the Hibernation of Marasmarcha. 413 
small piece of Ononis with a bit of root, which I put in a 
glass jar in clean sifted sand. 
These hatched in due course. About August 29th those 
in the tube spun themselves cocoons exactly like tuttz- 
dactyla, between the slide covers and bits of leaf and 
between the glass of the tube and bits of leaf. A bit of 
leaf seemed much more to their taste than two glass 
surfaces. 
On September 6th I examined the piece of plant in the 
sand, and at first was very unsuccessful. By searching on 
the plant above ground I could find nothing, then on that 
below I was equally unsuccessful. I finally, however, 
succeeded in finding half-a-dozen cocoons. These were 
found, as regards at least four of them, in the sand, 
with some attachment to dead leaves of Ononis that were 
on or partially buried in the sand; the two others were 
probably in similar places, as, though free in the sand 
when I found them, it was at the same time as I found 
the others, and a search in the sand all round the plant 
afforded no others. [Pl. XXVIII, figs. 1, 2, 3.] 
Later, however, having waited till the sand was quite 
dry, I carefully sifted it, and obtained by so doing eight 
further little cocoons, minute aggregations of sand particles 
on that side with the cocoon in the midst. [Pl. XXVIII, 
fig. 4. ] 
These cocoons had sand attached to them, so that the 
contents were not easily seen. But those built against 
glass were so thin on that side that the glass formed 
practically the wall of the cocoon. The cocoon is of course 
very small, and the larva is coiled up in it, so that it 
occupies hardly more space than it does in the egg. 
The larvee of Marasmarcha (pheodactyla, fauna, tutti- 
dactyla) always occur ov. plants that form a considerable 
mass, and I imagine that the young larve form their 
hibernating cocoons amongst the dead leaves and other 
material of the plant close to the ground, and not on the 
plant itself, but have, owing to the density of the plant, 
little difficulty in finding a growing point when they come 
out in the spring. 
The examination of the piece of Ononis pianted in 
sand, as well as the futile searches made on previous 
occasions, make me feel certain that the little cocoon is 
never made in touch with the living plant, but somewhere 
sufficiently close by. It is so small that one might easily 
