398 
lar opening is not economieal: for ‘the 
application of odorous, ill-tasting or ir- 
ritating fluids... If it is provided with 
means.for throwing its secretion, as it is 
in Cimbex and Harpyia, it is‘ somewhat 
more efficient at long range. The chanc- 
es, however, of a waste of. the secreted 
matter are still great, and, as Jérdens 
has observed; the larva of Harpyia uses 
its stream-throwing gland only when 
much disturbed... For the:application of 
an ill-tasting fluid the modified form. of 
gland found in the knobs of the larva of 
Attacus cecropia seems to .be the best. 
The secretion is set. free by:the very ac- 
tion of the enemy of the Jarva. ‘The use: 
of a strongly irritating fluid attains its 
highest development in. means of injec- 
ting the fluid into the flesh of the attack- 
ing animal, as is done by Hyperchiria 
and by many hymenoptera.. The great- 
est economy ip .the use: of:an odorous 
fluid is attained by.exposing suddenly a 
large surface moistened with the fluid.to 
the surrounding: air.. This is -accom- 
plished by osmateria, as in the larvae of 
Papilio, and by various evaginable ap- 
pendages, in other insects. 
with hair these appendages:expose still 
more surface to the air, and thus: give 
still better results in rendering the sur- 
rounding air odorous, whether for pro- 
tective or for sexual purposes. » For .sex- 
ual purposes a less penetrating, often to 
us a less disagreeable, odor, and. less of 
it, is necessary than for protective pur-: 
poses.: It is perhaps for this reason that: 
the evaginable appendages of Spilosomna, 
and of some other lepidoptera, appear 
to us always.dry. The reason for the. . 
Clothed. 
PSYCHE 
invagination of odor-distributing appen-°: \- 
dages is a double one; first .when inva- i 
ginated their necessarily delicate surface « « 
is less Hable to injury, and, second, their 
surface can be kept constantly moistened 
and ready for use without loss of the. 
odorous fluid.. Besides this,.too,-is the: 
general rule:that the appendages of loco-: - 
motory. animals are, as far as is: possible 
in attaining the purpose for which they re 
are developed, normally. invaginated or 
retracted to avoid hindrance to. locomo- : 
tion, while the corresponding organs of . 
sessile: animals and: plants attain their 
surface-devclopment by evagination. «. 
1 -Secretions which are intended to pros — - 
tect by «their odor usually..protect . by 
their taste also, for taste and smell.are so u 
nearly related senses in higher animals, 
while such glandsas secrete .ill-tasting u 
fluids, .if these fluids are volatile, .also 
usually: -protect by their odor. Still it - ‘ 
may be safely asserted, as a general rule, 
that odor protection is.accomplished by 
evaginable organs. 
The chemical substances: proved iby 
actual test to be secreted. by external - 
glands of insects are. few in number, and, 
as I hope sometime to make this the sub- 
ject of.a special paper, I should be glad ‘ 
to have my attention called to. any seere- 
tions of insects, especially to such: as 
can be obtained in sufficient quantity . 
for chemical examination. i + 
‘In the preceding notes I have sought 
to: call attention to glands of insects,:in ~ 
hope that some American, students of ve 
insects, who have not enlisted already «in. u 
that army of ‘species-describers, which a 
represents so-called American entomo-. . 
