CHEMISTRY: HARKINS AND ARONBERG 
711 
number of positive electrons in excess of the nuclear charge P. The 
fact that isotopes, with a constant value of P, but with a variable value 
of h, have almost exactly the same spectra, indicates that the non- 
nuclear electrons vibrate as might as expected if the h positive and h 
negative electrons were not present in the nucleus, that is as if it con- 
sisted of P positive electrons alone. Since, therefore, the vibration of 
the non-nuclear electrons depends almost entirely upon the net nuclear 
charge P, it is evident that both the effect (1) of the mass of the nucleus, 
and (2) of the bound h positive and h negative electrons, must be ex- 
tremely small. The work here described was undertaken to see if 
either of these effects is large enough to produce a measureable devia- 
tion in the spectra of isotopes. It might be expected, even if there is 
no effect due to the mass, that if the positive and negative electrons in 
the nucleus are not coincident, their space arrangement should cause a 
slight, though possibly unmeasureable effect upon the spectrum. 
The isotopes chosen for this work were ordinary lead and lead from 
radium (Radium G, or Uranio-lead 3). The atomic weight of the 
former is 207.18, and that of the latter, as determined by Prof. Theodore 
.W. Richards, is 206.34 for this particular specimen used in this investi- 
gation. If ordinary lead is a single isotope this specimen of Uranio- 
lead 3 contains about 25% of ordinary lead as an impurity, since the 
atomic weight of pure uranio-lead 3 is about 206.05. The value of N 
for these isotopes is 82. 
The spectra of isotopes were first investigated in 1912 by Russell 
and Rossi,2 and by Exner and Haschek,^ who examined the spectra of 
ionium and thorium (isotopes of atomic number 90) but found no meas- 
ureable differences. Isotopes of lead have been studied spectroscopi- 
cally by Soddy and Hyman,^ by Baxter and by Richards and Lembert,^ 
by Rutherford and Andrade,^ by Honigschmidt and St. Horovitz,^ by 
Merton,^ and by Siegbahn and Stenstrom,^ all of whom found the spec- 
tra of isotopes identical within the limits of accuracy of their work, 
except that Soddy obtained a slight difference in intensity for the line 
X = 4760.1. While Merton used a higher dispersion than any of the 
other workers, he does not give the atomic weight of the uranio-lead 
which he used. 
The mea^suf ements here presented were made by Dr. Aronberg, and 
the spectroscopic work was directed by Prof. H. G. Gale. Preliminary 
determinations with ordinary lead, and with ordinary lead mixed with 
about 20% of uranio-lead 3, when made on a six inch Rowland con- 
cave grating, gave no measureable differences in the spectra or in the 
