9S 
chiefly present on the basal half of each corallite, single or in groups. 
Fistula? present in the rejuvenescent corallites only. Fuga? fine. Calice 
deep and large, nearly straiglit-walled within. Septal laminae stout and 
strong, sixty-five to eighty, barely differentiated into primary and secondary; 
spines practically of one length, short, stout, occasionally laterally curved, 
but not denticulated. Tabulae highly developed, at very irregular distances 
apart, sometimes much crowded together, complete and incomplete ; complete 
tabulae are horizontal or oblique, but more often rolling; incomplete tabulae 
forming vesicles, either short and deep, and extending but a short distance 
from the wall, or elongately lenticular, and reaching nearly across the 
entire corallite ; longitudinal spines absent. Gemmation compound-calicinal- 
rejuvenoscent, copious, marginal or intra-marginal, and apparently only taking 
place in the gerontic calyx; buds at least thirteen, laterally united or 
connected by fistula?. 
Observations . — The corallites of T. princeps vary from cylindrical to 
cvlindro-ob-conical, i.e., a form varying between cylindrical and an elongate 
reversed cone (PI. XVII, fig. 1; PI. XX, fig. 1). The corallites are straight 
or gradually curved, and, in the latter condition, closely resemble in external 
appearance the large Campophyllum giganteum, Miclielin, of the European 
Carboniferous Limestone. The upward expansion is slow, and proceeds from 
a more or less obtusely-pointed base. The large radiciform processes at the 
bases are a very prominent feature, represented by their wart-like broken 
bases, occurring singly or in groups. At PI. XVII, fig. 1 c, will be seen a 
long process flattened against the side of the corallite and below the bud. I 
am undecided whether to regard this as a part of a very long radiciform 
process, or as a fistula; if of the latter nature it is the only instance in the 
mature state of this species. 
The septal lamina? are very numerous, and, so far as my sections 
reveal, do not appear to vary materially in number in different parts of the 
corallum. The septal spines are strongly developed, and sometimes laterally 
curved. The separation into primary and secondary is very arbitrary ; in 
some instances (PI. XVII, fig. 5) it is clear enough, in others it is quite lost, or 
it may appear in one portion of a corallite, and not in another ; this separation 
seems to be influenced by the amount of stereoplasmic thickening present. 
As a rule, there is one cycle of spines in each old. visceral chamber, or inter- 
tabular space ; but, in rare instances, where the tabula? are distant from one 
another, two, or even three cycles have been observed. 
