MORPHOLOGY OF THE ORTHOID SHELL 41 



of Schizophoria the low transverse ridge dividing the anterior adductors from the posterior ones is 

 oblique, being directed antero-laterally; in such the lateral pallial trunks extend in the same direc- 

 tion as the adductor ridge, thus giving rise to six subparallel trunks. 



Our studies show that in the present state of our knowledge not much taxonomic importance 

 can be attached to the dorsal pallial sinuses, because of the absence of marked variation, and also 

 because of the rarity of specimens in which they are preserved. 



MICROSTRUCTURE OF THE SHELL 



The orthid shell of the Middle Ordovician, as shown by Sardeson,^" consists of two calcareous 

 layers as follows: (1) an outer, thin, first or primary, non-fibrous layer that grows at the margin of 

 the shell only. It grows at an equal rate along the whole hinge, and elsewhere around the periphery 

 at an increasingly greater rate toward the front. It is this first layer that determines the general 

 outline and convexity of both valves, the folds and sulci and the plications. Sardeson erroneously 

 calls this primary calcareous layer the "epidermis," but in living brachiopods the thin chitinous 

 outer skin has long been known as the epidermis or periostracum and it covers the whole of the shell. 

 Such an epidermis undoubtedly also existed in the ancient brachiopods, but all trace of it has been lost 

 in fossilization. It is to this external chitinous skin that the term epidermis should be restricted, and 

 Sardeson's "epidermis" is the thin lamellar layer mentioned on an earlier page. 



(2) Sardeson next points out that inside the primary lamellar non-fibrous layer there is a much 

 thicker one composed of prisms or fibres of calcite, which are directed obliquely forward and inward 

 with respect to the whole shell. They are based against the lamellar layer and terminate upon the 

 inner surface of the valves. The growth of this fibrous shell matter is not at the edge of the valves 

 but upon the whole inner surface, and it is deposited by the mantle proper. This fibrous layer 

 thickens the valves more and more, obliterates the plications and other inequalities, and often thickens 

 the articulating and brachiophore structures. When the shell is endopunctate, it is this fibrous layer 

 that is perforate, due to the shell's growing around the minute and threadlike papillce on the outer 

 surface of the mantle and adhering to the inner surface of the outer non-fibrous lamellar layer. 



Young^" has also shown that in the genus Productus there are two layers, the inner one being 

 punctate (now known not to be true) and the outer impunctate. Kozlowski, however, points out 

 that other strophomenids are not punctate. Young also observed a similar condition in Eichwaldia 

 and Chonetes. In the impunctate genera it is probable that a similar condition exists, that is, there is 

 an outer calcareous layer growing only at the margins of the shell which determines the external 

 sculpture of the valves, and this is covered on the inner surface by the main mass of the fibrous or 

 prismatic shell. In punctate shells the primary shell matter of teeth, dental plates, apical plates, etc., 

 is impunctate. The deltidium is said to be impunctate in punctate shells.^^ 



In Cambrian shells Walcott reports a shell structure different from those of later periods. 

 According to him, the billingsellids and other Cambrian shells that he sectioned appear to have a 

 granular structure throughout. To the writers, this granular shell structure, restricted to the Cam- 

 brian, always appeared anomalous. Therefore we sectioned a specimen of B. Undstromi from the 

 Paradoxides zone of Sweden, and it proved to have a fibrous structure. Indeed, the fibers are clearly 

 visible in the specimen with a low-power hand lens (pi. 29, fig. 12). We have also rephotographed 

 one of Walcott's sections of B. coloradoensis and this too proved to be fibrous (pi. 29, fig. 13). The 

 photograph of this thin section was enlarged about 30 times, whereas Walcott used an enlargement of 

 50 to 100 times. We believe that this latter enlargement is too great and unduly emphasizes the 

 fractured condition of the mineralized shell. Furthermore, we found no evidence of puncta; in any 



"Amer. Geo!., vol. 19, 1897, pp. 92-93. 



=»In Davidson, Brit. Foss. Brach., Suppl. to Carb. Brach., vol. 4, 1874, pp. 296-302. 



'" In regard to this statement, the writers wish to say that they have studied but one punctate shell which has a del- 

 tidium {Kayserella), and unfortunately the one specimen studied was too mineralized to show whether or not the deltidium 

 was punctate. Beecher's statement that the deltidium is impunctate in punctate shells loses force in the light of Kozlowski's 

 recent work (1929), in which the latter shows that the supposed puncta: of the Strophomenacea are in reality granula: 

 formed by rods of crystalline substance in the shell. Since the Strophomenacea are (excluding the Clitambonitida:) the chief 

 deltidium-bearing shells, the anomaly emphasized by Beecher may after all not prove to be so important as he thought. 



