214 FAMILY GLOBIGERINIDA. 



Rotalia has been very accurately described by Prof. Williamson (cix), whose account of it is 

 confirmed in every particular by my own inquiries. When a horizontal section is taken near 

 the upper or flat surface of the shell, the whole spire is laid open, as shown in Plate XIII, 

 fig. 8 ; the finely-tubular structure of the lateral walls of its chambers, and the absence of 

 tubuli from the intervening septa, are apparent ; while a marked peculiarity here presents 

 itself in the septa being composed of two lamellae, between which there is sometimes a con- 

 siderable interspace. The intraseptal spaces are connected into a system of passages which 

 extends in a radial direction from the centre towards the periphery ; each passage first tra- 

 versing one of the intraseptal spaces, and then usually bifurcating as it enters the spiral lamina, 

 its two branches divaricating in such a manner as to pass to two septa in the next convolu- 

 tion. By comparing this, however, with such a vertical section as is shown in Plate XIII, 

 fig. 9, it is made apparent that what seem to be canals between the septa are in reality the 

 lower edges of intraseptal /55?«-e5, which intervene between the two lamellae through nearly 

 the whole area of each septal plane. From that edge of each fissure which is nearest the flat 

 surface, passages are given off which run towards that surface to open between the tubercles 

 on its septal bands ; whilst from that edge of the intraseptal fissure which abuts on the 

 oblique side of the cone, there is given forth a row of passages which open along the septal 

 band that passes vertically towards the umbilicus. Before reaching the surface these passages 

 usually divaricate, and the wedge-like space between each bifurcation (fig. 8, a) is occupied 

 by non-tubular substance, which forms, in fact, the base of one of the tubercles that rises 

 from the septal band. The exogenous deposit of which these tubercles are formed, extends 

 over the whole surface of the shell, adding considerably to the original thickness of the 

 walls of the chambers. It is shown by the vertical section (fig. 9) how the peculiar form 

 of the shell is engendered by the remarkable elongation of the later chambers between the 

 lateral surfaces, so that the upper side of each succeeding whorl is kept on a level with the 

 preceding whorl above, whilst it is produced far beyond it beloio. This production of each 

 whorl leaves a large umbilical fossa of irregular form, which is filled up with homogeneous 

 shell-substance, so as to become a sort of columella. 



373. Affinities. — The very marked participation in certain characters of the Nummuline 

 series which is exhibited by this genus, would suiBce, if taken by itself, to justify its claim to a 

 place among them ; the relationship of its most developed forms to those of PohjsiomeUa in 

 particular being so intimate, that the two cannot be justly considered as far removed from 

 each other. But, on the other hand, the general affinity of Bolnlia, to the other genera of the 

 Rotaline series is so close, that it is quite impossible to detach it from them ; and it must be 

 regarded, therefore, as the link which establishes the transition between the ordinary Rotaline 

 and the Nummuline series. It is impossible, moreover, not to be struck with the analogical 

 relationship between Roialina and Carpenteria, which hold the highest places in their respec- 

 tive series ; for Carpenteria would in all essential particulars be an outspread Rotalia attached 

 by its upper lateral surface, if its chambers communicated with one another, instead of open- 

 ing separately into a common vestibule. 



374. GeograpJdcal and Geolofjical Distribution. — The true RotaHa may be considered as 

 essentially a sub-tropical form ; its characteristic features being well exhibited in specimens 



