whose larve are external feeders. 345 
of similarity of structure and habits would be dis- 
coverable. 
The first point that occurred to me was that traces of 
abdominal legs, like those ef Hriocephala, might perhaps 
occur in newly-hatched larvee of some species of Zygxena 
or Limacodes. So: far, I have failed to detect such a 
structure, but find it recorded that the larva of Lagoa 
crispata possesses additional abdominal legs. I have not 
yet succeeded in obtaining eggs of this species. The 
pupa is unquestionably Limacodid, though Packard calls 
it a Liparid. Though many systematists consider Liparids 
and Limacodids to inosculate, the pupze prove them to be 
about as far apart, phylogenetically, as they well can be. I 
succeeded in obtaining eggs and young larve of Parasa 
chloris, and in rearing one larva; but though this is a 
most curious and interesting larva, it did not present any 
extra abdominal legs. It confirmed, however, the obser- 
vations on Limacodes testudo, which afforded me some 
very curious facts, and some very unexpected confirmation 
of the suspected relationship to Hriocephala. 
Liimacodes has suckers to the first eight abdominal 
segments, though the first and last of these are poorly 
developed ; these suckers are probably homologous with 
prolegs, and also with the eight pairs of abdominal legs 
of Hriocephala. 
When the larva of Limacodes testudo has completed 
its development within the eggshell, which is very easy 
to watch, owing to its transparency, flatness, and the 
facility with which the moth can be induced to lay them 
on glass, it is free from any spines or processes, but at 
the period of hatching certain long spines are rapidly 
developed; of these there are at least four on each 
segment, arranged as a dorsal and lateral series on either 
side. The ordinary tubercles can be detected as faint 
dots, but these spines are quite independent of the ordi- 
nary tubercles, and differently placed, and correspond in 
position to no larval processes I am aware of, except 
those of Hriocephala. 
In testudo the dorsal series on one side, though con- 
sisting of one spine on each segment, has them placed as 
though they were the double rows of Hriocephala, with 
alternate spines omitted, 7.e., the inner spine is wanting 
on the first abdominal segment, the outer one on the 
TRANS. ENT. SOC. LOND. 1894,—PART 11. (JUNE.) Z 
