80 THE OCEANIC HYDROZOA. 



those of the opposite nectocalyces, and thus enclose the ccenosarc in a sort of hydroecial 

 canal. Tlie nectosac occupies nearly the whole cavity of the organ. Near its mouth it is 

 subcylindrical, but almost immediately it becomes exceedingly wide and is divided into two 

 lobes, by an inflexion corresponding with the groove in the internal face of the nectocalyx. 



On very careful examination, the ends of four delicate canals could be traced opening 

 into a circular canal which surrounded the aperture of the nectosac, and the origins of 

 these canals could be seen radiating from the end of the canal of the pedicle, but in the fully 

 formed organs I could not satisfactorily trace their intermediate portions. 



M. Vogt says of these canals, which he calls " batonnets '' (though he recognises them 

 to be canals) : " Une entoure I'iris orbiculaire de I'ouverture en servant de support a ce 

 rideau musculaire,i tandis qu'un autre se porte d'arriere en avant. Deux courbes latcrales 

 embrassent le bouton median' par laquelle la cloche est fixee en arriere, se reunissent au 

 batonnet qui court dans le ligne mediane, tandis que deux autres se dessinent sur les ailes 

 posterieures proeminentes de la cloche" (p. 42). 



I do not quite understand this description, and I will, therefore, without further 

 comment, state the results of a careful study of the development of these organs in my 

 Pliysopliora. The smallest buds measured ^Jh of an inch in length, and were simple caecal 

 processes of the ectoderm and endoderm of the ccenosarc, immediately beneath the 

 pneumatophore, with an internal cavity, whose walls were ciliated, and which was in free 

 communication with the somatic cavity (PI. VIII, fig. 10«). 



In a bud of y^tli of an inch in diameter, the apical region of the wall had developed 

 a large, internal, spheroidal prominence, and its cavity had thus become cup-shaped (fig. 10 6). 

 In buds of one eightieth of an inch in length, the cavity had become divided into four canals, 

 and a central hollow began to appear in the spheroidal prominence (fig. 10 c). 



In a bud of one sixtieth of an inch long, the central hollow, or rudimentary nectosac, had 

 acquired a very large size, but was not open at its apex ; and the four canals were united by 

 a circular canal. The whole organ had become broader in proportion to its length, principally 

 by the widening of its base. This was still more marked in young nectocalyces of one 

 fiftieth of an inch in length, in which the nectosac had a larger cavity, and proportionally 

 thinner walls. In these the lateral canals had acquired a slight double curvature, so as 

 to have the outline of an italic s (fig. 10</). 



A. nectocalyx, one twenty-fourth of an inch long, consisted of an extremely broad and 

 expanded base attached to a short pedicle, and not yet excavated internally, surmounted by a 

 short, truncated, and subcylindrical apical portion. There was no aperture in this, nor were 

 any of the characteristic crests of the adult organ developed. While the median longitudinal 

 canals remained unchanged, the lateral ones had become singularly contorted (fig. 10 e). 

 Supposing the pedicle of the organ to be posterior, and its natural superior and external face 

 superior, the lateral canals pass at first outwards, and then turn sharply inwards on tiie 

 imder face, where each forms a loop by bending as suddenly outwards again. On reaching 

 the lateral face, it curves upwards and backwards to attain the superior face, on which it 

 passes inwards, and afterwards bends at right angles forwards ; it then passes inwards again 

 and finally forwards to the circular canal. I traced the development of the organ no further 



^ The membranous valve. " The pedicle. 



