114 



(5, c), which corresponds with the slightly raised line marking the 

 direction of the septa on the exterior of the shell. 



At the dilated external margin of each convolution (4, e e), the or- 

 dinary pseudopodian tubes suddenly disappear, giving place to a fan- 

 shaped arrangement of larger radiating passages, wliich communicate 

 between the outer angle of the cell and the exterior of the organism, 

 and which occupy nearly half a circle. These tubes do not proceed 

 at once to the margin, though such is their general direction ; but 

 under a high power, we find that they communicate with branches 

 springing from them at right angles, and running in a direction 

 parallel with the external outline of the organism. The orifices of 

 these tubes are seen at 4, e e, divided transversely, when a magnify- 

 ing power of 300 linear diameters is employed. Their direction will 

 be still better comprehended from fig. 6, which represents a horizon- 

 tal section of portions of two convolutions, with their transverse 

 double septa. Those in the inner convolution (6, a) exhibit the ordi- 

 nary appearance of these septa; whilst those of the outer one 

 (6, b) exhibit an irregular variation in their distribution, which 

 is not uncommon. In each of the spiral parietes (6, c and d) sepa- 

 rating contiguous convolutions, we see at least one continuous tube, 

 which sends off branches between the two layers of each septum. 

 In the external margin of the outer convolution (6, e) corresponding 

 with 4, e, in the vertical section, we see that these circumferential 

 tubes or canals are more numerous. They anastomose freely with 

 each other, and communicate with large ones (fig. 6, /), crossing 

 them at right angles ; the latter being the same as those which pro- 

 duce the fan-like radiations (fig. 4, c c'). Many of the latter can be 

 distinctly traced between the two layers forming the septa : where 

 this cannot be done, in the case of the layer radiating canals, smaller 

 branches given off from the anastomosing net- work are seen to be 

 thus distributed. 



Here we have evidently an analogous structure to that already seen 

 in Amphistegina Antillarum, and doubtless serving the same pur- 

 pose, whatever it may be. In this instance, however, we have, in the 

 radiating canals, more definite evidence that one object of the struc- 

 ture is to bring the surrounding water into closer connexion with 

 the interior of the animal than would otherwise be the case. The 

 radiating canals (4, e e) communicate directly between the exterior 

 and interior of the cell. Whether or not any pseudopodia have 

 escaped at these marginal apertures is doubtful, but I should be dis- 

 posed to question the probability of their having followed the indi- 



