MORPHOLOGY. 3 



tinct somatic cavity) for the distribution of nutriment — the jjoli/pite ; and, lastly, organs for 

 jn'chension — the tentacuJa. Furthermore, at particular seasons, tubercular elevations are 

 developed, which contain either an ovum or spermatozoa, and are the reproductive organs. 



A polypite and reproductive organs are, in fact, the sole essential constituents of any 

 Hydrozoon, but, so far as I know, no member of this group has yet been discovered of so 

 simple a composition. Organs of prehension and of fi.xatinn, or of flotation at least, are 

 always superadded, and, in the majority, there is more than one polypite. But when this is 

 the case it becomes necessary to distinguish between the polypites and the common trunk on 

 which they are supported. To the latter. Professor Allman's term of ccr.tiosarc is very use- 

 fully applicable ; and it will be found convenient, in treating of these more complex forms, to 

 speak of the hydrosoma as composed of a coenosarc and appendages, the latter being those 

 specially modified parts of the hydrosoma which subserve the functions of support, locomo- 

 tion, alimentation, and so forth. 



I will now proceed to point out the principal modifications vi'hich are undergone, first by 

 the coenosarc, and next by the appendages, throughout the Ilydrozoa. 



The Coenosarc. 



The coenosarc of the Cori/nida and Scrlalariada; has the form of a branching stem, 

 resembling that of a plant, and presenting almost as many diversities in form and habit. It 

 may be slender and creeping, or twining; or it may simulate a tree, with stout, erect trunk, 

 and multitudinous branches, arranged according to a definite pattern ; or its ramifications may 

 run into a sort of fleshy expansion, as in Hijdracilnia. 



In these orders, especially in those forms which possess an erect and branching stem, 

 the coenosarc is usually strengthened by the development of a strong cuticular layer upon the 

 exterior of its ectoderm. This structureless, or at most laminated, cuticular substance, may 

 remain in close contiguity with the rest of the ectoderm, or, as in the Campamdarim, may 

 become separated by a greater or less interval. In the latter case, it seems at first sight as if 

 the wall of the camosarc were composed of three membranes instead of two ; but the 

 examination of young organs will clearly show that the outermost or cuticular layer is nothing 

 but an excretion from, or metamorphosis of the outermost substance of, the ectoderm. 



In the Calycophoridce, PJii/sophoridce, Lucernariadce, and Mcdirsida, no such thick and 

 hard cuticular layer is developed, and, consequently, the coenosarc remains, throughout life, soft 

 and flexible. In the two former orders it is never tree-like, and when it gives ofi" branches 

 they are exceedingly short. In some few of the Lucernariadce, on the other hand [Rldzos- 

 toiiiida), the coenosarc is regularly branched, but, nevertheless, is extremely diff"erent from that 

 of the SertnJariadce. 



The coenosarc of the C all/ cop li or idee is slender 'and filiform; that of \S\& Fhysophorida 

 varies from extreme slenderness and elongation to a spheroidal or discoidal shape. 



In both these orders it is excessively contractile, a property which it owes to the 

 abundant muscular fibres developed in its walls, principally in the ectoderm. So far as I have 

 observed, these fibres are always disposed longitudinally (except perhaps in Slcphanomia) ; but 

 other investigators describe transverse fibres. 



