Part 2

Of the hyoid arch only the ceratohyals (see fig. 5A) are preserved in several specimens. These are long, curved bones with a posteroventral process and widened, flaring posterior margin. The medial (?) surface is concave in one specimen. The lateral (?) surface displays a distinct ridge on several specimens, arising on the dorsal surface opposite the posteroventral process and extending diagonally to the anteroventral end of the anterior limb. The impression of one other specimen appears to have a central ridge because of greater dorsal thickness and narrowness. Both surfaces are unornamented.

The urohyal (see fig. 5B) is an unornamented, Y-shaped bone, with the stem of the Y pointing anteriorly. Orientation with respect to dorsal and ventral surfaces is uncertain. In one view a faint ridge, also Y-shaped, occurs on the expanded posterior portion, and the surface is convex. The anterior process has a convex surface, sloping evenly off to the lateral margin; the opposite side of the process has a concave surface. The posterior portion has a slightly depressed area (see fig. 5B) at the junction of the "arms" of the Y.

The five branchial arches are represented by the ceratobranchials, several of which are preserved on K. U. no. 11431. These are long bones with anteriorly curving ventral ends. The medial surfaces are partly covered with minute granular teeth; only the dorsal part is without teeth. The dorsal articular surface is convex dorsally and rounded.

The mandible (fig. 3), the best specimens of which are K. U. nos. 788 and 11425, is seen only in lateral and ventral views, with only angular, splenial and dentary visible.

The angular forms the main body of the mandible, and is similar to that of _Spermatodus_. The dorsal margin of the angular is expanded in the central region, with some variation. One specimen has an expanded portion slightly anterior to that of the opposite angular. The articular surface near the posterior end has not been observed; the posterior end of the angular slopes off abruptly. The anterior sutures are seen in only two specimens, K. U. nos. 788, 11425. The dentary meets the angular in a long oblique suture; the dentary gradually tapers posterodorsally and ends on the dorsal surface of the angular. The splenial fits into a posteriorly directed, deep V-shaped notch on the ventral surface. The lateroventral surface of the angular contains sensory pores of the mandibular line. The ventral surface extends medially into a narrow shelf, approximately 1.0 mm. wide, which extends the full length of the bone; the external surface of this shelf is smooth and slightly concave dorsally. Ornamentation of the angular consists of tubercles and longitudinal or oblique striae, occurring mostly on the expanded portion. The medial surface is not seen. Several broken specimens show a central canal filled with a rod of calcite; in one of these the sensory pores are also calcite-filled and appear to be connected to the rod. Thus the pores originally opened into a central canal.

The dentary is an unornamented bone with the anterior half curving medially; the greatest height is anterior. This bone in specimen K. U. no. 11425 bears irregularly spaced, simple, recurved, conical teeth; nine were counted, but there is space for many others. One other specimen, no. 11429, seems to have tiny tubercles on the surface. The dentary meets the splenial dorsally in a straight suture.

The splenial also curves medially, and as stated, meets the dentary in a straight suture. Ornamentation on this bone was not observed. The posterior margin is V-shaped and fits the notch in the angular. The ventral surface bears three or more sensory pores of the mandibular line.

The gular plates are oval. The medial margin is straight to slightly curved, the lateral margin curved crescentically, the posterior end is blunt, and the anterior end somewhat rounded. Ornamentation varies greatly; some bones show only a few tubercles, whereas others exhibit an almost concentric pattern of closely spaced striae. Typically there are some tubercles in the anterior quarter or third of the total length; these pass into longitudinally oriented striae in the posterior section. A few have only randomly oriented, widely-spaced striae. The internal surface is smooth.

The coronoid (K. U. no. 11428) is a triangular bone, with the apex pointing dorsally. The lateral surface is smooth; no teeth were observed. Moy-Thomas (1937:292, 293) mentions several tooth-bearing coronoids in _Rhabdoderma_, but as yet these have not been seen in _Synaptotylus_.

_Axial skeleton_

Only three specimens (K. U. nos. 786F, 787F, 11450) show parts of the vertebral column, but isolated neural and haemal arches are numerous. All are of the coelacanth type, having Y-shaped neural and haemal arches, without centra. A total count of 38 was obtained, but this was incomplete; the actual number was probably near 50. Counts of 10 and 16 haemal arches were obtained in two of the specimens. Total height of neural arches ranges from 7.5 to 12.0 mm., and of haemal arches, from 9.0 to 12.0 mm. The shorter arches are anterior and the height increases gradually to a maximum in the caudal region. Height of the spines varies from 4.0 to 9.0 mm., or from twice the height of the arch in the anterior to three times the height in the caudal region. Total width of the base, measured in isolated specimens because lateral views in other specimens prevented measuring width, ranges from 0.7 to 4.2 mm. The short, broad arches having short spines occur at the anterior end of the spinal column; the narrower arches having tall spines occur toward the caudal end. Broken neural and haemal arches show a thin covering of bone with a central, calcite-filled cavity, which in life may have been filled with cartilage (Stensioe, 1932:58, fig. 20).

No ossified ribs have been observed, either isolated or in place.

For further description of the axial skeleton, see Hibbard (1933).

_Girdles and paired fins_

A nearly complete pectoral girdle on specimen K. U. no. 11433 (see fig. 6A) has only a cleithrum and clavicle. No evidence of an extracleithrum or supracleithrum has been observed, but the extracleithrum may be fused to the cleithrum. The two bones form a boot-shaped unit, with the anteroventral part turned medially to form a horizontal process which meets the opposite half of the girdle. In lateral view the surface is unornamented, and convex in the ventral half. The suture between the cleithrum and clavicle begins on the expanded posterior portion, the "boot-heel," at a point immediately below the greatest width on the posterior margin, passes anteriorly, then turns sharply and parallels the anterior margin. The shape of the cleithrum resembles that in _Rhabdoderma_ and the internal surface is not ridged (see Moy-Thomas, 1937:fig. 9). The exact orientation in the fish is uncertain, but if the median extension is really horizontal, then the posterior expansion is directed caudally. The medial surface is concave, steepest near the anterior margin, and then slopes outward evenly. In medial view one specimen (K. U. no. 11426) shows a small, caudally directed projection of bone, evidently for articulation of the fin-skeleton, at the widest portion of the cleithrum. Sutures on several specimens were indistinct. Broken specimens show sutural faces, but many nearly complete specimens show little or no indication of sutures, without regard to size of the girdles. The internal structure of the fin was not observed.

Numerous isolated basal plates of the pelvic girdle have revealed details of structure but no information on the orientation. Presumably the basal plates of _Synaptotylus_ had essentially the same orientation as those of other coelacanths (Moy-Thomas, 1937:395). The most complete basal plate is K. U. no. 788 (see fig. 6B). The three apophyses diverge anteriorly; the horizontal one is best developed and the dorsal one is least well developed. A median process (Schaeffer, 1952:49), denticulate on several specimens, articulates with the corresponding process of the opposite plate. The expanded part that articulates with the skeleton of the fin extends caudally. The posterior expanded part is nearly square in outline, resembling the dorsal, rectangular projection. One side bears ridges leading to the extremities of the apophyses, and faint crenulations on the median process. This may be the medial view. The other view displays a smooth surface, usually without indication of the ridges seen in the reverse view. These specimens differ somewhat from the basal plates of _Rhabdoderma_ and appear to be intermediate between _Rhabdoderma_ and _Coelacanthus_ (Moy-Thomas, 1937:fig. 10A, B). The apophyses are not free as in _Rhabdoderma_ but webbed with bone almost to their extremities, as in _Coelacanthus_.

The pelvic fin is seen in only two specimens (K. U. nos. 786F, 788). That on no. 788 is lobate and has 25 lepidotrichia, jointed for approximately the distal half, and 2.5 to 13.0 mm. in length. Total length of the fin is 25.0 mm. There is no trace of the internal skeletal structure or of the articulation to the basal plate in either specimen. For a description of the fin on no. 786F, see Hibbard (1933:281).

_Unpaired fins_

A few isolated bones on K. U. no. 788 (fig. 7) are interpreted as basal plates of the unpaired fins. For additional description of the unpaired fins on the type, K. U. no. 786F, see Hibbard (1933).

Two of these bones are flat, smooth and oblong, bearing a diagonal ridge that extends in the form of a projection. Orientation is completely unknown. These may be basal plates of the anterior dorsal fin. The fin on no. 786F that Hibbard (1933:281) interpreted as the posterior dorsal fin is now thought to be the anterior dorsal fin.

One distinctive bone may represent the basal plate of the posterior dorsal fin. This incomplete specimen shows two projecting curved processes, bearing low but distinct ridges, which diverge, probably anteriorly. The central portion is narrow. The two ridges continue onto the posterior portion. This has been broken off, but shows that the ridges diverge again. The surface is smooth, except for the ridges. As before, orientation is uncertain. On no. 786F this fin was interpreted by Hibbard (1933:281) as the anal fin.

Only part of one basal plate of the anal fin was preserved on K. U. no. 11450. That plate is oblong and has an expanded anterior end. The narrow, constricted part bears two oblique ridges and a few tubercles. The posterior part has nearly straight margins (represented by impressions) and the posterior margin is oblique, sloping anteroventrally. The flared anterior part has a smooth surface. This basal plate is more nearly like those of _Coelacanthus_, according to the descriptions given by Moy-Thomas (1937:399). The basal plate is associated with seven apparently unjointed, incomplete lepidotrichia. The anal fin on no. 786F is interpreted as the anterior dorsal fin (Hibbard, 1933:281).

The caudal fins are preserved on K. U. nos. 786F, 787F, and have a total of 24 lepidotrichia, 12 above and 12 below. These are jointed for the distal half or two-thirds, and are up to 16.0 mm. in length. In specimen no. 787F the supplementary caudal fin has at least seven lepidotrichia, the longest of which is 11.0 mm. but incomplete. Anterior lepidotrichia appear unjointed but the posterior ones are jointed for the distal two-thirds (?) (these are broken off). The supplementary caudal fin is approximately 1.5 mm. long and 8.0 mm. or more wide. The supplementary caudal fin on K. U. no. 786F described by Hibbard (1933:281) could not be observed; this part of the caudal fin is missing.

_Squamation_

In the suite of specimens isolated scales are numerous, but patches of scales are rare. Only two specimens (K. U. nos. 786F, 787F) are complete enough for scale counts, but preservation permits only partial counts. In general the scales resemble those of _Rhabdoderma elegans_ (Newberry).

The scales are oval. The exposed posterior part of each bears posteriorly converging ridges; the anterior part is widest and shows a fine fibrillar structure. There are at least six scale-rows on either side of the lateral line. Lateral line scales show no pores, and except for slight irregularities in the orientation and length of the posterior ridges, closely resemble the others. Central ridges on the lateral line scales are shorter and tend to diverge from the center of the impression of the canal. The lateral line canal shows only as the impression of a continuous canal 0.7 mm. in diameter. Preservation is poorest in scales along the line of the neural and haemal arches; therefore lateral line scales are rarely preserved. Isolated scales are of two types: those on which the posterior ridges converge sharply and form the gothic arch configuration mentioned by Hibbard (1933:282), and those which do not. Both types of scales can be present on one fish, as shown by specimen no. 788. This is not apparent on nos. 786F and 787F; all of the scales on these specimens appear to be much alike. Both Moy-Thomas (1937:385) and Schaeffer (1952:51, 52) have remarked on the variation of the scales on different parts of the same fish. Because the number of ridges and amount of convergence of the ridges is not related to size of the scale, it is concluded that these characters are not of taxonomic significance.

The strong resemblance of the scales of the Garnett specimens to those of _Rhabdoderma elegans_ (Newberry) caused Moy-Thomas (1937:399) to add Hibbard's two species to the synonymy of _R. elegans_. But at that time only the scales could be adequately described. If the shape of the scale and the number and pattern of ridges can vary with age, size and shape of the scale, it follows that assignment of isolated scales to a species should not be attempted. Assignment to genus should be made only with caution.

_Discussion._--The relationship of _Synaptotylus_ to other coelacanths is obscure at present. The knoblike antotic processes on the basisphenoid are unlike those of any other known coelacanth. The palatoquadrate complex is shaped like that of _Rhabdoderma elegans_ but consists of fewer bones, probably because of fusion. The scales resemble those of _Rhabdoderma_. With regard to general shape of fin girdles, the pectoral girdle resembles that of _Eusthenopteron_ more than that of _Rhabdoderma_, but the cleithrum is more nearly like the cleithrum of _Rhabdoderma_. The pelvic girdle appears to be midway between those of _Rhabdoderma_ and _Coelacanthus_ in general appearance. Regarding the basal plates of the remaining fins, those of _Synaptotylus_ appear to resemble basal plates of both _Rhabdoderma_ and _Coelacanthus_. Considering the structure of the sphenethmoid region of the braincase, _Synaptotylus_ is probably more closely related to _Rhabdoderma_ than to other known coelacanth genera.

COMMENTS ON CLASSIFICATIONS

Classification of Carboniferous coelacanths has been difficult, partly because the remains are commonly fragmentary, and significant changes in anatomy did not become apparent in early studies. In general, coelacanths have been remarkably stable in most characters, and it has been difficult to divide the group into families. As Schaeffer (1952:56) pointed out, definition of coelacanth genera and species has previously been made on non-meristic characters, and the range of variation within a species has received little attention. For example, Reis (1888:71) established the genus _Rhabdoderma_, using the strong striation of the scales, gular plates and posterior mandible as the main characters of this Carboniferous genus. Moy-Thomas (1937:399-411) referred all Carboniferous species to _Rhabdoderma_, redescribed the genus and compared it to _Coelacanthus_, the Permian genus. He cited as specific characters the ornamentation of the angulars, operculars and gular plates (Moy-Thomas, 1935:39; 1937:385). Individual variation in some species has rendered ornamentation a poor criterion. This variation is apparent in _Synaptotylus newelli_ (Hibbard), some specimens having little or no ornamentation; others having much more. The number of ridges and pattern of ridges on the scales also varies. Schaeffer (1952:56) has found this to be true of _Diplurus_ also. Moy-Thomas (1935:40; 1937:385) realized that the type of scale is a poor criterion for specific differentiation. In the search for features useful in distinguishing genera of coelacanths, Schaeffer and Gregory (1961:3, 7) found the structure of the basisphenoid to be distinctive in known genera, and thought it had taxonomic significance at this level. Higher categories should have as their basis characters that display evolutionary sequences. A recent classification (Berg, 1940), followed in this paper, reflects two evolutionary trends in endocranial structure of coelacanths: reduction of endocranial ossification and loss of the basipterygoid processes. Because there has been little change in other structures in coelacanths, Berg's classification is the most useful. Berg (1940:390) includes _Rhabdoderma_ in the suborder Diplocercidoidei because of the presence of the basipterygoid processes, and in the single family, Diplocercidae, but remarks that because of the reduced amount of endocranial ossification the Carboniferous Diplocercidae "probably constitute a distinct family." In considering this concept of classification, the subfamilies Diplocercinae and Rhabdodermatinae of the family Diplocercidae are proposed above. The subfamily Rhabdodermatinae includes at present _Rhabdoderma_ and _Synaptotylus_. The principal characters of the subfamily Rhabdodermatinae, named for the first known genus, are the retention of the basipterygoid processes and the reduction of endocranial ossification. Application of this classification based upon endocranial structure would probably change existing groupings of species of Carboniferous coelacanths; the entire complex of Carboniferous genera should be redescribed and redefined. It will be necessary to consider endocranial structure in any future classification.

The greater part of the evolution previously mentioned appears to have been accomplished during the Carboniferous; thereafter coelacanth structure became stabilized. The trend progressed from Devonian coelacanths which had two large unpaired bones in the endocranium, and both antotic and basipterygoid processes on the basisphenoid, to Carboniferous fishes in which ossification was reduced to a number of paired and unpaired bones embedded in cartilage, and retaining both processes, and then post-Carboniferous kinds with reduced ossification and no basipterygoid processes. The Pennsylvanian was evidently the time of greatest change for the coelacanths, and they have not changed significantly since, in spite of the fact that since the Jurassic they have shifted their environment from shallow, fresh water to moderate depth in the sea (Schaeffer, 1953:fig. 1). The changes in endocranial structure appear to be significant, and are perhaps related to higher efficiency of the mouth parts in catching and swallowing prey (see p. 482).

ENVIRONMENT

The coelacanth fishes from the Rock Lake shale are part of the varied fauna collected from Garnett. Peabody (1952:38) listed many elements of the fauna and flora, and concluded that the deposits are of lagoonal origin. In addition to numerous invertebrates (including microfossils) and arthropods, a number of vertebrates other than coelacanths have been found. These include at least one kind of shark, _Hesperoherpeton garnettense_ Peabody, one or more kinds of undescribed labyrinthodonts and the reptiles _Petrolacosaurus kansensis_ Lane, _Edaphosaurus ecordi_ Peabody, and _Clepsydrops_ (undescribed species). This is indeed a rich vertebrate fauna, and the earliest known reptilian fauna. Much of the rock contains plant remains. The flora that has been identified is adapted to growing in a well-drained soil; although it contains some elements considered characteristic of the Permian, it is of Pennsylvanian age (Moore _et al._, 1936). Peabody (1952:38-39) discusses the features of these lagoonal sediments. Much of the fauna and flora suggests continental origin, but the many marine invertebrates at some horizons indicate that at least some of the sediments were of marine origin.

Little can be said about the actual environment of the living fishes of the genus _Synaptotylus_. Remains of these fishes occur in layers containing marine invertebrates, as well as in those containing plant remains and vertebrate skeletal parts, and in those nearly completely composed of dark carbonaceous material. Most of the remains are fragmentary and consist of isolated bones, isolated scales, and dissociated skulls; only one specimen and half of another are nearly complete. Many published statements on _Rhabdoderma_, a related genus, indicate both marine and fresh-water environments. Wehrli (1931:115) regarded _Rhabdoderma elegans_ (Newberry) as a euryhaline species, and cited its occurrence with both marine and fresh-water fossils. Aldinger (1931:199) also found this to be the case with other species, and Fiege (1951:17) quotes others as giving the same information. Keller (1934:913) thought that few Carboniferous fishes were exclusively marine, and stated that the majority of them became adapted to fresh water during the late Carboniferous. Later, Schaeffer (1953:175) stated that all Carboniferous and Permian coelacanths were fresh-water fishes, and that many were from swamp deposits. If Keller is correct, then members of the genus _Synaptotylus_ may have inhabited the lagoon, the adjacent sea, or the streams draining into the lagoon. Perhaps these fishes swam upstream, as modern salmon and tarpon do, although there is no direct evidence for this. Possibly they lived in the lagoon at times of scant rainfall and little runoff, when the salinity of lagoon water approached normal marine values or the fishes may have lived in the streams, and after death were washed into the lagoon. As numerous remains of land plants and animals were washed in, perhaps this best accounts for the presence of the fish in nearly all layers of the deposits, not only the marine strata.

SUMMARY

A new genus of Pennsylvanian coelacanths, _Synaptotylus_, is described and a previously named species, _Coelacanthus newelli_ Hibbard, 1933 (_C. arcuatus_ Hibbard, 1933, is a junior synonym), is referred to this genus. All specimens of _Synaptotylus newelli_ (Hibbard) were collected from the Rock Lake shale member of the Stanton formation, Lansing group, Missouri series, six miles northwest of Garnett, Anderson County, Kansas. _Synaptotylus_ is distinguished from all other coelacanths by a basisphenoid having large, knoblike antotic processes each connected by a low ridge to a small basipterygoid process. _Synaptotylus_ is most closely related to _Rhabdoderma_, but is intermediate between _Rhabdoderma_ and _Coelacanthus_ in shape of the fin girdles and basal plates. Two new subfamilies, Diplocercinae and Rhabdodermatinae, of the family Diplocercidae, are proposed. _Synaptotylus_ and _Rhabdoderma_ are included in the subfamily Rhabdodermatinae, because both exhibit reduced ossification in the endocranium and retain basipterygoid processes.