142 C. R. Eastman — Asterolepid Appendages. 



rests upon the unproven hypothesis that Asterolepids are highly 

 modified Crossopterygians. Or, to state the converse proposi- 

 tion, the theory that these two groups are genetically related 

 depends upon whether the modified limbs of the one can be homo- 

 logized with thelobate pectoral fin of the other, there being abso- 

 lutely no other characters which can be said to indicate com- 

 munity of origin. 



The absence of a lower jaw in Ostracophores, the dissimilar 

 arrangement and structure of their head-bones, armouring of 

 the body, single dorsal fin without either dermal rays or basal 

 supports, heterocercal tail, and absence of both pelvic and 

 pectoral girdles, are characters which emphasize the violent 

 contrast between forms like Pterichthys and Crossopterygians. 

 That the latter are descended from a heterocercal ancestor is 

 not to be questioned, but if we admit the soundness of Mr. 

 Regan's conclusion that Asterolepids are highly modified 

 Crossopterygians, how are we to explain their reversion to the 

 primitive heterocercal condition, after having passed through 

 the homocercal? The anterior pair of limbs could hardly have 

 become so highly modified, without the hinder pair having also 

 undergone specialization. But even assuming, for sake of argu- 

 ment, that the pelvic fins have become lost, we should expect 

 to find remnants of a girdle, and in any case some indication of 

 a pectoral arch, corresponding to these structures in Crossop- 

 terygians ; whereas in fact we do not. 



The Asterolepid paddle is not made up of articulated rays, 

 but is simply a muscular extension of the body encased in der- 

 mal plates. An appendicular skeleton is wanting, and the 

 external covering plates are attached to the body-armour by a 

 complicated joint, one of the dermal plates being pierced for 

 the passage of nerves and nutrient canals ; hence it is clear 

 that in structure and mode of attachment these limbs differ 

 radically fr/>m normal Teleost conditions. The fact that Ostra- 

 cophores possess a much greater antiquity than Crossoptery- 

 gians also militates against the assumption that they are modi- 

 fied descendants of the latter. It is likewise impossible to 

 reconcile the geological occurrence of the groups to which 

 Cephalaspis and Pterichthys belong with the view expressed 

 by Mr. Regan that the former is a specialized Asterolepid. 



There remains finally the theory, which appears to be not 

 very generally accepted, that Asterolepid limbs have been 

 independently derived. The absence of an appendicular skele- 

 ton, and the peculiar mode of attachment of these organs, offer 

 such striking contrasts to the fins of fishes as to make it impos- 

 sible to conceive of a homology existing between them. More- 

 over, paired appendages are absent, so far as known, in all 

 other Ostracojmores. Either they were formerly present, and 



