246 
LYCAENIDAE 
is set on a very flexible, retractile neck, which can be protruded 
while in motion and withdrawn into the ventral surface of the 
segment while at rest. The ground colour is pale creamy 
ochreous with a pinkish-lilac tinge over the first and last 
segments and along the lateral ridge. The whole of the 
skin has a distended, shining appearance, appearing too 
tight for its bulky body. From the great increase in size, 
the fully-grown larva seems to differ greatly from its appear¬ 
ance just after its moult in the previous autumn. The black 
disc on the first segment now appears as a mere black speck. 
The dorsal furrow is less pronounced owing to the swollen-up 
body ; the lateral ridge is dilated, prominent and much rounded. 
Honey-gland. The honey-gland, which is the source of 
attraction to the ants, is on the dorsal surface of the tenth 
segment and excretes minute drops of clear, sweet fluid, 
which is very eagerly licked up by the ants. The gland is 
composed of flexible tissue and is surrounded by numerous 
pyriform, glassy-white processes varying in size; those 
bordering the edge of the gland bear extremely small bristles, 
each having four or five in the form of a fan with diverging 
points, the whole group forming a fringe round the gland. 
This arrangement is probably for holding the bead of liquid 
in place, and also may act as a protection to this highly- 
sensitive organ. 
Obviously, both the ants and larvae are instantly conscious 
of each other's presence, as immediately they meet the ants 
start caressing the larvae by stroking them with their antennae. 
These caresses usually bring into play the retractile tubercles 
situated behind the spiracles on the eleventh segment ; the 
functions of -the tubercles is unknown. Examination of the 
actions of the ant and larva under the microscope revealed the 
following. Directly they meet, the ant waves its antennae 
over and upon the gland, closes its jaws, and smells and 
licks it. Directly an ant's foot touches the gland, or goes 
very near it, it immediately throbs more violently, swells up 
and ejects a globule of clear white liquid, which is instantly 
licked up by the ant. In a few seconds another foot, perhaps, 
touches the gland, and another bead of liquid oozes out, which 
is again licked up as before. 
These observations I made upon the M. arion larva after 
