168 
A study of this table brings out clearly one very instructive fact, 
and hints at several others. We have, in these root-worms, three 
closely related species, attacking the same part of the same plant 
in precisely the same way, at the same stage in their development, 
and strictly dependent upon this plant, as far as is known, for their 
continued existence. One would say that here were all the con- 
ditions of a most determined struggle for existence between these 
three species, in which one or more of them must succumb. It is 
indeed interesting to see how this issue is evaded, and an adjust- 
ment reached by which competition is reduced toa minimum. The 
Colaspis larva makes the earliest attack upon the plant, begining its 
work upon the roots certainly as early as the first of May (half- 
grown individuals having been taken on the 15th), and finishing in 
June, all being full-grown and preparing to pupate by the end of 
that month. Next comes Paria, in July and August, neither ex- 
treme of its period being exactly defined by our observations; and 
finally comes Scelodonta, adults of which were copulating July 1, 
young larve occurring August 7. As far as now known, the Ncelo- 
donta larva is left in undisturbed possession during the remainder 
of the year, although there is a break in our observations during 
October. Certainly by November it has completed its work, and 
retired, full-grown and ready to transform, into its subterranean cell. 
It seems clear, furthermore, that this curious succession of periods 
is related to a difference of habit with respect to hibernation. Un- 
doubtedly, Scelodonta winters as a larva, and Paria as an adult. 
As Colaspis larve were only half-grown May 15th, they very probably 
hatched from the egg that spring; and as the adult Colaspis emerged 
about two months before the new brood of Paria, it seems hardly 
possible that both could have developed from eggs laid that spring. 
It is much more likely that Colaspis hibernates in the egg. On this 
hypothesis, we shall have the egys of Colaspis deposited in autumn, 
those of Paria in spring, and those of Scelodonta in midsummer ; 
the first genus hibernating in the egg, the second as an adult, and 
the third as full-grown larva, with the necessary result that their 
attacks upon their common food-plant are delivered successively. 
The advantage of this arrangement is evident where we reflect that 
by this means as many of each species are fed upon the strawberry 
as it would be possible to maintain of all three if they came into 
simultaneous competition. 
As a general rule, only the surplus structures of a plant may be 
destroyed by insects, since habitually to destroy more, would, in the 
natural order, eventually exterminate the plant, and with it the in- 
sect itself. As the strawberry plant grows continuously throughout 
the season, such an available surplus of root growth is continuously 
produced, but if all the root-worms were to attack it together, they 
would be limited to the surplus produced during one or two months; 
whereas by distributing their periods of activity, they are able to 
appropriate the whole, and avoid conflict altogether. 
‘he fact is worthy of notice that the times of the appearance of 
the adult beetles are not as widely separated as the periods of lar- 
val activity, the interval from the commencement of the advl 
<= -— 
