300 RHODAIJIDAE. 



My observations corroborate Chun's statement that there is no open porus. 

 But there is a small area at the top of the pneumatophore where the ectoderm 

 and entoderm of the pneumatocodon, and the entoderm of the pneumatosaccus 

 are confluent. Even in this region, however, the lining ectoderm of the latter 

 is as distinct as it is elsewhere. These facts point directlj' to the explanation 

 that the outer layer of ectoderm in the wall of the pneumatosaccus is the 

 primary ectoderm, and that its lining layer corresponds, as Chun believes, to 

 the secondary ectoderm 0/ Athorybia, Physophora, etc. The fact that the pri- 

 mary ectoderm is absent in the upper portion may be explained as a result 

 of the high degree of specialization of the secondary ectoderm. I may also 

 point out that Lens and Van Riemsdijk (:08), have recorded an individual 

 of Archangelopsis in which the secondary ectoderm lines the entire pneumato- 

 saccus. 



Color. At the apex of the pneumatophore the ectoderm of the pneumato- 

 saccus is densely pigmented with large granules, collected in clusters (Plate 21, 

 fig. 4). This results in a central circular area densely dotted with flecks arranged 

 somewhat radially and varying in color from brilliant red to dark brown. In 

 other portions of the colony the color is also variable, perhaps depending on 

 the state of nutrition. In some specimens the upper part of the conn is red- 

 dish in life, in others it varies from greenish to amber-yellow. The palpons 

 were either pinkish or yellowish. The siphons were pinkish in most specimens; 

 but in one large individual they were green with amber basigaster. 



Distribution. The records for this genus are from the Pacific south of the 

 Bonin Islands (Brandt), the Malaysian region ("Siboga"), the Eastern Tropical 

 Pacific ("Albatross"); Gulf Stream (Fewkes), the South Atlantic (Haeckel), 

 and the Sargasso Sea, 31, 5° N. 40, 7° W., (Chun). 



Rhodaliidae Haeckel, 1888. 



Aurmectae Haeckel, 1888. 



Auronectidae Chun, 1897. 



Angelidae Fewkes 1886, Schneider, 1898. 



The name Angelidae used by Schneider and by Lens and Van Riemsdijk to 

 replace Auronectae, the latter having been used by Haeckel under the supposi- 

 tion that the group was a distinct order, is unfortunate on nomenclatural grounds. 

 Fewkes ('8(5) supposed that Angela Lesson ('43) was closely related to his own 



