78 
The corallum is a large, sub-fasciculate and loose aggregation of long 
cylindrical corallites, either united by their lateral surfaces, or free and 
connected by fistulas, which are seldom on the same level in contiguous tubes, 
although occasionally so. These fistulse are enlarged at their junctions with 
the corallites, but contracted in their middle course, and do not appear to 
have any particular direction of curvature, although the horizontal pre- 
dominates. 
The corallites vary much in diameter, with eight milimetres as the 
approximate maximum width, and they are always wider at the offshoot of a 
fistula. Incomplete colonies only have been examined, but the corallites are 
known, even in their incomplete state, to attain six inches in length. 
In my first description of this coral, I said regular accretion rings or 
bourrelets were absent. This was, however, an error; they are present in 
some examples, but weak and inequidistant. 
The septal spines are highly developed and acicular, and the difference 
in size at once indicates the presence of primary and secondary lamellm ; 
lateral denticles are sometimes present. 
The tabulae are a very conspicuous feature in T. Lonsdalei, and are 
numerous in every corallite, the complete predominating largely over the 
incomplete. The clusters of closely packed tabulae are a noteworthy and 
interesting detail. This is one of the forms with occasional depressions on 
the upper surfaces of the tabulae, usually central, but sometimes excentric 
and then lateral ; as I have already explained, these depressions are in no 
way of the nature of a fossula. Vesicular tabulae do occur in T. Lonsdalei, 
but are small, marginal, and triangular in longitudinal section. The complete 
tabulae immediately below a gemmiferous bifurcation are distinctly sigmoidal, 
and, adopting a term from Crinoid terminology, may be termed axillary. 
Oblique floors arc not common as a rule, but one instance has come under 
notice in wdiich the whole of them in five corallites, forming the section, are 
oblique, or oblique and rolling, or oblique and convex. 
Gemmation has been observed but in few instances. The budding is 
certainly calicular, and may perhaps be parietal also ; so far as I can 
ascertain, two buds only are given off from a parent corallite, and not four as in 
T. dendroidea and T. vermiformis\ these branch obliquely from the parent, and 
present the appearance of a bifurcating stem. A specimen from the Sydney 
University collection exhibits a double bifurcation in the space of two and a 
quarter inches. It is possible that both parietal and calicinal gemmation 
