521. 
E THE GEELONG NATURALIST. 
the generative elements alone are produced. The group has, as 
already mentioned, a calcareous skeleton; this skeleton is 
pierced in every direction by ramifying tubes, which branch and 
join again in a complicated and irregular manner. The whole 
of the exterior of the skeleton is covered by a thin layer of living 
matter, which is connected with the deeper parts by connecting 
threads which penetrate the numerous pores scattered over the 
skeleton, and which are represented in the sketches by small 
elongated dots. : 
In some genera of this sub-order the zooids are scattered 
irregularly over the surface of the colony. In others, the gastro- 
zooids are arranged in two lines, one on each side of the line of 
gastrozooids. In others, again, the gastrozooid is surrounded 
by a circle of dactylozooids, and the small group thus produced 
is called a cyclosystem. ‘The cyclosystems are in some cases 
scattered irregularly over the colony, or in other cases arranged 
in a definite manner. 
The zooids project from small tubular pores with more or less 
definite walls, and are capable of complete retraction into these 
pores. The gastrozooids are called gastropores, and those con- 
taining the dactylozooids, dactylopores. 
In Stylasterids the gonozooids are situated in cavities called 
ampullae, which, in the two species I have described, project 
above the surface as small, rounded prominences. 
One of the sketches shows a cyclosystem of of Millepora, a 
form allied to the Stylasterids. The calcareous skeleton has 
been dissolved away, and the zooids are seen projecting from a 
mass of interlacing and branching processes, which form the 
common body. A colony would of course contain many of these 
cyclosystems. 
An examination of the sketch of Deontopora Mooraboolensis 
shows the cyclosystems arranged alternately on two sides of a 
cylindrical branch. The small sketch at the side shows a cyclo- 
system seen from above. The gastropore is in the centre, at the 
bottom of a small, cup-shaped depression; and the dactylopores 
are in grooves which run from the outer edge towards the gastro- 
pore, but are absent on the side next the branch. 
In Leptobothrus Spenceri the cyclosytems are irregularly scat- 
tered, and the dactylopores are ina complete circle round the 
gastropore, and are not placed in grooves. 
The natural size of the specimens is not much more than a 
quarter of an inch in length. ; 
The best way to find the specimens is to carry home a few 
pounds of earth. Dry it well in the sun or in the oven, and then 
ut it in a very fine sieve and move it slowly about ina tub of 
water till all the fine mud is washed away. It may be neces- 
sary to repeat the drying two or three times before all the lumps 
