have the same form, being high and acuminate in front, and concavely cut out behind; their rays are rather coarser than those of the ventrals, and their transverse joints are considerably longer than broad. In no case is the caudal fin well preserved, its rays being more or less disjointed and broken up, but it is clear that the prolongation of the body in the upper lobe was delicate, the rays slender, and with distant articulations.

Remarks. As already indicated, this species agrees with R. Geikiei in the general form of the body, and, along with it, differs from all the other species of the genus with which I am acquainted in the closer approximation of the anal fin to the caudal ; in both, also, the flank scales are comparatively small. But the action of the scale sculpture is essentially different. In R. brevis such ridges as are seen on the flank scales have not the same short, sharp, straight, and parallel character which distinguishes those which in R. Geikiei pass to the denticulations of the posterior margin; the ruga ornamenting the cranial roof bones are also coarser, more flattened, and less capable of being described as "striæ."

Geological Position and Localities..-From the Calciferous Sandstone series. Five specimens are in my possession, four of which are in ironstone nodules from Wardie beach; the fifth I found in a detached block of slaty carbonaceous ironstone, on the shore near the mouth of the Kenly Burn, between St Andrews and Kingsbarns.

Rhadinichthys carinatus, Agassiz sp.

Palæoniscus carinatus, Agassiz; "Poissons Fossiles," vol. ii. pt. i. p. 104 (1835); "Atlas," vol. ii. tab. 4b, figs. 1 and 2.

The form of the body is slender, the length of the head being contained about 43 times in the total up to the bifurcation of the caudal fin, and equalling the greatest depth of the body at the origin of the ventral fins. The head is very elegantly shaped, rather depressed above, but pointed when seen in profile. The bones of the cranial shield are ornamented with sharp, delicate, wavy ridges or "striæ." A specimen from Wardie in my collection shows most clearly the presence of extensive ossification in the periotic and alisphenoid regions of the cranium, and, moreover, a cast of the posterior semi-circular canal of the ear is most distinctly and

beautifully preserved. The opercular bones are rather small; the jaws remarkably slender and delicate; the external ornament of the facial bones consists, like that of the cranial shield, of delicate wavy ridges. No specimen I have as yet seen affords, however, the smallest trace of teeth. The paired fins are rather small. The pectoral is in no case well preserved, but there is sufficient evidence that its longer rays were not articulated till towards their terminations. The ventrals are still smaller, and are equidistant in position between the pectorals and the commencement of the anal. The median fins are, on the contrary, rather large in proportion to the size of the fish. The anal commences rather in front of a point equidistant between the origin of the ventrals and that of the lower lobe of the caudal; it is high and acuminate in front, and pretty deeply cut out behind; the length of its base is equal to that of its longest rays. The dorsal is similar in shape to the anal, and commences only slightly in front of the origin of the latter. The caudal is pretty large, and deeply cleft, the body prolongation along the upper lobe being, however, comparatively delicate. The rays of all these fins are slender, ganoid, and smooth externally; their articulations tolerably distant; the fulcra on their anterior margins obvious, though minute.

The scales are remarkable for their large size on the flanks, while on the belly they suddenly became very low and narrow; they diminish rapidly in size posteriorly, and on the tail pedicle are rather small. Externally they are brilliantly polished and nearly smooth; on the flank scales, however, a few shallow furrows extend a little way forwards from the posterior margin, which is, as usual, sharply denticulated.

Remarks. The original specimen, collected by Lord Greenock at Wardie, and figured by Agassiz, is in the collection of the Royal Society of Edinburgh. It is very imperfect, showing no fins save the mere origins of the anal and caudal, the former of which seems to have been overlooked by Agassiz. On the other hand, I cannot, by the most careful examination, verify his statement regarding the presence of small teeth, "en brosse" upon the jaw. I am indebted to Professor Turner for the loan of a specimen from Cornceres, in Fifeshire, which belonged to the late Professor Goodsir; it is possible that this may be the "Catopterus fusiformis" which Mr

Robert Walker states was described by Professor Goodsir in 1838 before a meeting of the Literary and Philosophical Society of St Andrews, and which, according to Mr Walker, he characterised as closely allied to Palæoniscus, but differing from it "in wanting the scaling or false rays along the anterior ray of the fins, and also in the dorsal fin being opposite the anal."* In this specimen the dorsal is, indeed, nearly opposite the anal (as in others of the same genus as well as species), but, on close examination, fulcra are distinctly observable. I have not been able to obtain Professor Goodsir's description, or even to ascertain exactly where it is to be found, but as it appears to have been published only in a provincial newspaper, the generic name Catopterus must remain with the Triassic semi-heterocercal Lepidosteids, to which it has been applied by the Messrs Redfield.

Geological Position and Localities. From the Calciferous Sandstone series. In ironstone nodules from the shore at Wardie, near Edinburgh; on the shore near Pittenweem, in Fifeshire, in the collection of Mr J. W. Kirkby; in a compact grey limestone at Cornceres near Kilrenny, collected by the late Professor Goodsir.

Rhadinichthys tenuicauda, sp. nov. Traquair.

Length, about 2 to 3 inches; form slender, narrowly tapering posteriorly. Head bones sculptured externally, with closely set flexuous ridges; dentition not visible. Scales rather large for the small size of the fish, and especially large on the sides of the abdomen, becoming smaller on the back and towards the tail. The posterior margins of the scales are denticulated, their external surfaces nearly smooth along the sides of the body and on the tail pedicle, but along the back, especially between the dorsal fin and the head, they are marked by a few rather prominent oblique ridges; a similar ornamentation, though rather finer and closer in character, makes its appearance also on the scales along the ventral margin. The pectoral fin is small; the non-articulation of its stouter rays seems to prevail over a less extent of the fin than is usual in this genus. The ventral is still smaller. The dorsal and anal are of medium size, and nearly opposite each other; their rays are fine

* Trans. Geol. Soc. of Edinburgh, vol. ii. pt 1. (1872), p. 124.

and distantly jointed. Between the anal and caudal fins there is a considerable interval. The caudal is inequilobate, with attenuated body prolongation; its rays are slender, smooth, and with distantly placed articulations. The fulcra are very obvious in all the fins, notwithstanding the small size of the fish.

Remarks. This species is allied to R. carinatus in general form, and in the large size of the scales of the flank, but differs obviously from it in the general larger proportional size of the scales, in the ornate character of those along the dorsal and ventral margins, and in the greater delicacy of the fins, and smaller number of their rays. The head seems also to be larger in proportion to the size of the fish.

Geological Position and Localities.-A fish of the middle or "Edge Coal" group of the Carboniferous Limestone series, the specimens above described being from Wallyford, near Tranent, in the collections of Mr D. J. Brown; and from Possil, near Glasgow, in the collection of Messrs James Thomson and John Young, F.G.S. To the kindness of these gentlemen I am indebted for the opportunity of describing this most distinct and interesting little form.

On the Cranial Osteology of Rhizodopsis, and on some points in the Structure of Rhizodus. By Dr R. H. Traquair.

(Abstract.)

Read May 21, 1877.

This paper contains an account of the cranial structure of Rhizodopsis as far as this can be elucidated by an examination of a suite of specimens from the Coal Measures of Fenton, Staffordshire, in the collection of Mr Ward of Longton, supplemented by additional specimens from various other localities.

The cranial shield of Rhizodopsis resembles in general form and structure that of the Saurodipterini, though the surface of the bone is elaborately sculptured instead of presenting the smooth glistening aspect characteristic of the external plates and scales of the last named group. The posterior part of the shield is composed of two elongated parietals, and having on its outer margin two smaller plates, anterior and posterior, which may be considered as the postfrontal and squamosal respectively. The anterior or fronto-ethmoidal

portion forms the depressed rounded snout, the front margin of which above the mouth, is formed by two small dentigerous præmaxillary bones entirely differing in form from the bones described as such by Messrs Hancock and Atthey. The gape is wide, the hyomandibular bone being slightly inclined backwards; there is apparently no symplectic. The form of the maxilla is well known; the structure of the mandible is exceedingly complex. Its dentary element is found to correspond with the bone interpreted by Messrs Hancock and Atthey as "præmaxilla," but turned in a contrary direction, i.e., with its toothed margin upwards instead of downwards as supposed by them; it bears the anterior laniary tooth and the outer row of small teeth. The laniaries posterior to the front one are borne on separate internal dentary ossicles, the presence of which is clearly proved by a portion of a jaw of a large specimen of Rhizodopsis in the Edinburgh Museum. The external aspect of the lower jaw is completed by at least two infra-dentary plates situated below the inferior margin of the dentary, behind which there is another covering externally the articular and angular region of the jaw, and which is probably the equivalent of the angular of other ganoids. On the internal aspect of the mandible there is, besides the internal dentary or laniary ossicles, a well-marked splenial. The orbit is very anteriorly placed; as in Megalichthys, three plates cover the cheek behind the suborbitals, corresponding to the single large one in Osteolepis. The author does not find Professor Young's statements that the jugular plates of Rhizodopsis are "in two pairs, principal and posterior," and that there is "no trace of median or of lateral plates," corroborated by the specimens under examination. On the other hand, he finds one pair of large principal jugulars with at least four lateral ones on each side, as well as very distinct evidence of the presence of a median plate behind the symphysis of the jaw. The shoulder girdle is provided with well-developed infraclavicular plates.

The lower jaw of Rhizodus Hibberti is found to have essentially the same structure as that of Rhizodopsis. The dentary element is of the same form, and bears the anterior large laniary tooth and the outer range of small teeth, the posterior laniaries being borne on separate ossicles, which are sometimes found entirely detached.

VOL. IX.

3 N