' 121 



conspicuous as each new segment is added, since the angle 15, b, 

 becoming recurved, verges towards the opposite portion of the corre- 

 sponding segments at c, giving an increasingly reniform contour to 

 the shell, in which condition it is the most frequently found. This 

 process of growth will be readily understood on tracing the direc- 

 tion followed by the successive segments in fig. 16, which represents 

 a specimen in which the process just described has gone on to such 

 an extent, that not only have some of the segments met at a, but 

 three or four of the outer ones are even completely cycloid, closely 

 resembling those of 0. complanata and the allied species from 

 Tonga. Here we see a mode of growth in which all the leading fea- 

 tures are identical with what we observed in 0. complanata (fig. 11) ; 

 only in the present example, instead of the convolutions being from 

 the first spread out in one uniform plane, they are primarily embracing. 

 In obedience to a general law which prevails amongst the Foramini- 

 fera, as illustrated by the genera Spirolina, Cristellaria, and others, viz., 

 that in their advancing growths they pass from a more to a less com- 

 plex type, we find that the cells of the later and more external con- 

 volutions of 0. adunca become spread out in one plane : towards 

 the exterior of this species they are quite as much so as in the disks 

 from Tonga. Thus we see that the cycloid form, which in the latter 

 example is exhibited almost from the first, and which soon becomes 

 the normal condition of 0. complanata, is only attained in the case 

 of 0. adunca at a very advanced stage of growth : still, it is attained ; 

 the type is the same ; the difference in the rate and extent of their 

 development is merely one of degree. 



On examining the internal structure of 0. adunca, the existence of 

 this affinity is thoroughly confirmed ; but in comparing it with the 

 other disks we must not select our illustration from the convoluted 

 portion, though such a selection would bring us to the same conclu- 

 sion. We must take a part of the organism where the other con- 

 ditions are the same, and where the new segments are spread out in 

 an uniform plane : fig. 17 represents a vertical section of such a por- 

 tion, made in the direction of the dotted line 16, b, and traversing 

 five of the outermost segments. On comparing this with the two 

 vertical sections (figs. 9 and 12), we find that all these are constructed 

 in conformity to a common type. The section is divided by vertical 

 partitions or septa (17, a) into separate segments, which frequently 

 communicate with each other through the open canals 17, b. La- 

 teral communications also exist between various parts of the same 

 segment, through the very large orifices 17, c, which are so large 



TRANS. MIC. SOC. VOL. III. R 



