290 
ble tubercles in the larvee in question are of the same color as the 
skin and are armed with fine hairs or sete. 
The third view is one which is incapable at present of proof, so 
that we are driven to provisionally regard these processes as per- 
sistently evaginate repugnatorial, or at least scent glands or osmateria 
which have possibly lost their power of ejecting a poisonous or 
disagreeable spray or fluid, owing to the fact that by a change or 
transfer of function the spine-like setz are poisonous, thus function- 
ally replacing a set of organs originally actively repugnatorial. 
It is also to be observed that the fact that in the Hemileucidz 
these eversible glands are restricted to but two of the abdominal 
segments, shows that in the ancestral forms these structures may 
have been developed on all, or at least nearly all the abdominal 
segments. 
Lacoa, aS REGARDS ITs LARVAL, PuPAL AND IMAGINAL CHARAC- 
TERS, A GENERALIZED TYPE. 
We have already seen that in respect to its general appearance 
the larva of Lagoa is in some respects intermediate between the 
Cochliopodidz and Liparide. It resembles the former group in 
the short thick body ; in the head being concealed by the pro- 
thoracic hood, and in the venomous spines. 
On the other hand it resembles the Liparide in the hairy body, 
the hairs being finely plumose, a peculiarity of more common oc- 
currence in the Liparidz than in the Cochliopodide. 
As regards the cocoon, this is intermediate in form and texture 
between that of Orgyia, etc., and the Cochliopodide, but it more 
closely approaches that of the latter; it varies somewhat in density 
in different species, being usually quite firm and dense like parch- 
ment, nearly as much so as in those of the Cochliopodids, and also 
approaching them in shape, being oblong-cylindrical, oval, con- 
tracted at the anterior end, and with a separately spun lid, closing 
the anterior end. As Dr. Lintner has shown with many interesting 
details: ‘‘ The lid is woven by the caterpillar separately from the 
rest of the cocoon, and is not a section cut from it after its com- 
pletion ’’ (p. 142). 
The pupa is much like that of Limacodes, etc., the integument 
or cast cuticle being remarkably thin, and after the exit of the moth 
the antennz and legs, as well as the wings, are free from the body ; 
