TWO GIGANTIC TYPES OE ARENACEOUS EORAMINIEERA. 
743 
the two genera already mentioned, Alveolina and Fusulina, its structural relations may 
be best understood by reference to the lower members of the Rotalian series, such as 
Planorbulina or JDiscorbina, or to the still simpler type Involutina. This relationship, 
though at first sight it may appear far-fetched, becomes obvious upon a comparison of 
their various sections. It is only necessary to imagine one of the simple Rotalians 
thickened and drawn out at the umbilici, so that what was before a convex disk should 
become a cylinder tapering more or less towards the ends ; and the analogy is at once 
apparent. The transverse section of a body so constructed would correspond to the 
horizontal section of the original disk ; and the ideal diagram B would represent equally 
well the arrangement of the principal chambers in the Rotalian types and in Loftusia, 
as seen in section. At the same time, the lines indicating the septa have in some 
respects a different significance in different cases. In theory the test of Loftusia may 
be said to consist primarily of a continuous lamina coiled upon itself, like a scroll con- 
stricted at the ends. The space enclosed by this ‘primary lamina’ is divided into 
chambers by longitudinal septa. The septa are of ‘ secondary ’ growth ; that is to say, 
they are not continuous with the principal wall or ‘ spiral lamina,’ but are rather offshoots 
from it. The chambers separated by the septa are long and very narrow, and extend 
from one end of the shell to the other. The septa are not perpendicular to the ‘ spiral 
lamina ’ as in Alveolina , but very oblique ; and they often take also a more or less oblique 
direction longitudinally. The longitudinal section is in this way somewhat confused, 
and less to be relied upon than the transverse, in its bearing upon the form and relation of 
the various parts. The chambers are further divided by numerous irregular extensions of 
the secondary or septal system, which it may be convenient to regard as subordinate to the 
rest. These ‘tertiary’ ingrowths are generally at right angles to the septa, or nearly 
so. — The diagrams A, B, and C will make somewhat clearer the general plan of 
structure ; but it must be borne in mind that they are purely ideal, and drawn without 
reference to scale. The first of the three, A, may be supposed to represent a transverse 
section of the spiral lamina, or ‘primary’ wall; B shows, in addition, the ‘secondary’ 
system to which the septa belong ; whilst C indicates roughly the subdivision of chambers 
by the ‘ tertiary’ ingrowths. 
31. We must now consider these various structures seriatim , in relation to the appear- 
