188 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1909. 
increase in mass due to the charge becomes not merely appreciable, 
but so great that, as the experiments of Kaufmann and Bucherer 
have shown, the whole of the mass of the corpuscle arises from its 
charge. 
We know a great deal about negative electricity; what do we know 
about positive electricity? Is positive electricity molecular in struc- 
ture? Is it made up into units, each unit carrying a charge equal in 
magnitude though opposite in sign to that carried by a corpuscle? 
Does, or does not, this unit differ, in size and physical properties, 
very widely from the corpuscles? We know that by suitable proc- 
esses we can get corpuscles out of any kind of matter, and that the 
corpuscles will be the same from whatever source they may be de- 
rived. Isa similar thing true for positive electricity? Can we get, 
for example, a positive unit from oxygen of the same kind as that we 
get from hydrogen ? 
For my own part, I think the evidence is in favor of the view that 
we can, although the nature of the unit of positive electricity makes 
the proof much more difficult than for the negative unit. 
In the first place we find that the positive particles—“ canal- 
strahlen ” is their technical name—discovered by our distinguished 
guest, Dr. Goldstein, which are found when an electric discharge 
passes through a highly rarefied gas, are, when the pressure is very 
low, the same, whatever may have been the gas in the vessel to begin 
with. If we pump out the gas until the pressure is too low to allow 
the discharge to pass, and then introduce a small quantity of gas and 
restart the discharge, the positive particles are the same whatever 
kind of gas may have been introduced. 
I have, for example, put into the exhausted vessel oxygen, argon, 
helium, the vapor of carbon tetrachloride, none of which contain 
hydrogen, and found the positive particles to be the same as when 
hydrogen was introduced. 
Some experiments made lately by Wellisch, in the Cavendish Lab- 
oratory, strongly support the view that there is a definite unit of 
positive electricity independent of the gas from which it is derived ; 
these experiments were on the velocity with which positive particles 
move through mixed gases. If we have a mixture of methyliodide 
and hydrogen exposed to Réntgen rays, the effect of the rays on the 
methyliodide is so much greater than on the hydrogen that, even 
when the mixture contains only a small percentage of methyliodide, 
practically all the electricity comes from this gas, and not from the 
hydrogen. 
Now, if the positive particles were merely the residue left when a 
corpuscle had been abstracted from the methyliodide, these particles 
would have the dimensions of a molecule of methyliodide; this is 
very large and heavy, and would therefore move more slowly 
