310 
hypothesis even in the case of liquids. Prof. Thomson has 
shown that the phenomena depend on the dryness of the gas, 
so that the conduction cannot depend on the gaseous molecule 
alone. In the case of conduction induced by a neighbouring 
discharge, this might be due to the expulsion of condensed 
vapour from the walls of the vessel. It would appear that in 
the dry state gases are not electrolytes. Mr. Enright said he 
thought it was not correct to say no work was done in electro- 
lysis. Prof. Silvanus Thompson said that the pursuit of the 
analogy between the conductivity in gases and liquids was apt 
to lead one too far. Thus, if you compare the conduction in a 
mixture of H and Cl with electrolysis, your analogy will be a 
false one unless you import into the term electrolysis the idea of 
chemical separation as taking place in the solution. If a current 
separated a mixture of powdered zinc and sulphur, it could not 
be called a case of electrolysis. Prof. Armstrong said an ex- 
periment of Prof. Dewar’s was very instructive. He had shown 
that if you cool the surface of a Crookes’ tube the discharge 
stops. It was quite inconceivable that at these low pressures 
the gas became liquefied, so that this experiment seemed to show 
that conductivity depends on the presence of a vapourous elec- 
trolyte. Mr. Enright asked if Prof. Armstrong knew how 
the presence of an electrolyte assisted conduction. In acom- 
munication, Prof. J. J. Thomson said that, in the decomposition 
-of steam bya spark, the fact that in the tube as a who/e the 
amount of steam decomposed is greater than the amount of gases 
liberated in a voltameter in series, was no objection to the con- 
ductivity being electrolytic. The only condition imposed by the 
laws of electrolysis was that the excess of H or O at one ter- 
minal, and of O or H at the other, should correspond to the 
amount of electricity passing through the tube. Thus, suppose 
in a water voltameter a number of metal partitions are fixed so 
that the current has to pass across these plates. Then at each 
_plate H will be given off on one side and O on the other, and by 
making the partitions sufficiently numerous, the total quantity of 
gases given off for the passage of a given current may be 
made as large as we please. The excess at the terminals 
would not be affected at all by these partitions. In the ex- 
periments made by Mr. Rutherford and himself (Prof. Thomson), 
they did not observe any polarisation when the conductivity was 
produced by Rontgen rays. With reference to Mr. Baly’s ob- 
jections to the electrolytic theory: (1) There is no reason to 
think that, under conditions other than in solution, the atom 
of hydrogen may not have a negative charge. (2) The electro- 
lytic theory leads us to expect that it would require a finite 
electromotive force to send a discharge through a gas. Before 
such a discharge can take place, the molecules must be split up, 
and this requires an electric field of finite strength. (3) In the 
case of a gas, the electric field has to ionise the molecules, so 
that an increase in the strength of the field will not only (as in 
the case of a liquid electrolyte) increase the speed of the ions, 
but it will also increase their number, and thus the current will 
increase faster than the electromotive force. (4) The ion once 
used can again combine, and, since the ionisation is done by 
the electric field, it can be again split up and used again. If, 
however, the ionisation has been: done by external sources—as, 
for example, by Réntgen rays—then we find that the conduc- 
tivity decreases as the current passes. (5) There seems to be 
no reason on the electrolytic theory why, in a mixture of Hcl 
and Cl, some of the current should not go through the chlorine. 
(6) A variable potential gradient would be produced if the ions 
moved with different velocities. Mr. Baly’s process in the 
,positive column appears to be the same as on the electrolytic 
theory minus the atomic charges. In a communication Prof. 
Schuster said : Mr. Baly criticises what he calls the electrolytic 
theory, but directs his arguments against a form of the theory 
which is, as far as the writer knows, advocated by no one. Mr. 
Baly appears not to have read the original papers in which the 
fundamental points of the theory, upheld by J. J. Thomson and 
the writer (Prof. Schuster), are explained. If he had done so, 
he could not have given, as an objection to the theory, that the 
conductivity of a gas increases with the E.M.F. The essential 
difference between a liquid and a gas is that in the liquid the 
number of ions is fixed by the chemical constitution of the liquid, 
while in a gas dissociation has, first of all, to be produced by 
the current itself, and hence the number of ions depends on the 
current. In the paper referred to by Mr. Baly, in which the 
fact that when a spark is passed through a gas the gas ceases to 
insulate for some distance round the spark is described, the 
explanation that this was due to a difficulty of passage of the 
NO. 1422, VOL. 55 | 
NATORE 
[January 28, 1897 
electricity from the electrode into the gas was especially dis- 
claimed. The explanation given being substantially the same 
as that now given by Mr. Baly. Mr. Baly asks what becomes 
of the ions that are set free? The answer, of course, is 
that they recombine. The view that stratifications are 
due to compound molecules, and do not probably occur 
in pure gases is not new. With reference to the author’s 
statement that ‘f measurements made by Wheatstone and J. J. 
Thomson prove that the electricity travels along the positive 
column from the anode to the kathode, and that its velocity is 
about half that of light,” Prof. Thomson’s results show that the 
break-down of the insulating power of air takes place in the 
manner described, but this does not show anything as to what 
happens when the discharge has reached the steady state. Mr. 
Baly is quite wrong in the excess charges he assigns to different 
parts of the vacuum tube. Experiments on the excess charges 
can count for nothing, unless they are done with continuous 
currents. Mr. Baly is further wrong in stating that the fall of 
potential is rapid in the glow. On the contrary it is 
very small in the glow, being very rapid in the dark 
space between the glow and the kathode. Mr. Baly adopts 
Prof, Thomson’s view as to the formation of molecular chains, 
but in a form very difficult to accept. The whole foundation of 
Mr. Baly’s theory is upset by his wrong assumptions as to the 
excess charges in different parts of thetube. The author, in his 
reply, said that on some points he had been misunderstood. He 
thought that the increase in conductivity could not be due to 
vapour driven off from the sides, for ultra-violet light also pro- 
duced such an increase. If Réntgen rays produce ionisation, 
then there ought to be a reduction in the density of the gas. 
Chemical Society, December 17, 1896.— Mr. A. G. Vernon 
Harcourt, President, in the chair.—The following papers were 
read :—On the experimental methods employed in the examina- 
tion of the products of starch-hydrolysis by diastase, by He. 
Brown, G. H. Morris, and J. H. Millar. From the results of 
a large amount of experimental work on starch-hydrolysis, the 
authors draw conclusions respecting the determination of solids 
from solution-density, the relation of [a]; to [a}p, the determina- 
tion of cupric reducing power, and discuss the limits of accuracy 
of the various methods.—On the specific rotation of maltose 
and of soluble starch, by H. T. Brown, G. H. Morris, and 
J. H. Millar. The specific rotation of 2 to 20 per cent. pure 
maltose solutions is constant, and at 15°5°, [¢l> = 137°93 5 
soluble starch in 2°5 to 4°5 per cent. solutions has, at 15°5°, 
[a], = 202'0°.—On the relation of the specific rotatory and 
cupric reducing powers of the products of starch-hydrolysis by 
diastase, by H. T. Brown, G. H. Morris, and J. H. Millar. 
The authors have established a definite relation between the 
specific rotation and the cupric reducing power of the products 
of starch-hydrolysis by diastase, which holds within very narrow 
limits.—The action of hydrogen peroxide and other oxidising 
agents on cobaltous salts in presence of alkali bicarbonates, by 
R. G. Durrant. Cobaltous solutions are turned green by 
hydrogen peroxide, hypochlorite, bromine, chlorine, or ozone 
in presence of alkali bicarbonates ; the green colour is dependent 
on the production of a cobaltic salt and on the presence of 
carbonic anhydride.—Electrical conductivity of diethylam- 
monium chloride in aqueous alcohol, by J. Walker and F. J. 
Hambly.—Formation of substituted oxytriazoles from phenyl- 
semicarbazide, by G. Young and H. Annable. Substituted 
oxytriazoles are obtained when mixtures of phenylsemicarbazide 
with benzaldehyde, meta- or para-nitrobenzaldehyde. metatoluic 
aldehyde, terephthalic aldehyde, or cinnamic aldehyde are 
oxidised, — a-Bromocamphorsulpholactone, by C. Revis and 
F. S. Kipping. Under certain conditions an a-bromocamphor- 
sulpholactone, Cy )HygBrSOs, is formed during the sulphonation 
of a-bromocamphor.—Dimethylketohexamethylene, by F. S. 
Kipping.—The localisation of deliquescence in chloral hydrate 
crystals, by W. J. Pope. Great differences have been observed 
between the speeds of deliquescence of the various forms present 
on crystals of chloral hydrate.—Enantiomorphism, by W. J. 
Pope and F. S. Kipping. A preponderance of either right- or 
left-handed crystals of sodium chlorate is deposited on crystallis- 
ing the material from aqueous solutions containing various 
optically actiye substances. 
Linnean Society, December 17, 1896.—Dr. A. Giinther. 
F.R.S., President, in the chair.—Messrs. James Green and J. 
H. Gardiner exhibited a series of sciagraphs of British batra- 
chians and reptiles in which the details of the skeleton were very 
