ON THE PELECYPODS OF THE SOLOVIKHINSKY LIMESTONE OF GORNY ALTAI

L. L. Khalfin

The Solovikhinsky limestone, exposed in the northern part of Gorny Altai, is of considerable interest: it is, so far, the only known occurrence in Siberia of deposits of the Gedinnian stage in the strict sense of the term—that is, a stage lying at the base of the Devonian system and characterized by a mixture of Silurian and Devonian fauna (in approximately equal amounts). Gedinnian deposits also occur in the Salair, but there they are still combined with the Ludlovian, forming part of the so-called Tom-Chumysh beds.

The fauna (chiefly brachiopods) of the Solovikhinsky limestone was first described, and its Gedinnian age established, in 1948 by L. L. Khalfin, from collections made by M. P. Nagorsky [Khalfin, 1948]. Only recently, however, did N. P. Kulkov (1960) establish that this limestone belongs to the Chagyr suite, forming its upper part; on the basis of more extensive and careful collections of fauna, N. P. Kulkov, from the study of brachiopods, confirmed the Gedinnian age of the Solovikhinsky limestone.

In the fauna of this limestone, brachiopods decidedly predominate; they are very numerous and diverse. The other groups (trilobites, stromatoporoids, tetracorals) are represented very poorly. Mollusks are no exception: L. L. Khalfin [1948, p. 169] described from the Solovikhinsky limestone Orthoceras sp. and Strophostylus nanus Khalfin; pelecypods from this limestone had not been described up to now. They were first collected by N. P. Kulkov and handed over to us for study.

This collection is very small: it consists of only 20 specimens, many of which are poorly preserved. Nevertheless, it is of great interest both because it permits the characterization of the fauna of this important subdivision of Siberian stratigraphy to be supplemented, and because, along with several new species, it contains very rare and interesting forms, the closest analogues of which are known from the Silurian and Devonian of Bohemia.

Quantitatively, among the pelecypods of the Solovikhinsky limestone, representatives of the genus Conocardium Bronn predominate; they belong to several, chiefly new, species. But the greatest interest among them is presented by a form that we regard as a local (Altaian) subspecies of the highly characteristic species Conocardium ornatissimum Barrande.

The distinctive features of this species (as well as of the species Conocardium protectum Barrande) are as follows: the narrow and tall, roof-shaped body of the valve, along whose ridge runs a groove, devoid of radial sculpture and crossed by the finest concentric ridges. This group of species deserves to be separated (if this has not already been done) into a special subgenus, which might be called Barrandeocardium, with the type species Conocardium ornatissimum Barrande, 1881.

From the Upper Silurian and Lower Devonian of Bohemia, Barrande figured a number of shells of the actinopterioid type, characterized by varied and extremely peculiar radial sculpture, on the basis of which they were later in part separated into special genera (Barrande assigned them to the genus Avicula Brug.). Among them, three groups are well distinguished, represented by: 1) the Lower Devonian Avicula palliata (Barrande, 1881, pl. 220), with chevron-like or zigzag radial sculpture; 2) the Upper Silurian A. mira, A. pseudomira, and A. cognata (ibidem, pl. 226), with rhombic microsculpture formed by the intersection of two systems of obliquely radial riblets; 3) the Upper Silurian A. scala (ibidem, pls. 226–227), with wavy, ribbon-like radial striae.

Representatives of the first and third groups have been found in the Solovikhinsky Limestone; these are:

Prantliella palliata (Barrande) subsp. altaica;
Actinopteria (?) scala (Barrande) subsp. altaica.

Thus, in the small collection of pelecypods gathered from the Solovikhinsky Limestone, three forms have been identified, evidently and most closely vicariant with pelecypods from the Bohemian Silurian and Devonian. Comparisons of these forms with their Bohemian analogues are given below, in their descriptions. Here we shall dwell on several questions of a general nature, namely: on the question of the qualification of differences analogous to the differences between the Bohemian and Altaian subspecies; on the question of the significance for stratigraphic correlation of forms possessing such differences; and on the question of the stratigraphic significance of pelecypods in general.

On the first question it must be noted, first of all, that in the present case the forms (subspecies) being compared display a commonality in the most important, very specific features; that is, the close relationship of these forms cannot be doubted. At the same time, there are also obvious differences between them. The question of how these differences should be evaluated and qualified is extremely complex, and in its solution the experience and tact of the investigator play no small role, which in turn cannot fail to impart a certain degree of subjectivity to the decision adopted.

For the purposes of correlating deposits and determining the age of a fauna, it is very tempting in such a case to suppose that these differences are due to differences in the state or degree of preservation of the specimens being compared, since the view has become firmly rooted that only on the basis of identical forms is stratigraphic correlation reliable and convincing. There is no need to prove that, for the complete identification of specimens coming from regions considerably distant from one another, very weighty evidence is needed, and any stretching in this matter can only do harm: a great number of cases are known in which, on the basis of hasty and unreliable identification, the age of deposits was determined incorrectly.

We have already had occasion to explain [Khalfin, 1958, p. 149; Khalfin, 1960, p. 19] that correlation on the basis of vicariant forms

does not exceed the limits of permissible accuracy in biostratigraphic correlations. The whole difficulty lies only in the fact that it is not always possible to separate similarity caused by systematic (or genetic) vicariance from similarity produced by convergence or parallelism—that is, from a case in which similarity provides no basis for stratigraphic parallelization. But precisely in our concrete case such doubts have no place: the diagnostic characters peculiar to all three species are so specific.

A question of fundamental importance in the stratigraphic aspect is whether the Altaian and Czech representatives of these species may be regarded as vicariant forms. To this question we give, without hesitation, an affirmative answer, and this is sufficient for stratigraphic correlations. In this sense it is of no importance whether the systematically vicariant forms are considered as geographic races, subspecies, or even as independent species. We believe that at present, and in ordinary cases, paleontology does not possess reliable criteria for deciding whether we are dealing with closely related species or with subspecies of a single species. In the future, when such criteria are found, taxonomic refinements will also be possible; but, as has already been said and as we emphasize once more, this has no significance for the accuracy of stratigraphic correlations. Therefore the important thing is the very fact of establishing vicariant forms; as for their nomenclature, this has no essential significance.

Previously we were inclined to regard vicariant forms as independent species, but perhaps it is more convenient, already in the very name, to reflect this property that is important for stratigraphic correlations—their vicariant character. If this point of view is adopted, then vicariant forms should be treated as geographic subspecies and these subspecies should be given corresponding geographic names reflecting their areas of distribution. This is exactly what we do with regard to the fauna that interests us in the present case, naming the vicariant species according to the regions of their distribution: subsp. bohemica and subsp. altaica.

The phenomenon of vicariance among the pelecypods of the Altaian Devonian is not confined to the cases cited above for the pelecypods of the Solovikhinsky Limestone. We noted it earlier [Khalfin, 1948] for some pelecypods of the Lower Emsian (Kondrat’ev beds) of Altai. Especially indicative in this respect are Tolmaia squamosa Khalfin, undoubtedly most closely related to Tolmaia lorana (Fuchs) from the Lower Koblenzian of the Rhine, and Archaeocardium scrobiculare Khalfin, just as close to Archaeocardium animans (Barrande) from the Czech Lower Devonian.

The presence of such, highly characteristic forms with distinctive and easily recognizable features makes it possible to recognize the Paleozoic pelecypods as a group of very great stratigraphic importance, not inferior in this respect to other faunal groups. In noting this, we consider it necessary to point out also another feature of Paleozoic pelecypods—the presence among them of many forms simulating resemblance, but belonging to different species and even genera. This applies especially to pelecypods with a very simple shell morphology. For example, among nuculids, lagoonal–freshwater pelecypods of the Upper Paleozoic (genera Anthraconauta Pruvost, Anthraconia Trueman et Weir, and others), and some other groups of pelecypods, one often observes a very great resemblance among species that are in fact different and exist-

...that existed at different times. This similarity has been the cause of repeated serious errors in the identification of various species and in determining the age of deposits in the past; and, unfortunately, such errors occur even now, as shown by works of some authors published in recent years. This question deserves special consideration, which we intend to give elsewhere.

Returning to the pelecypods of the Solovikhinsky Limestone, we may say that they confirm the characterization of the fauna of this subdivision given by us earlier [Khalfin, 1948, p. 22]: it contains Silurian and Devonian forms, and forms of the Czech–Uralian (Hercynian) facies type predominate.

Below is given a description of the three forms mentioned above from the pelecypod assemblage of the Solovikhinsky Limestone.

Conocardium ornatissimum Barrande, 1881, subsp. altaica subsp. nov.

Plate I, figs. 1, 2.

Compare: Conocardium ornatissimum Barrande, 1881, pl. 201, fig. III—1—6.

Diagnosis. A small shell, somewhat developed in height¹. The body of the valve is very high, but narrow (compressed in the anteroposterior direction), roof-shaped, with steep slopes (the posterior slope steeper). Along its crest there is a narrow, flattened groove, bounded by thin radial ribs. The anterior lobe is well developed, triangular, weakly convex; there is no furrow between this lobe and the body of the valve. Rounded radial ribs cover the anterior lobe and the anterior slope of the valve body, but are absent on the crest and on the posterior slope. The ribs on the lobe are the broadest. Very fine concentric ridges are clearly expressed; on the crest they form a weak bend, convex toward the umbo.

Holotype: specimen 38—8; left valve, figured on Plate I, figs. 1, 2; Devonian conglomerate northeast of the village of Solovikha.

Material. One specimen (holotype).

Description. A small left valve, somewhat damaged, but permitting observation of the characteristic features that sharply distinguish it from all other specimens of conocardiums from the Solovikhinsky Limestone. In this valve the posterior margin is damaged and the rostrum has not been preserved.

The principal feature of the valve is its narrow body (compressed in the anteroposterior direction) with strong inflation. The height of the valve is 10 mm; the length without the rostrum is 8 mm.

The narrow body of the valve represents a high saddle-like elevation, somewhat asymmetrical, with a blunted top, a very steep (but not vertical) posterior slope and a somewhat gentler anterior one. In longitudinal section the surfaces of both slopes form a dihedral angle of approximately 90°. There is no rib or keel (even a rounded one) separating the posterior slope, as is characteristic of many conocardiums, in the valve described. Along

along the crest of the body of the valve there is a narrow (1 mm wide) strip, bounded by fine riblets and lacking radial ribs.

The anterior lobe is broad, weakly convex; the furrow, which in many conocardiums separates it from the body of the valve, is not expressed.

The sculpture consists of rounded radial ribs and the finest concentric ridges, characteristic also of all the other conocardiums of the Solovikhinsky Limestone. The character of the ribbing changes in a peculiar way over the surface of the valve: the thickest and most distinct ribs are situated on the anterior lobe (in other conocardiums from the Solovikhinsky Limestone, on the contrary, the ribs on this lobe are the thinnest); in passing onto the anterior slope of the body of the valve they become thinner and weaker, disappearing completely on the crest of the body of the valve and on the posterior slope.

Fig. 1. Conocardium ornatissimum (Barrande) subsp. altaica. General view of the left valve, holotype. ×7.

Comparisons. The specimen described possesses very characteristic features proper to the conocardiums of the protectum—ornatissimum group, widespread in the upper Silurian and lower Devonian of Bohemia (Barrande, 1881, pls. 201, 202 and 204). The most important of these features are as follows: a high and narrow roof-shaped body of the valve, along the crest of which runs a flat furrow, and the complete absence of a diagonal keel and rib separating the posterior slope.

Especially close to the specimen described is the Lower Devonian Conocardium ornatissimum, distinguished by the presence of radial ribs on the posterior slope, which in our specimen is smooth, and by a narrower body of the valves. The Upper Silurian Conocardium protectum Barr. differs, in addition, in the broader posterior part of the shell.

Locality. The specimen described comes from Devonian conglomerates exposed northeast of the village of Solovikha (locality DK-3).

Actinopteria (?) scala (Barrande, 1881), subsp. altaica
subsp. nov.
Pl. I, figs. 3—6.

Comp.: Avicula scala Barrande, 1881, pl. 226, fig. IV.

Diagnosis. A small, moderately oblique shell, approximately equally developed in length and height. Hinge margin straight, long. The acute umbones are somewhat shifted forward from its middle. Auricles well developed; the posterior auricle larger than the anterior. Convexity of the valves insignificant. Diagonal elevation low, broadly rounded. The few interruptions in shell growth are sharply lamellar. On the lower half of the shell, the broad intervals between

they bear the finest radial striation and indistinct, fine concentric lines.

Holotype: specimen 38-10; right valve, illustrated in Figs. 3, 4, Pl. I. Northeastern end of the village of Solovykha.

Material. Two right valves of good preservation, although with somewhat damaged outlines; the sculpture is well preserved. In view of the fact that there are some differences in the character of the sculpture of these valves, they are described separately below.

Description of the holotype. A small right valve with a broken posterior margin, the outlines of which can be restored from the course of the growth marks; therefore, on the whole, the outline and shape of the valve are clear, especially since the general preservation of the specimen is good.

The height of the valve is 9 mm, the length approximately the same. The angle between the main diagonal and the straight hinge margin is 70°. The length of the hinge margin is approximately equal to the length of the shell. The sharp umbo, directed along the main diagonal, is considerably raised above the hinge margin; it is somewhat displaced forward from the middle of the latter.

The auricles are well developed; the posterior auricle is considerably larger than the anterior one. In the outline of the shell the auricles are weakly expressed; judging from the course of the growth marks, the posterior auricle below was bounded by a very slight concavity of the posterior margin.

The convexity of the valve is not great. Its umbonal part is the most convex; from there the surface of the valve gradually descends toward the lower-posterior margin, forming a very gentle and broad diagonal elevation. Toward the auricles the umbonal part of the valve descends smoothly, toward the posterior margin more steeply; the descent is gradual, so that the auricles become flat only near their apices.

The outline of the anterior margin is evenly convex; the lower margin is broadly rounded; the posterior margin in its upper part forms a weak concavity.

The surface of the valve is provided with a few, widely spaced, sharply lamellar growth marks; their number on the valve is six. The width of the bands between them increases from the umbo toward the lower margin; these bands are flat on the upper half of the shell, and slightly concave on the lower half (including the band between the last growth mark and the margin of the shell). On these bands there are finer, partly barely perceptible concentric lines.

Fig. 2. Actinoptera (?) scala (Barrande) subsp. altaica. General view of the right valve, paratype, × 8.5.

The valve possesses a very characteristic radial sculpture in the form of the finest, threadlike riblets. They are arranged in the intervals between the growth marks and in the interval between the last growth mark and the margin of the valve. They are expressed unevenly: quite distinctly in the three lower intervals over the entire width of the latter, in the fourth

in the interval are weakly expressed, chiefly near its lower margin; in the next (fifth) interval they cannot be detected, but are only guessed from a certain unevenness in the outline of the corresponding growth mark; at the very umbo there is no radial sculpture.

The riblets are most distinct in the interval between the margin of the valve and the last growth mark and in the next interval above it, the character of the riblets in these intervals being different. In the first of them the riblets are irregular, bend in a wavy fashion, and partly merge with one another; in the second, many riblets, as they approach the lower margin of the interval, merge in pairs and threes, forming thicker and higher riblet-swelling.

In each interval between growth marks the riblets arise independently, not passing into adjacent intervals across the growth marks.

Description of the paratype (specimen 38-11). Right valve approximately of the same dimensions as the holotype: its height is 10 mm. The outlines of the anterior margin are damaged, those of the posterior margin are hidden in the rock; the outlines of the valve are restored from the course of the growth marks. In all respects this specimen is similar to the holotype, but the number of growth marks, in accordance with the somewhat larger dimensions, is seven. The difference from the holotype consists in the fact that the radial striae are of one type: thin, slightly irregular, they do not form the nodular swellings that were noted above in the holotype.

Fig. 3. Actinopteria (?) scala (Barrande) subsp. altaica. Character of the sculpture, paratype, ×9.

Remarks. The differences noted above between the holotype and the paratype are hardly sufficient for the taxonomic separation of these specimens.

Comparisons. In the outlines of the shell, the character of the growth marks, and the character of the radial striation, our specimens may be referred to the Bohemian species Actinopteria (?) scala (Barrande); the chief difference of our specimens from the Bohemian ones (Barrande, 1881, pl. 226–227) is that in their upper part the posterior margin is concave, not convex. On this basis we consider it possible to distinguish within this species two subspecies—the Bohemian (subsp. bohemica) and the Altaian (subsp. altaica). In general shell form some species of the genus Rhobopteria Jackson are close (for example, Rh. mira—see Barrande, 1881, pl. 226), but they are readily distinguished by the rhombic-reticulate microstructure.

Distribution. Barrande described this species from the Upper Silurian of Bohemia (stage e 2).

Locality. Both specimens come from the Solovikhinsky limestone massif at the northeastern end of the village of Solovikha (holotype—from bed SI-2g, paratype SI-2b).

Prantliella palliata (Barrande, 1881), subsp. altaica subsp. nov.

Plate I, fig. 7.

Comp.: Avicula palliata Barrande, 1881, p. 24, pl. 220, fig. 1–10 (only!)

Diagnosis. Shell of small size, markedly convex, moderately oblique, sometimes almost straight, with massive beaks. Hinge margin straight, long. Growth marks few, coarse. Subradial ribs form a characteristic chevron-like sculpture on the surface of the valves.

Holotype. Left valve (specimen 38–13), illustrated on Plate I, fig. 7. Limestones at the northeastern end of the village of Solovykha.

Material. One left valve, on which the rafter-like sculpture characteristic of the species is clearly visible. A second specimen (38–14) is conditionally assigned to this species: it is also a left valve, the substance of which is preserved only in part, and on which the indicated specific sculpture is only with difficulty discernible.

Description. A small left valve, considerably damaged: its anterior and lower margins are damaged. Otherwise the preservation of the specimen is good; in particular, the chevron-like (zigzag) sculpture most characteristic of this species has been well preserved. The valve is 9–10 mm in height and length. The hinge margin is long; the posterior auricle is large, its posterodorsal angle close to a right angle. The beak is rather massive, considerably raised above the hinge margin.

The valve is markedly convex. Its umbonal part is the most convex; in longitudinal section this part represents an unequal-sided arc with a steeper anterior and a gentler posterior slope. Downward and backward, along the main diagonal, the surface of the valve descends very gradually.

Fig. 4. Prantliella palliata (Barrande) subsp. altaica. Left valve, holotype, ×2.

The valve is moderately oblique: the angle between the direction of the main diagonal and the hinge margin is about 60–70°.

The growth marks are widely spaced: four of them can be counted on the valve. The surface of the valve is covered with thin and flat subradial ribs, which disappear near the beak itself and are quite distinct on the lower half of the valve. On the posterior auricle, especially on the part adjoining its posterior margin, the ribs are also distinct, but considerably thinner than on the diagonal elevation.

The ribs are not radial; crossing radial directions, they form a very peculiar sculpture—zigzag, chevron-like, “herringbone.” The zigzags, alternating, are turned successively with their apices toward the beak and toward the margins of the valve. The angle formed by the zigzags is very acute (about 30°). These ribs pass across the growth-rest marks without being interrupted by the latter.

Remarks. Barrande assigned to this species 12 specimens from the Koněprusy Limestone of Bohemia. These specimens differ greatly in size, shell outline, and character of sculpture. In particular, the typical chevron-like sculpture is observed only in the four specimens indicated above in the synonymy; the remaining specimens, in our opinion, should be separated from the present species.

A few remarks on the growth marks observed in shells of this species. Both in the illustrations given by Barrande and in...

in our specimen the usual lamellar growth marks are absent, but it is clearly visible that the thin surface layers of the valves, at varying distances from the beaks, peel away and break off along extremely irregular, though generally to some extent concentric, broken lines. Apparently the growth marks were not only lamellar, but these lamellae were broad and thin, and each time the growth of the shell began beneath such a lamella overhanging the newly accreted part of the shell.

In the process of fossilization or preparation, these lamellae-scales break off along irregular lines. Alongside the latter, regular concentric lines are also noticeable on our specimen, corresponding to pauses in the growth of the shell; they have the character of shallow and narrow concentric furrows, across which the subradial ribs continue without interruption.

Comparison. The Czech specimens of this species cited in the synonymy differ from ours in having wider intervals between the ribs and in the presence, within these intervals, of concentric striae.

Distribution. Representatives of this species have been found in the Lower Devonian of Bohemia (Upper Koněprusy limestones—Chlupáč, 1955, p. 103) and in the Zhedin deposits of Altai.

Locality. Limestones at the northeastern end of the village of Solovikha (bed SI-2g); P. cf. palliata was found there as well, in bed SI-2v.

Literature

Kulkov N. P. On the question of the stratigraphic position of the Solovikha Limestone of northwestern Altai. Reports of the Academy of Sciences of the USSR, vol. 134, no. 6, 1960.

Khalfin L. L. Fauna and stratigraphy of the Devonian deposits of Altai. Proceedings of the Tomsk Polytechnic Institute named after S. M. Kirov, Order of the Red Banner of Labor, vol. 65, issue 1, 1948.

Khalfin L. L. Lamellibranch mollusks of the Ordovician of the Chuya-Ili Mountains. Transactions of the Geological Institute of the Academy of Sciences of the USSR, issue 9, 1958.

Khalfin L. L. The principle of biostratigraphic correlation. Transactions of the Siberian Institute of Geology, Geophysics, and Mineral Raw Materials, issue 8, 1960.

Chlupac I. Stratigraphicka studie o nejstarsich devonskych Barrandienu. Sbornik Ustr. Ustavu geolog. svazek XXI-1954, oddil geol., 2 dil., 1955.

Hall J. Palaeontology of New York, vol. V, pt. I. Lamellibranchiata, II. 1885.

Hind W. Monograph on the British Cardoniferous Lamellibranchiata, vol. 1. Palaeontographical Society, 1896-1900.

1. Following Hall (Hall, 1883, p. XXXIV), Hind (1896—1900, p. 450), and other authors, we accept that the obliquely truncated margin of the shell, usually provided with a siphonal tube (“rostrum”), is the posterior one. In connection with this, in the structure of the conocardium shell we distinguish the body of the valve, the anterior lobe, and the posterior rostrum (the latter is usually broken off). ↩