THE EMBRYOLOGY OF THE HONEY BEE 163 



ments of the corpora allata (Fig. 6iA, Cor All). At the next 

 stage (XI) the lumen of these invaginations has nearly disap- 

 peared and the corpora allata now appear as irregular cellular 

 outgrowths springing from the mesial side of the mandibular 

 apodemes (Fig. 6iB). These cell masses are next constricted 

 off from the apodemes, but still remain in close contact with them. 

 Meanwhile the corpora allata are carried dorsad by the growth 

 of the mandibular apodemes and at Stage XII lie between the 

 latter and the inner ends of the posterior arms of the tentorium 

 where these join the anterior arms to form the central body. At 

 Stage XIII the corpora allata become attached to the ventro- 

 lateral angles of the antennal somites (Fig. 6iC) and soon after 

 lose their attachment to the mandibular apodemes. This is their 

 position at the time of hatching and for at least a considerable 

 period of the life of the larva. The corpora allata acquire their 

 characteristic globular and compact form during the final stages 

 of embryonic development. 



Heymons (1895) was the first to call attention to the bodies 

 named by him "ganglia allata," describing their development in 

 some detail in Forficula, and more briefly in Gryllus. Three years 

 later Carriere and Burger (1897) found that in the mason bee 

 (Chalicodoma) the ganglia allata have the same origin as in 

 Forficula, but these bodies do not, however, later fuse to form a 

 median body as they do in Forficula but remain attached to the 

 ventro-lateral margins of the antennal coelomic sacs, as in the 

 honey bee. These investigators expressed a doubt as to the ner- 

 vous nature of the "ganglia allata," since no nerve fibres could 

 be observed within them. In 1899 Heymons published a brief 

 paper on the structure and development of the corpora allata 

 of a walking stick (Bacillus). In this form these bodies lie one 

 behind the other, above the oesophagus and caudad of the pharyn- 

 geal ganglia, from which they receive nerves. In this insect the 

 corpora allata have the form of capsules within each of which 

 is a concentric layer of chitin secreted by the capsular wall. Their 

 development was found to be the same as in Forficula. 



The function and homology of these bodies is unknown. Hey- 

 mons (1899) surmised that they had a static function, but exci- 

 sion of the corpora failed to produce any disturbance in the loco- 

 , motion of the insect (Bacillus). Janet (i899a, 1900) discusses at 



