22 
longitudiTial and transverse sections; the radiating arrangement of con- 
stituent fibres in the septa is seen in mirror image in the “ costae.”^ 
A consideration of the soft-parts that laid down the skeleton elements 
of Calapoecia is bound to be speculative, but it has interest for the zoologist. 
If the ends of the “ costae ” of a single corallite in transverse section are 
joined by an imaginary line, a polygonal shape is traced. The polygonal 
outlines thus formed about neighbouring corallites are in contact. Conse- 
quently, since the “ costae ” are presumably secreted by the same polyp 
that was responsible for septa with which they are continuous, the material 
deposited in the “ costal ” zone (that is, outside the stereozone) is not the 
true “ Common substance ” usually known as coenenchyme.- It is, rather, 
a structure laid down by the extra-stereozonal portion of the base of the 
polyp, which would be in continuity with that part of the base lying within 
the calyx. At a given moment these two zones of the base of the soft 
part would lie on the same horizontal level, passing through the notches 
between septa and “ costae,” but, when these notches w'ere closed by further 
secretion to give pores, the rim of the stereozone would mark the division 
between the outer and inner portions of the secreting polypal base. This 
w'ould account for tabulae sometimes passing straight through a pore. The 
laying down of incomplete tabulae over the mouth of a pore, inside the 
stereozone, and between “ costae ” outside it, is presumably evidence of the 
normal withdrawal of the base of the polyp to a higher level. It also proves, 
incidentally, that there was no communication between corallite and 
coenenchyme after the pore had been closed by secretion at the top. 
This attribution of the structures laid down in the “ costal ” zone to the 
peripheral portion of the polyp’s base is further borne out by the extreme 
similarity between the tabular elements of this zone and those of the 
corallite itself. The fact that no junction is seen between the “ coenen- 
chyme ” tabul® of one polyp and those of its neighbour is probably an ex- 
pression of the same state of close association between polyps that must be 
deduced from the study of C. canadensis. Here, more frequently than in 
var. anticostiensis, tabulse are seen to pass through a pore straight from one 
corallite to the next. In C. canadensis the pores are not closed by incom- 
plete tabulse; nevertheless, the possibility of communication between the 
soft parts of corallite and corallite through the completed pores is ruled out 
here, as in var. anticostiensis, by the formation of solid tabulse above these 
pores. 
In the descriptions this extra-corallite material is called coenenchyme, 
because its origin is surmised from deduction, and its appearance and situ- 
ation compare v/ith the general acceptance of this term in palseontology. 
^ Miss D. Hill, whose work reviewing the terminology of Eugose Corals will shortly 
be going to press, has examined the previous usages of the term ‘costa’ and finds them 
to be so diverse as to make it advisable that the word should no longer be used. She has 
suggested no alternative, and as it is not within the scope of the present work to do so, 
1 propose to use the term here on the understanding that no more is read into it than 
the description implies. 
2 Hickson (1924, 28) has written of “the oommoTi saiibstance which supports” the 
calices in Lophohelia polifera (Recent), which he calls “coenosteum”. ‘^fany 'WTiters on 
corals use the term ‘ coenenchyme.’ This is etymologically and historically inaccurate. 
The word coenenchyme was introduced by Milne-Edwards and Haime for the fleshy 
substance between the polyps in Alcyonaria”. This exemplifies the difficulty, often occur- 
ring. of equating the zoological and palaeontological points of view in a study of corals. 
