PRINCIPLES OF CLASSIFICATION. 47 



partially or wholly, not of a calcareous exudation from the sarcode-body, but of particles of 

 sand obtained from without, the cement with which they are attached together being all that 

 is furnished by the animal. The aspect of such " arenaceous " shells necessarily varies con- 

 siderably with the kind of material of which they are composed ; and the same t3'pe, inhabiting 

 several different localities, may thus present as many diverse fades. Thus, the TextularicB 

 dredged off the Canary Islands have their shells entirely composed of fine particles of a 

 black sand, apparently formed by the disintegration of volcanic rocks ; whilst those of the 

 Red Sea have shells of a grayish-yellow colour, the arenaceous particles of which vary much 

 more in their nature and origin. In some instances the particles are very uniform in size, 

 and are very methodically disposed (like those of which the tube of Pectinaria is made up) ; 

 so that the surface of the shell has almost the appearance, when sufficiently magnified, of a 

 tesselated pavement (Plate VI, fig. 41). This regularity — alike in size, form, and arrangement, 

 — is sometimes so remarkable as to have given rise to the supposition that the siliceous par- 

 ticles are not derived from external sources, but are formed by exudation from the surface of 

 the contained body. As, however, there is no siliceous " test " at present known to be formed 

 by an animal of this group, which is not obviously made up of an aggregation of distinct parti- 

 cles, instead of possessing the structural homogeneousness proper to the shells which are 

 undoubtedly formed by calcareous or siliceous consolidation, it would probably be correct to 

 say that the true " shell " of Foraiidnifera is imiformly calcareous, and that when this is 

 replaced by a siliceous " test " the materials of such test have been drawn together from 

 external sources. There are certain cases, on the other hand, in which the sandy paiticles 

 are less uniform in size and less regular in disposition, and are imbedded in a calcareous 

 cement which forms the essential constituent of the shell ; in these the arenaceous texture, 

 being superficial only, and to a certain extent accidental, has not that importance as a differ- 

 ential character which it bears when extending throughout the thickness of the shell.* 



* The surface of certain Foramiuiferous shells (especially Globigerince) is not unfrequentlj' 

 studded with the minute rounded or oval bodies to which the name " coccolitlis " was first given by 

 Professor Huxley. (See his Appendix to the ' Report on Deep-Sea Soundings in the North Atlantic 

 Ocean, between Ireland and Newfoundland, made in H.M. " Cyclops," Lieut. -Com. Dayman, in June 

 and July, 1857.") These bodies have been since observed by Dr. Wallich (civ b), not only in the free 

 state, but adherent to the siu-face of minute spherical cells which seem to constitute a rudimentary 

 type of Foraminifera ; the uniformity of this adhesion being such as to lead almost necessarily 

 to the conclusion that it is a normal condition. These " coccoliths " are described by Dr. Wallich 

 as of an oblong form, concave on their internal aspect, and convex externally (Fig. I, s, i), their 

 average length being about l-3700th of an inch ; in some specimens there is but a single aperture in the 

 centre ; in others the aperture is double, the two portions being separated by a 

 delicate transverse band; and the external marginal surface, which constitutes a ■''"'• ^• 



quoit-like oblong ring round the central perforated portion, is marked with radi- '•^^*; 



ating striae. The spheres, to the surface of which the coccoliths are found adherent %^ \ 

 at nearly regular intervals, are stated by Dr. Wallich to have a diameter from l-1.250th ale 



to l-1600th of an inch, and to be composed of a sarcode-like substance enclosed W ^ y V5>s^ 

 in a dehcate limitary wall, apparently consolidated by calcareous deposit (Fig. I, 1, 2, Cocoosplieres ; 

 1, 2). These bodies, to which he has given the name of " coccospheres," are 3, Jr, CoccoUtlis. 



