cement formed of the proper shell-substance, which is never wanting; and the close accordance in every other character between shells which are thus superficially altered and such us conform to the ordinary type, forbids our regarding the former as more than varietally distinct from the latter. The case is very different, however, with regard to certain genera in which the power of forming a proper shell seems to be altogether wanting; the testaceous envelope being essentially composed of substance directly derived from without, the only material furnished by the animal being the organic glue that holds them together, their substance is generally composed of a very fine cement in which coarser particles are imbedded; the former Sometimes predominating, so that the shell is smoothed off on the surface; whilst if the latter be in excess, the surface of the shell is rough. Of this family the genus Lituola is the most characteristic; and the variety of forms into which it passes, several of them so closely resembling those of other genera as to have been mistaken for them, would not be readily conceived by any but such as have made a special study of them.[1] In the genus Trochammina (Parker and Rupert Jones) We have an instance of a gradational transition from the monothalamous to the polythalamous type; for whilst its lowest form is a continuous vermicular spiral (the Spirillina arenacea of Prof. Williamson), this comes to present, in some instances, a degree of segmental division scarcely inferior to that which some of the most vermiculate forms of Rotalia are reduced. The genus Valvulina forms the transition between this group and the "perforated" series; for whilst the principal part of its "test" is uniformly made up of an aggregation of sandy particles, leaving no such pores for the exit of pseudopodia, as can be readily discerned in the arenaceous Textulariæ, this has a basis of true shell-substance in which pores can be distinguished.
In the whole of the sub-order Perforata, the shell is calcareous, and is formed of a dense hyaline or vitreous substance, which is traversed by tubuli running straight from the cavity of the chambers to the external surface, whose diameter usually ranges from 1-3000th of an inch (as in Rotalia and Planorbulina) to less than 1-10,000th (as in Operculina and Cycloclypeus). There can be no question that even the smallest of these tubuli are large enough to transmit the finest threads into which the protoplasmic substance may sub-divide itself: and looking to their remarkable continuity through successive layers of shell substance, when (as in Operculina) the earlier whorls are completely embraced by the later, there can, I think, he no reasonable doubt that, through their means, a direct communication is maintained between even the earliest and innermost segments and the surrounding medium.[2] This, of course, renders the successive
