304 DROMALIA ALEXANDRI. 



are wanting, and none were found in the bottle. The specimens now are color- 

 less, as might be expected after twenty years in alcohol. Fortunately their 

 histological condition is fairly good, and several are so perfect and so little dis- 

 torted that the more important external features are not only clearly visible, 

 but easily reproduced in photographs. 



In general aspect they all resemble Haeckel's ('88b, pi. 3, fig. 13, 14) figures 

 of Rhodalia miranda, except for external structure of the aurophore, and the 

 absence of nectophores. There is a large pneumatophore, a distinct nectosome 

 of considerable length, and, underlying the latter, an expanded bulbous sipho- 

 some (Plate 23, fig. 6, 7). 



Pneumatophore. An important systematic characteristic is afforded by 

 the fact that the pneumatophore, instead of being rounded and smooth, is some- 

 what flattened apically and bears 8-11 triangular gelatinous prominences on 

 its outer rim (Plate 23, fig. 8). These, though variable in number and size, 

 are present in all the specimens. So far as I am aware no such development of 

 the outer wall of the pneumatophore is known in any other Siphonophore. 

 It suggests the similar extreme gelatinous development in certain Medusae 

 (Halicreasidae) and would of itself be of sufficient importance to warrant the 

 establishment of a new genus. I may add that the present collection reveals no 

 prominences either in Angelopsis or in Archangelopsis. The pneumatosaccus has 

 no open porus. 



Aurophore. The most interesting external feature of the pneumatophore 

 is of course the peculiar structure named by Haeckel the aurophore. This 

 name has been abandoned by Lens and Van Riemsdijk ( : 08) on the ground 

 that the organ in question is not the peculiar medusoid Haeckel supposed, but 

 in reality something entirely different. However, it is convenient to retain the 

 term for the structure in question, though abandoning Haeckel's explanation, 

 because no single inclusive name, other than aurophore, seems to have been 

 applied to it as a whole. 



The aurophore lies on the surface of the pneumatophore just above the 

 junction of that structure with the nectosome (Plate 23, fig. 6, 7) in the same 

 position occupied by it in other members of the family. Externally it may be 

 described as a sac, with numerous papilliform appendages about 2 mm. long 

 on its surface. In all the specimens these papillae were well developed, but I 

 was unable to find any evidence of the interpolation of new ones. 



Fortunately the material was sufficiently well preserved to allow a study of 

 the internal structure of the aurophore, from serial sections, both radial and 



