[75] THE EVOLUTION OF THE FINS OF FISHES. 1055 



in the Teleosts, sliowiiig that the raiivscular metamerism at the base of 

 the dorsal is directly dependent n]i()n the metamerism which is observed, 

 in the lateral muscular system, which, as we have already noticed, has 

 determined the metaraeric arrangement of the interspiuous or intermus- 

 cular cartilages and the erectores and depressores spinfe muscles which 

 erect and xlepress the rays of the vertical fins of the Teleosts. The 

 muscular bundles which vibrate or abduct and adduct the rays of the 

 anal of Teleosts have a similar origin, and in fact seem to develop from 

 the bundles which also give rise to the erectores and depressores spinas, 

 and therefore exhibit a similar metameric order of arrangement. 



Turning now to the consideration of the recti ahdominales muscles of 

 fishes, these are metamerically segmented, the segments corresponding 

 to the lower end of the thoracic muscular segments, as is shown in lon- 

 gitudinal sections of embryos of several species (Amiurus^ Alosa). The 

 musculature of the pelvic fin, which is closely affiliated serially with the 

 history of the recti, does not in the stages which I have examined show 

 clear evidence of its origin from the muscular segments above it, but 

 such is doubtless the case. The lower ends of those broad membrane 

 bones, the coracoid and clavicle of J. wiwrw.s, develoj) in the septa be- 

 tween and above and below the muscular segments of the isthmus, 

 which are clearly the serial homologues of the segments of the recti 

 muscles alluded to above. The hinder coracoid element of the pecto- 

 ral girdle in AmiwrMS develops in cartilage; the clavicular portion in 

 membrane below the three first segments of the isthmus, the perichon- 

 drium of the coracoid above the fouith and fifth muscular segments of 

 the isthmus, counting from in front of to behind the hyoid arch.* It is 

 thus rendered evident that muscular metamerism influences the for- 

 mation of even the shoulder girdle; and when we bear in mind the 

 fact that such a metameric order was already established during the 

 enterocoelous stage of development, or just a little beyond the gas- 

 trula, wlieu the muscular somites were being constricted off from the 

 hypoblast, we may realize how far back in the life history of an em- 

 bryo we may begin to trace the influences which determhie the ultimate 

 form of the vertebrate skeleton. We saw, for example, that the inter- 

 spinous cartilaginous bases of what afterwards become the osseous in- 

 terspinous pieces bear a relation to muscular metamerism ; that, in fact. 



* In the Salmon the supraclavicle develops as a membrane bone in the septum be- 

 tween the third and fourth muscular segments behind the auditory vesicle ; the post- 

 temporal appears in membrane above the post auditory segment, just beneath the 

 skin. The muscles of the pectoral, that is, its abductor and adductor, it appears to 

 me, probably develop from the lower ends of the third and fourth post-cephalic so- 

 mites. A slender slip of muscle, which is developed from the lower inner side of the 

 upper end of the first post cephalic muscular somite, passes obliquely back from the 

 hinder aspect of the auditory vesicle to be inserted into the upper end of the clavicle. 

 The adductor and abductors of the pectoral are pushed out on either side of the car- 

 tilaginous pectoral plate and inserted on its anterior aud posterior faces, resembling 

 the Arrangement which is permanent in Gastrostomus. 



