. 
ON THE REPRODUCTIVE SYSTEM IN THE HYDROIDA. 425 
independent organisms. Further, we know that in the hydroid individual we 
may not only have independence of zooids, but polymorphism of zooids; so 
that a knowledge of any one zooid-form may give us a totally inadequate con- 
ception of the true individual form. It is on this account that, before we can 
expect to construct classificatory groups upon a solid basis, we must deter- 
mine the whole life-formula of the individual. 
Now it must be admitted that the practical application of this principle is 
by no means free from difficulty, and that in those cases where the medusa 
detaches itself from the rest of the hydrosoma in order to lead an independent 
existence, the synthesis of the individual by the reassociation of its consti- 
tuent zooids is often far from being an easy task. 
One of the chief sources of this difficulty is found in the fact that the me- 
dusa at the time of its liberation has not yet attained its adult condition, and 
may still undergo great changes of form before arriving at maturity, while we 
necessarily lose sight of it in the interval, and can therefore, in most cases, 
only by inference identify the free-swimming adult medusa with the young 
form which we know to have originated in some definite trophosome. 
But, besides this, numerous adult medusz are known of whose trophosome 
we have not as yet any indication, though there can be scarcely a doubt that, 
like the others, the great majority of these have also their polypoid tropho- 
some. It is plain that any attempt to arrange these medusz into species and 
genera must be purely provisional—so much so indeed that we may have two 
medusz between which it would be impossible to find specific differences, and 
yet their trophosomes may be widely different from one another, while, on the 
other hand, two very different medusee may haye trophosomes specifically in- 
distinguishable. 
In opposition however to this view, it has been asserted that similar tro- 
phosomes are always associated with similar meduse. Thus Agassiz, finding 
the same form of medusze—that, namely, which has been described under the 
generic appellation of Sarsia—originating from polypoid trophosomes which 
present no differences of generic value, and are accordingly included in our 
systems under a single generic name, that of Coryne, and finding, more- 
over, that this is the case in species gathered both upon the east and west 
shores of the Atlantic and upon the North American shores of the Pacific, 
concludes “ that medusz which are generically identical arise from hydroids 
bearing identical generic relations”’*. 
A little consideration, however, will show that this view is untenable; for 
a comprehensive survey of the Hyproma has distinctly proved that the amount 
of divergence between two different trophosomes is by no means a certain 
measure of the divergence between their gonosomes. 
If we were to imagine for a moment that a phanerocodonic and an adelo- 
codonic gonophore were independent organisms, instead of being mere zooids 
associated with others in a complex hydrosoma, we should not hesitate to 
regard the differences between them as of at least generic value, and yet we 
may have the trophosomes from which these two forms of gonophore respect- 
ively spring presenting no differences which would justify us in regarding 
them as more widely separated from one another than two species of a com- 
mon genus. We need only compare the Laomedea flexuosa, for example, 
and its simple fixed sac-like gonophores, with the Laomedea geniculata and its 
free disc-shaped meduse, in order to be convinced how widely the gonosomes 
may differ, and yet the trophosomes present no difference of importance. 
* Agassiz, Nat. Hist. of the United States, vol. iy. p. 217. 
