SEPTEMBER 4, 1913]| 
the cocoon and allowed them to develop on the wet 
cotton wool, the newly hatched larve congregated 
into a mass and remained so for a day or two, after 
which they became active in search of food. 
You will notice that the larva possesses on each 
body segment a pair of lateral processes, and on the 
last segment a pair of ventrally placed processes of a 
different kind. These latter, which are possessed by 
all water-beetle larvee which come to the surface tor 
their air, have probably some connection with raising 
the tail to the surface for breathing, but the hairy 
lateral processes have been called gills. Many larvee 
of the Palpicornia have lateral processes, usually 
smaller than those of Hydrocharis, but in no case are 
they really gills, and the larve quickly drown if 
prevented from bringing their tails to the surface to 
renew their air-supply. 
The larve of Hydrocharis, like those of Dytis- 
cus, will eat almost any kind of animal matter, and 
hence they are easily supplied. I fed them mostly 
upon chopped worms, but their method of feeding is 
very different from that of Dytiscus. They seize 
their food with the jaws, antennz, and the other 
mouth parts, and they then come to the surface, and 
raising their heads and part of the body out of the 
water, they proceed to chew up the food by opening 
and closing the jaws, turning it from time to time 
with the other mouth parts. The jaws are not per- 
forated, nor is there any mouth-lock as in Dytiscus, 
and they suck in the juices of the prey by the mouth, 
spitting up saliva at intervals, which actually froths 
over the food and digests it, the dissolved material 
then being sucked down. The external digestion is so 
complete that in the case of a thick piece of worm 
all that is ultimately rejected is the thin transparent 
outer nellicle. 
In the mouth parts of the larva I want to direct 
your attention to a curious want of bilateral sym- 
metry, noticeable not only in the jaws—one of which, 
the left, has a small extra tooth near its base—but 
also in the upper lip. In many species there is an 
absence of bilateral symmetry where a pair of organs 
are complementary. Thus in the jaws of the beetle 
itself, the base of the left one is hollowed out to 
receive the base of the right one, which is convex, 
the two being related as pestle and mortar for grind- 
ing up the food. The larva of another species of the 
same group also shows asymmetry of the jaws, but 
here again it is definitely associated with the method 
of feeding. This species feeds upon pond snails, and 
the left jaw holds the shell while the right jaw with 
its large double tooth cuts through it. 
The asymmetry of the upper lip, however, is at 
present inexplicable, and, curiously enough, it occurs 
in several other species. - 
The larva of Hydrocharis, like that of Dytiscus, 
passes through three stages, the first two of which 
occupy from five to eight days, and the third stage, 
up to the time the larva is full grown, occupies about 
four weeks. It then leaves the water and burrows 
into the earth, forming a cell, just as the Dytiscus 
larva did. I had many specimens of these larva, 
and so made many experiments with them, and one 
curious fact about them is that the instinct which 
leads them to burrow into the ground and make a 
pupal cell only lasts for one or, at most, two days. 
In no case, where I removed a larva even imme- 
diately after the completion of its cell, did it make 
any attempt to form another one, and if left on the 
surface of the soil it moved about listlessly and ulti- 
mately died, apparently of drought, since if placed in 
_a damp position, for instance, in an artificial cell, it 
survived and pupated. If a cell was damaged before 
completion the larva often completely destroyed it, 
NO. 2288, vor. 92] 
NATURE 23 
apparently in the attempt to repair the damage, and 
would be found sitting amongst the ruins. 
Once the cell is completed the larva rests for about 
three weeks, at the ena of which time the skin is cast 
otf and a greenish-white pupa appears. This is more 
spinose than that of Dytiscus; but it also prefers to 
lie upon its face, resting upon the two small tail 
projections and upon the “‘collar’’ of the prothorax. 
The perfect insect appears after about ten days, so 
that the whole life-cycle occupies about nine or ten 
weeks from the laying of the egg to the appearance 
of the perfect insect. This time, however, may be 
greatly prolonged under less favourable conditions. 
Thus, the later egg-cocoons produce larve which take 
twelve or fourteen weeks to grow up, and the cocoons 
built in July produce beetles which do not leave the 
pupal cell for six or seven months. The larve leave 
the water in September and even in October, and 
after three or four weeks turn into pupe. These 
pupz turn into beetles in late October or November, 
but the beetles remain, apparently torpid, until the 
following March or April, when they make their way 
out and to the water. 
I have mentioned that the larve of both Dytiscus 
and Hydrocharis breathe in the same manner by rais- 
ing the tail to the surface. The perfect insects, how- 
ever, assume very different positions when taking in 
their air-supply. 
Dytiscus floats up to the surface tail first, taking in 
air between the body and the great wing-cases which 
cover it, and it is in this cavity under the wing-cases 
that the whole reserve of air is carried. 
On either side of the body under the wing-cases is 
a row of pits, spiracles; the last pair of these are 
much larger than the others. When the insect rises 
tail first to the surface, the tubes connected with this 
last pair contract and expand, just as in the larva, 
renewing the air-supply in the whole tube system, 
while at the same time the body contracts and ex- 
pands, renewing the reserve supply under the wing- 
cases, 
Hydrocharis, on the other hand, comes to the sur- 
face head first, turns its head on one side, and pushes 
its short, club-like antenna through the surface-film. 
Now a large part of the under side of this beetle is 
covered with fine velvety hair, which retains a thin 
film of air upon it, just as a piece of velvet does when 
gently pushed under water. When the beetle raises 
its antenna above the water it brings this film of air 
into communication with the air above the water. 
It also has a reserve supply under its wing-cases, and 
this communicates at the sides with the ventral film, 
and by expansion and contraction of the body the 
used-up air is expelled above the water and fresh air 
is taken in. In Hydrocharis the most important 
spiracles are situated well forward, and thus the used 
air from the air-tubes is expelled and fresh air taken 
in at the front end of the body instead of the tail 
end. 
Anyone who examines Hydrocharis and compares 
it with Dytiscus will at once see great structural 
differences. In a ventral view of the two types, com- 
paring the heads, the most noticeable difference is in 
the antennz, which are filamentous in the former and 
clubbed in the latter, and the maxillary palpi, which 
are short in the former and long in the latter, in 
which they are used under water as feelers, just as 
are the antennz of Dytiscus. 
Passing over other less remarkable differences in 
the heads of the two types and coming to the body, 
one at once notices the different disposition of the 
legs : in Dytiscus the first two pairs are close together, 
in Hydrocharis the three pairs are about equidistant. 
In Dytiscus the basal segment of each hind leg—the 
