A. S. Woodward — Devonian Fishes, Canada. 5 



external face of the scale occasionally shows very delicate transverse 

 rugsB. Abraded scales, as the one figured, are very conspicuously 

 punctate, the pittings being arranged in transverse lines ; and ou 

 the inner aspect there are but feeble traces of a vertical rib. On 

 the whole, these scales are very suggestive of the so-called Porolepis 

 from the Lower Devonian of Spitzbergen.^ 



In the imperfect head and trunk just referred to, the caudal 

 extremity, with its small rhomboidal scales, is reflexed ; and close 

 to it, immediately in advance of the shield, there occur patches of a 

 fine shagreen-like material. Of this there are further traces between 

 the cornu and the squamationon the right side of the fossil. Though 

 much finer than the calcifications in the azygous fin-membranes of 

 the British species, the granules in question have probably served 

 a similar purpose. 



7. Cephalaspis, sp. 



A very small imperfect Cephalaspidian shield from Campbellton, 

 with well-separated tesseree in the middle layer, differs from the 

 corresponding shield of C. cnmpbelltonensis in the relatively larger 

 size of the denticles on the inner margin of the cornua. Unless this 

 be a character of immaturity, the fossil thus indicates a distinct 

 species. Fine granules mark the position of the opercular folds. 



DIPNOI {ARTSROBIRA). 



Genus PHLYCTiENASPis, Traquair. 



[Geol. Mag. [3] Vol. VII. 1890, pp. 60, IM.] 



8. Phlyctcenaspis acadica, Whiteaves, sp. PI. I. Figs. 7, 8. 



1881-89. Coecosteus acadieus, J. F. Whiteaves, Canadian Nat. n.s. vol. x. p. 94, 

 woodc, and Trans. Roy. Soc. Canada, vol. vi. sect. iv. p. 93, pi. ix. and 

 woodc. 



1890. R. H. Traquair, Geol. Mag. [3] Vol. VII. pp. 20, 60, PL III. figs. 1, 2, 



Of several specimens referable to the type species of Phlyctcenaspis, 

 two are especially fine— one exhibiting the outer aspect of the head- 

 shield, the other the inner or visceral aspect of the same. The 

 former is shown of the natural size in PL I. Fig. 7, and of the latter 

 the so-called " rostral " plate is separately represented in Fig. 8. 

 The first speciiTien is of great interest as having been crushed in 

 such a manner as to separate its component elements ; while both 

 specimens elucidate for the first time the precise nature of the 

 " rostral plate." 



The new specimens demonstrate that Dr. Traquair's determination 

 of the arrangement of the various elements of the shield is correct 

 in every particular ; and it is especially interesting to find that in 

 the original of Fig. 7 there is an anterior pair of bones (p.mx.), 

 additional to those previously discovered and evidently homologous 

 with the premaxill^ (Traquair) of Coecosteus. 



The statement that no median bone occurs over the pineal region 

 of Phlyctmiaspis, made in the Catal. Foss. Fishes Brit. Mus. pt. ii. 

 p. 277, must now be modified ; for both the new specimens under 

 1 A. S. Woodward, Ann. Mag. Nat. Hist. [6], vol. viii. (1891), p. 8. 



