87 
long, pipe-like corallites, one ten inches in length, retaining much the same 
diameter throughout. It the spines on the tabulae are not to he regarded as 
a specific point in T. columnar is , these corallites might then he included in 
that species from the want of evidence to the contrary. 
Locality. — Portion 77, Parish Eurimbula, County Gordon ( C. Cullen — 
Mining and Geological Museum). 
Tryplasma. dendroidea, sp. nov. 
(Plate XIV, Fig. I; PL XV, Fig. 5; PL XVI IT, Figs. 2-6; Pi. XIX, Fig. G ; PL XXII, 
Figs. 11-15 ; PL XXIII, Fig. 9 ; PL XXIV, Fig. 6 ; PL XXVII, Figs. 3 and 4.) 
Specific Characters. — Corail uni consisting in the first instance of single 
corallites, which, by repeated quadripartite gemmation, give rise to a modified 
form of dendroid growth. Corallites long, cylindrical, erect, or slightly curved, 
with an average diameter of twelve millimetres, increasing to fifteen or 
even twenty immediately previous to budding. Epitheca ornamented with fine 
concentric line-ridges, close, with here and there a coarser one. Eistulae seldom 
developed. Septal lamellae fine, about fifty-five, primary only ; spines thorn- 
like, short, very fine. Tabulae complete, at variable distances apart, horizontal 
or convex, forming plane, solid floors on natural fracture ; vesicular tabulae 
very exceptional. Gemmation very marked and persistent, four rejuvenescent 
buds in opposite pairs, rapidly diverging from one another. 
Observations. — -The corallum of T. dendroidea assumes a more definitely 
dendroid form of growth than that of any other of our Australian species. 
It appears to consist of one (PI. XVI II, figs. 2-1) or more long corallites in 
juxtaposition but separate, seldom ever united by fisfulac, increasing at 
intervals by the usual calicinal rejuvenescent budding in bunches of four, 
giving rise to a series of young corallites free laterally, conveying to the whole 
a peculiar and modified form of dendroid growth, but how frequently this 
takes place in a corallum I have no evidence to show. 
The complete or greatest length attained by any one corallite is un- 
known, but in one instance (PI. XVII f, fig, 3) the incomplete parent stem is 
seventy-five millimetres in length. It is, however, quite clear that the 
distance between points of gemmation varied considerably, for the offspring 
is but thirty millimetres long before it commences to again reproduce. 
In consequence of the usually complete nature of the tabulae, fracture 
across the stems easily takes place, and it is not uncommon to collect the coral 
li 
