144 Trans. Acad. Sct. of St. Louis 
before dusk, that this was the only one in any of the experi- 
ments to come in at dawn. It is easy to speculate that, since 
one promethea came in at the dawn period when cecropias 
usually fly, that phylogenetically the prometheas once had the 
same habit as cecropias and now occasionally revert to it. 
Exp. 4. June 15. At 11:30 p. m., I liberated 26 male 
prometheas 100 feet west of the house, to see if they would 
break their set habit of flying before sunset and come back to 
the cages of the females during the night. Eighteen of these 
were 14 day old and 8 were 2 days. None came back during the 
night or the period of dawn, but, true to their colors, 12 came 
back at the appointed hour, between 4 and 5 p.m. Of those 
which were 1% day old when liberated 50 per cent (9) came in, 
and of those 2 days of age, 27 per cent (3) returned. The 
strongest point here is that, while they were only 100 feet away, 
they did not bestir themselves to come until their accustomed 
time, seventeen hours later. 
The foregoing work on the prometheas does little more than 
to show that their time of flight is shortly before the close of 
day. Records of the bred and native males showed their period 
of flight to be as follows: 
3:40 to 4:00 
4:00 to 4:20 
4:20 to 4:40 
4:40 to 5:00 
5:00 to 5:20 
5:20 to 5:40 
5:40 to 6:00 
6:00 to 6:20. 
6:20 to 6:40 
peo OoONPANAH Ow 
Total 33 
It is at once apparent that of these 33 males which came in 
before sunset (we have socially ostracised the one who so far 
ignored conventionalities as to come in at dawn!) the majority 
eentered their activity about 4:20. It is exceedingly strange 
that promethea should fly by daylight; at present I know of no 
