28 BULLETIN 1426, U. S. DEPARTMENT OF AGRICULTURE 
of development may occur periodically, perhaps between matings, 
and thus afford the female the few days necessary for preparing a 
new egg gallery. Such females swept late in the season in flight 
would have a sexual development similar to that of those on their 
first flight. It is a doubtful point, therefore, whether the individual 
makes more than one flight. 
Examinations of individuals of both sexes of the parent generation 
indicate that some were reproducing as late as July 18, 1916, and 
August 13, 1920. On the latter date one male had testes still func- 
tioning and one female had been recently mated. Two females bad 
living spermatozoa in the spermatheca, but no sperm in the access- 
ory sac. In a cage experiment started with a male and a female 
March 20, 1918, the female was still alive July 25 and her ovaries 
were developing eggs, although visibly near depletion. It was found 
from a record made at Forest Grove that a male introduced into a 
cage April 24, 1917, still survived September 27, neither the female 
of the pair nor any offspring being recovered. 
NUMBER OF GENERATIONS 
There has been considerable difference of opinion as to the num- 
ber of generations produced annually by this species. American 
authors are unanimous in their conclusion that there is but one gen- 
eration a year. Most European authors, with the exception of 
Schmitt (39), have maintained that this species has two or more 
generations. Del Guercio (16, p. 271), the latest European writer 
to treat of the species, seems to conclude that there are three gener- 
ations per year on Trifolium pratense in Tuscany, although his sea- 
sonal notes agree closely with those recorded above for Oregon, 
western Maryland, and Ohio. 
American workers, notably Riley (37), Webster (46), and Davis (9), 
who have studied the species carefully, have noted that there are no 
well-defined periods in the seasonal development. The studies of the 
writer, the results of which are here given under the heads of sexual 
development, seasonal history, and life history, seem to afford the 
explanation of this fact. Records of other workers have afforded 
confirmatory evidence. 
All the evidence based on the facts observed in the development of 
the clover root borer tends to show that there is but one generation in 
a year, or even that a single generation overlaps the year. The female 
already noted, which had in all probability matured the previous 
August or September, was alive ina breeding cage and _ repro- 
ducing on July-25 following. There is also the case of the male pre- 
viously mentioned which survived in a cage from April 24 to Sep- 
tember 27. It is safe to assume that one of these individuals passed 
an adult life of almost a year. The male may have been produced 
in April from an overwintered larva, but even in that case it must 
have belonged to the parent generation, the offspring of the previous 
season. Adults of the parent generation have also been recorded 
as alive and reproducing as late as August 13. 
Eggs of the clover root borer have been found in clover roots in 
the field as late as August, and once on September 27. In the 
latter case the egg hatched the next day and must have been at least 
