PAPILIO III. 
ik 
On 15th Oct., I had several larvae feeding. Of these but one went to chrysa- 
lis before frost killed the leaves of the food plant and caused the loss of the re- 
mainder. This one yielded Telamonides $, 15th April, 1871. 
These observations therefore shewed that from Telamonides came Marcellus the 
same season and Telamonides in the following spring, that from Marcellus came 
successive broods of Marcellus the same season, and from the last brood Tela- 
monides in the spring. 
It also appeared that while there was a general limit to the duration of the 
chrysalis state, namely about 12 days, there were frequent exceptions, the imagos 
then emerging at irregular periods and some of nearly every brood living in chrys- 
alis till the following spring. 
The rapidity of growth from the egg was surprising, as compared with our other 
Papilios. In 1869, by obtaining very young larvae within a day or two after I 
had had butterflies emerge from chrysalis, I became satisfied that, besides the first 
brood from Walshii or Telamonides, there are three successive broods of Marcellus 
and the larvae of the fourth give chrysalids that go over the winter, thus making 
five broods per year. 
These observations failed to determine the connection between Walshii and 
the other two forms, though I had good reasons for feeling confident as to what 
that connection might be. F or example, from twenty-seven chrysalids obtained from 
larvae found on the food plant in Sept. 1868, and which it is now plain must have 
come from the last brood of Marcellus, emerged twenty-seven imagos between the 
2d and 21st April, 1868. Of these, twenty-six were Telamonides, and one was Wal- 
shii. Also from chrysalids of Sept. 1869, emerged two var. Abbotii in March, 1870. 
Out of fifty-seven chrysalids from larvae found on food plant in last of June 
and 1st of July, 1868, which were probably from first brood of Marcellus, forty- 
five produced Marcellus within the usual period; but five went over the winter, and 
between 24th March and 8th April, 1869, yielded five Telamonides. Another 
larva fed in August, 1868 (2d or 3d brood of Marcellus) produced Telamonides 
23d March, 1869. 
In the spring of the present year, (1871) Walshii was unusually abundant 
and it seemed to me, at the expense of Telamonides, which was comparatively 
scarce. On the 10th April, I confined three Walshii with the food plant, and by 
12th, had obtained one-hnndred and twenty-five eggs. On the 16th, I confined 
two others of same type separately and obtained many more eggs. The larvae from 
the last laying overtook in growth those of the first so that but one day intervened 
between the first chrysalis of each. Between 17 th and 23d May, all the survivors had 
changed, numbering seventy. From these emerged, between 1st and 6th June, 
fifty-eight butterflies, of which 22 $, 34 ?, were Marcellus, one $, Walshii and one 
