Oct. 6, 1881] 



NATURE 



549 



electricity to traverse it again. But the distance between the 

 terminals being very small, the electricity coming from the ma- 

 chine scon atlains the requisite quantity and tension, and the 

 discharge is r^ieated before the tacilities due to the preceding 

 di-charge are lost. The shortness of the interval between 

 the termmals consequently acts in a double manner to facditate 

 the discharge, ajid thus renders the transition fro u one form to 

 the other more rapid than it would otherwise have been. 



Observations with the spectroscope indicate that heating here 

 takes place, and the revolving mirror shows that either in the 

 discharge itself, or in the beating effects due to it, there is a 

 manifest duration, all of which tend in the direciion of the 

 explanation suggested above. 



The character of the discharge from the induction-coil both 

 when the terminals are widely separated and when they are near 



together, is generally similar to that from the machine : but the 

 durational character of the former is very much more marked 

 than that of the latter; so much so, in fact, that with large coils 

 the duration extends over a fraction of a second, perfectly appre- 

 ciable by the eye without any auxiliary apparatus. This is due 

 to the nature of the instrument, and is dependent both upon the 

 time occupied by the core in loving its magnetism, and also 

 upon the mutual induction of the convolutions of the secondary 

 coil. The flame which accompanies the spark proper is the part 

 of the discharge which persists; and it will have been noticed 

 particularly when the coil is excited by thcDe Meritens' machine. 

 The discharge produced from the secondary through the instru- 

 mentality of this machine is so remarkable, that it has been 

 considered worth a special study. It has al-o been of oreat 

 assistance in the examination of the action of a magnet upon a 

 discharge ; but the results of the latter experiments have not yet 

 been published. 



The form of discharge which we have now reached is sub- 

 stantially that which is known as the "arc," or comparatively 

 quiet and continuous discharge between two terminals near to 

 one another. 



Turning to the arc, let us take the form most familiar to our 

 mind.s, viz. that used in electric lighting. I now project on the 

 screen an image of the arc as u^ed in what are called "arc 

 lamps." The whole consists essentially of two rods of carbon 

 placed end to end, with a short interval between them. The 

 interval is of a length capable of being traversed by the current, 

 at all events after the discharge has been once established. }'y 

 the passage of the current, which, in fact, constitutes the arc, 

 the carbon becomes heated to a high degree. And it is im- 

 portant to understand that the main source of the light is to be 

 found, not in the arc proper, but in the heated carbons. It will 

 be noticed that, when a machine giving direct currents is used, 

 the two carbons are not equally heated, and that during the 

 combustion they acquu-e dissimilar configurations. This dis- 

 symmeti7 at the terminals is found to obtain in almost every 

 species of electrical disch.arge. 



With the construction and outcome of the various machines 

 employed for producing the current, and with the mechanical 

 contrivances used for maintaining the arc at its proper Itngth 

 and in its proper position, we are not here concerned. All that 

 need be here mentioned is that the carbon which would be con- 

 nected with the copper element of a Grove battery, if such were 

 used, and which is called the positive, is the one move rapidly 

 consumed. It becomes hollowed out, and incandescent particles 

 may be seen occasionally traversing the arc, and landing upon 

 the second or negative carbon. In the meantime the arc proper 

 flows steadily between the carbons, the colour being determined 

 by the nature of the terminals, or by that of any substance 

 placed on their ends ; and partly also by the nature of the gas 

 in which the discharge takes place. 



Let us now regard the terminals merely as parts of our 

 apparatus, subsidiary to the main purpose, and fix our attention 

 almost exclusively on the arc itself. If we had been working in 



^Miiiiitiiiiiiiliittiitiiiit 



the laboratory, I should have asked you to examine, with the 

 aid of a microscope, the minute structure or anatomy of the arc. 

 As it is, I mu-t be? you to accept as a substitute for the pheno- 

 menon itself the following series of photographs, for which we 

 are indebted to the skill and kindness of Mr. De La Rue, who 

 has done so much with his unrivalled battei-y in this field of 

 research. 



Figs. I to 5 are, in fact, magnified representations of the dis- 

 charge through air a* different pressures, beginning with that of 

 the atmosphere, and extending in a series of decreasing pressure 

 to about one 300th part of it. In Fig. 6 the pressure has been 

 reduced to al'out a 2000th part of an atmosphere. In all these in- 

 s'ances it will be noticed that there is a tendency on the part of 



the luminosity to break up into disconnected blocks, and that at 

 an early stage it begins to separate from the negative, and to cling 

 to the positive terminal. Also, that when the pressure is con- 

 siderably reduced, t ese blocks are replaced by the beautifnl 

 system of flakes or "stria:" delineated in the last figure of the 

 series. At this stage the dyssymmetry on which I have already 

 insisted is complete. 



The actual length of the discharges ot which you have just 

 seen the representations, varies in a tolerably regular manner 

 with the pressure, from half an inch to ten inches or more. Frorr 

 this we may gather the important fact that in the discharge 

 through gases at low pressures we have a magnified image of the 

 discharge at higher pressures. By this statement it is not of 



