39 
If a male we may see the testes, two oval bodies, one on each side of 
the fifth abdominal segment in the dorsal area. The larva, however, 
at this stage very soon loses its beautiful transparency, assuming a 
waxy appearance. In the position described it remains motionless for 
a period of 24 hours. Hoffmann ( l.c .) also gives 24 hours for this rest, 
but Liiders (l.c., p. 26) says “ about two hours.” Considering the 
great change which now takes place, I doubt whether the period, even 
in the most propitious circumstances, is ever so short as two hours. 
When lying thus, the thin upper skin of the leaf comes in contact with 
the sticky dorsal skin of the larva, and, as the moisture begins to 
evaporate, adheres to it, in the same way that a thin piece of paper will 
adhere if laid over wet glass. It may be asked, Why does the mois¬ 
ture in the mine begin to evaporate ? While the larva continues to 
mine juice is constantly supplied by the cells cut open by the larva, 
and moisture is also excreted by the larva itself. When, however, the 
larva becomes quiescent, both these supplies fail, if not entirely, at any 
rate in such a degree as to be unable to keep pace with the natural 
evaporation going on. As the loosened upper cuticle of the leaf, with 
the larva adhering to it, dries, it slightly buckles the mine, lifting the 
central portion of the larva up with it. While this is proceeding, the 
larva, in its strange new form, may be seen inside the now effete, still 
outstretched, old skin, shrinking to the proportions normal to the spin¬ 
ning stage. It withdraws the head and tail away from those portions 
of the old skin, which appears to be firmly wedged into the angles 
formed at the circumference of the mined space by the upper and lower 
surfaces. At the end of the 24 hours the cervical and thoracic por¬ 
tions of the old larval skin appear to split off from the remainder, and 
the head of the new larva appears. As soon as the head is well freed 
from the old skin the “prrepupa,” as the larva has been called at this 
stage (i.e., the prepupal stage) commences to line the lower surface of 
the mined space with silk, incessantly passing its head from one side 
to another, the threads running across the cocoon, more or less at right 
angles to the body of the larva. While incessantly spinning, it gradu¬ 
ally withdraws its body from the old skin. The larva in its fourth 
instar is so very dissimilar from its previous appearance that the first 
time I observed it, not having read the literature on the subject, I 
opened the cocoon to see how long, I think, it would take me to exe¬ 
cute vengeance on that horrid ichneumon. Perhaps the most noticable 
characteristic about the larva in the spinning stage is its extraordinary 
activity. The apparent eagerness with which it commences to spin, 
and the energy it displays during the work are really astonishing. The 
only thing I have seen at all equal in activity is the young larva of 
Gracilaria stigmatella building its first cone at the apex of a young 
willow leaf. Before long the praspupa turns its attention to the upper- 
surface of the cocoon. This it lines in the same manner, passing the 
spinneret rapidly from side to side. Here its legs and the adhesive 
discs of the thorax come plainly into view as it clings to the upper- 
surface of the cocoon. In about ten hours the cocoon is complete, and 
in another ten hours the larva has thrown off its spinning dress and 
donned that of the real pupa, which is already so fully formed that it 
can wriggle actively in the cocoon. 
Pr^epupa.— The prepupal form of the larva is more cylindrical 
when viewed from above than in the three previous instars, though, 
