175 



a long time before becoming a pupa, and the perfect insect 

 will then not be disclosed before the second summer after- 

 wards in the month of June. I should not forget to state that 

 when the caterpillar is at rest on the plant, it raises the fore 

 part of the body, and if disturbed wags it violently from side 

 to side. The head of the young larva is armed on the occiput 

 with two little corneous tubercles, which, however, disappear 

 after two or three moultino;s. This insect is the Notodonta sex- 



o 



guttata of my MS. ; but I wish some competent person would 

 compare the moth with Thyatira batis, and let me know if it 

 may not be near to that species in generic characters. 



My No. 38 (a very common insect here, and a form of which 

 we have several species) you referred in one of your letters to 

 OpJimsa crassiuscula. The discovery of the larva by your 

 little namesake removes the insect to another group, and proves 

 it to be a true Eudidia. Edward's specimens made cocoons 

 of a dried leaf, interwoven with a little silk, in 

 which the Iarva3 were changed to pupae, cov- 

 ered with a bloom like that of the pupa of 

 Catocala. The attitudes of the larvae, which 

 I have tried to represent, are very odd. Color 

 brownish white, with darker longitudinal 

 stripes, and two black dots on each side of 

 the body. The species is the Noctua gemma 

 Fabr., erectliea Cramer. 



I have also sketched three forms of Lima- 

 codes, making, with Smith- Abbot's pitJiecium and cippus, five 

 different forms of the American species of the genus. Some 

 writers state that the prolegs in Limacodes are retractile. 

 This is not true of our American species, which are entirely 

 destitute of prolegs. The alternate contraction of the muscles 

 attached to the abdominal folds serves, instead of prolegs, to 

 assist the insects in moving. 



The larva, represented on the left [PI. n, fig 10], stings by 

 means of the prickles wherewith the tubercles are beset, and 



