770 DR. E. B. WILSON ON THE DEVELOPMENT OF RENILLA. 
What the relation of these four septa is to the axial polyp is quite unknown, owing to 
the complete lack of anatomical studies of very young Pennatulids. Through what 
process the lower end of the axis comes to lie free in the peduncular cavity is also 
unknown. But it seems highly probable, as K6ri1KER remarks (J.c., p. 270) that the 
peduncular septum of Renilla is homologous with the single horizontal septum (septum 
transversale of KOLLIKER) of the posterior part of the peduncle in other Pennatulids. 
This homology appears especially clear in the case of Renilla amethystina (VERRILL) ; 
for in this case, the extreme anterior part of the peduncle is divided as in the 
Pennatulidz into four chambers, which clearly correspond to the four longitudinal 
canals of the latter. A section through this part of the peduncle (which is situated 
near its anterior end and forms in reality a part of the disc) is very similar except in 
the lack of an axis to a section through the four chambers of Pennatula (see fig. 72, 
plate 8, K6LLIKER). The two additional chambers of Renilla amethystina are 
laterally placed, and are developed apparently as a pair of cavities in the substance of 
the peduncular septum. The four partitions which thus arise are all continuous 
behind with the single horizontal septum. 
This comparison appears to me to be well founded, though it cannot be proved so 
without the aid of further embryological studies. If it is so, Renilla amethystina 
is a perfect connecting link, so far as the structure of the peduncle goes, between 
R. reniformis and the axis-bearing Pennatulids. 
In all the latter forms the axis, when present, is suspended by these four septa, and 
it is difficult to understand their appearance in Renilla amethystina, except on the 
supposition that in this form an axis once existed, but was subsequently lost. In 
R. reniformis, the four septa also have disappeared, leaving only the peduncular - 
septum as the representative of the septum transversale. This view is supported by 
the development of the colony which, as pointed out in section 19, indicates the 
derivation of Renilla from an axis-bearing form, resembling the Bathyptilee. 
As a matter of fact, we find the axis developed in very different degrees in the 
various genera of the Pennatulids; and in certain of the Veretilide, as Clavella or 
Cavernularia, the axis is very small] or, even in some species of the same genera, quite 
absent. Whether the rudimentary condition or total want of an axis in these forms 
is due to the gradual loss of an axis cannot be determined; but the probabilities 
certainly appear to be in favour of such a view, since these genera have a much less 
primitive structure in some other respects than some of the axis-bearing forms. We 
are perhaps able to get some idea of how the axis might be gradually lost. Since the 
axis ends at some distance from the tip of the peduncle, a certain amount of movement 
is still permitted to the latter; and the great development of the peduncular muscles 
indicates that this power of movement must be an important factor in the life of the 
organism. In Renilla the power of movement is of vital importance (see p. 784), and 
an axis would be of no conceivable use. It is easily conceivable that the power of 
movement might become of paramount importance to one of the axis-bearing forms, 
