Rau—Longevity in Saturniid Moths. 67 
third case, after a like period of 3 hours, showed a very 
low per cent. of infertility. 
ith so much individuality apparent we cannot yet 
point out any causal factors of this variability in the 
fertility of this species. We present the data of the 
‘Fertile Unhatched’’ column without any more com- 
ment than to say that here too we find the same variabil- 
ity as elsewhere; in the ova of some individuals the 
. embryonic mortality is high, and in others comparatively 
low. Whether or not this lack of viability of the fertile 
ova is due to some factor in the St. Louis conditions, 
temperature or humidity, we cannot tell. To be sure 
this species is typical of the Atlantic states and does 
not occur at St. Louis, and the environmental conditions 
here may have affected this material. But the fact that 
a considerable per cent. of the fertile eggs hatched nor- 
mally shows that the climate is not intolerable to some 
of them at least, so the problem remains: What are the 
special characteristics or weaknesses of certain other in- 
dividuals which predetermine that they shall not be able 
successfully to reproduce, or having been produced, to 
survive? 
Parallel work was done with 27 fertilized females from 
Philadelphia, with the results shown in the table below. 
(See Table 32.) 
In this lot we see an appallingly great degree of in- 
fertility, and also a much higher per cent. of the fertile 
eggs which failed to hatch. Of the 4416 eggs deposited 
by this lot only 1215, or 27.5%, gave forth caterpillars. 
Even the New York lot, which we considered very weak, 
gave 41% fertility. Almost one-half of these eggs laid 
were infertile, and about half of the fertile eggs failed 
to bring their development to completion. Thus we see 
this Philadelphia population in a sad plight in regard to 
their perpetuation. Elimination in the early stages of 
life is common in all branches of the animal kingdom, 
but this, an elimination of nearly three-fourths of the 
