33 
amount of silk (for a Noctuid), from which fact alone it might be sur- 
mised that hibernation took place within them as pupa, even did we 
not know, as previously stated, that all eggs that have come under 
notice hatched during the fall, and sufficiently early to permit of the 
larvae reaching full development. 
Prof. P. M. Webster writes, May 13, 1901, that he had a similar 
experience with this insect, meeting with no better success in rearing 
the moth. He placed a large number of larvae in a breeding cage, 
saw them feed and then go into the ground, but from that time onward 
they seemed to have totally disappeared, "as though an earthquake 
had swallowed them up." He further states that the records of the 
Ohio station do not show that this species was ever reared to the adult. 
These experiences, it seems to the writer, are proof conclusive that 
the species, being a Southern one, is unable to withstand our winters 
and our sudden changes. It seems equally certain that hibernation 
takes place exclusively as pupa in a climate like that of Wooster, Ohio, 
and the District of Columbia and northward, since it is hardly to be 
supposed that a Southern species could survive the rigors of a North- 
ern winter in the moth state. 
NUMBER OF GENERATIONS. 
The number of generations produced annually has been stated to be 
at least two (possibly three for central Illinois, according to Forbes), 
and three for the South. The number will vary according to climate 
and season, and the writer surmises that at least three generations are 
produced in a climate like that of the District of Columbia, and that 
four or more may develop farther south. This is a matter requiring 
investigation, particularly in the Southern States. 
SUMMARY OF LIFE HISTORY. 
Our limited knowledge of the life history of the fall army worm 
makes it impossible to present a complete summary, but the hypo- 
thetical life cycle is about as follows : 
It has been stated that larvae of the first generation appear in May 
and June (locality not mentioned), and this statement is undoubtedly 
true for some regions. The earliest appearance of the moths in any 
single locality does not appear to have been recorded, or if it has it 
has escaped the writer's observation. 
The period of the egg has been ascertained only for October in 
moderately cool weather, and it is ten days. This period will ■ vary 
according to temperature during the period of incubation. 
The larval and pupal periods of the different generations remain to 
be observed, but the former we know must be subject to extreme 
variation and the latter will, judging from analogy, not be much 
5525— No. 29—01 3 
