1890.] 
Ill 
[Packard* 
tera, appear to intergrade with the Pyralidae , 1 while the Tortricidae 
may be a direct offshoot from the Tineidae. 
Mr. Poulton has pointed out the origin of the Sphingidae from 
Aglia tau and Ceratoeampid-like forms, and our own observations 
on a number of ontogenies of the latter group confirm his opinion. 
It also seems quite evident from the ontogenies we now possess, that 
the Arctians and Lithosians have descended from the Liparidae, and 
that the latter may have sprung from the Lasiocampidae. The 
Zygaenidae(including the Glaucopidians) may be readily traced back 
to the Arctiidae. The Agaristidae may have had an independent 
origin from the Zygaenidae, as also the Castniadae. The Hepialidae 
and Cossidae have probably had a common origin, while the Thy- 
rididae and Sesiidae are so closely similar to the Sphingidae in their 
adult characters, that they may be late offshoots from the Sphin- 
gid phylum. 
As to the origin of the Hesperidae we need more light, but we 
would suggest that the other groups of Rhopalocera have been de- 
rived from Bombycine-like Lepidoptera. We have been led to this 
view from our studies on the ontogeny of the Bomlyces. We have 
for many years regarded this group as a generalized or ancestral 
one, though scarce^ able to give the reasons for the belief. But 
from a study of the freshlj’-hatched larvae, there now seems some 
foundation for the view that like the Ganoidea among fishes, the 
Labyrinthodonts among Batrachia, Theromorpha or Ichthyosaurus 
among Reptiles, and Creodonta among mammals, the Bombycine 
group is one rich in ancestral forms comprising tj-pes from which 
have arisen by an acceleration of development due perhaps to some 
more or less sudden change in their surroundings, isolated forms 
which have finally through geographical or other forms of segrega- 
tion given birth during the late mesozoic age to a series or con- 
stellation of secondaiy forms constituting our present more or less 
specialized groups of genera and species called families. Moreover 
the different families of Lepidoptera may not have originated one 
from the other in serial order, but in some cases independently of 
each other ; at all events there seems to be good ground for the 
view that not only the Sphingidae, but also the Agaristidae, and the 
Rhopalocerous families of Papilionidae, Lycaenidae and Nymphali- 
dae have more or less directly descended from at least Bombycine- 
like Lepidoptera. 
1 See Mr. Myriek’s paper Trans. Ent, Soc., London, 1889, 
