PTEBICHTHYS MILLBRI. 99 



Quite in front the cranial buckler is seen, with some remains of the ocular plates 

 in the large orbit, while at the left side the much displaced extra-lateral {e. I.) is 

 seen in impression. A very good view is here obtained of the form of the pectoral 

 limbs, the tapering contour of the distal segment characteristic of the present 

 species being well shown, while we may also note the fine denticulation of the free 

 edges of the two proximal external marginal elements. The somewhat broadly 

 ovate outline of the ventral surface characteristic of the " latus " type is also very 

 apparent, the anterior and posterior ventro-lateral plates being nearly of the same 

 length, and the distance from the posterior margin of the carapace to the middle 

 of the median ventral plate being equal to the breadth of the whole ventral surface 

 at the middle point. The median ventral (m.v.) looks large in comparison with the 

 same plate in the previously described specimen (PI. XIX, Fig. 3), but is due to its 

 entire internal surface, unoverlapped by any other element, being visible. 



Fig. 2 of the same plate represents a specimen in the Williamson collection in 

 the Manchester Museum, for the privilege of figuring which I am indebted to 

 Dr. W. E. Hoyle. It has been already alluded to at p. 69 of this work as having 

 furnished evidence for the form of the caudal fin in my restored figure on p. 65, 

 Fig. 36. Here we have a dorsal view of the body-carapace, which is, however, 

 obliquely cut off by the stone in front and on the right side, a portion of the left 

 pectoral appendage being, however, seen below and in front. The great interest 

 of this specimen centres, of course, in the heterocercal caudal extremity. At this 

 part the prolongation of the body axis turns slightly upwards, the dorsal margin 

 being set with narrow oblique fulcral scales which are distinctly seen to be dis- 

 tichous, or in two rows, while the fin-membrane on the ventral aspect has a low 

 triangular contour, getting gradually narrower posteriorly, and, like that of 

 Gejjhalaspis, devoid of any division into upper and lower lobes. 



Fig. 1, PI. XXII, represents a specimen from Achanarras, Caithness, lying 

 right on its side — a somewhat unusual position in which to find specimens of 

 Pterichthys. The tapering form of the pectoral appendage characteristic of PL 

 Milleri comes out very clearly, and the proportions of body to tail are well 

 exhibited. We note here the arrangement of the scales on the side in longitudinal 

 rows, the dorsal fin with its enlarged scales on the anterior margin, and the narrow 

 oblique fulcra on the upper border of the caudal body prolongation; unfortunately, 

 however, as is usually the case, the delicate caudal fin-membrane has almost entirely 

 disappeared, so that its contour is lost. 



Fig. 2 is the distal portion of the arm in a specimen from Tynet Burn, in which 

 the tapering contour is seen in a very typical maimer. 



The head-shield of Pterichthys Milleri seen from the internal surface, and minus 

 the loosely articulated extra-lateral plates, is given in PI. XX, Fig. 3, and displays, 

 in the middle of the orbit, the median or pineal plate and also some remains 



15 



