312 
from a primitive common ancestor (the germ- 
cell) and their evolution in the course of count- 
less generations into distinct genera and species. 
Coues, in fact, has gone so far, in writing of 
bird-anatomy, as to treat the different kinds of 
cells as pertaining to several genera and species, 
which he names. 
But we are here met by the extraordinary 
fact that all this complicated development and 
evolution is repeated anew in every individual, 
and that, speaking broadly, the course of cel- 
lular evolution is predetermined in the germ. 
This fact is so commonplace to us that we have 
ceased to realize the wonder of it, or its possible 
significance as a hint of the method of evolu- 
tion among species. 
Why may it not be that the evolution of spe- 
cies, to a greater or less extent, is similarly 
predetermined, and that here is to be found the 
explanation of the phenomena described in the 
beginning of this note? If life exists in Mars, 
a knowledge of it would go far toward answer- 
ing such a question. How much similarity 
would there be between creatures evolved on 
two planets, with all the diversity of conditions 
which this implies ? 
T. D. A. COCKERELL. 
East LAS VEGAS, N. M., 
January 29, 1901. 
NOTES ON PHYSICS. 
NON-PERMANENCE OF WEIGHT. 
EXPERIMENTS by Heydweiller (Phys. Zeitschr. 
Aug. 25, 1900), similar to those of Landolt 
(Zeit. fiir Phys. Chem., 12, p. 1, 1893), seem to 
show that a slight change of total weight ac- 
companies some chemical reactions. These 
experiments have been interpreted by some re- 
viewers as throwing doubt upon the axiom 
of the conservation of matter. This axiom is 
not, however, incompatible with variation of 
total weight in chemical or even in physical 
changes. If it should be found, for example, 
that the weight of a given amount of lead and 
of a given amount of oxygen varied with phys- 
ical and chemical conditions, a standard state of 
lead and a standard state of oxygen would have 
to be adopted in which state these substances 
would always have to be weighed, and the 
principle of the conservation of matter would 
SCIENCE. 
(N.S. Vou. XIII. No. 321. 
have to be stated thus: Given so much lead 
and so much oxygen, measured by weighing 
under standard conditions, then, whatever 
changes these substances undergo, the amount 
of each is found to be unchanged if both are 
brought back to standard conditions and 
weighed. 
Variation of weight with physical and chem- 
ical conditions would, no doubt, throw light 
upon the nature of gravitation, but if such a 
variation becomes established it will have but 
little disturbing influence upon the notion of 
the indestructibility of matter. 
In the light of Professor Fessenden’s elec- 
trical theory of gravitation, it would seem that 
the change of state most likely to produce a 
change of weight would be the dissolving of an 
electrolytic salt in water. For, assuming elec- 
trolytic dissociation to be a separating of posi- 
tively and negatively charged atoms or ions, 
the region throughout which the electric force 
of the atom is exerted would be greatly ex- 
tended by the dissociation. 
THE ELECTRO-MAGNETIC THEORY OF RADIA- 
TION, 
Proressor M. PLANCK, of Berlin, published 
some months ago a derivation of the formula 
connecting energy and wave-length in the spec- 
trum of a black body at a given temperature, 
the derivation being based upon the notion of 
an electrical resonator enclosed in a space sur- 
rounded by perfectly reflecting walls. . It is re 
markable that this formula should agree with 
the formula of Stefan obtained by thermody- 
namical considerations. In the Verhandlungen 
d. Deutschen Phys. Gesellschaft, for December 
1900, Professor Planck has given an outline 
of some work, soon to be published in full, in 
which he applies the method of probabilities to 
the determination of the partition of energy 
among a vast number of electrical resonators 
enclosed within a reflecting boundary. A con- 
sequence of the theory developed by Pro- 
fessor Planck, which gives some check upon its 
legitimacy, is a formula which permits the cal- 
culation of the number of actual molecules of 
“any salt in a gram-molecule (the number of 
atoms in a gram of hydrogen), the basis of the 
calculation being the energy curve of the spec- 
