312 Mr. J. J. Lister. On the Dimorphism of — [Mar. 2, 
biconvex tests without any corresponding increase in the volume of their 
contents. 
NV. perforatus, however, does appear to have a disproportionately large 
megalosphere, for the test of the microspheric form is nearly as biconvex 
as that of biarritzensis and discorbinus, yet the ratio of its volume to 
M? is not above that of the majority. The disproportion is especially 
marked in the case of the variety obesws of this species, from Beni Hassan, 
in which the corresponding megalosphere approaches the dimensions 
characteristic of the typical members of the species, although the micro- 
spheric form is much reduced. 
On the whole, the results of measurement appear to indicate, for the 
majority of the species examined, an approximately definite numerical 
proportion between the volume of the contents of the microspheric test and 
the volume of the megalosphere. WV. perforatus departs most widely from 
that proportion, and the existence of this outstanding case would render it 
probable that in the others the proportion, if it could be measured exactly, 
would be found to be only approximately uniform. 
In the account given above of the mode of origin of the megalospheric 
form of Polystomella crispa, the parent organism is described as microspheric. 
This is in accordance with a large series of observations on this species 
(13, p. 446), and I am not aware of any direct evidence showing a departure 
from the rule in Polystomella. 
In other orders of foraminifera, however, viz.,in Cornuspira, Miliolina, 
Peneroplis and Ortitolites among the Miliolide, and in Cristellaria (1, Plate 68, 
fig. 1, and 14, p. 116) among the Lagenide, it is the fact that the megalo- 
spheric form is capable of repeating itself, by giving rise to a brood of 
megalospheric young. 
It appears to me that the approximately close relation which has now 
been shown to exist between the size of the megalosphere and that of the 
microspheric form is an indication (though it cannot be considered a proof) 
that this latter mode of reproduction did not occur among the nummulites 
of the Eocene Period, as it does not occur (so far as the evidence goes) in 
their ally, Polystomella, at the present day. For if the small megalospheric 
forms of the species in which microspheric forms attain so large a size 
were thus to reproduce their like, there seems no reason why the megalo- 
spheres thus produced should be of a size proportional to the volume of the 
microspheric forms. 
In the account given in the first part of this paper of the three English 
species, it will be noticed that though the megalospheres attain very different 
