GRABAU : MONILOPORIDAE. 
423 
as is indicated by the regularity with which the calices were given 
off on all sides (PI. 2, tig. 2). In Ceratopora a different method 
of growth is indicated for each species. O. jacksoni has corailites 
with a circular cross-sectidn, indicating free growth. This is further 
shown in the type specimen by the tree-like character of the whole 
corallum, the branches diverging from the trunk on all sides, bend¬ 
ing outward and upward. The arrangement seems to have been 
such as would secure the greatest stability to a top-heavy structure. 
C. dichotoma , a type which rested on the bottom, shows a manner 
of growth which induced the flattening of the lower surface of the 
corallum. In muddy bottoms, such as probably formed the normal 
habitat of this species, the Hat resting-surface of the corallum pre¬ 
vented sinking into the mud. The calices of the adult individuals 
at the same time turned upwards, and were frequently long, so 
that the polypite was kept well above the mud on which its corallite 
rested. The normal circular section of the tube occurs in the last 
built, free projecting calices, indicating that the flattening of the 
side on which the corallite rested was primarily a dynamic feature. 
This supposition is borne out by the fact, that in C. distorta , which 
has normally a circular cross-section, with an erect growth, one side 
is occasionally flattened, the indications being, that the corallum 
rested on the flattened side (PI. 3, figs. 9 and 10). It is also 
strengthened by the peculiar form of the attached specimen referred 
to C. dichotoma (PI. 4, figs. 17, 18), in which the flattening does not 
appear in the older corallite, a subcircular cross-section, with 
irregular prolongations occurring. The similarity of this portion of 
the corallum to an attached specimen of C. distorta is very marked. 
C. distorta is intermediate in form, and probably was in habitat 
between C. jacksoni and C. dichotoma, and with sufficient material 
a series connecting the two extremes could probably be arranged. 
The origin and nature of the reticulate structure in Monilopora 
seem to be explained by the similar though less perfectly developed 
structures of this character in Ceratopora. The trabeculae of Moni¬ 
lopora are more numerous and more regularly disposed, at the same 
time being much shorter than those of Ceratopora. In both genera 
the trabeculae may be regarded as of the same nature as septa 
growing from the walls of the cysts, towards the visceral chamber. 
The addition of new lamellae over the trabeculae occurred at inter¬ 
vals, as the polyp prolonged its tube and contracted in its lower 
portion. From the newly added lamellae new trabeculae arose. 
