424, REPORT—1863. 
thread-cells, while the body has become much elongated and attenuated, 
and admits of being spontaneously thrown into conyolutions twisted spirally 
in a plane at right angles to the common basis of the colony. This form 
occurs only at the extreme edge of the colony, close to the orifice of the uni- 
valve shell on which the hydroid has fixed itself. 
Of the above four forms of zooids, it may however be doubted whether the 
long attenuated spiral polypites be not merely, notwithstanding their con- 
stancy, an abnormal departure from the normally developed polypite, due to 
the unfavourable conditions of constant attrition and other disturbances to 
which the colony is exposed at the extreme margin of the shell. I have in- 
deed, at a little distance within the margin, met with forms intermediate be- 
tween these and the ordinary polypites, being spirally twisted like the former, 
but having developed tentacles like the latter. Agassiz regards the spiral 
form as a modified reproductive polypite or gonoblastidium, but I cannot 
find that he has ever seen it bearing gonophores, while the intermediate form 
just mentioned seems to decide in favour of its belonging to the nutritive 
rather than to the reproductive zooids of the colony. According to Agassiz, 
the spiral polypites are occasionally branched in his H. polyclina*. 
Another remarkable example of tetramorphism is afforded by Plunularia 
and the allied genera, in all of which, besides polypite, gonoblastidium, and 
gonophore, we find entering into the composition of the colony the curious 
zooids to which the name of “nematophore” has been given, and which are 
described above. Here, indeed, heteromerism, as distinguished from mere 
polymerism,would seem to attain its maximum, the nematophores being formed 
upon a type which we have been hitherto in the habit of regarding as exclu- 
sively belonging to the Rutzoponpa. 
It will be at once perceived that the phenomena of heteromerism just de- 
scribed must be distinguished from the regular genetic succession of polymor- 
phic terms involved in the conception of “ alternation of generations; ” for 
any one of the “terms” which combine to form a “ period” (individual) in 
the life of the species (see above, p. 417) may be itself composed of hetero- 
morphic zooids, without thereby altering the particular type of alternation, 
whether binary, ternary, or quaternary. Thus, while in Hydractinia the 
type of alternation is ternary, the type of heteromerism is quaternary. 
I have already (see above, p. 359) referred to the impossibility of forming 
natural groups in any arrangement of the animals which form the subject of 
the present Report, without embracing the entire “ hydrosoma;” and I have 
endeavoured to show that neither trophosome nor gonosome can of itself, as 
long as we are ignorant of the other, afford characters sufficient for the defi- 
nition of any classificatory group. 
As long as we admit the axiom that the species is composed of individuals, 
itis manifest that, unless the individual be known to us, we have not data for 
the definition of the species. Now for the conception of individual in its 
biological sense (the only one which in this question we have anything to do 
with) absolute organic continuity is not necessary, for the individual may be 
itself, as we now know, composed of zooids which are equally parts of it, whe- 
ther they all remain in continuity with one another or become separate and 
* Wright, who was the first to call attention to the spiral polypites of Hydractinia, also 
describes (Edin. Phil. Journ.) what he views as a distinct form of zooid, in the shape of 
long tapering filaments, occupying, like the spiral polypites, the extreme margin of the 
shell. ‘These filaments, however, are certainly not constant, nor are they even usually 
present ; and I do not hesitate to regard them as an abnormal state of the ordinary poly- 
pite, due to the unfavourable conditions to which it has become exposed. 
