30 BULLETIN 1429, U. S. DEPARTMENT OE AGRICULTURE 
. 
insects. In the laboratory feeding took place very satisfactorily 
upon honey or sugar water, and individuals were kept alive for one 
month under these conditions. 
Dexia is decidedly crepuscular in habit and is therefore active 
largely during the period about sunset and at other times when the 
sky is overcast and direct sunlight is absent. One of the best places 
found for the collection of adults of this species was a young pine 
forest at the summit of a hill near the experiment station at Suigen, 
Chosen. Here they appeared abundantly resting upon the foliage of 
the broad-leaved oak, apparently preferring this almost to the 
exclusion of other types of foliage. 
Unlike Prosena, which deposits larva? exclusively, Dexia may at 
times produce eggs, which have a variable period of incubation 
externally, depending upon the extent of development attained 
prior to deposition. Some eggs kept under moist conditions in Petri 
dishes required two days' incubation, although the average period 
was much shorter. By far the majority, however, hatch within the 
ovarian sac and the eggshell is extruded with the larva, in this 
respect differing also from Prosena, which largely retains the shells. 
The larvae are very evidently scattered promiscuously over the 
surface of the soil, but observations in the collection of parasitized 
grubs would seem to indicate that they are deposited in some num- 
bers at each point rather than singly. It was frequently found that 
grubs within an area of about 1 square yard were very heavily 
parasitized, whereas those surrounding this area contained practically 
no parasites, and that these spots of heavy parasitism recurred in 
greater or less abundance throughout the zone of collection. No 
locality was found where the parasitism was at all uniform throughout 
the entire area. 
PROBABLE VALUE OF DEXIA VENTRALIS AGAINST POPILLIA JAPONICA IN AMERICA 
A consideration of the life history of Popillia japonica and of Dexia 
ventralis in Chosen still leaves open the question of the latter's 
ability to increase to a point where it will be of value as a check upon 
this particular host. Climatic conditions in New Jersey and Penn- 
sylvania are not markedly different from those in Chosen, and con- 
sequently may be regarded as a minor factor. The major difficulty 
is the occurrence of three full generations of the parasite per year 
under normal conditions, probably only one of which could be upon 
P. japonica. It appears probable that the overwintering forms would 
be contained in grubs of this host, and if such is the case the emergence 
of the first brood of adults would be delayed until the middle of June, 
as against a month earlier in Chosen, and the larvae produced by these 
females would be able to parasitize such grubs as had not yet pupated. 
This, however, calls for the utilization of the one host species to carry jj 
two generations of the parasite, a condition not conducive in this case M 
to the attainment of maximum numbers of the latter. The following 
generation must of necessity be upon some scarabaeid grub of proper 
size which pupates about the first of September. This condition is 
fulfilled principally in the subfamily Sericinae. Full data are not as 
yet available as to what may be had in the infested regions in this 
regard, but the general outlook is not good for the development of 
this species to an important position in the parasite series under 
course of establishment in America. 
