Review of the reptilia of the triassic

Because of their deeply amphicoelous articular surfaces and the hyposphene, these vertebrae give a strange impression, especially in the German Muschelkalk together with innumerable Nothosauridae. However such vertebrae could not belong to a sauropterygian.

Vertebrae of Pareiasaurus, Propappus, Tapinocephalus, Loxomma (Permian) have in the middle of the barely concave articular surface a deep hole that almost or completely passes through. The dorsal vertebrae of Dicynodon, Embolophorus, Dimetrodon, and Inostrancevia ⁽1) however have an hourglass shape in longitudinal section, exactly like the vertebrae in question, only the vertebrae of Dicynodon are shorter than those of the others named. The underside of the centrum from Mühlhausen, Thuringia, has almost exactly the same cross-section as a series of Anomodontia (Pareiasaurus, Tapinocephalus, Dicynodon, Sclerosaurus); the centrum is rather constricted in the middle, wedge-shaped ventrally, rather flattened right below, and again constricted laterally above the middle.

A hyposphene is known to me only from Dinosauria, Embolophorus, and Inostrancevia ⁽2). But the former have two-headed ribs and cannot be considered here. Thus the hyposphene seems to indicate Anomodontia. Other characteristics also lead to the same conclusion.

Because the vertebrae are found in the marine main Muschelkalk, it is very likely, even if by no means necessary, that they belong to a swimming animal or a land animal that was also able to swim. Anomodontia are in general land animals, however Placodontia have gone into the water; also close relations or descendants of the Anomodontia, the Sauropterygia and Chelonia, have adapted themselves entirely or partly to water life. The hyposphene might suggest that the vertebrae in question belong to Anomodontia adapted to a swimming way of life. This serves in all events for firmer articulation of the vertebral column and (considering the extraordinary length of the hyposphene: 1 cm in a vertebral length of 3.5 cm) especially prevents great lateral bending of the body—this is not required for swimming. In Triassic Dinosauria the hyposphene is only noticeable on the dorsal vertebrae whose articular facets are developed as saddle-shaped; both features indicate that vertical movement in raising the body or in bending down to the ground proceeded on the short forelimbs. In Dinosauria the structure of the hyposphene in any case connects with the way of life; it is as in Anomosaurus. That lateral bending was otherwise possible here is shown by the deep pits between postzygapophyses and hyposphene on the medial edge, whose purpose was to allow the medial edge of the prezygapophysis freedom to move right and left. In any case the hyposphene is a characteristic of secondary adaptation. The strengthening of the vertebral column for greater elasticity in a swimming way of life is always achieved in separate cases in the most economical way with the available means (e.g. ligaments, intercalary cartilage, zygapophyses, etc.). Two Nothosaurus vertebrae from Bayreuth (Tübingen Collection) illustrate best how the hyposphene may have arisen in this case. In Nothosauria the long neck certainly requires strong connection, particularly of the posterior vertebrae. The edges of the posterior opening of the neural canal and the medial edges of the postzygapophyses meet in a point and form an expansion there that perhaps already functioned as a hyposphene, so to speak.

A piece of skeleton in Breslau (from Gogolin), which will be described in detail by Dr. VOLTZ, shows vertebrae that are similar to those here, ribs that send a strange flat process posteriorly, a limb, a circular pubis, and thick abdominal ribs, as well as several small, thick dermal ossifications in connection.

It would be interesting to learn whether a relationship exists between Anomosaurus and Placodontia. In favor of this are the long diapophyses, the form of the prezygapophyses, abdominal ribs, pubis, and dermal ossifications; against are the high spinous processes, the shortness of the dorsal centra, and the occurrence in the main Muschelkalk together with entirely differently built vertebrae, which at any rate belong to Placodus (see above).

I think there is no doubt that these vertebrae belong to an anomodont. This occurrence, hitherto not appreciated or ignored, shows again the extraordinary plasticity of this noteworthy reptile group.

CTENOSAURUS KOENENI n. gen. n. sp.
A very noteworthy find has been made in the uppermost part of the middle Bunter Sandstone east of Reinhausen near Göttingen. There are four large slabs that contain dorsal vertebrae with enormous spinous processes and ribs. Such a vertebra with a spinous process is ca. 60 cm high, as I estimate from a four-times reduced photograph that the hon. VON KOENEN was kind enough to send to me. A simple outline sketch may be helpful. All the bones are flattened and the preservation is not very good. Thus e.g. the spinous processes seem to be flat from their base onward, but it cannot be said with certainty whether they originally had a round cross-section; in the upper half they are in any case flat and rather expanded. They are curved and bent rather posteriorly. The centra are nearly twice as long as they are high. The upper arch joins onto the anterior part of the centrum, but reaches the postzygapophysis far posteriorly.

I know of nothing better with which to compare these vertebrae than the Permian Clepsydropidae from Texas and Bohemia. The outline of the centrum recalls Naosaurus and Dimetrodon most, the structure of the upper arch recalls Naosaurus, Dimetrodon, and Embolophorus. But of these three, Dimetrodon has the greatest similarity. Whether the centra are amphicoelous as in Clepsydropidae is impossible to determine. Also the abnormal development of the spinous process is most to be expected in Clepsydropidae; I need only recall Naosaurus and Dimetrodon. For all that, I think it is more than likely that Ctenosaurus is a last straggler of the Clepsydropidae, hitherto only known from the Permian. Such an occurrence is very interesting for the completion of the Triassic land fauna.

4. SAUROPTERYGIA
Sauropterygia play a leading role in the marine Triassic, particularly in that of Germany, both in wealth of individuals and of forms. Sauropterygia are characterized by their high degree of adaptation to the swimming way of life. Many forms are more or less specialized, e.g. Lariosaurus is taken to be adapted or at least gives the impression of being adapted to the swimming life up to a certain point. One need only consider the flat humerus, the flat broad forearm and lower leg bones, the plate-shaped development of the carpals and tarsals, and the similarity of the fore- and hind limbs. The long neck with the fairly large head in proportion to the short limbs is sufficient to show that the animal was adapted, in the first place, for swimming, even if the structure of the feet with digits suggests that it may have moved on dry land. Nevertheless Lariosaurus is one of the most primitive representatives of Sauropterygia. In a Liassic Plesiosaurus this neck is very much longer, the body shorter, and the pectoral and pelvic girdles have become true pieces of armor, particularly on the ventral side. The limbs are changed into Ichthyosaurus-like fins; the forearm and lower leg bones can scarcely be distinguished from the other foot bones. In the Upper Jurassic and in Cretaceous this specialization goes even further. It is clear that we have here a succession of animals more and more adapted to swimming; the earliest limbs clearly show that they are derived from land animals, but the source is not immediately obvious. ZITTEL contented himself with a general remark on the resemblance to Rhynchocephalia.

Sauropterygia are divided into three ⁽1) families: Nothosauria, Plesiosauria, and Pistosaurus, which are all already found in the Triassic; the first and last exclusively, the Plesiosauria which continue up to the Upper Cretaceous only at its beginning and as a great rarity. Nothosauria are the rulers of the seas of the Triassic. They are very important because of their frequent occurrence, particularly in the German Triassic. Most genera and species are based on isolated skulls. Larger parts of the skeleton are seldom found in articulation. Thus there is often doubt as to which skull the widely differing skeletal parts belong. A monographic revision of the extensive material, scattered over all of Germany, is an urgent requirement, but can only be undertaken when all the specimens have bean examined.

Some Plesiosauria already appear in the Muschelkalk and Keuper. Several of these occurrences will be described briefly below. Pachypleura and Neusticosaurus lean toward Plesiosauria by the structure of the humerus.

Pistosaurus is known from a single skull which has strikingly small nasal openings lying far back. Pistosaurus can be compared best with Cymatosaurus and is in any case descended from it.

It is not the purpose of these short lines to throw light on the evolution of the genera and species, but I may go into the genetic relations of Sauropterygia a little.

Let us consider Anomodontia. The skull of Sauropterygia, and especially Nothosauria, is quite different in its appearance from the forms of Anomodontia. The elongated structure and uniform dentition indicate strong adaptations to water life. Both groups are alike in that they have only one temporal fossa, and indeed it is homologous to the upper one of other reptiles. The position of the outer nasal openings is again similar to Dicynodon; they are divided externally by the premaxilla and are found between the snout point and the orbits. The inner nasal openings lie paired in the anterior half of the whole bony palate, and in both groups they are separated by the vomer, which participates in the formation of the palate roof in rather small circles. The rudimentary palatal dentition mentioned by BOULENGER on the pterygoid of Lariosaurus is very interesting and extremely important in phylogeny; it recalls Thecodontia. The occiput presents many points of comparison despite many differences. Let us take a Nothosaurus (Cymatosaurus) from the Berlin Museum (after KOKEN ⁽1)) and a Ptychognathus figured by R. OWEN. In Nothosaurus the exoccipitals play quite a small part in the occipital condyle, in Ptychognathus considerably more; on the other hand in Dicynodon feliceps just as in Nothosaurus. The two diverging processes of the basisphenoid are found in Ptychognathus and the other Anomodontia and are present likewise, however directed more horizontally and less projecting. Laterally below is the pterygoid which is already to be seen to rise not so high in Anomodontia. Both the exoccipitals, which surround the greatest part of the foramen magnum, have in Ptychognathus a large broad continuation upward and a rather narrower one downward; the opisthotic appears between them with a small outer surface, it borders part of the canalis ossis quadrati. In Nothosaurus the opisthotic is much larger and the upper continuation of the exoccipital is less developed. In Ptychognathus the supraoccipital not seem to reach above the foramen magnum; also in Nothosaurus it is pressed into a small upward point by the exoccipital. The quadrate in Nothosaurus is much broader and projects more laterally than in Ptychognathus. The median ridge of the supraoccipital in Nothosaurus is only hinted at in Gorgonops torvus ⁽2), and otherwise it is normally missing in Anomodontia. But the lower position of the supraoccipital below the parietal corresponds exactly to the characteristics of all Anomodontia.

The skull of Deuterosaurus and Rhopalodon may be compared in many respects even better with Nothosaurus than with that of Anomodontia. The palatines close the largest part of the palate. The inner nasal openings are completely in front of them and are separated by the vomer. Behind, the pterygoid reaches the skull base rather higher than in Anomodontia. The supraoccipital meets the foramen magnum. Above it an interparietal with ridges inserts. By the indicated points of comparison, of which SEELEY has already mentioned several, I might only say that the principle of the skull structure is not so entirely different as it might seem at first.

The vertebrae of Nothosaurus are platycoelous or slightly amphicoelous and without tendon remains; in Therochelonia and Theriosuchia they are deeply amphicoelous; in Pareiasauria a tendon canal persists on both sides reaching deep into the centrum. Intercentra in Nothosaurus are only present on the atlas, while they have a greater distribution in Anomodontia. The length and height of the vertebrae are not very different. The structure of the upper arch is most like that of Rhopalodon and Deuterosaurus; the strong, short diapophysis even recalls Pareiasaurus. In Nothosaurus the cervical ribs are two-headed, the dorsal ribs single-headed, the proximal ends thick and rather S-shaped. This is the case in Anomodontia and Pareiasauria. The greatest difference lies in the amphicoelous nature of the centra in Anomodontia as compared to Nothosauria. It is quite conceivable that this was lost in the latter by extreme adaptation to water life. The number of sacral vertebrae—two in Nothosaurus, three in Anarosaurus—is the same as in Deuterosauridae. The arrangement of the sacral ribs is even strikingly similar, as the figures show.

The similarity of the ilium of Nothosaurus (two forms) with the several Pareiasauria and Anomodontia is to be seen particularly if one considers that Nothosaurus has only two sacral vertebrae instead of 6-8; thus the proximal end does not need the expansion. The distal part with the large, closed acetabulum in both recalls each other greatly. Ischium and pubis of Cynognathus crateronotus SEELEY ⁽1) agree strikingly with younger Sauropterygia, many Plesiosauria, but Nothosauria can also be compared with this.

The femur of Deuterosauridae is very like that of Pareiasauria, Theriodontia, and Clepsydropidae (Dimetrodon). On the other hand, the femur of Dicynodon may be placed alongside of that Nothosaurus, because it comes between both.

Shoulder girdle and forelimb are once again adapted to water life in Nothosaurus. In the former, therefore, all the bones are fused and only separated by sutures, to provide more strength. I dare express no opinion as to whether the distal spine-like process of the scapula of Nothosaurus may be compared with the meso-scapula SEELEY (= epiclavicle = cleithrum) of Pareiasaurus and the thick short part of the scapula with the disc-shaped scapula of Mastodonsaurus (which has much in common with Pareiasaurus). In any case the clavicle of Mastodonsaurus shows great similarity with that of Nothosaurus. But above all there is great agreement with Pareiasaurus, but at first glance one is struck by the curved clavicles with the small episternum fused in between them. SEELEY in particular draws attention repeatedly to the necessary previous existence of a cartilaginous procoracoid. The scapula has three articular surfaces, of which one receives the humerus and one the coracoid. There is no bone for the third. The coracoid has a notch on the articular surface for the scapula, and medially nearby an apparently useless process follows. The cartilaginous coracoid must have sat here; then the supracoracoid foramen would grow out from the notch, giving the third articular surface a purpose.

If such a procoracoid is present now, this gives another point of comparison with Anomodontia, because it does not occur as a separate bone in other groups of animals (it is recognized with certainty in Chelonia, but they also fuse in Palaeohatteria, Crocodilia, and Dinosauria).

The distal part of the humerus of Nothosaurus has an entepicondyle reminiscent of Dicynodon, Deuterosaurus, etc., with a matching foramen and an ectepicondyle with an ectepicondylar sulcus (rarely foramen). Again this may only be compared with Anomodontia.

The distal part of both limb pairs is very greatly modified (only well known in Lariosaurus and Nothosaurus). In Liassic Sauropterygia (Plesiosaurus) the forearm or lower leg bones have become similar to the carpals and tarsals and to an even greater degree in Upper Jurassic representatives.

Now, as far as Plesiosauria are concerned, one must not think that in general they follow the Nothosauria in time without a contemporaneous occurrence of both. Nothosauridae certainly never extended into the Liassic, but Plesiosauridae probably extended into the Triassic. Plesiosaurus is not uncommon in the English Rhaetic (Aust Cliff near Bristol). Termatosaurus albertii PLIEN. from the Swabian Rhaetic is a true plesiosaur. Several vertebrae of a Plesiosaurus from the Rhaetic bonebed of Bebenhausen are found in the Tübingen collection. Plesiosaurus bibractensis SAUVAGE and Plesiosaurus costatus OWEN from the Rhaetic of Autun also belong there. In the Zanclodon Beds of Schönthal near Basel, extraordinarily large Plesiosaurus vertebrae have been found (Fig. 51 and 52, p. 46). The distal end of a humerus from the Rhaetic Keuper is preserved in Göttingen (Fig. 53, p. 47). These are Rhaetic occurrences. But Plesiosaurus remains are also found, rarely, in the Upper Muschelkalk; H. von MEYER has already figured some without naming them. Such ones are found in most large collections which possess large series of Nothosaurus. Already they occur in the Upper Muschelkalk of Tarnowitz.

From this it seems that Plesiosauria must have evolved early on from a branch of Nothosauria, while others existed and evolved at the same time. These facts will only be observed in more detail in a monographic revision of Nothosauria.

At the beginning of this section attention was drawn to the fact that Sauropterygia are a group that adapt themselves to swimming in ever increasing degrees. But all the parts of the body, even the very primitive limbs, are already not inconsiderably adapted to it. Therefore, in the foregoing the comparison with a doubtless older reptile group is made, and one can only ask whether Sauropterygia are in direct genetic relationship with them or whether they represent an isolated group whose most primitive forms are still unknown. However, properly speaking, this is unlikely. I believe I have assembled sufficient material to show that Nothosauria are more closely related to certain branches of Anomodontia. Close agreements are established with Dicynodontia; but we must remember that these forms are themselves even more specialized. Deuterosauridae and Pareiasauria are primitive, they stand in closer relationship with the foregoing. In this connection it is important to remember that Deuterosauridae are of Permian age but Dicynodontia are of Triassic. However this fact makes the direct descent from the latter well nigh impossible. Therefore I will not be at all surprised if later more widely based studies of Nothosauria prove they are derivatives of Deuterosauridae that have returned to the water and they thus represent an equivalent branch to Dicynodontia from one stock. This is a point of view that may form the subject of a future monograph on Nothosauria of importance. The presence of a postorbital in some older Nothosauridae, such as Cymatosaurus and Nothosaurus (Eurysaurus (FRECH) latissimus), certainly seems to stand the way of the acceptance of this, but Therosuchia also possess one.

PRESENCE OF PLESIOSAURIA IN THE MUSCHELKALK
As mentioned already, there are found in the German Muschelkalk rare vertebrae of true Plesiosauria which differ only slightly from Nothosauria in some respects. HERMANN von MEYER has already figured a number of these in his “Fauna der Vorwelt”; they are mainly as follows: pl. 25, fig. 3; pl. 52, fig. 3, 7, 12; pl. 53, fig. 2-8; pl. 57, fig. 3, 4, 6.

Such vertebrae from the Upper Muschelkalk of Bayreuth lie before me. The cervical vertebrae are keeled along the middle with the characteristic foramina; the dorsal vertebrae are rounded and smooth. Once again I figure a posterior cervical centrum and a complete anterior cervical vertebra from Bayreuth (pl. VIII, fig. 4 and pl. V, fig. 5). Pl. V, fig. 5 represents an anterior cervical vertebra from Bayreuth. In the latter the high formation of the upper arch falls between the diapophysis and the zygapophyses, which are strongly developed. The spinal process is narrow and short. Another cervical vertebra from the lower bonebed of the Lettenkohle of Biebersfeld (pl. V, fig. 4) shows the two characteristic foramina very strongly marked on the underside. The upper arch is missing, thus a further two pairs of smaller foramina are visible from the base of the neural canal. The other pair lies right in front on the edge of the articular surface; each of the two openings is very elongate. The posterior pair is 2/5 of the length of the vertebra (= 10 mm of 25) distant from the rear edge. The base of the neural canal is much broader toward the rear than the front.

The Plesiosaurus vertebrae from the Muschelkalk may first of all be combined as
Plesiosaurus priscus
indeed it is suggested that they are distributed over more than one species (Tarnowitz!).

PECULIAR SACRAL VERTEBRAE FROM THE MUSCHELKALK