, 109 



not have been the case had the additions been made to the interior 

 of the cell. 



These facts led me to reject the first explanation that presented 

 itself, and to conclude that by whatever instrumentality the new 

 growths were formed, they were applied to the exterior, and not to 

 the interior, of the organism. Another way of accounting for the 

 facts still remained, and though rather startling, I believe it to be 

 the true explanation. At 2, i, we find that the laminae are consoli- 

 dated into one compact mass ; but on tracing the same lamellae down 

 to the point (2, c), we find that they separate into two divergent sets, 

 some forming an inner septum (2, b, corresponding with that seen 

 near fig. 1, a), whilst the greater number constitute the outer shell- 

 wall (2, c), with the narrowed posterior apex of a cell intervening 

 between them. Thus we learn that the thickness of the parietes at 

 2, i, is at least owing to the contiguity and amalgamation of the lami- 

 nated walls of two or more cells. 



Connecting this fact with that of the exceeding thinness of the 

 cell- wall of the last-formed cell (fig. 1, h), I have come to the conclu- 

 sion that the greater thickness of the earlier and older growths of 

 the shell is owing to the circumstance, that on the addition of each 

 new segment the cell-wall by which it is enclosed has not ter- 

 minated at the boundary of the individual segment itself, but has 

 been prolonged over a considerable number, if not the whole, of the 

 segments belonging to the outermost convolution of the animal. 



But before this explanation can be received we require the admis- 

 sion of another proposition, viz., that before the newly-formed seg- 

 ment enclosed itself within the limits of its own calcareous cell, it 

 must have had the power of spreading itself out in the form of a thin 

 gelatinous layer, investing the whole of the pre-existing organism. 



Startling as I acknowledge this proposition to be, I see no other 

 way of explaining the phenomena exhibited in this interesting spe- 

 cies. In the horizontal section (fig. 1) I can easily trace the more 

 external lamellae, extending continuously along the spiral wall from 

 / to c, covering once at least four chambers, and equal to nearly one- 

 third of the entire convolution. This fact requires an extension of 

 the soft animal, the secreting agent, at least proportionate to that of 

 the result produced, and which is clearly not to be found in the cir- 

 cumscribed segments of the soft animal, so long as they are shut 

 up within their calcareous cells : I can only conclude that in order 

 to the production of such a result, the animal must have had the 

 power either of leaving its cells to a considerable extent, or of 



