420 
ZOOLOGY: T. H. MORGAN 
F, G, and K type spectra. This would be in agreement with the hypoth- 
esis of Eddington, aheady referred to, that the relation between ve- 
locity and spectral type might be a relation between velocity and dis- 
tance, the stars nearest the sun, mainly the types F, G, and K, moving 
more rapidly than the distant stars. Eddington considered this hypoth- 
esis as disproved because an analysis of the A t5^e stars indicated no 
increase of radial velocity with increasing proper motion. The fact 
that such an increase exists in the case of the later type stars, however, 
is shown clearly in Table I. Because of the slight range in proper 
motion a similar variation for the B and A stars is less certain, although 
indicated on the face of the results. 
The principal feature of interest in this comparison of proper motion 
and radial velocity is the low average velocity found for the distant stars 
of types F to M. These stars are on the average stars of high absolute 
luminosity, and the possibility of a relationship between radial velocity 
and absolute luminosity has been considered in the communication by 
Kapteyn and Adams, to which reference has already been made. The 
observational material included here is much too limited to provide 
the basis for a discussion of this question. It may be noted, however, 
that the average radial velocity corrected for the solar motion of such 
absolutely faint stars as have been observed at Mount Wilson is ex- 
ceptionally great. The average velocity of sixteen stars with absolute 
magnitudes below 8 on a scale for which the sun is 5.5 is 36 km. Eight 
have velocities exceeding 40 km. Since these stars are probably of 
small mass the evidence so far as it goes is in favor of Halm's hypothesis'* 
of the equipartition of energy among the stars, their motions being a 
function of their masses. 
^ British Association Report, 1911. 
2 These Proceedings, 1, 14 (1915). 
3 Lick Obs. Bull., No. 196. 
^ Stellar Movements and the Structure of the Universe, p. 157. 
^ London, Mon. Not. R. Ast. Soc, 71, 634 (1911). 
LOCALIZATION OF THE HEREDITARY MATERIAL 
IN THE GERM CELLS 
By T. H. Morgan 
DEPARTMENT OF ZOOLOGY. COLUMBIA UNIVERSITY 
Presented to the Academy, May 17, 1915 
It has come to be recognized that it must be more than a coincidence 
that in each animal and plant there are two representatives of each he- 
reditary character (one derived from the mother and one derived from 
