878 
reduce friction. The bearing (Fig. 2) is a parallel ball- 
bearing for which the friction is less than 1/10,000 of 
the couple acting on the stator. To eliminate friction, 
the wheels over which the above-mentioned wires 
pass are carried on steel knife-edges resting on 
hardened steel planes. 
For the temperature measurements we use platinum 
thermometers which are connected differentially to 
a Wheatstone bridge made to Miiller’s design. During 
the course of a year the average variation of the 
fundamental interval from its mean value has been 
1/20,000 for one thermometer and 1/80,000 for the 
other. This would imply that the thermometry is 
of satisfactory accuracy. 
The evaluation of the heat lost has proved far the 
most difficult part of the experiment. 
In the earlier designs the loss in some experiments 


SEALING WAR ~~ - 



PLAN OF BASE—. 
SHOWING SECTION 
~—LINE THROUGH AB 
FULL SECTIONAL 
ELEVATION, | 
Fic. 2. 
was as high as 2 per cent of the heat developed. All 
attempts to determine it correctly, or to eliminate it by 
taking the difference between a heavy and a light ex- 
periment failed, for reasons which cannot be given here. 
To oyercome this difficulty the calorimeter system 
was reconstructed, by bringing the thermometers 
close to the vacuum flask and highly insulating them, 
as shown in the accompanying figures. This has re- 
duced the heat loss to about 1/10 of its previous value. 
Means have been provided for determining the loss, 
and the stator is being modified so that the loss may 
be eliminated in the usual manner in continuous 
flow calorimetry by taking the difference between a 
heavy and alight experiment. It is expected that this 
alteration will increase the heat developed, and so 
reduce the percentage of heat lost still further. 
T. Hi. VaR, 
University of Melbourne, April 4. 

The Transformation of Electronic into Electro- 
Magnetic Energy. 
Tue fundamental propositions given below, which 
do not refer to the excitation of characteristic but to 
that of the ordinary rays which}have been called 
NO. 2800, VOL. 111] 
“NATURE 













































[JUNE 30, 1923 
“independent ” X-rays, have sufficient experimental _ 
evidence supporting them to justify the following 
statements so that they may serve as guiding prin- 
ciples for further investigations. ‘ 
1. When a definite number of electrons in motion 
(cathode or §-particles) of definite velocity traverse 
very thin layers of different substances, the average 
fraction of their energy transformed into that of 
electromagnetic radiation (X- or y-rays) is, per atom — 
of any one substance, proportional to the square of 
its atomic number. 
2. In these circumstances, for a given layer, the 
energy so transformed depends only on the mass per 
unit area of the layer and on the number of cathode 
or §-particles traversing it, being independent of 
their velocity. 
By a very thin layer is meant one so thin that the | 
ratio of the number of particles emerging from the 
layer to the number entering it is very nearly equal 
to unity. ; 
The reason for the above statement is as follows. — 
If cathode or 8-particles of definite type and of total — 
energy E traverse a layer of a substance of unit area 
and mass dm, the energy of the X-rays formed in the 
layer may be written as \E.dm. We call » the 
mass transformation coefficient. The atomic trans- 
formation coefficient, a say, is then obtained by 
multiplying \ by A/N, where A is the atomic 
weight and N is the number of atoms in a gram 
of hydrogen. I find that a, which gives the average 
fraction of the energy transformed per atom, varies 
approximately as the square of the atomic number Z, 
while \ varies as Z?/A and both a and \ vary inversely 
as the energy of a single bombarding particle. Hence 
the above propositions hold approximately, since the 
total energy E is proportional to the energy of a 
single particle and their number 7, so that \E.dm 
varies as n.dm. 
With respect to the physical processes underlying 
the excitation of ‘‘independent”’ X-rays, certain 
considerations incline me to the provisional view that 
these X-rays are produced by a collision or by close 
interaction between the cathode or f-particles and 
the actual nuclei of the atoms rather than with 
electrons surrounding them. J. A. Gray. 
McGill University, Montreal, 
May 21. 
ae 
Dr. Kammerer’s Alytes. 
Pror. MacBripe’s letter in NATURE of June 23, 
p. 841, did not at first seem to require any rejoinder. 
But I find that some botanists, and perhaps others 
unfamiliar with zoological terms, suppose that 
quotations from Boulenger contradict my statement 
that rugosities are not formed on the palmar surfaces. 
Boulenger, of course with perfect accuracy, stat 
that rugosities in various genera appear on the inner 
side of the digits (italicised by Prof. MacBride). 
This is the radial side, as emphasised in both our 
letters, not the palmar surface, which was the part 
which bore the extraordinary structure visible in 
Dr. Kammerer’s specimen. W. BATESON. 
June 24. a 

The Breeding Period of Echinus miliaris. 
Tue breeding period of the sea-urchin, Echinus 
miliavis, is very interesting from many points 
view; and especially as this animal readily yields 
ripe eggs and sperm with which to carry out artificial 
fertilisation in inland laboratories for the observation 
