- 
1876.) Jumping Seeds and Galls. 217 
culent seed, which, in after growth, closes up the minute hole of 
entrance, just as in the case of the common pea weevil (Bruchus 
pisi). Toward the month of February the larva eats a circular 
hole through the hard shell of its habitation, and then closes it 
again with a little plug of silk so admirably adjusted that the 
future moth, which will have no jaws to cut with, may escape 
from its prison. A slight cocoon is then spun within the seed, 
with a passage-way leading to the circular door; and the hith- 
erto restless larva assumes the quiescent pupa state. Shortly 
afterwards the pupa works to the door, pushes it open, and the 
little moth escapes. When ripe, the shell is very light, and, as 
the worm occupies but about one sixth the inclosed space, the 
slightest motion will cause the seed to rock from one of the flat 
Sides to the other. But the seed is often made to jerk and jump, 
and, though this has been denied by many authors, Mr. Riley 
had had abundant proof of the fact, and had seen the seed jerked 
several lines forward at a bound, and raised a line or more from 
the surface on which it rested. If the seed be cut, the worm 
will'soon cover up the hole with a transparent membrane of silk ; 
and if two of the opposite angles be cut, the movements of the 
worm can then be seen, if the seed be held against the light. It 
thus becomes evident that the jerking motion is conveyed by the 
worm holding fast to the silken lining by its anal and four hind 
abdominal prolegs, which have very strong hooks, and then draw- 
Ing back the head and fore-body and tapping the wall of its cell 
with the head, sometimes thrown from side to side, but more 
often brought directly down, as in the motion of a woodpecker’s 
head when tapping for insects. In drawing back the fore-body 
the thoracic part swells, and the horny thoracic legs are with- 
rawn, so as to assist the jaws in receiving the shock of the tap, 
which is very vigorous, and often given at the rate of two a sec- 
ond and for twenty or more times without interruption. It is 
remarkable that this; of all the numerous seed-inhabiting Lepi- 
dopterous larvæ, ‘should possess so curious a habit. The seed 
will move for several months, because, as with most Tortricid- 
~~ larvæ, this one remains a long time in the larva state after 
coming to its growth and before pupating. 
t. Barnes gives the following account of the plant, received 
through Captain ‘Polhamus, of Yuma, A. T. It seems to 
called both Yerba de flecha and Colliguaja by the Mexicans : 
“ Atrow-weed (Yerba de flecha). This is the name the shrub 
that produces the triangular seeds that during six or eight 
