454 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1931 



that the procephalon is composed of the prostomium and three post- 

 oral somites, and that the gnathocephalon contains three somites is 

 currently accepted by most entomologists as approximately repre- 

 senting the facts, but there are several reasons for questioning if the 

 evidence of segmentation in each of the two parts of the head has 

 been rightly interpreted. 



A comparative study of the brain in the Arthropoda and Annelida 

 leads to quite a different concept of the head segmentation than that 

 derived from a study of the appendages. According to Holmgren 

 (1916) and Hanstrom (1928), the three divisions of the arthropod 

 brain, ordinarily assumed to correspond with three segments in the 

 procephalic head region, represent only two primary nerve centers 

 instead of three. The first of the two primary parts, these writers 

 claim, represents the prostomial archicerebrum of the Annelida, and 

 is secondarily differentiated into the definitive forebrain and mid- 

 brain (protocerebrum and deutocerebrum) of the Arthropoda; the 

 second primary part (tritocerebrum) is formed of the first pair of 

 ventral, postoral ganglia, which have united with the primitive 

 preoral ganglion, but retain their ventral commissure. 



As a corollary to this theory we should have to believe that the 

 cephalic lobes of the arthropod embryo (fig. 5, Pre) represent the 

 annelid prostomium, with the first (tritocerebral) segment generally 

 more or less united with it. The preantennal and antennal append- 

 ages, and the eye stalks of Crustacea then become organs equivalent 

 to the annelid prostomial tentacles (fig. 1 B, Tl)^ while the second 

 antennae are the first true segmental appendages. There is much 

 in the structure of the brain and in the innervation of the head to 

 support this theory. On the other hand, the preantennal and anten- 

 nal rudiments of the embryo appear to be true ventral, postoral 

 appendages homologous with the second antennae and the mouth 

 appendages, and the presence of three pairs of coelomic sacs in the 

 procephalic region indicates that there are here three true segments 

 in addition to the prostomium. 



Another problem in the theoretical morphology of the insect head 

 is one that concerns the number of somites that enter into the com- 

 position of the gnathal region of the definitive head capsule, and 

 the homologies of the mouth-part appendages of insects with those 

 of the Crustacea. 



The insect mouth parts, as we have seen, include a median, tongue- 

 like lobe of the ventral wall of the head, lying between the mandibles 

 and the maxillae, known as the hypopharynx. In some insects the 

 hypopharynx has a pair of lateral lobes, and in such cases the median 

 part of the organ is distinguished as the lingua (fig. 7 A, Lin), and 

 the lateral lobes as the superlinguae (Slin). The superlinguae are 



