62 



SCIENTIFIC NEWS. 



[July 20, 1888. 



the patent in question, ever appear to have been 

 sold, or indeed were ever made in this country 

 except for the purpose of evidence in litigation. To make 

 them was no easy task. Professor Crookes, whose skill 

 in delicate manipulations is as great as his reputation as 

 a chemist, was no more successful in his attempts than 

 were Professor Silvanus P. Thompson, Dr. Tidy, and 

 others. Filaments we»e reduced to ashes, putty of tar 

 and lamp-black refused lo be rolled out to a thinner 

 thread than one twenty-fifth of an inch, and lamps which 

 with the greatest difficulty were at length made and 

 tried, broke down after a minute or two. Mr. Giming- 

 ham, for the plaintiffs, however, seemed to have little 

 difficulty in making any number of the lamps. Professor 

 Dewar rolled out a fine filament in the witness-box from 

 a "tar putty" which he brought with him. With such 

 conflicting evidence, the judge ordered that the experi- 

 ments should be repeated by two experts from each side, 

 in the presence of counsel and of an independent " mode- 

 rator." Professor Stokes was selected for this office, and 

 he sent an extremely interesting report to the judge. It 

 turned out that this "putty" must be kneaded and 

 worked with a considerable expenditure of force for 

 about an hour and a half. It then assumes the consis- 

 tency of gutta-percha, and can be rolled out. But of 

 one hundred lamps made of this substance some broke 

 down as soon as the electric current was turned on. 

 Eleven lasted less than twenty hours, eighteen could not 

 stand forty hours' run, and after sixty hours only four 

 survived. 



In a most masterly judgment, in which Justice Kay 

 showed that he had made himself thoroughly conversant 

 both with the theory of the subject and with the most 

 minute details, he decided that Edison's patent was void 

 for six reasons. First, that the claims were too wide. 

 Second, that it does not describe a lamp which has met 

 with a commercial success. Third, that there is an in- 

 sufficient description of the processes of manufacture. 

 Fourth, that in one of the processes described, by which 

 volatile substances, such as camphor, are mixed with the 

 " tar putty," the presence of such substances would be 

 injurious. Fifth, that the use of certain " nonconducting 

 noncarbonisable substances," which it was said might be 

 used as a coating to the filaments, were of no utility. 

 And sixthly, that the coiling ol the filaments as described 

 by Edison, was also of no utility. Judgment was there- 

 fore given for the defendants with costs. 



The invention of Sawyer and Mann is for a most in- 

 genious and beautiful process of depositing carbon upon 

 a filament of an electric lamp. The bulb is filled with a 

 hydrocarbon gas or liquid ; coal gas is often used, and 

 the filament is " flashed " by passing a current through 

 it. If one part be thinner than another, this will offer 

 more resistance to the passage of the current than will 

 the thicker parts. The temperature will therefore be 

 greater, and the heat will decompose the hydrocarbon, 

 the carbon of which is deposited on the heated part. It 

 is clear that this process is perfectly automatic, and the 

 result is a great uniformity throughout the filament, 

 each part glowing with the same brilliancy. The de- 

 fendants say that their filaments are so regular that it 

 is very rarely that they could be improved as fur as uni- 

 formity is concerned, but that they use this process to 

 deposit a coat of carbon on the filaments in order that 

 their resistance may be made the same, so that a number 

 of lamps will give the same light. They admitted that 

 if there were by accident any flaw in a filament, this 



" flashing " would remedy it. They endeavoured to show 

 that the invention had been anticipated by the French 

 chemist Depretz, who, in trying to fuse carbon by the 

 passage of a strong current of electricity, surrounded it 

 with a hydrocarbon gas, and found that it increased in 

 size. Not only was this discovery of no apparent use, 

 but it interfered with Depretz's experiment. The 

 patent was therefore upheld, and an injunction was 

 given restraining the defendants from using the process, 

 and they were ordered to pay the proportion of costs. 



This case, which has excited a considerable amount ot 

 interest, thus throws open the manufacture of incandes- 

 cent lamps, as far as the elementary use of a high resist- 

 ance carbon filament in a vacuum is concerned, but 

 there are a considerable number of patents for such 

 details as the " flashing " process, which will probably 

 render it improbable that the fancy price of 5 s. each will 

 be at once reduced to a more reasonable sum. 



At an early stage of the case, Mr. Percy Sellon gave a 

 full account of the process of manufacture of the lamps 

 as made by the Anglo-American Brush Company. The 

 material from which the filament is prepared is cellulose 

 in an extremely pure condition. It is formed by dis- 

 solving ordinary cotton wool (which consists almost 

 entirely of cellulose), in a solution of chloride of zinc. 

 By the application of heat this produces a viscous semi- 

 liquid mass, the fibres and cellular structure being- 

 destroyed, as in the materials called "vulcanised fibre" 

 and " Willesden paper." The material, which is then not 

 unlike a strong solution of gum arabic, is then forced 

 through a small hole or die by the pressure of a column 

 of mercury. It is received as it issues from the hole in 

 alcohol, which both stiffens or congeals the thread, and 

 removes any chloride of zinc that may be present. It is 

 then left to soak in another jar of alcohol, and is allowed 

 to dry. It has the 'appearance of fine gut as used for 

 fishing. It is wound on wooden drums covered with 

 velvet, which yields to the thread as it contracts 

 while drying. It is then wound on a block of plum- 

 bago, provided with a piece of cardboard at one end, and 

 the whole is placed in a plumbago crucible, and is 

 packed with powdered plumbago, a little paraffin wax 

 being placed at the bottom of the crucible. It is then 

 closed and placed in a furnace, and gradually raised to a 

 white heat, which is attained after about six hours. The 

 paraffin vaporises and drives off the air contained be- 

 tween the interstices of the plumbago, and the piece ot 

 cardboard consumes, and allows the filaments to con- 

 tract. The temperature is raised to a fierce heat for 

 some hours, the total period in the furnace being about 

 eleven to fourteen hours. 



The crucible with its contents are allowed to cool, and 

 the filaments are taken out and separated. They are 

 " flashed " as has been described, the resistance being 

 measured between the flashings, which are repeated 

 until the right standard is attained. They are then per- 

 manently mounted on the platinum wires, which are 

 coiled at one end into a spiral, into which the carbon fila- 

 ment is pushed. A local flashing is then applied, the 

 current being diverted from the filament itself by a 

 bridge of wire. The junction alone, which offers some 

 little resistance, becomes heated, and the deposit of car- 

 bon takes place, sealing the filament to the wire. The 

 wires are then sealed into the glass bulb, and the air is 

 exhausted, the lamp being subjected to a current during 

 the latter part of the operation, thus driving off any gas 

 which is " occluded " by the carbon or the platinum. 



