606 
ALU RE 
[APRIL 29, 1897 
On these bases the following table is calculated :— 
Excess of the mean stature of the pedigree breed tn each successive 
generation, above that of the rest of the population, when 
both parents occupy the wndermentioned grades in their own 
generation of the pedigree-breed, 
| 
Generation. 99 |. 95 
inches, | inches. | inches. | inches, | inches. 
I 3 2 2:2) || Sr4ey'o%9 
2 61 43 3A ene: 1°4 
3 75) 5°34 ee a7 ate: 
4 8:4) G10.) 4°71) eget 
5 OT) 6:4 | SO Sasa aret 
&e. ; a &e. | &c: | &e. | &e. | &e. 
Limiting values LOWS |: °7°3'=|) 5s 7alleeens | 2°3 
The importance to the breeder, of using highly selected 
parents, is #zeaseved by these tables, and shown to be very great. 
Thus one generation of the 99° selection is seen to be more effec- 
tive than two generations of the 90° selection, and to have about 
equal effects with those of an So” selection carried on to per- 
petuity. Two generations of the 99° selection are more effective 
than four of the 95°, and than a perpetuity of the go”. 
It must be borne in mind, that there is no stability in a breed 
improved under the supposed conditions ; but that, as soon as 
selection ceases it will regress to the level of the rest of the 
population through stages in which the deviation at starting, 
sinks successively to w, w?. . . w” of its value. It may, how- 
ever, happen that a stable form will arise during the process of 
high breeding, that shall afford a secondary focus of regression, 
and become the dominant one, if the ancestral qualities that 
interfere with it be eliminated by sustained isolation and selec- 
tion. Then a new variety would, as I conceive, arise ; but into 
this disputable topic there is no need to enter now. 
We can thus understand the facility with which races of 
butterflies acquire mimetic forms, the severity of selection in 
their case being very great, while one of their generations 
occupies only a year. FRANCIS GALTON. 
The Effect of Rontgen Rays on Liquid and Solid 
Insulators, 
OWING to my absence from Cambridge in the Easter vacation, 
I have not until to-day seen the paper by Lord Kelvin, Dr. 
Beattie and Dr. M. Smolan (NATURE, March 25), on the 
influence of Rontgen rays on electric conduction through air, 
paraffin, and glass, in which the authors state that they cannot 
detect any influence of Rontgen radiation on conduction through 
solids. I think that the difference between this result and the 
one obtained by Mr. McClelland and myself arises from the 
temporary character of the effect of the radiation on solids. The 
increase in the conductivity of solids is only appreciable for a 
short time after the application of the electric force (see NATURE, 
July 30, 1896, p. 306); under long-continued electromotive 
forces the conductivity seems unaffected by the rays. The effect 
might perhaps be more accurrately described as an increase in 
the electric absorption, rather than as an increase in conduct- 
ivity. I have been for the past few months engaged in experi- 
ments on the effect of the rays on solids and liquids, particularly 
_liquids ; and, though the experiments have been much inter- 
rupted by the pressure of other work, I hope soon to have 
them ready for publication. There is one experiment, however, 
which may be of interest. Of all the liquids tried, that sold as 
vaseline oil has proved the best insulator ; in its pure state it is 
very transparent to Rontgen rays, so to increase the absorption 
of these rays I stained the oil with iodine, when it became very 
opaque to them. The oil does not insulate so well after stain- 
ing as it did before, but the effect of a slight amount of conduct- 
ivity is not of importance when the following method is used. 
Three electrodes, A, B, C, are placed in a leaden vessel filled 
with the oil. B, which is between A and C, is connected to one 
pair of quadrants of an electrometer, A and C to the terminals 
of a battery of 1000 small storage cells. If there is any leakage 
the potential of B will, in general, not remain zero after the 
battery is put on, but it will do so if an earth connection is 
made at the proper place in the battery. The base of the vessel 
below B C was cut out, and an aluminium vessel inserted, so 
NO. 1435, VOL. 55] 
that the liquid between B and C could be exposed to the Réntgen 
rays. A balance was obtained with the rays off; when the 
rays were turned on, the potential of B no longer remained zero, 
but changed in the way it would if the conductivity between B: 
and C had increased. This effect was small but well marked, 
and seemed to last however long the electromotive force was 
kept on. J. J. THomson. 
Cavendish Laboratory, April 24. 
The Theory of Dissociation into Ions, 
MR. SPENCER PICKERING has, in your number for January 7, 
brought forward certain difficulties which he says the advocates 
of the dissociation hypothesis have persistently ignored. I have 
been waiting in the hope that some one who supports the gaseous 
theory of solution as well as the theory of electrolytic dissociation 
would answer his letter. As no one has done so, I venture once 
more to trespass on your space. 
First let me say that the experiment described by Mr. Picker- 
ing, in which water or propyl alcohol exudes through the walls 
of a semi-permeable vessel containing a mixture of these liquids, 
according as propyl alcohol or water is placed without, appears 
to me, as it does to him, to be very strong evidence that it is 
complex molecules of solution to which the walls are impervious. 
The experiment is one which certainly needs explanation at the 
hands of those who uphold the gaseous impact theory of osmotic 
pressure, 
As I have already said, the idea that electrolytic conduct- 
ivity depends on dissociation of the ions from each other, does 
not involve, as is so often assumed to be the case, the gaseous 
view of solution. The evidence for such dissociation appears to 
me to be exceedingly strong, as I will explain very briefly below, 
so that some explanation of the second experiment described by 
Mr. Pickering is necessary. 
The experiment is this : The freezing point of a large quantity 
of acetic acid, to which is added a mixture of sulphuric acid and 
water in the proportions represented by too11,0 + H,SO,, shows 
that considerably less than 100 molecules have been dissolved. 
This result indicates that chemical union has occurred. Mr, 
Pickering says that, on the dissociation theory, the freezing 
point should be lowered by an amount corresponding to some- 
thing between 101 and 103 molecules. ; 
In such a case, however, we have conditions very different 
from those which hold when sulphuric acid is dissolved in water. 
In fact the liquid is in reality a mixed solution of water and sul- 
phuric acid in acetic acid, or possibly, as Mr. Pickering 
suggests, of the hydrate of sulphuric acid in acetic acid. It 
does not at all follow that because sulphuric acid is dissociated 
in water, it is, therefore, dissociated in other solvents ; in fact, 
the freezing points of its solutions in acetic acid show that, on 
the contrary, aggregation has occurred. We should, therefore, 
expect that dissolving sulphuric acid in acetic acid would have 
little or no effect on the conductivity; and this is also indi- 
cated by the low specific inductive capacity of acetic acid, which 
implies a low ionising power. There is no reason to suppose 
that the presence of a small quantity of water would modify the 
properties of the solvent enough to cause any appreciable change 
in the conditions. 
But, even if these considerations were insufficient to explain 
the facts, the dissociation theory would not be discredited. As 
I pointed out in your issues of October 15 and December 17, 
1896, dissociation of the ions from each other does not forbid the 
assumption that the ions are linked with one or more solvent 
molecules. Such a combination would explain Mr. Pickering’s 
observation. 
Mr. Pickering says that the dissociation theory depends. 
solely on the numerical agreement obtained when properties of 
solutions are interpreted by itsmeans. Although these numerical 
relations may have suggested the theory, they by no means 
furnish the only basis for it to rest upon, Other facts, to my 
mind, give much more conclusive evidence in its favour. As 
Mr. Pickering. has challenged the supporters of the theory to 
explain his experiment, I may be allowed to ask the opponents 
of the theory to explain the following phenomena in any other 
way than by a dissociation of the ions from each other :— _ 
(1) The velocity with which an ion travels through a dilute 
solution under an electric force is independent of the nature of 
the other ion present. . 
(2) ‘The conductivity of a dilute solution is proportional to its 
concentration, The alternative to the idea of dissociation is to 
