Vol.. 6, 1920 
PHYSICS: F. C. HOYT 
639 
might be accidental, just as Castle's non-linear results so obtained were accidental. 
The fact that the order really turns out to be hairy magenta forked illustrates the 
latter point to a nicety and falls in line with the other results to show the fallacy of the 
method of combining separate experiments for a determination of the nature of linkage. 
8 It is of course possible that larger counts would give double crossovers in these cases. 
But this is very improbable in the case of hairy, magenta, and forked, because of the 
shortness of the distance" involved; and even in the case of yellow, crossveinless, and 
vesiculated, where the distance is longer, there could scarcely be enough double crossing 
over to throw the factors much out of line. 
^ The sterility of rugose-glazed hybrids was first observed by Metz, who attributed 
it to "incompatibility," because he thought that glazed females were fertile. He had 
noticed their sterility when mated to rugose males, but he termed this also "incompati- 
bility," as he believed that glazed females were not sterile when mated to glazed or to 
wild-type males. Now that it is known that glazed females are almost always sterile, 
it is obvious that the failure of the hybrids to produce offspring is due to the dominance 
of the sterility of glazed. In my work, each class of females (glazed, wax, and hybrids) 
was tested with four kinds of males (wild-type, rugose, glazed, and wax); and in all 
cases, with the exceptions already noted, the matings produced no offspring, though 
all four types of males are otherwise fertile. Metz has, consequently, withdrawn his 
"incompatibility" theory and accepted the interpretation given above. 
10 Mulier, H. J., Amer. Nat., 50, 1916 (193-221, 284-305, 350-366, 421-434). 
" Gowen, J. W., Genetics, 4, 1919 (205-250). 
12 Bridges, C. B., Genetics, 2, 1917 (445-465); /. Gen. Physiol., I, 1919 (645-656). 
The yellow genes in mice and rats suggest themselves in this connection; but there 
is no proof that the yellow mouse case is actually like the hypothetical vermilion 
vermilion-deficiency combination described above, or that yellow in mice is homologous 
with either of the yellow factors in rats. 
1* Mulier, H. J., ^mer. iVa/., 48, 1914 (508-512). 
^^Altenburg, B., and Mulier, H. J., Genetics, 3, 1920 (1-59); Bridges, C. B., and 
Mohr, O.h; Genetics, 4, 1919 (283-306). 
THE INTENSITIES OF X-RAYS OF THE L-SERIES, III. CRITI- 
CAL POTENTIALS OF THE PLATINUM AND TUNGSTEN 
LINES 
By Frank C. Hoyt 
Rogers Laboratory of Physics, Massachusetts Institute of Technoi^ogy 
Communicated by Edwin B. Wilson, September 27, 1920 
This work is a continuation of that of Webster and Clark ^ and Webster.^ 
The methods used by them are applied to the classification of some lines 
that were doubtful, and the general validity of the intensity laws is tested 
by the use of a. different metal as anticathode (tungsten) . 
Apparatus. — ^The high tension outfit constructed at the Massachusetts 
Institute of Technology by Prof. D. L. Webster and described in these 
Prockedings^ was used without essential modification. The Chaffee 
