300 FAMILY NUMMULINIDA. 



diversified varieties of OrhUolites (If 1 78), they must be retained as distinct species. The cham- 

 berlets of the former are nearly as cylindrical as are those of the simple type of OrhUolites, 

 being sometimes slightly flattened against the margin of theannulus within, as shown in fig. 5, 

 and sometimes rendered hexagonal by mutual pressure, as shown in the lower part of fig. 9. 

 The chamberlets of the latter, on the other hand, are rectangular, and are generally con- 

 siderably longer in the radial direction than they are in the tangential, as is shown in figs. 

 1, 3, 14 ; the proportion of these two dimensions, however, varies greatly in different annuli, 

 and even in different parts of the same annulus, as I have shown to be the case in 

 CycJodypeus (1 495). The height of the chamberlets, that is to say, the thickness of the 

 median layer, is no less inconstant than in the simple type of Orbifolltes (^ 157); sometimes 

 being pretty uniform throughout the disk, especially in 0. Foriisii; sometimes, on the other 

 hand, increasing and sometimes diminishing from the centre to the periphery (see Plate XX, 

 figs. 1, 7, 10, 11*^).* The typical arrangement of the chamberlets in both species appears to 

 be alternate in successive rows, as in other cyclical Foraminifera ; but this arrangement, 

 shown in figs. 3 and 5, is very often departed from. When examined with a sufficient mag- 

 nifying power, the partitions between the chamberlets are seen (in well-preserved specimens) 

 to be double, each chamberlet being everywhere surrounded by its own proper walls (figs. 3, 5) ; 

 and indications of a canal-system are distinguishable in the interseptal spaces left between 

 these. Each chamberlet normally communicates by oblique passages with two chamberlets in 

 each of the adjacent annuli. These passages are shown at a b, a b, fig. 5, as they present 

 themselves in sections that traverse the median plane of 0. Maidelli ; and it is observable 

 that, like the two divergent passages which in like manner connect the columnar chambers 

 in the complex type of Orbitolltes (IT 168), those of the same pair are not on the same plane, 

 so that sometimes one of them is not brought into view by the section that displays the other. 

 1 have usually obser\ed three pairs of such passages, one above the other, between eacli 

 chamberlet and the two alternating chamberlets of the next annulus, as is shown in vertical 

 section at c, c, fig. G ; and these seem to be the six "radiating sarcodal channels" described 

 by Mr. Carter, (xxiii a, p. iiSO), of which he states that the orifices are distinguishable at the 

 margin of the test, arranged in zigzag one above the other, as in the complex type of 

 Orbitolitcs. Similar communications exist between the rectangular chambers of 0. Fortisii, 

 as represented at c, c, fig. 2, and as shown in the "cast" delineated in Plate XXII, fig. 5, 

 which seems very likely, from its geological position, to belong to this type, to which it is 

 referred by Professor Ehrenberg. I have not been able to distinguish in 0. Furti-s/i, any more 

 than in Cjjcloclijpeus, any indication of lateral communications between the adjacent cham- 

 berlets of the same annulus; and Mr. Carter seems to have been equally unsuccessful. In 

 0. Manfelli, however, such a lateral communication seems to exist close to the inner side of 

 each chamber, as shown in fig. 5 and in the lower part of fig. 8 ; and along this it may be 

 presumed that a continuous stolon of sarcode passed, like one of the annular stolons of 



* By Mr. Carter it is affirmed (xxiiio, p. 329), that tlie chamberlets of 0. Afffw^e/Ziincrease 

 considerably in height from the centre to the circumference, wliilst those of 0. Fortisii (liis 

 0. dispansa) present no sucli increase. I cannot but think, however, that the examination of a larger 

 variety of specimens would have caused him somewhat to modify this statement. 



