20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 8o 



infold of the sternal wall. An inflection of the integument between 

 the prothorax and the mesothorax, and one between the mesothorax, 

 and the metathorax form intersegmental folds between the thoracic 

 sterna. The fold between the first and second segments bears laterally 

 on each side a small process, the furcilla, and medially an unpaired 

 process (the spina) ; the fold between the second and third segments 

 has only a median process. The musculature of the Dytiscns larva, 

 Speyer shows, is primitive in many ways ; both abdomen and thorax, 

 for example, are traversed by continuous latero-ventral bands of 

 muscle fibers. In the abdomen the fibers of these muscles are attached 

 to anterior folds of the sterna, except in the first segment where all 

 but one pair are inserted anteriorly on posterior parts of the metaster- 

 num, either on the f ureal arms and the supporting ridge, or on lateral 

 points corresponding in position with the furcillae of the first inter- 

 sternal fold of the thorax. In the thoracic segments the ventral 

 muscles are attached to the f ureal arms, to the furcillae, to the median 

 process, and to a transverse ligament at the posterior edge of the 

 segment, except in the prothorax where one set ends on the back of 

 the head and another on the cervical sclerites. Without going farther 

 into details of Speyer's account, it is clear that the sternal thoracic 

 muscles in the larva of Dytiscus are attached either to posterior parts 

 of the sterna, or to processes (furcillae and spina) of intersegmental 

 folds. The folds, Speyer says, appear to be derived from the anterior 

 part (acrosternite) of the sternum following in each case. 



In the adult of Dytiscus (Bauer, 1910), the principal ventral 

 muscles of the thorax consist of paired bundles of longitudinal fibers 

 extending between the sternal apophyses, and from the prosternal 

 apophysis to the back of the head. This is the general condition of 

 the longitudinal ventral musculature in the thorax of adult Pterygota, 

 except that where median apophyses are present some of the muscles 

 are attached to them. 



A study of both the Apterygota and the Pterygota, therefore, ap- 

 pears to indicate that the first step in the evolution of the thoracic 

 sterna consisted of a union or close association of the points of 

 attachment of the ventral muscles with the sternal plates preceding. 

 It may be questioned whether the folds bearing the muscle attach- 

 ments in the thoracic region ever formed antecostae of the sterna 

 following, as they do in the abdf)men, but it is reasonable to suppose 

 that they did, considering that the sterna of the Chilopoda are of uni- 

 form structure in all the limb-bearing segments, and that the ventral 

 thoracic muscles of the Protura are attached to the anterior margins 

 of the sterna (Berlese). Cliitinizations of the intersegmental folds 



