200 PROF. W. J. SOLLAS OX SACCAMMIXA CARTEEI, [vol. Ixxvii, 



boundaiT is completed by a lateral extension of the bases of the 

 ])iers on each side, or, when these fail to meet, by the confluence of 

 the arches npon which they rest. The component fibrils run for 

 the greater part parallel with the course of the piers and arches. 

 In some regions, particularly near the base of the piers or pillars 

 and in the upper or under wall of a chamber, the fibrillar is re])laced 

 by a minutely granular structure. 



In a vertical radial section pillars presenting characters similar 

 to those just described are also present ; but frequently several of 

 them are confluent laterally (that is. in a vertical direction) to 

 form a more or less continuous wall to the vertically extended 

 chambers, and frequently also several are aligned along a radius to 

 form a radial strand. The component fibrils of the pillars main- 

 tain the radial direction, and where the pillars of one zone meet 

 those of another the ends of the fibrils are opposed along the line 

 of junction. This line corresponds to the boundary between 

 successive zones. 



Examined between crossed nicols the primordial chamber pi"e- 

 sents, like GJohirferiua, a dark cross, but with broader arms and 

 more vaguely defined boundaries : in accordance with this the 

 optical sign is found to be positive, whence we may conclude that 

 the optic axes of the constituent calcific elements are directed not 

 radially but tangentially. 



The surrounding chambers have been studied by Awerinzew 

 (<?;;. cit. 1903), who states that the radial walls (pillars) behave as 

 positive uniaxial crystals arranged radiately to the centre, while 

 the concentric walls (arches and wall of annular canals) and hori- 

 zontal Avails behave as similar crvstals tano-entiallv arrano-ed : 

 he adds, however, that the subject requires further investigation. 



The actual structure of the shell is indeed far too complicated to 

 be brought under so simple a generalization, and exception may be 

 taken to the terms in which it is expressed : for, since all the evidence 

 points to calcite as the only mineral component of the walls, an 

 interpretation of their structure in terms of positive uniaxial 

 crystals would seem to be precluded. 



The walls of the chambers, external to the central disc, do not 

 exhibit a dark cross when examined between crossed nicols. Such 

 figures as are observed vary in different specimens and different 

 parts of the same specimens. In offering a description of one of 

 the fonns most commonly met with we shall suppose that a 

 horizontal section is so orientated on the stage of the microscope 

 that the axis of one of the pillars coincides with the vertical cross- 

 wire of the eyepiece. The figure which is then seen consists of a 

 vertical bar and two curved arms proceeding from it as shown in 

 the diagram (fig. 1, p. 204, and in the illustrations, PI. VII, 

 figs. 2, -1, & 6). The vertical bar begins above,^ at the junction 



^ The terms ' aboTe ' and ' below ' refer here only to the image as seen in 

 the microscope : relative to the organism they should be ' distal ' and 

 ' proximal.' or ' outer ' and ' inner.' 



