5. The _crystallome_ of _Anthemodes_, &c., similar to the athorome but with the addition of a group of bracts. [Illustration: FIG. 72.--A, _Diphyes campanulata_; B, a group of appendages (cormidium) of the same _Diphyes_. (After C. Gegenbaur.) a, Axis of the colony. m, Nectocalyx. c, Sub-umbral cavity of nectocalyx. v, Radial canals of nectocalyx. o, Orifice of nectocalyx. t, Bract. n, Siphon. g, Gonophore. i, Tentacle.] _Embryology of the Siphonophora._--The fertilized ovum gives rise to a parenchymula, with solid endoderm, which is set free as a free-swimming planula larva, in the manner already described (see HYDROZOA). The planula has its two extremities dissimilar (Bipolaria-larva). The subsequent development is slightly different according as the future cormus is headed by a pneumatophore (Physophorida, Cystophorida) or by a nectocalyx (Calycophorida). (i.) Physophorida, for example _Halistemma_ (C. Chun, HYDROZOA [1]). The planula becomes elongated and broader towards one pole, at which a pit or invagination of the ectoderm arises. Next the pit closes up to form a vesicle with a pore, and so gives rise to the pneumatophore. From the broader portion of the planula an outgrowth arises which becomes the first tentacle of the cormus. The endoderm of the planula now acquires a cavity, and at the narrower pole a mouth is formed, giving rise to the primary siphon. Thus from the original planula three appendages are, as it were, budded off, while the planula itself mostly gives rise to coenosarc, just as in some hydroids the planula is converted chiefly into hydrorhiza. (ii.) Calycophorida, for example, _Muggiaea_. The planula develops, on the whole, in a similar manner, but the ectodermal invagination arises, not at the pole of the planula, but on the side of its broader portion, and gives rise, not to a pneumatophore, but to a nectocalyx, the primary swimming bell or _protocodon_ ("_Fallschirm_") which is later thrown off and replaced by secondary swimming bells, _metacodons_, budded from the coenosarc. From a comparison of the two embryological types there can be no doubt on two points; first, that the pneumatophore and the protocodon are strictly homologous, and, therefore if the nectocalyx is comparable to the umbrella of a medusa, as seems obvious, the pneumatophore must be so too; secondly, that the coenosarcal axis arises from the ex-umbrella of the medusa and cannot be compared to a manubrium, but is strictly comparable to the "bud-spike" of a Narcomedusan. _Theories of Siphonophore Morphology._--The many theories that have been put forward as to the interpretation of the cormus and the various parts are set forth and discussed in the treatise of Y. Delage and E. Hérouard (HYDROZOA [4]) and more recently by R. Woltereck [59], and only a brief analysis can be given here. [Illustration: After C. Gegenbaur.