NO. 2 DRAGONFLY LARVA — SNODGRASS 3 



said to be of ectodermal origin and to degenerate after the last moult. 

 From experiments on larvae of Aeschna cyanea, Deroux-Stralla 

 (1948) concludes that the ventral glands play an important role in 

 moulting and metamorphosis. She finds that larvae of the penultimate 

 stage from which the glands have been removed moult abnormally or 

 incompletely, while gland removal from last instar larvae results in a 

 complete suppression of metamorphosis. The presence of thoracic 

 endocrine glands has apparently not yet been noted in Odonata. 

 Wliedon (1938) suggests that the oenocytes associated with the aortic 

 diverticula, the activity of which is greatly increased in the adult, may 

 have some endocrine function. 



Hormones, of course, have nothing to do with the kind of structural 

 changes that will take in the tissues at metamorphosis. The course of 

 development may be inhibited or released from inhibition by hor- 

 mones, but it is determined in advance by the hereditary factors in- 

 herent in the tissues, which go into action when the hormonal balance 

 creates the proper environment. 



From the work of investigators on insect hormones in general it is 

 evident that the corpus-allatum hormone has the same function in the 

 young of ametabolous or hemimetabolous insects that it has in holo- 

 metabolous larvae, which is that of arresting development between 

 moults. We might conclude, therefore, that this hormone primarily 

 had no relation to metamorphosis, since it would appear that any ani- 

 mal having a nonexpansible integument during each of its develop- 

 mental instars must necessarily have some means of inhibiting the 

 growth of juvenile tissues between moults. However, Wigglesworth 

 (1953) finds that an adult of Rhodnius given both moulting hormone 

 and juvenile hormone will moult again, but shows a partial reversion 

 to nymphal characters. He concludes, therefore, that the corpus- 

 allatum hormone is not entirely inhibitory, but has a more positive ac- 

 tion on the juvenile tissues by actively favoring their differentiation, 

 since it does so even in an imaginal instar. 



Several writers have remarked that the metamorphosis of the 

 Odonata is essentially holometabolous, though no pupal stage inter- 

 venes between the larva and the imago. The dragonfly at least demon- 

 strates that as great a transformation can be accomplished by a direct 

 change of growth at one moult as many of the so-called holometab- 

 olous insects accomplish with two moults. According to Munscheid 

 (1933) the transformation in Aeschna cyanea occupies as much as 12 

 days before the adult comes out of the larval cuticle. The transforma- 

 tion period, therefore, is equivalent to the pupal stage of a holometab- 

 olous insect, and from this it might be argued that the pupa represents 



