58 THE APODID^: PARTI 



much more weakly developed than the longitudinal. 

 This is also the case in the Apodidae, where it is 

 almost entirely confined to the muscular bands which 

 run into the limbs, especially to those from the dorsal 

 surface. In the limbless part of the body, where the 

 longitudinal muscles form a complete clermo-muscular 

 tube, the circular muscle layer has entirely dis- 

 appeared. The commencement of the formation of 

 an exoskeleton renders it useless. We shall return 

 to this subject in discussing the musculature of the 

 limbs. 



The muscles attached on each side, just above the 

 ventral cord, to the membrane which encloses the 

 intestine and genital glands, and forms the intestinal 

 sinus, may perhaps best here be mentioned as in part 

 having arisen from the circular musculature. We 

 shall return to these also when we discuss the cir- 

 culation and the origin of the above-mentioned 

 membrane. 



Two especially interesting groups of muscles, of 

 unmistakably Annelidan origin, deserve particular 

 attention. These are the rows of dorso-ventral 

 muscles (Fig. 14, dv.) which pass between the intes- 

 tine and the genital glands in almost exact corre- 

 spondence with the longitudinal muscle dissepiments 

 so common among the Annelida (cf. Fig. 1 1, tni}. In 

 the Apodidae, these rows are composed of a kind of 

 lattice work of muscle bundles with definite points 

 of attachment, ventrally to the sinewy partitions of 

 the ventral muscle bands and thus indirectly to the 

 body wall, and dorsally to the segmental con- 



