312 ANGELOPSIS DILATA. 



suggestion (p. 309) that the genus may be characterized by the presence of 

 tentilla is based solely on Haeckel's ('88b) statement that the tentacles resemble 

 those of Rhodalia, where such organs occur. 



The gonodendron, in the example studied (Plate 21, fig. 8), is borne on a 

 gelatinous stalk, and has two main branches. On one of these the numerous 

 buds are evidently all female, but on the other the buds are too young for 

 their sex to be determined. On each branch there are also palpons in various 

 stages of development. 



Aurophore. Unfortunately the aurophore is somewhat damaged. But 

 although its cell layers are entirely destroyed in places, they can be traced 

 sufficiently to show that, with the important exception of the absence of papil- 

 lae on the outer wall (i. e. the evaginated portion of the pneumatocodon), it 

 agrees very closely in its structure with the aurophore of Dromalia (p. 304). 

 In Angelopsis, as in Dromalia, the evagination of the double wall of the pneuma- 

 tophore which forms the aurophore is more extensive than it is in Archange- 

 lopsis, with the result that the pneumatochone, or air-funnel apparatus, is much 

 longer than in the latter genus. The cliitinous ring, in the form of an elongated 

 cylinder with the lumen filled by a solid plug of secondary ectoderm, recalls 

 the corresponding structure in Dromalia. The secondary ectoderm encloses 

 several spherical cavities; owing to the condition of the material it is doubtful 

 whether giant cells occur in the primary ectoderm. The chitinous sheath as 

 seen in median radial sections apparently encloses the secondary ectoderm 

 even at its distal end (Plate 22, fig. 2). But in sections further to one side there 

 is an opening through the chitin in its distal region, though there is no actual 

 communication between the two ectoderm masses. And although this opening 

 looks like an artificial tear in the sections, conditions in allied genera make it 

 more probable that it represents the location where the primary and secondary 

 ectoderm were joined, rather than that the two ectoderm masses are secondarilj' 

 separated by the excessive formation of chitin. 



The septa connecting pneumatocodon and pneumatosaccus in the auro- 

 phore are not so numerous as in Dromalia. In this respect Angelopsis more 

 nearly resembles ^\rchangelopsis. The cavity is in free communication on the 

 one side with the pericystic cavity, and on the other with the voluminous hypo- 

 cystic space. The inner surface of the pneumatosaccus is so much damaged 

 that it is impossible to determine whether or not a disc-like expansion of the 

 secondary ectoderm occurs here as in other genera of Rhodaliidae. And the 

 chitinous pneumatocyst is entirely destroyed except in a few places. 



