6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 95 



or an arthropod (C) is invaded by the coelomic mesoderm, and may 

 contain one, two, or three pairs of coelomic sacs. 



According to this view, the preantennae and antennules of the 

 Arthropoda are comparable with the prostomial tentacles and palpi of 

 the Polychaeta, and the absence of individual coelomic sacs associated 

 with the cephalic appendages of the latter may be attributed to the 

 imperfect development or more primitive condition of the prostomial 

 mesoderm in the annelids. The known facts of comparative embry- 

 ology show clearly that the " segmentation " of the preoral mesoderm 

 is highly variable, and it is perhaps significant that the most complete 

 example of it has been found in an insect. In any case, the idea that 

 the prostomium, or acron, of the articulate animals is a preoral trunk 

 region inherently devoid of coelomic mesoderm is evidently a falla- 

 cious concept based on the condition in the specialized trochophore 

 larva of polychaete annelids and other invertebrates. There would 

 seem to be no morphological reason why the mesoderm bands should 

 not encircle the blastopore, surrounding the mouth anteriorly as they 

 do the anus posteriorly. On this theory it is easy to accept the claim 

 of histoneurologists that the oculo— antennal part of the arthropod 

 brain (protocerebrum and deutocerebrum) corresponds with the pro- 

 stomial brain of the annelids, and that the second antennal brain lobes 

 represent the ganglia of the first postoral somite. 



Following the concept thus developed that the first antennal and 

 preantennal coelomic sacs, when present in the Arthropoda, are pri- 

 marily adoral in position and lie in a region of the head (the acron) 

 corresponding with the prostomium of an annelid, the entirely prac- 

 tical plan is here adopted of designating numerically the truly postoral 

 segments beginning with the tritocerebral somite as Segment I (fig. 

 I C, /). The appendages of this segment are the second antennae of 

 Crustacea, the chelicerae of Chelicerata (C, 2 Ant, Chi), or the cor- 

 responding embryonic rudiments of these organs found in many 

 Hexapoda. In any case, there is no doubt as to the identity of the 

 tritocerebral somite, and there is no question that it is the first postoral 

 segment of the adult in all the Arthropoda. 



Some of the arthropods are epiniorphic, the definitive number of 

 body segments in such forms being established at the end of embry- 

 onic development ; others are anamorphic, in which case either a fixed 

 or an indefinite number of segments is added during postembryonic 

 growth. The generation of new segments appears always to take place 

 at one point, which is a zone of growth located between the last- 

 formed somite and the terminal periproct, or telson. The occurrence 

 of anamorphosis is well known in Crustacea, Diplopoda, Chilopoda, 



