THE DEVONIAN EOCKS OF CANADA. 87 



split longitvidiually, so that although their marginal outlines and internal structure are 

 for the most part well exhibited, both sides of the outer or tegumentary surface of each 

 of these fins is buried in the matrix. In consequence of the peculiar way in which 

 these fins happen to be exposed in the few specimens in which they are preserved at 

 all, it happens that no external lobation of either the pectorals or ventrals is actually 

 visible in any of the specimens of Eusthenopteron in the Survey collection, but it by no 

 means follows that these fins were not lobate. The paired fins of TrisHchopterus are 

 known to have been lobate, and, as will presently be shewn, there is such a close re- 

 semblance in internal structure between the pectorals and ventrals of that genus and 

 those of Eusthenopteron, that it is most likely that those of E. Foordi were lobate also. 

 The marginal outlines of both pectorals in the adult fish are well seen in specimen 

 No. 1. In all stages of growth the pectorals are a little larger than the ventrals, but in 

 the adult condition, as exemplified by specimen No. 1, the marginal outline of the former 

 is obovately subtriangular. 



The bony supports of the fin-rays of the pectoral are well exhibited in specimen No. 

 4. In this specimen, immediately behind the clavicle, there is some obscure, bony matter, 

 but it is impossible to decide whether this consists of the expanded end of one compara- 

 tively large bone, such as the base of the coraco-scapular ossification, or of two wholly 

 or partially separated small bones. From the posterior termination of this bony matter 

 there proceed two small, short and separate bones, which may be designated respective- 

 ly as Nos. 1 and 4, as No. 1 is the first of an upper series of three Ijones which it will 

 be convenient to enumerate as Nos. 1, 2, and 3, and No. 4 the first of a lower series of 

 .three, which will here be referred to as Nos. 4, .5, and 6, the whole being thus numbered 

 on Plate VI. The first bone of the upper series (No. 1), and the first of the lower (No. 

 4), are both coucavely constricted in the middle, but the former is much broader and 

 shorter than the latter. To the truncated, somewhat expanded and slightly bifurcated 

 posterior extremity of the first of these upper bones (No. 1), there are artici^lated two 

 bones (Nos. 1 and 4), and, finally, to the po.sterior termination of the second bone of 

 the upper series (No. 2), there are articulated two other bones (Nos. 3 and 6), which, in 

 their turn, bear a very close resemblance to Nos. 2 and 5. In this particular specimen, 

 therefore, there are six ossified segments of the archypterygium preserved, in addition to 

 the obscure bony matter next to the clavicle. Of these, the three that are uppermost 

 (Nos. 1, 2 and 3) are almost exactly alike, and are squarely truncated at both ends. In 

 Nos. 1 and 2, however, there are articulating surfaces at both ends, whereas in No. 4, 

 from which only the ordinary fin-rays appear to proceed, there is no well defined articu- 

 lating surface posteriorly. The three lower segments (Nos. 4, 5 and 6) appear to be 

 ]3recisely similar in shape and are each truncated in front and narrowly rounded behind. 

 From their relative position it might be assumed that these three upper segments repre- 

 sent the radiais and the three lower ones the basais, but judging by their shape and 

 from analogies with the internal structure of the pectoral in Ceraiodus, it seems more 

 probable that the three upper segments of the anterior portion of the corresponding fin 

 in this specimen oï Eusthenopteron, may re^jresent the axial or medial elements of the fin, 

 and the lower segments the radial element. However this may be, the fin-rays which 

 spring directly from these six bony segments are long, slender and closely arranged. 

 At their commencement on each side of the fin the rays are short, but they rapidly 



