- 
 nation of the thread at the point n to the 
 
 vertical = i ; horizontal angle through which P P' has moved, owing to 
 the repulsive force under examination = ', tensions of threads, m n and 
 m n' (which are not precisely the same) = T and T' respectively ; weight 
 of the suspended needle P P' = w ; horizontal force disturbing p p', and 
 acting at right angles, = /. 
 
 Then taking moments about n in the horizontal plane containing p p', we 
 
 Fig. 102. Bifilar Suspension. 
 
156 EXPLANATORY NOTES TO CHAP. IV. 
 
 have (T' sin. i) 2 r =/ (a -{- r) . . . . (1). Similarly, taking moments 
 about ri we have (T sin. i] 2 r = / (a r) . . . . (2). 
 
 Eesolving the forces of the system vertically, we have (T -{- T') cos. 
 
 *\V T 1 ' r/ I. T* T 1 ' 
 
 i = w; .-. T -}- T'= S but , dividing (1) by (2) ; .-. : ,= 
 
 cos. * T a r T -}-T 
 
 ; and T - = _r (T 
 
 2 a 2 a 2 a cos. i- 
 
 From (1) / = ' T' = T Sm ' l s/ = -W tan. <?'. 
 
 a + r a-j-r 2a cos. $ a 
 
 Now perpendicular distance of point of suspension above the index needle 
 p p' = I cos. i = p say, I being the length of each thread in the bifilar 
 
 balance, and r =p tan. i accurately = I cos. i X tan. * .-. tan. * = - x 
 sec. i. But i being a very small angle sec. i is nearly = 1, and .-. tan. i 
 may be considered = , this being too small by only a small quantity. 
 
 r rrflwr 2 w r 2 . 
 
 ' J = ~ w , tan. 1 becomes - w -r- = r 0> an " since is con- 
 a a I a I a I 
 
 stant, depending upon the dimensions and weight of index needle r p', 
 length of threads and distance between them, it appears that / oc ; that 
 is, the force deflecting p p' varies directly as the angle of deflection when 
 / acts at right angles to P p'. If the force be considered as acting in the 
 direction of the chord of the arc in which p moves, then the force / so 
 
 acting, = X 5 ' in which case/oc But since, in the practica 1 
 
 (I/ (f n s\ 
 
 cos. _ cos. y 
 
 2 2 
 
 application of this instrument the arc between the repelling disc and the 
 repelled end P of the needle is less as the force under consideration increases, 
 
 Q 
 
 cos. -. will differ very little from unity ; and we may, consequently, without 
 
 any sensible error, consider that / oc 9, or as the horizontal angle through 
 which the needle has moved owing to the repulsive force, which is the same 
 result as that arrived at by Coulombe. 
 
 (U) It may appear difficult to understand how, by turning the plate I 
 in the reverse direction of the torsion, we thereby increase the torsion force 
 of the suspending wire. It will however appear, on a little reflection, that 
 whilst the suspending wire is liable to torsion from one extremity in one 
 direction, it is acted upon by torsion on the other extremity in the opposite 
 direction ; that is to say, it is liable to twist by the action of the plate i 
 above in one direction, and by the resisting repulsive force below in the 
 opposite direction. It is, in fact, in the same condition as that of a rope 
 twisted at both ends in opposite directions. 
 
CHAPTER Y. 
 
 ELECTRICAL MANIPULATION. 
 
 THE most common articles required by the practical electrician 
 may be thus enumerated : 
 
 (1.) A moderately sized turning lathe, with turning tools, drills, 
 and other requisites. 
 
 (2.) A small work-bench of solid oak, fitted with parallel and 
 common vices, supported on strong iron levelling- screws, so as to 
 give it a firm and steady position. 
 
 This bench may be about 4 feet long, 18 inches wide, and about 
 3 feet high. It should have two drawers, one at each end, 
 immediately under its table surface, and two or three capacious 
 drawers in front. The drawers at the extremities are for small 
 tools, and other selected articles, such as fine broaches, and piercers 
 of various kinds ; delicate pliers, nippers, forceps, and the like. 
 
 (3.) Saws of various kinds and sizes, for wood and metal. 
 
 (4.) Planes of various kinds, small and large; spokeshaves. 
 
 (5.) Screwdrivers, chisels, gouges, gimblets, bradawls. 
 
 (6.) Files and rasps of different sorts and sizes, flat, round, half- 
 round, and square. 
 
 Slender round rasps and files, usually termed rat's-tail, are useful 
 in completing roughly perforated holes'. 
 
 (7.) Pincers, pliers, hand- vices, cutting nippers. 
 
 (8.) Callipers of various dimensions for taking measurements. 
 
 (9.) Cutting knives and scissors. 
 
 (10.) Large and small drills, with drill-bow and stock handles 
 to fix and work in a vice. 
 
 (11.) A set of punches from a quarter of an inch in diameter 
 upwards, including such as are used for gun wadding : a square 
 or circular mass of lead for punching on. 
 
 (12.) One or two portable anvils mounted on wooden blocks. 
 
 (13.) Hammers for fine and coarse purposes. 
 
 (14.) A stock, with bits of various sizes. 
 
 (15.) Oil stones, pressure oil-cans of a conical form for dropping 
 oil. 
 
158 
 
 ARTICLES REQUIRED IN 
 
 (16.) A diamond for scratching, or cutting glass. 
 
 (17.) One or two tobacco-pipe stems for blowing a stream of 
 hot air on any substance through their heated extremities. 
 
 (18.) A set of heating-irons (Fig. 103), for warming glass rods, 
 and other insulators. These should be inserted in handles, and 
 should be of various forms and dimensions. 
 
 The globular iron a is very convenient for drying the interior of 
 glass jars. 
 
 Irons having curvilinear or other surfaces, such as b, c, d, e y are 
 convenient for drying off cylindrical or other insulators. Two 
 curvilinear irons, such as b, c, or d, applied against each other 
 may be made to surround glass rods of various sizes. 
 
 Heating- irons of large mass, mounted upon a foot as represented 
 at /, are applicable in varnishing glass insulators of various 
 forms and sizes ; being fixed in a wooden holder they are easily 
 
 Fig. 103. Heating Irons. 
 
 heated in the fire, and subsequently supported by the wooden 
 holder on a stout wooden foot. The glass to be varnished is 
 held over them. 
 
 (19.) A set of burning-irons in handles g (Fig. 103), varying 
 from the size of a common steel knitting-needle to half an inch in 
 diameter, and from 6 inches to 1 foot or more in length. Such 
 irons, together with the rasps (6), are useful in perforating holes 
 through corks and other substances. 
 
 (20.) Earthenware pipkins of various sizes, chemical crucibles, 
 and such like. 
 
 (21.) T-squares, straight-edges, and graduated rulers, with 
 plumb-lines attached. 
 
ELECTRICAL MANIPULATION. 
 
 159 
 
 for 
 
 104. Dividing Board. 
 
 (22.) Case of drawing instruments. 
 
 (23.) One or two right-angled triangles for drawing. 
 
 (24.) Spring compasses and dividers, coarse and fine. 
 
 (25.) A square mahogany board set on levelling-screws, 
 obtaining a level surface when necessary. 
 
 (26.) One or two spirit-levels. 
 
 (27.) A dividing-board for graduating card-paper, or other 
 circles. This consists (Fig. 104) of a graduated brass circle, from 
 1 foot to 18 inches diameter, inlaid, centrally, in a square maho- 
 gany board. The meridian circle of a 
 globe may be employed for this pur- 
 pose. A central point is screwed into 
 the board, against which rests the 
 notch of a straight metallic rule, so 
 adapted as to be, when in place, a 
 prolongation of the radius of the 
 brass circle. The cardboard to be 
 divided having circles swept upon 
 it by accurate compasses, is pinned 
 down to the board, the centre of the 
 circle to be divided being in the 
 centre of the brass circle. The radius ruler is moved round 
 over the graduations of the brass circle, and the degrees marked 
 off by a drawing-pen on the card circle within. 
 
 (28.) A good double-barrelled air-pump, with receivers of various 
 sizes. 
 
 (29.) Balances with scales. A balance to weigh from 1 ounce to 
 3 pounds, and which will turn with a few grains when loaded with 
 its greatest weight. 
 
 (30.) A balance for more refined purposes, which will weigh 
 from half a grain to 3 ounces, and will turn with half a grain 
 when loaded with 1000 grains. 
 
 (31.) A balance for still more refined purposes, which will 
 weigh from the tenth of a grain to 1 ounce, and will turn with 
 one- tenth of a grain when loaded with 500 grains. 
 
 The first two may be suspended from a lever arm project- 
 ing from a brass standard rod, so that by a fine silk cord passing 
 over a pulley to a lever foot beneath, the beam may be raised, 
 and the scale-pans lifted off the platform upon which they 
 rest. 
 
 The brass standard in these cases is usually fixed on a neat 
 mahogany rectangular box, having a drawer for the reception 
 of the necessary weights. 
 
160 ARTICLES REQUIRED IN 
 
 The second, or finer balance, should be enclosed in a small glass 
 case, with front doors, or a sliding front plate of glass which 
 can be elevated to any height. 
 
 The third, or very refined balance, should be mounted in a way 
 similar to the fine chemical balances, and for which it is to be a 
 substitute for delicate purposes. This should be enclosed in a glass 
 case similar to the former. A chemical balance of the highest 
 quality and construction is not called for in ordinary electrical 
 investigations. 
 
 (32.) Weights. There are several kinds of British weights in 
 use, such as the avoirdupoise weight, troy weight, and apothecaries' 
 or chemists' weights. 
 
 The different divisions of avoirdupoise weight are tons, hun- 
 dreds, quarters, pounds, ounces, and drachms. In this weight 
 16 drachms make 1 ounce = 960 grains, and 16 ounces 1 pound 
 =; 15,360 grains. 
 
 The divisions of troy weight are pounds, ounces, pennyweights, 
 and grains. In this weight 24 grains make 1 pennyweight, 20 
 pennyweights 1 ounce = 480 grains, and 12 ounces 1 pound = 
 5,760 grains. 
 
 In apothecaries' or chemists' weight the divisions are pounds, 
 ounces, drachms, scruples, and grains. In this kind of weight 
 20 grains make 1 scruple, 3 scruples 1 drachm = 60 grains, 
 8 drachms 1 ounce = 480 grains, 12 ounces 1 pound = 5,760 
 grains. 
 
 These different denominations of weight are apt to lead to much 
 confusion, so that it becomes desirable, for the sake of simplicity, 
 to resort to a unit standard of weight, of which a grain is to be 
 considered the unit. Every denomination of weight, therefore, is 
 easily reducible to a certain number of grains. Thus we find by 
 the chemist's weight that 1 drachm contains 60 grains, being 3 
 scruples ; and if by avoirdupoise weight there are 16 drachms in 
 an ounce, an avoirdupoise ounce must be 60 X 16 = 960 grains, 
 and if there are 1 6 ounces in the pound, an avoirdupoise pound 
 must be 960 X 16 = 15,360 grains. In apothecaries' weight 
 the number of grains in the ounce is 480, being only half the 
 number of those in the avoirdupoise ounce. 
 
 The different species of weights are usually made in sets, so as 
 to fit one within the other, especially the larger weights pounds 
 and ounces. Small weights, such as grain weights, are made 
 of thin metal, such as sheet brass, silver, gold, or platinum, with 
 small stamped points on them indicating the number of grains, 
 No. 10 being marked with a cross ; or they are made of fine plati- 
 
ELECTRICAL MANIPULATION. 161 
 
 num wire, being a single piece to indicate 1 grain, and bent pieces, 
 indicating by the number of turns or bends their weight in grains : 
 these are generally from 1 to 5 (Fig. 105). Four weights of 
 1, 2, 3, and 4 grains are sufficient to weigh 10 grains ; and similar 
 weights of 10 grains will conveniently weigh up to 100 grains. 
 Fractions of grains are 
 generally made of small 
 platinum wire, bent in 
 the way just described ; 
 
 SO that with 1, 2, 3, "Fig. 105. Grain Weights. 
 
 and 4- tenths of a grain we may take any fraction of a grain in 
 tenths. It is perhaps needless to remark that these weights should 
 be preserved in small pill-boxes or cases, and placed in the drawers 
 of their respective balances. The beams of the balances should be 
 carefully preserved and frequently cleaned. 
 
 Besides these balances and weights, measures of capacity are 
 requisite, such as quarts, pints, half-pints, graduated glasses of 
 various kinds, one or two tubular measures, divided into cubic 
 inches. 
 
 (33.) A small collection of chemicals, such as alkalies, acids, 
 rectified spirit of wine, naphtha. 
 
 Pestles and mortars of porcelain and iron. 
 
 Glass, porcelain, and metallic funnels, small and large. 
 
 One or two spirit-lamps ; cotton for wicks. 
 
 A table bellows blowpipe, for melting, bending, or blowing 
 glass, with lamp and other requisite apparatus. 
 
 Blowpipe for the mouth. 
 
 Other kinds of blowpipes for exposing bodies to a high tempera- 
 ture will be found useful. 
 
 (34.) A few iron ladles of various sizes with lips. It would be 
 convenient to have a very large iron ladle for applying heat, 
 after the manner of a sand bath. The simplest form, however, of 
 a sand bath, is a small crock, such as used for melting pitch, filled 
 with sand and placed over the fire. 
 
 (35.) A few logs of a very light kind of wood termed white 
 cedar lath- wood, imported from Nova Scotia, will be found useful 
 in constructing electrical apparatus, more especially where levity 
 is an object. It is durable, and little liable to warp or change 
 in shape. Its levity is remarkable a cubic inch of it does not 
 weigh above 98 grains. A circular plate of 2'8 inches diameter 
 and one- tenth of an inch thick was found to weigh only 100 
 grains. A circular plate of 7 inches diameter and three-tenths 
 thick was found to weigh 700 grains only. A hollow sphere, 
 
 M 
 
162 
 
 ARTICLES REQUIRED IN 
 
 of 2 inches diameter and about one-tenth in thickness, was found 
 to weigh only 150 grains (154). A hollow sphere, of nearly 5 inches 
 diameter and about three- twentieths in thickness, was found to 
 weigh 600 grains. This wood is sightly in appearance when 
 got up in a lathe, varnished and polished, and is useful for deli- 
 cate electrical purposes. 
 
 (36.) A gilder's cushion for gold-leaf, with cutting-knife, palette- 
 knife, wooden forceps, together with a wide gilder's brush and a 
 hare's foot, for brushing away loose gold. 
 
 A gilder's cushion consists of a rectangular board, about 1 foot 
 long, 6 inches wide, and three-quarters of an inch thick, upon 
 which is laid several pieces of flannel, or cloth, of the same 
 dimensions. The whole is covered with calf- skin leather, such 
 as is used by bookbinders, the rough side upwards, secured to 
 the sides and ends of the board by fine tacks, the superfluous 
 edges of the leather being cut away to the size of the board. The 
 cushion is bound round at the edges with narrow silk ribbon 
 secured over the tacks by glue or paste. When thus finished the 
 leather surface should be hard and fair, and very little elastic. It 
 is usually rubbed over with a little prepared chalk, or very fine 
 whitening, which is well rubbed into the leather, and the super- 
 fluous particles brushed away. 
 
 (37.) A stout wooden bowl, suspended by three strong hempen 
 lines, containing a heavy iron shot from thirty to 
 forty pounds in weight, for pulverising (Fig. 106). 
 The lines of suspension are attached to the bowl 
 by means of small iron hooks and eyes screwed 
 to the bowl. The suspension lines converge to a 
 central hook after the manner of a scale-pan, by 
 which the whole is hung up to a beam, so as to 
 take the weight of the shot and bowl. The sub- 
 stance to be pulverised being broken into small 
 fragments, is put into the bowl with the shot. 
 The shot, by a dexterous circular impulse given 
 to the bowl externally by the hands, is caused to 
 spin rapidly round in it, by which the matter, 
 falling continually between the sides of the bowl 
 and the shot, is speedily reduced to a fine powder. 
 The bowl has a light cover to prevent any dust 
 from flying about. This is a convenient and use- 
 ful piece of apparatus for pulverising substances, 
 such as resin, lac, sealing-wax, and such like. 
 If the student be anything of a mechanic, and construct his own 
 
 Fig. 106. Pulverising 
 Bowl. 
 
ELECTRICAL MANIPULATION. 163 
 
 apparatus, lie will find all these things essential, either in whole or 
 part, so that, with the occasional assistance of clever workmen, he 
 will have little difficulty in carrying out any experimental research 
 in which he may be engaged. 
 
 (38.) Glass in almost every form is essential, as rods of glass, 
 glass tubes, and plates of various dimensions. Glass rods (some- 
 times called glass cane), vary from the eighth of an inch to 
 1 inch in diameter, and from 1 to 2 feet in length. Tubes of 
 glass from one quarter of an inch to 1 inch in diameter, and 
 upward. 
 
 Mica or Muscovy talc, in bulk and in thin plates. Tourmaline. 
 
 (39.) Kesinous substances of various kinds, such as amber, lac, 
 brimstone, common resin, beeswax, turpentine, sealing-wax (red 
 and black), and other resinous bodies. Lac and brimstone 
 are useful in electrical excitation ; as are also gutta percha, 
 caoutchouc, and other gums ; also vulcanite and ebonite new 
 combinations of these substances with brimstone and other elec- 
 trics, are important, all these being good insulators. 
 
 (40.) Silk, being an electrical substance, should be at hand in 
 all its forms such as soft silk stuff; plain white silk, as employed 
 for fine silk handkerchiefs ; soft white and other silk ; old silk 
 handkerchiefs; silk ribbon of different widths; silk thread; long 
 filaments of unspun thread from the silk- worm ; common silk- 
 worm gut, such as used in fishing, and which will be found an 
 excellent insulator; silk, oiled on one side only, so as to give 
 a somewhat rough surface on the other ; oiled silk general^. 
 "Where a firm silk surface is required, plain white or other silk 
 may be strained over plates of wood with intermediate layers of 
 thick cloth or flannel. 
 
 (41.) Dry and baked wood, more especially mahogany, which 
 becomes insulating after immersion in hot oil. 
 
 (42.) Animal furs and skins of various kinds, such as cat- skin, 
 hare-skin, tanned with the fur on. Soft new flannel and 
 woollen stuffs, fine Basil leather ; skins of chamois leather, which 
 are admirable for cleaning glass. 
 
 (43.) The substance termed spongio-piline a combination of 
 sponge and wool is firm, elastic, and useful to the practical 
 electrician. The best exciter for rough glass tubes, brimstone, 
 and other negative electrics, is soft, new, dry flannel. Cushions 
 constructed of these substances will be very useful. 
 
 (44.) Light substances of various kinds, such as straw reeds, 
 used in bonnet-making ; also other kinds of reeds and light 
 tubular vegetable substances generally ; quills of small birds, and 
 
 M2 
 
164 ARTICLES REQUIRED IN 
 
 other quills ; fine downy feathers ; bog down ; very fine cotton 
 wool ; corks of all descriptions fine bottle corks, bungs, and 
 phial corks of all sizes. Elder-pith, pith of the sola plant, and 
 such like. Gold and silver leaf ; Dutch metal leaf ; gold-beaters' 
 skin. 
 
 (45.) Fine threads of various kinds ; fibres of unspun silk 
 from the silkworm, just mentioned (40), and of very fine flax ; 
 fine lengths of human hair ; spiders' web wherever it can be 
 collected (Chap. I, note A). All these are frequently required for 
 the suspension of light bodies requiring easy motion. 
 
 (46.) Balls of various kinds and sizes ; balls of smooth fine 
 cork ; light balls, made from the pith of elder, or the sola plant, 
 varying from one-tenth to half an inch in diameter ; a cylinder 
 of the pith of the sola plant, 10 inches long by loths dia- 
 meter, was found to weigh 100 grains, so that a cubic inch of it 
 would weigh about 10 grains. A ball of the sola plant of half 
 an inch diameter weighs only about 1 grain. 
 
 A ball of elder pith, three-tenths of an inch diameter, was found 
 to weigh one-fifth of a grain. 
 
 A ball of fine cork seven-tenths of an inch diameter was found 
 to weigh 7 grains. Small wooden balls of various sizes, neatly 
 turned; light, thin, hollow metallic balls, of different sizes, 
 generally of sheet brass or copper. 
 
 (47.) Drawn metallic tubes, of various sizes, from one- eighth to 
 half an inch in internal diameter, and from 1 foot to 18 inches or 2 
 feet in length. These tubes are usually of brass or copper, sometimes 
 silvered, especially the smaller ones, and slide easily one within 
 the other. Solid brass rods and wires, of various diameters and 
 sizes ; small rings of fine iron, copper, brass, and other wires. 
 
 (48.) Stability being essential to the success of many important 
 inquiries in electricity, it is desirable to be provided with some 
 rings of lead of different dimensions, from 2 to 5 inches in 
 diameter, and from half an inch to 2 inches thick, and having a 
 round hole in the centre from half an inch to an inch in diameter, 
 or more. 
 
 (49.) Small lumps of pure zinc and tin ; mercury. These are 
 the ingredients in electrical amalgams (94). 
 
 (50.) Aurum Husivuw, already mentioned, Chap. I. (7), also 
 termed Mosaic gold, is efficient as an amalgam for the rubber of the 
 electrical machine (100). 
 
 (51.) Circular and other foot pieces nicely got up in a lathe. 
 These should be of various diameters and thicknesses, and turned 
 with as few ornaments and projections as possible, they should 
 
ELECTRICAL MANIPULATION. 165 
 
 have a central socket or hole to receive and uphold any vertical 
 rod or glass support which may be required. A wide deep 
 groove should be turned in the under surface of the foot to 
 receive a ring of lead which should be run into it with a view to 
 stability. Whilst in the lathe the foot should be nicely varnished 
 in the way described (70). 
 
 (52.) Preparation of electrical balls. Hollow, metallic balls for 
 electrical purposes are usually prepared by casting two half spheres 
 as thinly as possible, and then soldering them together. The junc- 
 tion is neatly effected by previous preparation in a lathe, so that 
 the edges may exactly apply one over the other. They are then 
 finished off and lacquered in the way described (69) ; one of the 
 hemispheres is usually left a little thicker at its apex to receive 
 a hole and tap for a screw. The joint should not be exactly 
 in the centre of the ball, for which purpose one of the hemi- 
 spherical castings should be somewhat larger than the other, so that 
 we may drill small holes, if necessary, in the equator of the ball 
 without meddling with the joint. We may obtain at the iron- 
 monger's shops light, three-quarter, hollow spheres in brass pressed 
 up by machinery, and having an iron screw in the centre, generally 
 employed as supports for articles of furniture. These are con- 
 venient in constructing hollow metallic balls for electrical appa- 
 ratus. The iron screw, with the attached part of the hollow ball, 
 being removed, the deficient portion of the sphere is supplied by 
 a brass casting from a mould accurately turned to the required 
 segment. This segment is nicely fitted by means of a lathe, and 
 soldered to the remaining part of the ball, the whole being turned 
 off so as to be completely spherical. In this way we have three- 
 fourths of the ball without a joint. For the convenience of tap- 
 ping, the added segment should be thicker at its apex than the rest 
 of the ball. 
 
 Another method of producing light hollow metallic balls, is by 
 beating up two hemispheres of sheet copper in hemispherical 
 moulds, and then uniting them by soldering as before. When united 
 they are carefully finished off in the lathe, and lacquered. Yery 
 beautiful electrical balls may be obtained in this way, and, if very 
 thin sheet copper be employed, are extremely light. 
 
 (53.) Wooden balls. These are convenient for many purposes, 
 and may be of any kind of wood ; boxwood, ebony, and mahogany 
 are three kinds of wood often employed for solid balls intended to 
 cover the extremities of brass or glass rods, and may be from a 
 quarter of an inch to an inch in diameter. They must of course 
 be got up in a lathe, and should be nicely varnished whilst in it. 
 
166 ARTICLES AND PROCESSES IN 
 
 Light, hollow balls of wood are best made of the white cedar 
 lath- wood already described (35). A block of the wood is first 
 roughed out in the lathe of a spherical form, and then cut by 
 a saw into two hemispheres ; each of these hemispheres is then 
 hollowed out in the lathe to given dimensions. The two hemispheres 
 are then firmly united by means of a rabbet joint, and glue ; after 
 which, the rough sphere is again returned to the lathe, and 
 turned down to a fair spherical surface, and of such a thickness 
 and dimensions as to leave the hollow sphere required. Yery 
 light spheres of this description are important in the measurement 
 of electrical attractive forces by delicate balances to which they 
 are usually suspended. (Chap. IV., 154). 
 
 Hollow wooden balls may be formed of other woods, the weight 
 being diminished by removing the interior portions as in the pre- 
 ceding case. 
 
 (54.) Cork balls , being valuable in electrical experimental 
 research, should be got up in a lathe from very select pieces of 
 cork of the finest texture, but great care is requisite in turning 
 them. 
 
 In preparing cork and other soft balls, the following method may 
 be occasionally resorted to. A hemispherical cup is to be turned 
 in a piece of freestone, common brick, or very hard wood, and 
 being adapted to the lathe, so that the cup may revolve centrally, 
 the cork previously roughed into a spherical form is held in the 
 revolving cup, and turned' round in it until it becomes fair, and 
 smooth, and rubbed down into a sphere of the same diameter as 
 the hemispherical cup. 
 
 A lathe is not absolutely necessary for this purpose ; the cork 
 being neatly fashioned by eye into the form of a sphere with 
 the assistance of a sharp pen-knife, is perfected by turning the 
 ball round in all directions by hand in the hemispherical cup. 
 If a hard wood be used for the hemispherical cup a small quan- 
 tity of emery, or other abrasive powder, rubbed into the cup will 
 greatly shorten the process. 
 
 Light substances of soft delicate texture, such as the pith of 
 elder or the pith of the sola plant, may be readily formed into 
 balls in a similar way ; that is to say, by first fashioning the mass 
 by means of the hand and a sharp pen-knife into a spherical form, 
 and then turning it round in a hemispherical cup of common 
 house-brick or freestone. 
 
 (55.) Piercing holes through corks. The employment of cork 
 in the construction of electrical apparatus is frequently required, 
 cork being a most valuable substance, especially when of fine 
 
ELECTRICAL MANIPULATION. 167 
 
 grain and compressible. We frequently require cork plugs with 
 holes through them for the admission of glass rods or tubes. 
 The best method for piercing holes through cork is the follow- 
 ing : Having selected a piece of thin-drawn brass tube of the 
 required size, file one of its extremities to a cutting edge, and 
 with this bore centrally from each end of the cork. We shall 
 speedily get through it, leaving a clean hole in the cork. The 
 brass tube in leaving will have taken out a cylindrical core, 
 which will be found remaining in the tube, and may be usefully 
 employed as a plug of small size. If the thread of a fine screw 
 be cut upon the boring extremity of the brass tube, the operation 
 will be greatly facilitated. It enables us, as it were, to screw 
 the tube through the cork. 
 
 The common method of piercing corks is by means of a red hot 
 iron ; or, having first bored a small hole through the cork with a 
 gimblet or some other boring tool, we may proceed to enlarge it 
 with a rat-tail or other round file (6). The first method, however, is 
 the best, and it will be found convenient to prepare a series of 
 boring tubes in the way above-mentioned, varying from 3 to 6 
 inches in length, and from a quarter of an inch to 1 inch in 
 interior diameter. 
 
 Very light balls may be obtained from the gall-nut. These are 
 often found in the forest, penetrated by an insect, and hollow 
 within. When turned truly in the lathe, so as to be perfect 
 spheres, they form extremely light balls. 
 
 (56.) Metallic conducting rods of various lengths and sizes. These 
 are convenient for uniting detached insulated bodies, as, for ex- 
 ample, electrical jars, or in connecting them with the conductor 
 of the machine, and such like. They are constructed of a central 
 tube, having two pointed sliding wires passing with friction 
 into it from each end, so as to admit of adjusting the length 
 of the conducting-rod to any extent required. These conduc ting- 
 rods may vary from 4 inches to 2 feet in length, having central 
 tubes of from one-tenth to a quarter of an inch in diameter. 
 A great number of these connecting-rods should be prepared of 
 various lengths and sizes. 
 
 (57.) Varnishing, Varnishes, and Cements. The art of covering 
 glass insulators and other pieces of electrical apparatus with a film 
 of dry, hard, non-conducting varnish, impervious to moisture, is 
 an important process in practical electricity. The liability of 
 insulators, more especially glass, to condense moisture on their 
 surfaces, is fatal to its perfect insulation. It may be proper here 
 to observe that any kind of varnish employed for obviating this 
 
168 ARTICLES AND PROCESSES IN 
 
 source of error must be free from all clamminess, so that nothing 
 may adhere to the varnished surface. Hence, any preparation 
 containing balsam, oil, and the like, is inadmissible. Varnishes 
 for electrical purposes should dry speedily, and solidify into a hard, 
 thin, transparent film. 
 
 (58.) The great basis of varnish for electrical apparatus is a 
 resinous gum called " lac," usually obtained in commerce under 
 the three following forms " stick-lac," " seed " or " gumJac," 
 and " shell-lac " (note V). 
 
 Seed-lac and shell-lac are the forms usually resorted to 
 in preparing electrical varnishes. Shell-lac is best adapted for 
 glass. Seed-lac is commonly used in preparing varnish, com- 
 monly termed lacquer, for guarding metallic substances against 
 oxidation. 
 
 (59.) Shell-lac and seed-lac varnishes are thus prepared : 
 
 Formula 1. Put 1 ounce of the shell-lac of commerce into a wide- 
 mouthed 8 -ounce phial, containing 5 ounces of well- 
 rectified naphtha. 
 
 Close the bottle with a cork, and let it stand in a warm 
 atmosphere until perfectly dissolved. Shake the mixture fre- 
 quently, and pass the fluid through a paper filter, such as is used 
 by chemists. Add rectified naphtha to the solution from time 
 to time in such quantities as will enable it to percolate freely 
 through the filter. The filter must be changed when neces- 
 sary for a new one. We obtain in this way a thin, transparent 
 fluid, perfectly clear, and of a light colour. This must be pre- 
 served in a clean phial, well stopped. The quantity of rectified 
 naphtha to be employed must be regulated by the fluidity of the 
 original solution, so that it may be easily filtered. The naphtha 
 must be very highly rectified, otherwise we do not get a clear 
 solution ; it will appear muddy, and not settle finely, nor will 
 it filter. 
 
 The kind of lac best adapted to electrical varnish is the common, 
 coarse lac employed by hatters and in other trades ; it is rather 
 of a brownish cast. The very refined preparations of shell-lac 
 sold by the chemists do not always produce a clear solution ; but 
 frequently one of a thick, muddy character, which does not filter 
 clearly. 
 
 We may occasionally substitute, for naphtha, highly rectified 
 methylated spirit. 
 
 (60.) Varnish or lacquer for metallic surfaces, such as polished brass, 
 copper, and other metals. 
 
ELECTRICAL MANIPULATION. 169 
 
 Formula 2. Dissolve 1 ounce of bruised seed-lac in 6 ounces of 
 rectified naphtha. 
 
 Keep it, as before, in a warm atmosphere, in a closed Sr-ounce 
 phial, until the solution is complete. Shake the mixture from 
 time to time. Pass the solution, as in the last case, through 
 a filter, and preserve the clarified solution in a clean phial, well 
 stopped. This is the best and most simple form of lacquer for 
 metallic electrical apparatus. 
 
 (61.) Varnish for Wood. Both seed-lac and shell-lac are 
 employed as the bases of varnishes for wood. The lac is usually 
 dissolved in highly proof- rectified spirit, or rectified naphtha. A 
 very small portion of Venice turpentine is sometimes added to the 
 solution to render the varnish less brittle. 
 
 Formula 3. Dissolve If ounce of shell-lac in 5 ounces of rectified 
 naphtha, or highly rectified spirit, as in the preceding 
 formula. The solution being complete, add 1 scruple 
 of Venice turpentine, or about a drachm of common 
 resin. When perfectly dissolved, shake the mixture, 
 filter, and strain off, 
 
 (62.) There is a kind of varnish, called quick-drying cabinet 
 copal, very valuable for wood apparatus, and which dries with a 
 fine glossy surface. 
 
 (63.) French Polish. The celebrated French polish is a solu- 
 tion of shell-lac in naphtha, sometimes combined with gums. 
 
 Formula 4. Gum-copal, and gum-arabic, of each 1 drachm. Ee- 
 duce these to a fine powder, and sift them through a 
 muslin sieve. Put the powders into a capacious, wide- 
 mouthed bottle, together with half an ounce of fine 
 shell-lac coarsely powdered. Add half a pint of recti- 
 fied naphtha. 
 
 Close the mouth of the bottle and expose the mixture to a gentle 
 heat for several days> occasionally shaking it. The simplest 
 method of heating is to fill a common iron saucepan with fine 
 sand ; immerse the bottle in the sand, and set it on a stove or 
 a hot plate until the whole is sufficiently warm, when it may 
 be put aside for a time. The heat must not be great. When 
 the solution is complete, which may be easily perceived by shaking 
 the bottle, filter and strain off. 
 
 (64.) Electrical Varnishes. Mode of Application. In the 
 application of electrical varnishes we must carry on the opera- 
 tion in a warm, dry atmosphere, the presence of moisture 
 being fatal to success. If moisture be present, the varnished sur- 
 face, instead of appearing transparent and clear, assumes a 
 dirty, opaque appearance, and becomes covered with irregular 
 
170 
 
 ARTICLES AND PROCESSES IX 
 
 striae and scoriae, which not only injure the value of the var- 
 nish, but spoil its appearance. Besides this precaution, it is 
 requisite to get the surface to be varnished perfectly clean 
 and free from, grease. The best way of effecting this is to wash 
 the article to be varnished in a weak solution of soda, and 
 subsequently in warm clean water, after which dry and wipe 
 off with clean linen free from lint ; finally, expose it to a 
 gentle heat, so that the surface of the applied varnish may not 
 chill. 
 
 The bodies to be varnished may be exposed to the heat of a 
 common fire in any convenient way. A common plate- warmer, 
 such as used for domestic purposes, having a door in its back 
 which can be thrown open when required, is admirably adapted 
 to varnishing purposes. It should have one or two round holes in 
 
 its sides for the reception of 
 glass-rods, tubes, or other 
 articles. These may be passed 
 into and through it, and ex- 
 posed to heat. The varnish 
 is to be applied by camel- 
 hair brushes of various de- 
 scriptions, flat and round; 
 and it would be convenient 
 to prepare a few with double 
 holders, one straight, the 
 other oblique (Fig. 107), so 
 that by means of a curved 
 
 Fig. 107. Varnish Brushes and Cup. j^^ we may apply ^^ 
 
 readily to an interior concave glass surface. 
 
 Previously to applying the varnish, a portion of it should be 
 poured out in a well- dried and warm conical cup, similar to an 
 egg-cup (Fig. 107), having a ring, with a small wire strained 
 across it, fitted to its mouth, so that, when the brush has been 
 dipped into the varnish, the superfluous fluid is readily disengaged 
 by drawing the brush across the wire. The cup, previously to 
 receiving the varnish, may be gently heated before a fire, in order 
 to free it from all moisture, and render the varnish as fluid as 
 possible. It is convenient to have several metallic rings of 
 various diameters fitted on different sized varnish cups. 
 
 (65.) Application to Glass-rods, Tubes, and Glass generally. The 
 rod or tube being duly prepared in the way above described (64), 
 and its temperature raised to a gentle heat, is held over a con- 
 cave red-hot iron, such as already alluded to (18). The brush 
 
ELECTRICAL MANIPULATION. 171 
 
 having been dipped in the varnish, and freed from its superfluous 
 fluid, is swept lightly along the rod from end to end, turning the 
 glass round over the hot iron during this process, and taking care 
 not to go over the same place twice. The varnished surface should 
 then be exposed to the heat of the iron from end to end, turning 
 it round until the varnish is nearly dry ; it should then be removed 
 into the warm atmosphere of the fire, where it should remain until 
 completely hard, which will be in a very short time. The operator 
 must be cautious not to employ the varnish except in a dilute and 
 thin state, so as to obtain an attenuated transparent coating upon 
 the glass. If a first application does not appear to be sufficient, 
 let it be varnished a second time. 
 
 (66.) Application to Glass Tubes. It is not always easy to 
 varnish the interior surface of a glass tube, in consequence of the 
 difficulty in applying the brush. 
 
 If the tube be sufficiently large, a full round brush, with a long 
 handle slightly curved forward close to the brush, may be em- 
 ployed. As soon as the varnish has been applied, a red-hot iron 
 wire terminating in a little round knob, and held by a convenient 
 handle, as shown at a, Fig. 103, is to be carefully passed into and 
 through the tube, so as to harden the surface by heat, and dissipate 
 all moisture ; or a stream of heated air may be passed through 
 the tube by holding it perpendicularly over a bright fire. The 
 tube should then be placed in the drying oven as before. If the 
 tube be so small as not to admit of the entrance of the brush, the 
 following method may be employed : Enclose a small lump of fine 
 cotton-wool in a piece of very fine thin muslin, and secure it 
 so as to form a kind of ball a little exceeding the diameter of the 
 tube ; pass a piece of stout twine through the tube, and attach 
 this ball to the end of it ; moisten the muslin surface freely with 
 the varnish so as to make it sufficiently wet, and then by means 
 of the cord draw the ball rapidly through the tube, the tube 
 having been previously heated. Let a current of warm air be 
 passed through the tube immediately after. If this first applica- 
 tion be not sufficient, repeat the operation. 
 
 (67.) Application to Glass Jars. The uncoated interval of the 
 electrical jar that is to say, that portion of the glass which is left 
 exposed to the air requires to be covered with a thin coating of 
 lac varnish both inside and out. This operation is a delicate and 
 sometimes a difficult one, more especially as concerns the interior 
 surface of the jar. In large jars varying from 8 to 14 inches in 
 diameter, having wide open mouths, the course to be pursued is as 
 follows : The coating of the jar having been completed up to 
 
172 , ARTICLES AND PROCESSES IN 
 
 the height required, place the jar horizontally with its mouth 
 towards a common fire, turning it round from time to time until 
 the glass becomes fairly warm. This will be better effected through 
 the back of the warming-oven already described (64), the mouth 
 of the jar being within the oven. Finally, the red-hot heating- 
 iron a, Fig. 103, may be passed carefully into the jar through 
 its mouth, taking care to avoid contact with the glass. In this 
 way the exposed glass becomes readily and evenly warmed, both 
 inside and out. 
 
 The interior surface of the glass jar should be first varnished, 
 for which purpose an assistant should hold the jar obliquely 
 over a red-hot drying-iron (18), while a wide camel-hair brush, 
 mounted on a curvilinear handle (Fig. 107), and moistened 
 with varnish, is being introduced into the mouth of the jar, 
 and rapidly swept round its interior surface until the whole of 
 the surface has been lightly covered. Immediately this is done 
 the globular drying-iron (18) should be heated, and carefully 
 introduced into the jar, and moved round and about it until 
 the coat of varnish has been well dried off. The jar should then 
 be exposed to a gentle heat by holding it in its first position 
 over the red-hot drying-iron. The outer surface should next 
 be varnished, which is a comparitively simple process. The jar 
 is to be placed upright before the fire. A wide brush, moistened 
 with shell-lac varnish, is then rapidly swept round the exterior 
 surface, after which the varnished surface is to be exposed to a 
 moderate degree of heat, which is best effected by a heated iron, 
 moved gradually round it at a moderate distance, until the varnish 
 appears perfectly dried off, and becomes clear and transparent. 
 
 Small electrical jars are more easily handled, and placed in any 
 position than large ones, but the mode of treatment is the same, 
 only the brushes and heated irons must be adapted to the size 
 and dimensions of the jar. "Where the jars are very small, the 
 mouths not exceeding three-eighths of an inch in diameter, it 
 will be necessary to employ small camel-hair pencils at the end 
 of curved handles, in order to varnish the interior surface. By 
 this method, and careful manipulation, a hard transparent surface, 
 very repellent of moisture, may be obtained. Some judgment 
 and practice will still be necessary to enable the operator to arrive 
 at the precise temperature to which the glass should be exposed, 
 and the regulation of the distance from the sources of heat. The 
 temperature of the glass previously to varnishing should be moderate 
 just sufficient to enable the varnish to run freely over the surface 
 without becoming chilled. On the other hand, the glass must 
 
ELECTRICAL MANIPULATION. 173 
 
 not be so hot as to drive the varnish into vapour, and cause the 
 iac to solidify in streaks. 
 
 (68.) Application of Varnish to Plates of Glass. Glass plates are 
 to be treated as in the preceding cases, the glass being first gently 
 warmed, and the varnish quickly applied by means of a very wide 
 camel-hair brush, rapidly swept over the surface. Two very thin 
 coats are generally requisite, the brush being carried for the first 
 coat in one direction, and for the second coat in a transverse 
 direction. It is not easy to apply the varnish to a glass plate so 
 as to give it a uniformly transparent appearance ; but it may be 
 done if proper precautions be taken. Plates of mica and other 
 electrics requiring varnish are to be treated in a similar manner, 
 taking care in all cases to operate in a dry, warm atmosphere, 
 and avoid exposure to moisture. 
 
 (69.) Application of Varnishes to metallic surfaces. The kind of 
 varnish best adapted to metallic surfaces is given in Formula 
 2 (60) ; but the method of application is the same as before, 
 with similar precautions. The surface must be quite clean, 
 free from grease, and be well got up, nicely polished, and other- 
 wise prepared. For this purpose the surface is to be washed in 
 a weak solution of soda and then in hot water. The process known 
 as lacquering is simple, but requires care and practice. A thin 
 coat of varnish should be laid on, the thinner the better, so long 
 as the metal is covered. A great point for appearance sake is the 
 previous preparation of the surface. The beautiful appearance of 
 philosophical instruments in opticians' shops mainly depends on 
 this. For ordinary practical purposes, however, it will be 
 sufficient for the electrician, if his apparatus be prepared in a less 
 costly manner. The bodies to be varnished must be exposed to 
 a moderate degree of heat, and the varnish applied to them by fine 
 camel-hair brushes. "Well polished copper and brass should, after 
 varnishing, be slightly rubbed over with a clean rag or flannel 
 moistened with finely drawn linseed oil. 
 
 (70.) Varnishing Wood Surfaces. The formula No. 3 (61) is a 
 good varnish for wooden apparatus generally. It is to be applied 
 by means of wide camel's-hair, or other brushes, as in the former 
 instances. The piece to be varnished should be well got up, 
 and gently warmed before varnishing, after which it should 
 be placed in a warm atmosphere. The surface should be finally 
 rubbed over and polished off with a clean rag, moistened with 
 finely drawn linseed oil. Wooden pillars, and apparatus which 
 admit of being turned in a lathe, are best varnished before removal 
 from it; being finally polished off with a clean oiled rag 
 
174 ARTICLES AND PROCESSES IN 
 
 while in motion. The most perfect varnish for flat surfaces is 
 French Polish (63). The polishing process is somewhat tedious ; 
 there are two or three different methods. The following is very 
 effectual. Roll up a piece of thick flannel, cloth, or listing into a 
 cylindrical wad. Moisten one end of this wad freely with the 
 varnish given in formula No. 4, or any other kind of lac- varnish. 
 Formulae Nos. 3 and 4 answer very well. The wad wetted 
 with the varnish, is to be covered with clean linen, moistened with 
 good drying linseed oil ; the surface to be polished is now rubbed 
 with this wad, first in the direction of the fibres of the wood, and 
 then in small circles, completing a small portion at a time, until the 
 surface is covered. Finally the surface is to be rubbed in a similar 
 way with a clean cushion moistened with alcohol, by which the 
 polish is perfected. This is termed spiriting. There are two 
 other methods of applying the polish. A cylindrical wad is to 
 be made as before, faced up with some folds of fine linen. A 
 little linseed oil is applied by the finger to that part of the 
 wad intended for polishing. A small quantity of the varnish is 
 then allowed to drop out from the mouth of the bottle contain- 
 ing it, upon the oiled portion of the rag. The surface to be 
 polished is rubbed with this as in the former case, repeating the 
 application of the varnish and oil when necessary, that is, so soon 
 as the wad is rubbed dry. This operation requires some patience 
 and practice. The surface; as in the previous case, is to be rubbed 
 first in the direction of the fibres of the wood, and then in small 
 circles. This is the English method. The French first moisten 
 the cushion or wad with the varnish and a little olive oil, and 
 rub the wood in the direction of the fibres, repeating the opera- 
 tion several times until the wad becomes dry. The wood is then 
 rubbed with a cushion moistened with olive oil and a little tripoli. 
 The finishing lustre is arrived at by friction with a soft skin, such 
 as chamois leather. The process of varnishing by French polish 
 being tedious, and in many instances not required, it will be 
 sufficient for all ordinary purposes to apply the polish, Formula 4, 
 by a wide camel-hair brush in the usual way. When dry 
 the surface is polished off with a clean rag moistened with quick 
 drying linseed oil. 
 
 There are several other varnishes applicable to wood ; those 
 before enumerated, however, will be sufficient for general 
 purposes. The following, nevertheless, may be occasionally em- 
 ployed with advantage : 
 
 Formula 5. Copal, coarsely powdered 3| ounces 
 
 Camphor 10 grains. 
 
 Highly rectified methylated spirit . . 28 ounces. 
 
ELECTRICAL MANIPULATION. 175 
 
 These ingredients should be put into a clean bottle, and digested 
 in a sand heat, until the solution is complete. The heat should 
 be such as to cause bubbles to rise from the bottom of the 
 bottle. Put the solution by for a few days to settle, and pass 
 the supernatent clear liquor through a paper filter. We have 
 then a fine bright solution, which is to be applied by means 
 of a brush in the usual way. This varnish does not dry so quickly 
 as lac- varnish, but when dry has a very bright appearance, 
 especially after being rubbed over with an oiled cloth. The 
 residue may be treated with more rectified naphtha, and a second 
 produce obtained. 
 
 Formula 6. Seed-lac 3 drachms. 
 
 Gum mastic 2 ,, 
 
 Venice turpentine 1 
 
 Highly rectified methylated spirit . . 2 ounces. 
 
 Digest carefully in a sand heat, and proceed as before. 
 
 It may be as well to observe that in the application of varnishes 
 to wood by the brush in the usual way, the grain of the wood, 
 however smooth the surface, is liable to become raised, giving it 
 a coarse appearance. If we wish to obtain a very smooth varnished 
 surface, we must get rid of this by rubbing down with fine glass- 
 paper until a new and perfectly smooth surface is obtained. This 
 done, a second coat is to be applied as before. To avoid the 
 absorption of the varnish by the wood, a coating of thin isinglass 
 is sometimes first applied and, previously to varnishing, rubbed 
 smooth when dry. Lac being rendered soluble in water by means 
 of borax, we may thus obtain a species of lac varnish of a simple 
 and most useful kind. 
 
 Formula 7. Shell-lac 1 ounce. 
 
 Borax ith 
 
 Water 6 ounces. 
 
 Put these ingredients into a clean wide-mouthed bottle, and 
 raise the temperature to nearly a boiling heat by means of a sand 
 bath. In a short time a clear solution of the lac and borax is 
 obtained. Set it by to cool, and then pass the solution through a 
 paper filter. This is a useful varnish, equal to many spirit var- 
 nishes. It is valuable as a vehicle for water colour and is the 
 basis of Hindoo ink. When once dry, water has no effect 
 upon it. 
 
 (71.) Varnishing Paper. It is often requisite to protect paper 
 surfaces against dirt and moisture, more especially graduated paper 
 
176 ARTICLES AND PROCESSES IN 
 
 scales, circles of cardboard, and such like, as also to render a dry 
 paper surface more or less insulating. 
 
 Formula 8. Dissolve 1 ounce of Canada balsam in 2 ounces of spirits 
 of turpentine. Put these ingredients into a clean, 
 wide-mouthed bottle, and digest at a gentle heat. 
 
 Before the solution is quite cold, pass it through a paper filter 
 into a clean bottle. A heated iron (18) should, be held over the 
 filter in order to preserve the solution in a sufficiently liquid state. 
 
 Mode of Application. The paper to be varnished must first be 
 prepared with a coating of clear, thin, isinglass size, and well 
 dried. The varnish being poured out into a small cup with a 
 cross*wire (64), should be exposed to a gentle heat, and then laid 
 on rather quickly, taking care to sweep the brush in one direction 
 only, and not to touch the same part twice. If one coat be not 
 sufficient, repeat it. 
 
 (72.) The following is a fine varnish for paper : 
 
 Formula 9. Gum anime and gum sandaric, of each . ^ ounce. 
 Gum mastic ......... ,, 
 
 Highly rectified methylated spirit . . 8 ounces. 
 
 Reduce the gums to a very fine powder, put them into a clean 
 capacious bottle with the methylated spirit, and digest in a gentle 
 sand-heat, keeping the bottle closed. When the solution is com- 
 plete, pass it through a paper filter* 
 
 Mode of Application. The paper to be varnished must first have 
 a thin coating of isinglass size. As this varnish dries quickly, 
 care must be taken in laying it on to rapidly pass the brush over 
 the surface and not to touch the same part twice. 
 
 Isinglass size is obtained by dissolving small strips of isinglass 
 in boiling water. The solution should be thin, and filtered through 
 paper whilst hot. As the solution solidifies in cooling, it will be 
 necessary to keep it in a liquid state by standing the vessel con- 
 taining it in boiling hot water, as in the case of common glue. 
 
 (73.) Amber has been occasionally employed as a basis for 
 electrical varnishes. The preparation of Amber Yarnish is diffi- 
 cult, tedious, and often dangerous. It is best procured from the 
 varnish maker. 
 
 Amber, when fused and dissolved in oil of turpentine, constitutes 
 a brilliant electrical varnish. It dries speedily, does not crack, 
 and is very durable and hard when dry ; it resists boiling water 
 and friction, withstands the stains of ink or coloured liquids, and 
 maintains its splendour for any length of time. 
 
 The difficulty, however, of preparing and even obtaining good 
 
ELECTRICAL MANIPULATION. 177 
 
 amber varnish is such, as to greatly discourage its use, more 
 especially as the lac varnishes previously described (59, &c.), are 
 efficient. 
 
 (74). Application of Varnish and Oil to Silk. Fine oiled or 
 varnished silk is of great value in electrical experiments, and is 
 much used in the construction of the electrical machine. The silk, 
 however, for this purpose should be oiled or varnished on one 
 side only. 
 
 The following method of varnishing silk for electrical appa- 
 ratus is very effectual and simple. A stout, open, rectangular 
 frame of wood, similar to a picture- frame, is prepared, of any 
 required dimensions say 4 feet long by 15 inches wide in the 
 clear. The silk is to be first strained upon this open frame, secur- 
 ing its edges to the wood by means of fine tacks. The silk being 
 thus strained is to be brushed over on one surface with a mixture 
 of boiled oil and turpentine, in the following proportions : 
 
 Formula 10. Boiled oil 6 ounces. 
 
 Clear spirits of turpentine .2 ,, 
 
 Let these ingredients be well stirred together in a clean vessel; then, 
 with a fine painter's tool dipped in the mixture, and the super- 
 fluous fluid allowed to run off over a wire (64), brush over one sur- 
 face of the silk from end to end, and set it by to dry. When 
 dry, give it a second coat, if necessary. The mixture of oil and 
 turpentine should not be very thickly laid on. Silk may also be 
 varnished over with the Canada balsam varnish (71) in a similar 
 manner. For the purpose of the electrical machine the thinnest 
 kind of silk, called Persian, should be employed. Thicker kinds 
 of silk strongly varnished are requisite for other purposes. 
 
 (75.) Electrical Cements. Cements of different kinds are 
 requisite in electrical manipulation, for attaching brass to glass, 
 or glass to wood, &c. For cementing glass rods into wood or brass 
 caps, and such like, there is no cement equal to the best sealing- 
 wax, which consists of shell-lac, boiled Venice turpentine, and 
 vermilion, or some other colouring matter. A coarser kind of 
 sealing-wax may be occasionally employed, for which the follow- 
 ing is a good formula : 
 
 Formula 11. Shell-lac 2 ounces. 
 
 Powdered resin, of the finest quality . 1 ounce. 
 
 Fine Venice turpentine 1 ,, 
 
 As much vermilion, or other colouring matter, as 
 required. 
 
 Mode of Preparation. First dissolve the lac in a clean pipkin ; 
 when nearly fluid add the resin, stir well together, then add the 
 
 N 
 
178 ELECTRICAL MANIPULATION. 
 
 turpentine, and, lastly, the colouring matter. Stir all the time, 
 and pour it out into a mould of oiled tin, or upon an oiled marble 
 slab, to cool. 
 
 The fine polish usually seen on the best sealing-wax is obtained 
 by exposing the sticks, when removed from the moulds, to the 
 heat of a fire, just sufficient to liquefy the surface. 
 
 The composition of black sealing-wax is the same as the red, 
 with the exception of the colouring-matter, which for black seal- 
 ing-wax is usually lamp-black, or ivory-black ; but we may use 
 the composition without any colouring matter. For general use 
 it will be well to procure sticks of the very best sealing-wax from 
 the makers. "Whatever colouring matter be employed for sealing- 
 wax it should be ground down on a marble slab with a fine muller, 
 as in the preparation of delicate oil colours. 
 
 (76.) Mode of Application. In cementing brass caps, sockets, 
 or other metallic surfaces to glass, the glass must be heated over 
 a charcoal fire, or a hot iron, until the sealing-wax cement rubs 
 off freely upon the part to be cemented. The brass cap, or other 
 surface, must be treated in the same way, and applied whilst hot 
 to the glass. The two surfaces are then rubbed together by a 
 gentle movement of the brass and glass upon each other, apply- 
 ing external heat to the metal to further liquefy the cement if 
 necessary. The exact position of the metallic socket, or other 
 body, being determined, the whole is left to cool. Before it is 
 fixed, all superfluous cement should be carefully removed. When 
 cold the joint is cleaned off with a little rectified naphtha. In 
 this way a perfectly air-tight joint may be obtained. The 
 metallic parts of the Thermo-Electrometer (178) are secured in 
 this way, as also wood to glass or metal. All joints to be cemented 
 are well- secured by the same process. 
 
 Formula 12. Yellow resin 16 ounces. 
 
 Bees'-wax 2 ,, 
 
 Linseed oil 2 ,, by measure. 
 
 Melt these together in an earthenware pipkin so as not to 
 boil and become frothy. Stir the ingredients well, and add, by 
 degrees, red ochre, 4 ounces ; or colour with any other better 
 colouring matter if desired. If required to be of a very fine colour 
 we may add vermilion. The best way of melting these ingre- 
 dients together is to place the pipkin containing them in a sand 
 heat. This cement is useful for ordinary purposes, and in cases 
 where sealing-wax is not required. 
 
 Formula 13. Yellow resin 5 ounces. 
 
 Yellow Lees' -wax 1 ounce. 
 
 Venetian red 1 ,, 
 
 Plaster of Paris 
 
PREPARATION OF PLATES, CYLINDERS, ETC. 179 
 
 Prepared and applied as before. 
 
 Formula 14. Common resin 5 ounces. 
 
 Bees' -wax 1 ounce. 
 
 Plaster of Paris ....... 2 tablespoonfuls. 
 
 This last is used for coarse purposes, and when applied the sur- 
 faces must be gently heated. 
 
 In preparing these cements, the colouring matter and the plaster 
 of Paris must be well-dried and gradually mixed with the resin 
 and bees' -wax whilst in a state of liquefaction. The heat should 
 be continued just above the boiling-point, until the mixture 
 becomes perfectly tranquil ; it is then allowed to cool, but before 
 it is solid it is cast into cakes or rolls on an oiled marble slab, 
 and preserved for use. 
 
 For roughly cementing glass to wood a compound of plaster of 
 Paris, white lead, and spirits of turpentine may be employed. 
 
 Formula 15. White lead 2 ounces. 
 
 Spirits of turpentine ..... 1^- ounce. 
 Plaster of Paris 2| ounces. 
 
 (77.) Preparation of Electrical Plates and Cylinders. Cylinders 
 and plates of brimstone, lac, resin, and other resinous bodies are 
 often called for in experimental electricity, more especially in the 
 construction of the Electrophorus (92). Brimstone is particularly 
 available, as it easily detaches itself, by contraction, from the 
 moulds in or on which it is cast. If the moulds be very smooth 
 it comes away with a beautifully polished surface. Glass tubes 
 of various sizes furnish very efficient moulds for cylinders of 
 brimstone. If a little taper, so much the better. 
 
 (78.) Preparation of Brimstone Cylinders, Cones, fyc. Put 
 some common brimstone, broken into pieces (37), into a clean earth- 
 enware pipkin ; place the pipkin in a strong sand-heat, raising the 
 temperature gradually until the brimstone liquefies. The heat 
 must not be pressed too far, but must be just sufficient to render 
 the brimstone freely fluid. It is then ready for casting. The moulds 
 being duly prepared, pour the fluid brimstone into them and let 
 it cool. After a short time the mass contracts and may easily be 
 removed from the mould, when it will have a beautifully polished 
 surface. If cast in conical glass moulds, such as wine-glasses, 
 we obtain very good insulating supports. Brimstone cylinders 
 are excellent for exciting negative electricity. When cast in a 
 short cylindrical lamp- glass, and a glass rod put vertically into 
 
 N2 
 
180 ELECTRICAL MANIPULATION. 
 
 the fluid mass before cooling, to serve as an. insulating handle, 
 we obtain a convenient form of electric for negative excitation. 
 
 (79.) Brimstone Plates. Prepare the brimstone in the way just 
 described (78), and place a wooden ring of any given dimensions 
 upon a plate of flat sheet glass, placed pretty level on a table. A 
 few leaden weights may be put upon the ring, so as to press it 
 firmly upon the glass. The wooden ring may be about half an inch 
 or more in thickness, and should be well polished within ; the brim- 
 stone being perfectly fluid, pour it upon the glass within the ring 
 until level with its surface. When completely cool, the casting may 
 be separated from the glass, and on being turned over will present 
 a circular plate with a remarkably fine, smooth, polished surface. 
 This casting should be supported on a plate of wood, having a 
 raised edge. Brimstone castings produce negative electricity very 
 freely, and the only objection to them is their liability to crack. 
 
 (80.) Cylinders and plates of resin, wax, and other substances 
 of that kind, are produced in a similar manner ; the moulds, how- 
 ever, in which they are cast must be oiled to prevent adhesion. 
 
 Resinous plates, compounded of various resinous bodies, are of 
 great value to the practical electrician; as in the case of sulphur (79), 
 they may be cast upon a polished marble surface within a ring of 
 wood or metal a little more than half an inch deep, or upon a glass 
 or polished metallic surface. Block tin may be employed for this 
 purpose. The surface on which the composition is to be poured, 
 together with the sides of the ring, should be rubbed over with a 
 piece of clean cloth slightly smeared with oil, to prevent the casting 
 from sticking. If the ring be of wood, its interior surface should 
 be covered with some very thin tinfoil : this admits of the plate 
 when cast being readily disengaged from the wood ring. 
 
 The following composition for a resinous plate for ordinary 
 purposes will be found sufficient : 
 
 Formula 16. Resin 8 parts. 
 
 Gum lac 1 part. 
 
 Venice turpentine 1 ,, 
 
 As much ivory black as will give the mass a good 
 appearance. 
 
 Break up the resin and lac into small pieces, or reduce to a 
 coarse powder (37) ; put the powdered resin and lac into an iron 
 kettle, or into an earthenware pipkin, and expose them to a sand 
 heat, or to the heat of an oven, of a sufficiently high temperature 
 to liquefy the resin and lac ; as soon as the resin and lac begin 
 to melt, add the Venice turpentine, and continue stirring the 
 mixture with a clean glass rod. When the whole is perfectly fluid, 
 
CYLINDERS AND PLATES OF RESIN, WAX, ETC. 181 
 
 add colouring matter by degrees until a sufficiently deep colour 
 is obtained. The fluid should be retained in a state of quiet 
 fusion for a short time, so as to expel all the air from it. It is 
 now poured out into the mould, and allowed to cool. When 
 perfectly cold, the casting may be detached from the surface or 
 mould, and turned over on a circular plate of wood. 
 
 A resinous plate of this kind, about 15 inches in diameter, is 
 well adapted to experimental purposes ; amongst others, to the 
 separation of certain mixed powders, as brimstone, red lead, &c. 
 &c., by positive and negative electricity. 
 
 The following is well adapted to an electrophorous plate : 
 
 Formula 17. Shell-lac, Venice turpentine, and resin, in equal parts. 
 
 Melt these ingredients in a clean, covered iron crock or pipkin 
 placed in a sand heat, and run the liquid out upon a polished 
 surface within a metallic or other ring of full half an inch in depth, 
 taking care to prepare the surfaces by wiping them with an oiled 
 cloth. 
 
 A plate of this kind, of from 12 to 15 inches in diameter, con- 
 stitutes a very powerful electrophorus plate. The plate may be 
 coloured with ivory black, prussian blue, vermilion, or red lead, 
 as may be desired. 
 
 The following is another formula for an electrophorus plate : 
 
 Formula 18. Gum lac 10 parts 
 
 Eesin 3 ,, 
 
 Venice turpentine 2 ,, 
 
 White wax 2 ,, 
 
 Pitch of a part. 
 
 Liquefy these ingredients as before, and cast them within a ring 
 on a polished surface. This is an efficient combination. The 
 liquid may be coloured with a little vermilion if desired. 
 
 Either of the formulae for the coarser kinds of sealing-wax (81) 
 answer very well for electrophorus plates. The ordinary ingre- 
 dients for an electrophorus plate are resin and bees' -wax. 
 
 The following, by PfafF, is a simple formula : 
 
 Formula 19. Gum lac 1 part. 
 
 Eesin 8 parts. 
 
 Venice turpentine 1 part. 
 
 There is some little disadvantage in casting the plate upon an 
 oiled surface, which if possible should be avoided. It is, however, 
 necessary, to prevent adhesion. We may often succeed by casting 
 the plate upon a highly-polished marble surface, especially if 
 the liquid be cast at a low temperature. Compounds contain- 
 ing a large proportion of shell-lac congeal rapidly. In all 
 
182 ELECTRICAL MANIPULATION. 
 
 cases the surface should be freed as much, as possible from 
 the oil upon them, which is best done by washing them over 
 with a little soft soap and warm water. The common method 
 of casting electrophorus plates is to pour the liquid upon a tin 
 plate with a raised edge, or upon a wooden plate covered with tin- 
 foil, allowing the liquid to cool. If this operation be carefully 
 performed, we may obtain a very level, brilliant surface. The 
 following method of preparing an electrophorus plate will be 
 found efficient and available : Melt the ingredients in a clean, 
 covered, iron crock, which is best done by exposing them to the 
 heat of a close oven, until they become perfectly fluid and free from 
 air-bubbles. Let a polished plate of marble be now closely 
 covered with a thin sheet of tinfoil, and carefully levelled. Place 
 upon the wooden ring- also covered with tinfoil, and which is to 
 serve as a mould some leaden weights, so as to press the ring 
 closely upon the surface of the marble below. Let the melted 
 ingredients, now in a perfectly fluid state and free from air- 
 bubbles, be poured gently within the ring until the whole interior 
 be filled up to its edge. A perfectly level and brilliant surface 
 will in this case be the result. Sufficient time must now be given 
 for cooling, and nothing must be brought near the recently lique- 
 fied surface until it is perfectly hard. When cold, the resinous 
 plate is easily disengaged from the tinfoil surface and from the 
 ring surrounding it. The ring may be divided in one point so 
 as to admit of slight extension, it being temporarily held together 
 by a few turns of fine twine. 
 
 To produce an electrophorus plate by the last method, select a 
 composition such as Formula 16, which when cold will congeal 
 into a very hard substance. When the cake is removed from the 
 mould, the surface may be rendered perfectly level and fine in a 
 lathe. Brimstone, cast in the way above described, presents the 
 finest surface ; and is perhaps, of all substances, the most efficient 
 for an electrophorus plate. It ' is, however, very liable to crack 
 from change of temperature. This may be in some degree avoided 
 by repeated liquefaction of the brimstone before casting, and by 
 gentle annealing. With the view of annealing, the cake, on 
 removal from the mould, should be exposed to a moderately 
 low temperature in an oven. In this way we obtain a plate having 
 a very excellent exciting surface derived immediately from con- 
 gealation. The opposite surface is not usually so perfect. The cake, 
 however, if sufficiently hard, may be fitted into a wooden platter, 
 and polished in a turning lathe, when it will be easily adapted 
 to the purpose of an electrophorus, or to any other use for which 
 
CEMENTING GLASS PLATES. 183 
 
 a resinous plate is required. We should be aware, however, of the 
 liability of plates of this kind to yield or buckle under their own 
 weight. Hence it is necessary, if we wish to preserve their form, 
 to keep them supported upon a flat surface, which is best done by 
 placing them on a circular plate of wood with a raised edge. 
 
 Whatever colouring matter be employed should, previously 
 to mixing it with the other ingredients, be ground down with 
 a muller upon a marble slab, as in the case of sealing-wax (75). 
 No more colouring matter should be employed than is just suffi- 
 cient for the purpose, more especially where black is used : a very 
 little will suffice to colour the liquid mass. Light blue is perhaps 
 the best colour for an electrophorus plate, although black has a 
 very brilliant appearance, and for many experiments is to be pre- 
 ferred. 
 
 (81.) Junction of Plates of Glass ly Cementing. This is an im- 
 portant process in experimental electricity. Many kinds of thin 
 glass, such as window- glass, being very excitable, two thin plates of 
 window glass,when firmly united by means of good sealing-wax or 
 electrical cement, may be substituted with advantage for plate- glass 
 in the plate electrical machine, and may also be employed as a 
 glass electrophorus. The best cement for the purpose is the com- 
 mon kind of sealing-wax, black or red. Formula JSTo. 11 may be 
 employed. The following formulae for an inferior class of sealing- 
 wax may also be used : 
 
 Formula 20. Shell-lac 8 parts. 
 
 Venice turpentine ...... 4 
 
 Colophone, or black resin .... 3 ,, 
 
 Ivory black, sufficient to colour the mass. 
 
 Formula 21. Shell-lac 4 
 
 Black resin 1 
 
 Venice turpentine 2 ,, 
 
 Vermilion 1 ,, 
 
 Add the vermilion gradually, when the other ingredients have become 
 fluid. 
 
 Formula 22. Shell-lac 2 parts. 
 
 Venice turpentine 8 ,, 
 
 Black resin 4 ,, 
 
 Vermilion 1 ,, 
 
 Eemove from the fire and add ~ ounce of rectified spirit. 
 
 There are a great variety of formulae for sealing-wax, but the 
 ingredients in nearly all of them are shell-lac, resin, Venice tur- 
 pentine, and colouring matter. These are united in various propor- 
 tions. The best sealing-wax is made without resin, the proportions 
 being : 
 
184 ELECTRICAL MANIPULATION. 
 
 Formula 23. Shell-lac 4 parts. 
 
 Venice turpentine . . . . 1 to If ,, 
 Vermilion 3 ,, 
 
 Melt the lac in a clean copper pan suspended over a clear charcoal 
 fire; then add the turpentine slowly, and afterwards the ver- 
 milion, stirring briskly all the time. 
 
 This forms a beautiful electrical cement for refined purposes. 
 For inferior kinds of wax we have less lac. The colour may be 
 chosen according to taste. 
 
 The colouring matters commonly employed are vermilion, cin- 
 nabar, red lead, red ochre, carmine, lamp black, ivory black, 
 Prussian blue, artificial ultramarine, and English umber. King's 
 yellow and Prussian blue produce a green colour. 
 
 The colouring matter must be rubbed down with a little linseed 
 oil and turpentine to a fine impalpable powder, by means of a 
 muller and marble slab. As already mentioned (76), the colouring 
 matter must be gradually added to the melted ingredients, and 
 carefully stirred in. The liquid wax thus prepared is to be run 
 into oiled moulds, which may be hollow cylinders of block tin from 
 half an inch to 1 inch in diameter, rather conical ; or it may be 
 run out upon an oiled marble slab, and formed into rolls before 
 being quite cold. If the congealed wax does not readily escape 
 from the tin moulds, it may be easily detached by a gentle 
 heat. In cementing together glass plates of 18 inches to 2 feet 
 in diameter, the surfaces of the plates should be clean and dry ; 
 the sealing-wax by which their union is to be effected should be 
 reduced to a moderately fine powder by the method already 
 described (37). The glass plates should be supported upon a 
 wooden platform, constructed of pieces of deal board, three- 
 quarters of an inch thick, screwed upon two stout cross-bars of 
 wood, at least two inches wide and one inch thick, placed edge- 
 ways. This enables us to get the hand under the platform, and 
 to move the whole in and out of a heating oven or drying-closet. 
 A few small circular or square pieces of deal, about 1 inches in 
 diameter and half an inch thick, being now distributed upon 
 various points of the platform, one of the circular glass plates 
 to be cemented, covered with pulverised sealing-wax, uniformly 
 and carefully spread over its surface to a depth of about the 
 eighth of an inch, is laid upon the pieces of support, and the 
 whole placed in an oven, sufficiently heated to liquefy the wax. 
 When the wax is just fused and covers the glass, it is to be 
 removed, together with the platform upon which it rests. The 
 other circular plate to be united is subjected to the same process. 
 
ISINGLASS, GLUE, SIZE, ETC. 185 
 
 As soon as the wax is fluid this plate is also removed and the first 
 plate inverted upon it, so as to bring the two wax surfaces together. 
 A circular plate of glass of 3 or 4 inches in diameter, if the compound 
 plate be intended for the plate of an electrical machine, should be 
 now placed centrally between the two plates, so as to strengthen 
 the centre. Four small circular plates of glass, of about one inch 
 in diameter, should be also placed between the two plates near the 
 circumference at cross diameters, in order to preserve an equable 
 thickness of the compound plate ; a few leaden weights should be 
 now laid upon thin pieces of deal, and placed at various points upon 
 the upper glass plate, so as to press the plates gently together ; 
 the whole should be now restored to the heating oven in order to 
 completely liquefy the intermediate cement and run the plates 
 together, as it were, under the gentle pressure of the leaden 
 weights. The whole is then to be again removed from the oven, 
 and allowed to cool gradually, taking care that the circumferences 
 are fair one with the other. We may, during the cooling, easily 
 move one of the plates upon the other, and adjust the position, 
 giving gentle pressure with the finger if there should be any lack 
 of contact with the glass. The union of large plates in this way 
 requires much care. The temperature must not be pushed too far, 
 only sufficient to melt the wax, and very little more. The joined 
 plates, after being removed from the oven, should be meddled 
 with as little as possible. When cold the superfluous edges of the 
 wax are to be cut away with a warm knife, and any small inter- 
 stices between the edges of the plates filled in with a little melted 
 wax run out with a lip-ladle. Finally, we may smooth over 
 the wax at the circumferences of the two plates with a heated 
 iron. 
 
 The excitability of a plate of glass thus compounded is most 
 remarkable, and is very superior to that of ordinary plate- 
 glass. The vitrefied surface of flatted blown glass appears to 
 be very favourable to electrical excitation. The plate may be 
 drilled with care, and mounted on an axis in the usual way. 
 The best colours for such a plate are either red or black. Black 
 is perhaps the best for an electrophorus plate. A plate of this 
 kind answers well. Smaller plates, from six inches to a foot or 
 more in diameter, are easily joined by the same process. As 
 also small square plates, from a few inches to a foot square. The 
 best kind of glass for electrical excitation appears to be crown- 
 glass, the pieces being selected as flat as possible. 
 
 (82.) Isinglass, glue, and size form a useful cement. It is easily 
 obtained by dissolving the isinglass of commerce in clean water. 
 
186 ELECTRICAL MANIPULATION. 
 
 The solution should be gently boiled, and for nice purposes 
 filtered ; for coarse purposes, strips of parchment may be sub- 
 stituted for the isinglass. The proportion of isinglass will depend 
 on the required strength of the size. As a preparatory size for 
 wood or other bad conductors, previously to the attachment of 
 metallic leaf by paste, one part of isinglass dissolved in twenty 
 parts boiling water by weight will be found sufficient ; for the more 
 adhesive purposes of glue, one part of isinglass may be dissolved 
 in ten parts water. 
 
 (83.) Rice Glue, termed Japanese Cement. Mix together inti- 
 mately rice in fine powder, and cold water. The rice flour and water 
 should be well rubbed up in a clean porcelain mortar, and be 
 then gently boiled. This forms a highly adhesive cement, beauti- 
 fully white, and when dry nearly transparent. 
 
 (84.) Paste. This is also a useful cement for practical electrical 
 purposes, such as for attaching metallic leaf to glass, as in the 
 preparation of the Ley den-jar ; as also in covering wooden and 
 other substances with tinfoil. Good flour paste may be made as 
 follows :-^- 
 
 Formula 24. Kub up some good flour in a porcelain mortar with as 
 much cold water as will produce a liquid fluid. 
 
 Care must be taken not to leave any lumps of flour, but the 
 whole must be perfectly smooth. Add about a drachm of finely 
 powdered alum. Put into a clean tinned iron saucepan nearly 
 as much clean water as there is of flour and water, and set it over 
 the fire to boil ; when boiling pour the mixture previously pre- 
 pared into it, and keep stirring the whole with a large wooden 
 spoon until it becomes an adhesive mass, which it will speedily do. 
 When it shows symptoms of boiling remove it from the fire, con- 
 tinuing to stir the mass. Transfer it by the aid of a clean wooden 
 spoon into a clean white open-mouthed jar, and set it by to cool. 
 The paste used by curriers and shoemakers, which is especially 
 good, is prepared somewhat in this way. It is adapted to the 
 coating of the electrical jar with tin-leaf: 
 
 Formula 25, Good flour 2| ounces. 
 
 Coarse brown sugar 20 grains. 
 
 Water 5 ounces. 
 
 Isinglass size 5 ,, 
 
 Essential oil of lavender .... 20 drops. 
 
 Rub up the flour, water, and sugar in a clean "Wedgwood ware 
 mortar, adding the water gradually, until the whole is in a perfectly 
 liquid state ; add now the isinglass size which should be hot 
 and perfectly fluid together with the essential oil ; rub the whole 
 
ELECTRICAL SUSPENSIONS. 187 
 
 together so as to unite the ingredients more completely ; pour the 
 mixture into a clean saucepan containing a little boiling water, 
 and boil the whole gently over a slow fire, stirring it until it 
 thickens into a paste. Finally remove the paste into a covered 
 jar for use. 
 
 Paste is usually a very perishable article, but made in this way 
 it will keep in a covered vessel for a long time. The sugar 
 renders it pliable, and the essential oil secures it from the minute 
 vegetable growth which turns it mouldy. This kind of paste is 
 very adhesive, and is especially adapted for covering wood with 
 tin-leaf, the wood being previously coated with isinglass size. 
 Even if this paste dry up into a hard mass, it may be recovered 
 by diluting it with a little hot isinglass size. 
 
 (85.) To cover Wood with Tm+Zeaf.-^lt is desirable to prepare 
 the wood by first giving it a coating of fine isinglass size, and 
 allowing the size to dry. The tin- leaf with which the wood is to 
 be covered is then cut to the required dimensions and form best 
 adapted to different parts of the surface. The portion to be laid 
 on the wood must be brushed lightly and equably over with the 
 paste. The pasted foil is now laid on the wood surface, and 
 pressed down upon it by means of a damp cloth or sponge, and 
 then further rubbed down upon the wood by gentle friction with 
 the handle of a bone paper-knife, wetted with water. When the 
 whole surface has been covered in this way, clean 'it off with a 
 wet cloth or sponge, and then wipe it dry with a soft nmslin 
 cloth. All projecting edges must be rubbed until they disappear. 
 The surface will then be perfectly even and metallic, as if entirely 
 of metal. 
 
 (86.) Electrical Suspensions. For the perfect development of 
 minute attractive force, great sensibility of movement is requisite. 
 This is obtained by balancing light needles on fine agate centres, 
 or suspending light bodies by filaments of unspun silk or other 
 fibres ; or otherwise, by the suspension of strips of leaf gold. 
 The suspension upon hardened points and centres, after the 
 manner of a delicate and fine compass-needle, admits of great 
 freedom of motion. The centres may be of hardened brass or 
 agate. The points or centres of motion may be of hardened steel, 
 or hardened brass, which is often preferable. The points of sus- 
 pension may either descend from the needle into the glass or agate 
 centres, or the centres may be fixed in the needle, and the points 
 ascend into them, so as, in either case, to have the centre of gravity 
 of the needle a little above the point of suspension. The Chinese 
 have an ingenious method of suspending a needle, so as to obtain 
 
188 ELECTRICAL MANIPULATION. 
 
 remarkable delicacy of movement in their small compasses. It 
 is represented in Fig. 108. In this construction, the centre of 
 gravity of the needle is above the point of sus- 
 pension ; d e is a small cap of brass, very thin 
 rig. IDS. and light ; the needle e s n passes through 
 
 a small ring e of support, formed of a light slip of brass, and fixed 
 on the top of the cap. The whole is sustained on a pin. 
 Although the needle is in this case above the point of suspension, 
 yet the centre of gravity of the whole system, namely, the needle, 
 cap, and ring of support is somewhat below it. The Chinese 
 needle is very small and light ; it is not above 1 inch in length, 
 and the one- fortieth of an inch in diameter ; it is, however, very 
 sensitive. 
 
 (87.) Suspensory Filaments. The most sensitive and delicate of 
 these is the spider's thread, which admits of being twisted some 
 thousand times without exhibiting torsion. It is, however, 
 difficult to handle and apply. The following method will be 
 found successful. Construct two movable radii of light wood 
 united upon a centre pin, after the manner of a pair of common 
 compasses. Each leg should be about 12 inches or more in 
 length, and tapered towards the extremities, and so closely united 
 as to admit of being turned upon each other with some degree of 
 friction. We may thus adapt the opening to any required length 
 within the limits of the legs. When we are fortunate enough 
 to discover a long and strong spider's thread hanging from the 
 ceiling, or one of the long stretching lines of the geometrical web 
 of the garden spider, we open the legs of the compasses to the 
 length of the thread we require and can command. Then touch- 
 ing the extremities of the legs with a little weak gum- water, we 
 intersect the thread between the legs of the compasses which 
 immediately adhere to it ; then by means of a little weak gum 
 attach a fine pith ball to each extremity of the intersected thread, 
 (which now becomes easily detached from its fixed points), holding 
 one of the balls at the extremity of a fine needle, and allowing the 
 other to remain pendent. The other extremity of the needle is 
 now inserted in any fixed arm of support ; and thus we obtain 
 the most delicate suspension. A light pith ball may be success- 
 fully applied to the thread as it hangs from the ceiling, con- 
 stituting a very delicate means of exhibiting attractive force. 
 
 A filament of unspun silk, as unwound from the cocoon of the 
 silkworm, is another very delicate means of suspending light bodies, 
 and may be managed in pretty much the same way as the former, 
 especially where the thread is fine. It generally admits of being 
 
ARRANGEMENT OF GOLD LEAF. 189 
 
 applied through the medium of a very fine sewing needle which 
 may be passed through a fine reed, or may be otherwise attached 
 by a little cement or gum, after being wound round the centre 
 of the needle. A light needle of gum-lac is best constructed by 
 inserting a fine thread of gum-lac in the extremities of a short, 
 delicate reed, about half an inch in length, and then passing the 
 suspension silk through the centre of the reed by means of a fine 
 sewing needle, or otherwise attaching it by a central loop of stouter 
 silk filament supporting the needle on each side of its centre. Fila- 
 ments of glass thread drawn out in the lamp constitute very 
 delicate suspensions for light bodies. They have an extremely 
 small degree of torsion force, which is sometimes of advantage. 
 
 Yery fine metallic wire may also be employed. The suspension 
 wires employed by Coulombe were so fine as to admit of being 
 twisted through eight circles without interfering with its elastic 
 reaction. It was so fine, that 1 foot of the wire weighed only 
 one-sixteenth part of a grain. 
 
 (88.) Management and Handling of Gold-leaf. The suspension of 
 strips of gold-leaf requires much attention and practice. The leaf 
 is handled through the medium of the leather cushion and spatula 
 knives already described (36), and in the following way: The book 
 of gold is laid flat on the cushion towards one of its extremities. 
 Then lifting up the paper of the book by means of the thumb and 
 fingers, so as to expose a leaf of gold, we insert the edge of the 
 spatula for a short distance beneath the edge of the leaf, which 
 must be carefully and gently done ; then, giving the spatula half 
 a turn with the hand so as to raise the edge of the leaf off the 
 paper upon the spatula for a short distance, we drag the gold gently 
 forward until it rests on the leather cushion. We may now blow 
 very gently upon the leaf, which will press it equally upon the 
 leather. This, however, requires a great deal of care and precision, 
 for unless the breath fall perpendicularly upon it, the leaf is liable 
 to be blown off the surface of the cushion. The leaf of gold being 
 thus secured, and the width of the required slip determined on, 
 we adjust the edge of the spatula to the given width or distance 
 from the edge of the leaf, taking care to keep it parallel to the 
 edge, which must be done by an accurate eye ; after which the 
 edge of the spatula is pressed down on the leaf, and drawn along 
 its surface so as to divide it completely through ; then taking a piece 
 of gummed paper, such as is used for letter stamps, and cutting it to 
 the required width and length, which may be about half an inch, 
 and having very gently moistened it at one extremity with the lip, 
 we lay the paper carefully upon the end of the strip of gold, and 
 
190 ELECTRICAL MANIPULATION, 
 
 thus it can be easily detached from the square of gold-leaf to which 
 it was previously united, and raised off the cushion. It may then be 
 placed in the holder of a Gold-leaf Electrometer (Chap. I., ISTote G), 
 or any other required point of suspension. Where two slips of gold 
 are required, as in Bennet's Gold-leaf Electrometer (Chap. I., 33), 
 a second equal slip of gold is prepared in exactly the same way, and 
 whilst one of the slips is resting on the cushion the other slip is laid 
 carefully on it, the two slips with the paper-holders being nicely 
 adapted one to the other so as to lie evenly and parallel, taking care 
 to place the surfaces of the paper to which the gold slip is attached 
 outward, so that the gold attachments may not touch. The double 
 slip of gold with the paper-holders are now transferred to any 
 given point of support, as in the holder of the Gold-leaf Electro- 
 meter above mentioned. If the strips are longer than necessary, 
 they may may be shortened by the same means as before. Gold 
 slips are attached to any other medium of support, such as small 
 wooden cylinders or other surfaces in much the same way, but in 
 all cases the gold-leaf must be handled upon the leather cushion 
 and with the gummed paper. The gold-beaters in handling their 
 leaves commonly rub a little dry whitening over the surface of the 
 leather cushion, and that in such way as to whiten the surface 
 equally over and then removing all superfluous whitening with 
 a dry leather. They also prefer to cut the gold with the sharp 
 edge of a piece of split cane. 
 
 The management of gold-leaf in the way described, although 
 requiring some practice and care, is easily attained. The squares 
 of leaf- gold as prepared by the gold-beaters are about 3f inches 
 square, Squares of silver-leaf are 4'5 inches square. A square of 
 leaf- gold, therefore, contains about 10-8 or nearly 11 square inches. 
 A square of silver-leaf contains 20 -25 square inches. 
 
 (89.) Selection of Glass for Electrical Tubes, and Electrical 
 Machines. All kinds of glass are not well adapted for electrical 
 excitation, indeed some kinds are not excitable at all. The best 
 kind of glass for electrical excitation appears to be white flint glass 
 of good quality ; this should have a fine vitreous polish, and be 
 clear and hard, without blemish. The common white bottle glass, 
 of which apothecaries' phials are made, will be generally found 
 to be very freely excitable some apothecaries' phials powerfully 
 so. Common window or crown glass is also very excitable. 
 
 The constituents of glass generally are silica, lime, oxide of iron, 
 a little alkali, and smaller quantities of other matters. Flint glass 
 contains a large proportion of oxide of lead, which renders it 
 heavier and more fusible than crown-glass. Silica and borate of 
 
CONSTRUCTION OF ELECTRICAL MACHINES. 191 
 
 lead are also employed as being the best ingredients. Glass for 
 electrical purposes should contain as little alkali as possible, since 
 it renders the glass soft and very liable to attract moisture from 
 the atmosphere, and is consequently but little excitable. Glass 
 plates containing an excess of alkali have been found after a short 
 time to crumble in pieces. In olden times the glass of Cherbourg, 
 the crystal glass of England, and the glass of Bohemia were much 
 preferred for electrical excitation, and were observed to be more 
 excitable when they had been long exposed to a high temperature. 
 It was, for example, found by Bose, of Wittemburg, that glass 
 exposed to heat in chemical distillations is incomparably more 
 excitable than glass not so exposed. Hence it was recommended 
 by Priestley to blow cylinders and globes for electrical machines, 
 when the metal had been for some days in a state of fusion. 
 
 (90.) Precautions to be observed in the Construction of the Cylindri- 
 cal and Plate Machines. Almost every kind of glass is more or less 
 applicable in the construction of electrical machines, but since 
 there is much difference in the excitability of different kinds of 
 glass, it is important to select a kind of glass found by experience to 
 be the most excitable. Glass for electrical machines should be well 
 annealed, that is to say, exposed to heat for a long time after 
 manufacture. The crystal or white flint glass of England has great 
 excitable power. Some differences of opinion have arisen amongst 
 the older electricians relative to the glass of different countries, 
 some giving a preference to the clear crystal glass of England, 
 others preferring, more especially for cylindrical machines, the 
 blown glass of Cherbourg, many recommend the clear yellow glass 
 of Bohemia, others the white glass of Brittany. The excitation of 
 green bottle-glass is difficult and weak. The vitreous surface of 
 blown glass has certainly an advantage over the polished surface 
 of glass plate, hence it is that the cylindrical machine has in some 
 instances been found more powerful than the ordinary plate glass, 
 the surface rubbed being of the same extent. 
 
 (91.) Cylindrical Machine. Having been careful in the selection 
 of the glass of which the cylinder consists, we have next to deter- 
 mine the relative proportions of the diameter to the length between 
 the shoulders, that is of the surface to be rubbed. If the diameter 
 be to the length between the shoulders as 8 to 10, that pro- 
 portion will do extremely well. A cylinder of 10 inches in 
 diameter, by about 12^ inches in length, would constitute a 
 powerful machine of a moderate size ; but 12 inches in diameter by 
 15 inches in length between the shoulders, would form a very 
 powerful machine. The cylinder should be carefully blown, so as 
 
192 ELECTRICAL MANIPULATION. 
 
 to turn as equably as possible upon its two necks, when a temporary 
 axis of wood is passed through it. It should be rather flattened at 
 the shoulders. The neck should be straight and open, but not more 
 open than is necessary to admit of nicely cleaning it on the inside. 
 In preparing the cylinder for mounting, it must be made clean 
 and dry, especially upon its inner surface. There should not 
 be a particle of dirt or dust within the glass. Two soft 
 silky corks should then be fitted so closely within the open 
 necks as to completely close them air-tight. Before closing 
 the necks, the cylinder should be preserved in a very dry atmo- 
 sphere, and a day should be selected for shutting up the cylinder 
 when the air is peculiarly dry, and the wind in a northerly or 
 easterly quarter. When the cylinder is closed, the ends of 
 the corks should be covered with melted sealing-wax, which 
 should flow freely over them by the aid of heat, so as to exclude 
 the possibility of any damp air getting through their pores, which 
 it cannot do if the corks be firmly pressed within the necks after 
 the manner of bottle corks. The closed necks should be covered 
 with caps of brass, which are to support the axis of the cylinder, 
 firmly cemented on them. This should be done with the best red 
 sealing-wax (81). Closing up the cylinder, having a perfectly dry 
 atmosphere within, is a most important part of this manipulation ; 
 the success of the whole business depends on it ; for if the slightest 
 moisture be condensed upon the inner surface of the glass, or 
 eventually condense on it from any cause, the action of the machine 
 will be weak, and its efficiency all but ruined. The inner surface 
 becomes more or less conducting, and hence carries off the electri- 
 city excited upon the outer surface. This was probably the great 
 source of failure recorded by Dr. Priestley and the old electricians, 
 in the construction of their apparatus. The most promising and 
 perfect of their glass globes would frequently fail to give a 
 single spark. All possible means were tried with a view to pro- 
 mote the excitation.* The author remembers a case in which a 
 cylindrical machine, very perfect in form, of very large dimen- 
 sions, and blown of the best possible glass, totally failed in its 
 action, and could hardly be caused to emit a spark an inch in 
 length. After numerous trials it was determined to remove the 
 caps covering the open necks, after which the cylinder was com- 
 pletely cleaned out with rectified spirit of wine, and wiped very 
 dry, at the same time very dry warm air was blown into it with 
 the bellows. The necks were then again replaced, and cemented. 
 The consequence was, an extremely powerful development of 
 * Priestley's "History of Electricity," pp. 553, 554, &c. 
 
THE RUBBER OR CUSHION. 193 
 
 electricity, and sparks were obtained from 12 to 15 inches in length, 
 or more ; thus showing the vast importance of preserving the inner 
 surface of the cylinder in a perfectly dry non-conducting state. It 
 is most probably owing to the ruinous effects of moisture, or the 
 humidity of the inner surface of the glass, that the old electricians 
 observed a vast improvement in their globes on coating them inter- 
 nally with resinous cement, which was made to flow over the 
 surface by means of heat. It is worthy of consideration even at 
 the present day, how far it might be desirable, previously to 
 closing the cylinder, to varnish over its inner surface with shell- 
 lac varnish, and dry it off by internal heat. Sigaud de la Fond, 
 a French philosopher who wrote about the year 1781, says 
 that globes for electrical machines are greatly improved by 
 returning them for a few days to the heating furnace, supposing 
 they do not act at first very well ; a remark which applies to glass 
 cylinders. 
 
 (92.) The Construction of an Electrical Rubber or Cushion. The 
 best facing for a rubber is either red basil leather, the rough side 
 of stout black oiled silk, or fine Morocco leather. The surfaces 
 should be fair and even, and have some little degree of elasticity. 
 The cushion is best constructed in the following way : An oblong 
 piece of mahogany about a quarter of an inch in thickness is 
 first provided ; then a second similar, but somewhat thicker, piece, 
 of the exact dimensions required for the cushion, is placed 
 immediately upon it. One edge of the leather or silk to form the 
 face of the rubber is then secured by very fine tacks along the side 
 of the thinner piece of mahogany which is to form the back of the 
 rubber, and then turned and strained firmly over the thicker piece 
 of mahogany resting upon it as a mould. It is now secured 
 in a similar way upon its opposite edge. The mould may now be 
 drawn out from beneath the facing, leaving a fair and regular 
 hollow within. Two or three layers of thick woollen stuff, making 
 up the thickness of the hollow case, being now loosely and evenly 
 sewn together and temporarily secured at one end to a little 
 stout cord, are forcibly drawn by it into the hollow casing. The 
 extremities of the rubbing surface are now neatly folded in and 
 secured to the ends of the back piece. We have in this way a 
 very regular and somewhat elastic surface, which, when in place, 
 applies equally and fairly to the glass. Although thick woollen 
 cloth may be employed with advantage, yet a still better substance 
 is that known as spongeo-piline, prepared for medical purposes, and 
 to be had at any druggist's, consisting principally of sponge and 
 wool, to which are added various fibrous materials, backed by an 
 
 o 
 
194 ELECTRICAL MANIPULATION. 
 
 impermeable surface of gutta-percha or india-rubber ; the whole 
 constituting a light, porous, elastic substance, admirably adapted, 
 when cut into slips, for filling the cushions of electrical machines 
 in the way just described. The oiled silk flap (68), invented by Dr. 
 Nooth, should be carefully applied to the rubber about a quarter 
 of an inch from its edge, should be first secured by very strong 
 paste (84), and the edge of the silk should lie flatly and evenly 
 upon the cushion from one end of it to the other. It should be 
 finally fixed to the rubber by two rows of long stitching of waxed 
 sewing silk. To cover the heads of the tacks securing the facing 
 of the rubber, the edges of the cushion may be bound round with 
 narrow silk bind, fixed by paste, which gives the rubber a neat 
 appearance. The silk flap should be applied with its rough side 
 next the glass, and neatly rounded at its termination opposite the 
 row of points projecting from the conductor. The best way of 
 constructing the row of points is by the insertion of a thin edge 
 of brass along the line of the conductor, projecting from it about 
 one quarter of an inch, upon which are soldered short dumpy 
 points, about three-tenths of an inch apart. 
 
 (93.) Plate Machine. Plate machines are generally constructed 
 of polished glass plate ; the plate being first cast, and then 
 polished. In selecting glass plate for an electrical machine, the 
 polish of the surface should be the most perfect possible, and the 
 glass of the most clear and flinty kind. 
 
 Although well-polished plate is very efficient and generally 
 available in the construction of the plate machine, yet it must 
 be allowed that a polished surface is not, upon the whole, so 
 excitable as a vitreous surface, the vitreous surface of well- 
 blown glass, such, for example, as the best blown glass of Cher- 
 bourg, or the best blown flint-glass of England, being peculiarly 
 susceptible of excitation. Hence it becomes a question how far a 
 compound glass plate for an electrical machine, constructed in the 
 manner already described (81), is not preferable to a single plate of 
 polished plate-glass as usually employed, more especially if the pro- 
 cess of cementing be improved by experience and practice, and the 
 blown glass be manufactured and selected expressly for the purpose. 
 The plates of several electrical machines, from 18 inches to 2 feet 
 in diameter, constructed in this way by the author, were found to 
 be wonderfully powerful, and certainly exceeded ordinary plate 
 machines of the same diameters. The process for obtaining an 
 electrical plate by the junction of two thin plates, in the way just 
 described (81), is as yet open to much further improvement. If 
 the best methods of uniting thin plates of blown glass by inter- 
 
THE AMALGAM. 195 
 
 mediate cement were well understood, investigated, and practised 
 by skilful mechanics conversant with, such kind of manipulation, 
 we should no doubt obtain plates for electrical machines of extra- 
 ordinary power. A compound plate being thus obtained, is to 
 be finished off, nicely rounded, drilled, and mounted on an axis in 
 the usual way. 
 
 (94.) Amalgam for exciting the Electrical Machine. The employ- 
 ment of an amalgam of tin and mercury by Mr. Canton, so long 
 since as the year 1768 (a notice of which will be found in the 
 52nd vol. of the Phil. Trans.), was a most important step in 
 practical electricity. Mercury amalgamated with tin, or zinc, or 
 both, increases the excitation of glass in a most surprising manner. 
 Canton's amalgam was of a very simple form. It consisted of an 
 alloy of tin and mercury, which he directs to be well rubbed into 
 the cushion with a little chalk. This will excite the glass very 
 powerfully, and with little friction, especially if the rubber be some- 
 what damp. 
 
 (95.) Higgins's amalgam, as proposed by him, and noticed in the 
 Phil. Trans, for 1778, consisted of mercury and zinc, in the propor- 
 tion of 4 or 5 parts of mercury to 1 of zinc. This, he says, is 
 better than Canton's amalgam of tin and mercury. The zinc and 
 mercury is reduced by fusion or trituration to the consistence of 
 butter, and converted into a powder by the addition of very dry 
 chalk, well rubbed up in a mortar. An amalgam made of tin 
 and mercury was treated by Canton in the same way. 
 
 (96. ) The Baron Kienmayer, who compounded a powerful elec- 
 trical amalgam, consisting of tin, zinc, and mercury, says, that in 
 every kind of amalgam he had hitherto tried he found three inconveni- 
 ences : 1st. The mercury became separated from the metal with which 
 it was alloyed, and deposited small globules upon the electrical ma- 
 chine. 2nd. The excitation often decreased under the operation of the 
 amalgam. 3rd. The friction was inconveniently great. This led him 
 to employ a combination of the three metals in the proportion of zinc 
 and tin each 1 part, mercury 2 parts, the mercury being equal to the 
 tin and zinc together. This he converted into powder without the aid 
 of chalk, and found it to produce a very powerful excitation. This 
 method of preparation is as follows, and is continued to a greater or 
 less extent up to the present day : The tin and zinc having been 
 purified according to the best chemical process, are melted together 
 in a clean iron ladle or an earthenware crucible, and well stirred 
 together with an iron rod. The mercury, equal to the weight of 
 the two metals taken together, must be gently heated, and be poured 
 into a cylindrical wooden box having a round hole in the cover. 
 
 o 2 
 
196 ELECTRICAL MANIPULATION. 
 
 The alloy of zinc and tin are now removed from the fire and poured 
 upon the mercury through the hole in the box, which must be in- 
 stantly closed with a cork, and the whole kept in a state of ag-itation 
 by shaking the box until the amalgam is cold. It is then poured out 
 upon a marble slab or table. The amalgam will be hard and of a 
 silvery colour. It is to be reduced to a very fine powder in a glass or 
 stone mortar. The mass must not be left long untriturated, other- 
 wise it becomes so hard that there is a difficulty in reducing it to 
 powder. If we triturate the amalgam for a long time, and at 
 several repetitions, it becomes grey, and may, by frequent tritura- 
 tion, become black. But there is no rule for this ; all we have to 
 do is to obtain a fine powder, free from lumps. It should run 
 through the fingers like dust. This powder may be preserved 
 many years in a bottle. The mercury never separates, even by 
 new trituration, which shows that it is intimately combined with 
 the other metals. It is not desirable to make a large quantity 
 at once ; about half a pound is quite sufficient that is to say, 2 
 ounces of zinc, 2 ounces of tin, and 4 ounces of mercury. 
 
 Kienmayer's amalgam has been more or less accepted by modern 
 electricians ; the different metals being amalgamated in different 
 proportions. Mr. Singer observes that amalgam in which the 
 proportion of mercury is considerable is more transient in its 
 effect than when the proportion of mercury is less, at least within 
 certain limits. It is therefore necessary to ascertain the best pro- 
 portion, so as to obtain a steady and long- continued excitation 
 without a frequent renewal of the amalgamated surface. 
 
 The following are a few formulae which have been resorted to 
 with success, by which it will be seen that the metals, tin, zinc, 
 and mercury, form the basis of electrical amalgams, and that they 
 have been combined in all sorts of proportions. 
 
 Formula 26. Tin, 1 
 
 pulver sing CANTON. 
 
 27. Tin, 1 
 28. Zinc, 1 
 29. Zinc, 2 
 30. Tin, 1 
 31. Tin, 1 
 32. Tin, 1 
 33. Tin, 1 
 34. Tin, 1 
 35. Tin, 3 
 
 36. Tin, 3 
 37. Tin, 2 
 38. Tin, 1 
 
 mercury, 2 ; fine chalk sufficient for 
 
 mercury, 1 ,, 
 
 ; mercury, 4 or 5 HIGGINS. 
 
 ; mercury, 5 KEENMAYER. 
 
 zinc, 1 ; mercury, 2 
 
 zinc, 2 ; mercury, 5 ,, 
 
 zinc, 2 ; mercury, 6 SINGER. 
 
 zinc, 2 ; mercury, 3^ , , 
 
 zinc, 2 ; mercury, 3 ,, 
 
 zinc, 5 ; mercury, 9 ; with the addition 
 
 occasionally of a little sulphur HEARDER. 
 
 zinc, 5; mercury, 7 
 zinc, 1 ; mercury, 5 
 zinc, 1 ; mercury, 1 
 
THE AMALGAM. 197 
 
 (97.) Much, care is required in compounding electrical amalgam. 
 The manipulation is not by any means easy. In many instances 
 the zinc oxidizes rapidly in the course of fusion, and some part 
 may fly off, hence a change in the proportion of zinc. Kienmayer's 
 method, as already noticed, is no doubt efficient, but it is still 
 open to failure unless it be carefully carried out. 
 
 (98.) The following process for an electrical amalgam may be 
 relied on. Pure tin and zinc are to be carefully weighed out 
 and put into a covered earthenware crucible, or into a clean, 
 covered iron ladle. The zinc should be first heated, and when 
 about to fuse the tin added. In the meantime the mercury is 
 placed in a separate crucible or small iron ladle, and heated to 
 something above the boiling-point of water. "When the zinc 
 and tin have become fluid, they may be removed from the fire, 
 and the mercury poured gradually on the combined metals, 
 stirring the whole at the same time with a clean iron or steel rod. 
 We must, however, allow the melted metals to cool a little, for if 
 the mercury be poured upon the alloyed tin and zinc at anything 
 like a red heat, it may be driven into vapour before amalgamation 
 takes place, and this vapour not only occasions a loss of mercury, 
 but is very dangerous to inhale. The exact time when to add the 
 mercury is a nice point in the process. The amalgamated mass is 
 now allowed to cool further, and is then poured into a wooden box, 
 as recommended by Kienmayer (96). The box should be kept in a 
 constant state of agitation by shaking until cold. If the amalgam 
 does not contain a large quantity of mercury, it will be commonly 
 found as a fine black powder, which is a proof of the success of the 
 operation. This powder is to be now rubbed down in an iron 
 mortar, and passed through a fine muslin sieve, and then set by 
 in a closed wide-mouthed bottle, or a clean covered box, for future 
 use. If the amalgam contains a larger portion of mercury, this 
 result does not ensue, but the mass is more or less of a pasty cha- 
 racter, and of a bright colour. This is the case with Formula 32. 
 It becomes more compact after some hours' cooling. It is desirable 
 in this case to knock it to pieces in an iron mortar before it has 
 become greatly consolidated. If not containing a very large 
 portion of mercury, it will at length become so pulverised as to 
 admit of being passed through a fine muslin or gauze sieve. All the 
 different forms of amalgam specified (96) should be treated in this 
 way as far as possible. It is not, however, always possible, where the 
 portion of mercury is very large. In this case the amalgamated mass 
 is very soft, and for some time after its formation may be spread 
 almost like butter upon the rubber. The amalgam, Formula 32, is, 
 
198 ELECTRICAL MANIPULATION. 
 
 as just stated, of this character. The actions of these amalgams 
 being not always very certain, it is desirable to keep a number of 
 them prepared in separate bottles, and occasionally combine them. 
 If, for example, an amalgam containing a very small portion of 
 mercury be found too dull, it may be stimulated by adding to it a 
 portion of one of the amalgams containing a large portion of 
 mercury. In this way we obtain an amalgam well adapted to our 
 purpose. It is not always easy to prepare the amalgam to a certainty, 
 whereas by this method we may ensure success. In every case of 
 combination of different amalgams we should rub them well to- 
 gether in an iron mortar, and sift them immediately before using. 
 The author, after a series of careful experiments, found the fol- 
 lowing amalgam very powerful and effective, it being prepared in 
 the way above-mentioned : 
 
 Formula 39. Tin, 1 ; zinc, 2 ; mercury, 4. The ingredienta of tin, 
 zinc, and mercury being double of each other. 
 
 (99.) Mode of Application of Electrical Amalgam. There are 
 two modes of applying amalgam to the cushion of the electrical 
 machine. 
 
 First, by smearing the rubber over with a thin film of pure 
 lard, so as to give it a somewhat greasy surface, and then by means 
 of a fine sieve sprinkling the surface freely with the powdered 
 amalgam. We then pass a spatula, or thin, wide, stiff knife, with 
 pressure over the surface, so as to smooth it down. The knife may 
 be very slightly smeared with fine lard. The surface of the rubber 
 will very often assume a bright metallic lustre, and will then be 
 in a fit state for exciting the machine. Care must be taken not to 
 allow the amalgam to spread over the edge of the silk flap attached 
 to the cushion. 
 
 Second. Rub the powder up in an iron mortar with as small a 
 portion of purified lard as will enable it to spread over and adhere 
 to the surface of the cushion. It should be lightly and equably 
 spread, and not passed beyond the edge of the silk flap. 
 
 This method of application is often a difficult one. The amalgam 
 will frequently break off the surface and not adhere with sufficient 
 freedom, so that the surface becomes irregular and lumpy. In order 
 to spread the amalgam equably and thinly, a large proportion of 
 grease or lard is necessary. This is by no means favourable to 
 excitation. To correct this defect we should finish by sifting some 
 dry powder upon the surface, and smooth it down with a spatula. 
 When the amalgam has been applied for some considerable time, 
 
MOSAIC GOLD. 199 
 
 the excitation often decreases. In this case we find on examining 
 the rubber that the coating of amalgam has become a hard, solid 
 surface, and requires to be renewed. "We may frequently restore 
 the power of the machine at the instant by passing a piece of coarse 
 brown paper or fine glass paper between the glass and the cushion, 
 drawing it a few times forwards and backwards so as to rub up the 
 hard surface and restore it to a more compressible state by a 
 renewal of the original surface. When fresh amalgam is required, 
 the old amalgam must be completely removed, and the surface of 
 the rubber made clean. Mr. Wilson, an electrician of about the 
 middle of the last century, was in the habit of gilding the rubber 
 of his machine, or coating it with silver or some other metal, which 
 he says answers very well. 
 
 The Professor of Natural Philosophy in the Hoyal College of 
 Chartres, France, so long since as the year 1748, moistened his 
 rubbers with water to promote excitation. Dr. Watson also, about 
 the same time, moistened his rubber with a view to promote con- 
 duction. 
 
 (100.) Aur-um Musivum, commonly termed Mosaic gold, has been 
 occasionally employed with success in promoting electrical excita- 
 tion. It is a bisulphide of tin ; and is obtained by preparing an 
 amalgam of 12 parts tin and 6 of mercury. This is reduced 
 to a powder, and mixed with 7 parts of sublimed sulphur and 
 6 of sal-ammoniac. This mixture is exposed to a gentle heat in a 
 flask with a long neck, and sulphurated hydrogen is driven off. 
 When the smell of the sulphurated hydrogen is completely gone, 
 the temperature is raised to a low red heat ; eventually a scaly mass 
 remains. This is a bisulphide of tin, and is of a beautiful yellow 
 colour. If the chemical process has been properly conducted, this 
 yellow mass acts powerfully as an electrical amalgam, and is a very 
 clean preparation. On the contrary, if not perfectly prepared, 
 the compound is a total failure in this respect. The quantity of 
 sal-ammoniac which it may contain is very objectionable from its 
 liability to attract moisture. To avoid this defect the mass may 
 be reduced to a powder, and washed on a filter until all the 
 sal-ammoniac is got rid of. It is then to be carefully dried 
 upon heated paper, and applied as a powder without the aid of 
 grease. It has been found advantageous to smear over the surface 
 of the rubbers with aurum musivum. If the preparation be very 
 pure it may be applied at once to the rubber from the lump, which 
 soon yields up a golden surface. As yet, however, no other pro- 
 cess has been devised for producing this substance by direct union 
 
200 ELECTRICAL MANIPULATION. 
 
 of the ingredients. In this process part of the ingredients employed 
 are driven off by heat, and seem to act partly mechanically in 
 giving the required texture to the product, and partly chemi- 
 cally by absorbing a portion of the heat which accompanies the 
 sulphuration of the tin and mercury, and which would otherwise 
 rise high enough to partially decompose the product. 
 
 (101.) Excitation of the Cylindrical Electrical Machine. First 
 remove the cushion and conductors, and then proceed to clean 
 the glass, which must be effectually done. If the machine has 
 been long in use, black spots and lines will be found frequently 
 adhering to its surface. If they be very numerous and rigid, 
 they require to be removed by mechanical means. The fine edge 
 of a bone paper-knife may be employed. Mr. Higgins (Phil. 
 Trans., vol. 68) recommends the skin of a dog-fish, which he says 
 will not scratch the glass as many powders will. The glass may 
 be wiped over with finely-prepared chalk as used for medical 
 purposes, wetted with rectified or methylated spirits of wine, and 
 to which a little benzine, or a small quantity of solution of soda 
 or potash, has been added, in order to remove any grease which 
 may adhere to the glass. The surface of the cylinder should now 
 be rubbed off with a clean linen or muslin rag until all the chalk 
 has disappeared. The preparation called Powdered Blue, which is 
 very finely-powdered starch coloured with cobalt, has been also 
 used for this purpose, and was employed by Cuthbertson. The 
 cylinder should be finally wiped over with a fine muslin rag wetted 
 with rectified spirits, and then rubbed dry with a clean silk hand- 
 kerchief, or some other soft substance. 
 
 Having effectually cleaned the glass cylinder, the cushion and 
 flap next demand attention. All the old amalgam which is 
 adhering to the rubber should be carefully scraped off, and its 
 surface wiped very clean. The dirt adhering to the silk flap 
 must likewise be removed. This should be further cleaned, if 
 necessary, by a muslin rag dipped in rectified spirits, with which 
 a few drops of benzine have been mixed, the object being to 
 remove any grease and dirt which may be adhering to the surface 
 of the flap; the silk should be finally wiped dry with a clean 
 muslin rag. ~VVe may now proceed to cover the surface of the 
 rubber with the amalgam already described; taking care not 
 to plaster it, as it were, upon the rubber in a thick and irregular 
 coat, but rather incorporate it with the surface of the rubber by 
 gentle pressure and friction, so that we may obtain a smooth and 
 equable exciting surface. Proceed now to fit the rubber in its place, 
 
GLASS FOR THE PLATE MACHINE. 201 
 
 the negative conductor being wiped clean and dry, as also the insu- 
 lating rod upon which it is supported. Finally, interpose a wide slip 
 of foolscap paper between the rubber and the cylinder, and turn the 
 glass round against it about a dozen times. The paper should 
 have been previously made hot against the fire, so as to deprive it 
 of all moisture. The glass will be now in a very excitable state, 
 and will emit faint sparks, attended by a low rustling sound. If 
 the paper be now drawn out from between the rubber and the 
 cylinder, brilliant luminous sparks will soon fly round the glass, 
 especially on connecting the negative conductor with the ground ; 
 and if the knuckle be presented to the surface of the cylinder at 
 the termination of the silk flap, the evolved electricity comes like 
 a wind from the glass upon the knuckle, and is often attended by a 
 crackling, rushing noise. This is good evidence of the success of 
 our manipulation. 
 
 We may now apply the prime conductor, taking care to wipe it, 
 and its insulator, very clean. Brilliant sparks will then fly from 
 the prime conductor to the knuckle, or to a large brass ball or 
 other metallic body brought near to it, if the presented body be in 
 connection with the earth or the negative conductor. 
 
 The machine is now in a fit state for experiment. 
 
 (102.) Excitation of the Plate Electrical Machine. In the construc- 
 tion of the Plate Electrical Machine, care must be taken in the selec- 
 tion of the glass, avoiding that which contains a large proportion 
 of alkali. The more metallic the glass is, the better is it adapted for 
 electrical purposes. Alkaline salts give glass a strong disposition 
 to absorb moisture. This defect is said by the old writers to be 
 remedied, in great measure, by returning the glass to the furnace for 
 a short time, or by exposing it in some way to heat. Old looking- 
 glasses have been found well adapted for glass-plate machines, from 
 the circumstance of having been long exposed to a seasoning tem- 
 perature, by which the alkaline salt in the glass has been dried out. 
 It is to be further remembered that the surface of the glass of a 
 Plate Electrical Machine is not what may be termed a natural vitre- 
 fied surface, but is a polished vitreous surface, produced after the 
 glass has been manufactured. This is a most important difference 
 in the plate and cylindrical machines, the natural vitrefied surface 
 being superior to a polished surface in its adaptation to the purposes 
 of electrical excitation. In the construction of the plate machine, 
 therefore, we must be careful to select glass of a high and 
 perfect polish. Common crown or window glass is admirably 
 adapted to the purpose of electrical excitation, and consequently 
 
202 ELECTRICAL MANIPULATION. 
 
 to the Plate Electrical Machine. It is, however, too thin and 
 fragile for the purpose in its ordinary state ; but when two 
 plates of crown-glass are joined together by means of a liquefied 
 sealing-wax, we obtain a very efficient plate for the Plate Machine. 
 A plate thus prepared, with black sealing-wax, has a fine polished 
 appearance. The author constructed a plate machine of 20 
 inches diameter in this way so long since as the year 1820. 
 The power was remarkable, although the glass was by no means 
 flat, having been cut out of the common tables then employed for 
 window glass. It would be difficult to construct a plate machine 
 in this way, of a very large diameter, without very considerable 
 experience. It is, however, to be accomplished, and would be no 
 doubt superior to ordinary plate-glass. In mounting the plate 
 upon its axis, the compressing flanges must be set remarkably true 
 and perpendicular upon the axis. One of the flanges, as observed 
 (77), is a fixture ; the other is a nut which, by a screw upon the 
 axle, can be turned up against the fixed plate, and confine the 
 plate of the machine with pressure between them, and enable the 
 axle to hold it fast whilst revolving. The screw of the flange to be 
 turned up against the glass, together with the screw upon the 
 axle itself, should be extremely fine, so that the pressure may 
 increase very gradually. 
 
 Glass-plate machines have been found very liable to crack by 
 pressure somewhere upon the central hole, which has been drilled 
 through them for the reception of the axis. Too much care can- 
 not be taken to avoid this, both in the drilling of the hole (which 
 should be smooth, without the sign of a splinter) and checking 
 the pressure of the flange beyond a certain point. The pressure, 
 it is clear, will go on constantly increasing in turning the machine, 
 since we turn in one direction against the resistance of the rubbers 
 in the opposite direction. The most effectual way of securing the 
 plate against fracture from the pressure, is to drill a hole through 
 the rim of the flange perpendicularly into the axle, tap it, and 
 turn up a fine screw, which will prevent any further movement 
 of the screw flange (a sufficient pressure for turning the plate 
 round having been obtained), as also by the judicious interposition 
 of a circular pieoe of fine, elastic leather between the flange and 
 the glass. It is also very desirable to thicken the centre of the 
 plate by small plates of glass about 4 inches diameter, and from 
 one- eighth to one-fourth of an inch in thickness, having holes in 
 them to take the axle. These plates are cemented to the glass 
 plate, one on each side, by adhesive varnish. 
 
CUSHIONS, RUBBERS, SILK FLAP. 203 
 
 Fig. 47 (page 71 ante) represents the brass axle upon which the 
 plate is mounted. It is about 20 inches in length, and from eight- 
 tenths to one inch in diameter, a b are the brass flanges ; c d, 
 the glass plates cemented to the plate of the machine. The axle 
 has a little play in its bearings, which are fixed to the transverse 
 bars of support, Fig. 46. That part of the axle carrying the 
 handle is encircled by a fixed brass socket, screwed fast to the 
 frame of the machine, about 1 inch in diameter and 2 inches 
 in length. A solid ring of brass is adapted to this cylindrical 
 socket so as to turn upon it with easy friction, and gives support 
 to the vertical conductor a R b, Fig. 51, and by which the elec- 
 tricity of the glass is carried off when the prime conductor is 
 transformed into a negative conductor. [See Chap. III. (80)]. The 
 axle projects on the side of the handle about 4 inches clear of the 
 frame. The glass handle is full an inch in diameter at its thickest 
 part, is a little taper, and about a foot in length between its brass 
 caps. 
 
 In exciting the plate machine we have first to get the plate per- 
 fectly clean by the same means as recommended for the cylindrical 
 machine. The skin of the dogfish may be employed with advan- 
 tage for removing any black lines or spots upon the glass. A 
 half- sheet of foolscap paper made hot before the fire, and inter- 
 posed between the cushions and the glass, and the plate revolved 
 between them, is a very beneficial process in the final preparation 
 of the plate for excitation ; indeed, papers of this kind should be 
 always kept between the rubbers and the glass, or the rubbers 
 themselves when not in use. 
 
 The preparation of the cushions or rubbers with the silk flap in 
 the electrical plate machine demands considerable attention. In the 
 machine now under consideration, the rubbers may be in length 
 about one-third of the diameter of the plate, and in width about 
 one-fifth their length. Fig. 48 represents the rubbers of the 
 3-foot plate as constructed by the author. They are 1 foot long, 
 by 2J inches wide, and consist of layers of thick cloth or felt, or 
 better of spongeo-piline (92), laid upon a flat piece of mahogany, 
 and faced up with fine morocco leather, or thick black oiled-silk, 
 the rough sides of each being next the glass. The leather or silk is 
 secured to the edges of the mahogany back by very fine copper 
 tacks, as already described for the cylinder machine (92). The 
 rubbers are loosely held in a central position in a sort of spring 
 frame by projecting pins, screwed into the mahogany back. The 
 frames supporting the rubbers are represented in Fig. 48, and 
 
204 ELECTRICAL MANIPULATION. 
 
 consist of two oblong side pieces, z , of mahogany, slightly 
 hollowed to give them elasticity. These spring pieces fit loosely 
 by a dove-tail joint in a mahogany block, B ; and are held 
 together with pressure upon the rubbers by a short cylindrical 
 bolt, a b, passing freely through them immediately in front of 
 the block, and two milled balls screwed one upon each end of it, 
 so as to press the mahogany holders together upon the rubbers 
 and the intermediate glass plate with any required degree of force. 
 All these points should be easy and movable not by any means 
 rigid so that the whole may easily play with the movement of 
 the plate. In the old plate machines it was customary, instead of 
 allowing the compressing bolt to go freely through both the holders, 
 to fix it to one of them, and by a nut in the other they were com- 
 pressed. The rubbers cannot accommodate themselves too easily to 
 the movement of the plate. The flaps should be applied to the cushion 
 in the same way as described for the cylindrical machine (92), 
 that is to say, first securing its edge upon the face of the rubber at 
 a distance of about one- third of an inch, by a little strong adhesive 
 paste, and finally, by a double row of running stitches of very fine 
 waxed silk, so that the edge of the flap may be well secured to 
 the rubber, and lie fairly and evenly upon its surface. The flaps of 
 each pair of rubbers are eventually united over the edge of the 
 plate. In applying the amalgam to the cushions, care must be 
 taken to lay it on thinly, smoothly, and evenly. The amalgam 
 should not pass beyond the edge of the flap where it is secured 
 to the rubber. 
 
 OCCASIONAL MEMOEANDA AND EXPLANATOEY NOTES. 
 
 (Y) Lac is a gum resin, the product of an insect termed the " Coccus 
 lacca," which deposits it on the branches of certain trees in India, especially 
 in Assam and Thibet, where it is found in a regular cellular structure, con- 
 taining the eggs of the insect. It is one of the best electrical insulators 
 we have, and plays a most important part in electrical investigations. Its 
 constituent parts are resin, gluten, wax, and a peculiar red colouring 
 matter. 
 
 It is imported into Europe under three forms, Stick-lac, Seed-lac, and 
 Shell-lac. 
 
 Stick-lac is the first or rude state, as found encrusting the twigs and branches 
 of the trees on which it is deposited. For purification it is broken into 
 small pieces, put into a long narrow canvas bag, and exposed to a suffi- 
 cient heat to liquefy the gum. The liquefied gum is forced out by twisting 
 the bag, and is allowed to flow over a plane smooth surface, to which it 
 
OCCASIONAL MEMORANDA. 
 
 205 
 
 does not adhere. The inhabitants of India allow the liquid to flow out and 
 consolidate upon the convex surface of a plantain-tree expressly prepared 
 for the purpose. The mucilaginous and smooth surface of this tree prevents 
 the gum from adhering. In 100 parts of stick-lac in its crude state we find 
 about 68 resin, colouring matter 10, wax 6, gluten 5*5, extraneous matter 
 10-5. The colouring matter is a valuable product, and forms the basis of the 
 lac-lakes and lac-dyes. Stick-lac thus purified and consolidated, being 
 pounded in a mortar, reduced into small grains, and a further portion of the 
 colouring matter extracted by the process of boiling, constitutes the substance 
 called Seed-lac. 
 
 Shell-lac is the lac in its natural state after the process just described of 
 simple purification by heat, and is produced by liquefying, straining, and 
 forming the liquid lac into thin plates, from whence its name Shell-lac. 
 
 LIBRARY 
 
 'UNIVERSITY OF 
 
 CALIFORNIA. 
 
PART II. 
 
 ON THE LAWS OF ELECTRICAL FORCE. 
 
 CHAPTER I. 
 
 BRIEF ENUMERATION OF FACTS AND PHENOMENA OF ELECTRICAL 
 ACTION, AS DEDUC1BLE FROM AN INVESTIGATION INTO THE THEO- 
 RETICAL AND PRACTICAL NATURE, OPERATION, AND LAWS OF 
 ELECTRICITY. 
 
 1. THE nature of electricity has always been a source of much 
 intricate discussion and theoretical investigation. JSTo solid con- 
 clusion, however, has ever been arrived at, calculated to determine 
 the precise nature of this wonderful invisible agency. Many 
 philosophers, both ancient and modern, have considered electrical 
 force as depending on a highly elastic fluid. Gilbert thought 
 that electrical attraction was the same as the attraction of cohe- 
 sion, and was occasioned by effluviae proceeding from bodies 
 excited by friction. 
 
 An hypothesis by Digby, as set forth by Brown, supposes that 
 a kind of electrical effluvium extends by friction from an electrified 
 substance which, after extension, again retracts, and brings with 
 it light particles of matter. Grassendi adopts a notion of this 
 kind. Similar views have been countenanced by Boyle and other 
 philosophers, both ancient and modern. 
 
 The Abbe Nollet supposes that in electrical excitation there is 
 an efflux of electrical effluviae from, and an afflux toward, the 
 electrified body. By the afflux toward the excited body, attrac- 
 tion is the result ; by the efflux from, repulsion ensues. Nollet 
 further supposes that in all bodies there are two kinds of pores, 
 one for receiving, the other for giving out the effluviae. 
 
 2. Notions of this kind, however, can only be considered as 
 mere wanderings of the imagination, and we are left to base our 
 views of electrical action on substantial experiment. 
 
208 ELECTRICAL FORCE. 
 
 3. "We are indebted, as we have seen (Part I., 17), to M. du Fay, 
 Intendant of the Gardens of His Majesty the King of France, in 
 1730, for the first great step in theoretical and practical electricity. 
 This ingenious philosopher discovered that electrically- excited 
 bodies develop two kinds of force, one produced by excited 
 vitreous, the other by excited resinous substances. Hence he 
 terms the one vitreous, the other resinous electricity. On further 
 investigation these different electricities were found to be attrac- 
 tive of each other, but repulsive of themselves (Part I., 27). 
 
 4. The wonderful discoveries of Newton, and his determination 
 that the laws of gravity were universal, naturally led philosophers 
 to entertain the notion that every other force associated with 
 ordinary matter was subject to the same law ; that the law of 
 gravity was, in fact, the dominant law of all attractive forces, of 
 whatever kind or species, of electricity and magnetism more espe- 
 cially ; and a great variety of experimental inquiries were insti- 
 tuted with the express view of verifying this conjecture. Cavendish, 
 after ^Epinus, was certainly the first philosopher who investigated 
 experimentally, and threw light on this question,- as appears by 
 his celebrated paper in the sixty-first volume of the Philosophical 
 Transactions for the year 1772, and likewise by his MS. but as 
 yet unpublished papers, handed down to us by the Earl of 
 Burlington, a descendant of the Cavendish family. Cavendish 
 directly proves that on the supposition of electricity being an 
 elastic fluid, the mutual repulsion of its particles is in the inverse 
 duplicate ratio of the distances. That in the case of a sphere 
 being charged with electricity, all the charge should be found on its 
 surface, whilst the action upon any interior point would be nothing 
 (Part L, 55). He further describes a great variety of original 
 and very beautiful experiments in support of this hypothesis. It 
 is, therefore, a misapprehension on the part of some writers when 
 they say that " Cavendish had assumed no particular law of 
 electrical force, but merely that the action decreases as the dis- 
 tance increases." Dr. Robison, in 1769, read a paper professing 
 to show that the law of electrical attraction was identical with 
 that of gravity : the same result had been arrived at by Mayer, 
 and had been eagerly sought for by other philosophers of that 
 period. The beautiful and profound physico-mathematical re- 
 searches of Coulombe, in the Memoirs of the French Academy 
 for 1785, all went to confirm this law of the inverse duplicate 
 ratio of the distance as stated by Cavendish. The views of 
 electricity which Coulombe has handed down to us have since 
 received the sanction of the most eminent philosophers of Europe. 
 
LAW OF ELECTRICAL FORCE. 209 
 
 The experimental inquiries of Coulombe have been made the basis 
 of a comprehensive analysis by Poisson, as also by Le Place, Biot, 
 and other eminent mathematicians. Coulombe's theory now 
 stands pre-eminent as the Coulombian theory of electricity. 
 Such a galaxy of eminent names, countenanced by so wide and 
 general a reception, would necessarily go far to discourage doubts 
 of the accuracy of Coulombe's views ; still the progress of phy- 
 sical discovery must not be limited ; since from the moment we 
 establish a limit, and say that limit has been attained, from that 
 same moment we close the door upon all further research into 
 principles, and we remain immersed in a sort of philosophical 
 orthodoxy, very unfavourable to a more complete knowledge of 
 those unseen yet astounding powers of nature, the effects of which 
 we daily experience. (Note A.) 
 
 That the law of the inverse square of the distance, as deduced by 
 Coulombe and other philosophers, is a law of electrical force, and 
 a very general and important law in its consequences, we cannot 
 for a moment doubt ; although the experiments by which it was 
 sought to establish this law were few, and not always satisfactory, 
 being derived principally from the operation of what is termed 
 electrical repulsion, an element liable, in many instances, to much 
 uncertainty. Admitting, however, this law of the inverse dupli- 
 cate ratio of the distance to be a general law, it is still important 
 to examine under what circumstances the law obtains ; and the 
 peculiar conditions requisite to its full development. Such an 
 inquiry is calculated to bring us further acquainted with the 
 nature and operation of electrical action, and to develop numerous 
 laws of this force, of which we might otherwise remain in ignor- 
 ance. 
 
 5. It may not be hence altogether unprofitable to define clearly 
 what we really understand when we say that a force increases or 
 decreases in the inverse duplicate ratio of the distance of its 
 action, since upon this a large amount of sound reasoning upon 
 the laws of physical forces mainly depends. It is highly 
 probable, if not morally certain, that every physical effect is 
 in a simple proportion to its cause. If the cause increase in 
 any proportion, the effect will increase in the same proportion. 
 The effect cannot, for example, be as the square or the cube of the 
 cause ; for that would be, as remarked by the learned Dr. Samuel 
 Clarke, to refer the effect partly to the cause, and partly to nothing. 
 The amount of illumination, for example, of a given point, will 
 depend on the quantity of light concentrated in that point. If 
 we double the quantity of the agency in operation, we should 
 
 p 
 
210 ELECTRICAL FORCE. 
 
 double the illuminating effect ; we should certainly not quadruple 
 it. In referring, therefore, the decrease of a force to an increase 
 of the distance, or more especially an inverse square of the distance, 
 we are fairly led to inquire into the cause of so wonderful, and 
 apparently so inconsequent, a phenomenon. Suppose, for example, 
 we endow distance, that is to say, a mere imaginary line or space, 
 with all the properties of entity, and consider its extension as the 
 immediate cause of the observed decrease in the force ; still we 
 have no ground for supposing that a double extension in length 
 would cause more than a proportionate reciprocal decrease of the 
 force. We could have no ground for assuming that the force 
 would become four times, nine times, &c., less, as the distance 
 became twice, three times, &c., greater ; that is, upon the hypo- 
 thesis that mere distance was the cause of the decrease of the 
 forces. Reasoning much in this way, many philosophers eminent 
 for their intellectual powers looking upon gravity as a central 
 force, or an emanation from a centre, have considered this emana- 
 tion to expand in all directions, and to become more feeble in 
 proportion to the increased spherical surfaces over which the 
 emanation may be supposed to expand ; and since the concave 
 surfaces of the imaginary spheres, intercepting the emanation 
 at given distances, would increase directly as the squares of 
 those distances that is, as the squares of their radii therefore 
 the quantity of the emanation or force in any point of space 
 would be in the same ratio ; so that at double the distance from 
 the centre we should only have one-fourth the force in any one 
 point, and, therefore, only one -fourth the effect, and thus the 
 force would be said to vary in an inverse ratio of the squares of 
 the distances. 
 
 Dr. flallev, for example, reasoned in this way ;* and the reason- 
 ing, it must be admitted, is perfectly clear and intelligible ; the 
 effect here is in a simple ratio of its cause, not as the square of its 
 cause. The force, for example, decreases with the cause of the 
 force; and we see so far how it is that distance enters as an 
 element into the general expression of the law of the force of 
 gravity. 
 
 6. The question then is, whether electrical or magnetic force 
 is such a force as this ? 
 
 Some philosophers object to this kind of reasoning, and contend 
 for the absolute existence of force, varying in an inverse dupli- 
 cate ratio of the distance between indefinitely minute points of 
 vanishing magnitude, in which they say that no attenuation of 
 * Whewell's " Hist, of Inductive Sciences," vol. ii. p. 158, 2nd edition. 
 
DISTRIBUTION OF ELECTRICITY. 211 
 
 the agency upon which the force depends can be supposed to 
 obtain ; they take the points as points, without magnitude or 
 parts. How far this involves a metaphysical subtlety is left to be 
 determined. "We certainly know nothing of forces, or the laws of 
 their action, except through the medium of observation and physi- 
 cal experiment; but as yet no physical experiment in either 
 electricity or magnetism, on the attractive or repulsive force of 
 points of vanishing magnitude, has been ever realised. Then, 
 again, we must not leave out the peculiar condition of power on 
 such points. If we suppose, for example, all the electricity upon 
 a given sphere condensed into a point, and all the electricity of 
 another sphere condensed into a point, still we may conceive the 
 physical powers of these two points to be very different. 
 
 Reasoning in this way, all the matter of the sun condensed into 
 a point would necessarily constitute a very different point of power 
 from that of all the matter of the earth condensed into a point ; 
 and, therefore, at last we have a somewhat complicated question 
 to deal with, when we apply this sort of analysis to such forces as 
 those of electricity and magnetism, exerted at different distances 
 from each other. 
 
 7. Our present object, however, is to investigate the physical con- 
 ditions under which the forces of attraction and repulsion apparent 
 in electricity obtain ; what are the general laws of the operation of 
 such forces ; how far we may consider them as central forces, such 
 as gravity ; or whether they are to be considered more in the 
 light of parallel forces destructive in their character, in all their 
 relations to common matter, and in the elementary conditions 
 of their operation. 
 
 8. If we examine the physical data upon which our general views 
 of the laws of these forces rest, we do not find the experimental 
 investigations to be extensive, nor are they always satisfactory ; 
 certainly they are not at all comparable with the precedents of 
 other branches of science. The data nearly, if not all, depend on 
 a few selected and especial cases of repulsive force, and are very 
 dependent on the perfect operation of a small insulated disc termed 
 a tangent plane (Part I., 143), which, being brought into contact 
 with various portions of a charged body, is assumed to be an element 
 of the surface of that body, so that when withdrawn from it, the tan- 
 gent plane is considered as coming away charged with a quantity 
 of electricity proportionate to the quantity absolutely existing in 
 the point to which it has been applied. From such experiments a 
 certain distribution of electricity, considered as a highly elastic 
 fluid, is inferred ; all of which is, to a greater or less extent, still 
 
 p 2 
 
212 ELECTRICAL FORCE. 
 
 within the limit of conjecture, and it must be considered as being 
 a somewhat remarkable fact, that up to a late period there 
 existed scarcely any electrical instrument of research, based upon 
 attractive force. A very large range of electrical phenomena 
 remains, therefore, to be investigated by varied experiment, and 
 which does not appear as yet to have been brought under the 
 dominion of experimental analysis. 
 
 9. Now it is to be here observed, that if we imagine an insulated 
 conductor to exist in space perfectly freed from the influence of 
 common matter supposing, for example, it were the only body 
 in space then it would really be impossible to say supposing it 
 charged with what we call electricity how the charge exists 
 upon it. For anything we should know to the contrary, the 
 electrical agency might be uniformly distributed throughout its 
 substance or over its surface. We could really know nothing, nor 
 could we predicate anything of its electrical condition, except 
 through the instrumentality of other bodies brought to share in 
 the charge ; we might, indeed, upon a variety of hypotheses and 
 refined methods of analysis, arrive at what should be its especial 
 electrical condition supposing the hypothetical basis of such cal- 
 culations to be correct ; but we should still have to verify the 
 results upon the broad basis of observation and experiment. 
 
 10. It may here be as well to offer a few observations on the opera- 
 tion of what has been called the Tangent Plane, as employed by 
 Coulombe and the French philosophers, with a view of determining 
 the distribution of electricity on charged bodies, on the assumption 
 that electricity is an agency depending on two fluids of high elas- 
 ticity, repulsive of their own particles, but attractive of each other. 
 
 One of the most striking and important features of electrical 
 attractive force is not the phenomenon of the apparent attraction 
 itself observed to arise on presenting a charged to a neutral 
 surface, or to another body in an opposite electrical state, but the 
 previous or sympathetic, or what is termed the induced actions 
 which take place between the bodies, and upon which the sub- 
 sequent attractive force altogether depends. In fact, both in 
 electricity and magnetism, bodies are first rendered attractable 
 or repellent, before attraction or repulsion can ensue ; and if 
 such changes do not obtain, no attractive or repulsive force 
 is apparent. If, for example, we suspend an extremely thin 
 conducting disc from an arm of a balance, by insulating threads 
 of shell-lac, we shall in vain attempt to attract it by opposing 
 to it a similar disc charged with electricity ; or at least the 
 force will be extremely small ; nor will such an insulated neutral 
 
INDUCTION NECESSARY. 213 
 
 disc be in a perfect condition to receive or abstract electricity 
 from a charged disc (Part L, 44). 
 
 11. Now we find on increasing the thickness of the suspended 
 disc the attractive force becomes more fully developed ; and if we 
 give it a conducting communication with the earth, we then 
 complete its susceptibility of being rendered attractable ; and the 
 force, with a given charge at a given distance, is then the greatest 
 possible ; but if we take the suspended disc insulated and in- 
 definitely thin, we have really no development of attractive force 
 at all. 
 
 Take the tangent disc in any way you please, it really cannot 
 be considered in any other light than that of a neutral con- 
 ductor, brought to share in the electricity of a charged conductor. 
 
 12. Now I observe, in touching various points of a long cylin- 
 drical and charged conductor with a small insulated conducting 
 body, that the proportion of the charge it receives by contact with 
 different points of the charged body is very dependent on its 
 thickness ; so that the touching body takes up less and less elec- 
 tricity as we diminish the thickness. When the thickness of the 
 disc is inconsiderable, the charge which it receives at the centre 
 vanishes, or nearly so. Now I should wish to have it considered 
 whether at a certain limit of thickness this touching body might 
 not come away altogether neutral, or nearly so, from any or every 
 point of the charged body. That is a point which I think 
 deserves great consideration before we are entitled to draw any 
 legitimate conclusion as to the distribution of the electricity on the 
 charged body. Now it is notorious that if you take a small 
 insulated conductor of great extension in thickness, such a con- 
 ductor comes away equally charged with whatever point of the 
 charged body it be brought into contact. If we coat a small disc 
 of talc with tinfoil so as to complete the conditions of the Ley- 
 den jar or coated pane and apply it to various points of a long con- 
 ductor charged with electricity, it will be charged equally high from 
 every point of such a conductor ; or if we bring an electrometer 
 in contact with such a conductor, it will evince the same degree 
 of power at all points of application ; so that the vital source of the 
 power of one body to receive electricity from another body, is the 
 susceptibility of induction in the recipient body ; and I cannot 
 conceive the possibility of any body taking electricity from another 
 without undergoing previous inductive change. This consideration 
 leads me to ask attention to another series of facts which I conceive 
 to be intimately associated with the operation of the tangent disc. 
 It is well known that if a small insulated carrier ball be plunged 
 
214 ELECTRICAL FORCE. 
 
 within a charged hollow sphere, it comes away perfectly free from 
 all charge ; its contact with the interior surface of the sphere has 
 been productive of no electrical change in the ball whatever ; it has 
 been hence inferred, as we all know, that no electricity exists on the 
 interior surface of the sphere. That, I admit, may possibly be the 
 case ; but certainly this result is not conclusive of the fact. The 
 failure of the small ball to become electrified, may as well be sup- 
 posed to arise from the total incapability of all inductive change 
 in the small ball under such circumstances as from the absence 
 of electricity on the interior surface of the sphere. In evidence of 
 which I made the following experiment, and which will be found 
 fully described in the Royal Society's Transactions. A hollow globe 
 of glass, with a long neck, was arranged so as to have the coatings 
 of mercury. The glass was then charged, and the mercury drawn 
 off. The outside of the jar was bristling with electricity, yet no 
 electricity was to be obtained on the inside. (B.) 
 
 13. It is I think evident from this, that the forces here so operate 
 upon the carrier ball in opposite 'directions when placed within the 
 charged sphere, as to fetter completely its inductive suscepti- 
 bility ; hence it fails to share in the electrical charge. That there 
 is really no essential difference between this and the case of the 
 hollow conducting sphere, as regards electrical charge, and that the 
 absence of inductive change is really the cause of the neutrality of 
 the ball, is shown by a repetition of the same experiment, with" 
 the exception of substituting a plane of glass for the hollow sphere. 
 If we communicate a charge to a coated pane, and remove the 
 coatings, we may take up electricity freely from the charged side, 
 but in this case the carrier ball can, by external exposure, undergo 
 inductive change. 
 
 That the induction upon the ball should be zero within the 
 sphere, supposing the force to be in an inverse duplicate ratio of the 
 distance, is admitted, and that is so far an evidence of the truth 
 of such a law, which I beg leave again to state I am not disputing ; 
 but it is still consistent with the phenomena of the hollow charged 
 conducting sphere, in which also the inductive change on the 
 carrier ball is zero, and hence it cannot receive electricity from the 
 interior surface, even supposing the interior surface to be capable 
 of communicating it. 
 
 14. I think we may conclude from these facts that a tangent plane 
 will receive more or less electricity, in proportion as it is more or 
 less freed from the opposing forces of surrounding electrical 
 particles. According to Yolta it would be the greater or less 
 emersion of the body in the electrical atmosphere of the charge. 
 
INFLUENCE OF CHANGE OF FORM. 215 
 
 In illustration of this, let the square or circle A, Fig. 109, and the 
 rectangle B, represent two plane conductors of equal area, charged 
 with the same quantity of electricity, and let us imagine for the 
 moment that the electricity is equally distributed, as designated 
 by the crossed straight lines on those figures. Now it is quite 
 evident that if we apply a small in- A 6 B 
 
 sulated plane to the centre of the [TTT] rr i i i i i ! i I i i i I I i i 
 square, the small plane would be 
 more immersed in the surround- ri s- 1 9 - 
 
 ing electrical particles than if applied at the middle of the 
 rectangle B, and certainly more embarrassed in its inductive 
 efficiency than if applied at the points o 6 of either of these con- 
 ductors. In fact, at the points o 6 it would have as it were one 
 exterior portion free from electrical particles ; hence, on this 
 account only, it could take up more electricity at the extremities 
 than at the centre of the conductor, independent of any consi- 
 deration of distribution. 
 
 15. That the grouping together of electrical particles to a greater 
 or less extent influences their activity, or what may be termed 
 their state of quietude, is evident from the fact that a given 
 quantity of electricity will repose with more quietude on a 
 given surface, in proportion as the surface becomes extended 
 in length. Thus, although the areas of the circle or square A, 
 and of the rectangle B are the same, yet B has a greater linear 
 extension than A, and an electrometer will evince a very 
 much greater effect when connected with the circle than when 
 connected with the rectangle. As I have already observed, 
 in calling attention to these facts. I would by no means be con- 
 sidered as professing to determine authoritatively the ques- 
 tion of the accuracy of the proof plane as an instrument of 
 research, and as employed by Coulombe and other philosophers. 
 It would be unwarrantably presumptuous in me to pronounce 
 dogmatically upon the absolute merits of such investigations as 
 those to which I have already adverted. If I venture, therefore, 
 on submitting the question of the proof plane for further con- 
 sideration, I trust it will be believed that I do so with all becoming 
 diffidence, and a due appreciation of the many beautiful researches 
 in which the method of the proof plane has been so largely 
 employed. 
 
 16. Having now adverted to the question of the proof plane 
 as important in the investigation of electrical force, I will briefly 
 consider the way in which the forces of attraction and repul- 
 sion in electricity and magnetism manifest themselves, together 
 
216 ELECTRICAL FORCE. 
 
 with the laws to which they are subjected. The dependence of 
 the electrical and magnetic forces of attraction and repulsion on 
 the elementary changes by induction, which the attracting or 
 repelling bodies evince, at once gives a distinctive and separate 
 character to these forces, as compared with the force of gravity. 
 In the force of gravity we know nothing of what may be termed 
 a dual force ; nor do we perceive that the attractive power exerted 
 between one mass of matter and another is at all dependent on 
 any previous or elementary change in the attracting masses, or 
 in the existing condition of the agency on which the attraction 
 depends, that is, so far as we at present know ; and, therefore, so 
 far the forces of electricity and magnetism are essentially different 
 in their very nature from the force of gravity. That these forces 
 become less in some inverse ratio of the distance of the bodies 
 through or upon which they are exerted, is certain, but according 
 to what law they thus increase or decrease was for a long time a 
 problem of some difficulty. 
 
 17. The early ^Experimentalists. Sir Isaac Newton conceived 
 magnetic force to vary as the cubes of the distances, inversely. 
 Hawksbee, Whiston, Brook Taylor, Muschenbrock, Martin, and 
 several other eminent philosophers, all arrived at different experi- 
 mental results. Hawksbee found the force to vary in an inverse 
 sesquiduplicate ratio of the distances, that is, as the -J power of 
 the distance inversely. Martin deduced the law of an inverse 
 sesquiplicate, or -J power of the distance. Brook Taylor con- 
 ceived that the force did not change according to any regular 
 law, but varied at different distances from the magnet. Whiston, 
 however, and subsequently Brook Taylor and Hawksbee, again 
 found it as the -J power of the distance. -ZEpinus, and after him 
 the Dutch experimentalist, Muschenbrock, found the force to be 
 in the simple inverse ratio of the distance. In some cases, how- 
 ever, Muschenbrock obtained all sorts of laws. Finally, Caven- 
 dish, Lambert, Mayer, and Coulombe arrived at the laws of the 
 inverse square of the distance, as in the case of gravity. Now, it 
 has been, as I think, too prevalent a practice to discard the 
 experiments of the earlier experimentalists, in which other laws 
 appear to have been deduced. It is not likely that such men as 
 Hawksbee, Brook Taylor, and Whiston would have instituted 
 inaccurate experimental investigations. It remains, therefore, to 
 conciliate these differences, and show the source of the differences 
 in the respective results. 
 
 18. Now, in examining the operation of these forces as exerted 
 between given masses, we have first to examine the laws of those 
 
LAW OF ELECTRICAL FORCE. 217 
 
 changes we have termed inductive, upon which the forces depend. 
 Now I think I have shown in my former papers, contributed to 
 the Transactions of the Eoyal Society, that when an electrified 
 conductor is placed in the presence of a neutral conductor in a 
 free state, a change takes place in the neutral conductor, such as 
 to induce in it an opposite force, which force increases in a simple 
 inverse ratio of the distance between the opposed surfaces. Then 
 this force, so called into operation, re-acts upon the charged 
 body, by what I have termed a species of reflection, or a reflected 
 induction, which reflected induction also varies in an inverse 
 simple ratio of the distance between the opposed surfaces, that 
 is, if the charged surface have likewise great capacity. When 
 these two elementary forces increase or decrease together, the 
 total force varies with the two conjointly, and thus we obtain a 
 law of force in an inverse duplicate ratio of the distance ; but if 
 from any cause either of the two elementary inductive forces 
 becomes impeded, as in the case of the force between a charged 
 conductor and an insulated neutral conductor of small extension, 
 then we have no longer the law of force just stated ; the law may 
 be any of the laws arrived at by the early experimentalists, depend- 
 ing entirely on the law of variation of the previous inductive 
 changes. 
 
 These direct and reflected inductions in magnetism I am enabled 
 to illustrate, to a certain extent, by the apparatus before me,* and 
 this is equally true for electrified bodies. 
 
 19. This understood, it is easy to see that the limit of the law of 
 electrical or magnetic force is a simple inverse ratio of the distance 
 that is to say, if we take two masses affected electrically or mag- 
 netically to saturation (that is to say, so affected that neither 
 undergoes any further inductive change by approximation), we have 
 then a force varying as the simple distance inversely. In approach- 
 ing the limit of saturation, therefore, we may readily conceive the 
 induced force to change gradually ; still the convergence may be 
 so slow that for a long series of terms the induction may be taken 
 as constant. It is only, then, between certain limits, and under 
 certain free states or conditions of magnetic or electrical change, 
 that we obtain the law arrived at by Coulombe. In the case of 
 electrical force between a charged and neutral conductor, the force 
 will be always in the inverse ratio of the squares of the distances 
 between the attracting surfaces, so long as the neutral, or what is 
 considered the attracted body, has free communication with the 
 earth, and the charged body has also an unlimited extent of charge. 
 * See " Rudimentary Magnetism," part iii., chap. i. 
 
218 
 
 ELECTRICAL FORCE. 
 
 Thus, for example, if a plane surface be in communication with a 
 charged Leyden jar, and another plane and equal surface in con- 
 nection with the earth be opposed to it, the force will vary in an 
 inverse duplicate ratio of the distances with great precision, and it 
 will be as the number of attracting points directly. If the opposed 
 surfaces be spherical, as in the case of two spheres, we have then 
 to determine a point within each hemisphere in which we may con- 
 ceive the whole force to be collected, and to be the same as if operat- 
 ing from every part of the opposed hemisphere. These points, 
 designated by p p' in the annexed figure, may be 
 easily determined on the demonstrable hypothesis that 
 the force is as the areas of attraction directly, and 
 as the squares of the distances inversely. When the 
 spheres touch, these points will be in the surface ; 
 when at an infinite distance they will be at the 
 centres of the spheres. I have exhibited, by means 
 of the Electrical Balance, the precision with which 
 the attractive force of spheres varies in an inverse 
 duplicate ratio of the distances between these points, 
 Fig. 110. so as to leave no doubt on the mind of the truth 
 of this law. 
 
 20. Now it is a remarkable fact, that in order to obtain this law of 
 force it is essential that both the attracting surfaces have unlimited 
 electrical capacity., We cannot, for example, obtain this law with 
 two thin conducting planes, nor can we obtain it by opposing to 
 each other mere positive and negative electricity accumulated on 
 a non-conducting surface. I have lately contrived an experi- 
 ment which seems to verify this perfectly. If two light discs of 
 talc be furnished with temporary coatings, and be charged as in 
 the case of the Leyden pane, we may, on removing the coatings, 
 obtain a purely positive and negative surface. If these two sur- 
 faces be opposed to each other, and the forces of attraction at given 
 distances measured by some adequate instrument, the force will 
 be in an inverse simple ratio of the distance. 
 
 21. I have already stated that the measurement of electrical 
 and magnetic force, through the instrumentality of repulsion, 
 is by no means satisfactory ; and that will, I think, be appa- 
 rent on a very little consideration. When two bodies similarly 
 charged, either electrically or magnetically, are brought within 
 each other's influence, the elementary inductive forces, already 
 adverted to, are still to be considered, and the immediate ten- 
 dency of each body is to reverse the existing electrical or mag- 
 netic state of the other that is to say, the inductive and repulsive 
 
OCCASIONAL MEMORANDA. 219 
 
 forces are in opposite directions. From the induction, attraction 
 ensues ; and hence it is that two bodies similarly but unequally 
 charged will attract each other at some distances, and repel at 
 others ; and it is only, therefore, in cases in which the bodies are 
 so permanently charged as to maintain their existing state in 
 opposition to the inductive actions, that we can expect to arrive at 
 consistent results. The inductive forces in opposition to the exist- 
 ing state of the bodies is in fact the immediate source of the 
 apparent repulsion. The effort really is attractive force after all; 
 but in the effort to bring about this attractive force, the bodies, 
 if movable, recede from each other. Hence it follows that the 
 amount of the observed repulsion is only the difference between 
 the repulsion maintained and the attractive force induced ; and 
 this is really a very serious consideration in operating with the 
 Torsion Balance. 
 
 OCCASIONAL MEMOEANDA AND EXPLANATORY NOTES. 
 
 (Note A. p. 209.) The following note by the author appeared in the Philo- 
 sophical Magazine for August, 1857 : 
 
 It is not without regret that I am led to offer a few remarks on the several 
 notices of my papers on electricity by Professor Thomson and other writers, 
 which have from time to time appeared in the Philosophical Magazine. In 
 the first place, I observe (Phil. Mag., vol. viii., p. 42) Mr. Thomson states, 
 that Dr. Faraday and myself have undertaken researches with a view of 
 invalidating the theory of Coulombe, certainly a gratuitous and unproved 
 assumption to begin with. Then the experimental result I obtained, F <*Q 2 , 
 is represented as having been adduced by me in opposition to the theory ; 
 and the Eeport made to the British Association in 1837, by the Eev. Dr. 
 Whewell, is referred to in support of Mr. Thomson's views. In this report 
 I am represented as having been surprised at finding the force to be as the 
 square of the quantity ; and in another place (" Hist, of Induct. Sciences," 
 vol iii., p. 28, new edition), it is said that I considered the result as " inex- 
 plicable." Now if Mr. Thomson or the Eev. Dr. Whewell, for whom I must 
 ever entertain the greatest regard and respect, will be so good as to point out 
 where, in any of my works, even the shadow of all this is to be found, I 
 should feel myself greatly obliged. So far from having been surprised at the 
 result, or considering it inexplicable, I distinctly say in my paper in the 
 Philosophical Transactions for 1834, referred to by Mr. Thomson, p. 266, 
 that "I do not advert to these experiments as containing any unexpected 
 results, but rather in explanation of particular methods of research," &c. ; 
 and at page 236 I refer the law in question to electrical induction upon the 
 attracted disc, &c., being just what Mr. Thomson has been so obliging to 
 repeat, virtually in as many words, and with a view of exposing the delusion 
 he assumes I labour under in regard to the Coulombian theory. In Mr. 
 Thomson's paper (Phil. Mag. for 1854, vol. vii., p. 193), I observe the follow- 
 
220 ELECTRICAL FORCE. 
 
 ing remark: "The amount of heat is proportional to the square of the 
 quantity discharged, as was first demonstrated by Joule, although it had 
 been announced by Sir Snow Harris, as an experimental result, to be simply 
 proportional to quantity." Riess, Joule, and Clausius appear to agree as to 
 rny having made this announcement, and confidently allude to the inaccu- 
 racy and to the refutation of my deductions. Now I do here most em- 
 phatically and most positively deny ever having stated anything of the 
 kind ; but since these gentlemen say I have done so, and pretend to have 
 corrected my error, I call upon them, as a point of honour, to say when I 
 made such an announcement, or where, in any of my published works, it is 
 to be found. The fact is, that so far from having stated the law of electrical 
 heat to be simply as the quantity of electricity discharged, I was really the 
 first to discover with precision, and demonstrate with exactitude, the law in 
 question, and that, too, long before Mr. Thomson and the other gentlemen 
 above mentioned were at all known in the world of science. If Mr. Thomson 
 will turn to pages 67 and 68 of the Transactions of the Plymouth Institution, 
 published in 1830, and quoted in the Journal of the Royal Institution, 
 1830-31, p. 380,* he will find in the former work a series of original experi- 
 ments on the heating effects of the electrical discharge, and at page 68 the 
 following announcement : "It may be hence inferred that the effects of an 
 electrical discharge on a metallic wire, all other things remaining the same, 
 is directly as the square of the quantity." See also Journal of Royal Institu- 
 tion, p. 381 (vi.). Moreover, the date of my paper is so far back as Novem- 
 ber, 1825 ; and I may say, in the words of Dr. Eiess (Phil. Mag., 1854, vol. 
 vii., p. 348), that " an assertion of such a general character as that ventured 
 by Mr. Thomson ought to be the consequence of a careful examination," 
 especially of what has been done at home. 
 
 As this subject is of importance, I may further remark, that the experi- 
 ments of Cuthbertson and others, referred to by Becquerel, and quoted by 
 Joule (Phil. Mag., October, 1841), can scarcely be said to have established 
 the law in question, or even its probability. In the first place, they had 
 no accurate measure of the quantity of electricity accumulated ; they were 
 unaware that twice the quantity accumulated on a given coated surface 
 would counterpoise four times the weight, regulating Cuthbertson's own 
 steel-yard electrometer, in which a charge of 30 grains was taken as twice 
 that of 15 grains. Their methods of research by the fusion of wires appears 
 to have been anything but exact, and all sorts of tricks were played with the 
 battery, as by breathing into the jars, &c. Cuthbertson says ("Practical 
 Electricity," pp. 180 to 186), "If 18 inches of wire be taken, and a given 
 charge just causes it to run into balls, much shorter lengths will still be only 
 converted into balls ; if only 7 inches be taken, nothing but balls will 
 appear ; the only difference will be that the balls will be smaller, and will 
 be dispersed to a greater distance, which may be easily overlooked." Now 
 what confidence can be placed in all this as experimental research ? Yan 
 Marum found his batteries produce a heating effect proportional to the 
 coated surface. Cuthbertson, by his most exceptionable method of breathing 
 into his jars, when he wanted a greater effect, did, it is true, obtain a higher 
 ratio, yet no direct satisfactory comparison between the quantity discharged 
 and the heat produced was ever arrived at, and various results ensued. This 
 we see admitted at page 182 of Mr. Cuthbertson's work, Exp. 149 : here the 
 law in question evidently failed. At page 185 we observe that when the 
 quantities of electricity were said to be as 2:3, another result ensued ; for 
 experiments 150 and 151 show that the lengths of wire fused, instead of 
 
 * See also a copy of my paper (printed in 1828), in the Library of the Royal 
 Society. 
 
OCCASIONAL MEMORANDA. 221 
 
 being as 4 : 9, were as 2 : 6 ; that is, as 1 : 3. It is quite impossible to repose 
 any confidence in such a state of things. Indeed, we have only to examine 
 Mr. Cuthbertson's experiments attentively, as given in his work, and in 
 Nicholson's Journal (4to., vol. ii., p. 218), to be assured of the inexactitude 
 of the experimental processes. In the latter we find the lengths of wire 
 melted to be as the quantity of electricity (see p. 218). However true, there- 
 fore, it may be that Cuthbertson obtained results which led him to imagine 
 that twice the quantity of electricity would melt four times the length of 
 wire, he cannot be said to have demonstrated and established that law ; and I 
 may, therefore, without any philosophical injustice, claim to have been the 
 first to have clearly developed that law by exact electrical measurements, 
 and by new methods of research, as my paper, dated 1825, and quoted in the 
 Journal of the Eoyal Institution, 1830-31, fully shows. There, also, will 
 be found the hypothesis advanced by Mr. Joule, that increased velocity is pro- 
 bably the source of the quadruple heat. I endeavour to show that the heat 
 is as the velocity with which the unit of charge traverses the wire, that a 
 double quantity passes with a double velocity, and the effect is "as the 
 momentum," or quantity into velocity. I again, therefore, ask Mr. Thomson 
 to state when or where I ever announced, "as an experimental result," 
 that the heating effect of an electrical discharge was " simply proportional 
 to the quantity of electricity." Mr. Joule, in referring to my paper in the 
 "Transactions" for 1834, appears to have confounded this question with 
 that of the same quantity accumulated under different electrometer inten- 
 sities, a question discussed in my late differences with Dr. Kiess. What I 
 said in my paper in the "Philosophical Transactions," and which I still 
 insist on, was simply this, viz., that under whatever electrometer intensity 
 you accumulate a given quantity of electricity, provided the battery surface 
 be undivided, that quantity, when dischared through a metal wire, will still 
 excite in it the same degree of heat. Thus, if a quantity of electricity = A, 
 for example, be collected on a Leyden jar, exposing 2 -5 square feet of coating, 
 and then be collected on ajar exposing 5 square feet of coating, I say that, 
 notwithstanding the electrical intensity in these two cases may be as 4 : 1 
 nearly, yet that the discharge of the quantity A in each case will excite the 
 same heat in a metallic wire ; that is, if the same charging rod and circuit 
 be employed. And I ask the gentlemen to whom I have just alluded to 
 make that experiment. Until they do so, they have no right to talk of " the 
 memoir in which Eiess refuted my statement," &c. (Phil. Mag., 1854, vol. 
 vii., p. 297). 
 
 (B. p. 214.) *A hollow globe of glass about five inches in diameter, with a 
 short neck about an inch in diameter, is filled up to the neck with dry mercury, 
 and placed in an outer glass vessel also filled with mercury. We have here all 
 the conditions requisite for imparting a charge to the interior of the glass. 
 For this purpose let a light insulated charging rod, surmounted by an elec- 
 trometer of repulsion, be introduced within the globe, and the mercury in 
 the exterior vessel be connected with the ground by a metallic communica- 
 tion. Let this system be charged in the usual way : when charged to any 
 degree of intensity, as shown by the electrometer, remove the communica- 
 tion with the earth, and also the charging rod, by means of its insulating 
 
 * The substance of this note is from a paper by the author, entitled " Researches 
 in Statical Electricity," contained in the Philosophical Magazine for 1857- The 
 paper is illustrated by an engraved Plate containing 22 Figures. Figs. 5, 6, and 7 
 refer to the experiment alluded to at p. 264, and described in this note. ED. 
 
222 ELECTRICAL FOftCE. 
 
 support ; run off the mercury first from the outer vessel, and then from the 
 interior of the globe by means of a glass siphon, and place the now empty 
 globe on an insulating stand. We may be now assured, on the faith of 
 Franklin's celebrated experiment of the electrical jar with movable coat- 
 ings, that all the interior surface of the globe is covered with electricity. 
 Introduce now a small insulated carrier-ball into this charged globe, so as 
 to touch the interior electrified glass, and again withdraw it. The carrier- 
 ball comes away quite neutral, as in the case of the hollow metal globe, not- 
 withstanding that it has been actually brought into contact with a dense 
 stratum of electricity, the presence of which may be made evident by simply 
 attaching the carrier-ball to the lower point of the insulated charging rod, 
 and introducing it as before. The electrometer will, if delicately hung, be 
 immediately affected ; or otherwise the charge may be shown by the medium 
 of an ordinary gold leaf. 
 
 That the failure of the carrier-ball to take up electricity is in no degree 
 dependent on the circumstance of what may be considered in the light of 
 electrical accumulation on an insulating surface as distinguishable from the 
 case of the hollow metal sphere, may be clearly shown by charging a plane 
 glass surface having movable coatings, and treating the plane charged glass 
 in a similar way. In this case electricity is freely taken up by the carrier- 
 ball from the charged side : it is hence evident that the globular form of the 
 surface is the immediate cause of the failure of the carrier-ball to take up 
 electricity from the glass. The carrier-ball, in fact, cannot assume that 
 induced electrical state requisite for its reception of free electricity ; the 
 forces operating on it being in contrary directions, its natural electricity 
 cannot recede from any point of its surface, hence all induced change in the 
 distribution of its own proper electricity is impossible. 
 
 This experiment with charged glass may be effectively managed by em- 
 ploying water instead of mercury ; or we may envelop the outer surface of 
 the globe in tinfoil and electrify the glass internally by means of a point 
 connected with the electrical machine, and projecting within the globe. 
 
 That induction would go on within the globe were it free to do so, may be 
 exemplified by one or two striking illustrations. 
 
 Let a charged hollow globe of metal or glass be placed on an insulating 
 support as before ; introduce within it the small insulated carrier-ball, and 
 whilst within the globe touch the carrier-ball with a light insulated wire, 
 projecting freely into the air ; remove this wire, and then withdraw the car- 
 rier-ball ; the carrier-ball will be found charged with electricity opposite to 
 that of the globe ; if the globe be plus, the carrier-ball will be minus, as 
 might be expected. In this state introduce the carrier-ball again within the 
 globe, and so as to touch the interior surface ; it comes away now quite 
 neutral, that is to say, it has taken up positive electricity from the interior 
 surface, either by immediate contact with the electrical particles in the 
 case of a charged globe of glass, or through the medium of the metal sur- 
 face in the case of a charged metal globe. 
 
 The same result ensues if we touch the carrier-ball with the free wire when 
 in contact with the interior surface. The insulated free wire comes away 
 positively charged with the electricity which had retired from the carrier-ball. 
 But if both the touching wire and the carrier-ball be raised together and re- 
 moved without the globe, then the whole evinces positive electricity ; for the 
 carrier-ball, whilst in contact with the interior of the sphere, having first 
 become negative, immediately takes positive electricity from the charged 
 sphere and becomes neutral, and probably remains so whilst in contact with 
 the sphere ; whilst the exterior ball of the touching wire evinces positive 
 electricity, being necessarily charged with the electricity superinduced upon 
 
OCCASIONAL MEMORANDA. 223 
 
 it by the first induction. On raising the whole system out of the sphere, 
 however, this superinduced electricity expands over the whole ; for the 
 original conditions are restored, whilst the new electricity taken up remains ; 
 hence the carrier-ball will now evince positive electricity. And this is really 
 what happens when an insulated wire and ball are introduced within the 
 globe of sufficient length to project into the air. 
 
 Although the two cases of charged globes to which I have thus called 
 attention may at first appear different, the one being a case of a hollow 
 globe of metal, the other of glass, yet a very little reflection will show that 
 both cases are virtually the same thing ; the - difference is a difference of 
 degree, not of kind ; they are, in fact, both reducible to the elements of the 
 electrical or Leyden jar. Indeed, every case of what is commonly called a 
 charged conductor, resolves itself into the form of a coated electric, and is 
 the result of a peculiar disposition and combination of electrics and con- 
 ductors ; it is, in fact, the accumulation of electricity upon the terminating 
 strata of a dialectric medium, bounded by, and in direct contact with, con- 
 ducting matter, either near or distant. Thus, in the case of what has been 
 termed a charged hollow metallic globe, in which all the charge is conceived 
 to be impelled, as it were, by the repulsive force of its particles from a centre 
 of force, and so finds its way to the surface of the metal, we find on an atten- 
 tive examination the following arrangement of conductors and electrics, and 
 into which every case of electrical charge may be finally resolved. We have, 
 first, a metallic surface ; secondly, exterior to this an insulating medium, 
 viz., atmospheric air, in a stratum of which, immediately surrounding the 
 globe, there is a dense electrical accumulation ; thirdly, beyond this stratum 
 we have in continuation other air not so immediately electrified, and suscep- 
 tible of further inductive change : the external air is in its turn bounded by 
 other conducting matter. When, therefore, we impart free electricity to the 
 hollow sphere, we do nothing more than cause an electrical accumulation to 
 ensue upon the dense stratum, according to the well-established principles 
 of the Leyden experiment. We do not, in fact, charge the sphere at all, 
 any more than we charge the coating of an electrical jar ; indeed, it is doubt- 
 ful if in any case we could charge a metallic conductor taken apart from, or 
 in the absence of, a dialectric boundary. The globe itself can be regarded 
 in no other light than that of the inner coating of a given dialectric bounded 
 by distant conducting matter, and which we may consider as the opposed 
 coating ; the metal of the globe is merely the conductor to the charge. The 
 inductive action upon which the charge depends, may be shown to extend 
 to great distances. Cavendish traced it from the centre to the walls of a 
 room 16 feet in diameter.* Faraday traced it from a ball suspended in the 
 middle of a room to the walls, 26 feet distant, f 
 
 If we examine the experiment of the glass globe with mercury coatings, 
 we find the elementary conditions precisely the same. Here we have an 
 interior coating of mercury, then an external dialectric, which in this case 
 is glass ; finally, an outer coating of mercury. Here, as in the preceding 
 case, we do not charge the interior mercury, as is well known, although it 
 may possibly, on being removed, be slightly electrified : the result of the 
 operation is to cause an electrical accumulation on the interior surface of 
 the glass. The great difference in the two cases simply consists in the more 
 or less perfect application of the coatings to the dialectric medium upon 
 which the charge depends. If we could suppose our hollow metal globe 
 surrounded by a second external globe of metal, the two globes being near 
 each other, but not anywhere touching, then we should have the two cases 
 
 * Cavendish, MS. f " Experimental Researches," 1303. 
 
224 ELECTRICAL FORCE. 
 
 identical, as seen in the very ingenious apparatus employed by Faraday.* 
 If, further, we imagine that, subsequently to the charging of this system, 
 both the globes were removed, as in the case of the mercury coatings, and 
 the intermediate air to remain, as it were, fixed and immovable, then 
 would be developed upon the boundaries of this stratum all the phenomena 
 of the hollow glass globe above described ; the charge would remain with 
 the air. If, in the case of the charged glass globe, we allowed the interior 
 metallic coating to remain under the form of thin metallic leaf attached to 
 the glass, then the final experimental conditions would be identical. As it 
 is, we operate upon the charge in the case of the hollow metal sphere, 
 through the medium of the coating ; in the case of the glass globe, we come 
 into contact more immediately with the charge itself. 
 
 The theoretical view, therefore, of the celebrated experiment of what has 
 been termed a charged hollow sphere, and which appears the best adapted to 
 explain the phenomena, is that of charged electrics generally. The free 
 electricity first communicated to the inner coating, viz., the metallic sphere, 
 operates by induction upon the nearest matter susceptible of electrical 
 change, and thereby develops and calls into operation the opposite elec- 
 tricity. The opposite forces thus brought into life tend to combine and 
 exhibit attractive force, and consequently come as near together as want of 
 conducting power in the intervening restraining dialectric will permit : the 
 imparted electricity must therefore necessarily find its way upon the exterior 
 surface of the hollow metal globe without the aid of any kind of repulsive 
 force to which the phenomenon has been hypothetically attributed. It is 
 in virtue of this kind of action that we are said to charge simple insulated 
 conductors generally. The amount of charge, however, or quantity of elec- 
 tricity which can bo sustained by them under a given electrometer indication, 
 can never be so great as in the case of systematically coated electrics of com- 
 paratively small thickness. The case of simple conductors is much the 
 same thing as the case of extremely thick glass, or the limiting of the free 
 action of one of the coatings of an electrical jar ; in either case the quantity 
 of electricity which can be accumulated under a given degree of the electro- 
 meter is greatly diminished. The experiment with an electrified hollow 
 globe therefore appears to have been expressed in very inexact terms, and 
 the phenomenon of the exclusive appearance of the charge upon the ex- 
 terior surface somewhat misapprehended. 
 
 It follows from these demonstrable conditions of electrical charge, that a 
 stratum of what may be termed electrical particles must always necessarily 
 exist upon the surfaces of a charged conductor, as is clearly demonstrable 
 by experiment ; and the electrical agency, whatever it be, penetrates to a 
 greater or less degree the substance of the air itself, or other dialectric 
 medium in contact with the conducting surface, as is well shown in Fara- 
 day's most comprehensive researches (1245). This is really the acceptation 
 of the term electrical atmosphere ; a term correctly applied by the celebrated 
 Volta, who most thoroughly comprehended the practical nature of electrical 
 force, notwithstanding that his power of rigorous thought has been ques- 
 tioned, and his theoretical views of electricity rather severely remarked on 
 by an eminent writer not altogether unbiassed by theoretical opinion, and 
 evidently not a little impatient of dissent, t That Volta was most perfectly 
 correct in attributing the phenomena of charged electrical conductors to the 
 presence of electrical atmospheres surrounding them, taken in the sense in 
 which I have just explained the term, is absolutely demonstrable by the 
 most conclusive experiments : we remove the metal, and there remains the 
 
 * " Experimental Researches," 1195. 
 
 t " Bibliotheque TJniverseUe," article "Volta." 
 
OCCASIONAL MEMORANDA. 225 
 
 charged stratum ; in other words, the atmosphere of electrical particles, as 
 it may be termed, exists entirely without and independent of it. 
 f The electrical stratum thus found to exist on electrified conductors appears 
 firmly held to the surface by attractive force, and is inseparable from it by any 
 movement of the body. Franklin whirled a charged ball attached to a silk 
 cord many times round in the air, and with great velocity, still the ball re- 
 tained its charge.* An electrified conductor, therefore, when transferred 
 from one place to another, may be supposed to carry the electrical stratum 
 along with it, just as the metallic coating of charged glass would do. It is 
 true that the term electrical atmosphere has been occasionally used in a 
 vague and unsatisfactory sense, and has hence been justly discountenanced 
 by many eminent physicists and mathematicians. When taken, however, 
 in the sense in which I have applied it as expressive of a demonstrable fact, 
 the question assumes quite another form. Volta, therefore, in referring the 
 phenomena of electrical attraction and repulsion, and the operation of elec- 
 trified bodies generally on each other, to the immediate action of the elec- 
 trical particles themselves held firmly on the surfaces of bodies, did in no 
 way violate sound deduction from rigorous thought, or evince in any degree 
 a vague and imperfect apprehension of the probable nature of electrical 
 force. When we remove the coatings of a charged electric, something is 
 evidently left behind some agency or source of power. What is that some- 
 thing ? In what does it consist ? It is evidently external to the metallic 
 surface with which the dialectric surface was previously in contact, although 
 inseparable from it so long as the two remain combined ; and it is really 
 from this something, which we express by the term electrical stratum or 
 atmosphere, that the phenomena we observe arise, and not from a hypo- 
 thetically charged conductor, f 
 
 * If we charge a plate of glass through plane movable coatings of gilt wood, the 
 coatings will adhere to the glass. 
 
 f The experiments of Beccaria and Franklin with the smoke of resin and colo- 
 phonia, and which they observed to collect about electrified conductors, so as to 
 envelop them, although perhaps no very satisfactory evidence of the existence 
 of a similar atmosphere of electricity, are still not without very considerable interest. 
 It is not easy to explain the adherence of these atmospheres of smoke to the charged 
 surface, admitting the theory of electrical repulsive force. 
 
 LIB.RAK i 
 
 UNIVERSITY OF 
 
 I CALIFORNIA. 
 
CHAPTER II. 
 
 ON THE NATURE OF THE ELECTRIC FORCE. 
 
 THE most approved view of electricity as a physical agency, not 
 only in remote but in modern times, assumes the existence of 
 an extremely subtle fluid form of matter of high elasticity, termed 
 the electrical fluid. So long ago as the year 1748, M. le Morin, 
 Professor of the Royal College of Chartres, in France, had, after 
 Lucretius's view of magnetic agency, and in common with many 
 other philosophers, the idea that this electrical fluid displayed 
 itself in the shape of exhalations from the surfaces of bodies ; and 
 he endeavoured to show that all bodies exhale and inspire this 
 electrical fluid.* The Hon. Robert Boyle, in 1760, seems to have had 
 a similar notion. He imagined that electrical bodies emit a gluti- 
 nous effluvium which seizes upon small and light particles of matter, 
 and then carries them back to the electrically excited substance. 
 Wilson, in 1756, Muschenbrock, in 1769, M. Sigaud de la Fond, 
 in 1781, together with many other eminent philosophers, all agreed 
 in the assumption of an existing material agency which they termed 
 an electrical fluid. M. Avogardo, an acute and distinguished 
 physicist, who wrote in more recent times, viz., in 1806, thought the 
 assumption of the existence of an electrical fluid diffused through- 
 out common matter, not only a necesssity, but a great theo- 
 retical fact since without it we can form no adequate conception 
 of electrical phenomena. Coming down to the present day, 
 philosophers adhere for the most part to the same notion ; they 
 are, however, divided in opinion as to the more immediate consti- 
 tution of this fluid. They all appear satisfied of its extreme, I may 
 say, vanishing rarity, and of its almost immeasurable elasticity and 
 expansive power ; but they differ as to its character whether it 
 be of a simple elementary kind, or compounded of other elementary 
 electrical fluids, On the hypothesis of its being an elementary 
 fluid, repulsive of itself, but attractive of common matter, whilst 
 the particles of common matter, deprived of this fluid, are also 
 
 4 
 
 * Nouvelle dissertation sur L'Electricite des Corps. 
 
EARLY NOTIONS OF THE ELECTRIC FLUID. 227 
 
 repulsive of each other, a large amount of electrical action is expli- 
 cable. So, on the other hand, by considering the electric fluid 
 as a compound of two elementary fluids, repulsive of them- 
 selves, but attractive of each other, and without any affinity for 
 common matter, we are enabled also to elucidate an equally large 
 amount of electrical phenomena. The objection to the first hypo- 
 thesis, adopted by -ZEpinus, appears to be the necessity for attribut- 
 ing a repulsive force to the particles of common matter, in order 
 to obtain a new force requisite to complete the condition of what 
 has been termed electrical equilibrium, but which inconvenience 
 would be obviated by conceiving the electrical fluid as made up 
 of two other elementary fluids each repulsive of its own particles 
 and attractive of the other. But since those who advocate this 
 latter doctrine do not pretend to have proved the actual existence 
 of an electrical fluid at all, but place it in theory with its two ele- 
 mentary fluids, merely as aiding us to predict results and helping 
 us to better comprehend electrical phenomena,* it seems of very 
 minor importance, taken in this point of view, whether we make 
 up the number of required forces in the one way or the other. 
 Hence, any rigid criticism of the theory of a single elementary 
 fluid, as interpreted by ^Epinus, is certainly uncalled for, and 
 superfluous. The time, as it appears to me, is fast approaching, 
 if not actually arrived, when it may be found desirable to abandon 
 the theoretical notion of an existing electric fluid or fluids, and 
 base our researches on a much less hypothetical and more solid 
 ground. 
 
 Newton, in his masterly researches, examined and deduced the 
 laws of gravity, but he did not stop to inquire whether gravity 
 were a fluid or fluids endowed with given properties; he was 
 content to treat gravity simply as a physical force, and deal with 
 it accordingly, upon principles of a sound practical philosophy, 
 and a profound geometry. We gain but little, so far as I see, 
 by resorting to any hypothetical view of the nature of electricity 
 as an occult power. Take the case of force, as exemplified in 
 the action of an elongated or compressed spring, which either 
 expands or contracts. No one would, I believe, build much on 
 the assumption of a fluid elasticity as the source of the observed 
 forces. Far better would it be to deal with electrical phenomena 
 as evidence of some, at present, inexplicable physical power of 
 nature, the laws of whose action we seek to determine ; and so 
 build upon the substantial ground of demonstrable facts, involv- 
 ing no assumption whatever. Moreover, this pretended electrical 
 
 *Hauy, vol.i., p. 103. 
 Q 2 
 
228 ELECTRICAL FORCE. 
 
 fluid or fluids, repulsive of their own particles, involve very 
 anomalous and contradictory elements, inconsistent with the nature 
 of things. We assume, for example, the existence of an extremely 
 rare species of matter, devoid of gravity, and yet exerting forces 
 of repulsion and attraction, through a mechanical agency, greater 
 than any similar force observed in ponderable matter. We give 
 to the particles of this imponderable electrical fluid a species of 
 repulsive force, essentially different from any kind of repulsive 
 force of which we have the slightest cognizance from experience. In 
 all elastic fluids, such as air, steam, and the like, the repulsive force 
 of the particles only extends to small distances, and to particles 
 in the immediate vicinity of each other ; whilst their attractive 
 power, by which they obey the law of a universal gravitation, 
 extends to all distances. Now electrical repulsion seems the very 
 reverse of this. Here the force of repulsion extends to great 
 distances, and we observe a detached mass of electrical particles in 
 one place, exerting repulsive force upon another mass of electrical 
 particles in another, without any apparent proximity ; and whilst 
 thus exerting an apparent force at very sensible distances, the 
 force between the molecules themselves at insensible distances is 
 so feeble as to be incapable of expansion into space, and often 
 inoperative between similarly charged bodies when brought into 
 ordinary contact with each other. 
 
 It is to be regretted that so few experiments have been instituted 
 of a simple and direct kind, with a view of elucidating the laws of 
 electricity, considered simply as a physical power, and without 
 reference to any theoretical assumption as to its occult qualities. 
 The experimental investigations upon which our fine mathematical 
 superstructures are based, are really very few, and certainly not 
 very satisfactory. They consist principally in determining the 
 law of distribution of a hypothetical electrical fluid on or between 
 electrified bodies of different forms, and also what has been 
 termed the density or thickness of the stratum in various 
 points of the charged surface. This is supposed to be effected 
 by touching the charged body in different parts with a small 
 insulated circular metallic disc of little thickness, and then 
 by transferring the disc to a balance of torsion, so determine the 
 quantity of electricity with which it has become charged, upon 
 the assumption that it is virtually an element of the surface it has 
 been caused to touch, and hence informs us of the proportionate 
 quantity of electricity at the point of contact. Now, although it 
 cannot be questioned but that the repulsive force indicated in the 
 torsion or bifilar balances is a true measure of what may be 
 
THE AUTHOR'S MODE OF INQUIRY. 229 
 
 called the quantity of force actually in operation, yet it is 
 extremely doubtful whether the tangent disc can be considered as 
 an element of the surface of the charged body in the point of 
 contact, and still more doubtful whether it comes away charged 
 with a quantity of electricity proportionate to the quantity actually 
 existing in that point, as I shall presently endeavour to prove. 
 
 The basis, therefore, of our present knowledge of electricity 
 being so far limited, it appears to me desirable to treat the question 
 of electrical force in a more general and less hypothetical manner, 
 and to deal with it by simple and more direct experiments, much 
 in the way that we are accustomed to examine other physical forces 
 of whose peculiar nature we have no adequate conception. It may 
 be, perhaps, convenient and useful to employ analogical expressions 
 in interpreting the phenomena observed, and to facilitate descrip- 
 tion, as when we talk of the quantity of electric matter, of its 
 tension, intensity, thickness of stratum, density, the addition or 
 abstraction of electricity considered as a fluid, and so forth : and 
 it is, nevertheless, most important to remember, that it is with 
 force simply, and the laws of force, that we are dealing, and not 
 with electrical fluids and their supposed distribution on the surfaces 
 of bodies. 
 
 It has been well observed by an eminent writer* that number, 
 weight, and measure, form the foundation of all exact science ; that 
 no branch of human knowledge can be held as being out of its 
 infancy which does not in some way or other frame its theory or 
 correct its practice with reference to these elements. Impressed 
 with this great truth, I have for a long time endeavoured to 
 contrive and perfect such quantitative measures and processes in 
 electricity as shall give us, in a speedy, direct, and simple way, the 
 information we require relative to the laws and operations of 
 electricity, taken simply in the light of a physical force, without 
 any care about its precise nature or its occult quality. The 
 instruments I employ, with the exception of the bifilar balance, 
 are now before us in their most recent and perfect state. I am 
 aware that I have before called attention to some of these instru- 
 ments, and to some few experiments connected with them; but 
 their more perfect and refined state, with a great number of new 
 results, I have still to notice. 
 
 These instruments have already been described in the previous 
 
 pages, and they consist 1st. Of the Scale-Beam Electrometer, 
 
 with new and recent improvements for estimating force, distance, 
 
 and other elements. 2nd. The Hydrostatic Electrometer. 3rd. 
 
 * M. Quetelet on " Probability." 
 
230 ELECTRICAL FORCE. 
 
 The Unit Measure, with its mode of operation. 4th. The Quan- 
 tity Jar. 5th. The Thermo-Electrometer. 
 
 Having explained the nature of these instruments for bringing 
 the observed phenomena of electricity under the dominion of 
 number, weight, and measure, I proceed to some of the leading 
 features of electrical action. 
 
 First. It is to be observed that electricity, considered as a simple 
 physical power, assumes two forms or modes of action. These 
 have been commonly termed vitreous and resinous electricities, and 
 have been further designated by the mathematical signs plus and 
 minus. These two forms or modes of action are always present in 
 every electrical development. They are really not distinct and 
 separate elements, except when taken in a certain sense, but are in 
 themselves identical as to active power. Thus, for example, when 
 we shoot an arrow from a bent bow, considered as a spring, the 
 particles of matter of the bow upon the concave surface are com- 
 pressed ; those upon the convex surface are strained and extended. 
 In this case there are two opposite forces. We cannot have one 
 without the other. The expression for the kind of tension of the 
 one may be different from the expression for the tension of the 
 other ; but still the force which propels the arrow is resolvable into 
 one thing, viz., the force of what we call elasticity. As I have 
 already remarked, we should gain little or nothing by assuming the 
 existence of two elastic forces in order to investigate and explain 
 the nature or mode of action of this exertion of mechanical 
 force. 
 
 Secondly. We have to observe that electrical force does not 
 operate in several directions at once with the same degree of power 
 as if exerted in one direction only. In the planetary system, the 
 attractive force of the sun on the earth is, so far as I understand 
 it, quite independent of its force on the other planets. The sun 
 would, for example, attract the earth at a given distance with the 
 same force as at present were the other planets all annihilated, and 
 the sun and earth the only bodies in space. Now, this is not the 
 case with electrical force. In the exertion of electrical force 
 between two bodies, the introduction of a third body causes the 
 force to decline, as has been already proved by experiment. 
 
 Let us now see on what this depends. With this view we 
 observe 
 
 Thirdly. That in electrical attraction the bodies are first 
 rendered attractable by a peculiar operation of force termed 
 induction, and then it is that attraction follows, but not before; 
 and if any disturbing force or impediment to the full and free 
 
REFLECTED INDUCTIONS. 231 
 
 exertion of this influence arise, the amount of force is either greatly 
 limited or does not appear at all. I wish the reader to observe 
 that I am now taking the phenomena as they present themselves 
 without any relation whatever to the possible occult qualities of 
 electricity, and that he should dismiss all impression with refer- 
 ence to the actual nature of electricity, of electrical fluids and such 
 like. 
 
 This species of inductive action admits of being illustrated by 
 experiment, or possibly in a more direct way by an approxima- 
 tion of what may be termed the charged to the neutral body, 
 instead of the neutral to the charged body. 
 
 Now, it is this kind of influence, apparently a sympathetic 
 action, exerted between the bodies at very sensible distances, that 
 constitutes the great characteristic of electricity as a physical 
 power ; and, although most probably resolvable into a species of 
 propagation of action through the particles of the medium inter- 
 posed between the attracting bodies, is still the great and first 
 phenomenon to be examined. 
 
 According to what law, then, does this action between the bodies 
 progress ? That is the first point to which I call attention. 
 
 Now, we are enabled to determine this by the instruments 
 before us, and by the torsion or bifilar balances : observing, 
 as stated in my Bakerian Lecture in the Philosophical Trans- 
 actions for 1839, that not only is the inductive influence exerted 
 by the electrified on the neutral body, but that the previously 
 neutral body so influenced reflects back a similar action upon 
 the electrified body; and that it is actually upon the simul- 
 taneous operation of these two elementary forces combined that 
 the whole force of attraction between the two bodies depends. If 
 either be interfered with, if both be not freely and fully exerted, 
 we have no longer an invariable amount of force. Murphy, in his 
 fine mathematical tract on the Laws of Heat and Electricity, 
 embraces a view not very dissimilar to this, alluded to in my paper 
 in the Royal Society's Transactions, tinder an hypothesis of a 
 principle of successive influences. He obtains numerical approxi- 
 mations to the representation of inductive forces between two 
 bodies, by calculating the effects of four or five successive acts of 
 influence. 
 
 Professor Thomson has, subsequently to all this, placed the 
 
 question under the form of a principle of reflected images, and has 
 
 obtained a series easily applicable, in some instances, to the exertion 
 
 of electrical attractive force between spheres and o ther bodies. 
 
 I have elsewhere described my method of measuring these 
 
232 ELECTRICAL FORCE. 
 
 inductions, and have shown that they are inversely as the simple 
 distance between the bodies. Calling H the force of induction, 
 we have 
 
 Ho.-], 
 
 These previous inquiries lead us more especially to a considera- 
 tion of the law of force as observed between two bodies one 
 electrified, the other neutral, and both susceptible of what may be 
 termed electrical change ly influence. 
 
 Now some considerable anomalies appear to have arisen in the 
 experimental results obtained by philosophers at different periods 
 relative to this question, some finding the force to vary in 
 the inverse ratio of the simple distance, others in the inverse 
 sesquiplicate ratio of the distance, that is, the square root of the 
 cube of the distance ; others in the inverse sesquiduplicate ratio of 
 the distance, that is, the square root of the 5th power of the dis- 
 tance ; others inversely as the square root of the distance. Cavendish, 
 Coulombe, Lambert, and some other philosophers, asserted that 
 electrical force was in the inverse duplicate ratio of the distance 
 in common with gravity ; others, such as Brook Taylor and 
 Muschenbrock, were led at one time to the conclusion that the 
 operation of these forces did not follow any particular law. The 
 question now is, how are we to reconcile such anomalies ? It is 
 next to impossible to imagine that men so distinguished in 
 physico-mathematical science should have made clumsy and ill- 
 contrived experiments, or that their results should not have been 
 faithfully recorded by them. It has, I believe, been too much the 
 practice to call the experimental skill of these philosophers in 
 question ; more especially in every case in which they failed to 
 establish the favourite law of force, viz., the inverse square of 
 the distance. I believe, however, I can conciliate all these apparent 
 anomalies, and show that both in electrictity and magnetism, any 
 law of force is possible so long as the bodies are in any way so 
 circumstanced as to limit the inductive change of which they are 
 susceptible. For pure electrical attraction we require both bodies 
 to be fully and freely susceptible of inductive change. In this 
 case the actual force will always vary in the inverse duplicate ratio 
 of the distances not as a simple elementary force but as the 
 result of a combination of the elementary inductions. 
 
 When a magnetic bar, A, Fig. Ill, is opposed to a similar, but 
 smaller, bar of iron, B, then a new polarity, n, is induced in the 
 near parts n of the iron, opposite in kind to that of the opposed 
 polarity A, whilst another polarity, q, arises in its more distant 
 
REFLECTED MAGNETIC ACTIONS. 
 
 233 
 
 parts, similar in polarity to that of the polarity A, but opposite to 
 that of the induced polarity n. This, however, is not all. On 
 further examination, we find that the temporary polarity n, thus 
 
 ~j 
 
 induced in the near surface of the iron, 
 
 operates in its turn on the near surface p of 
 
 the magnet, producing there, by a species of 
 
 reflection or reverberation, what may be 
 
 considered as a new polarity, p, opposite in 
 
 kind to that of the induced polarity n, but c- 
 
 similar to that of the permanent polarity A ; 
 
 that is to say, a portion of the force, which 
 
 under the ordinary conditions of magnetized 
 
 steel is directed towards the centre of the 
 
 magnet, becomes now determined toward 
 
 the iron in the direction p n. Fig> 1IL 
 
 The sublime discoveries of Newton, and his verification of the 
 general law of universal gravitation, naturally enough led philo- 
 sophers to entertain the notion that all forces in nature were subject 
 to the same law, and that this law was a dominant law of every species 
 of attractive force. Hence a strong bias existed in favour of such a 
 conclusion. Philosophers have not unfrequently gone out of their 
 way, and twisted experiments, solely with a view of verifying this 
 law in the exertion of electrical and magnetic force ; and have, as I 
 think, betrayed some want of respect for the investigations of those 
 philosophers who failed to establish it. But let us see how this 
 case stands. 
 
 Since the total force between the bodies suppose between the 
 two bodies p N will be dependent on the direct 
 and reflected actions we have just described; 
 then supposing they both go on sine limite ; that 
 p is a charged body and N a neutral body, the 
 first operation is a direct action of influence of 
 p upon N. Now let the number of particles of 
 excited force in the proximate surface n of N = 
 a, and the number of particles of reflected force in 
 the opposite surface j-; of p = b, the total force at 
 the distance, which we may call p n = unity or 1, 
 will be represented by a X b = a b. For the attrac- 
 tion of one particle in the proximate surface n of 
 the body N, to all the particles on the surface p of P, 
 will be as b. The attraction of two particles will be 
 2 b, of three particles 3 b, &c., and so on until we have the attrac- 
 tion of the whole number = a b> for the direct force. Well now, 
 
 Fig. 
 
234 ELECTRICAL FORCE. 
 
 suppose we diminish the distance between these bodies, say to 
 the line c d, and reduce it by one half in this case ; the induced 
 force is doubled, a becomes 2 a, the reflected force b becomes 2 b, 
 and the total force is 2 a X 2 b = 4 a b = 2 2 a b. If we again reduce 
 the distance p n, say to J n, the direct force a becomes 3 , the 
 reflected force = & becomes 3 b, and the total force is3aX3b = 9ab 
 = 3 2 a b. Thus, whilst distances decrease in the series 1, j, |, ^, 
 &c., forces increase in the proportion 1, 2 2 , 3 2 , 4 2 , that is, inversely, 
 as the squares of those distances. 
 
 So far, the results are plain and intelligible. But now let us 
 imagine that by an imperfection of insulation, or a limited surface, 
 we curb these elementary forces, and that the law of inductive force 
 between them changes, which, by experiments similar to those we 
 have described, it may be shown to do, we may, for example, with 
 a limited induction obtain a force in the inverse ratio of the square 
 root of the distance. If instead, therefore, of a free state of the 
 neutral body, we insulate it, and its limited capacity is so reduced as 
 to cause the induced and reflected forces to vary in the simple 
 inverse ratio of the square root of the distance ; then, reasoning 
 as before, and taking a unit of force = a b } at a unit of distance 
 = p n, we have no longer at the distance J c d, the induced force 
 $ 2 a, but equal 1*4 a, so that the reflected force instead of being 
 2 b, is 1*4 b, that is, inversely as the square root of \. In this case the 
 total force is 1'4 a x 1-4 b 2 a b y and the total forces are as 1 : 2, 
 that is to say, the force varies as the simple distance inversely a 
 result arrived at by Muschenbrock and ^Epinus ; so that we may 
 actually obtain almost any law of total force, depending on the 
 limiting elements of the inductive actions upon which the force 
 depends. 
 
 Suppose, by way of further example, the induced force should 
 approach the inverse ratio of the power of the distance ; then at 
 the distance j, the forces of induction are 1*68 a, instead of 2 a ; and 
 1-68 b, instead of 2 b. Then the total force being 1-68 a X 1'68 b, 
 will be 2 -8 a b nearly. So whilst the distances are I and 2, the 
 forces are as 1 and 2 -8 ; that is, in the inverse sesquiplicate ratio of 
 the distances, as found by Martin in three very unexceptionable 
 experiments on magnetic force. 
 
 I cannot but here express my conviction that, taken as a simple 
 elementary or fundamental law, there is no such law in nature as 
 that of an inverse duplicate ratio of the distance. That we can 
 have an effect in any other proportion than that of its cause, is, as 
 observed by the learned Dr. Samuel Clarke, to suppose the effect to 
 depend partly on its cause and partly on nothing. Take the case 
 
OCCASIONAL MEMORANDA. 
 
 235 
 
 of gravity or of light, which is but a force emanating in all direc- 
 tions from a centre. We have the light in a given point directly 
 proportional to the quantity of light in that point ; that is to say, 
 the effect is directly proportionate to the cause, but still represented 
 
 by the expression 
 
 Now one hitherto difficult physical problem in electricity is the 
 law of force as exerted between spheres and other curvilinear bodies, 
 and many theoretical views of this case have been advanced by M. 
 Biot and others. Nevertheless, taken in the way just explained, 
 the question becomes a very simple one, and we arrive at very 
 beautiful experimental results. 
 
 We may demonstrate with perfect success, that the quantity of 
 force being constant, the total force will vary directly as the areas 
 of the attractive surfaces, and in the inverse duplicate ratio of the 
 distance between them. (C.) 
 
 OCCASIONAL MEMORANDA AND EXPLANATOEY NOTES. 
 
 (C. p. 235.) The distribution of force in the charge and discharge of the 
 electrical jar admits of geometrical representation in a way calculated to 
 elucidate in a satisfactory manner the precise condition of this question. 
 
 Let the lineAC, Pig. 113, move for- 
 ward parallel to itself upon the line 
 CD, and generate the equal rectangu- 
 lar spaces A m, en,fo,gD; then these 
 spaces may stand for and represent 
 equal and successive quantities of 
 electricity communicated to the elec- 
 trical jar ; and the total rectangu- 
 
 Fig. 113. 
 
 lar space A B C D may stand for, and 
 represent, the total charge. Now, 
 since the force between the coatings continually decreases as the force through 
 the external circuit increases, therefore these forces may be represented by any 
 magnitude supposed to flow, and at the same time continually increase or de- 
 crease. Let therefore the line A C move forward parallel to itself as before, 
 and continually decrease by the linear magnitudes eh, fi, gk, B D (which are 
 proportional to the lines A e, A/, Ag, A B, that is, to the respective quan- 
 tities of electricity, G e, C/, G g, C B), and so generate the triangular space 
 ADC, which may stand for and represent the total force in the direction of 
 the coatings up to the point D, where it vanishes in explosive discharge. 
 Similarly, let the extremity A of the line C A flow and gradually increase by 
 the same proportional lines e h, fi, gk, B D, generating the triangular space 
 A B D, which may stand for and represent the total force in the direction of 
 the circuit up to the same point of explosion, D. This being understood, we 
 may observe that the first quantity accumulated being represented by the 
 
236 ELECTRICAL FORCE. 
 
 space e, the force in the direction of the coatings will be represented 
 by the space A C m h, and the force in the direction of the external circuit 
 (that is, the electrometer indication) by the triangular space A e h. Similarly 
 a double quantity of charge will be represented by the double space G f = 
 twice C e the force in the direction of the coatings by the space A n i ; 
 and the force in the direction of the external circuit by the triangular 
 space A fi, and so on. But the triangles A e h, A/*, are to each other as 
 the square of A. e to the square of A/; and since A /is double of Ae, these 
 triangular spaces are as 4 : 1. In the same way it may be shown that the 
 triangles A g k, A e h, are as 9 : 1 , and so on. Now, as these triangles stand 
 for and represent the force through the external circuit, that is to say, the 
 respective electrometer indications, whilst the sides A e, A/, A g, &c., are 
 proportional to the respective accumulations C e, G f, &c., we see that the 
 force in the direction of the circuit is (all other things remaining the same) 
 directly proportional to the square of the quantity accumulating. It is easy 
 to see by the diagram that the decrements of force upon each added quantity 
 will be as the rectilinear spaces AGmh, hmni, in ok, okD, whilst the in- 
 creasing force in the direction of the circuit will be represented by the 
 triangular spaces A e h, Afi, A g k, A B D. 
 
 In like manner let the lines A e, A/, Ag, &c., stand for the intervals, or 
 explosive distances between the balls of the Lane's discharger, which 
 the force in the direction of the external circuit Can break through, 
 the increasing force being represented by the triangles A e h, A f i, &c. : 
 then, since the attractive forces between the exploding points of the balls of 
 the discharger, with a given accumulation, are in the inverse duplicate ratio 
 of the distances A e, A/, Ag, &c., these attractive forces will be inversely 
 proportional to the same triangular spaces, A. eh, Afi, Ag k, &c. If, there- 
 fore, when force in the direction of the circuit is A e h, discharge takes place, 
 quantity being C e, distance A e, and attractive force between discharging 
 balls 1, then supposing A eto become A/ = twice Ae, the force between 
 the balls at distance A / would be only one-fourth as great ; that is to say, 
 it would be inversely as triangle A e h to triangle A f i. Hence with the 
 same quantity accumulated = C e, discharge coufd not occur at distance A/. 
 Let now the first quantity accumulated = e become twice as great ; that 
 is to say, let it be represented by rectangular space C / = 2 C e : in this case 
 the force in the direction of the external circuit would be represented by 
 triangle A / i = 4 times A e h ; and since the attractive force between the 
 exploding points of the balls of the discharger is as the squares of the quan- 
 tity accumulated, the attractive force through the external circuit with the 
 double accumulation C / becomes four times as great ; and is the same at 
 the distance A / with a double accumulation as at the distance A e with a 
 single accumulation ; in this case explosive discharge again ensues. 
 
 In a similar way it may be shown that when the distance A e is extended 
 to A g, the attractive force between the balls with the single accumulation C e 
 is reduced to one-ninth, in which case no explosive discharge could occur at 
 the distance A g. Let the quantity accumulated, however, become three times 
 as great, that is to say, let the rectangular space C e become C g ; in this case 
 the force through the external circuit is represented by the triangular space 
 A g k = 9 times the triangular space A e h ; but since these spaces are in- 
 versely proportional to the attractive forces at distances A e, A g, the attrac- 
 tive force at the distance A g is the same with a treble accumulation, as at the 
 distance A e with a single accumulation ; explosive discharge will therefore 
 again occur, and so on. Hence the interval at which discharge occurs, as 
 measured by a Lane's discharger, L, Fig. 7, will be directly as the quantity 
 accumulated ; whilst the electrometer indication or force through the external 
 
OCCASIONAL MEMORANDA. 237 
 
 circuit will be as the square of the quantity, being as the triangular spaces 
 A e h, A / i, A g k, &c. M. De la Eive, in his comprehensive treatise on Elec- 
 tricity, considers this result as somewhat remarkable.* It is evident, how- 
 ever, from the geometrical diagram above referred to, that it could not be 
 otherwise, and is a necessary result of the forces in operation. We may infer 
 from the equality of the triangles ADO, A D B, that at the instant of 
 explosive discharge the force A B D through the external circuit has super- 
 seded the force A D in the direction of the coatings. If, therefore, we 
 suppose discharge to occur with the successive accumulated quantities e, 
 C /, C g, &c., then discharge with first quantity = C e taking place at distance 
 A e, the force through the external circuit will be represented by the triangle 
 A m e, and the force in the direction of the coatings by the triangle A m C. 
 When discharge occurs at the distance A / = 2 A e with a double accumula- 
 tion C / = 2 C e, the force through the circuit will be represented by triangle 
 A nf, and force in the direction of the coatings by triangle A n C, and so on. 
 We have here to observe, however, that, in estimating the forces through 
 the circuit, the force for the quantity e, with the distance of discharging 
 balls A/ = 2 A e, will not be represented by the triangle Km e (when the 
 quantity becomes C/ = 2 C e), but by the triangle A.r e, or one-fourth of 
 the triangle A n f = force through the circuit at the instant of discharge with 
 a double quantity. When, therefore, the distance A e is increased to A / 
 with the unit of charge represented by C e, then force acting through the 
 external circuit is represented by triangle Are, and not by triangle Am e. 
 
 In order, therefore, to measure a double accumulation by means of a 
 movable electrometer, we must oppose to the force through the circuit four 
 times the resistance: thus, in the operation of Cuthbertson's ingenious 
 steelyard discharging electrometer, we must, in order to obtain a double 
 accumulation, set the slider of the balance-arm to four times the number of 
 grains ; in order to obtain a treble accumulation, we must set the slider of 
 the balance-arm to nine times the number of grains, and so on. When, in 
 the application of Lane's discharger, therefore, discharge occurs at a double 
 distance, the quantity of electricity accumulated is twice as great, and the 
 respective forces through the circuit as 1 : 4, as represented by the triangles 
 A e r, A/w; when discharge occurs at a treble distance, the quantity of 
 electricity accumulated is three times as great, and the force through the 
 circuit nine times as great, as represented by the triangles A e s, A g o, and 
 so on, the force through the circuit being, as already observed, as the 
 square of the quantity. 
 
 It may perhaps be as well to further remark that, although, according to 
 the diagram, the forces A B D, A D C are at the instant of discharge con- 
 sidered as equal, we cannot however infer their precise equality, or suppose 
 the decreasing force A D C in the direction of the coatings to absolutely vanish 
 at the point D : hence some little residuum, commonly called residual 
 discharge, may remain in the jar in consequence of the vanishing attractive 
 force which the coatings exert upon each other, as is found by experiment. 
 
 * " Ce qu'il y a d'assez remarquable, c'est que la distance alaquelle une ddcharge 
 entre deux balles chargees d'electricites contraires pent avoir lieu, est simplement 
 proportionnelle aux quantity's d'electricite, tandis que les forces attract! ves sont 
 proportionnelles aux carres de ces forces." Traite d 1 Electricity tome i. p. 66. 
 
CHAPTER III. 
 
 FURTHER INQUIRIES CONCERNING THE LAWS AND OPERATIONS 
 OF ELECTRICAL FORCE.* 
 
 1. BY quantity of electricity we are to understand the amount 
 of the unknown agency, whatever that may be, constituting elec- 
 trical force, and which, being accumulated, may be represented 
 by some arbitrary quantitative electrical measure. 
 
 The actual quantity, as thus represented, which any insulated 
 conductor can sustain, under a given electrometer indication, was 
 not unaptly termed by Cavendish the charge of that body, or 
 electrical charge ; this, in fact, is the only true measure of what 
 may be called electrical capacity. 
 
 Electrical charge, therefore, is the quantity of electricity which 
 can be placed upon a given insulated conducting surface under a 
 given electrometer indication. 
 
 Electrical intensity, on the contrary, is the electrometer indica- 
 tion answering to a given quantity upon a given surface. 
 
 2. The term electrical intensity has often been employed and 
 received in a very indefinite and vague sense, as I have endea- 
 voured to show else where, f and has been made to signify certain 
 hypothetical qualities assumed to be peculiar to the occult nature 
 of electrical force. It has been used to express, for example, 
 different degrees of elasticity or tension, different degrees of 
 density, and such like; thereby taking it for granted that the 
 electrical agency is capable of altering its state or condition under 
 certain circumstances. We have, however, no substantial evidence 
 of this. 
 
 I have in vain sought for experiments which could at all lead us 
 to infer the possibility of effecting a change in the constitution of 
 electrical force without, at the same time, varying the quantity of 
 the agency in operation, or the surface on which it is distributed, 
 
 * This forms the subject of a paper read before the Eoyal Society, June 
 16th, 1864. The reference numbers within brackets refer to this chapter 
 only. ED. 
 
 f Phil. Mag. for December, 1863, Supplement. 
 
CAPACITY OF CONDUCTORS. 239 
 
 and have been hence led to conclude that the term intensity, when 
 rigidly interpreted, is, after all, only another form of expression 
 for quantity, signifying nothing more or less than the quantity 
 of electrical force operating at a given point of the charged 
 surface. It is in this sense I employ the term intensity in the 
 following pages. 
 
 3. Electricians have observed, from a very early period of the 
 progress of modern electrical discovery, that, although the quan- 
 tity of electricity which insulated conductors can receive depends 
 not upon their solidity or volume, but upon their surfaces ; yet 
 bodies will not take up electricity in proportion to their surfaces. 
 This important fact was first noticed by Le Monnier, in " L'His- 
 toire de 1' Academic," so long since as the year 1746. Cavendish, 
 in his manuscript papers (1770), notices the same fact. The most 
 interesting notice, however, of past times relating to quantity, as 
 regards surface, is to be found in Volta's memoir on the capacity 
 of electrical conductors, which appeared in Rozier's Journal in 
 1779. Volta shows that if any plane surface be extended in 
 length, it can sustain a greater quantity of electricity, although 
 the area of the surface remains the same; Exp. 10 (25). Thus 
 a conducting plate, A, Fig. 114, 1 foot square, of small thickness, 
 would not sustain so great a charge as a rectangular conducting 
 plate, B, 48 inches long by 3 inches wide (and of the same thick- 
 ness), although the areas do not differ. Yolta attributes this 
 result to the circumstance that the electrical particles are, in the 
 case of the elongated surface, " placed further without each other's 
 influence," in consequence of which the electrical excitation or 
 
 Fig. 114. 
 
 action of the particles upon each other is diminished ; it being, in 
 fact, demonstrable, by familiar experiments, that when two simi- 
 larly charged bodies, with attached electroscopes, are brought near 
 each other, the angular divergence of the electroscopes increases, 
 and the accumulated electricity does not rest upon them with the 
 same quietude as it does when the two bodies are at a distance from 
 each other. Let, for example, two equal spheres, equally charged 
 (having attached electroscopes), be caused to approach each other, 
 the angular divergence of the electroscopes will be greatly in- 
 creased ; conversely, when the same two spheres are widely 
 separated, the angular divergence of the electroscopes will be 
 
240 ELECTRICAL FORCE. 
 
 greatly diminished. Supposing the equal spheres, therefore, to 
 represent or stand for approximated or separated particles, Yolta's 
 view, as must be admitted, is very consistent with experiment. 
 The greatest intensity (2), according to Yolta, is when the elec- 
 tricity is accumulated on a given circular area, and the least when 
 the same area is expanded into a right line of small width. In 
 the former case the electrical particles are congested or packed, as 
 it were, more closely together ; in the latter this congestion is 
 relieved, and the particles are at greater distances from each 
 other's influence. Yolta has, no doubt, shown much ingenuity in 
 this explanation of the phenomena. 
 
 4. Some further elucidation of the question, however, is de- 
 rived from more recent inquiries into the operation of what has 
 been termed neutral electrical induction. We cannot possibly 
 extend the charged surface A, Fig. 114, in length as B, without at 
 the same time exposing it to a greater amount of neutral in- 
 ductive action, arising from the influence of surrounding neutral 
 matter, which, on the principles of the condenser, diminishes the 
 electrometer indication or intensity, hence the intensity is less : 
 and the given surface can receive an increased quantity, before the 
 electroscope again rises to the same angle. We have to consider, 
 therefore, not only how the particles are disposed or grouped in 
 relation to each other, but how they are disposed or grouped in 
 relation to external bodies, by which they are subjected to a greater 
 or less amount of neutral inductive action. If, for example, we 
 have an insulated charged sphere, with an attached diverging 
 electroscope,* and an adjacent neutral surface, the divergent elec- 
 troscope, under the influence of the neutral surface, approximated 
 in a greater or less degree towards the charged surface, declines, and 
 the given surface receives a higher charge before the electroscope 
 would again diverge to the same angle ; showing the influence of the 
 adjacent neutral matter on a charged surface, and the consequent 
 increase of charge, under the same electrometer indication. In 
 accordance with this fact, Yolta shows, as already stated (3), that 
 a given conducting surface extended in length takes an increased 
 charge. We may in fact perceive, by reference to Fig. 114, that 
 a given surface under the form of a square plate, A, is exposed to 
 a less amount of exterior inductive action than when elongated or 
 drawn out into a rectangular plate, B, whose length equals four 
 times the side of the square, although the surfaces are in each 
 case actually the same ; the intensity, as illustrated by the sphere, 
 
 * Figs. 27, 38, and 39, in Part I., show the kind of figures referred to in 
 this paper. ED. 
 
LAW OF ELECTRICAL CHARGE. 241 
 
 is consequently diminished, and the quantity under the same 
 electrometer indication increased. This latter condition involves 
 the extent and disposition of the linear boundary, or perimeter, of 
 the given surface, as we shall presently show. 
 
 5. No very satisfactory experiments appear to have been in- 
 stituted, showing the relation of quantity to surface, at least none 
 from which the general laws of this relation have been deduced : 
 the quantity upon a given surface has been often vaguely estimated, 
 without any regard to a constant electrometer indication or in- 
 tensity. For example, we can scarcely infer, from the beautiful 
 experiment of Coulombe, in which a charged sphere was brought 
 into contact with an insulated neutral circular plate of twice its 
 diameter, that because the charge was shared between them in the 
 proportion of the two exposed surfaces of the plate to the one ex- 
 posed surface of the sphere, that therefore the capacity of the 
 circular plate is twice the capacity of the sphere. If the electrical 
 capacity, or charge of the plate, were fully twice that of the sphere, 
 we could, under the same electrometer indication, place twice the 
 quantity on the plate that we could on the sphere ; but this is 
 not possible, as we shall eventually show (26, Exp. 16). 
 
 In the meantime we may observe, that the sharing of electricity 
 between two bodies is a different question from that of electrical 
 charge, according to our previous definition of it (1). 
 
 6. In further investigating the laws of electrical charge and 
 intensity, together with Yolta's discovery of the increased capacity 
 of conductors when the surface is extended in length, I was 
 led to infer that the quantity of electricity which any plane rect- 
 angular surface could receive under a given intensity (2) depended, 
 not only on its surface, but also on its linear boundary extension, 
 or perimeter. Thus, the linear boundary of 100 square inches of 
 surface under a rectangle of 37*5 inches long, by about 2*66 inches 
 wide, is about 80 inches ; whilst the linear boundary of the same 
 100 square inches of surface under a plate 10 inches square is 
 only 40 inches. The linear extensions of the two surfaces in this 
 case may be taken as 2 : 1 ; so that the linear boundary of the 
 rectangle is twice that of the square, in accordance with which 
 fact we find, as already noticed (3), that the charge of the rectangle 
 is much greater than that of the square, although the surfaces are 
 equal, or nearly so. 
 
 7. A rigid experimental examination of this question (25, 
 Exps. 7, 8, 9, and 10) led to the conclusion that electrical charge 
 (1) in certain cases depended on the surface and linear extension 
 conjointly. 
 
242 ELECTRICAL FORCE. 
 
 It appeared from a series of careful experiments with, improved 
 instruments, to be hereafter described, in which rectangular con- 
 ducting plates of small thickness and various forms and extensions 
 were employed 
 
 1st. That there exists in every plane surface what may be 
 termed an electrical boundary, having a relation to the grouping 
 or disposition of the electrical particles in regard to each other 
 and surrounding matter. In circles and globes the perimeters, 
 or boundaries, are represented by the circumferences. 
 
 2nd. That if this boundary be constant, the charge varies with the 
 square root of the surface ; if the surface be constant, the charge 
 varies with the square root of the boundary ; if the boundary and 
 surface both vary, Ihe charge varies with the square root of the 
 surface multiplied into the square root of the boundary. Thus, 
 calling C the charge, S the surface, B the boundary, and u some 
 arbitrary constant, depending on the electrical amount of charge, 
 
 we have C = u\/S B (25), which will be found, with some excep- 
 tions, a general law of electrical charge. 
 
 It is evident from the above formula, C = >/ 8 B, if when we 
 double the surface we also double the boundary, the charge will 
 be twice as great ; that is to say, the charge in this case may 
 be said to vary with the surface, but not otherwise, as we shall 
 presently see (25, Exp. 10). According to this formula, if /and b 
 represent the length and breadth respectively of a plane rect- 
 angular surface, then the charge of such a surface is expressed by 
 u v2 I b (I -f- b), which is the law agreeing with experiment. 
 
 We have, however, to remember in all these cases the differ- 
 ence between charge (1) and intensity (2). 
 
 8. On examining the electrical intensity of plane rect- 
 angular surfaces, that is to say, the electrometer indication of a 
 given quantity in operation at any one point of the charged 
 surface, it was found to vary inversely as the boundary multiplied 
 into the surface. Thus, if the surface be constant, the intensity 
 is as the boundary inversely ; if the boundary be constant, the 
 intensity is inversely as the surface ; if both vary, it is inversely 
 as the surface, multiplied into the boundaries. Thus, calling 
 
 intensity, that is, electrometer indication = E, we have E =T=T- 
 the surface being constant ; if the boundary be constant, we have 
 E = -^ ., If, therefore, both vary alike and together, E may be 
 said to vary inversely as the square of either. If, for example, 
 
CHARGE UNDER GIVEN INTENSITY. 243 
 
 when S is double, B is also double, E may be said to vary as 
 -i (25, Exp. 10). 
 
 The electrical intensity of a plane rectangular surface being 
 given, we may always deduce therefrom its electrical charge under 
 a greater given intensity, since we only require to determine the 
 increased quantity requisite to bring the electrometer indication 
 up to the given required intensity. This we may readily effect, 
 the intensity being by a well-established law of electrical force 
 as the square of the quantity (21, Exp. 1). 
 
 Let, for example, the intensity of two surfaces, which we will call 
 A and D, each charged with a given quantity, m, be as 1 : 4, the in- 
 tensity of A with quantity, m being equal to 1, and the intensity of 
 D with the same quantity m being equal to 4 ; and let it be required 
 to determine what quantity must be accumulated on A to produce 
 the same electrometer indication = 4 as that shown by D, with 
 the given quantity m. Then if m represent the total quantity to 
 be accumulated on A, in order that the intensity may become four 
 times as great, that is, equal to the intensity of D, intensity 
 of A with quantity m is to its intensity with quantity m r as is m 2 : m' 2 ; 
 or m 2 : m' 2 : : 1 : 4; we have consequently m : m' : : Vf~: \/4~; 
 therefore m' = 2 m, which is twice as great as that with which 
 the surface D is charged ; so that the charge of A is twice that of 
 D ; or in other words, A can receive twice the quantity under the 
 same electrometer indication, as is demonstrable by direct experi- 
 ment (25, Exp. 10) ; and we have the charge as the square root 
 
 of the intensity inversely, or C = -r^r 
 
 9. It is important to observe that these laws relating to charge, 
 intensity, surface, &c., apply more especially to continuous surfaces 
 taken as a whole. If a surface be divided into equal separated 
 parts, the formulas we have just quoted require correction. Let, 
 for example, a plane rectangular surface, A, Fig. 115, of a given 
 length and breadth, be divided into two equal portions, a, b, sepa- 
 rated and placed distant from each other, as at B ; further, suppose 
 a given quantity of electricity = 1 accumulated on A to be disposed 
 on the two equal half surfaces a and b, so that each of the separated 
 half surfaces, a, b, may in this case have one-half the total quantity 
 accumulated on the total continuous surface A ; on comparing the 
 two half surfaces, a, b, with each other, it is clear that the total 
 quantity = 1 disposed upon them, occupies two equal surfaces, 
 having separate and equal linear boundaries ; so that the given 
 
 R 2 
 
244 ELECTK I CAL FORCE. 
 
 quantity equal 1, may be considered to be accumulated on a 
 double surface, a -j- b> as compared with one of the half surfaces 
 a, taken singly. Hence the quantity upon the two half surfaces 
 united as at A, would be double that of the quantity upon one 
 
 
 
 
 
 a 
 
 
 
 6 
 
 
 
 Fig. 115. 
 
 
 of the half surfaces, a, taken singly ; and the intensity of the total 
 quantity accumulated upon the separate surfaces, a, b, would, 
 when taken together, be the same as the intensity of one-half the 
 quantity upon one of the half surfaces, a, only. We therefore 
 perceive, that if as we increase the quantity we also increase the 
 surface and boundary, the intensity does not change. If three 
 or more separated equal spheres, for example, be charged with 
 three or more equal quantities, and be each placed in separate 
 connection with the electrometer, the intensity of the whole is not 
 greater than the intensity of one of its parts. Similar results 
 ensue in charging any united number of equal and similar elec- 
 trical jars. An electrical battery, for example, consisting of five 
 equal and similar jars, charged with a given quantity = 5, has 
 the same intensity as a battery of ten equal and similar jars 
 charged with a quantity = 10 ; the intensity of the ten jars 
 taken together is no greater than the intensity of any one of 
 the jars taken singly. Eef erring to B, Fig. 115, suppose the 
 whole quantity = 1 to be now accumulated on one of the half 
 surfaces, a, only, the other half surface, b, remaining charged 
 with quantity 1 as before, then, since the intensity varies as 
 the square of the quantity (21, Exp. 1), the intensity of a with 
 the quantity 1 will be to the intensity b with the quantity i as 
 I 2 : (i) 2 or as 1 : J ; that is to say, the intensity of a with quan- 
 tity = 1 will be four times the intensity of b with quantity = i, the 
 surface being in each case the same. When, therefore, the quantity 
 = 1, instead of being accumulated upon one of the half surfaces 
 a only, is accumulated on both the half surfaces, a + b, so that 
 each may contain quantity , the intensity of the two separate half 
 surfaces, a, b, is only one-fourth the intensity of the total quantity 
 accumulated on one of them. It appears from this, therefore, that 
 
EFFECTS OF VARYING SURFACE, 245 
 
 in accumulating a given quantity of electricity upon a double 
 surface, B, in separate and equal jars, a, fr, the boundary being 
 also double, the intensity varies as the square of the surface in- 
 versely, and we have E 2 (8) ; hence the surfaces, a, b, can 
 
 receive, when taken together, under the same electrometer indica- 
 tion, twice the quantity which either can receive alone, and the 
 charge varies with the surface, the boundaries being equal. In 
 this case, as we increase the surface in separate equal parts, we 
 also increase the linear boundary, and consequently we increase 
 the charge. 
 
 Taking, therefore, a given surface in equal and divided separate 
 parts, we may conclude that the intensity varies with the square 
 of the quantity directly, and with the square of the surface in- 
 
 Q 2 
 
 versely, from which we derive the formula E * , which fully 
 
 represents the phenomenon of a constant intensity attendant upon 
 the charging of equal separated surfaces, with quantities increasing 
 as the surfaces in the case of charging a united series of equal 
 electrical jars. Thus, if Q = 1, and S == 1, we have E = T = 1 ; 
 
 2 2 
 if Q = 2, and S = 2, we have E = -^ = 1, and so on. In cases, 
 
 & 
 
 however, in which, from any circumstance, E does not vary with 
 - j ? this formula does not apply. 
 
 If, for example, E varied as , which under certain circum- 
 
 D 
 
 stances is found to be the case (26, Exp. 15), the intensity is no 
 longer constant, as the quantity increases with the surface. This 
 leads to a consideration of the case in which separate sur- 
 faces, a, by Fig. 115, are united into one continuous whole, as at 
 A. Let, for example, the two half surfaces, a, b, be joined at 
 their extremities so as to form a continuous surface, A, and 
 suppose a quantity = 1 to be accumulated upon them thus com- 
 bined, whilst either of the separate half surfaces, a, b, remains 
 charged each with a quantity -| ; we have then, as in the pre- 
 ceding case, a double quantity accumulated on a double surface, 
 A, the only difference being that the double surface in the former 
 case, B, is made up of two equal separate parts, whilst in the 
 present case, A, it is made up of one continuous whole. We 
 might hence be led to conclude that the intensities of a and A will 
 be the samo, and that we could accumulate on A, under the sam 
 
246 ELECTRICAL FORCE. 
 
 electrometer indication, twice the quantity we could accumulate 
 on a or b. Such, however, is not the fact ; the intensity of a 
 double quantity on A exceeds the intensity of half the quantity on 
 a, whilst the charge of A is considerably less than twice the 
 charge of a ; or the charge of a and b when apart from each 
 other. The cause of this appears to be the difference in the 
 relative linear boundaries of the figures. We have already ob- 
 served that, in order to accumulate a double charge, we must have 
 not only a double surface, but also a double linear boundary, as 
 shown (25, Exp. 10). JN"ow the linear boundary of A cannot 
 possibly be double the linear boundary of a. Thus, calling L the 
 length, and U the breadth of the figures, we have for the linear 
 boundary of A, 2L -f- 2U, and for the linear boundary of a, 
 2 (i L + U) = L + 2 U. But it is evident that 2 L + 4 U is equal 
 to twice the boundary of a ; the boundary of A, therefore, is less 
 than twice the boundary of a by 2 U ; that is, by twice the breadth ; 
 consequently, the intensity of A with twice the quantity, is greater 
 than the intensity of a with one-half the quantity. The charge 
 (1) of A, therefore, is less than twice the charge of , although the 
 surface of A is twice the surface of a. If we suppose the two half 
 surfaces a and b, instead of being joined at their extremities, as A, 
 to be joined at their sides, as D, Fig. 115, the linear boundary of D 
 is still less than twice the linear boundary of a by L ; that is, by 
 twice the length of a ; and is less than the linear boundary of A 
 by L 2U ; that is, by the length minus twice the breadth. The 
 intensity therefore of the two half surfaces a and b, when joined 
 at their sides, as at D, is much greater than when joined at their 
 extremities, as at A ; the charge therefore of D is less than the 
 charge of A, and is in either case much less than the charge of 
 the two equal half surfaces a + b apart from each other, as at B ; 
 so that, as we have observed (3), (Fig. 114), a given conducting sur- 
 face, B, will, as stated by Yolta (3), receive a greater charge when 
 extended in length, as at A. From these analytical inquiries also 
 it is further evident, as observed by the early electricians (3), 
 that conducting bodies do not take up electricity in proportion 
 to their surfaces, except under certain conditions of surface and 
 boundary (25, Exp. 10). If the breadth U and the length L of a 
 given surface, A, be capable of unlimited variation, the surface 
 remaining constant, then, as observed by Yolta (3), the least 
 quantity that can be accumulated on a given conducting service 
 under a given electrometer indication, is when the given surface is 
 under the form of a circular plate, that is to say, when the 
 boundary is a minimum; and the greatest when extended in 
 
SPHERES. OF UNEQUAL SURFACE. 247 
 
 a right line of small width, that is, when the boundary is a maxi- 
 mum. 
 
 10. From these facts it is further evident that in the combina- 
 tion of two similar surfaces by contact, as for example, two circles, 
 two spheres, two rectangular plates, two cylinders, &c., we fail in 
 either case to obtain twice the charge of one of them taken 
 separately. If we bring them into contact either at their extre- 
 mities or sides, we fail to decrease the intensity, the quantity 
 being constant, or increase the charge, the intensity being con- 
 stant. It being evident that whatever tends to decrease the inten- 
 sity or electrometer indication, must increase the charge, that is 
 to say, the quantity that can be accumulated under that intensity ; 
 conversely, whatever tends to increase the electrometer indica- 
 tion must decrease the charge, that is to say, the quantity 
 that can be accumulated under that electrometer indication. As 
 a further illustration, let two equal separated spheres, for example 
 (5), equally charged, be made to touch each other. In this case 
 the electrical particles of the charge will be more congested 
 and within each other's influence when touching each other, 
 and differently grouped in relation to external bodies, than 
 when separated and at a distance from each other (3) (4) ; con- 
 sequently, the intensity of either sphere in a separated state will 
 be much less than the intensity of either of the spheres in contact. 
 The total intensity, therefore, of the two spheres in a separated 
 state will be much less than the total intensity of the two spheres 
 in contact ; the sum, therefore, of the charges of the two spheres, 
 when taken together in a separated state, will be greater than the 
 sum of the charges of the two spheres when in contact. Hence, 
 two equal spheres when in contact, similarly charged, have not 
 the same intensity as one of the spheres taken alone ; the charge of 
 the two spheres in contact is consequently not double the charge 
 of one of them taken alone a result which applies to any two 
 similar figures, either separated or in contact, as is found by experi- 
 ment. Take, now, the case of two spheres in which the surface 
 of the one is double that of the surface of the other, and suppose 
 a given quantity of electricity = 1 to be accumulated on each ; 
 then, in any one point of the larger sphere, there will be only 
 one-half the quantity there is in one point of the smaller sphere. 
 We might infer from this that the intensity in any one point of 
 the double surface is only one-fourth the intensity in any one point 
 
 of the single surface, according to the formula E = (8) ; so that 
 if we double the quantity upon the double surface, the intensity 
 
248 ELECTRICAL FORCE. 
 
 would be the same as half the quantity upon the single surface. 
 Such, however, is not the result of experiment, owing to the 
 relative congested states or groupings of the electrical particles in 
 regard to each other and external bodies in each of the spheres ; 
 as also from the circumstance of the total quantity upon the 
 smaller sphere being grouped upon a double surface on the larger, 
 taken as a whole, and not upon a double surface in two separate 
 e^ual parts, as with spheres of equal size. 
 
 In the small sphere the electrical particles are more congested, 
 or within each other's influence relative to its circumference, than 
 they are in the large sphere relative to its circumference ; the cir- 
 cumferences of the two spheres, in fact, are not double of each 
 other, but are as 1 : V 2, that is to say, as their diameters ; the 
 surfaces of the spheres being double, whilst the quantity in any 
 one point of the double surface affecting the electrometer is only 
 one-half the quantity affecting the electrometer in any one point 
 of the single surface : a sphere of twice the surface, therefore, does 
 not take a double charge. The charges of spheres, in fact, are 
 found to be as their diameters, that is, as the square roots of the 
 surfaces (11) ; hence, the charge of a sphere of diameter equal 1 : 
 the charge of a sphere whose diameter equals 2 : : 1 : 2, as just 
 observed, whilst the intensity varies as the surface inversely, that 
 
 is, as . If we could obtain spheres whose surfaces varied with 
 
 S 
 
 their circumferences or boundaries, then the intensities with 
 quantities increasing as the surfaces would be constant ; but this 
 is not possible. We cannot, for example, obtain a sphere whose 
 surface and circumference are each double the surface and circum- 
 ference of another sphere. 
 
 11. It follows from these and similar considerations that if in 
 any case it should so happen that the grouping or disposition of 
 electrical particles, in regard to surrounding matter, be such as not 
 to materially influence the amount of external inductive action on 
 the electrical boundaries of the surface, then this external induc- 
 tion would not materially affect the result ; for example, in all 
 similar figures, as squares, circles, &c., the electrical boundary of 
 the figure is, in relation to surrounding matter, pretty much 
 the same, whatever be the extent of the surfaces. 
 
 Thus, for example, if we suppose two circular plates to be sur- 
 rounded by an outer circle of inductive rays, both circles will be 
 similarly placed in regard to those rays, so that when charged 
 with electricity, the particles would be grouped, or disposed upon 
 their surfaces, in regard to this induction, pretty much in the 
 
VARIATIONS OF SURFACE. 249 
 
 same way, and little or no difference might ensue ; the difference 
 depending merely upon a slight increase or decrease of distance. 
 We should hence only have to consider the grouping or disposi- 
 tion of the particles upon the surfaces in regard to each other, and 
 might, therefore, take into calculation the surfaces only, without 
 regard to the boundaries or circumferences of the circles ; a con- 
 sideration which materially influences charge, surface, and intensity, 
 as regards similar figures ; thus, in all similar figures, such as 
 square plates, circular plates, globes, &c., their relative charges 
 were found to be as the square roots of the surfaces only, without 
 regard to the boundaries. The charges of square plates were 
 found to be as their sides ; the charges of circular plates as their 
 diameters, as also the charges of globes (26, Exps. 11, 12, 13, &c.). 
 In cases of rectangles also, in which the boundary extension is the 
 same whilst the surfaces vary, the same result ensues ; the charges 
 are as the square roots of the surfaces simply. In cases of square 
 or circular plates and globes it is easy to see, as we have already 
 observed, that the boundary extensions of each are similar and 
 parallel. A circular plate, for example, of 18 inches diameter is, 
 with reference to its surface, not under a much greater inductive 
 influence in respect of its boundary than is a plate of half that 
 diameter in respect of its surface and boundary, the surrounding 
 influential bodies being in each case similarly circumstanced. The 
 difference, as is evident, consists merely, in a slight change of 
 distance from surrounding matter ; hence the induction arising 
 from the boundary extension does not materially affect the result. 
 In cases of hollow cylinders and globes, in which one of these 
 surfaces is shut out from external influences, only half the surface 
 may be considered as exposed to the inductive action of neutral 
 bodies. If, therefore, we suppose a square plate of any given 
 dimensions to be rolled up into an open hollow cylinder, by which 
 we shut out one of the surfaces from, the influences of external 
 bodies, the charge of the cylinder will be to the charge of the plate, 
 into which we may suppose it to be expanded, as vj. : v 2, that 
 is, as the square roots of the exposed surfaces, being as* 1 : 1'4 (28, 
 Exp. 17). In like manner, if we take a hollow globe and a cir- 
 cular plate of twice its diameter, in which the surfaces are the 
 same, surface for surface, the charge of the globe will be to the 
 charge of the plate as the square root of the one exposed surface 
 of the globe is to the square root of the two exposed surfaces of 
 the plate ; that is, as in the former case, as 1 : 1'4, as already ob- 
 served (5) relative to the ingenious researches of M. Coulombe. 
 This is the general relation of the charge of closed to the charge of 
 
250 ELECTBICAL FOUCE. 
 
 open surfaces. The charge of a circular plate is to the charge of 
 a square plate, whose side is equal to the diameter of the circular 
 plate, as 1 : T13. According to Cavendish, it is as 1 : 1-15, which 
 is not far different. It is not unworthy of remark that the rela- 
 tions of a square plate to a circular plate, as determined by 
 Cavendish nearly a century since, are in near accordance with 
 the formula we have already given for circles, squares, and similar 
 
 figures generally, C oc v/S (7). Since, on examining the relation 
 given by Cavendish, 1 : 1-15, we see that is very nearly the pro- 
 portion of the square roots of the surfaces of a circular and a square 
 plate whose diameter and side are the same; for example, the sur- 
 face of a circular plate of 10 inches diameter is 78 -54 square inches, 
 whilst the surface of the square plate of 10 inches side is 100 square 
 inches. Now, -v/78-54 : v/100 ' : 1 : 1*5; that is to say, we have 
 8*85 : 10 : : 1 : 1-15 very nearly. If we substitute the proportions 
 given by my experiments, we have 8*85 : 10 : : 1 : 1-13, which is 
 very nearly exact. Cavendish further states that the electrical rela- 
 tion of a circle whose diameter is equal to 1 is to that of a circle of 
 the same area as a square whose side is equal to 1 as 1 : .1*02 ; not a 
 very appreciable difference. I have, therefore, not been enabled 
 to detect any very sensible difference by direct experiment. 
 Cavendish further states that the relation of a circle to a globe of 
 the same diameter is as 12 : 18 - 5 ; that is, as 1 : 1*54. 
 
 My inquiries give the relative charges as 1 : 1'4. In fact, the 
 surface of a globe is four times one of its great circles ; hence the 
 two exposed surfaces of a circular disc of equal diameter are exactly 
 one- half of the surface of the sphere ; that is, the surfaces exposed 
 to inductive action in the disc and sphere are as 1 : 2, and there- 
 fore the charges under a given intensity are respectively as >/l : 
 1/2, as in other similar cases. The charge of a cube approaches 
 the charge of a globe whose diameter equals the edge of the cube, 
 but not exactly ; the charge of the globe being to the charge of 
 the cube as 1 : 1-2 nearly ; which is not far from the relation of a 
 circular plate to a square plate whose side is equal to the diameter 
 of the circle, as just observed. 
 
 12. On examining the intensity of circular plates, globes, and 
 closed and open surfaces generally, the charges being constant, I 
 found the intensity (2) to vary inversely as the exposed surfaces 
 (26). The intensity of a circular plate, for example, is only half 
 the intensity of a globe of the same diameter ; the exposed surfaces 
 being as 1 : 2 (26, Exp. 15). The intensity of circular planes 
 are in a simple inverse ratio of their surfaces ; thus, a circular 
 plane of 18 inches diameter has only one-fourth the intensity of a 
 
FOEMTJL^E FOR QUANTITY, ETC. 
 
 circular plane of 9 inches, or half that diameter, the surfaces being 
 here as 4 : 1. The intensities of globes are inversely as their sur- 
 faces. Thus, calling E the intensity and S the surface, we have 
 
 13. The following enumeration of formulae deducible from 
 these inquiries embrace the general laws of quantity, intensity, 
 surface, and boundary extension, and will be found practically 
 useful in deducing the laws of statical electrical force. 
 
 Symbols. Let C = electrical charge ; Q = quantity ; E === in- 
 tensity, or electrometer indication ; S = surface ; B = boundary 
 extension, or perimeter ; A = direct induction ; 3 = reflected 
 induction ; F = force ; D = distance. 
 
 FOKMUL^. 
 
 C x* S when S and B vary equally together . . . ..- (25) Exp. 10) 
 
 x Q, E being constant or equal ! . ,".- . . . *. (1) 
 
 C. * VS, B being constant or equal 1- '.. ". ...... . (25) (26) 
 
 C x VB, S being constant or equal ! . . . (25) 
 
 C x VSB when S and B vary together . ,.:. , . , (25) 
 
 E OB ~^D, Q being constant for all plane rectangular surface (8) 
 
 E cc , S being constant . V. . ,Y .. . (8) 
 B 
 
 E 7 i, B being constant . .. . ' . ",' . . (8) (26) 
 
 o 
 
 Ex , when S and B vary together . , ' , c (8) 
 
 E x Q2 ? S being constant, or equal ! ;' '. (21) 
 
 Ex^i . . . .... ;;'. ' . . V (9) 
 
 82 
 
 In square plates C x with side. of square . . . (11) 
 
 In circular plates C x with diameter . .. . (H) 
 
 In globes C x with diameter , . . .' . (11) 
 
 A, or induction x with S, all other things remaining j ^ 
 
 the same ; the same for , or reflected induction . j 
 In circular plates, globes, closed and open surfaces. 
 
 Ex- or as (12) (4) 
 
 S A 
 F or E x Q 2 (21) 
 
 F or S x -^-g, Q being constant (21) 
 
 Generally we have F - . . . (22) 
 
 JJ2 
 
 14. It is easy, from the laws of charge for circles and globes (11), 
 
252 
 
 ELECTRICAL FORCE. 
 
 to calculate a series of circular or globular measures of definite 
 values, taking the circular inch or globular inch as unity, and 
 calling, after Cavendish, a circular plate of an inch diameter 
 charged to saturation a circular inch of electricity ; or otherwise 
 charged to any degree short of saturation, a circular inch of elec- 
 tricity under a given intensity. In like manner a small globe of 
 an inch diameter may be designated as a globular inch of electri- 
 city. In the following table are given the quantities of electricity 
 in particles or units of charge, contained in circular plates and 
 globes, together with their respective intensities for diameters 
 varying from 0'25 to 2 inches. 
 
 In calculating this table, the circular inch that is, a circular 
 plate of an inch in diameter and one-fifth of an inch thick is 
 taken as unity, and supposed to contain 100 particles or units of 
 charge. 
 
 Diameters 
 or 
 Units of Charge. 
 
 Circle. 
 
 Globe. 
 
 Particles. 
 
 I .tensity. 
 
 Particles. 
 
 'I .tensity. 
 
 0-25 
 
 25 
 
 062 
 
 35 
 
 0-124 
 
 0-50 
 
 50 
 
 0-250 
 
 70 
 
 0-500 
 
 0-75 
 
 75 
 
 0-560 
 
 105 
 
 1-120 
 
 1-00 
 
 100 
 
 1-000 
 
 140 
 
 2-000 
 
 1-25 
 
 125 
 
 1-560 
 
 175 
 
 3-120 
 
 1-40 
 
 140 
 
 1 960 
 
 196 
 
 3-920 
 
 1-50 
 
 150 
 
 2-250 
 
 210 
 
 4-500 
 
 1-60 
 
 160 
 
 2-560 
 
 224 
 
 5-120 
 
 1-75 
 
 175 
 
 3-060 
 
 245 
 
 6-120 
 
 2-00 
 
 200 
 
 4-000 
 
 280 
 
 8-000 
 
 15. These elementary data being premised, I shall now en- 
 deavour to submit them to the rigid philosophy of experiment. 
 It will, however, be first requisite to explain the nature of the 
 processes resorted to, together with the instruments employed. 
 
 I must here crave permission, as being essential to the progress 
 of this inquiry, to enter upon a brief review of the Hydrostatic 
 Electrometer, recently perfected and improved. This instrument, 
 I am aware, has been before adverted to, in its primitive state, in 
 my paper in the Philosophical Transactions for 1839, plate 3. I 
 find it, however, requisite to particularly explain it under its 
 iccent and more perfect form. 
 
 [The author here describes the instrument, already noticed 
 at sufficient length in the previous pages, Figs. 86 to 89, 
 para. 147. But the drawings in the paper in the Royal 
 Society's Archives are even more minute and elaborate than 
 those given in the text just referred to. Should any inquirer 
 
APPARATUS EMPLOYED. 253 
 
 at some future time wish to construct the hydrostatic electro- 
 meter, it maybe interesting to him to consult the manuscript, 
 although, for all general purposes, the description given at 
 p. 122 is sufficiently ample. The instrument, as applied to 
 magnetic force, is described as the hydrostatic magnetometer 
 in "Rudimentary Magnetism/ 7 published in Weale's Series, 
 in 1850. ED.] 
 
 16. The nature of the electrometer as an instrument of measure 
 being fully apprehended, it remains now to notice certain instru- 
 ments of quantitative measure to be employed in connection with it. 
 First. A series of globular and plane circular transfer measures, 
 specified in the foregoing table (14), varying from 0*25 to 2 inches 
 in diameter, and being about one-fifth of an inch thick (14). 
 These globes and plates were carefully turned in a lathe, out of 
 very light wood, and smoothly gilded with extra thick gold 
 leaf. Each transfer measure has a loop of fine silk gut, firmly 
 secured to it, by which it may be easily taken up on a long 
 slender rod of varnished glass or vulcanite.* In circular plates it 
 is requisite to secure the suspension loop effectually in the follow- 
 ing way : A fine thread of silk gut being selected, a small open 
 loop is tied centrally in it ; small holes are drilled obliquely upwards 
 from the lower edge of the plate at each extremity of a diameter, 
 so as to extend from the edge of the plate to a little within its 
 surface : the extremities of the suspension loop are passed through 
 these holes in the surface towards the circumference, so as to bring 
 the loop in the centre. Being drawn tight, the thread of silk gut, 
 with the loop of suspension, is secured by means of fine wooden 
 pegs. The extremities of the pegs and ends of the gut are cut 
 oft by fine nippers, fair with the circumference, so as not to leave 
 the least projection. In globes, this loop of silk gut is easily 
 secured by drilling a hole in the globe, and inserting both ex- 
 tremities of the gut in it, after which the gut is firmly secured by 
 a small wooden peg. The vulcanite rod for suspension is slit open 
 by a thin saw-cut at one of its extremities, which is pointed, in 
 which slit the loop of suspension is held fast ; or the extremity of 
 the glass rod is a little bent up, by applying the heat of a spirit 
 lamp, to prevent the suspending loop from slipping off it. Thus 
 prepared, the given measure is charged to saturation, or to a given 
 intensity, by contact with the ball of a charged electrical jar. This 
 jar I term a " quantity jar," the construction of which it is impor- 
 tant to notice. 
 
 [The quantity jar is described at p. 119, Fig. 85, par. 141. 
 * See Fig. 85, page 119. ED. 
 
254 ELECTRICAL FORCE. 
 
 The author then goes on to describe the unit jar, which has 
 already been figured and described at p. 106, Fig. 75, 
 par. 120. These descriptions occupy paras. 17 and 18 in 
 the original paper. ED.] 
 
 The quantity jar being charged with a given number of measures 
 from the unit jar, and the unit jar, or the quantity jar itself, with- 
 drawn, a transfer measure, suppose a circular or globular inch, insu- 
 lated on its vulcanite rod (16), is brought into contact with the ball 
 of the jar, a communication being made at the same time with its outer 
 coatingj through the slip of tinfoil passing over the edge of the salver 
 on which it rests ;* the charged circular plate, or other measure, is 
 now transferred to the subject of experiment ; for example, an 
 insulated conducting sphere or plate, in connection with the elec- 
 trometer. This being effected, the transfer measure is returned 
 to the quantity jar, contact being made again with its outer coat- 
 ing : a second measure is now deposited on the plate, and so on ; 
 and thus, by a repetition of the operation, we may deposit on the 
 plate any number of circular or globular inches of electricity of a 
 given intensity we require. The error arising from the small 
 quantity of charge left upon the transfer measure after each de- 
 posit may in most cases be neglected, or otherwise calculated and 
 avoided altogether. In the example just given, suppose the re- 
 lative diameters of the transfer measure and the plate to be as 
 1:12; in that case their respective charges would, by the preceding 
 data (14, Table 1), be twelve times as great upon the large plate 
 as upon the small one. To put in evidence how small a quantity 
 is in this case left upon the transfer plate, after contact with the 
 plate, we may observe, as just stated, that the charge of which the 
 larger plate is susceptible is at least twelve times that of the smaller 
 plate ; and therefore, after contact, the charge of the smaller disc 
 will be shared between the two ; the larger one consequently taking 
 twelve-thirteenths of the whole charge, leaving only a residue of 
 one- thirteenth part of the transfer charge on the circular inch ; 
 which, according to the preceding table (14), would be considerably 
 less than could be sustained by a disc of one-tenth of an inch diameter. 
 There will of course be a slight increase after each transfer, but 
 for the few transfers generally employed, the error arising there- 
 from would be of no moment. It is easy to estimate the amount 
 of the total deficient quantity, by placing a hollow sphere having 
 an open mouth in connection with the electrometer, and after each 
 contact of the transfer measure with the given insulated conduct- 
 ing plate, deposit the transfer measure within the hollow sphere. 
 * This is seen at N, Fig. 85. ED. 
 
APPARATUS EMPLOYED. 255 
 
 We know in this case it will be deprived of all its remaining elec- 
 tricity, and may be withdrawn in a perfectly neutral state : the 
 electrometer will consequently indicate the total sensible quantity 
 of electricity which the small transfer measure has carried away after 
 any given number of transfers to the plate, the distance between 
 the electrometer discs remaining constant. This kind of experiment 
 will be quite sufficient for estimating the greatest error which 
 could possibly arise from the small quantity carried off by the 
 transfer measure, after any given number of deposits. 
 
 19. These mechanical conditions being fully apprehended, the 
 experimental manipulation may next be discussed. 
 
 [For this ( which occupies paras. 20 and 21 in the original 
 paper) we may refer to Part I., par. 120, pages 150-152. 
 For the details of experiments made to illustrate the law of 
 electrical attractive force, as regards quantity, to par. 152, 
 Exp. 38 ; and as regards distance, to Exp. 39. ED.] 
 
 22. Similar results ensue when, instead of connecting the elec- 
 trometer with a simple conducting surface charged by transfer 
 measures in the way just explained (18), we connect it with an 
 electrical jar, charged through a unit measure from the electrical 
 machine. The force between the attracting plates will be in this 
 case also as the squares of the quantities directly, and as the 
 
 Q 2 
 squares of the distances inversely, according to the formula ^ 
 
 (13), and these laws are rigidly exact, and are obtainable with the 
 greatest ease and precision. Experiments in this way in connection 
 with the electrical jar are rather more certain and manageable 
 than with simple and open exposed surfaces ; the electrical charge 
 being more readily retained by the jar than by a simple surface. 
 The index of the electrometer, under the operation of the jar, 
 remains unmoved for a considerable time, evincing little or no 
 dissipation of charge. This is very important in the employment 
 of every instrument of quantitative measure, whether of attraction 
 or repulsion, such as the instrument in question, or the " Torsion" 
 or " Bifilar " balances ; and we must be especially cautious to per- 
 fect the insulations, and in the case of simple surfaces not to employ 
 quantities of electricity which, under a given insulating state 
 of the air, cannot be retained on them without dissipation, at 
 least for a longer time than we require for the experiment. The 
 best method of perfecting the insulations is to dry them off before 
 making the experiment, with a curved iron heated to redness.* 
 
 23. Since in the ordinary application of the electrometer we 
 
 * See Fig. 103, page 158. ED. 
 
256 EtECTRICAL FORCE, 
 
 connect the attracting plate p with a given surface, p,* it is 
 requisite to further ascertain what differences may possibly arise 
 in consequence of the extension of the surface under exami- 
 nation, as well as the influence of the reflected induction from 
 the suspended neutral plate N. We may conclude, however, that 
 in cases in which the plates p n are small in proportion to the 
 charged surface, no sensible error will arise in such cases. If, 
 for example, we add to a circular plate, P, of 1 foot in diameter, a 
 small plate, p, then the total surface will be the surface of the plate 
 p plus the surface of the attracting plate P ; but as the surface of p 
 is very small in proportion to P, the addition of p, together with 
 the reflected induction of the suspended neutral plate n, may be in 
 most cases neglected. The reflected induction of the suspended 
 disc n, however, must, as is evident, tend to diminish the intensity 
 of the charged surface under examination, and enable it to sus- 
 tain a somewhat greater charge than it could otherwise, but as 
 this in the ordinary course of experiment may be considered as 
 common to every case, it cannot introduce any considerable error 
 in our calculation ; moreover, the result of the induction must be 
 necessarily small. Suppose, for example, the circular plate p in 
 connection with the electrometer were under the influence of a 
 similar neutral plane of equal diameter placed at a very short 
 distance from it, with a resisting dielectric between, after the 
 manner of a condenser. Even in this case we should fail to 
 neutralise all the electrometer indication, and this influence of one 
 plate upon the other would decrease rapidly in some inverse ratio 
 of the distance between the plates, and in some direct ratio of 
 the area of the condensing plate. Now, as the area of the sus- 
 pended electrometer disc would in any case bear a small proportion 
 to the surface under experiment, the intensity of the charged 
 surface would be but little changed by the influence of the small 
 neutral electrometer disc n ; for example, a disc of only l inch in 
 diameter, and whose area does not greatly exceed T76 square inch, 
 operating by induction at a distance of half an inch, or perhaps at 
 double that distance, on a circular plate, p, of eight times its dia- 
 meter, exposing a surface of 113 square inches (at least sixty-four 
 times the surface of the small plate), could not increase the charge 
 of such a plate by any very appreciable quantity. If, for example, 
 we communicate 5 circular inches of electricity to an insulated 
 plate, which we may call P, of 9 inches diameter, apart from the elec- 
 trometer, and then connect it with the attracting plate p of the 
 electrometer, the intensity is the same, or very nearly so, as when 
 
 * See Fig. 90. ED, 
 
ACCURACY OF THE AUTHOR'S ELECTROMETER. 257 
 
 charged in connection with, the attracting plate, p, under the 
 influence of the suspended disc n. Nevertheless cases may possibly 
 occur in which it might be desirable to estimate the increased 
 charge liable to arise from the reflected induction of n. Suppose, 
 for example, the augmented charge were less in proportion to a 
 large surface than to a small one, then it might perhaps be 
 necessary in certain cases to take this into calculation. JSTow 
 the reflected induction of the suspended disc is constant, whatever 
 be the extent of the charged surface, it being limited to the area 
 of the suspended disc. Imagine, by way of illustration, the 
 reflected induction of the suspended electrometer disc, n, on a cir- 
 cular plate, P, of 9 inches diameter, to be such as to enable the 9-inch 
 plate to receive under the same intensity (2) one-half a measure 
 more upon five measures than it could if no such auxiliary induction 
 were present, and that a circular plate of 18 inches, or double that 
 diameter, with a double charge or 10 circular inches, receives 
 under the same influence only a quarter of a measure more ; cases, 
 therefore, might arise in which the augmented charge of a large 
 plate consequent upon the electrometer induction, may at last 
 become extremely small in regard to the augmented charge of a 
 small plate, which would in such a case interfere with an exact 
 result ; for since the half measure bears a much greater proportion 
 to the charge of the 9-inch plate, the charge of the 9-inch plate 
 would not be actually as great in proportion to that of the 18-inch 
 plate as it appears : still it is not difficult to estimate this correction. 
 When, however, the approximations are so close as to involve such 
 small differences, we can have but little doubt of the law we seek. 
 24. The following experiments may be adduced in evidence of 
 the small influence of the operation of the instrument on the 
 accuracy of its indication. 
 
 Exp. 3. Withdraw the slider M,* together with the fixed 
 electrometer plate p, and let the suspended plate n hang imme- 
 diately over an insulated circular plate, P, of 1 foot in diameter, 
 insulated on one of the travelling stages, a, the two plates, p, n, 
 being accurately adjusted to a given distance, suppose half an inch. 
 Charge plate p with a given quantity, suppose 8 circular inches, 
 so as to produce an attractive force of say 10 degrees, then by a 
 movement of the electrometer on the rails, or by a movement of 
 the travelling stage, a, supporting P, bring the suspended plate n 
 over different points of p in order to observe if the electrometer is 
 differently affected by different points of the charged surface. 
 
 * In the description of these experiments, the letters refer to such figures 
 as Figs. 86 to 89, 90, and 91, in Part I. ED. 
 
 S 
 
258 ELECTRICAL FORCE. 
 
 Result of tjiis Experiment. The force appeared invariable ; it is 
 everywhere the same, being 10 at the distance of five-tenths. 
 
 Exp. 4. Discharge and withdraw plate p. Eeplace slider M 
 and attracting electrometer plate p. Connect the plate p with 
 the electrometer plate p in the usual way, and let the dis- 
 tance between p n be set accurately to five-tenths, as in the 
 former case. Charge the plate p with 8 circular inches as before. 
 Result. The indicated force is the same, that is, 10, showing 
 that the addition of p does not in this case sensibly interfere with 
 the electrometer indication, and that it is of no consequence what 
 part of the combined surfaces is opposed to the suspended plate n 
 of the electrometer. 
 
 Exp. 5. Repeat the last experiments 3 and 4, substituting 
 for the circular plate p a long rectangular plate, R, of copper, 
 or any other conductor, which may be from 2 to 3 feet long, and 
 from 2 to 3 inches wide. It is requisite for this experiment, to place 
 the rectangle upon strong insulating supports, D, before alluded to, 
 (15 Section K), on the platform at the side of the rail, and turn the 
 electrometer a little aside upon its pillar in order to admit of the 
 suspended plate n being immediately over the surface of the 
 rectangle; the suspended disc n may be thus easily moved 
 along over the surface of the rectangular plate by the move- 
 ment of the electrometer upon the rails. The result in this 
 case is the same as in the last experiment ; the force is in- 
 variable throughout the whole length of the rectangle, the 
 distance of the attracting surfaces being everywhere the same ; 
 and this is observed to be the case at any distance, however small, 
 at which we can make the experiment, a result perhaps not to 
 be anticipated, but nevertheless in strict accordance with ordi- 
 nary electroscopic indications ; thus, at whatever point of a charged 
 surface any ordinary electroscope of repulsion be placed, its regular 
 divergence is the same. 
 
 Exp. 6. Let the attracting plate p and the suspended disc n be 
 now replaced in position as before, and the rectangle R connected 
 directly with the electrometer as in ordinary cases. Adjust 
 the distance of the plates p n to five- tenths as in the last case ; 
 charge the rectangle R with the same quantity : here the result is 
 again the same as when the electrometer disc n hung immediately 
 over R, being still 10, so far showing, as in the preceding case, 
 Exp. 4, that the additional surface p has little or no influence 
 on the result ; and that it is of little or no consequence what part 
 of the combined surfaces operate on the electrometer. It may, 
 however, be well to remark, that in any case of the fixed plate 
 
LAWS OF SURFACE AND BOUNDARY. 259 
 
 p of the electrometer bearing a large proportion of the surface 
 under experiment, the suspended plate n of the electrometer 
 should be immediately over the surface itself, that is, the surface 
 under experiment should be substituted for the fixed plate p. 
 
 25. Laws of surface and boundary as regards rectangular 
 plane surfaces. Verification of the formula C = >/SB (13) and 
 
 In order to investigate the charges of plane rectangular surfaces 
 together with the verification of the formula C = \/ SB (13), a 
 series of smoothly polished plates of copper were employed, vary- 
 ing from 10 inches square to 40 inches in length, by from 2^ to 6 
 inches wide, and about one- eighth of an inch thick, exposing from 
 100 to 200 square inches of surface. The charges of these plates 
 were carefully determined, the attracting plates of the electrometer 
 being at a given distance, suppose half an inch. 
 
 E'jcp. 7. Comparison of a copper plate, A, Fig. 116, 10 inches square, 
 with a rectangular plate, B, 40 inches long by 2'5 wide, the thickness 
 
 Fig. 116. 
 
 of the plates being the same. In these plates we observe that the 
 surfaces are each 100 square inches, whilst the boundaries are 
 40 and 85 inches, so that the boundaries may be taken as 1 : 2, 
 without any very sensible error, the difference not being of any 
 practical moment to the experiment. 
 
 On connecting these plates with the electrometer successively, 
 the charge of the square plate A was found to be 7 circular inches 
 under an intensity of 8. The charge of the rectangular plate B 
 under the same intensity of 8 amounted to 10 circular inches 
 nearly ; by which we perceive, that the surface being the same, 
 the charges were 7 and 10, whilst the boundaries are as 1 : 2. The 
 charges, therefore, are as 1 : 1*4, or as \/ 1 : A/ 2 that is, as 
 the square roots of the boundaries. 
 
 Charge as determined ~by Intensity (2). Seven measures deposited 
 upon the square plate A had an intensity of 8, as in the last ex- 
 periment ; the same seven measures deposited upon the rectangle B 
 had an intensity of only 4, or one-half the former intensity, being 
 
 inversely as the boundaries, according to the formula E =^r (13). 
 
 s2 
 
260 ELECTRICAL FORCE. 
 
 When about three additional measures were added to the rectangle, 
 the intensity was increased to 8, or double the former intensity, 
 the intensity being as the square of the quantity (21, Exp. 1) ; so 
 that the relative charges as thus determined were 7 and 10 nearly, 
 or as 1 : 1-4, as in the former case. 
 
 Exp. 8. Rectangular plate c, Fig. 117, 37 -5 inches long by 2'7 
 inches wide, surface 101 inches, boundary 80 '5 inches ; compared 
 with a rectangular plate D, 34*25 inches long by 6 inches wide, sur- 
 face 205 square inches, boundary 80*5 inches. Here the boundaries 
 are the same, whilst the surfaces may be taken as 1 : 2 nearly. 
 
 These plates being connected with the electrometer, the charge 
 of the plate c, surface 101, was 8 -5 circular inches under an inten- 
 
 sity of 8 ; the charge of the plate D, surface 205, was found to be 
 12 circular inches nearly, under the same intensity of 8 that is 
 to say, whilst the surfaces are as 1 : 2, the charges are as 8 '5 : 12 ; 
 that is as 1 : \/ 2 nearly, or as the square roots of the surfaces. 
 Charge as determined by Intensity. 8*5 measures deposited on the 
 rectangle c, surface 101, evinced an intensity of 8, as already 
 observed. The same 8*5 measure on the rectangle D, surface 205, 
 evinced an intensity of 4 C only, or one-half the former, being 
 
 1 
 
 inversely as the surfaces ; according to the formulae E * -~ (13) ; 
 
 when 3*5 measures, or a little more were added, the intensity 
 became twice as great, or 8, the intensity increasing as the square 
 of the quantity (21, Exp. 1). The relative quantities or charges, 
 therefore, as thus determined, for the same intensity (= 8), 
 were 8'5 and 12 circular inches nearly, being as 1 : 1*4; that is, 
 as 1 : \/ 2, or as the square roots of the surfaces, the surfaces 
 .being in the ratio of 1 : 2. 
 
 Exp. 9. Comparison of a rectangular plate, E, Fig. 118, 26'25 
 inches long by 4 inches wide, surface 105, boundary 60"5, with a 
 rectangular plate r, 40 inches long by 5 inches wide, surface 200, 
 boundary 90. Here the surfaces are as 1 : 2 very nearly, whilst 
 the boundaries are as 2 : 3. 
 
 On examining the charges of these plates, the charge of the 
 rectangular plate E, surface 105, boundary 60 -5, was 7 circular 
 
CHARGE AS DETERMINED BY INTENSITY. 
 
 261 
 
 inches under an intensity of 10, and the charge of the rectangular 
 plate F, surface 200 inches, boundary 90 inches, was 12 circular 
 inches, nearly under the same intensity of 10. The charges here 
 
 Big. 118. 
 
 are as 7 : 12 nearly, or as 1 : 1*7, being as the square roots of the 
 surfaces multiplied into the square roots of the boundaries of the 
 two rectangles, very nearly. 
 
 Charge as determined by Intensity. Seven circular inches on 
 the rectangle E evinced an intensity of 10, as just stated. The 
 same 7 circular inches on the rectangle F gave an intensity of 
 something more than 3 '5, being inversely as the surfaces multi- 
 plied into the boundaries according to the formula E = (13). 
 
 When 5 circular inches more were added, the intensity became 
 10 as before, being as the square of the quantity (20, Exp. 1). 
 The relative charges, therefore, as thus determined for an intensity 
 of 10, were 7 and 12 nearly ; that is to say, as the square roots 
 of the surfaces multiplied into the square roots of the boundaries, 
 as just seen. 
 
 Exp. 10. Plate A, Fig. 119, 10 inches square, surface 100 inches, 
 
 K 
 
 LIBRA i.- > 
 
 UNIVERSITY OF 
 
 CALIFOUXLv. 
 
 Fig. 119. 
 
 boundary 40, compared with a rectangular plate, F, 40 inches long by 
 5 wide, surface 200 inches, boundary 90. Here the surfaces are 
 
262 ELECTRICAL FOKCE. 
 
 double each other, and the boundaries nearly double each other, 
 but not exactly ; we may, however, take them as double ; for 
 since the charge varies as the square root of the boundary, Exp. 7, 
 a difference of a few inches in the boundary would not introduce 
 any very sensible difference in the experiment ; so that in taking 
 the boundaries as 40 and 80, instead of 40 and 90, we do so 
 without any sensible error. 
 
 On examining the charges of these plates, the charge of the 
 square plate A, surface 100, was found to be 6 circular inches 
 under an intensity of 10 ; the charge of the rectangular plate r, 
 surface 200, was 12 circular inches nearly under the same intensity 
 of 10, or double the charge of the former ; so that taking the 
 boundaries as 40 and 80, the charges are as the square root of the 
 surfaces and boundaries conjointly, according to the formula 
 C = \/ 8 .B (13), as just determined, Exp. 9. A double surface, 
 therefore, having a double boundary, takes a double charge, but 
 not otherwise ; neglecting all considerations of the boundary in 
 each, the surfaces and boundaries varying equally together, the 
 charge will vary as either directly, and therefore may be said to 
 vary with the surface. 
 
 Charges as determined by Intensity.- Six circular inches on the 
 square plate A had an intensity of 10 Q , as already stated ; the same 
 6 circular inches on the rectangle F of double surface and double 
 boundary, gave an intensity of 2*5 only, or one-fourth the former in- 
 tensity, being inversely as the boundaries multiplied into the surfaces ; 
 or the surface and boundary varying equally together, inversely 
 
 as the square of the surface, according to the formula E = ^ (L3). 
 
 When another 6 circular inches were added, or twice the former 
 quantity, the intensity 2'5 of the rectangle became 10, being as 
 the square of the quantity (21, Exp. 1). 
 
 The relative charges, therefore, as thus determined, for an in- 
 tensity of 10, were 6 and 12, being as 1 : 2 ; that is, as the square 
 roots of the boundaries, multiplied into the square roots of the 
 surfaces of the two rectangles ; or neglecting all considerations of 
 the boundaries, directly as the surfaces, according to the formula 
 C x S (13). 
 
 A variety of similar experiments, which, to avoid tedious detail, 
 I omit, confirmed these general laws of electrical force, in all cases 
 of rectangular plates whose boundaries and surfaces involve con- 
 siderable difference of extension and exposure to external inductive 
 influence. 
 
CHARGE AS DETERMINE!} BY 1NTE:N 7 SITY. 263 
 
 Charge and intensity of square plates, circular plates, spheres, 
 and closed and open surfaces. Verification of the formula; 
 
 26. Exp. 11. A copper plate, A (Fig. 119), 10 inches square, sur- 
 face 100, boundary 40, compared with a copper plate, K, 14 inches 
 square, surface 196, boundary 56.- Here the surfaces are as 
 100 : 196, that is, as 1 : 2 nearly, whilst the boundaries are as 40 
 and 56, hat is, as 1 : \/ 2 nearly. 
 
 On examining the charges of these plates, the charge of the 
 plate A, surface 100, was 8 circular inches under an intensity of 
 10. The charge of the plate K, surface 196, was 11 circular 
 inches under the same intensity of 10. Here the charges are as 
 8 : 11, whilst the surfaces may be taken as 1 : 2 ; that is to say, 
 the charges are as the square roots of the surfaces, neglecting the 
 difference of boundary (11), or as 1 : 1-4. 
 
 Charge as determined by Intensity. Eight circular inches upon 
 the plate A, 10 inches square, surface 100, evinced an intensity of 
 10 Q , as already stated. The same 8 circular inches upon the plate, 14 
 inches square, surface 196, evinced an intensity of 5 only, or one- 
 half the former intensity, being as the surface inversely, according to 
 
 the formula E = - (13). When another 3 circular inches were 
 
 added to the plate K, the intensity became 10, as in the plate A, 
 being as the square of the quantity (21, Exp. 1). The relative 
 quantities, or charges, therefore, under an intensity of 10, as thus 
 determined, were 8 and 11, being as 1 : 1*4, as before ; that is, as 
 the square roots of the surfaces nearly. 
 
 Exp. 12. A circular plate, which we will call M, of 9 inches in dia- 
 meter, surface 63 '6 square inches, compared with a circular plate N, 
 of 18 inches, or double that diameter, surface 254. Here the sur- 
 faces are as 1 : 4, being as the square of their diameters, whilst the 
 circumferences or boundaries are as 1 : 2. 
 
 The charge of the 9-inch plate M was found to be 6 circular 
 inches, under an intensity of 10, the charge of the 18-inch plate N, 
 12 circular inches, under the same intensity of 10. Here the 
 charges are as 1 : 2, whilst the surfaces are as 1 : 4. The charges, 
 therefore, neglecting the difference of circumference or boundary 
 (11), are as the square roots of the surfaces, according to the 
 formula (13) C = */ S, B being neglected. 
 
 Charge as determined by intensity. Six circular inches upon the 
 9-inch plate M evinced an intensity of 10, as already stated. The 
 
264 ELECTRICAL FORCE. 
 
 same quantity deposited upon the 18-inch plate N, evinced an inten- 
 sity of only 2 '5, being inversely as the surfaces, according to the 
 
 formula E = -^- (13). Another 6 circular inches being added to the 
 
 18-inch plate, the intensity became 10, as at first, being as the 
 square of the quantity. Here, as in the last experiment, the 
 quantities or charges are as 1 : 2, whilst the surfaces are as 1 : 4. 
 The charges, therefore, are as the square roots of the surfaces. 
 
 Exp. 13. A circular plate, M, of 9 inches in diameter, 
 surface 63'6, compared with a circular plate o, of 12'72 inches in 
 diameter, surface 127'2 square inches. Here the surfaces are as 
 1 : 2, whilst the boundaries are as 3 : 4, nearly. On examining 
 the charges of these plates, the charge of the 9-inch plate, M, was 
 5 circular inches, under an intensity of 8. The charge of the 
 12.72-inch plate, o, was 7 circular inches under the same intensity 
 of 8. The charges here are as 5 : 7, whilst the surfaces are as 
 1:2; the charges, therefore, are, neglecting the difference of cir- 
 cumference or boundary (11), as the square roots of the surfaces, 
 being as I : 1*4, as before. 
 
 Charge as determined by Intensity. Five circular inches on the 
 9-inch plate M evinced an intensity of 8. The same 5 cir- 
 cular inches deposited on the plate o, 12'72 diameter, evinced an 
 intensity of 4 only, or one-half the former, being inversely 
 as the surfaces ; when 2 more circular inches were added, the 
 intensity of the plate o became 8, being as the square of the 
 quantity (21, Exp. 1). The relative charges or quantities, 
 therefore, under the same intensity (= 8), were 5 and 7; the 
 surfaces being as 1 : 2. The charges, therefore, are as the square 
 roots of the surfaces, as in the former case ; that is, as 1 : 1 *4. 
 
 Exp. 14. Comparison of a sphere p, Fig. 31, of 4'5 inches 
 diameter, surface 63 '5, with a sphere R of 9 inches, or double 
 that diameter, and surface 254 square inches. Here the sur- 
 faces are as 1 : 4, whilst the circumferences or boundaries are as 
 1 : 2. 
 
 On examining the charges of these spheres, the charge of the 
 sphere p was found to be 4 circular inches under an intensity of 
 9. The charge of the sphere R, 8 circular inches under the 
 same intensity of 9. Here the charges, neglecting the dif- 
 ference of boundary (11), are as 4 : 8 ; that is, as 1 : 2, the surfaces 
 being as 1 : 4. The charges, therefore, are as the square roots of 
 the surfaces, being as 1 : V 4. 
 
 Charge as determined by Intensity. Four circular inches, de- 
 posited upon the sphere P, surface 63*5, evinced an intensity of 
 
CHARGE AS DETERMINED BY INTENSITY. 265 
 
 9. The same quantity of 4 circular inches deposited upon the 
 sphere R, surface 254, evinced an intensity of 2-5 only, being 
 inversely as the surfaces, very nearly. Another 4 circular inches 
 being added, the intensity of the sphere R became also 9. The 
 quantities, therefore, under the same intensity, were as 1 : 2, that 
 is, as the square roots of the surfaces. 
 
 Exp. 15. A circular plate M, of 9 inches diameter, compared with 
 a sphere, R, of the same diameter. Here the surfaces are respec- 
 tively, for the plate and sphere, 63'6 square inches, and 254 square 
 inches, being as 1:4; whilst the boundaries are the same. 
 
 On examining the charges of these surfaces, the charge of the 
 plate M was 8 circular inches under an intensity of 12. The 
 charge of the sphere R was 11 circular inches under the same 
 intensity of 12. 
 
 We have here to observe, that one surface of the sphere is 
 closed or shut up ; consequently, the two exposed surfaces of the 
 circular disc are exactly one half the exposed surface of the sphere. 
 The charges, therefore, being as 8 : 11, or as 1 : 1'4, are as the 
 square roots of the exposed surfaces. 
 
 Charge as determined by Intensity. Eight circular inches on the 
 plate M had, as just noted, an intensity of 12. The same 8 circular 
 inches upon the sphere R evinced an intensity of 6, being in- 
 versely as the surface, according to the formula E = -^- (13). 
 
 When 3 circular inches or a little more were added, the intensity 
 of the sphere became 12, or double the former, being as the 
 square of the quantity (21, Exp. 1). The charges, therefore, 
 as thus determined, for an intensity of 12, were as 8 : 11, or 
 as 1 : 1*4, being as the square roots of the exposed surfaces, as 
 before. 
 
 Exp. 16. Comparison of a sphere s of 7 inches diameter, with a 
 circular plate, T, of 14 inches, or double its diameter. In this case 
 it is to be observed that the surfaces of the sphere and plate are 
 actually the same ; that is to say, the inner and outer surfaces of 
 the sphere will be equal to the two surfaces of the plate. It is, 
 however, to be observed, as in the last experiment, one surface of 
 the sphere is closed or shut up. The surfaces of the sphere and 
 plate, therefore, electrically considered, are not equal; the plate 
 having twice the exposed surface of the sphere. The surfaces open 
 to induction, therefore, are as 1 : 2. 
 
 On examining the charges of these surfaces, the charge of the 
 sphere s was 10 circular inches under an intensity of 20. The 
 charge of the plate T was 14 circular inches, under the same 
 
266 ELECTRICAL FORCE. 
 
 intensity of 20. The charges here are as 10 : 14, or as 1 : 1-4. 
 The charge of the sphere, therefore, as compared with the charge 
 of the plate, is as 1:^2; that is, as the square roots of the 
 exposed surfaces. 
 
 Charge as determined by Intensity. Ten circular inches deposited 
 upon the sphere, evinced an intensity of 20. The same 10 
 circular inches upon the plate evinced an intensity of 10, being 
 inversely as the exposed surfaces, according to the formula 
 
 E = -r- (13). An additional 4 circular inches raised the intensity 
 
 of the plate to 20, being as the square of the quantity very 
 nearly, so that the relative quantities or charges under the same 
 intensity = 20, were 10 and 14 ; being as the square roots of 
 the exposed surfaces. 
 
 27. This experiment has immediate reference to the beautiful 
 and ingenious investigation of Coulombe, before referred to, re- 
 lative to the electrical capacity of a sphere and circular plate of 
 twice its diameter, and from which it has been inferred that the 
 capacity of the plate is twice that of the sphere ; Coulombe 
 having shown that when the plate in a neutral state and insulated, 
 was made to touch the sphere, also insulated, but charged with 
 electricity, the plate took away from the sphere two-thirds of the 
 accumulation, leaving only one-third with the sphere. It is to be 
 here again especially observed, that both surfaces of the plate are 
 open to external inductive action, whilst one surface of the sphere 
 is closed, and not so exposed. Consequently, the inductive in- 
 fluence of external bodies upon the two exposed surfaces of the 
 plate, is double that of the same influence upon the one exposed 
 surface of the sphere ; and thus, in the sharing of the charge, the 
 plate is in a condition to take from the sphere a greater quantity 
 under the same electrometer indication. The plate no doubt acts 
 inductively from both its surfaces, as compared with the sphere, 
 but this is not, however, electrical charge (1), as applied to a given 
 surface. 
 
 In order to estimate correctly the relative capacities of a globe 
 and a circular plate of twice the diameter of the globe, it is neces- 
 sary to refer each to a third body, as a standard of comparison ; 
 for example, to another sphere which may be of equal diameter 
 to the given sphere. Let, therefore, this third body, or trial 
 sphere, be carefully insulated; bring the charged sphere, also 
 carefully insulated, into contact with it. Then, because the two 
 spheres are equal and similar, the accumulation (as admitted by 
 Coulombe) will be shared equally between them, and the neutral 
 
COMPARISON OF PLATE AND CYLINDER. 267 
 
 sphere will have taken away one-half the electricity from the 
 charged sphere, so that the capacity of the charged sphere relative 
 to the trial sphere will be represented by the fraction J- 
 
 Let this experiment be now repeated, substituting the neutral 
 plate, carefully insulated, for the trial sphere. In this case, 
 according to Coulombe, the plate takes from the sphere two-thirds 
 of its electricity. The capacity of the plate, therefore, as referred 
 to the trial sphere, is represented by the fraction ; so that the 
 relative capacities of the sphere and the plate will be as J : f, or as 
 3:4; that is, as 1 : ^/2 nearly, as just shown, Exp. 16. Hence, 
 if three measures of electricity deposited on the sphere produce an 
 electrometer indication of 10, then four measures deposited on the 
 plate would give the same intensity of 10, or very nearly, as is 
 demonstrable by experiment. Now, if the capacity of the plate 
 were double that of the sphere, the relative quantities should have 
 been 5 and 10, instead of 3 and 4, which is not the case, as we 
 have just shown. We see, therefore, that the plate has really 
 not twice the capacity of the sphere ; and that it is not possible 
 to place upon the plate twice the quantity of electricity under 
 the same intensity. The conclusion, therefore, arrived at by 
 Coulombe relative to the double action of the plate is only exact 
 for the relative surfaces of indication. 
 
 28. The following experiment may be further adduced in support 
 of the preceding. 
 
 Exp. 17. A copper plate, which we will call A, 10 inches square, 
 compared with the same plate, rolled up into an open cylinder, u, 
 10 inches long by 3 -2 inches in diameter. Here, as in the last case, 
 the surfaces are actually the same ; but the plate having twice the 
 exposed surface of the cylinder, the surfaces of the plate and of the 
 cylinder, electrically considered, are as 1 : 2, as in the case of the 
 globe and the plate. On examining the charges of these surfaces, 
 the charge of the plate was 9 '76 circular inches under an intensity 
 of 11. The charge of the cylinder 7 circular inches, under the 
 same intensity of 11. The charges, therefore, being as 7 : 10 nearly, 
 or as 1 : 1.4, whilst the surfaces are as 1 : 2, we have the charge of 
 the cylinder to the charge of the plate, into which it may be ex- 
 panded, as 1 : \/ 2 (11) ; that is to say, as the square roots of the 
 surfaces of induction. In determining the charge by intensity, we 
 have seen that 7 circular inches upon the cylinder gave an inten- 
 sity of 11 ; while the same quantity upon the plate gave an 
 intensity of 5J, being inversely as the exposed surfaces. When 
 about 3 circular inches were added to the plate, the intensity 
 advanced to 11. The quantities or charges, therefore, under the 
 
268 ELECTRICAL FORCE. 
 
 same intensity, were as 7 : 10 nearly, as before, or as 1 : \/ 2, being 
 a general law of charge for closed and open surfaces. 
 
 Exp. 18. A hollow copper cube having the edge = 5-7 inches, 
 the surface 195 inches, compared with a hollow copper sphere, with 
 diameter edge of the cube, surface 103 square inches nearly. On 
 examining the charges of these surfaces, the charge of the sphere 
 was 9 circular inches, under an intensity of 10. The charge of 
 the cube 10 circular inches, under the same intensity of 10. The 
 relative charges of the sphere and of the cube, therefore, may be con- 
 sidered as approaching each other. We may, therefore, consider 
 the charge of a cube as approaching that of a sphere of the same 
 diameter as the edge of the cube, notwithstanding the difference of 
 the surfaces (11), owing to the six surfaces of the cube not being 
 in a separated or disjointed state (9). 
 
 29. I have to observe here, in conclusion, that the numerical 
 results of the foregoing experiments, although not mathematically 
 exact in every instance, yet were, upon the whole, so nearly accordant 
 as to leave no doubt upon the mind as to the law in operation. In 
 very delicate experiments of this kind it is next to impossible to 
 determine forces to within a nearer approximation than that of a 
 degree of the electrometer, or to within quantities less than that of 
 25 of a circular inch. In cases in which the attractive forces are 
 considerable, it will be sufficient if the approximations are within a 
 degree or two, or within one measure of quantity. Nevertheless, 
 I found the numerical results generally exact. If the manipulation 
 be skilfully conducted, and the electrical insulations duly provided 
 for (25), it is quite astonishing to find how rigidly exact the nume- 
 rical results come out. 
 
4TMO SPHERIC ELECT El CITY 
 
 AND 
 
 PROTECTION PROM LIGHTNING, 
 
 [These Lectures were, prepared expressly for the Artillery and Engineers 
 at Woolwich and Chatham, and were to have been delivered in September, 
 1861. The apparatus was packed and everything prepared, when just before 
 the author's intended departure, he was seized with an attack of acute 
 inflammation in the eyes, the commencement of more than a year and a 
 half of much suffering. ED.] 
 
FIRST LECTURE. 
 
 GENTLEMEN, 
 
 I have the honour of addressing you under the sanction of 
 the Hight Hon. the Secretary of State for War on a scientific 
 subject of great physical and practical public interest ; the ques- 
 tion of the security of ships and buildings and other elevated 
 structures on the earth's surface, against the terrible agency of 
 lightning ; a question which, as you may possibly have already 
 learned, has been, ever since the middle of the last century, when 
 the American and French philosophers identified the agency of 
 lightning with that of our ordinary electrical apparatus, a very 
 vexed question and remains even now at this time not altogether 
 set at rest. The immediate practical result of the new discovery 
 of ] 750 was a proposal by the celebrated American philosopher, 
 Franklin, to lead metallic rods or wires along the sides of towers 
 and down the rigging of ships, which being found to transmit 
 easily the matter of lightning, he assumed would parry its ap- 
 palling thrust, and so avoid damage to the structure to which 
 they were applied. These lightning rods of Franklin excited, 
 however, very severe, and, as you will presently see, very absurd 
 controversy. The laws of electrical force and discharge not being 
 at that time well understood ; moreover, the tendency of lightning 
 to settle on pointed metallic bodies, as on spires and ships' masts, 
 on the spears of soldiers, and such like, presenting to the spectator 
 so many luminous points, led to the notion that these bodies 
 attracted the matter of lightning, and that hence the lightning 
 rods of Franklin were to be rather dreaded as a source of danger, 
 than hailed as a means of protection. On the other hand, Frank- 
 lin's hypothesis assumed that the influence of the rod was such as 
 to charm to rest the otherwise unruly spirit of the thunder- 
 storm, and by attracting to itself the matter of lightning, discharge 
 it silently or in harmless corruscations into the earth. Upon this 
 assumption there arose much difference of opinion relative to the 
 
272 FIRST LECTURE. 
 
 amount of security the lightning rod could ensure, as compared 
 with the quantity of lightning which its active influence might 
 draw down upon the structure to which it was applied. 
 
 You will, of course, see, that in an early stage of electrical 
 discovery, when a large induction of facts was wanting, either to 
 settle the matter in favour of, or against, the new invention, such 
 grounds of dissent were by no means either unreasonable or un- 
 tenable. No great wonder then that men's minds, especially those 
 of mariners, who of all others were frequently exposed to terrific 
 thunderstorms, should have become confused and often supersti- 
 tiously unsettled upon the question. We accordingly find very 
 extraordinary means resorted to both at sea and on shore, with a 
 view of keeping off lightning ; means which forcibly point out the 
 extremely unenlightened state of mankind upon this great and 
 important question. In many countries, not excluding this, 
 it was customary on the approach of a thunderstorm to agitate 
 the air by ringing the bells of church towers, more especially on 
 the Continent, where the bells had been commonly blessed by a 
 religious ceremony. Occasionally the practice of a discharge of 
 cannon or artillery was resorted to. In ships it was customary to 
 apply wet swabs over the pumps and doors of the magazines, 
 with a view to prevent the entrance of lightning. In houses, 
 during a thunderstorm, it was customary, up to a recent period, to 
 throw up the sashes and enclose glass articles, especially looking- 
 glasses, in- blankets. Balls of glass were placed on the mast-heads 
 of ships and on steeples, as having the power to repel lightning. 
 This superstition, so grossly absurd, has been revived even in 
 the present day, when science has made such great advances 
 as would lead one to imagine so ridiculous an assumption im- 
 possible. 
 
 It will be my object on the present occasion to bring under your 
 notice some striking practical deductions and experimental illustra- 
 tions bearing on the question of security against lightning, and by the 
 assistance of three great lights of Baconian Philosophy induction, 
 observation, and experiment so to simplify and clear the subject 
 of irrelevant and sophistical argument, as to leave no doubt on 
 your minds as to the course we should pursue in placing buildings, 
 magazines, ships, and other structures, beyond the reach of danger 
 from one of the most terrible elements of nature. 
 
 Before, however, entering more especially upon the laws and 
 mode of operation of the peculiar physical force active in atmos- 
 pheric electrical discharge, it may not, perhaps, be amiss to call 
 your attention to some of its distinctive effects. 
 
DAMAGE TO SHIPS, BUILDINGS, ETC. 273 
 
 A writer in " Nicholson's Journal of the Progress of Science " 
 calculates the loss by lightning, in Great Britain alone, at 50,000 
 sterling annually. I believe this to be under the mark. In the 
 Royal Navy, during the late war, the country was losing at least 
 10,000 a year upon cases of damage found recorded in the 
 journals of H.M.'s ships. We had, between the years 1810 and 
 1815, that is, within about five years, no less than forty sail of the 
 line, twenty frigates and twelve sloops and corvettes, placed hors 
 de combat by lightning. In two hundred and fifty such cases one 
 hundred seamen were killed, and two hundred and fifty, at least* 
 severely hurt. In the Merchant Navy, within a comparatively 
 small number of years, no less than thirty-four ships, most of 
 them large vessels with rich cargoes, have been totally destroyed 
 being either burned or sunk to say nothing of a host of vessels 
 partially destroyed, or severely damaged. Damage to H.M.'s 
 ships by lightning has happily ceased : it is now not known in the 
 British Navy. In the Merchant Navy, however, it unfortunately 
 continues, no adequate means being taken, as in the Royal Navy, 
 to check it. 
 
 If we turn to buildings on the land, here also we find records of 
 destruction of the most frightful character. Fuller, in his " Church 
 History, " says, that " scarcely a great abbey exists in England 
 which once, at least, has not been burned by fire from heaven." He 
 quotes, as examples, the abbey of Croyland twice burned, the 
 monastery of Canterbury twice, the abbey of Peterborough twice, 
 the abbey of St. Mary's, in Yorkshire, the abbey of Norwich, and 
 several others. The number of churches and church spires wholly 
 or partially destroyed by lightning is beyond all belief, and would be 
 too tedious a detail to enter upon. Within a comparatively few 
 years, in 1822 for instance, we find the magnificent cathedral of 
 Rouen burned, and so lately as 1850, the beautiful cathedral of 
 Saragossa, in Spain, struck by lightning during divine service and 
 set on fire. So lately as March of the last year, a despatch from 
 our Minister at Brussels, Lord Howard de Walden, was forwarded 
 by Lord John Russell to the Royal Society, dated 24th February, 
 stating that on the Sunday preceding a violent thunderstorm had 
 spread over Belgium, that twelve churches had been struck by 
 lightning, and that three of these fine old buildings had been 
 totally destroyed. 
 
 Explosions of gunpowder magazines by lightning have been 
 frequent. Lately, at Sondpore in India in 1857 lightning fell 
 on the magazine, fired it, and nearly a thousand people lost their 
 lives. In the preceding year, 1856, lightning struck the church of 
 
274 FIRST LECTURE. 
 
 St. John's in the Island of Rhodes ; the electrical discharge exploded 
 a large deposit of gunpowder in the vaults beneath : the whole 
 building and part of the town were laid in ruins. In 1829 lightning 
 blew up the magazine at Navarre, and destroyed everything around. 
 Admiral Bosimel reports this case to his Minister of Marine at 
 Paris. A lamentable occurrence of this kind in 1769 laid a fourth 
 part of the beautiful city of Brescia in ruins. About twenty such 
 cases may be quoted. 
 
 Without going more at length into instances of destruction by 
 the terrible agency of lightning through a long series of years, I 
 will merely mention a few cases that have occurred within the last 
 two or three years. 
 
 In 1859 the General Hospital at Jersey was struck and set on 
 fire ; all the centre of the building was destroyed. In 1857 a 
 goods shed on the South Eastern Eailway was struck ; 350 feet of 
 the roof fell in. In 1860 several churches were struck and severely 
 damaged. 
 
 I cite these few cases merely with a view of showing how much 
 more frequent and ruinous are the effects of lightning in this 
 country than is commonly supposed ; and now having, as I think, 
 sufficiently adverted to this, I will, in conclusion, ask your atten- 
 tion to two instances of the actual effects of this agency, as exem- 
 plified on sea and on land. These will better enable you to under- 
 stand what sort of agency we have to deal with. 
 
 In this Diagram, No. 1, you have a faithful representation of one 
 of our frigates, the Thisbe, of 36 guns, struck by lightning off 
 Scilly, in January, 1786. I will read you the extract from the 
 ship's log, and I think you will say that nothing can be more 
 frightful: 
 
 " Four a.m. Strong gales ; handed main-sail and main top-sail ; 
 hove to with storm stay-sails. Blowing very heavy, S.E. 4.15, a 
 flash of lightning with tremendous thunder disabled some of our 
 people. A second flash set the main- sail, main- top, and mizen 
 stay-sails on fire. Obliged to cut away the main-mast ; this carried 
 away mizen top-mast and fore top- sail yard. Found fore-mast 
 also shivered by the lightning. Fore top-mast went over the side 
 about 9 a.m. Set the foresail." 
 
 Well you see here was a ship reduced to a mere wreck. I put 
 this case before you as a powerful practical illustration of what 
 lightning may effect ; and I beg you not to take it as an isolated 
 case, happening once in a hundred years ; such cases are numerous. 
 Here is another similar case in a sister ship the Loicestoffe, struck 
 
NATURE OF ELECTRICAL FORCE. 275 
 
 in the Mediterranean, in March, 1796, some few years after. I 
 will read you an extract from the log of this ship. 
 
 "North end of Minorca, S. 46 W. ; heavy squalls, E. by S. ; 
 hail, rain, thunder, and lightning. 12.15, ship struck by lightning, 
 which knocked three men from the mast-head, one killed. 12.30, 
 ship again struck. Main top-mast shivered in pieces ; many men 
 struck senseless on the decks. Ship again struck and set on fire in 
 the masts and rigging. Main-mast shivered in pieces, fore top- 
 mast shivered ; men benumbed on the decks and knocked out of 
 the top, one man killed on the spot. 1.30, cut away the main- 
 mast ; employed clearing wreck. 4, moderate ; set the foresail/' 
 
 Here is another ship left a wreck on the sea by lightning. 
 
 The Diagram No 2 is a faithful representation of the remains of 
 the beautiful spire of St. Michael's Church, at the Black Rock, 
 Cork, as it appeared on the morning of the 30th January, 1836, 
 after being struck by lightning on the preceding night. Imagina- 
 tion cannot form a picture of a more awful ruin. The side of the 
 steeple was fairly torn out, the top of it swept away, and the stones 
 scattered in all directions. You will, I trust, excuse my troubling 
 you with these introductory remarks ; but I am desirous you 
 should fully comprehend the kind of agency with which we are 
 dealing, and the vast importance of the question of security against 
 its effects, under the form of lightning, to the national interests 
 and to the public service generally. 
 
 Well, then, having by a sort of Discours Preliminaire so far 
 called your attention to the nature of the subject of which we are 
 about to treat, I shall proceed with a few remarks on electrical 
 force generally, at least so far as involving atmospheric electricity 
 and the action of lightning, and electrical discharge. In doing 
 this, I shall not pretend to develop what is usually termed a 
 course of elementary electricity. I have no doubt but that you 
 are fully aware of the ordinary elementary phenomena of electrical 
 action, so that what I shall lay before you will be easily compre- 
 hended. We will, therefore, step on to great leading facts and 
 principles of a practical kind, bearing, more especially, on the great 
 question we have in hand. 
 
 In the first place, I must request you to dismiss from your minds 
 all notion of an electric fluid or fluids, as the immediate or occult 
 power in which electrical action originates. It is a very common 
 and a very learned error to refer the tremendous power of which 
 I have given you such striking examples to an immaterial fluid of 
 high elasticity ; an assumption involving such manifold absurdities 
 
 T 2 
 
276 FIRST LECTURE. 
 
 as to be utterly untenable in treating a practical question such, as 
 we have now before us. 
 
 When Newton by a sublime geometry and a profound analysis 
 treated the movements of the planetary system, he was content to 
 recognise gravity, in his doctrine of a universal gravitation, as 
 a physical force taken in the abstract ; and in order to study its laws 
 and mode of operation, he did not for a moment, at least in his more 
 immediate researches, seek to determine, or pretend to explain, 
 the occult nature of gravity, considering such an inquiry with- 
 out the limit of legitimate physical research. Now, this is just 
 the course I propose to you to follow. We shall view the agency 
 which rent open the steeple already referred to in this dia- 
 gram as a physical force of nature, taken as force merely, and 
 we shall study its laws and its mode of operation. Now it is no 
 sound objection to this to say that, being ignorant of the nature 
 of the agency with which we are dealing, we can scarcely treat it 
 with clearness and precision, because the great end of modern 
 philosophy is rather to trace and apply the uniform relations of 
 certain facts, than speculate upon the source of a mysterious causa- 
 tion, most probably by a decree of the Almighty placed beyond 
 the reach of that mental power with which he has endowed us, 
 and which tells us pretty plainly how far we can not reach, at all 
 events. 
 
 Suppose I bend a spring a Kentish bow, for example, with 
 which our forefathers shot their arrows, a cloth-yard long what 
 do I effect ? You will immediately perceive that all the particles 
 on the concave side of the arc are compressed ; all those on the 
 convex side become expanded. I relax the string : the particles 
 on the convex side contract, which force, for distinction sake, I 
 may call positive; those on the concave side expand, hence another 
 kind of force which I may call negative. Or, reciprocally, these 
 forces make up the total power of the bow, and, in combining, 
 operate so forcibly on the arrow, through the bow-string, as to 
 shoot away the arrow with enormous power and great velocity. 
 Now we are accustomed to designate this power by the term 
 elastic force, or elasticity. And really the fact itself is all we 
 know about the matter. No one can say in what the elasticity 
 consists, or really how the motion is induced ; and I think you 
 will see how little we shall be likely to gain by assuming the 
 existence of an elastic fluid of high tension as the source of it. 
 Similar reasoning applies to the assumption of an electric fluid 
 as a source of electric power. For any good, sound, practical 
 purpose we had better come at once to consider electricity as a 
 
ELECTRICAL INDUCTION. 277 
 
 physical force, the nature of which, is quite unknown to us, but 
 the laws and modifications of which are within our reach. 
 
 Observing then a peculiar series of phenomena which lead us 
 directly to recognise the existence of a peculiar unknown power of 
 nature which we term electricity, the first important fact we observe 
 is the dual nature of this force : it is made up as it were of two 
 distinct and inseparable modes offeree .these have been for the sake 
 of clearness termed positive and negative >as we named the forces 
 of the bent bow. You cannot have one of these alone, the other 
 is always present. Hence the analogy of the bent bow helps us in 
 some degree, although it may not help us in other cases. In the 
 bent spring we have, as already observed, compression of particles 
 upon the concave surface, extension upon the convex, from which 
 results a force of expansion on the one side, a force of contraction 
 on the other ; both forces are operative in bringing the bent bow 
 to its normal state. You cannot in the bent bow have one force 
 without the other. Now such is the ease in electrical action. 
 Electricity, therefore, like elasticity, you will regard as a peculiar 
 property of common matter, or a peculiar force impressed on it. 
 The immediate sources of the phenomenon that we term electrical, or 
 the means of bending, as it were, our electrical bow, are numerous ; 
 for instance changes of temperature ; changes of form ; me- 
 chanical actions, such as friction, pressure, percussion, and the like ; 
 to which may be certainly added the contact of bodies ; any of these 
 may superinduce upon common matter an apparently attractive 
 force, as you see in all common electrical experiments. In this 
 case the bodies are said to be electrified, or charged with electricity. 
 
 A most wonderful and striking fact, characteristic of electrical 
 force, is its apparently sympathetic action or foresight, as it 
 were. It pioneers, in fact, for its subsequent operation ; it finds 
 out, in advance and beforehand, what it can effect, what course 
 it may best follow, and it impresses upon common matter a 
 sort of physical preparation for its reception, in the shape of an 
 active power. This species of electrical action, or, to all appearance, 
 action at a distance, has been commonly termed induction; and 
 by our scientific neighbours, the French, electricity par influence. I 
 must beg of you before entering upon a few elementary experimental 
 illustrations of these and other electrical phenomena, to take the 
 electrical apparatus before you to a certain extent upon trust, 
 or rather upon the faith of your general knowledge of ordinary 
 electricity, or upon the ground of a broad generalisation, without 
 further elaborate explanation and detail. You will be so good, 
 for example, as to consider what is termed the electrical machine 
 
278 FIRST LECTURE. 
 
 as an instrument which enables us to call up, through the aid of 
 friction, the force we term electricity, that is, the true electrical 
 powers of which we have been speaking, and apply either of them 
 or both to the purposes of experiment. You will look at the 
 Ley den or electric jar as a means of accumulating one or both 
 of these forces, and causing them to assume a dense explosive form, 
 imitating a discharge of lightning. Other kinds of apparatus or 
 auxiliary instruments you will view as instruments for facilitating 
 experiments on electrical accumulation and discharge, or other 
 electrical investigation. We shall occasionally offer a few ex- 
 planatory remarks on all these instruments as we proceed, many 
 of them not being generally known. 
 
 It may, perhaps, be as well to observe in passing, that the 
 physical agency or power we term electricity, is subservient in 
 a certain sense to the kind of common matter with which it is 
 associated ; some bodies, for example, propagate or transmit it with 
 rapidity from one substance to another. Such bodies have been 
 since termed electrical conductors. The metals, acids, carbon, &c., 
 do this ; other kinds of matter, such as vitreous and resinous bodies, 
 arrest its progress, and stop its course ; they, in fact, imprison it, 
 as it were ; these have been termed insulators of electricity. 
 
 We will give a few practical illustrations of this. 
 
 Exp. 1. Electrical machine-conductors and non-conductors. 
 All bodies have conducting power ; none are perfect insulators. 
 Division into most perfect, less perfect, and imperfect. See Table. 
 
 We have further to observe that an electrified body is any 
 insulated substance having either of the electrical forces domi- 
 nant in it, or in different parts of it ; and it is said to be 
 charged positively or negatively, according to the particular 
 modification of force with which it is thus electrified. In 
 common parlance, the two forces have been designated positive 
 and negative electricities; positive electricity being arbitrarily 
 associated with the electrical development induced by the friction 
 of glass with silk; whilst negative electricity is identified 
 with the development induced by the friction of resinous bodies 
 with woollen ; hence, as you know, the terms, vitreous and 
 resinous electricity, often called the vitreous fluid and the resinous 
 fluid. We may certainly employ this figurative language as a 
 matter of convenience, but we must take very especial care not 
 to build on it as fact. As already noticed in the case of the bent 
 bow, we may for convenience call the force of the convex surface 
 positive elasticity, and the force of the concave surface negative 
 
ILLUSTRATIONS OF INDUCTION. 279 
 
 elasticity ; but by this we are after all only designating force, or 
 kind of force. 
 
 Thus premised, we will first take a case of what is termed ekc- 
 trical attraction, and I propose to show you that in this the pre- 
 vious induction or foresight of the electrified body is present. 
 The phenomenon we term attraction, you will observe, is the first 
 and most palpable evidence of the presence of electrical force. It 
 was in fact from this apparent power developed by the friction of 
 amber, that the Greek philosopher, Thales, inferred that mineral 
 to possess a species of animation ; hence, as you are aware, the term 
 electricity, from the Greek rjXtKTpov, signifying " amber." 
 
 Exp. 2. Suspended movable arm and disc with gold leaf on 
 opposed electrified bodies ; the leaf diverges, and attraction 
 ensues. Explain the experiment. The quantity jar, 
 carrier-ball, &c., &c. 
 
 We see here a striking instance of what is termed electrical 
 induction apparently action at a distance. We will take it as such, 
 without entering upon theoretical explanations. So that you see 
 before a body is attracted electrically, it is first rendered attract- 
 able ; the twin power or force is first developed in it by the 
 influence of the conjugate or attracting force ; and this is a uni- 
 versal law of electrical action. But we have apparently a second 
 species of electrical force to bring under your notice, evident in the 
 divergent leaves, also important to illustrate. In the previous 
 experiment we see the immediate result of a development of the 
 two electricities, viz., attractive power. But let us see what would 
 happen, supposing each of the bodies were electrified alike, either 
 positively or negatively. In this case the tendency to a final attrac- 
 tive force is still present, and an effort is made by each body, as in 
 the former case, to develop an opposite power in the other, without 
 which it may be shown that no attraction could ensue. Before 
 this attraction can be effected, the electrified state of each must be 
 more or less changed, their states must be in some considerable 
 degree not merely reduced to zero, but positively reversed. The 
 consequence of this effort is an apparent repulsion of the two 
 bodies, if free to move. The already existing electrical states 
 refuse, as it were, under existing circumstances, to be coerced in 
 this way ; they resist, and the bodies separate. Still here, as 
 before, the advanced influence is first operative, before divergence 
 or separation of the two electrified bodies ensues. 
 
 Exp. 3. Show the increased divergence of the electrified leaf 
 on the movable arm at the instant of repulsion. 
 
280 FIRST LECTURE. 
 
 It is from this and other phenomena that we have the well- 
 known electrical formula dissimilar electricities attract similar 
 electricities repel. That is to say, two positive or two negative 
 forces repel ; a positive and a negative force attract. In fact, 
 attractive force cannot exist except in the case of these opposite 
 forces. 
 
 Now whatever be the source or occult cause of electrical 
 force, certain it is, the two forces it develops are singularly and 
 peculiarly associated with common matter. Different kinds of 
 matter have a disposition to develop positive electricity; others 
 negative electricity. Thus sulphur in falling through the air 
 becomes negatively electrified ; red oxide of lead positively 
 electrified. The following experiment, whilst it shows the 
 attraction of opposite forces, illustrates in a remarkable way this 
 disposition of certain bodies to assume a positive or a negative 
 electrical force. 
 
 Exp. 4. Sift red lead and sulphur on a resinous plate, and 
 explain. 
 
 We will here stop to illustrate two important laws of electrical 
 force immediately bearing on our present subject. We have 
 already shown the tendency of two bodies charged with opposite 
 electricities to approach each other. Now such is precisely the 
 case with electrified clouds such, more especially, as those we 
 term thunder- clouds and the opposed surface of the earth. It 
 is well we should understand, therefore, the law according to 
 which the force increases, as we increase the quantity, as we may 
 term it, of the electrical charge, or as we diminish the distance. 
 
 Now it may be shown that the force between a mass of 
 electrified clouds and the earth increases with the square of the 
 quantity of electrical charge directly, and decreases with the 
 square of the distance inversely. Yery much intricate discussion 
 and experimental and analytical calculation has been expended 
 upon these laws of electrical force ; but I believe I shall be 
 enabled to show you, by the balance apparatus before us, that, 
 as mere experimental facts, the demonstration of this is simple 
 and easy. 
 
 Exp. 5. Discs and electrical balance. Accumulate a certain 
 number of measures of charge on discs. Constant distance. 
 Note weights for these measures. Double the measures, 
 and again note the force. F oc 2 2 . 
 
 Exp. 6. Double the distance for the last noted quadruple 
 
EXPERIMENTS ON INDUCTION. 281 
 
 force. The force will be now one-fourth and same as at 
 the unit of distance. 
 
 It would be doubtless interesting to pursue this question of elec- 
 trical attractive force, and unfold all its very wonderful phases 
 and physical bearings on bodies of various forms, as, for example, 
 on spheres, planes, and other surfaces; but, as I have already 
 remarked, we cannot possibly on this occasion treat the whole 
 science of electricity. I must, however, call your attention to an 
 extremely important phenomenon of electrical force as thus exem- 
 plified, viz. wherever induction can go on, attractive force follows. 
 If it cannot, or at least, to any extent, we have little or no attrac- 
 tive force. Be so good as to fix attention on this, because it bears 
 immediately on the question of protection from lightning. "We 
 have all heard of the notion entertained by the generality of persons 
 of the danger incident to metallic bodies on the occurrence of 
 lightning storms, because of their superior attractive force, vanes, 
 spindles, knives, and spears, to say nothing of metallic substances 
 generally, have all been commonly deemed dangerous in the ex- 
 treme. This, however, is an unlearned and a very vulgar error, as 
 you will presently find. Pointed metallic bodies, or other metallic 
 bodies, have no more power of drawing or attracting lightning than 
 non-metallic substances, or even insulators of electricity, taken in 
 the sense implied, that is to say, as having a specific affinity or 
 inherent attractive force for the matter of lightning. Both Caven- 
 dish and Coulombe found electrical charge to be quite independent 
 of the nature of the substance charged. An equal division of elec- 
 tricity always obtains between bodies sooner or later without any 
 regard to the kind of substance electrified. I proceed to one or 
 two illustrations of this. 
 
 Exp. 7. Electrical balance wood disc metal disc equal. To 
 show attractive forces equal, or spheres. 
 
 In these and the preceding similar cases you perceive, as indi- 
 cated in the third experiment with the movable disc, a free 
 induction upon the earth, or upon the source of charge, is open to 
 the attracting and attracted bodies. But let us arrest this induc- 
 tion in some considerable degree, by insulating the attracted 
 body, that is, cutting off its free induction upon the ground, 
 then the opposite electrical force is no longer developed to an 
 extent sufficient for attractive effect, that is, in the especial case 
 I am about to show you ; for you are to recollect that it is upon 
 the opposite electrical forces the action depends not upon the 
 kind of matter absolutely with which they are associated. 
 
282 FIRST LECTURE. 
 
 Exp. 8. Suspend the thin disc by silk or gutta-percha or fine 
 silk gut ; compare, first, the attractive force of a given 
 number of measures, when uninsulated by a fine wire. Now 
 remove wire and weight of wire. Little or no attraction. 
 
 You perceive in these instances how ill supported by experiment, 
 and, as you will presently find, by the great physical operations 
 of nature, are the common notions of lightning as being especially 
 attracted by certain kinds of matter in preference to other kinds 
 of matter. There is, however, one peculiar kind of action which 
 appears to favour the conclusion that attractive force is resident in 
 pointed metallic bodies, seeing that lightning falls on the metallic- 
 pointed terminations of church spires, the vane spindles of ships, 
 and such like, and, as it is assumed, with a disregard of other 
 bodies. In treating this question, which is of a somewhat vexed 
 character, you will have to draw a very nice line between absolute 
 attractive force and apparent attractive force, and limit by clear 
 definition, what you really understand by attraction. We will 
 first proceed to one or two experimental illustrations. 
 
 Exp. 9. Let a current of sparks pass between two balls, as in 
 Exp. 1 ; present a pointed uninsulated metallic conductor at a 
 distance ; sparks stop ; take away, they return, &c. The 
 point becomes luminous at a distance. 
 
 It may be here inferred that this pointed body has actually 
 attracted the matter of lightning towards itself ; well, let it be so ; 
 we will by way of a redmtio ad absnrdum, grant it. Then comes 
 the question, is this result peculiar to pointed metallic bodies ? Is 
 it not also an effect producible by other bodies not metallic ? Let 
 us see : 
 
 Exp. 10. Repeat former experiment, and present a wooden 
 point : same effect ensues. 
 
 You see common wood has the same property ; it is just as 
 attractive of the matter of lightning as -metal. Well, now comes 
 the question (for you see I am treating this matter pretty plainly 
 and practically, and with an eye to our subject), will the metal 
 point always, under all circumstances, act in the way just exem- 
 plified ? No, it will not. 
 
 Exp. 11. Repeat stream of sparks as before ; present an in- 
 sulated point ; sparks do not stop. 
 
 We may fairly question the assumption, upon the faith of these 
 
ACTION OF POINTS. 283 
 
 facts, of the peculiar affinity or attractive influence of metallic bodies 
 for the matter of lightning. This leads us to inquire further into 
 the immediate source of the phenomenon in question, viz., this 
 peculiar influence of pointed bodies. 
 
 On a careful investigation of this phenomenon, we find it to 
 depend on a peculiar adaptation of acute terminations to the 
 purposes of electrical induction. Acute terminations in free con- 
 ducting bodies so dispose and prepare the particles of any resisting 
 medium, intermediate between attracting bodies in opposite 
 electrical states, as to greatly facilitate electrical action in the 
 direction or position in which they happen to be placed ; when they 
 are not enabled to do this, then, as you have just seen, they are 
 inert. Thus, in Exp. 10, as we have just witnessed, the force is 
 originally exerted between the discharging balls. The two forces 
 are developed between these ; their union is attended by a breaking 
 down of the particles of the resisting atmosphere intermediate 
 between them. Light and heat are both developed in this rupture, 
 and a succession of explosive sparks ensues, as we continually 
 evolve electricity by the electrical machine. But now, when I 
 bring a distant free point into competition with the receiving ball 
 of this system leading to the earth, then by its peculiar action on 
 the arrangement of the particles of the intermediate resisting air, 
 it facilitates the inductive force on the direction of its position to 
 so great an extent, as to supersede the explosive discharge between 
 the discharging balls altogether, and thus cause what we may 
 term the current of discharge to pass in a different direction, and 
 in a thin attenuated stream upon the point, without explosion. 
 This is an important phase of electric force. You may call it 
 attraction if you like ; but it is not that species of attraction as 
 commonly understood. If we place a large hollow pipe in com- 
 munication with the gutter leads of a building, which, by its 
 facilitating the discharge of heavy rain, prevents its accumulation 
 and discharge in some other direction, we may, in a certain sense, 
 and if we choose to indulge in such a figure, attribute the operation 
 of this pipe to a species of attractive force upon the water, by 
 causing the water to gravitate in a certain direction ; but we 
 surely should not dream of attributing to such a pipe a power of 
 attracting the rain from the clouds, and so drawing down the water 
 upon the building! 
 
 Let me for an instant, by way of concluding this part of our 
 subject, ask the favour of your momentary attention to a final 
 experimental illustration bearing on this question. We have seen 
 that without the presence of the two electrical forces, without this 
 
284 F1KST LECTURE. 
 
 pioneering influence of induction, we have little or no result ; let 
 us now see how this may be promoted by a pointed free body I 
 mean a conducting body in connection with the earth, and acting 
 at a distance ; and also, when this induction is perfected, how 
 immediately the attraction ensues. 
 
 Exp. 12. Repeat Exp. 9. Collect a certain number of measures 
 sufficient to induce attraction in the free state ; then present 
 a point over the suspended disc ; attraction immediately 
 ensues by completing the induction ; so it does with wood. 
 
 We see here a case of an apparent repulsion of a point. 
 
 We shall conclude this First or Introductory Lecture here, our 
 time being pretty well expended. Before I do so, however, it 
 may be well to pass over again, by a brief recapitulation, the path 
 we have been following. 
 
 RECAPITULATION. 
 
 We have, first, in the way of introductory matter, called atten- 
 tion to Franklin's invention of the lightning-rod, and the ob- 
 jections to it ; we have next offered remarkable instances of the 
 effects of lightning and its ruinous consequences to public and 
 private interests ; we have taken a general view of electrical force, 
 on which the destructive effects we have called attention to 
 depend : we find it a sort of dual power, and made up of two 
 modes of force ; we have seen that in every case of what is termed 
 electrical action, both these kinds of force, termed, for distinction's 
 sake, positive and negative, are called up by a peculiar operation, 
 termed induction, apparently action at a distance ; and we have 
 seen this to be a sort of pioneering action, and precedes the 
 electrical phenomena of attraction and repulsion. 
 
 Considering attraction as the immediate consequence of electrical 
 development, we have called attention to two important laws ; we 
 have found the force to be directly as the square of the develop- 
 ment, or what for convenience we may term the quantity of 
 electricity accumulated, and inversely as the square of the distance 
 between the attracting bodies. We have shown these laws to 
 depend on a free induction : if either body be limited in this 
 respect and the induction arrested, we obtain little or no attractive 
 force. We have called attention to the peculiar action of pointed 
 bodies, considered as sources of attraction, and shown upon what 
 their operation depends. 
 
SECOND LECTURE. 
 
 HAVING in my last lecture considered certain elementary principles 
 of ordinary electrical action, bearing on the phenomena of thunder- 
 storms, I now proceed to the more immediate object of these lectures 
 the laws and operation of electrical discharge under the form of 
 lightning, and the means of protection from its destructive efiects. 
 With a view to continuity, however, you will, I hope, excuse a 
 brief recapitulation of my former observations. 
 
 It will be convenient and perhaps judicious, in continuing these 
 remarks, to commence with a general view of Atmospheric Elec- 
 tricity and the nature of thunder-storms. By certain natural 
 actions, such as changes of temperature, chemical or other opera- 
 tions, the agency which we term Electricity becomes developed in 
 our atmosphere, and some thousands of acres, perhaps, of condensed 
 vapour, under the form of cloud, become charged with electricity. 
 The air intermediate between the plane of this area of cloud and 
 the earth's surface participates in the inductive force upon the 
 plane of the opposed surface of the earth beneath. The two elec- 
 trical forces thus disjointed and brought into play seek, by a law 
 of nature, to reunite and recover their normal state of repose, just 
 as water, by a law of gravity, seeks its level of rest. But as you 
 saw in the preceding lecture, air is an insulator, and will not admit 
 of this immediate union of the two forces, until at length the 
 attractive force between the plane of the area of cloud and that of 
 the earth's surface beneath is so powerful as to break down the in- 
 tervening air with a terrific and dense explosion. Now it has been 
 a very common misapprehension, in ordinary views of lightning 
 discharge, to consider strokes of lightning on buildings as a species 
 of force or action set up between the building, or some part of it, 
 and the thunder-sky which sends forth its terrific flash, and which 
 alarms us by the power of its dreadful artillery. This is a great 
 mistake : the action is not between the clouds and the building, but 
 
286 SECOND LECTURE. 
 
 between the clouds and a large area of the earth's surface ; the 
 building is only acted on and struck by the electrical discharge, as 
 being a point in the earth's surface, and at the moment placed in 
 a given electrical position. 
 
 Well, then, in this point of view, we have to take into account 
 the area of watery vapour, the area of the earth's surface opposed 
 to this, and the intermediate air ; and we have to consider the 
 peculiar conditions of these three elements. First, we have a 
 development of force by induction between the clouds and the 
 earth, by which opposite forces, as already observed, are called into 
 play. These may be taken as the terminating planes of the action. 
 Now the condition of the intermediate air, termed by Faraday the 
 Diakctric Medium, or medium through which the forces operate, 
 is an extremely interesting question. It is really difficult to 
 realise action at a distance, as if by influence or sympathy ; and yet 
 we can scarcely tell how to avoid it. According to Faraday, the 
 intermediate particles of the atmosphere that is, the dialectric 
 become polarized. Electrical forces in these particles developed by 
 influence become disposed as in the diagram. The positive and 
 negative forces, which originally existed in a state of union or 
 repose, are separated and opposed to each other in a series, and so the 
 force becomes propagated to or from the earth's surface. So long 
 as the particles can sustain this forced state, so long insulation is 
 the result, and the forces upon the terminating planes of the action 
 that is, the area of cloud and the earth's surface are maintained ; 
 but if they break down, as it were, or give way to reversion of the 
 polarized state, the whole system restores itself to its normal con- 
 dition by what is termed Disruptive Explosive Discharge. Light 
 and heat are evolved, attended by violent expansive power. Some 
 little illustration of this by means of experiments may perhaps be 
 useful here. 
 
 Exp. 13, with illustrations. Two circular discs, 18 inches and 
 1 foot, fitted with electroscopes, show induction, insulation, 
 and discharge, and condition of air between the plates, &c. 
 
 Conceiving the fixed large circular disc to be a portion of the 
 earth's surface, and the lesser disc an area of cloud over it, let us 
 see first what is the immediate result of an electrical development 
 on this upper area. I may be supposed to hold in my hand here 
 some thousand acres of charged cloud ; the divergent electroscope 
 is evidence of the presence of this charge ; I bring it immediately 
 over the lower fixed plane. You see its inductive effect : all the in- 
 tervening air is now in a state of great excitation, and if the force 
 
POLARIZED AIR. 287 
 
 induced between the two planes were sufficiently great, it would 
 break down with an explosive spark, termed by Faraday, as already 
 noticed, the disruptive discharge. This explosive disruption would 
 occur in the points of least resistance ^that is to say, in points in 
 which, from a great many incidental circumstances, such as the 
 presence of foreign matters, &c., the particles of the polarized 
 atmosphere would most readily yield. Imagine the small body I 
 place upon the lower disc to represent a building : this would be a 
 mere point in the great surface. It would not produce discharge 
 as upon itself. It would be nothing more nor less than the door- 
 way, as it were, through which the electrical forces operated between 
 the opposed planes. An easier course, not far distant from the 
 building, might arise from many causes, in consequence of the 
 easier yielding of the air in that line. We have represented in 
 this diagram a mass of clouds, c, a portion of the earth's surface, 
 2 N, with an intermediate atmosphere ; R, the position of a building 
 or ship; and suppose a thunder- stroke to originate at c, we at once 
 perceive that resistance from many causes, through c N or c o, 
 might be much less than through the course c R, so that the 
 thunder- stroke is in no sense determined by attractive force in the 
 building ; it is the total force of the whole surface which deter- 
 mines it. And when we consider how small comparatively is the 
 magnitude and elevation of any building or other artificial structure, 
 when taken in relation to a vast mass of clouds and electrified air, 
 we see that it vanishes in the general surface. We find accord- 
 ingly that lightning often strikes near a building or ship directly 
 through the air without touching it ; and so definite and determined 
 is its course, that instances have occurred, especially in the navy, 
 in which the sleeve of a coat has been rent from the arm of the 
 wearer without damage to his person, so that this is purely a 
 question of resistance in this or that direction, as we shall presently 
 see. 
 
 Before I proceed to illustrate the laws of destructive electrical 
 discharge, and the means of palliating or avoiding its terrible 
 effects by an efficient and systematic application of metallic con- 
 ductors, it may be interesting to exhibit, by very conclusive and 
 pretty experimental illustrations, the operation of a charged atmo- 
 sphere such as we have alluded to. I have little doubt but that 
 highly charged clouds and a polarized air are the sources of many 
 of the phenomena classed as whirlwinds, waterspouts, and such-like, 
 and certainly of the luminous phenomena frequently seen on the 
 vane-spindles of ships and towers, termed by sailors comazants, and 
 noticed on soldiers' spears by ancient writers. Thus Caesar, " On the 
 
288 SECOND LECTURE. 
 
 African War," De BettdAfricano, says, "That same night the points 
 of the spears of the fifth legion burned with fire," &c. Seneca, 
 again, in his "Natural Questions/' remarks, "The spears in the 
 Roman camp seemed to be on fire." 
 
 Franklin, after his great experiment with the kite, in which he 
 obtained electricity from the atmosphere as we do now artificially 
 by the conductor of the electrical machine, resorted to an extremely 
 elegant and simple contrivance, by which he obtained notice of the 
 passage of charged clouds over his house. I have in my last 
 discourse shown and explained the operation of pointed free con- 
 ductors. I have here so modified Franklin's beautiful contrivance 
 as to bring it within the limits of ordinary experiment. 
 
 Exp. 14. The Thunder-cloud and Bells. 
 
 The source of the electrical accumulation is here our electrical 
 machine, the means of accumulation the electrical jar, a swinging 
 arm poised on the charging rod carries an artificial cloud, which, 
 being composed of light cotton wool, freely receives the concen- 
 trated electricity, and spreads its fibres in the way of the electrified 
 vapours of the atmosphere. The air under this charged substance 
 becomes affected by induction, which reaches to the bodies beneath, 
 just as in the case of a charged atmosphere, and will hence in- 
 fluence a building with a pointed conductor when passing over it. 
 If the conductor be insulated as in the model, and be connected 
 with an insulated bell, the charge it receives will accumulate on 
 the bell, and will readily attract a small insulated ball of metal 
 freely suspended near it by a silk thread. But when this ball 
 becomes, by contact with the electrified bell, also electrified, and 
 with the same kind of electricity, then, as we have seen, it tends 
 to separate from it by a species of repulsion ; so that if a second 
 bell uninsulated that is, in connection with the earth be placed 
 on the other side of the suspended ball, it will pass to that, and 
 discharge its electricity upon it, in which case it will return 
 again to the electrified bell, and so continuously as long as 
 the artificial cloud throws off electricity and charges the inter- 
 vening air. We have in this way a continued sounding of the 
 bells, a constant ringing. Such was Franklin's arrangement 
 through his house. The passage of a cloud charged with electri- 
 city over his conductor invariably gave notice of its presence. 
 
 The phenomena of whirlwinds are easily imitated in this way, 
 as in the whirling columns of sand in a desert, and such like. 
 
 Exp. 15. Some light bran is to be strewed on a conducting sur- 
 
POLARIZED AIR. 289 
 
 face, and an artificial cloud made to swing over it. The 
 bran will be caught up in a sort of whirlwind. 
 
 These experiments apply more especially to certain phenomena 
 of a charged atmosphere unaccompanied by condensed explo- 
 sion or disruptive discharge. This we will now proceed to 
 consider. 
 
 When the disjointed electrical forces are so powerfully attractive 
 of each other as to break down all insulating resistance, and unite 
 with spasmodic violence, then it is that the phenomena termed 
 Lightning and Thunder present themselves. Lightning is the 
 extrication of brilliant light, accompanied by heat, from the air or 
 dialectric medium, whose particles are violently broken through ; 
 that is, depolarized. Thunder is the crash of this effort, prolonged 
 by repeated echoes and reverberations from the surface of clouds 
 or other bodies, just as the sound of artillery reverberates amongst 
 distant hills. The crash itself, heard near at hand and apart from 
 reverberation, is something very horrid : it is as if an enormous num- 
 ber of earthenware jars of great magnitude, suppose ten thousand 
 in number, were smashed by falling from a great altitude upon a 
 reverberating stone pavement. When the thunder- stroke is close, 
 a vivid gleam of light immediately precedes this. If distant, the 
 crash is modified and lost in a roar of reverberation, and is not 
 heard until some seconds, more or less, after the lightning flash ; 
 so that by noting the time which elapses between the flash and the 
 sound, we may, as in the case of cannon, estimate the distance : 
 the more immediate is the crash upon the light, the nearer is the 
 discharge. 
 
 The distance of air through which disruptive discharge can 
 break is directly as the electrical disturbance ; or, to use other 
 but more figurative language, directly as the quantity of electri- 
 city accumulated between the terminating planes of the action 
 viz., the clouds and earth or artificially between the coatings of 
 a fulminating square of glass, or the Leyden jar. The following 
 experiment will pretty fairly demonstrate this. 
 
 Exp. 16. If two smooth balls be opposed to each other, one of 
 them connected with the rod of a charged jar, the other 
 with its outer coating, and measured quantities of electricity 
 be accumulated, the jar will discharge by breaking through 
 the air intervening between the balls ; and the distances 
 between the two balls will measure the accumulated quantity, 
 or verv nearly. 
 
290 SECOND LECTURE. 
 
 We saw in a former experiment that the attractive force of the 
 accumulation, as measured by a delicate balance, was as the square 
 of the quantity of electricity ; and we have to reconcile these 
 apparently opposite results, and show how it is that whilst the 
 attractive force is as the quantity squared, the explosive force or 
 distance is as the quantity simply. A very little examination 
 will suffice to elucidate this. You will immediately perceive, by 
 reference to this diagram, that since the force increases with the 
 quantity squared directly, and decreases with the square of the 
 distance inversely, the force for explosion must remain the same 
 at all distances, and must be as great as is requisite to overcome 
 the atmospheric resistance, which is a constant quantity. 
 
 Imagine discharge to occur at distance D, between two balls, 
 p, N, with a given quantity of electricity, which we will call 1 or 
 unity ; then whilst undischarged, or before discharge takes place, 
 let the receiving ball N be removed to o, or at distance D 2 from p. 
 In this case the force between P o will be only one-fourth of what it 
 
 was before, since the force is as . Discharge could not, therefore, 
 
 D 2 
 
 take place at this distance, p o, with the unit of quantity. But 
 now suppose we double the quantity and make it 2, or twice as 
 great ; then, because the force increases as the square of the quan- 
 tity directly, the force in p o becomes four times as great, and is 
 hence from its reduced state of one-fourth raised up to one that is, 
 to what it was before and now the explosive discharge again 
 ensues, so that what we lose by distance we gain by increased 
 quantity. 
 
 I have now to call attention to certain laws of explosive elec- 
 trical discharge very important to keep in view in all our 
 arrangements for obviating the destructive effects of lightning. 
 On examining numerous instances in which ships and buildings 
 have been struck and damaged by lightning, we find the electrical 
 discharge always pursuing a determinate line or lines, which upon 
 the whole present the least resistance to the neutralisation of the 
 two opposite electrical forces. Both space and time appear to be 
 economised as it were in its progress. For however small we 
 assume the duration of the thunder- stroke, or however limited the 
 distance through which the lightning strikes, both these quantities 
 would become still less were other lines of transit provided 
 opposing still less resistance. This is the leading character of 
 electrical disruptive discharge. Hence a stroke of lightning seizes 
 upon bodies which lie convenient and ready for its progress, abso- 
 lutely avoiding other bodies, however near, from which it can 
 
POLARIZED AIR. 291 
 
 receive no assistance, as I have already remarked ; and I must again 
 here observe, as a most marvellous and interesting fact, that at the 
 instant before electrical explosive discharge occurs, the explosive 
 stream, in the act of moving between the opposed planes of action 
 to a condition of neutrality, feels its way as it were in advance, 
 and marks out by a species of foresight the course it is about to 
 take. The course of lightning is not, as many imagine, left to the 
 chances of the instant, to be drawn aside this way or that by 
 adjacent metallic or other matter. On the contrary, the course of 
 a thunder-stroke is already fixed and absolutely settled before the 
 lightning appears and the actual discharge takes place. The fol- 
 lowing experiments will serve to illustrate this selection of a least 
 resisting course, and how electrical discharge seizes upon some 
 bodies and avoids others, according as they lie in positions favour- 
 able or unfavourable to its progress. 
 
 Exp. 16 and Exp. 17. Disjointed gold on papers, and also 
 illustrating ships' conductors. 
 
 These separated fragments of gold leaf may be taken to repre- 
 sent detached perfect or imperfect conducting masses anyhow 
 placed in a building or ship : the course of a discharge of lightning 
 through them will be found to obtain on precisely the same 
 principles. 
 
 THE END, 
 
 VIR1UE AM) CO., PE1STEJRS, CllV ROAD, LoM>UX 
 
14 DAY USE 
 
 RETURN TO DESK FROM WHICH BORROWED 
 
 LOAN DEPT. 
 
 This book is due on the last date stamped below, 
 or on the date to which renewed. Renewals only: 
 
 Tel. No. 642-3405 
 
 Renewals may be made 4 days priod to date due. 
 Renewed books are subject to immed^a^ tysall. 
 
 BtU 
 
 LD21A-60m-8,'70 
 (N8837slO)476 A-32 
 
 General Library 
 
 University of California 
 
 Berkeley 
 
 REC'Q I 
 
 LD 21-100m-7,'40 (6936s) 
 
 V 
 
UNIVERSITY OF CALIFORNIA LIBRARY 
 
If 
 
 s 
 
 i>; <*