224 THE AMERICAN X. 1 Tl R A LIST. [Vol. XXXVI I . 



been attached to the head, near the anterior end of the series of 

 so-called -ill openings. The anterior pair of incisions, or open- 

 ings (of which I have found at least two additional pairs in front 

 of the six pairs described by Rohon) are found on the lateral 

 margin of the anterior part of the head, in precisely the position 

 where the hypothetical oar-like appendages should be attached 

 provided any were present. Moreover in Tolypaspis, Cyathaspis 

 and Pteraspis, there is a very conspicuous rounded incision on 

 the margin of the dorsal shield that has usually been regarded as 

 the site of the lateral eyes, apparently because there was no 

 other interpretation for these openings at hand, and because 

 otherwise the lateral eyes would appear to be absent. But, 

 these incisions also occur at points corresponding to the ones 

 where the oar-like appendages in Pterichthys and Bothriolepis, 

 and the anterior pair of incisions in Tremataspis, are located. 

 The conclusion to be drawn from these facts is that Tremataspis, 

 Cyathaspis, and Pteraspis were probably provided with jointed, 

 oar-like swimming appendages similar in character and location 

 to those of Pterichthys and Bothriolepis. 



Moreover, and this is the important point, if Tremataspis has 

 a pair of jointed, armored appendages attached to the anterior 

 margin of the ventral surface, they must have been attached to 

 one of the anterior pair of marginal incisions. If this is the 

 case, there is no escape from the conclusion that behind the oar- 

 like appendages there are as man}' more pairs of serially homol- 

 ogous structures (but not necessarily similar to them in form or 

 function) as there are pairs of marginal incisions. 



The assumption that Tremataspis is provided with jointed 

 pectorals is strengthened by the fact that Lindstrom 1 has 

 described what he considers to have been an appendage in 

 Cyathaspis. He states on p. 5 : "Along with some other frag- 

 ments was found, detached and broken into three pieces, what 



(PI I, fig. 9-12) It is plain that it is not a fragment of the 



