94 THE MICROSCOPE AND ITS REVELATIONS. 



connecting the successive segments of the latter, as seen at c l . There 

 can be little doubt that this remarkable development of the canal- system 

 has reference to the unusual amount of shell-substance which is deposited 

 as an ' intermediate skeleton ' upon the layer that forms the proper walls 

 of the chambers, and which fills up with a solid ' boss 'what would other- 

 wise be the depression at the umbilicus of the spire. The substance of 

 this 'boss' is traversed by a set of straight canals, which pass directly 

 from the spiral canal beneath, towards the external surface, where 

 they open in little pits, as is shown in PI. xv., fig. 16; the umbilical boss 

 in P. crispa, however, being much smaller in proportion than it is in P. 

 craticulata. There is a group of Foraminifera to which the term 

 Nonionina is properly applicable, that is probably to be considered as a 

 sub-genus of Polystomella; agreeing with it in its general conformation, 

 and especially in the distribution of its canal system; but differing in its 

 aperture, which is here a single fissure at the inner edge of the septal 

 plane (Plate xv., fig. 19), and in the absence of the ' retral processes' of 

 the segments of the sarcode-body, the external walls of the chambers 

 being smooth. This form constitutes a transition to the ordinary 

 Nummuline type, of which Polystomella is a more aberrant modification. 

 488. The Nummuline type is most characteristically represented at 

 the present time by the genus Operculina; which is so intimately united 

 to the true Nummulite by intermediate forms, that it is not easy to sep- 

 arate the two, notwithstanding that their typical examples are widely 

 dissimilar. The former genus (Plate xvi., fig. 2) is represented on our 

 own coast by very small and feeble forms; but it attains a much higher 

 development in Tropical seas, where its diameter sometimes reaches 

 l-4th of an inch. The shell is a flattened nautiloid spire, the breadth of 

 whose earlier convolutions increases in a regular progression, but of 

 which the last convolution (in full-grown speciirfens) usually flattens 

 itself out like that of Peneroplis, so as to be very much broader than the 

 preceding. The external walls of the chambers, arching over the spaces 

 between the septa, are seen at #, #; and these are bounded at the outer 

 edge of each convolution by a peculiar band , termed the ( marginal cord.' 

 This cord, instead of being perforated by minute tubuli like those which 

 pass from the inner to the outer surface of the chamber-walls without 

 division or inosculation (Fig. 335), is traversed by a system of compara- 

 tively large inosculating passages seen in cross section at a' ' ; and these 

 form part of the canal-system to be presently described. The principal 

 cavities of the chambers are seen at c, c; while the ' alar prolongations ' 

 of those cavities over the surface of the preceding whorl are shown at 

 c', c'. The chambers are separated by the septa, d, d, d, formed of two 

 laminae of shell, one belonging to each chamber, and having spaces 

 between them in which lie the * interseptal canals,' whose general distri- 

 bution is seen in the septa marked e, e, and whose smaller branches are 

 seen irregularly divided in the septa d', d', whilst in the septum d" one 

 of the principal trunks is laid open through its whole length. At the 

 approach of each septum to the marginal cord of the preceding, is seen 

 the narrow fissure which constitutes the principal aperture of communica- 

 tion between the chambers; in most of the septa, however, there are 

 also some isolated pores (to which the lines point that radiate from e, e) 

 varying both in number and position. The interseptal canals of each 

 septum take their departure at its inner extremity from a pair of spiral 

 canals, of which one passes along each side of the marginal cord; and 

 they communicate at their outer extremity with the canal-system of the 



