114 



in many of the different species of the minute recent concamerated 

 shells. 



In none of the multilocular shells are such differences observable, with 

 respect to the size of the siphunculi, as in this genus. In some the 

 siphunculus does not equal one tenth, whilst in others it exceeds a third, 

 and is sometimes nearly one half, of the diameter of the orthoceratite 

 itself. It generally assumes that tumid form, which a membraneous 

 part might be expected to assume, when dilated between the septa : as 

 may be seen in the large siphunculus, Plate VIII. Fig. 2, and in Fig. 7, 

 where the siphunculus is seen intersected by the septa of the orthoce- 

 ratites ; the whole being imbedded in a mass of whitish marble. 



The siphunculus, or rather the cast of the siphunculus, of this shell, 

 very often yields a striking appearance, from an obliquity in its 

 form, as in Plate VIII. Fig. 2, and Fig. 6. This appearance has 

 contributed somewhat to the degree of error which has existed, re- 

 specting the original nature of tbese bodies; since, even those who 

 might be disposed to consider them as deriving their form from the 

 siphunculus of an orthoceratites, might expect to find a corresponding 

 obliquity of form in the orthoceratites itself. But a slight investigation 

 will show, that there exists no reason for supposing, from this circum- 

 stance, that the shell to which it belonged possessed any particular 

 degree of obliquity of form. 



The septa, by which the siphunculus is intersected, being a seg- 

 ment of a hollow sphere, the divisions of the siphunculus must neces- 

 sarily vary with the direction with which the siphunculus passes through 

 them. If the siphunculus passes through the centre of the septa, and in a 

 line nearly parallel with the parieties of the tube, it will be marked by 

 segments of a hollow sphere, the sections of which are disposed in 

 nearly transverse lines, as at Plate VIII. Fig. 3. But if the siphun- 

 culus does not pass through the centre of the septa, and yet keeps a line 

 nearly parallel with the parieties of the tube, it will be marked by seg- 



