256 MR. F, BALFOUR-BROWNE ON THE { Mar. 2, 
and A. puella, which very closely resemble each other, is the form 
of the notum of the prothorax, and I give drawings of this part of 
both these species. It will be seen that the males also differ some- 
what in this respect, but to a less extent than the females. 
Hither while a pair are flying about attached per collem, or 
while so resting, the female curves her abdomen underneath her 
in such a way as to bring the underside of her 9th segment into 
contact with that of the 2nd abdominal segment of the male ; and 
the parts are so constructed that a firm attachment can take place, 
and the spermatozoa are transferred from the sperm-sacs of the 
male to the oviducts of the female. 
After a contact of shorter or longer duration, from a few seconds 
to a minute or more, the female once more straightens herself out 
and sooner or later oviposition commences. This is most commonly 
done while the male is still holding the female per collem, but, 
though I believe less commonly, the female will oviposit without 
being held bythe male. The male remains entirely passive during 
oviposition, apparently obeying the will of the female. When she 
flies, he also flies; but she directs, or seems to direct, the course. 
She seeks the aquatic vegetation and deposits her eggs as already 
explained. The male anenmernille rests either with his legs on the 
leaf or more commonly standing erect in the air, his body rigid, 
his legs folded, and himself supported only by the claspers attached 
to the prothorax of the female. 
I kept the prisoners in the cage for about a week, feeding them 
on flies and other insects captured with the sweeping-net. The 
prey crawled about on the muslin, and the dragonflies, flying about 
in the cage, picked them off and either ate them while still flying 
or after settling onthe sides. A large number of eggs were mean- 
while deposited in the frog-bit leaves, besides very many which I 
found in the bottom of the dish. The eggs were laid during the 
first four days of August, and I placed the frog-bit leaves in 
tumblers of clean water and also collected a number of the loose 
eges and dealt similarly with them. 
After about a fortnight the leaves showed signs of decay ; and 
towards the end of August nothing of them remained but a mass 
of decayed vegetation at the bottom of each of the tumblers, and 
my hopes of seeing any nymphs from the eggs vanished. How- 
ever, on the 28th ‘of the month numbers of nymphs appeared on 
the top of the mass of decay—almost four weeks after the depo- 
sition of the eggs. Although the majority of the eggs hatched at 
one time, nymphs were appearing for a fortnight or three weeks 
after that ; and the same fact was noticed with the eggs of /schnura 
elegans on another occasion. 
The eggs which were dropped in the bottom of the dish never 
developed at all, and were all decomposing within 10 days. I 
have found that the eggs of the water-beetle Dytiscus also failed 
to hatch when cast loose instead of being buried in plant-tissues, 
as they normally are; and Miall (1895, p. 40) mentions that 
Lyonnet also found that loose eggs of Dytisews never hatched. 
