POLYPI. 133 



than Sars in preserving the life of the medusoid individuals, appears 

 to have traced the phenomena of the development of the ova in them, 

 and arrived at the conclusion that they are not directly transformed 

 into the polype as Van Beneden believed to occur in the Tubularia 

 Dumortieri. The ova so developed would give origin to " planulae," 

 and these, by the metamorphosis so well described by Dalyell, would 

 produce the rooted zoophyte, from which a compound ramified group 

 might result by successive incomplete gemmations. So that with 

 regard to the generation of the marine Hydrozoa, we find these modi- 

 fications : — either the ovicapsules are developed as buds on the 

 nutrient polype which is cast off, and the young polype is afterwards 

 developed in and escapes from the ovicapsule ( Tubularia indivisa) ; 

 or the ovicapsules are transformed polypes or pinnae on different 

 and determinate parts of the compound Hydrozoon, give issue to 

 ciliated planulae, and are cast off (most Sertularia) ; or bell-shaped 

 Medusae are developed in similarly formed ovicapsules, and escape 

 {Eudendrium ramosum) ; or a small nursing polype is developed 

 from the summit of the ovicapsule which incubates the ova, and 

 falls like a flower (Campamtlaria geniculata); or Medusoids may 

 be directly produced by gemmation, as in Syncoryne, and these, 

 becoming detached, continue to move for some time (fifteen or 

 eighteen days) before they fully develope the ova, which produce the 

 planulae, which are metamorphosed into the compound zoophyte. 



The medusoid brood, which V. Beneden saw to be the produce of 

 the fixed and composite Campamdaria gelatinosa, closely resemble 

 those minute acalephae which Sars* has described as the Cytaeis 

 octopunctata, and Willf as Cytaeis polystala. With regard to the 

 special capsule in which certain Hydrozoa produce locomotive larvae 

 of the acalephoid character, although it is called " ovicapsule," from 

 its analogy to those that give issue to planulae, it does not contain 

 a proper "ovum" with its germinal vesicle, vitellus, and chorion, 

 requiring the distinct act of impregnation before development of the 

 medusoid embryo can begin. The development of such forms of 

 embryo in the ovicapsules of the Eudendrium ramosum and Cam- 

 pamdaria dichotoma seems to be the result rather of the last remnant 

 of the old spermatic force in the slightly modified cell-tissues of the 

 parent, and to be a modification of the same gemmiparous faculty as 

 that to which the previous multiplication of the nutrient form of 

 polype and the compound character of the whole was due. 



The development of the true ova and the reproduction of the 



* CXIIL p. 28. tab. vL f. 14. t CXIV. p. 68. tab. ii f. a. 



K 3 



