THE WHITE COFFEE-LEAF MINER. 339 
of the sexes. Guérin says (Mém. etc., p. 17), that the insect is 
reproduced several times in the year, in the Antilles, once in about 
every forty to forty-eight days. This would allow for the Eggs, 
T to 8 days; Larva, 15 to 20 days; Pupation, 2 days; Pupa, 6 
days; Imago, 10 to 12 days; total, 40 to 48 days. 
Habits of the larva.—As soon as the larva is hatched (if I mis- 
take not), it cuts through the upper epidermis of the leaf, and begins 
toeat the parenchyma. Usually it may be found under an edge 
oran end of the blotch, eating. I found no cast skins in the mines. 
The larvæ can not be considered social, although several are 
often found in one mine when several mines have become united. 
They show no signs of pugnacity or mutual destructiveness. When 
the larva is full-grown it escapes from the mine, and often, or 
even generally, goes to another leaf to make its cocoon. This 
it can do by letting itself drop with a thread of silk. It then 
makes its cocoon across one of the furrows at,the edge of a leaf, 
on either the upper or the under surface, but oftener on the under 
Surface. The larva places itself across the furrow, and begins a 
Mehi by spinning a series of threads from one side to beyond the 
middle of the furrow, swinging the fore part of its body back and 
forth sidewise. When it has made one side of one end of its web 
thus, it spins a like series of threads to make the other side, with- 
out changing the position of the hind part of its body. Thus an 
pening is left in the middle of this end of the web, in the space 
occupied by the body of the larva. It then turns around, and 
places its body across the furrow in the opposite direction. Here 
ìt spins a like series of threads on each side of it, from the leaf to 
> ao part of the web, leaving a similar opening in this end. 
„en retires beneath the web, and lays a flooring of silk. On 
om it spins its cocoon, laying the outside threads length- 
ath cocoons are found in the greatest abundance on the leaves 
are near the ground, and frequently on leaves which have 
never been injured. : 
Habits of the Pupa.— The larva-skin is split longitudinally over 
Mai middle line of the head and first two thoracic segments. The 
agas extends from the very foremost extremity of the head to the 
- rd thoracic segment (metathorax), but does not enter this 
m * tor the certain observation of this fact, and of the manner 
Which the pupa-skin is burst, I am indebted to Dr. Hagen, 
