no THE EMBRYOLOGY OF THE HONEY BEE 



while a straight line connecting them would pass close to the 

 posterior edge of the stomodaeal imagination. Examination of 

 cross sections through the region of this segment at a stage a 

 trifle older than Stage VIII demonstrates that the bases of these 

 swellings are nearly in line with the bases of the rudiments of 

 the mandibles and maxillae, but that these swellings do not resem- 

 ble the rudiments of the mouth parts in their structure. If the 

 appendage-like swellings on the tritocerebral segment at Stages 

 VIII or IX are compared with the rudiments of the antennae or 

 the mouth parts a difference in their histological character be- 

 comes at once apparent. The rudiments of the true appendages 

 are ectodermal evaginations composed of numerous small and 

 slender cells with a core of mesodermal tissue; those of the trito- 

 cerebral swellings on the other hand are composed of an outer 

 layer of cells destined to form hypodermis, and an inner layer 

 of neurogenic cells, continuous cephalad with the rudiments of the 

 protocerebrum and deutocerebrum, caudad with those of the ven- 

 tral cord. In brief, the tritocerebral swellings are composed princi- 

 pally of neurogenic tissue, and at a later stage the tritocerebrum is 

 actually split off from the ectoderm of this region. With these facts 

 in view the probability that these swellings represent appendages 

 diminishes and it therefore seems more appropriate to consider 

 them as exaggerated ganglionic swellings. At Stage X the trito- 

 cerebral swellings have become almost invisible, this region of the 

 head having flattened out, and before Stage XI is reached the 

 swellings have totally disappeared. Biitschli (1870) states that 

 these rudiments fuse to form a sort of lower lip for the larva. 

 An inspection of Stages VIII to X shows that this is not the case, 

 but that there is on each side a fold which appears to be a con- 

 tinuation of the cephalic edge of the mandibular rudiments and 

 which extends cephalad and mesiad and terminates on the ex- 

 ternal side of one of a pair of rounded eminences. An inspection 

 of the transverse sections shows that these eminences are only 

 the anterior ends of the neural ridges, as shown in the figures of 

 these stages, and it seems also probable, judging from the rela- 

 tion of the parts, that these swellings belong rather to the mandi- 

 bular than, to the tritocerebral segment. Recently (1909) 

 Hirschler has described in Donacia a pair of similar swellings, 

 situated on the tritocerebral segment near the ventral mid-line, 



