Skip Navigation

Late Paleozoic Pelecypods: Mytilacea

Prev Page--Family MYTILIDAE || Next Page--Foreign species


Systematic Paleontology, continued

Family MYALINIDAE Frech, emend. Newell

Shell inequivalve, the right valve slightly less convex and slightly smaller at the margin than the left; beaks at or near the small and projecting anterior end; posterior margin subovate, quadrate, or extended in a posterodorsal auricle; ligament external, duplivincular, mainly opisthodetic, but with amphidetic remnant before the beaks in some genera; hinge edentulous, or with weak cardinal teeth 1/(2a,2b)at the front end of hinge; inner shell layer lamellar, probably aragonite; outer layer composed of calcite, either homogeneous or finely prismatic, the prisms and optic axes nearly normal to the shell surface; musculature characteristically anisomyarian, but monomyarian in a few species. The genera that unquestionably belong here do not possess well defined radial ornamentation, and all pass through a form stage in the early ontogeny similar to adult Modiolopsis.

Remarks—The phylogeny of this family cannot as yet be determined with any degree of completeness, because the shell structure, ontogeny, and internal characters are unknown for the majority of Mississippian and older species that may belong here. Several of the Ambonychiidae resemble some of the myalinas so closely as to suggest that the similarities are more than superficial. Furthermore, some of the genera tentatively included under the Myalinidae have not yielded adequate data on the early growth stages as yet, and it may be that further information will indicate that some of these genera do not belong in this family.

The origin of the family is not known, but presumably both the Mytilidae and Myalinidae were derived from an Ordovician or Silurian pelecypod having a configuration like Modiolopsis and a parivincular ligament. The oldest genera known to me that I would tentatively include in the Myalinidae are Myalinoptera and Hoplomytilus from the Devonian. Neither of these is well known.

The family is well represented in the Mississippian and, particularly, is an important element in the Pennsylvanian faunas, undergoing a marked decline in the Permian. Triassic and Jurassic species in many parts of the world have been referred to Myalina, and almost certainly the few forms known to have had a duplivincular ligament should be referred to the family. So far as I am aware, members of the Myalinidae are not known in Cretaceous or younger rocks.

Genus MYALINA De Koninck, 1842

Myalina De Koninck, 1842, Description des animaux fossiles qui se trouvent dans le terrain carbonifère de Belgique, p. 1251. Hind, 1897, Palaeontographical Soc., vol. 51, p. 103; Newell, 1940, Am. Jour. Sci., vol. 238, pp. 286- 295, pl. 1, text figs. 1, 2.

[1 This work bears the date of imprint 1842-1844. I am assuming that the first part of the book containing descriptions of pellecypod species, was published in 1842, because that is the date given for these fossils in De Koninck's later writings.]

Genolectotype, Myalina goldfussiana De Koninck, 1842, designated by Stoliczka, 1871, Palaeontologia Indica, vol. 3, p. 366. Geologic range of genotype, Viséan.

Mytiliform shells with terminal beaks and a pronounced forward obliquity, generally unornamented, but the left valve in some species roughened by relatively prominent growth lamellae; right valve slightly smaller (discordant) and less convex than the left valve; ligament area opisthodetic, relatively broad, traversed by several coarse parallel ligament grooves, which descend posteriorly from a place under the beak toward the hinge axis; area bounded anteriorly on the right valve by a large ridge, which extends posteroventrally from the beak across the end of the ligament area to form a kind of tooth whose ventral extremity articulates with the ventral end of a corresponding furrow of the left valve; anisomyarian; umbonal musculature consisting of three pits in a closely spaced posterodorsal-anteroventral series, of which the middle impression is the largest; pallial line consisting of closely spaced, discrete pits, which tend to coalesce and form a continuous line in the posterior half of the shell; posterior impressions elongated parallel to the rear margin, extended dorsally into two processes, of which the rear one is the longer; a series of closely spaced muscle pits extending from the inner margin of the posterior muscle impression forward into the umbonal cavity.

The shell microstructure in the type species is unknown. In various species that are clearly con-generic with Myalina goldfussiana, the shell consists of a relatively thick inner lamellar ostracum, probably originally composed of aragonite, and a thin outer ostracum made up of very fine, irregularly polygonal prisms of calcite arranged normal to the shell surface. The prisms of the right valve are much coarser than those of the left valve in the typical subgenus.

For the sake of documentation, the measurements of three specimens of the genotype are given. Numbers 1 and 2 are right and left valves, respectively, of the holotype; number 3 is a left valve in the Burrow collection at Cambridge (no. 463).

Measurements of Myalina goldfussiana, in millimeters
  no. 1 no. 2 no. 3
Convexity 11 14  
Length 58 60 45
Height 48 50 37
Hinge length 42 43 33
Greatest dimension     49

Remarks—Accompanying the original diagnosis of Myalina were the descriptions of three new species—M. goldfussiana, M. lamellosa, and M. virgula. In the absence of designation of a genotype, these species automatically were genosyntypes. All of them were originally found in the Viséan rocks of Belgium. The earliest legal designation of a genolectotype that comes to my attention was made by Stoliczka in 1871. He selected M. goldfussiana as the type of Myalina, thereby specifically establishing the characteristics of the genus.

Without exception, writers subsequent to De Koninck have turned to the published diagnoses of that author to learn the fundamental characters of Myalina. Modern writers are learning that this procedure, though convenient, leads all too often to perpetuation of erroneous concepts. Hardly any descriptions of fossils are wholly objective, nor is it necessarily desirable that they should be. Each writer is obligated to evaluate his specimens to the best of his ability, and to avoid cluttering up the literature with extraneous details that seem not to be significant. It is fairly well understood today that an author's conclusions regarding his own genera or species have no validity where they are at variance with the characters exhibited by the type specimens.

De Koninck's diagnosis of Myalina is at variance with the characters of the subsequently designated genotype. How he arrived at the characters of the genus as reported by him has no legal bearing on the case. The original diagnosis of Myalina is as follows (free translation) :

Shell equivalve, not equilateral, elongated or oblique, straight or slightly arched at the cardinal edge. Hinge without teeth. Ligament internal, covering completely a large facet, traversed longitudinally by a number of narrow furrows incised parallel to the cardinal border. Beaks terminal, acute, ordinarily small and slightly recurved. Internally, and immediately adjoining these, occurs a small septiform lamina similar to that observed in certain species of Mytilus.
The external form of the genus Myalina resembles that of Mytilus. In the form of its ligamentary facet it resembles Pterinea. Generally it seems to be intermediate between the two and seems to form a natural transition from one to the other. The valves of the species that we report on here are hermetically closed on all sides. It is then not probable that the animal possessed a byssus by which it could attach itself to submarine objects. Their shell is rather thick. Although we have not been able to inspect the muscle impression of this genus we suppose that it must be placed among the dimyarian mollusks since it is rather probable that the small lamina situated next to the beaks served as a surface of attachment of the anterior muscle. (De Koninck, 1842, p.125)

At a later date De Koninck corrected some of the errors in the original diagnosis:

Valves subequal, mytiliform; ordinarily higher than long; beaks generally pointed, terminal, provided with an anterior septum; cardinal margin broad, flat, with several longitudinal cartilage furrows; two muscle impressions, pallial line simple...
When I created this genus I had at my disposition only a small number of fragmentary Carboniferous examples, which all appeared to be equivalved. However, W. King and F. McCoy, having had occasion to study entire well preserved Permian species related to the genus, have made the observation that the right valve is nearly always smaller than the left. Since then I have had occasion to note the same on some Carboniferous species, and I am in accord with these paleontologists in concluding that the shell must not necessarily be equivalved to belong to the genus, providing that it possesses other characters indicated in the diagnosis. (De Koninck, 1885, p. 167).

Through the good auspices of Dr. Pierre Bonnet I have had the privilege of studying the holotype of Myalina goldfussiana, property of the École des Mines, in Paris. The holotype consists of incomplete portions of two massive complementary valves lacking the posteroventral one third of the shell. The substance of the shell is recrystallized, so that the original microstructure cannot be determined. The shell is mytiliform, and the right valve is both smaller and slightly less convex than the left valve. Preparation of the interior of the valves has revealed that this species does not possess an umbonal septum. The ligament area is relatively broad, and is traversed by four, coarse ligament grooves which converge with the hinge axis posteriorly at an acute angle. On the right valve the anterior terminus of the ligament area is marked by a large, vertical ridge or tooth, which corresponds in position to a shallow furrow, or socket, in the left valve. When the valves are in tight apposition, only the ventral portion of the tooth is in contact with the socket; and the two diverge upward at an angle corresponding with the divergence of the ligament areas of the two valves above the hinge axis.

The shell form is not well shown by the imperfect holotype, so observations were made on a supplementary specimen of the same species from the Carboniferous limestone of Settle, England. It was lent to me by Dr. O. M. B. Bulman, of Cambridge University. This specimen, a left valve, was erroneously identified by Hind as Myalina peralata De Koninck (Hind, 1895, p. 119, pl. 7, fig. 3). The muscle pattern within the umbonal cavity of both valves of the holotype is well shown (pl. 9, figs. 5a-6b) but much of the general musculature had to be interpreted from another Viséan specimen of M. goldfussiana described by De Koninck (1885, pl. 29, fig. 7). Photographs of this specimen were sent me by Dr. F. Demanet, of the Musée Royal d'Histoire Naturelle, in Brussels.

As herein defined, the genus Myalina seems to be restricted to Carboniferous and Permian rocks. Similar Devonian and post-Paleozoic forms commonly referred to the genus probably should be differently classified.

Three subgenera of Myalina are recognized, Myalina s.s., Orthomyalina, and Myalinella. Critical examination of the evidence reveals that these groups, for all their superficial resemblance, followed different paths of evolution during the Late Paleozoic. They are not form groups, but represent true genetic series.

Figure 12—Palingenesis of some species of Myalina (Myalina). The outlines of a succession of species are shown in stratigraphic succession. These outlines are correlated with ontogenetic stages of one of the most highly specialized species, M. copei. Only a critical examination will reveal form differences in the two series. The gerontic stage of M. copei illustrates the interesting fact that gerontism is not reached simultaneously in all bio-characters. The cessation of expansion of the rear auricle, and general slackening of peripheral increase are interpreted as gerontic characters. However, the forward curvature of the umbonal ridge, responsible for decreasing obliquity, seems to be unchecked.

Palingenesis of some species of Myalina (Myalina).

Table 5—Species of Myalina (Myalina), Showing Age and Minimum, Average, and Maximum Dimensions1, in Millimeters

Species Age Length Height Greatest
Dimension
Form Ratio Angle α Angle β
M. (M.) sinuata2 Lower
Guadalupian
        60°  
85.0 (1) 75.0 (1)   1.10 (1) 67° (4) Very acute
        75°  
M. (M.) wyomingensis
var. thomasi
Lower
Guadalupian
32.0 28.0 39.0 1.00 47° Obtuse,
broadly
rounded
47.0 (10) 43.0 (11) 56.0 (10) 1.11 (10) 55° (16)
59 57.0 70.0 1.27 73.0°
M. (M.) copei Wolfcampian 68.0 71.0 74.0 0.83 76.0° 14.0°
78.0 (17) 83.7 (17) 80.9 (17) 0.94 (17) 84.0° (17) 32.0° (17)
92.0 94.0 101.0 1.01 104.0° 48.0°
M. (M.) pliopetina Wolfcampian 56 59.0 70.0 0.95 59.0° 40°
76 (2) 10.4 (5) 81.2 (5) 0.98 (2) 62.3° (5) 47° (2)
97 95.0 114.0 1.02 70.0° 55°
M. (M.) aviculoides Wolfcampian 37 (1) 41 (1) 44 (1) 0.90 (1) 78° (1) 95° (1)
M. (M.) petina Virgilian 46.0 54.0 63.0 0.82 64.0° 75°
47.0 (2) 55.0 (2) 64.0 (2) 0.85 (2) 68.0° (2) 76° (2)
48.0 56.0 65.0 0.88 72.0° 76.5°
M. (M.) miopetina Virgilian 52.0 48.0 63.0 0.98 59.0° 82.0°
56.6 (3) 41.2 (3) 70.6 (3) 1.03 (3) 61.0° (3) 89.0° (3)
60.0 61.0 75.0 1.08 63.0° 96.0°
M. (M.) glossoidea Virgilian
and
upper Missourian
26.0 30.0 33.0 0.83 73.0° 93.0°
48.3 (8) 55.7 (8) 61.5 (8) 0.87 (8) 76.3° (8) 109.0° (8)
65.0 74.0 82.0 0.90 83.0° 118.0°
M. (M.) arbala Missourian 26 27 33 0.71 58.0° 80.0°
52 (4) 60 (14) 69 (14) 0.88 (14) 71.3° (14) 94.0° (13)
69 77 92 1.10 87.0° 117.0°
M. (M.) wyomingensis Desmoinesian
to
Wolfcampian
13.0 12.0 16.0 0.87 65.0° 95.0°
32.5 (11) 32.6 (11) 39.4 (11) 1.05 (11) 70.5° (12) 116.0° (11)
51.0 48.0 58.0 1.12 81.0° 139.0°
M. (M.) lepta Desmoinesian 26.0 23.0 30.0 1.1 55.0° 108.0°
28.2 (4) 24.7 (4) 33.5 (4) 1.1 (4) 55.5° (4) 119.0° (4)
33.0 28.0 40.0 1.2 59.0° 124.0°
M. (M.) copha Lampasan 5.0 4.5 6.0 0.9 43.0° Obtuse,
broadly
rounded
10.4 (16) 8.1 (16) 11.5 (16) 1.2 (16) 55.5° (16)
25.0 20.0 27.0 1.5 75.0°
1 Figures in parentheses indicate number of specimens used. Dimensions given in millimeters.
2 Known only from internal molds; therefore, measurements are not reliable.

Subgenus My ALINA s.s.

Thick-shelled species like the genotype, which have a well developed anterior lobe, at least in the left valve; primitive species have a marked prosocline obliquity and a marked homeomorphic resemblance to Promytilus, advanced species become acline, or opisthocline, and develop a posterior auricle, reminiscent of some Pteriidae and some Ambonychiidae; in primitive forms the valves are nearly equal; in advanced forms the valves are markedly unequal, left valves commonly more rugose than right valves, being particularly roughened on the anterior lobe.

Geologic range—Carboniferous and Permian.

Remarks—Species of the subgenus Myalina present one of the most complete records of subgeneric evolution known to me among Paleozoic pelecypods. There is a graded series of forms from species like Myalina wyomingensis to highly specialized species like Myalina copei. The progression of form displayed in the ontogeny of some of the Permian species corresponds with the stratigraphic sequence of the fossils, so that the record seems to be unusually complete (text figs. 12, 14, and 15). Inasmuch as some species have the anterior lobe much more pronounced in the left valve than in the right, isolated right valves may be mistaken for representatives of the subgenus Myalinella or Orthomyalina, in which the anterior lobe is lacking, or nearly so. Species of Myalina, s. s., are among the most useful guide fossils among Late Paleozoic pelecypods, and rank favorably with many of the brachiopod families. It is unfortunate that the record of the myalinas seems to be extremely scant outside of the United States.

It is difficult to make any generalizations about the ecology of Myalina. The shells are found in all sorts of limestones and limy calcareous shales, less coromonly in sandstones. Bivalved shells are common, as are shells that obviously were torn apart by current or wave action. It is possible that species of this subgenus lived under a variety of ecological conditions.

MYALINA (MYALINA) COPHA Newell, n. sp.

Plate 7, figure 3

Valves almost, but not quite equally convex, relatively small for a Myalina, shaped like Promytilus, with an elongate form, marked prosocline obliquity, pronounced anterior lobe and deep, welldefined sulcus; left valves quite rugose with projecting shell lamellae, right valves relatively smooth. Comparative measurements are given in Table 5. Measurements of the holotype are as follows: Length, 13.5 mm; height, 10.0 mm; greatestdimension, 14.0 mm; angle α, 49°.

Comparison—This species is closely similar in form with the juveniles of geologically younger myalinas, so presumably it is more like the ancestral stock than are the majority of myalinas. Comparison of Myalina (Myalina) copha with the undescribed Mississippian form shown on plate 7, figure 4, suggests that the latter is even more primitive than M. (M.) copha in the smaller value of the angle α. The close resemblance between M. (M.) copha and other primitive myalinas with Promytilus is probably homeomorphic. The two can be distinguished, of course, by differences in ligament areas and shell microstructure, distinctions by which the Myalinidae and Mytilidae are differentiated. Species of Promytilus known to me have very smooth shells and are equivalve. The possibility must be recognized that the material on which M. (M.) copha is based is not fully mature, in which case the small size of the individuals, and their primitive characters are juvenile features and do not have the significance that I have ascribed to them. However, the species is founded on quite a number of specimens, the largest of which shows the characters believed to be distinctive of the new species.

Material—The species is based on 16 specimens, holotype and topoparatypes, from thin, impure limestones. The material is not very well preserved, but the critical characters of the Myalinidae can be observed. The holotype (no. 7568) and some topoparatypes are at the University of Kansas. Other topoparatypes are at Yale University.

Occurrence—Associated with other pelecypods and brachiopods in thin impure limestones at the top of the Webber's Falls sandstone member of the Atoka formation (Lampasan), in the bed of Verdigris river at Okay, northeast of Muskogee, Okla.

MYALINA (MYALINA) LEPTA Newell, n. sp.

Plate 3, figures 6a, b, 9a, b

Shell slender, markedly prosocline, with nearly parallel anteroventral and posterodorsal margins; anterior lobe well developed in left valve, but wholly lacking in the right valve; hinge line about two thirds as long as the shell length; the two valves only slightly and about equally rugose, except for the anterior lobe of the left valve, which is roughened by imbricating growth lamellae; distinguished especially by the relatively great length (average form ratio, 1.1), the small angle α (average 55.5°), and a large value for the angle β (average, 119°); the outer ostracum in both valves is prismatic. Comparative measurements are given in table 5. Measurements of the holotype are as follows: length, 33 mm; height, 28 mm; greatest dimension, 40 mm; angle α, 59°; angle β, 120°.

ComparisonMyalina (Myalina) lepta is quite similar to the young of the majority of myalinas, and individuals are relatively small, characteristics which indicate that the species is a primitive representative of the genus. In many respects M. (M.) lepta is intermediate in characters, as well as geologic age, between M. (M.) copha and M. (M.) wyomingensis. From M. (M.) copha the present form is distinguished by having a less prominent anterior lobe on the left valve and no perceptible lobe on the right valve. The shell is somewhat shorter than in M. (M.) copha (form ratio 1.1 compared with 1.2), but the principal difference lies in the relative prominence of the umbones in the latter species. In M. (M.) lepta the right valve is relatively much smaller and flatter than the right valve in M. (M.) copha.

Myalina (Myalina) wyomingensis is distinctly more advanced than M. (M.) lepta in being shorter (form ratio 1.05 as against 1.1), in having a larger angle α (70.5° as compared with 55.5°), and a smaller value for the angle β (116° as compared with 119°).

Material—This species is based on four well preserved specimens that exhibit details of form and structure. The holotype (no. 21205) and three topoparatypes are in the collections of the University of Wisconsin.

Occurrence—Associated with brachiopods and other mollusks in fine, argillaceous shales of the middle Boggy formation (Desmoinesian) at the center of sec. 17, T. 2 N., R. 7 E., Oklahoma.

MYALINA (MYALINA) WYOMINGENSIS (Lea)

Plate 3, figures 1-4, 7, 10; plate 7, figure 6

Modiola wyomingensis Lea, 1853, Philadelphia Acad. Nat. Sci., Jour., vol. 2, p. 205, pl. 20, fig. 1a.
Myalina recurvirostris Meek and Worthen, 1860, Philadelphia Acad. Nat. Sci., Proc., p. 456; ——, 1866, Illinois Geol. Survey, vol. 2, p. 344, pl. 26, figs. 9a-c.
Myalina wyomingensis Girty, 1903, U. S. Geol. Survey, Prof. Paper 16, p. 422, pl. 8, figs. 8-13; ——, 1908, U.S. Nat. Museum, Proc., vol. 34, p. 290, pl. 19, figs. 2, 3.

Shell volselloid, with a well developed anterior lobe on the left valve and a distinct, though relatively small, lobe on the right valve; left valve quite rugose, right valve relatively smooth and somewhat smaller (discordant) and flatter than the left valve; beak of left valve prominent, rising well above the hinge line, and markedly prosogyre, right beak relatively blunt and not especially prominent; outer ostracum of both valves prismatic, composed of short polygonal prisms of calcite arranged normal to the shell surface; especially characterized by a relatively large angle α (average 70.5°) as compared with other volselloid myalinas. Dimensions of the holotype are as follows: length, 44 mm; height, 42 mm; greatest dimension, 52 mm; angle α, 67°; angle β, 125°; comparative measurements are given in table 5.

Comparison—Because of the well developed anterior robe this species has frequently been compared with species of Naiadites (Hind, 1895, p. 132), with which it has very little in common. There has been a tendency to refer species of Myalina having a prominent recurved left beak to M. recurvirostris, on the assumption that these characters of the beak are of specific value. Actually, nearly all of the myalinas have a relatively prominent left beak, which tends to overhang that of the right valve.

In any critical study Myalina (Myalina) wyomingensis is likely to be confused only with M. (M.) lepta, which has a markedly smaller angle α (55.5° as compared with 70.5°), or with the variety of M. (M.) wyomingensis thomasi, which has a more massive and distinctly larger shell.

Material—The material used in this study includes the holotype of Myalina (Myalina) wyomingensis (U.S. Nat. Museum, no. 33900), and the holotype and a topoparatype of M. recurvirostris (Univ. Illinois, no. x-214), as well as several hypotypes from several localities and horizons. The material is sufficiently well preserved for the study of all significant characters. In the original illustration of M. recurvirostris by Meek and Worthen, the holotype of that species was shown as a bivalved individual. The right valve is now lost but may be in the collections of the University of Illinois.

Occurrence—The species seems to have a fairly long range for a Myalina, being identified with some confidence from various horizons ranging from early Desmoinesian (Myalina bed, McCoy formation, near Bond, Colo.) through the Missourian, well into the Virgilian (Rico formation, 125 feet above base, Scotch Creek, Rico Quadrangle, Colorado; the associated Myalina (Orthomyalina) subquadrata is not known elsewhere above upper Virgilian). The horizon that yielded the holotype is unknown, the specimen having come from a coal mine near Wilkesbarre, Pa. The types of M. recurvirostris seemingly are Missourian in age ("about the horizon of coal number 11, of the Upper Coal Measures, La Salle, Illinois"). The species is common in the Rock Lake shale (sandstone facies) of the upper Missourian, in northeastern Kansas, and is known in the Graham formation (lower Virgilian) of Young county, Texas.

The majority of material available to me was collected from argillaceous shales, and some of the specimens retain both valves in apposition, denoting burial under quiet-water conditions. Other material occurs in a matrix of fine quartz sandstone, with the valves separated. The holotype is imbedded in a hard carbonaceous shale. Probably the conditions of burial were markedly different in the various occurrences. In those instances where the valves remain in apposition the burial apparently was in situ.

Figure 13—Comparison of form in (A) Myalina (Myalina) wyomingensis thomasi, n. var., from the Permian, with (B) Myalina (Myalina) wyomingensis, from the Pennsylvanian.

Comparison of form in (A) Myalina (Myalina) wyomingensis thomasi, n. var., from the Permian, with (B) Myalina (Myalina) wyomingensis, from the Pennsylvanian.

Figure 14—Evolutionary trends in the genus Myalina, s. s. With decreasing geologic age the size of the angle α increases, the shells become relatively shorter and higher, the posterodorsal angle β decreases, and the shells increase in absolute size. 1, Myalina sp., St. Louis limestone; 2, M. (M.) goldfussiana, the genotype, Viséan; 3, M. (M.) copha, Atoka; 4, M. (M.) lepta, Boggy; 5, M. (M.) wyomingensis, Desmoinesian to Wolfcampian; 6, M. (M.) arbala, Missourian; 7, M. (M.) glossoidea, Lansing to Douglas; 8, M. (M.) miopetina, Shawnee; 9, M. (M.) petina, Wabaunsee; 10, M. (M.) pliopetina, Wolfcampian; 11, M. (M.) copei, Wolfcampian. The points plotted represent the earliest occurrence in the stratigraphic record.

Evolutionary trends in the genus Myalina, s. s.

MYALINA (MYALINA) WYOMINGENSIS var. THOMASI Newell, n. var.

Plate 14, figures 15, 16

This variety is very like Myalina (Myalina) wyomingensis, s. s., differing from the typical form principally in being considerably more robust. The larger size of mature individuals, taken alone, would probably be insufficient grounds for the erection of a new variety. However, the new variety has a considerably smaller angle α, 55° on the average, as compared with about 70° in the typical form. M. (M.) wyomingensis thomasi is considerably younger than the youngest known specimens of M. (M.) wyomingensis, a fact that does not harmonize well with the general tendency for geologically younger myalinas to exhibit a more upright form than their older relatives. Comparative measurements are given in table 5. Dimensions of the holotype are as follows: length, 59 mm; height, 57 mm; greatest dimension, 70 mm; angle α, 70°. The posterodorsal extremity in the material at hand is broadly obtuse, and the angle of intersection between the hinge axis and the posterior margin is even greater than in M. (M.) wyomingensis. In general, the configuration is more primitive than that of M. (M.) wyomingensis, except perhaps for the fact that the anterior lobe in both valves of M. (M.) wyomingensis thomasi is less pronounced than in the typical variety. The ventral part of the shell seems to be slightly more slender in var. thomasi than in M. (M.) wyomingensis, s. s.

Figure 15—Inferred phylogeny of Myalina, s. s., determined by ontogeny and stratigraphic position. 1, Myalina sp., St. Louis limestone; 2, M. (M.) goldfussiana, Viséan; 3, M. (M.) copha, Atoka; 4, M. (M.) lepta, Boggy; 5, M. (M.) wyomingensis, Desmoinesian to Wolfcampian; 6, M. (M.) arbala, Missourian; 7, M. (M.) glossoidea, Lansing to Dougas; 8, M. (M.) miopetina, Shawnee; 9, M. (M.) petina, Wabaunsee; 10, M. (M.) pliopetina, Wolfcampian; 11, M. (M.) copei, Wolfcampian; 12, M. (M.) aviculoides, Wolfcampian; 13, M. (M.) wyorningensis thomasi, Satanka shale (Word?); 14, M. (M.) sinuata, Phosphoria (Word).

Inferred phylogeny of Myalina, s. s., determined by ontogeny and stratigraphic position.

Material—The new variety is based on a large collection of poorly preserved specimens from a single locality. Only 16 specimens were sufficiently instructive to be used in this study. All of the material occurs in the form of molds, and none of the original shell remains. The holotype, no. 21193, and most of the topoparatypes are at the University of Wisconsin. Some topoparatypes are at the University of Wyoming.

Occurrence—Local dolomitic limestone at the base of the Satanka shale (Word?), Red Mountain, sec. 9, T. 12 N., R. 76 W., Wyoming. The majority of specimens retain both valves in apposition.

MYALINA (MYALINA) ARBALA Newell, n, sp.

Plate 7, figures 9, 10

Obliquity slightly prosocline to acline; both valves provided with a prominent anterior lobe, that of the right valve being somewhat more pointed than the left; left valve only slightly rugose, covered with nonprismatic outer ostracum of seemingly homogeneous calcite, beak prominent, extending well above hinge line; right valve nearly smooth, slightly flatter than the left, with prismatic outer ostracum, beak pointed and not projected above hinge line; a broad, slight sinus just below the posterodorsal angle produces a fairly well defined auricle. Comparative measurements are given in table 5. Dimensions of the holotype are as follows: length, 63 mm; height, 77 mm; greatest dimension, 83 mm; angle α, 82° angle β, 84°.

Comparison—The upright form of this species and the small posterior auricle are reminiscent of Myalina (Orthomyalina) subquadrata. The resemblance is entirely superficial, however, because in the present form anterior lobes are well developed in both valves; whereas, in Orthomyalina, the anterior lobe is entirely lacking in right valves, and is missing or very obscure in left valves. In the majority of acline species of Myalina, s. s., as in M. (M.) arbala, the ventral margin is narrowly rounded, instead of broadly rounded as in species of Orthomyalina, tending to give the shell a somewhat triangular outline, the ventral margin being the apex of the triangle. M. (M.) arbala is closely akin to M. (M.) glossoidea, n. sp., and in absence of reasonably complete collections it might be difficult to distinguish them. There is a tendency for M. (M.) glossoidea, the younger of the two, to have slightly larger angle α, and especially a larger angle β. The situation is somewhat anomalous in that the older species, M. (M.) arbala, invariably exhibits a small rear auricle in adults, whereas such an auricle is more commonly missing in M. (M.) glossoidea. Judging from other myalinas, the development of the rear auricle is a mark of specialization, becoming most marked in the youngest species. Probably M. (M.) glossoidea is not derived from M. (M.) arbala, since the former species, although geologically the youngest, is in some respects more primitive than the latter. The two species are distinguished also by the fact that the length of that part of the shell halfway between the hinge and the ventral margin is longer in M. (M.) arbala than in M. (M.) glossoidea, giving to the former a relatively massive appearance.

Material—This species is based on fourteen specimens, all from the Winterset limestone (early Missourian). The material is well preserved, although part of the shell substance is recrystalized. The holotype is no. 7569, University of Kansas. Topoparatypes are at the University of Kansas, Yale University, and the U.S. National Museum.

Occurrence—Carbonaceous phase of the Winterset limestone (Bronson group, Missourian series) in the Kansas City area, associated with a fauna of brachiopods, pelecypods, nautiloids, and gastropods. The shells most commonly occur as separate valves, suggesting fairly strong current action in the depositional environment.

MYALINA (MYALINA) GLOSSOIDEA Newell, n. sp.

Plate 7, figures 1, b, 5, 7, 8a, b

Most of the comments on Myalina (Myalina) arbala apply to this species. Like the species just mentioned, M. (M.) glossoidea superficially resembles Orthomyalina, but can be distinguished from species of that subgenus by the prominent anterior lobe on each valve. The majority of specimens of M. (M.) glossoidea lack the posterior auricle that characterizes adults of M. (M.) arbala, and the lower part of the shell in M. (M.) glossoidea is shorter, therefore more slender than in the older species. Another species that is closely comparable with both M. (M.) glossoidea and M. (M.) arbala is the lower Permian M. (M.) aviculoides Meek and Hayden. Judging from the holotype, M. (M.) aviculoides is almost truly acline, having no conspicuous forward or backward obliquity, being more distinctly upright than either of the other two species in question. Seemingly, M. (M.) aviculoides is like M. (M.) glossoidea in lacking a posterior auricle.

Comparative measurements are given in table 5. Dimensions of the holotype of M. (M.) glossoidea are as follows: Length, 65 mm; height, 74 mm; greatest dimension, 82 mm; angle α, 75°; angle β, 100°.

Material—This species is based on eight specimens, all of which are relatively well preserved. The holotype is no. 7573, University of Kansas. Topoparatypes and paratypes are also at the University of Kansas.

Occurrence—The holotype and associated topoparatypes were found as isolated valves in calcareous shale in the upper Vinland shale, Stranger formation (early Virgilian), 1 1/2 miles south and 1/4 mile east of Homewood, Franklin county, Kansas. Paratypes from the Rock Lake shale (Stanton formation), upper Missourian, were collected near the center of the south side of sec. 27, T. 11 S., R. 22 E., Kansas. These specimens also occur as separated valves in calcareous shale.

MYALINA (MYALINA) AVICULOIDES Meek and Hayden

Plate 7, figure 2

Shell nearly acline, subquadrate, with a narrow, sharply rounded ventral margin, and prominent extended beaks; anterior lobe well defined; left valve relatively smooth, right valve and shell microstructure unknown. Comparative measurements are given in table 5. Dimensions of the lectotype are as follows: length, 37 mm; height, 41 mm; greatest dimension, 44 mm; angle α, 78°; angle β, 95°

Comparison—This species clearly belongs to the group of Myalina (Myalina) arbala and M. (M.) glossoidea, and is more closely akin to the latter. The three species are relatively advanced in having lost much of the primitive obliquity, but are somewhat anomalous in the weak development, or lack of a posterior auricle. In general form these species are similar to Orthomyalina, but the possession of a well developed anterior lobe, which is absent or obsolescent in Orthomyalina, indicates that these species are better classed with Myalina, s. s. From M. (M.) glossoidea the present species is different in having a truly quadrate posterodorsal extremity.

Material—This species is not well founded. The only very instructive specimen that I have seen is designated here as the lectotype, no. 2526 at the U.S. National Museum. The most critical part of the shell, the posterodorsal extremity, is not entirely complete. However, extrapolation of the growth lines indicates that it originally had a quadrate angle β. The shell is not especially well preserved.

Occurrence—The lectotype came from "South Cottonwood Creek," in central Chase county, Kansas. The most extensive outcrops in this area are in the Big Blue (Wolfcampian). Study of the rock matrix suggests that the type collection was taken from the Herington limestone, a formation that is known to have yielded other material described by Meek and Hayden. The rock in which the types occur is a light buff, somewhat cavernous dolomitic limestone, in which the majority of fossils are preserved as molds. A single specimen was collected from the upper nautiloid beds of the Casper sandstone, 1 mile south of Forelle, Wyo.

MYALINA (MYALINA) MIOPETINA Newell, n. sp.

Plate 3, figures 5a, b, 8a, b

Form roughly that of an isosceles triangle; the apex at the posterodorsal extremity; anteroventral margin only slightly sinuate, lobe well developed on both valves; left valve somewhat rugose, with homogeneous outer ostracum, right valve nearly smooth, with prismatic outer ostracum; posterior margin provided with an obscure sinus about one fourth of the height below the hinge margin, delimiting a rounded and obscure posterior auricle; ventral margin narrow, subangular; characterized by subquadrate posterodorsal angle (average value of angle β, 89°). Comparative measurements are given in table 5. Measurements of the holotype are as follows: length, 60 mm; height, 61 mm; greatest dimension, 75 mm; angle α, 63°; angle β, 82°.

ComparisonMyalina (Myalina) miopetina is a distinctive shell and need not be confused with other species. M. (M.) petina is similar in a general way, but has a more pronounced auricle and a broadly rounded ventral margin. The latter species has a shorter shell (form ratio. 0.85, as against 1.03 for M. (M.) miopetina), and a more acute angle β (76° as compared with 89° in M. (M.) miopetina). M. (M.) pliopetina is a considerably more advanced shell in having a very acute posterior auricle (47°).

Material—This species is based on three relatively complete, separate left valves and one fragmentary right valve, all from one locality. The holotype, no. 7565, and topoparatypes are at the University of Kansas.

Occurrence—From a shale somewhere near the horizon of the Lecompton limestone (lower Virgilian), probably the Stull shale, in sec. 7, T. 30 S., R.12 E., Kansas, associated with abundant Myalina (Orthomyalina) subquadrata and Septimyalina perattenuata.

MYALINA (MYALINA) PETINA Newell, n. sp.

Plate 6, figures 2a-3b

Shell relatively upright for a Myalina, with a short posterior auricle, and a subcircular ventral margin; anteroventral margin only slightly sinuate, lobe well developed in both valves; left valve somewhat rugose, with homogeneous outer ostracum, right valve smooth, with prismatic outer ostracum; characterized by a relatively small form ratio (length/height), about 0.85, and a subquadrate posterodorsal extremity (angle β about 76°). Comparative measurements are given in table 5. Measurements of the holotype are as follows: length, 46 mm; height, 56 mm; greatest dimension, 65 mm; angle α, 64°; angle β, 75°.

Comparison—The poorly developed posterior auricle of this species causes a superficial resemblance to Myalina (Myalina) miopetina. The subangular ventral margin, greater form ratio (1.03 as compared with 0.85), and larger angle β (89° as compared with 76°) in the latter species distinguishes the two. M. (M.) petina is conspicuously more upright in form than the relatively oblique M. (M.) miopetina.

Material—The species is based on two well preserved specimens, the holotype, no. 1526, and a paratype, no. 12003, Texas Bureau of Economic Geology.

Occurrence—Beds formerly classified as Harpersville, now thought to be upper Thrifty (Virgilian), 6 miles west of Cisco, Tex. The shells occur as isolated valves in calcareous shale.

MYALINA (MYALINA) PLIOPETINA Newell, n. sp.

Plate 6, figures 1a, b, 6

Form trigonal, with relatively straight anteroventral margin; anterior lobe well developed in both valves; left valve rugose, with homogeneous outer ostracum, right valve smooth to only slightly rugose, with prismatic outer ostracum; posterior margin marked by a relatively deep sinus, which varies in position from about a fourth of the height down from the hinge margin to an intermediate position between the ventral extremity and the hinge; rear auricle slender, extended; ventral margin narrow, nearly semicircular; characterized by the nearly straight anteroventral margin and correspondingly straight, oblique umbonal ridge. Comparative measurements are given in table 5. Measurements of the holotype are as follows: length, 56 mm; height, 59 mm; greatest dimension, 70 mm; angle α, 59°; angle β, 40° ±.

Comparison—In a general way Myalina (Myalina) pliopetina is similar to M. (M.) copei, with which it is associated locally, and it is like the much younger species M. (M.) sinuata, From M. (M.) copei it differs principally in the relatively acute angle with the dorsal margin (angle α about 62.3°), as compared with M. (M.) copei (average, about 84°). Unfortunately, M. (M.) sinuata is not well known, so that in some respects a detailed comparison with M. (M.) pliopetina is impossible. However, topotypes of M. (M.) sinuata are characterized by a relatively high convexity, correlated with an unusually thick visceral cavity. The living animal of M. (M.) sinuata must have been twice as thick as the animal of M. (M.) pliopetina. In other respects the two species were apparently quite similar, although some of the fragmentary specimens of M. (M.) sinuata that are available seem to have had a somewhat longer posterior auricle than in M. (M.) pliopetina.

Material—The species is based on five well preserved specimens from Texas. The holotype, no. 8364/4, and a topoparatype, no. 8364/1A, are at the American Museum. Other paratypes are at the University of Texas.

Occurrence—The holotype came from McKenzie's Trail, near Ft. Griffin, Shackelford county, Texas, and is one of the paratypes of Whitfield's Myalina copei. It is probable, but not certain, that Whitfield's collection came from a single horizon and locality. It seems significant that in none of the various other collections available to me do the two species M. (M.) copei and M. (M.) pliopetina occur in direct association. According to F. B. Plummer (personal communication) it is probable that the Whitfield material from near Ft. Griffin came from the Moran or the Pueblo formation (Wolfcampian). The preservation of the shells is similar to that found in other material known to have been collected from the Moran formation. The species also is represented in collections from the Pueblo of Texas (Wolfcampian) and the Fort Riley (Wolfcampian) of Kansas.

The species is represented in argillaceous shales and in dolomitic limestone. A few of the specimens retain both valves in position indicating burial under fairly quiet conditions.

MYALINA (MYALINA) COPEI Whitfield

Plate 4, figures 1a-2c; plate 5, figures 1-3.

Myalina copei Whitfield, 1902, Am. Mus. Nat. History Bull., vol. 16, pp. 63-66, text fig. 2, not text fig. 1.

Shell of adults acline or nearly so, with narrow, rounded ventral margin and extended anterior and posterior terminations of the dorsal part of the shell, giving the shell roughly the form of a "T"; anterior lobe distinct, but relatively small on both valves; posterodorsal extremity in an elongate and slender auricle having a rounded termination, delimited below by deeply incised sinus, the apex of which is from a fourth to a third of the shell height below the dorsal margin; anterior and posterior margins of the shell nearly perpendicular in the lower third of the shell; anterior margin concave and broadly arcuate, paralleling the regularly arcuate umbonal ridge; left valve rugose, outer ostracum homogeneous, right valve nearly smooth, with a prismatic outer ostracum; right valve flatter and slightly smaller than the left valve (discordant), particularly along the anterior border, where the margin of the right valve fits within the edge of the left valve; mature shells show a tendency for the socket of the left valve to develop an obscure median ridge, or tooth, the tooth of the right valve being correspondingly divided by a narrow and shallow furrow which mayor may not occur at the middle of the tooth; mature shells relatively large, compared with other myalinas, ranging up to a maximum of around 100 mm for the greatest dimension. Comparative measurements are given in table 5. Dimensions of the lectotype are: length (restored), 82 mm; height, 82 mm; angle α, 77°.

Comparison—The greatly extended posterior auricle of this species distinguishes it from all known myalinas, excepting Myalina (Myalina) pliopetina and M. (M.) sinuata. Both of the last two forms are characterized by oblique shape and a nearly straight anterior margin, differing rather markedly from the arcuate anterior margin and upright form of M. (M.) copei.

Material—The type collection consists of several separated specimens and one bivalved individual (lectotype, no. 8364/1) at the American Museum. Although more or less fragmentary, the material is well preserved. Together with various supplementary hypotypes, the collection of instructive examples available for this study numbered 17 specimens. Several scores of specimens, too incomplete for critical measurements, were examined.

Occurrence—This species is highly characteristic of the Wolfcampian series and equivalents. It has been recognized in many of the fossiliferous Wolfcampian limestones in Nebraska and Kansas, is abundant at several horizons in the Pueblo, Moran, and Putnam formations of north-central Texas, and occurs sparingly in the Hueco formation of the Hueco Mountains, West Texas. The types were collected on McKenzie's Trail, near Fort Griffin, Shackelford county, Texas, possibly from the Pueblo formation, but more probably from the Moran formation. Myalina (Myalina) pliopetina occurs in the type collection of M. (M.) copei. Presumably all of the material came from the same horizon. However, the two species are not associated together in any of the other collections that I have seen. In the northern Mid-Continent region of Kansas and Nebraska the species has been recognized in the Neva, Funston, Florence, Fort Riley, Gage, Cresswell, Krider, and Herington limestones. In all of these occurrences the shells are found in argillaceous limestones or dolomitic limestones. In Texas the shells occur in argillaceous shales and impure limestones. Bivalved specimens are not uncommon.

MYALINA (MYALINA) SINUATA Branson

Plate 6, figures 4 and 5

Myalina sinuata BRANSON, 1930, Missouri University Studies, vol. 5, p. 45, pl. 12, fig. 1.

Although this species is almost certainly distinct from the other myalinas herein described, it is not adequately known. All of the specimens of Myalina (Myalina) sinuata known to me are so fragmentary that a satisfactory description cannot yet be made. The available collection of fragmentary topotypes indicates a very large and massive shell, larger in adults than any other myalinas known to me. Some of the specimens seemingly were originally over 100 mm in height and had an even greater length. The most distinctive feature of the internal molds is their very great convexity, indicating a visceral cavity of much greater depth than that of M. (M.) pliopetina and M. (M.) copei. Judging from the fragments, the complete shell probably was very like that of M. (M.) pliopetina in form, with a rather marked prosocline obliquity. Adults, however, were seemingly nearly half again as convex as in M. (M.) pliopetina and probably had a slightly more extended posterior auricle. The measurements given in table 5 are for internal molds and therefore cannot be relied on to give an accurate impression of external shell characters. Nevertheless, the characters which seemingly distinguish the species are easily recognized.

Figure 16—Anterodorsal-posteroventral sections of (A) Myalina (Myalina) pliopetina, n. sp., and (B) Myalina (Myalina) sinuata Branson, drawn parallel to the greatest dimension of the shell, to show the relative shell convexities of the two species. The diagram of M. (M.) sinuata is a reconstruction, based on internal molds.

Anterodorsal-posteroventral sections of (A) Myalina (Myalina) pliopetina, n. sp., and (B) Myalina (Myalina) sinuata Branson.

Material—The holotype, which is supposed to be in the collections of Colorado College, at Colorado Springs, cannot be located. A dozen or so fragmentary topotypes, chiefly internal molds, have been made available for this study by Alfred Fischer. These specimens are at the University of Wisconsin. Two inadequate internal molds, from the upper part of the Kaibab limestone of northern Arizona, were lent by the curators of the Museum of Northern Arizona.

Occurrence—Upper limestone of the Phosphoria (Word?) of the Wind River Mts., Wyoming; and the upper part of the Kaibab limestone (upper Bellerophon beds, Word?), northern Arizona. The type specimen was collected by Branson at Bull Lake, in the Wind River Mts., near Du Bois, Wyo. The Phosphoria specimens occur in cherty light gray to bluish-gray limestone; those from the Kaibab occur in a matrix of cavernous dolomitic limestone.

Subgenus ORTHOMYALINA Newell, n. subgen.

Subgenotype, Myalina slocomi Sayre. Geologic range of subgenotype, Desmoinesian-Virgilian, Pennsylvanian.

Subquadrate, nearly acline, massive-shelled myalinas, in which the anterior lobe is obsolete in both valves, and the posterior adductor system is placed far down toward the posteroventral margin of the valves (text fig. 6G).

Geologic range—Desmoinesian-Wolfcampian.

Remarks—The early ontogeny of shells of Orthomyalina, as revealed in growth-lines on the umbones, shows clearly that this phyletic group of shells sprang from a prosocline, lobate Myalina like Myalina (Myalina) lepta, possibly in the early Desmoinesian epoch, because representatives of Orthomyalina are as yet unknown in rocks older than mid-Desmoinesian. As in other Myalinidae, there is a marked tendency for the geologically younger species to develop a posterior auricle, and an auriculate form, O. subquadrata, replaces the quadrate varieties in the Wabaunsee and Wolfcamp horizons.

Species of Orthomyalina are by far the most abundant representatives of the Myalinidae in the Missourian and Virgilian series of the Pennsylvanian in the Mid-Continent region. Commonly the shells are crowded by the thousands in a matrix of limestone or calcareous shale, to the near exclusion of other fossils. Less commonly they occur in a sandstone matrix. Juvenile shells are practically unknown in these Orthomyalina beds. Inspection of mature shells reveals that they reached a large size before the shell became thickened (pl. 8, figs. 1, 3). New ligament grooves were added as the shell was thickened. Consequently, the number of ligament grooves is an index of the relative maturity of a shell. Probably the very small juveniles were so fragile and, like Mytilus, so heavily charged with organic material that they were destroyed during the fossilization process.

Bivalved shells of Orthomyalina are so common that it is certain that the living animals favored relatively quiet waters. The construction of hinge and ligament was such that even moderate current action after death would have quickly separated the valves. It does not follow that the water inhabited by these animals was relatively deep, because the very flat bottom that seemingly characterized so many of the Pennsylvanian seas in the Mid-Continent area would tend to preclude extensive and vigorous surf action.

MYALINA (ORTHOMYALINA) SLOCOMI Sayre

Plate 8, figures 1-6; plate 10, figure 6

Myalina (?) slocomi Sayre, 1931, Kansas Geol. Survey, Bull. 17, p. 117, pl. 11, figs. 1-la. (Date of imprint, 1930) Myalina subquadrata of authors, in part.

Shell massive, approximately acline, sublinguate, and commonly slightly expanded ventrally, higher than long in about the ratio of 3/2; beaks slender, projected forward beyond the front margin of the body of the shell below, anterior lobe absent on both valves except in the very early juvenile stages; posterodorsal extremity broadly obtuse to rounded (compare pl. 8, figs. 1 and 4), rarely subquadrate (fig. 6). In 25 well-preserved specimens from the Stull shale, Shawnee group, Virgilian, near Waverly, Kans., the dimensions are as follows: length, maximum 63 mm, average, 55 mm, minimum, 44 mm; height, maximum, 90 mm, average, 78 mm, minimum, 64 mm; greatest dimension, maximum, 93 mm, average, 81 mm, minimum, 67 mm; angle α, maximum, 91°, average, 87°, minimum, 76°.

Comparison—The typical variety of Myalina (Orthomyalina) slocomi is readily distinguished from the variety M. (O.) slocomi parvula and the species M. (O.) ampla by notable differences in size. M. (O.) slocomi parvula seems to be a dwarf, reaching full maturity (as shown by the thickened shell and large number of ligament grooves) at about half the size of typical M. (O.) slocomi. Shells of M. (O.) ampla are not only somewhat larger than M. (O.) slocomi, but are relatively longer, with shorter beaks, and they show no tendency for the ventral expansion of the shell typical of M. (O.) slocomi. There are rare specimens in the lower Wabaunsee (Virgilian) rocks that are nearly intermediate in form between M. (O.) subquadrata and M. (O.) slocomi. Such specimens form less than one per cent of the collections, however, and the possession of an obscure posterior auricle distinguishes M. (O.) subquadrata from the older form.

The holotype of Myalina (Orthomyalina) slocomi Sayre (pl. 8, fig. 6) is a young, therefore thin-shelled representative of the species from the Westerville oolite (Missourian) of the Kansas City region.

Material—The collections at my disposal are more voluminous than for any other species. There are more than a thousand valves in the assembled collection, representing almost every fossiliferous horizon in the mid-continent region from Wewoka (late Desmoinesian) to early Wabaunsee (Virgilian). Naturally there is considerable variation exhibited in this material. Intensive studies indicate the impracticability of attempt to recognize stratigraphically useful subdivisions of the typical variety of Myalina (Orthomyalina) slocomi as herein defined.

The holotype, a poorly preserved immature left valve, is in the University of Kansas collections, no. 50.1.

OccurrenceMyalina (Orthomyalina) slocomi is probably the most abundant of all the myalinas known to me. Numerous beds in the Missourian and Virgilian series are crowded with the shells of this species. Septimyalina perattenuata is a common companion of M. (O.) slocomi in these myalina beds. Horizons in the northern Mid-Continent area, at which these species are, at least locally, extremely abundant are: Farley limestone (Wyandotte formation), Rock Lake shale (Stanton formation), Vinland shale (Stranger formation, Heumader shale (Oread formation), Stull shale (Kanwaka formation), Ost limestone (Tecumseh formation). These horizons are in the Missourian and lower Virgilian series.

In the northern Mid-Continent area the species is at least sparsely represented in almost every mollusk-bearing formation from the Marmaton group (upper Desmoinesian) to the Howard limestone (lower Wabaunsee). In Texas the species is well represented in the Mineral Wells, Graford, Brad, and Graham formations, of Missourian and early Virgilian age. In Illinois the species has been recognized in the Shoal Creek, Centralia, and Macoupin formations of Missourian age.

The shells occur in various kinds of sediments, but most coromonly are found in calcareous shales. In this instance the valves commonly are in direct apposition, except where they have been separated by weathering.

MYALINA (ORTHOMYALINA) SLOCOMI var. PARVULA Newell, n. var.

Plate 10, figures 2 and 3

Shell small, commonly less than 60 mm high and 40 mm long, otherwise like the typical variety of Myalina (Orthomyalina) slocomi. The following measurements are taken from 10 valves of the type collection, collected from the Rakes Creek shale (Tecumseh formation), Shawnee group, 2 1/2 miles northwest of Michigan Valley, Kan. Height, maximum, 63 mm, average, 49 mm, minimum, 40 mm; length, maximum, 41 mm, average, 36 mm, minimum, 30 mm; angle α, maximum, 93°, average, 84°, minimum 76°.

The holotype has a height of 50 mm, length, 35 mm, angle α, 80°.

It is probable that this variety is simply an ecological variant of M. (O.) slocomi. The large number of ligament grooves (up to 15) shown by some of the specimens demonstrates their advanced maturity, so they are not to be regarded simply as the young of the typical variety.

Material—The new variety is based on more than a score of specimens from a single locality. They are not associated with larger myalinas of the M. (O.) slocomi type. The holotype is numbered 7582 at the University of Kansas, where the entire collection of types is stored.

Occurrence—Calcareous shale of the Rakes Creek member, Tecumseh formation, early Virgilian, where the variety is rather widely distributed through central eastern Kansas. It was found also at a slightly lower (?) horizon, in the fossiliferous shales of the Jacksboro limestone, Caddo Creek group, 4 1/2 miles east of Jacksboro, Tex. In all of the known occurrences the shells seem to have been buried in relatively quiet water.

MYALINA (ORTHOMYALINA) AMPLA Meek and Hayden

Plate 10, figure 5

Myalina ampla Meek and Hayden, 1864, Paleontology Upper Missouri, Smiths. Contr. Knowledge, 172, p. 32.

Shell robust, with relatively short beaks and only slightly incurved anterior margin; compared with other orthomyalinas the shell is relatively longer, therefore not so slender as commonly is the case in species of the subgenus. The holotype has a height of 98 mm and a length of 65 mm. The angle α is approximately 90°.

Comparison—This species is quite similar to Myalina (Orthomyalina) slocomi, and it is possible that the shells referred here to M. (O.) ampla may be ecological variants of M. (O.) slocomi. The differences between the two are sufficiently distinctive, however, to warrant retention of both names. M. (O.) ampla is characterized by a larger, more massive shell, which is appreciably less slender than in M. (O.) slocomi. The beaks are less produced and somewhat more blunt in the latter species. Also, there is a less pronounced tendency in M. (O.) ampla for the ventral edge of the shell to be expanded than in M. (O.) slocomi or M. (O.) subquadrata.

Material—The only moderately complete valve of this species at my disposal is the holotype, no. 998, at the U.S. National Museum. However, the same species is abundantly represented in my collections from the Lansing group (Merriam limestone, Plattsburg formation) by fragmentary internal molds. It is almost certain that the holotype came from the same horizon.

Occurrence—Judging from the field occurrence this species preferred clear and quiet waters. All of the material even tentatively referred to M. (O.) ampla occurs in a limestone matrix; consequently, it is very difficult to collect good specimens. The majority of material is bivalved.

The holotype was collected along the Missouri River, below Ft. Leavenworth, Kan., from an undesignated horizon. The entire Lansing group is exposed in the river bluffs in this place, but there is only one horizon of Myalina (Orthomyalina) ampla—the Merriam limestone, basal member of the Plattsburg formation. Southwestward, about as far as Miami county, Kansas, large specimens of M. (O.) ampla are common in the Merriam limestone but are rare or absent farther to the southwest. In northwestern Missouri the species is common at this horizon. Questionable representatives of the species have been found in the Winterset limestone around Kansas City and in the lower limestone of the Fort Scott formation, Fort Scott, Kan.

MYALINA (ORTHOMYALINA) SUBQUADRATA

Plate 9, figures 1-4

Myalina subquadrata Shumard, 1855, Missouri Geol. Survey, Ann. Repts.1 and 2, p. 207, pl. C, fig. 17.
Myalina deltoidea Gabb, 1859, Philadelphia Acad. Nat. Sci., Proc., p. 297; ——, 1860, idem., p. 55, pl. 1, figs. 1a-c.

Quadrate orthomyalinas characterized by a relatively long hinge, a shallow, broad sinus at the posterior margin, and a more or less distinct, but short posterior auricle; anteroventral margin somewhat produced, giving some shells a slightly opisthocline obliquity, which is not indicated in measurements of the angle α because the umbonal ridge becomes flattened and indistinguishable before the margin is reached in adults. Dimensions of 10 specimens from the Caneyville formation (Kansas and Oklahoma), upper Virgilian, are as follows: length, maximum, 78 mm, average 64 mm, minimum, 53 mm; height, maximum, 97 mm, average, 84 mm, minimum, 63 mm; angle α, maximum, 97°, average, 85°, minimum, 75°.

Comparison—The rear sinus and auricle in Myalina (Orthomyalina) subquadrata distinguish this species from the similar M. (O.) slocomi. Both species show some variations in form, and it is not surprising that a very few upper Shawnee and lower Wabaunsee forms are so nearly intermediate between the two species that they might with equal propriety be referred to either species. Experience shows that a small proportion, less than one per cent, of the specimens of M. (O.) slocomi from lower Shawnee beds possess a quadrate posterodorsal extremity, as shown in plate 10, figure 4. Even these specimens show litttle or no posterior sinus. Upper Wabaunsee collections of M. (O.) subquadrata contain a very few specimens, like that of plate 9, figure 4, which closely resemble M. (O.) slocomi in form. All of these specimens are relatively quadrate and show at least a slight posterior sinus. Our collections of orthomyalinas are not so complete for intervening horizons, i.e., upper Shawnee and lower Wabaunsee. Probably separation of the two species must be based on statistical studies in these transition beds. Ontogenies of M. (O.) subquadrata reveal without question that this species developed directly out of M. (O.) slocomi. Judging from the phylogenies of other myalinas we are justified in anticipating the discovery of a long-winged descendant of M. (O.) subquadrata in upper Wolfcampian or Leonardian strata. However, thus far M. (O.) subquadrata is the youngest Orthomyalina known.

Material—As nearly as I can determine the holotypes of Myalina (Orthomyalina) subquadrata and M. deltoidea are lost to science. Because of that fact, I would not hesitate to urge abandonment of both species were it not for the fact that the horizon and locality for the type of each species can be fairly well determined. Furthermore, illustrations of the type specimens were published so that their significant features are readily evident. M. (O.) subquadrata was collected "on Missouri River, two miles below the mouth of the Little Nemaha" at Aspinwall, Nemaha county, Neb. According to Condra (1927, p. 112) the horizon of abundant Myalina in this area is zone 4 of the Table Creek shale, upper Wabaunsee.

Myalina deltoidea seems to be synonymous with M. (O.) subquadrata on subjective grounds, and came originally from about the same horizon. According to Gabb, his holotype came from just above a coal bed at Ft. Belknap, Tex. This horizon clearly is the lower part of the Harpersville formation of the older literature or the upper Thrifty of recent usage (Cheney, 1940, p. 66). The horizon of abundant myalinas at Ft. Belknap occurs just above the coal in the upper Thrifty formation. The age of these beds in terms of the northern Mid-Continent section is uppermost Wabaunsee.

OccurrenceMyalina (Orthomyalina) subquadrata has been reported in the literature from various formations throughout the Pennsylvanian and Permian. Needless to say, the great majority of these citations are in error. Carelessness in the identification of myalinas has become the rule, so that traditionally almost every large Myalina in the Late Paleozoic has been called Myalina subquadrata without regard to form or other significant characters. This species is abundant only in a limited part of the upper Wabaunsee. Early in the progress of this study I was under the impression that the species is confined to the Caneyville formation. Acquisition of further collections, totaling some hundreds of specimens, from Nebraska, Kansas, Oklahoma, Texas, and New Mexico, reveals that the species appears at least as low as the Auburn shale (mid-Wabaunsee) and ranges at least as high as the Foraker limestone (lower Wolfcampian). So far as I am aware. the species occurs in persistent "Myalina beds" only in the Friedrich, French Creek, and Caneyville formations of the upper Wabaunsee and in the equivalent upper Thrifty formation of Texas. Other occurrences are for the main part sporadic. Dr. M. L. Thompson contributed typical representatives of the species from the upper Magdalena limestone of central New Mexico, where it is associated with species of Schwagerina.

Figure 17—Ontogeny of Myalina (Orthomyalina) subquadrata, correlated with the phylogeny. The form stage represented by A is nearly identical in shape with the Lower Carboniferous Myalina (Myalina) goldfussiana; B corresponds to M. (M.) lepta, from the lower Desmoinesian; C corresponds to M. (M.) wyomingensis, which appears first in the lower Desmoinesian; D and E correspond to M. (Orthomyalina) slocomi, which appears first in the upper Desmoinesian; F represents the adult form of M. (O.) subquadrata, appearing first in the Wabaunsee equivalents, upper Virgilian.

Ontogeny of Myalina (Orthomyalina) subquadrata, correlated with the phylogeny.

Subgenus MYALINELLA Newell, n. subgen.

Subgenotype, Myalina meeki Dunbar. Geologic range of the subgenotype, Desmoinesian (?) to lower Leonardian.

Shell strongly prosocline, triangular, with nearly straight anteroventral posterior and dorsal margins; beaks slender, not lobed; left valve only slightly more rugose than the right valve, outer ostracum of both valves composed of irregular polygonal prisms about 10 to 15 microns in diameter; shell very thin, seldom exceeding 1 mm. even in the largest shells; shells commonly less than 20 mm in the greatest dimension; musculature unknown.

Geologic range—Desmoinesian-upper Guadalupian? (Quartermaster).

Remarks—The marked prosocline obliquity and triangular form of shells of Myalinella are "primitive" features when considering the family as a whole. Yet, the near or complete lack of an anterior lobe below the beaks is clearly a mark of specialization. Juveniles and the geologically older forms generally display an anterior lobe. Myalinella certainly is not as closely related to Myalina, s.s., as is Orthomyalina. Even in the very early ontogeny, the form of the shell does not closely duplicate any of the typical myalinas. The lack of differentiation in shell structure between the two valves also sets this group apart from Orthomyalina and Myalina. s.s. For these reasons, it might logically be argued that Myalinella is deserving of separate generic rank. Because of the fragile character of the shells, good material is very rare and my information on this group is far from complete. Consequently, a more conservative course has been adopted; that is, Myalinella is classed under Myalina, as a subgenus.

The best evidence that Myalinella constitutes a genetic series lies in the discovery that one of the characteristic evolutionary trends of the Mytilacea is displayed by the myalinellas as they are traced through the Late Paleozoic. In all significant respects but one, the shells of this subgenus are constant in various horizons. The only conspicuous change is seen in progressive reduction of the posterodorsal angle (angle β). The oldest species has a remarkably obtuse angle, larger than any other myalinid known to me. At the higher horizons this angle is progressively reduced, though none of the species thus far recognized reaches the place where the angle β is quadrate.

The subgenotype was first recognized with associations. that suggest a freshwater or brackishwater habitat (Dunbar 1924, p. 201), and it has been noted at other horizons in association with a well preserved terrestrial fauna and flora. Myalina (Myalinella) meeki and some other species of the subgenus commonly occur in association with marine invertebrates, however. I am impelled to the conclusion that species of Myalinella were tolerant of rather great variations in salinity of the water. That these animals lived at least partly in quiet waters is indicated by the fact that many of the specimens from various horizons and localities retain the two valves in close association.

Figure 18—Form series in the subgenus Myalinella. A, Myalina (Myalinella) cuneiformis Gurley, Desmoinesian; B, M. (M.) meeki Dunbar, Desmoinesian?-Leonardian; C, M. (M.) acutirostris Newell and Burma, Guadalupian. All figures X 1.25. [Image scaled and magnifications adjusted for web presentation.]

Form series in the subgenus Myalinella.

MYALINA (MYALINELLA) MEEKI Dunbar

Plate 14. figures 7-14

Myalina meeki Dunbar, 1924, Am. Jour. Sci., vol. 7, p. 201, fig. 3.
?Myalina pernaformis Cox, 1857, Geol. Rep. Kentucky, vol. 3, p. 569, pl. 8, fig. 8.
?Myalina? exasperata Beede, 1899, Kansas Univ. Quart., vol. 8, p. 128, pl. 32.

Shell small, fragile; form slender, subtriangular, posteroventral margin, however, being rather broadly rounded; anteroventrallobe almost imperceptible in both valves, except where shells have been distorted by mashing; beaks extended, slender; umbonal ridge nearly straight except in the umbonal region, where it is slightly arched; inequivalve, the right valve being slightly less convex and smaller around the margin than the left valve; surface of both valves relatively smooth; outer ostracum of both valves composed of short irregular polygonal prisms of calcite 25 to 40 microns in diameter; internal characters unknown. Measurements based on six type specimens are as follows:

Measurements of Myalina (Myalinella) meeki
  Length
(mm)
Height
(mm)
Angle
α
Angle
β
Greatest
dimen.
(mm)
Maximum 52 43 47° 125° 65
Average 26 22 43° 127° 33
Minimum 17 16 40° 131° 21
Holotype 52 43 45° 129° 65

Comparison—This form is distinguished from other myalinellas by the distinctive size of the angle α. From primitive myalinas like Myalina (Myalina) wyomingensis, which it superficially resembles, this species differs in the relatively smooth left valve and absence of an anterior lobe. The form of the beaks is quite similar to that of many species of Septimyalina. The resemblance is heightened by the fact that left valves of some species of Septimyalina, such as S. scitula, are nonrugose. The posterior margin of the shell in S. scitula is generally rounded, instead of straight, and the posteriodorsal corner of the shell is not as angular as ordinarily is the case with Myalina (Myalinella) meeki. S. scitula, like other species of Septimyalina, is a relatively thick-shelled form and possesses an umbonal deck, or septum, a feature which is lacking in all myalinellas.

Two poorly known species are similar to Myalina (Myalinella) meeki and possibly are identical with it. Myalina pernaformis Cox was described from the Providence limestone above Coal No. 11 at Providence, Ky. The horizon is in the Allegheny formation, approximately equivalent to the Pawnee limestone (Marmaton) of Kansas. The original description and figure are not adequate for recognition of this species in any critical work, but discovery of topotype specimens will some day solve the status of this species.

Myalina? exasperata Beede (pl. 10, figs. 1a, b) was based on very poor specimens from an unknown horizon associated with a coal bed (Nodaway coal?, Wabaunsee) west of Topeka, Kans. In the absence of good authentic specimens, this form is unrecognizable.

Material—The specimens on which this description is based include some dozen or so shells, most of which show the original form quite well. A few of the specimens retain the original shell, and are undeformed. All of the type specimens occur in soft, calcareous shale, and are somewhat flattened by compaction. In all of these latter specimens, the inner ostracum of the shell has been removed by solution.

The types are the property of Yale University, no. 3496.

Occurrence—The species seems to be longranging. A specimen from the lower Fort Scott limestone (Marmaton group), (pl. 14, fig. 11) is the oldest shell of the Myalina (Myalinella) meeki type in my collections. This shell is roughly like the form described by Cox as M. pernaformis, from about the same horizon (limestone above Coal No. 11) at Providence, Hopkins county, Ky. In the absence of topotype specimens, however, it is by no means certain what kind of a shell Cox described. Other specimens have been collected from the Walchia beds in the Rock Lake shale near Garnett, Kans., where they are associated with a peculiar mixture of plant remains, reptiles, fishes, scorpions, and macerated remains of marine invertebrates. Many of the fossil myalinellas are bivalved and therefore occur near the spot where they lived. The best specimens occur associated with a normal marine fauna, in a thin, sandy limestone in the Calhoun shale (Shawnee), at the NW cor. sec. 1, T. 21 S., R. 13 E., Coffey county, Kansas. The type specimens came from the famous fossil insect beds in the Carlton limestone, 3 1/2 miles southeast of Elmo, Kan. The Carlton limestone occurs in the Wellington shale (upper Sumner) above the great salt beds of the Wellington, lower Permian. These shells probably were deposited in quiet waters, because many of them are bivalved. The ordinary marine species are absent. The salinity of the water may have been higher than normal sea water, or on the other hand it may have been relatively low, or even fresh (Dunbar 1924, p. 201).

MYALINA (MYALINELLA) CUNEIFORMIS Gurley

Plate 14, figures 1-4

Myalina cuneiformis Gurley, 1883, New Carboniferous fossils, Bull. 1, p. 4; ——, Girty, 1903, U.S. Geol. Survey, Prof. Paper 16, p. 420, pl. 8, figs. 14-17.

Shell small for a Myalina, mature specimens measuring less than one inch in length, fragile; form slender, subtriangular, the cardinal and posterior margins forming two nearly equal sides and the fairly straight anteroventral margin the base of an obtuse isosceles triangle; anteroventrallobe absent in right valve, barely perceptible in the left; beaks extended, very slender, depressed slightly at their extremities; angle α relatively small, around 35°; angle β broadly obtuse, about 130°; umbonal ridge slightly arched in the region of the umbones, but tending to become straight in the mature part of the shell; inequivalve, the right valve being distinctly the smaller of the two; surface of both valves relatively smooth, that of the left being only slightly more rugose than the right valve; ligament area narrow, seemingly with only two or three relatively fine ligament grooves, in harmony with the thinness of the shell; right valve provided with a single prominent but thin hinge tooth, which occurs just. behind the beak; the axis of elongation of this tooth is about 125° or more to the hinge axis, being directed downward and forward; left valve provided with a deep and narrow socket hardly more than a notch in the shell margin, corresponding to the tooth of the complementary valve, and occuring immediately below and behind the beak; musculature and shell microstructure unknown.

Measurements of two specimens are as follows: (1) lectotype, a juvenile right valve, length, ±13 mm; height, 8 mm; convexity, ±3 mm hinge length, ±9 mm; greatest dimension (measured parallel to the umbonal ridge), ±14 mm; angle α, 40°; angle β, 130°; (2) hypotype, mature bivalved specimen, length, 20 mm; height, 13.5 mm hinge length, 14 mm; convexity left valve, ±3.5 mm; convexity right valve, ±3 mm; greatest dimension, 23 mm; angle α, 45°; angle β, 125°.

Comparison—This species is readily distinguished from other myalinellas by the remarkably small angle α.

Material—Gurley's types before me are two in number, a juvenile, which, being the better speciman, is designated here as the lectotype (Univ. Chicago, no. 6318), and a larger badly crushed bivalved paratype, about 25 mm long. Both specimens occur in dark gray to black micaceous, sandy limestone. Without good supplementary material, I would have been uncertain about the recognition of this species. Such material, fortunately, is available in several silicified specimens from the lower part of the Tensleep sandstone of western Wyoming (Yale University, nos. 16342 to 16344). Most of the above description was based on the Tensleep material. These specimens occur in a siliceous dolomitic limestone matrix and were collected by Dr. J. D. Love on Black Mountain, eastern end of the Washakie Range, Wyoming (Love, J. D., 1939, p. 30). Only one of the specimens at hand is not fragmentary.

Occurrence—Girty (1903, p. 421), who studied the Carboniferous faunas from Ouray, Colo., whence the types of Myalina (Myalinella) cuneiformis came, concluded that Gurley obtained the original specimens from the lowest exposed Carboniferous beds at Ouray, probably from the cliffs on either side of a little cascade just east of town. He classed these beds as Hermosa? formation. He also recognized the species in the "Weber" limestone in the Crested Butte district, Colorado. In the collection at hand, the species is represented by a few silicified specimens, mainly fragments, collected by J. D. Love, 30 feet above the base of the Tensleep sandstone in section 1 or 2, T. 6 N., R. 4 W., Wyoming. The Hermosa and Tensleep formations are probably about the same age and are generally correlated with some part of the Wedekindellina zone of the Cherokee of the Mid-Continent region.

MYALINA (MYALINELLA) ACUTIROSTRIS Newell and Burma

Plate 14, figures 17 -19

Myalina sp. Newell, 1940, Geol. Soc. America, Bull., vol. 51, p. 256, pl. 2, fig. 1.
Myalina acutirostris Newell and Burma in Roth, 1941, Jour. Paleontology, vol. 15, p. 315, pl; 45, figs. 11-15.

Shell subtriangular to subrhombic, relatively small, fragile, slightly more upright in form than other myalinellas; beaks slender, extended anteriorly; surface of both valves relatively smooth; hinge relatively long, about 0.6 to 0.7 as long as the shell length and about equal to the shell height; angle α increasing during ontogeny to around 60° in mature specimens angle β decreasing during ontogeny to about 110° in mature specimens; ventral margin of shell tending to be narrowly rounded and placed well back of a midpoint on the shell. Measurements, based on five specimens, are as follows:

Measurements of Myalina (Myalinella) acutirostris
  Length
(mm)
Height
(mm)
Angle
α
Angle
β
Maximum 25 19 60° 128°
Average 16 12 53° 118°
Minimum 7.5 6.5 45° 110°
Holotype 25.0 19.0 45° 110°

Comparison—The angle α is larger in adults of Myalina (Myalinella) acutirostris than in other myalinellas and the angle β is smaller. Otherwise, the species closely resembles M. (M.) meeki and M. (M.) cuneiformis. The general appearance of M. (M.) acutirostris recalls Septimyalina scitula (Myalina perattenuata of authors), particularly the shape of the slender, projecting beaks. The latter form, however, in addition to possession of an umbonal deck, is a more massive, thick-shelled form, and tends to be distinctly more upright.

Material—The species was based on four left valves and one right valve. In general, the specimens are fairly well preserved, although features of hinge and musculature are not well shown. The holotype, no. 20846 and three topoparatypes, nos. 20843-20845, are at the University of Wisconsin. One paratype, no. 16326, is at Yale University.

Occurrence—Restricted, so far as is known, to sandy and oolitic dolomite in the red beds of northern Texas. Dozier lens, Whitehorse group (Guadalupian), south side sec. 78, block 19, H. G. and N. Survey, Texas; basal Doxey shale, Quartermaster group (Guadalupian?), center sec. 63, block E-2, D. and S. F. R.R. Survey, Texas. The shells are separated, and somewhat macerated. Cross-lamination of the enclosing sediments is coromon, suggesting that the waters at the site of burial were strongly agitated.

Genus SELENIMYALINA Newell, n. gen.

Genotype, Myalina meliniformis Meek and Worthen. Geological range of the genotype, upper Desmoinesian to Missourian or Virgilian?

Shell strongly prosocline to acline, nearly equivalve, the right valve being almost imperceptibly less convex than the left, apparently not discordant; both valves smooth, lacking any trace of an anterior lobe in adults, and the beaks are neither extended nor acuminate but form a well defined acute angle at the anterior extremity of the hinge, and the umbones are flattened and ill-defined; posterodorsal margin broadly rounded, never angular; dorsal broadly rounded, never straight; ventral margin semicircular; ligament area traversed by exceedingly fine and numerous opisthodetic ligament furrows; dentition similar to Myalina, except that the single cardinal tooth of the right valve and the single triangular socket of the left valve are situated on a sort of hinge plate which partly covers a very shallow umbonal recess; musculature similar to Myalina (text fig. 6B), but differs in having a cordate posterior adductor impression; outer ostracum alike in the two valves, consisting of irregular polygonal prisms of calcite 15 to 20 microns in diameter and more than 100 microns in length; inner ostracum relatively thick and apparently like that of Myalina.

Geological range—Desmoinesian-Virgilian.

Remarks—The group of shells placed under Selenimyalina is so different from other Myalinidae that its probable relationships are not at all clear. Selenimyalina is closely comparable with the Devonian Mytilarca Hall. Although similar forms are not known from the Mississippian rocks, it seem probable that the Late Paleozoic genus was derived from the Devonian form. The shell microstructure and musculature is not known in Mytilarca, but there seems to be very close agreement between the two genera in form and characters of the ligament area.

The only obvious difference lies in characters of the dentition. As illustrated by Hall (1884, pls. 31-33), Mytilarca is characterized by several cardinal teeth and sockets, as well as an equivalent number of posterior lateral teeth and sockets. Selenimyalina possesses only one cardinal tooth and socket, and seemingly has no posterior laterals. Mytilarca is commonly classed with the Ambonychiidae, and its subgenus Plethomytilus certainly looks more like the ambonychiids than any of the Myalinidae. Unfortunately, none of my specimens of Selenimyalina is sufficiently well preserved to show the form at early ontogenetic stages. In the material at hand there is no trace of an anterior lobe, which is so characteristic of at least the young of Myalinidae. Considering the scanty evidence now available, it is quite possible that Selenimyalina was derived from a stock different from the Myalinidae and that its resemblance to that family is superficial.

Species of this genus are not yet known in the Pennsylvanian below upper Desmoinesian beds, nor above the lower Virgilian. Permian representatives are unknown. A great host of Lower Triassic myalinas closely resemble Selenimyalina, and close scrutiny may reveal that they should be so classed. I refer here to such species as Myalina schamarae Bittner, M. vetusta Bittner, and M. platynotus White.

Figure 19—Form series in the genus Selenimyalina n. gen. A, S. dotti, n. sp., Desmoinesian; B. S. meliniformis (Meek and Worthen), Desmoinesian-Virgilian? ; C, S. quadrata, n. sp., Missourian. All figures X 1.5. [Image scaled and magnifications adjusted for web presentation.]

Form series in the genus Selenimyalina n. gen.

Nearly all of the specimens of Selenimyalina that I have seen retain both valves in apposition, and all were collected from argillaceous shales. These circumstances suggest that members of the genus favored a muddy bottom in quiet waters.

SELENIMYALINA MELINIFORMIS (Meek and Worthen)

Plate 2, figures 1, 2

Myalina meliniformis Meek and Worthen, 1866, Illinois GeoL Survey, vol. 2, p. 343, pL 27, fig. 3.

Shell relatively small, greatest linear dimension less than 30 mm, moderately prosocline, with a slightly concave anterior margin; dorsal, posterior, and ventral margins almost uniformly convex, without angulations; umbonal ridge poorly defined, beaks small and short; hinge characters unknown; outer ostracum in both valves consists of slender, polygonal prisms of calcite 15 to 20 microns in diameter, or possibly larger in some individuals; inner ostracum composed of laminated calcite, probably originally aragonite; musculature as in text figure 6B. Measurements for three specimens are as follows: [Note: The holotype and topoparatyps are internal molds; therefore. the measurements are not so reliable as for the hypotype, which is testiferous.]

Measurements of Selenimyalina meliniformis
  Length
(mm)
Height
(mm)
Greatest
dimension
Angle
α
Holotype 20 21 26 62°
Topoparatype 23 23 28 65°
Hypotype 20 22 26 82°

Comparison—This species differs from the similar Selenimyalina dotti, n. sp., in having a shorter shell and less oblique form.

Material—Seemingly, this species is quite rare. Only three specimens were available for this study, all of them bivalved. Two primary types, University of Illinois X-210, are internal molds composed of limonite; the hypotype, U.S. National Museum no. 51146, is a relatively well preserved, testiferous specimen.

Occurrence—The types were collected from a ferruginous shale on Saline Creek, Gallatin county, Illinois, from a horizon correlated roughly with no. 11 coal of western Kentucky (Marmaton equivalent). The single hypotype was found at an unknown horizon in Gentry county, Missouri. According to the geological map of Missouri, the Pennsylvanian rocks exposed in that county range from the Missourian series into lower Virgilian. The preservation of this specimen indicates that it was collected from a shale bed. The retention of both valves in all of the specimens at hand suggests a quiet water environment at the site of burial.

SELENIMYALINA DOTTI Newell, n. sp.

Plate 2, figures 7, 8

Shell of moderate size, greatest linear dimension being less than 40 mm, markedly prosocline, with an angle α around 65° to 70°; anterior margin markedly concave in mature individuals, nearly straight in young ones; dorsal, posterior, and ventral margins smoothly convex, with only a trace of an angulation at the posterodorsal extremity; umbonal ridge broad and poorly defined, beaks rather slender in mature specimens, and extended well forward; general shell form relatively slender and elongated along the anterodorsal-posteroventral axis as compared with other species of the genus; characters of musculature, shell microstructure, and hinge unknown. Measurements for two type specimens are as follows:

Measurements of Selenimyalina dotti
  Length
(mm)
Height
(mm)
Greatest
dimension
Angle
α
Holotype 31 30 37 68°
Topoparatype 24 21 29 65°

Comparison—This form seems to be more primitive than Selenimyalina meliniformis and S. quadrata in the marked obliquity and resembles somewhat the early growth stages in the last two species. S. dotti is distinctly more slender than the other two selenimyalinas.

Material—The new species is based on two well preserved bivalved specimens from a single horizon. They are nos. 7560 (holotype) and 7561 at the University of Kansas.

Occurrence—Middle part of the Holdenville formation (upper Marmaton equivalent), above a heavy sandstone, center south line sec. 14, T. 7 N., R. 8 E., Oklahoma. The specimens were found in a soft argillaceous shale, associated with a well preserved fauna of pelecypods and brachiopods.

SELENIMYALINA QUADRATA Newell, n. sp.

Plate 2, figures 3a, b

Shell acline, subquadrate, seemingly a little higher than long, with a broadly concave anterior margin, and convex dorsal, posterior, and ventral margins; intersection of anterior and dorsal margins forms nearly a right angle; musculature unknown, features of shell structure and hinge as in other species of Selenimyalina. The only known specimen is incomplete. Estimates of the original dimensions are as follows: Length, 19 mm; height, 25 mm; greatest dimension, 25 mm; angle α, 90° (more or less).

Comparison—In general form this species recalls Myalina (Orthomyalina) slocomi, particularly the dwarf variety M. (O.) slocomi parvula. However, the resemblances are only superficial. Left valves of Orthomyalina are rugose, whereas those of Selenimyalina are smooth; the character of the ligament area and dentition are different in the two groups, a hinge plate being absent in Orthomyalina.

Material—Based on a single, fragmentary left valve, which, however, is particularly well preserved, showing the hinge characters and sufficiently distinct growth-lines to permit reconstruction of the probable adult form. The holotype is no. 7562, at the University of Kansas.

Occurrence—Argillaceous shale, Vilas formation (Lansing), brick plant of Kansas State Prison, Lansing, Kan.

Genus SEPTIMYALINA Newell, n. gen.

Genotype, Myalina perattenuata Meek. Geologic range of genotype, Lampasan-Virgilian, Pennsylvanian.

Prosocline to acline, thick-shelled Myalinidae, with slender, extended prosogyre beaks, and an obsolescent anterior lobe just under each beak; both valves smooth, or rugose, in the latter instance the surface of the left valve being markedly more rugose than that of the right; inequivalve, the right valve being slightly flatter and smaller around the ventral periphery than the left; umbonal cavity extending far into the umbones, and partly floored over by large or small umbonal deck or septum, which joins the anterior part of the hinge axis with the anterior margin of the shell; septum in majority of species having no obvious function other than the mechanical strength that it imparts to the shell; dentition similar to that of Myalina, consisting of a functional tooth at the anterior end of the hinge of the right valve, which fits into a pivotal depression at a similar position in the left valve, the trace of both tooth and socket in the growth lamellae being marked by narrow ridge extending from the apex of the umbonal deck to the beak; ligament and shell microstructure as in Myalina; musculature similar to Myalina (text fig. 6C) except that the anterior adductor has been completely lost, at least in the species of Pennsylvanian and Permian age. [Note: The umbonal deck shows a small but distinct anterior adductor impression in S. redesdalensis (Hind), from the British Lower Carboniferous.]

Geologic range—Viséan to Wolfcampian.

Remarks—The mention of an umbonal septum in the original diagnosis of the genus Myalina by De Koninck shows that this structure was known to him. Unfortunately, the genotype of Myalina does not possess the umbonal septum, and De Koninck evidently drew his diagnosis from several Myalinidae having diverse structures.

Figure 20—Shell form in species of Septimyalina. A, S. orthonota (Mather), X 1, from the Morrowan series; B, S. perattenuata (Meek and Hayden) (Myalina kansasensis of authors), X 1/2, Atoka to Wabaunsee; C, S. burmai, n. sp., X 1/2, from beds of Wolfcampian age; D, S. orbiculata, n. sp., X 1/2, from Marmaton beds; E, S. scitula, n. sp. (Myalina perattenuata of some authors) X 1/2, from the Wabaunsee group; F, S. sinuosa (Morningstar), X 1/2, an auriculate species characteristic of beds of Desmoinesian age.

Shell form in species of Septimyalina.

The ontogeny of species of Septimyalina indicates that the genus was derived from typical Myalina, and the evolutionary trends, arrived at independently, parallel those of Myalina, s. s., to a surprising degree. The majority of species of Septimyalina, even the more specialized ones, retain the primitive obliquity, but the tendency to develop a posterior auricle has produced several species that bear a marked homeomorphic resemblance to species of Myalina. For instance, the Chester Septimyalina angulata (Meek and Worthen), the British Lower Carboniferous S. redesdalensis (Hind), and the Desmoinesian S. sinuosa (Morningstar ) are strongly reminiscent of the Wolfcampian species Myalina (Myalina) copei.

Although the more completely known species of Septimyalina herein described are unquestionably monomyarian, a typical Septimyalina from the British Lower Carboniferous, S. redesdalensis (Hind) (1897, p. 109), shows a small but distinct anterior adductor pit near the center of the umbonal deck. Much of my material from the Pennsylvanian rocks is so perfectly preserved that the faintest muscle impressions are perfectly shown; and, furthermore, the muscle marks, including the pallial line, are pigmented so that there is no reasonable possibility that I have overlooked the anterior adductor pit. The trace of the pivotal pit, or socket of the left valve in Myalina, is a well defined furrow extending to the beak. Except at its distal end this furrow is not functional, because it does not come in contact with the corresponding ridge of the opposite valve. The left valve of Septimyalina differs in that a well defined ridge occupies the position of the dental furrow of other Myalinidae. The trace of the pivotal pit in the growth lamellae produces a ridge instead of a furrow.

It is interesting to note that the auriculate septimyalinas seem to be confined to Upper Mississippian equivalents and to the Lower Pennsylvanian, being unknown above Desmoinesian beds. The representatives of the genus ranging into the Upper Pennsylvanian and Lower Permian are decidedly more primitive in general aspect. There is nothing extraordinary about this situation, however, because the more primitive animal stocks in the geologic record commonly are comparatively long ranging.

Much of the confusion over recognition of species in this group of shells in the past is seemingly due to the fact that internal molds commonly give a very erroneous impression of the original shell form. The shells are relatively thick as compared with the visceral space. Noteworthy is the tendency for the space occupied by the living animal to decrease in volume in the gerontic stage (pl. 13), as in oysters. Unlike most myalinas, the mantle withdrew from the ventral margin in old age.

It is not easy to obtain a general idea of the range of habitats of members of the genus. Species of Septimyalina are found in a wide range of rock types—sandstone, various types of limestone, and shales. Some of the shells are bivalved, indicating burial in quiet waters, whereas specimens of some species, particularly S. perattenuata, are common in shell breccias, which indicate burial in the zone of active wave action. Of course, it is possible that the broken and fragmentary shells were concentrated and removed by waves and currents from the life environment. In any event, it seems that the environmental range of species of Septimyalina was somewhat greater than for the majority of other myalinids. Although shells of Septimyalina are generally found associated with other myalinas, they also occur in places where other representatives of the family are absent.

SEPTIMYALINA PERATTENUATA Meek and Hayden

Plate 12, figures 3-12

Myalina perattenuata Meek and Hayden, 1858, Albany Inst., Trans., vol. 4, p. 77; —— 1859, Philadelphia Acad. Nat. Sci., Proc., p. 28; ——, 1864, Paleontology Upper Missouri, p. 32, pl. 1, figs. 12 a-b.
?Myalina kansasensis Shumard (in Shumard and Swallow), 1858, St. Louis Acad. Sci., Trans., vol. 1, p. 213.

Prosocline slender thick-shelled forms with extended beaks and prominent umbones; hinge margin straight, forming a broadly obtuse angle with the straight or slightly curved posterior margin; ventral margin rather narrowly rounded and linguiform, anterior margin strongly concave above and generally slightly convex below; left valve covered by rugose, more or less regular projecting lamellae, which in well preserved adults develop hemicylindrical prone spines that project from the shell margin; right valve marked only by non-projecting, somewhat irregular growth lamellae; right outer ostracum composed of polygonal' calcite prisms ranging in diameter between about 7 and 15 microns, left outer ostracum composed of "fibrous' calcite in which individual fibers are of the order of 1 micron in diameter, alignment in both instances being normal to the shell surface; inner ostracum finely lamellar, probably originally aragonite; ligament area as in other Myalinidae, except that the furrows are very fine, 15 to 20 occupying a space of 5 mm; internal characters as indicated in the generic diagnosis. Measurements are indicated in the following tables:

Measurements of Septimyalina perattenuata
Thirty specimens from the Stull shale, Shawnee group, Virgilian
  Length
(mm)
Height
(mm)
Angle
α
Angle
β
Maximum 53 60 74° 139°
Average 44 49 63° 125°
Minimum 35 41 51° 114°
Thirty-two specimens from the Vinland shale, Douglas group, Virgilian
  Length
(mm)
Height
(mm)
Angle
α
Angle
β
Maximum 60 71 75° 140°
Average 49 51 62° 122°
Minimum 39 40 53° 110°
Thirty-one specimens from the Missourian series1
  Length
(mm)
Height
(mm)
Angle
α
Angle
β
Maximum 54 58 78° 133°
Average 41 (16) 42 (17) 59° (31) 121° (12)
Minimum 28 30 47° 105°
1. Various horizons. Numbers in parentheses indicate the number of specimens on which the significant measurements could be taken.

Comparison—From the two most similar species, Septimyalina orthonota and S. burmai, this form differs in its obliquity, being almost intermediate between the very oblique S. orthonota and the upright S. burmai. It is probable that the three species form a genetic series, inasmuch as S. orthonota is geologically the oldest as well as the most primitive, and S. burmai is the youngest and most specialized. S. scitula and S. orbiculata are relatively less rugose and shaped differently. S. orthonota is rather poorly known, however, and may represent the young of S. perattenuata.

Material—The above description was based on more than one hundred instructive specimens, many of them perfectly preserved, from many horizons and localities in the Mid-Continent Pennsylvanian. The majority of these specimens are at the University of Kansas.

Occurrence—This species is one of the most ubiquitous of the myalinas. It occurs, associated with various other myalinas, throughout the Pennsylvanian, and in some places, is the only representative of the family. The species is sparingly represented in a number of collections from the Atoka and Desmoinesian equivalents, and it becomes particularly abundant in the later Pennsylvanian, being found in greater or less abundance in every molluscan facies of the Missourian and Virgilian series.

Shells of this species are found in a wide range of lithologic types, indicating life under variety of ecologic conditions. Specimens are rarely found, however, in carbonaceous rocks. Bivalved specimens are less coromon than separated valves, and the common occurrence of the species in brecciated limestones and relatively coarse sandstones suggests that the animal thrived in agitated waters near the zone of effective wave action.

SEPTIMYALINA ORTHONOTA (Mather)

Plate 13, figures 1, 2

Myalina orthonota MATHER, 1915, Denison Univ. Bull., Jour. Sci. Labs., vol. 18, p. 221, pl. 15, fig. 4.

All of the shells in the type collection are relatively small, the largest being only slightly more than one inch in the greatest dimension; thin-shelled; form slender, subtriangular, hinge line relatively short, anterior and posterior margins relatively long and subparallel; anteroventrallobe inconspicuous or absent; beaks slender but not markedlyextended; angle α relatively small, about 45°; angle β broadly obtuse, about 145°; umbonal ridge almost straight except in the beaks, broadly rounded, becoming rather obscure below the middle of the shell; surface of left valve coarsely lamellose, right valve unknown; hinge, musculature, and shell microstructure unknown.

Measurements of two specimens are as follows: (1) holotype, a juvenile left valve; length, 13 mm; height, 13 mm; hinge length, ±9 mm; convexity, ±3.6 mm; greatest dimension, ±17 mm; angle α, ±45° angle β, ±130°; (2) paratype, a larger left valve, collected with the holotype; length, 23 mm; height, 18 mm; hinge length, 14 mm; convexity, 6 mm; greatest dimension, 27 mm; angle α, ±45°; angle β, ±145°.

Comparison—The specimens available to me do not permit a wholly satisfactory evaluation of this species. It is not certain that the specimens are fully mature, nor does the preservation allow detailed comparison with juveniles of similar but larger species in my collections. The general aspect recalls the form described above as Septimyalina perattenuata, and the similarity is heightened by the rugosity of the surface of left valves. This species may simply have been founded on the young of S. perattenuata. Additional topotype material is needed for a final judgment.

Material—Four of Mather's types from the University of Chicago are before me. They are preserved in dark gray, crystalline limestone. Only the two specimens mentioned above are sufficiently complete to exhibit pertinent characters. The holotype is no. 16045 at Walker Museum, University of Chicago. The three paratypes bear the catalogue number 16583.

Occurrence—The types were collected from the Hale formation (Morrow), East Mountain, Fayetteville, Ark. Other occurrences are unknown.

SEPTIMYALINA BURMAI Newell, n. sp.

Plate 12, figures 1-6

?Myalina apachesi Marcou, 1858, Geol. North America, p. 44, pl. 7, figs. 6, 6a.
Myalina apachesi Girty, 1909, U.S. GeoL Survey, Bull. 389, p. 81, pl. 9, figs. 6, 7.

Shell rather robust, about as long as high in adults, somewhat prosocline, but less markedly so than several species of the genus; both valves marked by prominent, regular lamellae of growth, which tend to become projected in short, hemicylindrical spines; anterior margin only slightly concave, nearly parallel with the broadly convex posterior margin; ventral margin broadly convex, dorsal margin straight; particularly characterized by a relatively large angle α, and a relatively great shell length. Dimensions of the holotype are as follows: length, 48 mm; height (restored slightly), 46 mm; greatest dimension (restored slightly), 57 mm; angle α, 70°; angle β, 122°.

Comparison—This species is similar to Septimyalina perattenuata, but adults differ from the latter in being less slender in form and in having a distinctly larger angle α. At a given shell size the hinge margin is considerably longer in S. burmai than in S. perattenuata. Juveniles of the two species are almost indistinguishable. In fact, ontogenetic stages in S. burmai simulate closely the adults of S. orthonota (pl. 12 figs. 2 and 5) and S. perattenuata (pl. 12, fig. 4). Specimens of S. burmai less than 35 mm in height are likely to be confused with S. perattenuata. It is probable that the three species just named form a natural gens, for they appear successively in the geologic column in the same order in which their form equivalents are introduced in the ontogeny of the youngest species, S. burmai. The remaining species of Septimyalina are sufficiently distinct that they probably belong in separate tribes of the genus.

It seems that Girty was not justified in reviving Marcou's Myalina apachesi for this species. The species is not recognizable from Marcou's description and illustrations. The types are lost, and the exact horizon and locality are not known.

Material—This species is abundantly represented in my collections from rocks of Wolfcampian age. The overwhelming majority of specimens are so poorly preserved, however, as to supply no basis for a statistical analysis. Most of the material is in the form of internal molds, which are grossly misleading. The size and shape of the visceral cavity of the shell is much more variable than the shell itself. Inasmuch as this is a thick-shelled species, the shell form can be determined only from shells or external molds. The specimen selected for the holotype (U.S. Geol. Survey, no. 2497) is characteristic of the species and is exceptional only in its unusually good preservation.

Occurrence—Characteristic of almost every molluscan facies in the Council Grove, Chase, and Sumner groups, Lower Permian, in the northern Mid-Continent region. The species has been recognized in the Burr limestone, Cottonwood limestone, Florena shale, and Eiss limestone, of the Council Grove group; the Oketo shale, Chase group; the Krider and Herington limestones, Sumner group. Girty found the species in the Abo sandstone of New Mexico (Wolfcampian), about 50 feet above the base of the red beds at Abo canyon.

SEPTIMYALINA ORBICULATA Newell, n. sp.

Plate 11, figures 1-3

Shell robust, massive; rear margin of shell broadly rounded, so that in some instances there is no distinct angulation at the posterodorsal margin; anterior margin broadly and distinctly concave; beaks very slender, extended far forward; ligament furrows relatively coarse, as compared with other septimyalinas, eight furrows lying in a space of about 5 mm; right valve smooth, left valve slightly rugose. Measurements of the holotype are as follows: length, 57 mm; height (somewhat restored), 46 mm; greatest dimension (somewhat restored), 64 mm; angle α, 65°; posterodorsal extremity rounded. The angle α in three other topoparatypes measures 60°, 58°, and 64°, respectively.

Comparison—This species needs comparison only with Septimyalina scitula. Undoubtedly the two are closely related, and it may be that the former lies directly in the lineage of S. scitula. The latter species, at equivalent size, is distinctly less oblique and less massive. Furthermore, the surface of both valves in S. scitula is about equally smooth.

Material—The species is based on well preserved but very fragmentary material, a half dozen specimens, from a single locality. The holotype (no. 7584) and topoparatypes are at the University of Kansas.

Occurrence—Shaly limestones at about the horizon of the Lenapah limestone ("Eleventh Street limestone"), upper Marmaton group, near Beggs, NW sec. 19, T. 15 N., R. 12 E., Oklahoma. The specimens occur both in the limestone and in calcareous shale.

SEPTIMYALINA SCITULA Newell, n. sp.

Plate 11, figures 4-14

Myalina perattenuata of authors, in part.

Shell relatively small, upright, almost acline, surface of both valves nearly smooth; beaks slender, extended forward; anterior margin markedly concave in adults, subparallel to convex rear margin, ventral margin relatively narrow, rounded, dorsal margin straight, intersecting rear margin at a well defined obtuse angle of about 1000 in most specimens; umbonal deck relatively broad, ligament grooves relatively fine, comparable with those of S. perattenuata. Measurements, based on six specimens, are as follows:

Measurements of Septimyalina scitula
  Length
(mm)
Height
(mm)
Angle
α
Maximum 35 37 77°
Average 20 20 60°
Minimum 14 13 54°
Holotype 23 21 55°

Comparison—Juveniles of Septimyalina scitula are very similar to S. orbiculata in general form; although they tend to have a straighter anterior margin; and, of course, they are smaller and less massive than adults of the latter species. More confusing is the similarity between some specimens of S. scitula and Myalina (Myalinella) meeki. In some instances poorly preserved material can not be differentiated. In addition to the presence of an umbonal deck in the former and its absence in the latter, there are minor differences in form between the two species. The majority of mature specimens of S. scitula exhibit a concave front profile. The anterior margin in M. (M.) meeki generally is nearly straight. There is a tendency for the beaks in S. scitula to be very slightly more acuminate than in M. (M.) meeki. Large (fully mature) specimens of S. scitula are distinctly more upright in form than mature individuals of M. (M.) meeki.

Material—The holotype and several topoparatypes are located at the University of Kansas, nos. 7587-7592; other paratypes are at the National Museum, no. 7035.

The material on which the species is based is well preserved, but unfortunately the majority of specimens are not fully mature.

Occurrence—Undoubted representatives of this species are recognized only in Virgilian horizons. The types were collected from the Iowa Point shale (Shawnee) at the NW sec. 1, T. 21 S. R. 13 E., Coffey county, Kansas. Other specimens are from unspecified Wabaunsee beds in northeastern Kansas. Fragmentary and otherwise unsatisfactory material, about which the identification is not positive, comes from Wolfcampian strata in northern Texas and in Kansas. Specimens have been found in separated condition in sandy limestone, and bivalved pyritized specimens have been collected from carbonaceous shales, associated with coal. Too little is known about field occurrences, in general, to warrant drawing conclusions about the environments in which S. scitula lived.

SEPTIMYALINA QUADRANGULARIS Newell, n. sp.

Plate 12, figures 7-9

Shell relatively upright, almost acline in adults, subtriangular, with the narrow ventral margin forming the apex of an inverted triangle; anterior margin only slightly concave; posterior margin slightly convex, hinge margin straight; septum relatively small, ligament furrows fine, as in Septimyalina perattenuata, surface of both valves relatively smooth. Measurements of two type specimens are as follows.

Measurements of Septimyalina quadrangularis
  Length
(mm)
Height
(mm)
Angle
α
Angle
β
Holotype 38 49 72° 75°
Topoparatype 21 25 72° 87°

Comparison—In a general way, this species recalls Septimyalina burmai and S. scitula in the erectness of the shell. From both it differs in the narrowness of the ventral margin and the relatively small value of the angle β. The relative smoothness of both valves removes any likelihood of confusion with S. burmai. S. quadrangularis is relatively specialized, as shown by its upright form. The ontogeny, as revealed by growth lines, suggests derivation from a shell shaped like S. perattenuata, which, however, is quite rugose, unlike the present form, and may not lie directly in the lineage of S. quadrangularis.

Material—This is a very rare species. Only three specimens occur in my collection, all of them from a single locality. They are preserved as molds in very fine, calcareous sandstone. The holotype, no. 21194, and two topoparatypes, nos. 21195, 21196, are at the University of Wisconsin.

Occurrence—Savanna sandstone (middle Desmoinesian), near Pittsburg, Okla. The valves are separated, indicating deposition in agitated waters.

SEPTIMYALINA SINUOSA (Morningstar)

Plate 12, figures 10-13

Myalina recurvirostris var. sinuosa Morningstar, 1922, Ohio Geol. Survey, Bull. 25, p. 219, pl. 11, figs. 10-12.

Shell of moderate size for the genus, and relatively massive; mature form su brhombic except for the posterodorsal extremity which is extended in a prominent but slender auricle; beaks very slender and extended anterodorsally above the hinge, virtually without lobation; beaks markedly prosogyre, the left one being somewhat more conspicuous and rising dorsally above the right beak; umbonal ridge relatively straight below the region of the umbones, so that the anterior margin of the shell also is nearly straight; angle α ranges from about 53° to 58°; angle β acute, ranging from about 40° to 50°; valves apparently somewhat discordant, the right valve being slightly smaller and less convex than the left; both valves marked by relatively uniformly spaced edges of growth lamellae, which are somewhat rugose in the left valve, less so in the right, and distributed about four in the space of 10 mm along the umbonal ridge, at a shell height of 50 mm or more; ligament area relatively broad, measuring up to 7 mm in one mature individual in which there are at least 20 very fine ligament grooves. These grooves descend to the hinge axis at a low angle so that at least six intersect the axis in one incomplete specimen within a horizontal distance of about 36 mm; dentition in right valve unknown, that in the left valve consisting of a shallow pivotal socket occurring at the posterior end of a narrow ridge, which extends downward and backward from the beak at an angle to the hinge axis of about 16°; umbonal deck or septum rather massive in left valve lying nearly in a plane with the ligament area, that of right valve not observed; outer ostracum in right valve composed of short irregularly polygonal prisms of calcite about 15 to 20 microns in diameter, that of the left valve consisting of fibrous calcite with fibers about 1 micron in diameter normal to the shell surface; inner ostracum of both valves lamellar, probably originally, aragonite; musculature unknown.

Measurements of specimens of Septimyalina sinuosa, in millimeters
  Lectotype
(no. 15273)
Topoparatype
(no. 15272)
Topoparatype
(no. 9819)
Length 51    
Height 45   60±
Convexity 14 16 20
Hinge length 51    
Greatest dimension 54   78±
Angle α 53° 55° 58°
Angle β 30°-40° 35°  

Comparison—The parallelism between this species and Myalina (Myalina) copei is very striking. Critical inspection, however, reveals that the resemblance is superficial and largely produced by the possession of a rear auricle in both species. M. (M.) copei possesses a conspicuous anterior lobe, and the septum of Septimyalina is lacking. S. angulata (Meek and Worthen) (1866, p. 300, pl. 23, figs. 7a, b) from the Chesterian series, near Chester, Illinois, is a closely related form but has a smaller angle α and less prominent auricle, as might be expected of an older and more primitive species. The British species S. redesdalensis (Hind, 1897, p. 109, pl. 3, figs. 1-5,7-13; pl. 5, fig. 1; pl. 7, figs. 1, 1a) from the Lower Carboniferous, is another auriculate Septimyalina, which also has a smaller angle α and less extended rear auricle than the present form. Judging from the relatively great value of the angle α and the extended auricle, S. sinuosa is the most highly specialized species within the genus. It very likely is one of the terminal members of a series of septimyalinas, which, during the Late Mississippian and Early Pennsylvanian, progressively increased the length of the hinge by development of the rear auricle.

Material—The most instructive representatives of this species at my disposal are part of the original suite on which the species was based. There are five specimens from the University of Ohio, three good left valves and two crushed specimens retaining both valves in opposition. They bear the catalogue numbers 15273 (designated here the lectotype), 15272, 9819, 4621. The shell material is well preserved and occurs in a calcareous gray shale or argillaceous limestone. Other fragmentary specimens occur in argillaceous shale.

Occurrence—Lower Mercer limestone (upper Pottsville), northern part of Madison Township, Muskingum county, Ohio, one fourth mile south of the county line near Adams Mills (the types); questionably from the upper Pottsville or lower Allegheny equivalents on Otter Creek, near Brazil, Ind.; Labette shale (Marmaton), St. Louis, Mo.; from an unknown horizon in the McCoy formation (Desmoinesian), near Bond, Colo.

Genus NAIADITES Dawson, 1860

Naiadites Dawson, 1860, Acadian Geology, Supplement, 1st edition, p. 43; Dawson, 1894, Quart. Jour. Geol. Soc. London, vol. 50, pp. 435-437; Hind, ibid., pp. 437-442; Hind, 1895, Palaeontographical Soc., vol. 49, pp. 126-131; Davies and TRueman, 1927, Quart. Jour. GeoL Soc. London, vol. 83, pp. 236-237; Dix and TRueman, 1930, The Naturalist, pp. 15-18; Newell, 1940, Am. Jour. Sci., vol. 238, pp. 292-295.

Genolectotype, Naiadites carbonarius Dawson, 1860, designated by Hind, 1894, Quart. Jour. Geol. Soc. London, vol. 50, p. 439.

Modioliform shells having the beaks situated distinctly behind the anterior terminus of the hinge; surface devoid of radial ornamentation; hinge without well defined teeth, ligament areas external, as in other Myalinidae, but slightly amphidetic, with relatively fine ligament furrows; anisomyarian, with two relatively small pedal scars located in the umbonal recess just behind and above the larger anterior adductor; pallial line differentiated into a series of separate pits, at least in the anterior part; posterior adductor probably as in other Myalinidae (text figure 6F); shell microstructure alike in the two valves, consisting of a relatively thick inner lamellar ostracum and thin outer ostracum of fine (about 30 microns in diameter), short, irregularly polygonal prisms of calcite arranged normal to the shell surface.

The left valve of Naiadites carbonarius, on which most of the above observations were made (McGill College, no. 2.1172), has the following measurements: length, 18 mm; height, 13 mm; hinge length, 15 mm; greatest dimension, 20 mm; angle of hinge with umbonal ridge, about 60°; posterodorsal angle, 125°.

The zone of Naiadites, which seems to include the entire known range of the genus, embraces the Lanarkian and Westphalian stages (Pottsville) of European Upper Carboniferous. Davies (1930) has called attention to the fact that Naiadites, together with its customary associates Anthracomya and Carbonicola, although not common, are rather widely distributed in the Pottsville and Allegheny beds of the Appalachian region. In view of the great stratigraphic interest of these forms, it is unfortunate that exhaustive collections have not been made and studied in the United States. Since I do not now have access to satisfactory collections, an extended treatment of the species of Naiadites must be deferred. At present, it seems that the genus is represented in North America only in the eastern areas, in deposits of Early Pennsylvanian age. Future studies may reveal that Naiadites is represented also in the coal-bearing deposits of the Mid-Continent region.

Remarks—Much has been written about Coal Measures pelecypods that are referred to the genera Naiadites, Anthracomya, and Carbonicola. Their remains are common in the roof shales of coal beds in various Upper Carboniferous basins of western Europe. The high order of stratigraphic usefulness of these shells has led to the publication of several outstanding monographic studies (Amalizky, 1892; Hind, 1894-1896; Davies and Trueman, 1927; Tchernyshev, 1931; and other important works) with the result that a succession of zones based on these "fresh-water" pelecypods of the late Paleozoic is recognized in all of the important coal basins of the Upper Carboniferous of Europe.

The majority of students of these shells generally are agreed that the species of Naiadites, Anthracomya, and Carbonicola were of estuarine or freshwater habit, although Tchernyshev (1931, pp. 105- 111) finds reason to regard them as near-shore marine. His observations in the Donetz basin reveal an occasional association of these pelecypods with forms usually regarded as typically marine (Lingula, Aviculopecten, Myalina, Bucania, Murchisonia, and goniatites.) The association of freshwater and marine shells in the deltaic and estuarine environments of modern coasts is too coromon to merit discussion here. It is pertinent, however, to remark that the exceptional occurrences of shells of the three pelecypod genera with undoubted marine shells does not necessarily imply that Naiadites, Anthracomya, and Carbonicola invaded the normal marine habitat during their life.

Since the publication of Hind's monograph (1894-1896), Anthracomya and Carbonicola have been referred to the Unionidae by many authors and to the Cardiniidae by others. Since these two genera are not related to the myalinas, they will not be discussed further here.

The genus Naiadites erroneously has been classed with Carbonicola and Anthracomya by many authors. Hind and his followers, however, recognized the affinities of Naiadites with the Mytilacea and classified the genus along with Myalina in the Mytilidae. Although I prefer to place these genera in the family Myalinidae. it would be ungenerous not to point out that Hind's assignment was perfectly proper in consideration of the broad limits given the family Mytilidae in his day.

While making his pioneer geological studies in Nova Scotia, Sir William Dawson found large numbers of small, distinctive pelecypods in the roof shales of coal beds of the Upper Carboniferous. Association of the mollusks with plant remains and fossil amphibians suggested to him that the pelecypods had lived in a brackish or fresh-water habitat. To these shells he gave the name Naiadites, although he later admitted that they represented more than one genus. A genotype was not designated in the original publication of Naiadites. Three years later, in 1863, Salter (Quart. Jour. Geol. Soc., vol. 19, p. 80) introduced the generic name Anthracoptera for Dawson's Naiadites carbonarius, ignoring the priority of Naiadites. Salter's genus has no valid claim for recognition, as has been pointed out by subsequent investigators, and therefore need receive no further attention here. Hind apparently was the first to designate a genotype for Naiadites, and his selection of N. carbonarius in 1894 is legally correct. In the last analysis, our concept of the genus must be determined largely by the character of N. carbonarius.

Through the kindness of Dr. T. H. Clark of McGill University, I have been able to make some pertinent observations on specimens that may be syntypes of the species. Doctor Clark supplies the following information:

We have nearly all of Dawson's material here; but, inasmuch as Dawson's illustrations were, for the most part, sketches and inasmuch as he did not mark his type specimens, it is almost impossible today to be absolutely sure which specimen he used as a type. Occasionally an imperfection reproduced in the drawing or the shape of the rock helps. From time to time various people have tried to determine Dawson's type specimens among our collections, not altogether with success, however. I am sending you four such specimens determined by various museum workers here, but I am not satisfied that Dawson must have used these specimens.

Although some doubt exists regarding the identity of Dawson's type specimens, the specimens that have been made available to me were identified by Dawson, and were collected by him from the same locality as the types, South Joggins, Nova Scotia, from beds generally correlated with some part of the Pottsville series.

Students of Pennsylvanian faunas often have remarked on the close resemblance of Naiadites to Myalina. For instance, Hind says,

The question .. arose as to whether de Koninck's Myalina might not even supersede Naiadites, ... but there still remained the fact that de Koninck affirmed a rostral plate or myophore in his description, which was, however, conspicuously absent in his figures, and more markedly so in the illustrations in pl. 29 of his later work, "Faune du Calcaire Carbonifère de la Belgique," 1885. ... I have therefore adopted the name Naiadites Dawson for the Mytiliform shells of the Coal-measures for a twofold reason. First, because they do not possess myophorial septa; and second, because they do not occur with a typically marine fauna as a rule ... (Hind, 1895, pp. 129-130.)

Had Hind understood that the type species of Myalina actually does not possess the septum postulated by De Koninck, it is highly probable that he would have relegated Naiadites to the synonymy of Myalina. As it is, I have been confronted by the problem of searching for significant features in the respective genotypes of the two genera to determine whether or not Naiadites can in practice be distinguished objectively from Myalina. With this end in view many species of Naiadites from the Lower Coal Measures of England were borrowed from the U.S. National Museum and from Yale University, and comparisons were made directly with several species of Myalina that can be regarded as being closely like the genotype. These observations indicate that Naiadites and Myalina are persistently different.

Species of Myalina, including the genotype, are characterized by the possession of an anteroventral lobe or emargination of the shell, producing a somewhat superficial resemblance to Volsella. In spite of the similarity, there is no Myalina, to my knowledge, that can truly be called volselloid, because the beaks of Myalina are located at the extreme anterior end of the cardinal margin, whereas in Volsella the beaks are situated distinctly back of the front extremity of the shell. In respect to the beaks and the anteroventrallobe, Naiadites is like Volsella, though differing from that genus, of course, in characters of musculature, ligament, and shell microstructure.

Species of Naiadites, such as N. carbonarius, N. modiolaris, and N. carinatus are alike in certain distinctive characters of the anterior muscles. Three muscle impressions occur in the umbonal cavity. The largest of these is at the terminus of the pallial line and occurs just back of the most anterior projection of the shell. It is evidently the anterior adductor impression. A pit about half as broad occurs just behind and above the anterior adductor, and, judging from its position, this represents the point of attachment of one of the pedal muscles. Behind this scar, and almost in line with the other two, occurs a still smaller pit, the superior pedal attachment.

Like Naiadites, typical species of Myalina have three anterior muscle pits, and the impressions are arranged approximately the same. However, there is a marked tendency for the three pits to be more closely spaced in Myalina, and the anterior adductor impression is relatively small, becoming obsolescent in certain species.

The exact character of the posterior adductor impression has not been determined in the type species of Naiadites, but the pallial line is like that of Myalina in being broken anteriorly into a series of small muscle pits.

The shell microstructure in Naiadites carbonarius seems to be the same in both valves, consisting of a relatively thick, lamellar inner ostracum and a thin outer ostracum composed of very small, irregularly polygonal calcite prisms arranged normal to the shell surface.

The shell form in Naiadites is similar to very primitive or very young Myalina. The assumption that Naiadites is the most primitive member of the Myalinidae seems a fair one. The conclusion is further borne out by the fact that this is the only genus that had not yet acquired a completely opisthodetic ligament; therefore, Naiadites presumably is very similar to the ancestral radicle of the family Myalinidae. Unfortunately for this theory, undoubted Naiadites is unknown below the Upper Carboniferous. It is possible that Naiadites was derived from a more distinctly amphidetic form like the Devonian Ptychodesma Hall.

Henderson (1935, pp. 60, 61) cites nine species of Naiadites in American paleontological literature. They are N. carbonarius Dawson, 1860; N. fragilis (Meek and Worthen), 1866; N. longus Dawson, 1894; N. mytiloides Dawson, 1894; N. ohioensis Morningstar, 1922; N. politus White, 1878; N. triassicus Pilsbry, 1926; N. wanneri Pilsbry, 1926. In my opinion N. fragilis, from Mississippian rocks, and the Upper Triassic species, N. wanneri, and N. triassicus, should be referred to other genera. The following specific descriptions are limited to the two species—N. carbonarius and N. ohioensis—of which I have seen typical material.

Naiadites CARBONARIUS Dawson

Plate 15, figures 2a-b

Naiadites carbonarius Dawson, 1860, Acadian Geology, Supplement, 1st edition, p. 204. ——, Newell, 1940, Am. Jour. Sci., vol. 238, pp. 292-295, text fig. 1b, pl. 2, figs. 4a-c.

Shell volselloid, the beaks located about onesixth of the shell length behind the anterior extremity of the shell; relatively small, having in a typical specimen a length of 18 mm, height, 13 mm; hinge length, 15 mm; greatest dimension, 20 mm; umbonal angle, about 60°; posterodorsal angle, about 125°; a well defined sulcus marks off a prominent anterior lobe of the shell, but the posterodorsal margin of the umbonal ridge is only obscurely delimited by a very broad sulcus, which corresponds in position with nearly all of the posterodorsal part of the shell; ligament area relatively narrow, about 0.5 mm in breadth at the widest place under the beaks, and traversed by seven or eight very fine ligament grooves in the specimen at hand; dentition seemingly lacking, nor is there any trace of an umbonal septum such as is found in some genera of Myalinidae; the musculature is not wholly determinable from material at hand, but so far as shown is quite distinctive as compared with species of other known genera, the anterior adductor impression being relatively large and deep, situated directly below the beak in the anterior lobe of the shell; anterior byssal retractor impressions relatively small, being located slightly behind and above the adductor; pallial line and posterior musculature unknown; shell microstructure identical in both valves, consisting of a relatively thick, lamellar ostracum of aragonite, and a thin outer ostracum of small (about 30 microns in diameter), irregularly polygonal calcite prisms arranged normal to the shell surface.

Comparison—Typical representatives of this species have been directly compared with Naiadites ohioensis Morningstar and found to be very similar. The two can be distinguished, however, by differences in the posterior profile and by the fact that the umbonal ridge in the Ohio species is more sharply delimined by rather narrow, well-defined sulci; probably also the Ohio form is relatively more ventricose. This species is very like the European N. modiolaris (Sowerby), and I am by no means certain that they are specifically different. Statistical studies on adequate collections of both species will be necessary to solve this problem.

Material—The above observations were made on several fragmentary specimens and one nearly complete left valve reposing in the McGill College collection, no. 2.1172. This collection is thought to be part of Dawson's original suite of specimens and was collected in the "Lower Coal Measures," probably upper Pottsville, of South Joggins, Nova Scotia, where the species is said to be very abundant.

Naiadites OHIOENSIS Morningstar

Plate 14, figures 5a-6b

Naiadites ohioense MORNINGSTAR, 1922, Ohio Geol. Survey, Bull. 25, p. 221, pl. 12, figs. 10-13.

Shell rather fragile, relatively small for the genus, volselloid, the beaks located about one-sixth of the shell length behind the front end of the shell; umbonal ridge delimited on both sides by a rather narrow and well defined sulcus, the intersection of the posterodorsal sulcus with the rear margin of the shell being marked by a shallow notch or sinus in the profile; shell very ventricose, the two valves in apposition being only slightly less thick than the shell height; posterior end somewhat truncated in appearance so as to produce a roughly subquadrate posterodorsal angle; right valve very slightly less convex than the left but apparently not discordant; musculature unknown, except for the anterior adductor which is relatively large for the genus, covering of the inner surface of the anterior lobe of the shell.

Measurements of specimens of Naiadites ohioensis, in millimeters
  Lectotype
(no. 15281)
Topoparatype
(no. 15279)
Topoparatype
(no. 15280)
Length 11.5 13.0 12.5
Height 8.0 8.0 9.0
Thickness 6.0 7.0 6.5
Hinge length 10.0 10± 10
Greatest dimension 12.0 12.5 12±
Angle α 48°± 45° 45°±
Angle β 65-75° 70-87°  

Comparison—Although similar to Naiadites carbonarius in lateral profile, this species is significantly smaller, has smaller angle α and β, and it is especially distinguished by a slight indentation of the posterior margin corresponding to the location of a narrow sulcus that borders the umbonal ridge posterodorsally. The marked obesity of the shell seems to be distinctive.

Morningstar (1922, p. 221) has stated that the shell gapes in this species along the ventral and posterior margins, but examination of the types leads me to the conclusion that there is no appreciable gape. The two valves are very slightly and about equally rugose, but this characteristic does not seem to distinguish the species from other Naiadites.

Material—The above observations were made on the type collection, lectotype 15281, and topoparatypes 15279 and 15280, University of Ohio. The specimens are rather well preserved and excepting no. 15279, which is an internal mold, retain the original shell. The material was collected in nodules of "iron ore" and seems to have a matrix of clay ironstone.

Occurrence—Morningstar states that the shells were found associated with Anthracomya elongata (Dawson) but no other fossils were present. The fossils were found in concretions that lie in the shale above the Sharon coal in the roof of the John Alexander mine near the head of Higgins Run, northern part of sec. 15, Madison Township, Scioto county, Ohio. This horizon occurs low in the Ohio Pottsville and is correlated by Wanless (1939, p. 64) with some part of the Morrowan series of Oklahoma and Arkansas.

Genus LIEBEA Waagen, 1881

Liebea Waagen, 1881, Palaeontologia Indica, ser.13, pp. 292- 295; Newell, 1939, Jour. Paleontology, vol. 13, pp. 589- 593, pl. 68.

Genolectotype, Liebea squamosa (J. de C. Sowerby), 1829, designated by Cox, 1936, Geol. Soc. London, Quart. Jour., vol. 92, pp. 38-39.

Shell small, rarely attaining a length greater than 25 mm, mytiloid, subrhombic, with arched dorsal margin; anterior margin lacking anterior lobation, except in the earliest juvenile stages; relatively biconvex, subequivalve, with right valve almost imperceptibly less convex than the left valve; surface of right valve relatively smooth, left valve more or less conspicuously lamellose; anTopopamtype terior marsrin of both valves crenulated below beaks and in front of angular ridge, which bounds front margin of ligament area, crenulations consisting of one crest and one sinus each, the crest of the left valve being above the sinus and that of the right valve being placed below a sinus, crests and sinuses interlocking to produce a sort of "dysodont" dentition seen in some species of Mytilus; hinge like that of Septimyalina otherwise; anisomyarian, with anterior adductor lodged in a deep pit along the anteroventral edge of a small umbonal septum, or deck, which walls off a rather deep umbonal cavity; posterior adductor indistinct, seemingly located rather low under posterior extremity of hinge; position of other muscles uncertain, but probably similar to Septimyalina.

Figure 21—Hinge characters and dentition in Liebea. A, Interior of left valve of L. squamosa showing myaJinid ligament area, umbonal septum, and anterior adductor pit on septum; b front view of bivalved specimen showing interlocking denticulations of front margin; C, crest; S, sinus. Approximately X5. [Image scaled and magnifications adjusted for web presentation.]

Hinge characters and dentition in Liebea.

From Septimyalina this genus is distinguished by the interlocking crenulations of the anterior margin below the beaks, by the deep adductor pit on the septum, and by the mytiloid contour. It is probable that Liebea arose from a primitive Septimyalina in which the anterior adductor was not completely lost.

Geologic range—Permian of Greenland and the eastern hemisphere. Unknown in the United States.

Dimensions of Liebea squamosa from Zechstein near Gera, Germany, in millimeters. All the specimens are bivalved internal molds except No. 1, which is a cast.
Specimen 1 2 3 4 5 6
Length   18 16 17 22 15
Height   15 14 14 15 12
Thickness 10 10 12 11   9
No. lig. grooves 8 7 7 8   8

Remarks—A small mytiliform pelecypod commonly referred to Liebea squamosa (J. de C. Sowerby), or its more widely employed synonym L. hausmanni (Goldfuss) (1837), is widely distributed and locally abundant in the late Permian Zechstein-Kazan provinces of northern and northwestern Europe, and is associated with a Zechstein fauna in the east Greenland geosyncline (Frebold, 1931). The species has been reported also from the Italian Alps (Merla, 1931) and, doubtfully, from the Karoo beds of Tanganyika (Cox, 1936), and the Bone Springs and Capitan horizons of West Texas (Girty, 1908). I think that the last mentioned citation is in error, and the material on which Girty based his identification is too poorly preserved to warrant generic identification.

Similar species described under Liebea have been recorded from the "Bellerophon beds" of the Carnic alps (Liebea dieneri Gortani, 1906) and the uppermost Permian of Madagascar (Liebea dieneri Treat, 1933, not Gortani, 1906). Other species referred to Liebea have been described from the Lower Productus limestone (Parafusulina beds) of India (Waagen, 1881) and the Maping limestone of China, "early" Permian (Grabau, 1936).

Unfortunately, most of the references to Liebea have been based on external form without information on the ligament and internal characters that distinguish Liebea from other Late Paleozoic mytiloid shells. The fact that the Permian rocks of many regions contain these diminutive shells, which agree closely in form, is of great stratigraphic interest and suggests that they may indeed all be closely related. It is urgent, however, that an attempt be made to determine whether or not all of the above-mentioned forms possess the critical characters of Liebea.

Characteristics of Liebea—Waagen (1881, pp. 292-295) established Liebea on small mytiloid shells from the German Zechstein, characterized, as he thought, by the following features: (a) Byssal notch lacking, (b) ligament internal, lodged in a number of resilifer pits observable only in old specimens, and (c) by certain peculiarities of the anterior "auricles." His interpretation of the ligament was erroneous. The original diagnosis is quoted:

Shell inequivalve, the right valve smaller than the left one: inequilateral, mytiIiform, with a small, not very distinct, posterior wing in both valves, and with a very small anterior wing in the left valve only. The position of this latter wing is not as in other Aviculidae, but is placed about vertically to the plane of the valve, turning round the smaller right valve and embracing it partially. This wing is compressed from both sides, bearing on its top a sharp ridge which projects at about right angles from below and in front of the beak of the left valve, treminating the area of that valve anteriorly and giving it a squarish shape.
In the right valve the existence of an anterior wing is barely indicated. In its place there is an incision of the shell, into which fits the bent-over anterior wing of the left valve.
Both valves are provided with a more or less distinct sharply defined area, which is linear in young specimens and very broad in old ones. It is smooth and only covered by distinct striae of growth, parallel to the hinge margin.
The area in the left valve is broader than that in the right one.
The entire hinge line serves for the attachment of a quite internal ligament. This ligament seems to be linear in young specimens, but in old ones it is distinctly divided in several parts, which are attached to a number of pits, between which the hinge margins of both valves seem to have nearly touched each other. The number of pits is generally four or five. The first is situated just below the beaks, the others are distributed over the posterior part of the hinge line.
The hinge seems to have been quite edentulous, as no trace of teeth has been detected up to the present.
Internally, below and in front of the beaks, there is in each valve a small septum, which separates the anterior wing from the remainder of the shell.
The muscular and pallial impressions cannot be observed in any of the specimens at my disposal. They have been described accurately by King, only the position of the anterior adductor, as indicated by him, seems to be somewhat doubtful. If the anterior adductor had been attached to the anterior wing of the shell, it must have been much larger in the left than in the right valve, as in the latter this wing is so rudimentary that there was not much place for the attachment of a large muscle. (Waagen, 1881, pp. 293-294).

Kings observations on the same species are pertinent:

As J. de C. Sowerby and Goldfuss have overlooked two or three important characters of this interesting species, it is necessary for me to add, that it is inequivalved; has recurved umbones, a horizontal septum within the umbonal cavity of each valve, and a wide longitudinally-grooved cartilage-furrow or fulcrum. Its muscular system has already been noticed; but it may be added, that the anterior visceral or pedal muscles, judging from the position of their impressions, must have been situated much further back than is usual in Mytilus (King, 1850, p. 160.)

It should be noted that King described the ligament area ("cartilage furrow") as being longitudinally grooved in Liebea squamosa; whereas, Waagen described it as being smooth, marked only by growth lines. There can scarcely be any question as to King's meaning in reference to the ligament grooves. because he described and illustrated the same feature in another species generally referred to the same genus, L. septifer (King) (1850, p. 161, pl. 14, fig. 13). L. septifer, as described by King, certainly had a ligament system similar to that of Myalina de Koninck, and judging from King's description of L. squamosa, this species also had a similar ligament.

Maslennikov, while studying Kazanian specimens from the Timan Mountains of northern Russia, was puzzled at being unable to confirm Waagen's observations on Liebea (Maslennikov, 1935, pp. 63-66). After cleaning some of the Russian specimens, he concluded that the hinge does not possess multiple resilifers such as those described by Waagen, Instead, he found the hinge area marked only by parallel striations, which he interpreted as growth lines. Maslennikov's plate 2, figure 11, indicates ligament grooves rather than growth lines on the hinge area of one imperfect specimen. He expressed somewhat hesitantly the possibility that the Russian species may differ from the Zechstein L. hausmanni (— L. squamosa), pointing out that the Russian form lacks the peculiar anterior auricle on the left valve and that the musculature varies slightly from descriptions of the western European specimens. He might have added also that the specimens illustrated by him differ considerably in form from the Zechstein representatives, and have a much narrower ligament area. Although the Russian form was described under the name Liebea hausmanni, Maslennikov intimated that perhaps his material should be classed with Myalina instead of Liebea.

When Waagen framed the original diagnosis of Liebea, he had under consideration two species, L. hausmanni (Goldfuss), from the German Zechstein, and L. indica Waagen, from the basal beds of the lower Productus limestone of India (horizon of primitive Parafusulina). Although most of Waagen's observations were made on the German species, he failed to designate it as the genotype; therefore the two species were genosyntypes. The first formal designation of a genotype for Liebea, to my knowledge, was made by Cox (1936, pp. 38-39) when he indicated L. squamosa as the type of Liebea. This designation is valid, in the absence of earlier designations, inasmuch as Cox placed L. hausmanni in the synonymy of the earlier described L. squamosa. It is my opinion that the two names, as Cox also believed, are subjective synonyms. Although the species L. squamosa would not be recognizable from Sowerby's description of 1829, King (1850) redescribed the species with Sowerby's specimens before him. Seemingly no one has questioned the identity of the form described by King as L. squamosa with the German L. hausmanni of Goldfuss.

Through the generosity of Mr. Kenneth Oakley of the British Museum of Natural History, I have examined four specimens of Liebea squamosa. Two of them are from the Magnesian limestone of England and are essentially topotypes of Sowerby's species. The other two are from the Zechstein of Gera, Germany, the locality that yielded Waagen's specimens. All of the specimens are internal molds, as were most of Waagen's. One of the English specimens is extraordinary in retaining a portion of the external mold of the ligament area of the left valve. This specimen, although imperfect, clearly had a ligament like that of other Myalinidae.

Another collection of specimens was made available for this study from the collections of the Geologisches Landesmuseum at Berlin, through the kindness of Dr. O. H. Schindewolf. These specimens are all from the Zechstein dolomites of the Gera district. Some, perhaps all of them, are from the middle Zechstein horizon. The more instructive specimens consist of six internal molds, some of which exhibit features of musculature, and one incomplete but superb external mold, which shows perfectly the ligament characters and form of the shell (pl. 15, figs. 3a-d; text fig. 5). This specimen is unique among all described Myalinidae in showing perfectly the imprint of the original conchiolin structures of the ligament, and verifies my prediction (Newell, 1938, pp. 34-35, pp. 26-30, fig. 6) that the ligament of Myalina and similar Paleozoic shells was comparable structurally to that of modern Arca.

Genus ATOMODESMA Beyrich, 1864

Atomodesma Beyrich, 1864, Konigl. Akad. Wiss. Berlin, Abhand., p. 68; ——, C. Wanner, 1922, Pal. von Timor, Lief. XI. no. XVIII, pp. 60-70.

Genolectotype, Atomodesma exarata Beyrich, 1864, designated by C. Wanner, idem., p. 63; range of genotype, Lower Triassic, Timor.

Wanner's diagnosis: Shell equivalve, transversely oval, more or less elongated diagonally, flat to strongly convex. Beak terminal, bluntly pointed, bent forward and inward, either free or fused with the cardinal margin. Cardinal margin straight, toothless, on the average up to half as long as the shell, with an equally long ligament groove, which is not divided by vertical septa into individual ligament pits, but occasionally carries fine longitudinal ridges. Prismatic layer of the shell very strongly developed, as in Inoceramus. Surface ornamented with irregular concentric wrinkles, and radial furrows of variable strength and number.

Geologic range—Permian and Lower Triassic of Timor; uppermost Permian (zone of Cyclolobus), Salt Range, India.

Remarks—Unfortunately Wanner's observations on the hinge characters do not conclusively establish the type of ligament possessed by Atomodesma. The general expression of the shells suggests that they are indeed myalinids, and presumably the ligament was of the Myalina type. I am more impressed by the similarity of Atomodesma with Aphanaia than was Wanner (1922, p. 62), the principal difference being that the typical species of Aphanaia is said to be strongly inequivalve, the right valve being more convex and having a longer beak than the left valve. Externally, Marwick's Maitaia is very like Atomodesma, and has a similar shell microstructure. Maitaia possesses a true myalinid hinge and, in addition, possesses an umbonal deck. Unfortunately, the presence or absence of an umbonal deck in Aphanaia or Atomodesma has not been disclosed.

Figure 22—Myalinid genera Atomodesma Beyrich and Maitaia Marwick. A, Diagram of Atomodesma exarata Beyrich, X 1, genotype of Atomodesma, from the Lower Triassic of Timor (redrawn from Wanner, 1922); B, diagram of Maitaia trechmanni Marwick, X .5, from the Maitai beds (Permian?) of New Zealand (redrawn from Marwick, 1935). [Image scaled and magnifications adjusted for web presentation.]

Myalinid genera Atomodesma Beyrich and Maitaia Marwick.

Genus MAITAIA Marwick, 1934

Maitaia Marwick, 1934, Fifth Pacific Science Congress, Proc., p. 948; ——, 1935, New Zealand Roy. Soc., Trans. and Proc., vol. 64, p. 295.

Genotype, by monotypy, Maitaia trechmanni Marwick, 1934; range of genotype, Maitai beds (Late Paleozoic, Permian?), New Zealand.

Marwick's diagnosis: Shell of moderate size, mytiliform, winged posteriorly, equivalve; beaks terminal, slightly raised above the long, straight dorsal margin. Hinge edentulous, margin thickened and flat, and having about three well-defined longitudinal, ligamental grooves. The shell margins, towards the beaks, thickened and bevelled, and below the beaks, internally, a septal plate was well developed in large shells. Shell substance consisting chiefly of the prismatic layer. Surface with low, rather irregular, concentric waves. Muscle-impressions weak, posterior adductor scar semicircular, of moderate size, placed far back and somewhat nearer the dorsal than the ventral margin.

Geologic range—Late Paleozoic (Permian?), New Zealand.

Remarks—Like the typical species of Atomodesma, Maitaia trechmanni is a relatively upright, nearly acline form. The external resemblance between the two genera, including peculiarities of the shell microstructure, suggests that they are closely related, or even congeneric. Unfortunately, the hinge and other internal characters of Atomodesma are not yet adequately known for a detailed comparison with similar genera. Aphanaia, so far as can be determined at the present time, seemingly did not possess an umbonal septum.

Genus APHANAIA De Koninck, 1877

Aphanaia De Koninck, 1877, Soc. royale Sci. Liege, Mem., 2nd ser., vol. 2.; ——, 1898, Geol. Survey, N. S. Wales, Mem. 6, p. 238.

Genolectotype, Inoceramus mitchelii M'Coy, 1847, herein designated; Geologic range of genotype, Permian, Australia.

De Koninck's diagnosis: This shell is inequivalved, inequilateral, gibbous, with a posterior obtuse wing, and an apparently edentulous, straight hinge. Its beaks are anterior and separated by a hollow area, and having also a ligament. The surface is marked with large concentric rings, usually very unequal, and like those of some species of Inoceramus, to which some authors have assigned examples of Aphanaia. Their chief characteristic lies in the number and shape of their muscular impressions; the impression of the adductors is double, very large, placed posteriorly, and very much nearer the ventral margin than the hinge line; the diameter of one is nearly double that of the other; the larger, which is usually closest to the ventral margin, is generally reniform, while the smaller is suborbicular. A very much smaller impression than the preceding, probably that of the foot, is seen just behind the beaks, very close to the hinge line. (De Koninck, 1898, pp. 238-239.)

Range of the genus, Permian of Australia, Siberia, ?South Africa.

Remarks—De Koninck described two species of Aphanaia with the original diagnosis of the genus. They are A. mitchellii (M'Coy), and A. gigantea De Koninck. These two species were genosyntypes. I am not aware of any formal designation of a genolectotype; therefore, I am designating A. mitchellii, the holotype of which is said to be in the Clarke Collection in the Sedgwick Museum, Cambridge.

It is unfortunate that the genus Aphanaia still is hardly any better understood than it was in De Koninck's day. Characters of the hinge are not adequately known and the shell microstructure is unknown in A. mitchellii. Externally the species might be confused with Posidonia, Inoceramus, Atomodesma, Mataia, and probably some other similar genera. In discussing a supposed Aphanaia (A. haibensis Reed,) from the Upper Dwyka beds of South Africa, Reed (1936, p. 162, 163) remarks that

It was at once recognized that this shell bore a great resemblance to De Koninck's genus Aphanaia from the Permo-Carboniferous beds of New South Wales, and a comparison with the type specimen of Aphanaia mitchellii (M'Coy) in the Clarke Collection in the Sedgwick Museum, Cambridge, strongly supported this view, though our African specimen does not show the striated ligamental area and we cannot be certain that the shell is inequivalve.

By implication, I would judge that Reed has recognized a "striated ligamental area" in the type specimen of A. mitchellii, even though such a feature was not mentioned either by M'Coy or De Koninck. The bilobed adductor scar as described and illustrated by M'Coy and De Koninck is very like that of Septimyalina Newell, n. gen. (text fig. 6, C). Circumstantial evidence suggests that Aphanaia mitchellii is indeed one of the Myalinidae, but there seems to be one striking feature unknown in other myalinids. That is, according to De Koninck (idem, p. 239), the right valve in A. mitchellii, as well as A. gigantea, is conspicuously larger than the left valve. In all other myalinids known to me the reverse is the rule. Any conclusions regarding the shell microstructure in Aphanaia are, at the present time, assumptions. Without more complete knowledge of Aphanaia, based on authentic specimens of A. mitchellii, I do not see how the genus can be recognized with any certainty. The genus Atomodesma Beyrich is very similar, at least superficially, to Aphanaia, and was established first and is better known. However, if Aphanaia is really strongly inequivalve, as indicated, this character will distinguish the two genera.

Licharew (1934 and 1939, p. 142, pl. 34, fig. 14) has reported Aphanaia from the Kolyma region in Siberia. I have had the privilege of examining this material in Professor Licharew's laboratory in Leningrad. The specimens are noteworthy for having a very massive shell consisting principally of very long calcite prisms, recalling the shell of Inoceramus. However, the beaks are not terminal, and the ligament area does not exhibit the furrows characteristic of the Myalinidae.


Prev Page--Family MYTILIDAE || Next Page--Foreign species

Kansas Geological Survey, Geology
Placed on web Dec. 18, 2017; originally published 1942.
Comments to webadmin@kgs.ku.edu
The URL for this page is http://www.kgs.ku.edu/Publications/Bulletins/Vol10_2/04_system2.html