1849.] CARPENTER ON THE STRUCTURE OF NUMMULINA. 23 



My examinations, it will be as well to state, have been made in 

 two ways ; first, by making sections thin enough to be viewed with 

 a high power by transmitted light ; and second, by breaking the shell 

 in various modes, and examining by reflected light the parts laid open 

 by the fracture. Both of these methods are necessary, in order to 

 gain a satisfactory idea of the complicated arrangement of which I 

 shall now proceed to give an account ; since each affords information 

 which the other is unable to yield ; and the errors into which we 

 might be led by the inspection of one set of appearances, are corrected 

 by those presented by the other*. 



When we have made a horizontal division of N. IcEvigata, either by 

 section or fracture, through the medial plane {a a', fig. 4, PI. III.), a 

 cursory inspection may satisfy us that the character of its spire, and 

 the form and arrangement of its chambers (fig. 2), are very different 

 from those of the chambered shells of the Cephalopodous Mollusks. 

 The increase in the breadth of the spire in successive whorls, though 

 tolerably regular at first, becomes afterwards very slow, and seems to 

 cease altogether ; and it not unfrequently happens that a spire en- 

 larges rather suddenly, as at a, or that a large part of one spire (as 

 that marked c) is narrower than those on its interior and exterior 

 {b and d) . The division into chambers, also, is far from being re- 

 gular, the distance of the septa from each other being subject to 

 great variation even in the same whorl ; and cells much smaller than 

 the rest, and apparently abortive, being not unfrequently seen, as at 

 e, fig. 2, and at a and b, fig. 3. In specimens whose chambers have 

 been filled by calcareous infiltration, it not unfrequently happens that 

 the central portion of this is deeply tinged, in every chamber, with 

 dark brown matter, which has every appearance of being the residue 

 of some organic substance. These facts are important, as tending to 

 show that the anim.al of Nummulite must have borne a resemblance 

 to that of the existing Foraminifera, and have been composed rather 

 of an aggregate than of a single body. 



This inference is confirmed by an examination of the septa which 

 divide the chambers, under a higher magnifying power. For it then 

 appears that each of these septa is double, so that every chamber has 

 its own proper wall ; and further, that there is a space between the 

 two layers, which, being filled up with crystalline infiltration in the 

 fossil Nummulite, must have been vacant in the recent shell, unless 

 occupied by the soft parts of the animal itself. This, which may be 

 termed the interseptal space, is well seen in several of the septa whose 

 broken edges are delineated in fig. 3. That each septum is perfo- 

 rated by an aperture, close to its junction with the margin of the pre- 



* It appears to be from having followed only the second of the above methods, 

 that MM. Joly and Leymerie, who have recently been giving their attention to 

 the same subject with myself, have arrived at results which will prove, I beheve, 

 to be in some respects erroneous and in others imperfect. As my investigations 

 have been pursued quite independently of theirs, I think it better not to interrupt 

 my account of them by continual references to the memoir of MM. Joly and Ley- 

 merie (Mem. de I'Acad. des Sciences de Toulouse) ; but sh 11 confine my notice of 

 their labours to a brief indication of those points in which their results coincide 

 with my own, or in which they differ from them. 



