166 Mr. H. J. Carter on the Form and Structure 



We now come to the structure of the shells, to whieh^ of all 

 others^ both in description and illustrations, Dr. Carpenter ap- 

 pears to me to have contributed most. 



MM. Joly and Leymerie seem to have gone no further, than 

 to have shown, that in fossil Nummulites exist hemispherical 

 granulations or little circular depressions, corresponding to gra- 

 nulations both on the external and internal surface of the shell, 

 and that these are nothing more than perforations with which 

 the shell was pierced during the existence of the animal. Also, 

 that there existed a semilunar hole in each septum arching over 

 the margin of the preceding whorl, and that the rest of the par- 

 tition was imperforate. (Mem. sur les Nummulites, Sect. B. 

 p. 20.) 



Dr. Carpenter, however, whose investigations were carried on 

 independently of those of MM. Joly and Leymerie, has gone 

 much further than this, and therefore it will be as well to give a 

 short summary of all that he has observed. 



Commencing with the septa, he states [lac. cit.) that each con- 

 sists of two layers, by which every chamber has its own proper 

 wall, and that the intervening portion, which he terms the ^' in- 

 ierseptal space/' " must have been vacant in the recent shell, un- 

 less occupied by the soft parts of the animal itself ;" — " that each 

 septum is perforated by an aperture, close to its junction with 

 the margin of the preceding whorl " (as he believes was first ob- 

 served by D'Orbigny, and figured first by Mr. Sowerby) ; and, 

 '^ that these perforations pass through both layers of each sep- 

 tum, so as to establish a free communication between one cham- 

 ber and another.^' That this case is difierent, however, "with 

 regard to certain more minute apertures, which may be seen by 

 a careful examination, under a sufficient magnifying power, to 

 exist on the surface of every septum, though not consistent either 

 in number or position ;" " they penetrate that layer only of the 

 septum on whose surface they open,'^ " and establish a commu- 

 nication between each chamber and the adjoining interseptal 

 spaces. ^^ " Other apertures of the same kind may be generally 

 traced, on careful examination, in the walls of the chambers that 

 form the surface of the whorl ; and these too appear to commu- 

 nicate with the interseptal spaces by channels burrowed into 

 those walls.^^ 



" Thus the cavity of each chamber communicates with that of 

 the one before and behind it in the same whorl, by the large 

 aperture first mentioned, which frequently appears as if made up 

 by the coalescence of a number of smaller perforations (fig. 7 6), 

 suggesting the idea that the animal substance which originally 

 passed through it was not a single large canal, but was com- 

 posed of a bundle of minute tubes or threads. This idea is 



