DESCRIPTION OF CHALK CORALS. 263 
placed obliquely by the side of another fully produced ; and a similarly inclined 
rudimentary structure (fig. 37+) occurred about midway in a band, the boundary 
circle resting partly on the fibres, and partly on the Echinoderm itself. So far as 
could be ascertained, young cavities formed in advance of the perfected, were not 
based upon the axis, but being secreted contemporaneously with it, were equally 
situated on the subjacent general surface. Many of the older hollows, however, 
had the bottom clearly raised above that level; and a certain amount of up- 
ward growth was therefore inferred, with successive internal additions, to adapt 
the depth to the dimensions of the abdominal organs. If this be correct, the 
obliquely placed imperfect cavities would indicate subsequent developments, 
being wholly or partly based on the axis, but whether they should be regarded 
as normal or abnormal interpolations, no opinion is offered. Their existence 
nevertheless proves, it is conceived, that the axis was invested by the polype. 
The breadth of mature bands varied from one to half a line, a slight contraction 
also frequently occurring midway between two visceral centres. The length, as 
already stated, was sometimes three lines, though occasionally jimited to one ; and 
the greatest intervals existed apparently where the divergences were fewest. Fibres 
often extended uninterruptedly from one abdominal hollow to another, and even 
further, passing around the side of those structures; but they were frequently 
shorter, especially near the cavities ; and as they became less numerous the outer 
fibres inclined inwards, to maintain a due compactness. The greatest number 
counted was ten, and the least five, allowing for local defects or injuries. So far 
as was ascertained they were united at the base; the intermediate furrows how- 
ever were deep and well-defined, while no changes dependent upon age were 
detected, nor any signs of a laminated texture. 
With respect to the systematic position of Epiphaxum, the eight mdentations, 
and the proofs of an investing living crust suggest, that it ought to be associated 
with those eight tentaculated Anthozoa, which constitute the ‘‘ Polypiers Corti- 
ciferes”” of Lamarck (Phytocorallia Octactinia of Ehrenberg). Among the known 
zoophytes of that division, the abdominal cavities are however confined to the 
cortex, even when it is evanescent, on a specimen being removed from the water. 
Other eight tentaculated Anthozoa which have the visceral receptacles within the 
fabric or axis, according to Dr. Grant’, of the coral, as the Halcyonina and Tubi- 
porina, differ so essentially in structure from Epiphaxum, that it is impossible to 
regard the fossil genus as a member of either of those families. On the con- 
‘ On the Structure and Nature of Flustre, Edinb. New Phil. Journ., no. 5. p. 116. 1827. 
2m 2 
