164 Trans. Acad. Sci. of St. Louis 
mained within, in either temporary or final rest. Again 
on September 23 only one carpenter-bee was seen. At 
this date all the other life was proportionately reduced, 
as was to be expected amid unfavorable weather condi- 
tions. A visit to the spot ten days later, October 3, re- 
vealed several dead specimens of these bees and other 
species about the bank, but no living ones. In all prob- 
ability they were either dead in their burrows or closely 
huddled together, ready to hibernate for the winter. 
In the year 1918, as I shall relate elsewhere, all the 
more important species in the unit appeared earlier than 
in the previous year. In 1917, the first carpenter-bees 
were sawing their tunnels on June 15; in 1918, on May 
28 not only were they at work, but they had already com- 
pleted their burrowing and were busily gathering the 
yellow pollen from the wild roses growing some distance 
away. The big shining black bee, heavily laden with 
bright yellow pollen, made a conspicious bit of color, and 
was as pretty as the blossom itself. An examination 
showed that no new tunnels had been made, but the old 
ones from the last year had been again used. 
My next visit on June 28 found no carpenter-bees out; 
this was probably the period between the two genera- 
tions as was July 30 in 1917, when the first generation 
was late by a corresponding margin of time in getting a 
start. On July 17, when it seemed time to expect the 
second generation, I visited the colony, and found one 
lone carpenter-bee at work; July 31 likewise found only 
one bee out. 
All of the evidence so far in 191% indicated a marked 
reduction in the population of carpenter-bees. An abun- 
dance of shedding-skins of the silver-winged parasite, 
Argyromoeba tigrina, adhered to the wooden rafters 
