May 17, 1912] 
eight pages in all, will be disappointing to old- 
time ecologists, to whom, however, its brevity 
should be a suggestion that they have hitherto 
given overmuch emphasis to this phase of the 
subject. 
A thoughtful chapter on “ adaptation,” in 
which the author gives his personal views on 
the subject, closes the book in such a manner 
as to leave the student in a properly humble 
state of mind, since it makes it clear that 
many of the “cock sure” conclusions of yes- 
terday are improbable, or quite impossible. 
A most useful, ten-page appendix contains 
a classified bibliography which will prove very 
useful to the student who wishes to go farther 
than the study suggested in the text. 
Cuartes E. Bessey 
THE UNIVERSITY OF NEBRASKA 
SPECIAL ARTICLES 
THE PHOTOELECTRIC EFFECT 
Reapers of SclENCE may be interested in 
the following brief summary of some of the 
principal results of an investigation of -the 
magnitude and distribution of the total kinetic 
energy of the electrons emitted when light 
falls on metals, considered as a function of 
the frequency of the light and of the nature 
of the metal. A fuller account of the investi- 
gation was communicated to the meeting of 
the American Physical Society at Boston on 
April 27. 
Monochromatic ultraviolet light of various 
wave-lengths from a quartz-mercury arc lamp 
was allowed to fall on a small strip of the 
metal to be tested placed at the center of an 
exhausted conducting sphere. Measurements 
of the currents against various opposing po- 
tentials enable the distribution of the energy 
among the emitted electrons to be obtained 
directly. The experimental results may be 
analyzed and exhibited graphically by plotting 
the number of electrons having a given 
energy against the energy. These curves are 
nearly symmetrical about the axis of mean 
energy. The mean energy is very close to the 
most probable value of the energy. The prob- 
ability of an electron having energy within a 
SCIENCE 
783 
given range changes very rapidly in the 
neighborhood, both of the maximum energy 
and of zero energy. The maximum energy, 
and also the range of energy, of the electrons 
emitted by light of a given frequency is ap- 
proximately a linear function of the fre- 
quency. 
For different substances the relation be- 
tween the mean energy 7 and the frequency 
v of the exciting light is found to be 
Tv=k,(v—v,). For sodium, magnesium, 
zine, aluminium, tin and platinum k,= 
2.9 X10 erg. sec. v,isa constant character- 
istic of the substance. The above formula is a 
particular case of a more general relation 
Tv=v¢(v,/v), where @ is a universal func- 
tion of the argument, which was deduced 
theoretically by one of the writers. Accord- 
ing to the theory the values of v, should be 
ealeulable from Planck’s radiation constant h 
and the intrinsic potentials of the substances. 
The ecaleulated values of X,=c/v, are com- 
pared with those given by the photoelectric 
measurements in the following table: 
Na al) Me Zn | Sn | Bi ae 
Xp (calculated). ...|52.6 
57.0 
36.0 34.6/38. ap 
No (photoelectric) 3 
29.4 
39.5/36.5/36.1| 33.1 
1.0 
3.8 
Our measurements of the maximum energy 
Tm are probably less accurate and certainly 
more irregular than those of the mean energy; 
but they are all fairly near the linear rela- 
tion Im—k,(v—v,). The values of v, are 
the same as before and k, is very near to 
6X10” erg. sec. k, is thus about 10 per 
cent. less than Planck’s constant h. We do 
not, however, wish to emphasize this differ- 
ence, pending further investigation, as we 
realize that the accurate measurement of the 
maximum energy is a rather difficult problem. 
Bismuth and copper appear to have smaller 
values of both &, and k, than the other metals, 
but here again it is possible that further re- 
search will remove the difference. 
Tf the laws which we have found to connect 
the frequency of the light with the maximum 
and mean energy of the liberated electrons 
hold up to the highest frequencies, it follows 
