August 1 6, 1S83] 



NA TURE 



?8i 



during the last few year?, devoted much attention to the meteoro- 

 logy of Siberia, and, besides the meteorological observations 

 made at its stations, it has collected materials for ascertaining 

 the dates of the freezing and breaking up of the ice in Siberian 

 rivers. The li^t of these dates for the rivers of Siberia for the 

 jears 1874 to 18S0 will certainly he consulted with profit, as 

 also several brief notes on amber in Siberia, on chemical analyses 

 of salt from various salt lakes, and of coral from the Nerchinsk 

 district, and from the banks of the Amur, as also other smaller 

 notes. 



EXPERIMENTAL RESEARCHES ON THE 

 ELECTRIC DISCHARGE WITH THE CHLO- 

 RIDE OF SILVER BATTERY 1 



' r HE authors recall that at the conclusion of the third part of 

 1 their researches (Phil. Trans, for June 1 1, Part I. vol. clxxi. ) 

 they stated that they intended to make an investigation on the 

 dark discharge, and the special conditions of the negative dis- 

 charge ; this paper contains a number of experiments, more 

 especially on the latter subject, and also others intended to 

 throw light on the general nature of the electric discharge 

 through gases. 



The first part of the paper de^ribes some experiments mnde 

 with vessels of different forms in order to ascertain whether the 

 dimensions and shape of the vessel have any effect on the 

 pressure of minimum resistance to the electric discharge. This 

 was found to be the case ; for example, with a residual air charge 

 in a spheroidal vessel 7 inches ( 1 7"8 centims.) long, and 5 inches 

 (127 centims.) diameter (Fig. I), the pressure of minimum 

 resistance was as high as 3 millims. , 3947 M ; while in a tube 

 22"5 inches (57 centims.) long, and 1-625 inches (4-1 centims.) 

 diameter, it was only 069 millim., 908 M ; again in a smaller 

 tube 23 inches (58'4 centims.) long, and 075 inch (1/9 centims.) 

 diameter, it was 1 millim., 1316 M. It is evident, therefore, 

 that not only the dimensions of the tube, but possibly also the 

 shape of the terminals, have an influence on the pressure of 

 least resistance, and it is very probable that in the atmosphere, 

 where lateral expansion is practically unlimited, the conditions 

 of minimum resistance are different from those which exist even 



r> 0== 



given off to form a new mirror -like coating with a fresh dis 

 charge ; this properly has continued since March, 1875. 



The paper next describes experiments to ascertain the length 

 of the spark in dry air and in air saturated with moisture. It 

 was found to be practically the same in both cases. With 10,860 

 cells the mean length of the spark between two paraboloidal 

 points was found to be in dry air 045 inch (I'l centims. ), in 

 moist air 0447 inch (i"i centims.). 



The next subject taken up is the discharge in a tube from two 

 batteries, first in the same and then in contrary directions. In 

 the tube are two terminals at each end, one pair at opposite ends 

 being inclosed in two short pieces of tube 9 inches (22'S cen- 

 tims.) long and h inch (1 '27 centims. ) diameter; the main tube 

 being 31 inches (95-2 centims.) long and iij inch (4'4 centims.) 

 diameter. The various phases of the stratified discharge are 

 represented in an engraved mezzotint steel plate copied from 

 photographs, and show the effect of the one stratified discharge on 

 another stratified discharge produced by a second battery. It is 

 seen that two discharges in contrary directions may take place in 

 the same tube, and that the one modifies the aspect of the other. 



Experiments are also described in a tube in the form of a cross 

 with four arms at right angles (Fig. 4), with two separate bat- 

 teries connected in various ways with the different members. 



Fig. i. 



in very large tubes, and that this may influence the height of the 

 aurora. 



The paper next deals with the discharge in miniature tubes 

 I inch (2-2 centims.) long, and £ inch (063 centim.) diameter, 

 with terminals nearly touching ; at first it required 2400 cells to 

 pass, then a single cell would do so, but after standing a short 

 time it required 4800 cells to reproduce a discharge. In 

 another tube ij inch (4-4 centims.) long and | inch (0-95 

 centim.) diameter), with the terminals distant 0-00104 inch 

 (0-0264 millim.), it required 2240 cells to produce a discharge, 

 then the potential had to be increased to 11,240 cells to do so. 

 Ultimately even this number failed, but after the tube had lain 

 by for some days 600 cells could pass. It is very possible that 

 the strong discharge in the first instance volatilised a portion 

 of the terminals which were of platinum, and that this volati- 

 lised metal condensed afterwards, or else that the terminals ab- 

 sorbed the gas so completely as to produce a vacuum too perfect 

 to admit of a discharge taking place; and that uhimately suf- 

 ficient of the occluded gas was again given off to render it again 

 possible. 



In connection with the occlusion of gas by terminals a case is 

 described in which the terminals are of palladium and the charge 

 hydrogen (Fig. 2). After a few discharges the terminals oc- 

 cluded some of the gas, and when a fresh one was produced a 

 volatile compound of hydrogen and palladium was given off, 

 especially from the negative, and produced a dense, mirror-like 

 coating on the inside of the tube (Fig. 3) ; this was reoccluded 

 by standing for a couple of days, leaving the tube free, and again 



1 Abstract of a paper read at the Royal Society on June 14. by Warren 

 De La Rue, M.A., D.C.L., F.R.S., and Hugo W. Muller, Ph.D., F.R.S. 



The experiments were made both in air and in hydrogen. By 

 the introduction of external resistance of one of the batteries, the 

 discharge could be readily identified as belonging to that battery 

 by the effect of the resistance on the character of the stratifica- 

 tion. In one of the mezzotint plates are several figures copied 

 from photographs which show clearly the phenomena produced. 

 Calling the poles P and N of one battery, A, and p' and N' of 

 the other, B, it is shown in one case when two currents were 

 equal 0-0083 ampere, that a discharge from A battery goes from 

 p in the direction of N only so far as the junction at the cross, 

 and then turns off to n', the negative of the other battery B ; 

 while, on the other hand, the discharge of the B battery goes 

 from p' to N of the A battery. The case is different if an external 

 resistance is introduced in one of the discharges, reducing it to 

 o 00087 ampere, then the discharge of the A battery goes from P 

 to N, and that of the B battery from p'to N'. There is a bending 

 down, however, of the strata of the weaker discharge of the 

 cross junction, in consequence of the action of the stronger one. 

 The authors remark that one cannot but be impressed, from 

 the experiments described in the paper, and others in their 

 former papers, by the apparent plasticity of the aggregate assem- 



Fig. 3. 



blage of molecules constituting a stratum which yields to 

 external influences that modify its form. 



The authors describe and figure a case of complex strata in 

 the form of an outer bracket convex towards the negative (Fig. 5), 

 and close to it an inner chord ; also discharges in various gases 

 in tubes of large dimensions, 37 inches (94 centims.) long, and 

 5J4 inches (14-8 centims.) diameter. In these the stratification, 

 which is comparatively narrow at the terminals, extends in a 

 conical form from the terminals to the full diameter of the tube. 



They have found that the dark space in the discharge in 

 vacuum tubes is only relatively actinically dark in comparison 

 with a stratum, and they succeeded in obtaining a photograph 

 of the dark space in thirty-five minutes as strong as that from a 

 stratum in two and a half seconds ; consequently they conclude 

 that the dark space is 840 times less actinically bright than a 

 stratum. 



The authors next describe a number of experiments, by means 

 of a Thomas-Becker electrometer used on a method, to avoid 

 leakage, proposed to them by Prof. Stokes, to ascertain the 

 difference of potential in different parts of a vacuum tube having 

 a number of rings sealed within it, also in other tubes of special 

 construction. These bring out instructive information, in refer- 

 ence not only to the relative resistances of different lengths of a 



