98 



of the pneumatophore of Angelidae and how this family fits in quite with the others in the order 

 Physonecta. The pneumatocyst has now its definite shape and the region situated proximally to 

 the dub-shaped enlargements ("Chitinring") constitutes the pneumatopyle or "Trichterpforte". 



The position of the young buds of appendages is of course different from that of the 

 mature ones, the former being situated on both sides of the air-funnel, the latter having a 

 position on the right side of the pneumatophore. It is clear that if ever Archangelopsis had 

 shown a porus as in Rhisophysa, this would have been on the opposite side of the bunch of 

 appendages. That is why we consider the appendages to be on the right side of the pneuma- 

 tophore. I)Ut it may be possible that those situated on the more, so to say, "apical" part are 

 young buds of future nectophores, those situated on the lower, basal side perhaps develop into 

 appendages of the siphosome. 



We hope that our conception of the structure of the pneumatophore in Angelidae w'ill 

 also remove the difficulties which Chun mentions (97a p. 76). He says that when in SipJionopIiores 

 the stem is shortened the appendages are situated in the neighbourhood of the air-funnel and 

 that the principal axis of the primary^ siphon corresponds with the same in the air-funnel. It 

 seems to us that this difficulty no longer exists. 



To continue our description of sketch VI (PI. XVH, fig. 1 30 VI) we may add that the 

 secondary ectoderm of air-funnel and pneumatosaccus have melted together. In this secondary 

 ectoderm small clefts {cav. sec.) are to be seen, which correspond exactly to those sketched by 

 Chun in PJiysophora called by him "Spaltraume im Lufttrichter". They occur both in the 

 cushion-like elongated secondary ectoderm tissue of the pneumatosaccus and in the same tissue 

 in the air-funnel. They are distinctly different from the cavities occasioned b)- the more or less 

 empty giant-cells. 



These clefts are also found again in sketch VII (PI. XVII, fig. 131 VII) and have a 

 more rounded appearance; it seems as if a porus were being formed in the secondary ectoderm. 

 In fig. VIII (PI. XVII, fig. 132) one sees yet a further communication inside, going deeper into 

 the air-funnel. It is of great importance that the cavities thus formed are always circular and 

 that in sketch VIII there are three, such cavities situated one above the other. We are fully 

 convinced that they are cavities created by the accumulation of gas in the secondary ectoderm. 

 It is a known phenomenon in other Physonecta (see Chun 97a, sketch oi Athorybia PI. I\', fig. 7). 

 In our sketches these cavities have just been halved and constitute therefore no real opening. 

 We compare with this sketch the transverse section of Angelopsis globosa given by Fewkes 

 (84 see our fig. 136 of PI. XVIII). It is now clear that the opening ("<?") of the 

 bud [g.ni.) into the float is simply such a circular cavity and that the 

 s u r r o u n tl i n g tissue which was drawn purposely distinctly clearer amidst 

 darker surroundings constitutes the cushion of s e c o n d a r )• ectoderm. 



A further proof of the great probability of this interpretation of the opening "</' is the 

 fact of our having seen, when making the same transverse section through the three specimens 

 of Archatigelopsis., exactly the same black opening "tf" its surrounding light coloured area on 

 a darker back-ground, accompanied this time by one more opening. This opening corresponds 

 to the cavity formed by the secondary ectoderm on the left in the cushion-like part (sketch VII), 



