Feb. 2, 1903.] 
THE TROPICAL AGRICULTURIST. 
565 
called hypeimetamorphosis, that is it passes through 
more than the three normal stages o£ metamorphosis. 
When first hatched it is an active little creature 
with well-developed legs. Bat soon it settles down 
into the degraded condition of parasitism in the 
egg-cocoons of certain epiders, where it assumes 
the form of a fat fleshy grub and loses the use of 
its limbs. From thence it passes into the cormal 
pupal stage, and eventually renimes active life as 
ihis delicate net-winged little insect that you see 
before you. It preys upon small flies, catching and 
eating them in exactly the same manner as does the true 
Mantis. We find the fame form of limb recurring in the 
2-winged flies (Diptera). It is very noticeable in that 
remarkable little species the " Telescope Fly " 
{Diopsis), so-called because its eyes are placed on stalk- 
like projections from the sides of the head. In this 
insect the femur of the front leg is armed with a row 
of sharp spines upon which the tibia closes. The 
Diopsis runs about ou the leaves of plants iu shady 
places and pounces upon smaller insects that happen 
to be exercising themselves in the same situations. In 
another group of flies, the Empidue, it is the hind 
legs that are modified for grasping, and in these we 
find a somewhat similar arrangement of the femur 
and tibia of the third pair of limbs. The Empis fly 
often pounces upon its victim and bears it along with 
it, taking its meal in mid-air. The b'g " robber flies" 
(Asilidae; catch and carry, their prey in the same 
manner, but use their feet only in holding it. Their 
legs are consequently without any special modification. 
These flies suck the juices only of thtir victims. 
Amongst 
THE WASP TRIBE 
(IJyinencptera), grasping apparatus is found both 
on the front and hind limbs. Of the former type, 
the Gonatopus may be taken as an example. In this 
instance the prc-bensible apparatus is a modifica- 
tion of the claws of the front-leg to form a strong 
pair of nippers. This picture was made from a spe- 
cimen caught by sweeping with a butterfly net 
amongst coarse grass. Although it figures so large 
on the screen, the Gonatopus is really a very minnte 
insect, only a Email fraction of an inch iu length. 
The feeding habits of this insect are not known. 
In its larval stage it is a parasite on the bodies of 
small insects popularly known as " frog-hoppers," 
The eggs are laid on the body of the frog-hopper, 
and it is in holding on to these acrobatic little insects 
that the gripping apparatus of the Gonatopus isbionght 
into play. In some other parasitic Hymenoptera, e.g., 
S iephaims indicitts—ihe h'^d Wrnha are modified in the 
same way as those of the Empis fly already de- 
scribed. This character is very well marked in a little 
wasp that breeds in the egg-cocoons of the com- 
mon Mantis. It is a brilliant little insect with a 
burnished metallic green body. Its extravagantly long 
ovipositor is required to allow of the eggs being 
placed in position in the heart of the cocoon. But 
little is known of the feeding habits of these minute 
insects and it is uncertain whether this structure 
is employed for the capture of prey or solely for 
grasping the insects in which they wish to deposit 
their eggs. It can hardly be required for this pur- 
pose iu the case of the insect that oviposits in 
the egg-cocoon. I must now bring to your notice 
some insects in which 
THK JIOUTII PARTS 
are especially modified for reptorial pur- 
poses. One of the most remarkable of these 
modifications is the apparatus carried by the larvae 
of dragon-flies. The adult dragon fly is well-known 
to all of yon. One species is perhaps 
THE COMMONEST INSECT IN COLOMBO 
or, at any rate, the one most in evidence. No-one 
can have driven frpm the Colombo station to the 
Fort without being stiuck by the sight of the 
thousands of dragon-flies hovering iu the air along 
the edges of the lake, or resting in long lines on the 
telegraph wires. The adult drngou-fly takes its food 
on the wing, its prey consisting of all sorts of smaller 
winged insects. Those dragon flies that we see along 
the lake edge are doubtless growing fat on a plentiful 
diet of that troublesome Colombo pest of which we 
have lately heard so much — the Lake-fly. But 
pethapa very few of you have seen the insect in its 
larval stage, or would recognise it, if yon did see it. 
The dragon-fly larva lives in the water and feeds 
upon worms and small aquatic insects, and perhaps 
picks up an occasional young fish. Some years ago 
1 was present when the stew-ponds at Nnwara Eliya 
were emptied Hundreds of young trout fry had 
been turned into the ponds some months previously : 
but they had completely and mysteriously disap- 
peared. The drag-net brought up not a single 
troutlet. But it contained numerous ugly looking 
larvae which I identified as the grubs of one of our 
largest dragon-flies. I think it very probable that 
these larvae were responsible for the disappearance 
of the young fish. The dragon-fly larva has a broad 
head with prominent goggle eyes : six well developed 
jointed legs : and a body that is short and stout or 
long and slender — -according to the species. But 
they may always be recognised by the presence — on 
the underside of the head — of an elbowed process, 
provided with a strong pair of forces at its extremity, 
and capable of being shot out in front to seize any 
desired object. Its application reminds one of the 
lazy- tongs used by bur graudmothers to pick up pins 
or other small objects without rising from their seats. 
When at rest this apparatus is folded close against 
the underside of the head, the front part almost 
completely covering the face of the insect, on which 
account it is often referred to as the mask. I have 
here a photographic representation of a young dragon- 
fly larva with the apparatus retracted : another 
showing it fully extended : and a third giving a 
larger view of the head and 'apparatus itself. It would 
be difficult— without entering into minute technical 
details of the mouth-parts of insects in general — to 
explain the exact relations of this organ, and I will 
merely say that it consists of a modification of the 
lower lip (labium) and the chin (mentum) with the 
labial palpi. The dragon-fly larva is a very common 
insect and may be dredged up. in a hand-net, from 
the bottom of any shallow pool nr along the edges of 
the lake. It would repay anyone to keep one or two 
of these insects in a shallow vessel — such as a finger- 
bowl or Eoup-plate — to observe their habits. Drop a 
fly or a small worm into the water, at a little distance 
from, but within sight of the larva. You will see the 
latter creep stealthily towards the object until it is 
within striking distance, when the extensile pincers 
will be suddenly shot orward and the victim will 
be seized, pulled in and promptly devoured. Another 
most interesting insect — also a member of the family 
Neui'optera — is 
THE APHIS-LION 
(the larva of Miirmdeo). It is of special interest 
more from its habits than its structure. It has no 
extensile apparatus, nor does it run after its prey. It 
waits until its food actually tumbles into its jaws. 
To ensure this happy result, the ant-lion constructs 
small funnel-shaped pits in loose snnd or fine earth, 
in spots sheltered from rain. The ground beneath 
the eaves of a bungalow or outhouse is an almost 
sure find for them. At the bottom of the pitfall lurks 
the little larva, its body completely buried, its head 
alone exposed — with the sickle-shaped jaws widely 
extended- There the little creature waits, patiently 
motionless, until some wandering ant falls headlong 
into the trap, when it is seized, dragged under the 
sand and sucked dry. If the ant escapes the first 
snap of the jaws and tries to climb the sloping walls 
of the pit, the shifting send gives way beneath its feet 
and its discomfiture is completed by a volley of saud 
jerked at it by the concealed insect. A second tumble 
to the bottom usually results iu its capture. The 
empty skin of the ant is finally jerked out of the hole 
and the trap repaired for the reception of the next 
