There are several bivalve shells which pre sent what may be termed a sub-nacreous struc ture, their polished surfaces being covered with lines indicative of folds in the basement membrane ; but these folds are destitute of that regularity of arrangement which is neces sary to produce the iridescent lustre. This is the case, for example, with most of the Pecti nidce, also with some of the Illytilaceo, and with the common Oyster. Where there is no indication of a regular corrugation of the shell-membrane, there is not the least approach to the nacreous aspect ; and this is the case with the internal layer of by far the greater number of shells, the presence of nacre being exceptional, save in a small number of families.
The membranous shell-substance, some form of which constitutes the internal layer of most bivalve shells, is occasionally traversed by tubes, which seem to commence from the inner surface of the shell, and to pass towards the exterior. These tubes vary in size from about the I-20,000th of an inch, or even less, to about the 1-2000th ; but their general diameter, in the shells in which they most abound, is about 1-4000th of an inch. The direction and distribution of these tubes are extremely various in different genera. Thus, in Anomia Ephippium they are scantily dis tributed in the internal nacreous lamina ; but in the yellow outer layer they are very abundant (fig. 415.), forming an irregular net work, which spreads out in a plane parallel to the surface. In Cleidothcerus chamoides, on the other hand, the tubes are abundant in the internal layer of the nacreous lining, where they form an intricate but irregular net-work parallel to the internal surface of the shell ; and from this arise a series of straight tubes, which pass nearly at right angles with the surface, at a considerable distance from each other, through the external portion of the nacreous layer, towards the cellular structure which constitutes the exterior of the shell. This, however, they do not penetrate; stopping short as they approach it, just as the tubes of dentine cease at its plane ofjunction with the enamel. The diameter of the tubes is toler ably uniform, even vvhen they divaricate ; the trunk not being much larger than either of the branches. In other instances, however, no such net-work is formed, but the tubes run at a distance from each other, traversing the shelly layers obliquely, and are then usually of' comparatively large size : this is the case, for example, with some species of Arca and Pectuneulus. That these tubes are not mere channels or excavations in the shell substance, is proved by the fact that they may be frequently seen very distinctly in the decalcified shell-membrane. They often pre sent, in their beaded aspect, indications of a cellular origin, as if they had been formed by the coalescence of a series of cells arranged in a linear direction. They are generally
has everywhere a similar origin; and if one variety of membranous shell-substance be thus most abundant in shells whose exterior has a foliated or sculptured character ; and not un frequently they may be distinctly seen to pass directly towards the prominences of the sur face, — as in Lima scabra and various species of Chama. They are by no means restricted, however, to shells thus characterised; nor are they universally present in them.
Of the origin and mode of formation of the membranous shell-structure, it is rather diffi cult to give an exact account. Possibly, after the epidermic cells have undergone calcifica tion, so as to form the external cellular layer, the basement membrane itself may become detached from the surface of the mantle, in combination with a layer of calcareous matter. Even in nacre, however, which may be con sidered as the most perfect form of this sub stance, indications of cellular structure are not unfrequently to be seen, especially in univalve shells: these are particularly evident in Ha!Roth., the nacreous lamina of which, when carefully examined under a sufficiently high magnifying power, are found to be com posed of minute cells of a long oval form (fig. 416.), their short diameter not being above 1-5000th of an inch. Their boundaries in many parts are very indistinct, or even disappear altogether, so that every gradation can be traced, from the obviously- cellular ar rangement shown in fig. 412., to the homo geneous aspect presented by the nacre of bivalve shells. The same cellular structure, and the same gradation to a homogeneous stratum, may be made apparent in the decal cified membrane ; so that here we seem to have evidence that even the membranous shell substance is originally formed by the agency of cells, although the boundaries of these have usually been subsequently obliterated, so that the structure comes to present a homoge neous aspect. Indications of the same cel lular organisation may be detected in the na creous lining of the shell in Turbo and Nautihis. We seem justified in concluding that nacre proved to have been formed by the agency of cells, little doubt can be entertained as to the corresponding organisation of others. The fact may probably be, that, as maintained by Professor Goodsir 4', the basement mem brane is itself composed of cells more or less perfectly developed, the boundaries of which usually disappear. Of this view a very good illustration is afforded by the va rious examples of shell-membrane ; which pre sent every gradation, from the most per fectly homogeneous pellicle, to a distinct stra tum of cells.