REPORT ON THE DEEP-SEA KERATOSA. 25 



genera is so widely different, and partly so insignificant, that they may belong to very 

 different groups. 



Arenaceous shells of cylindrical or urceolate form, with a simple mouth-opening at 

 the distal end, occur in very different classes of the animal kingdom, viz.: — 1. Fora- 

 minifera (Perforata as well as Imperforata) ; 2. Physemaria (Prophysema, Gastro- 

 physema); 3. Spongias (Ammoconidaa) ; 4. Hydroida (Atractylis, Perigonymus, &c); 

 5. Anthozoa (Cerianthus, &c.) ; 6. Rotatoria (Melicerta) ; 7. Gephyrea ; 8. Annelida 

 (Oligochaeta and Polychasta) ; 9. Insecta (larvse of Phryganidae, &c). In all these cases 

 the determination of the group is difficult, or even impossible, when only the shell is 

 known, and not the animal producing it. Sometimes the recognition of tbe shell is 

 possible by comparison, or by means of secondary circumstances. But in other cases it 

 is quite impossible. 



The majority of the gigantic deep-sea Foraminifera described by Brady and others 

 are Imperforata, and possess a solid arenaceous shell ; these are therefore not sponges. 

 But a number of arenaceous genera are Perforata, and there may be true sponges among 

 them. It is possible (or even probable) that many arenaceous tubes regarded hitherto 

 as Rhabdamminidae are indeed Ammoconidae. Brady himself rightly calls many of his 

 Astrorhizidae doubtful oro-anisms, of which it is difficult to determine the zoological 

 origin and position. Indeed, his Sagcnella is so similar to our Ammoconia, his Rhiz- 

 ammina to our Ammosolenia, and his Rhabdammina to our Ammolynthus, that they 

 may be easily confounded. If w T e assume that, in the well-known calcareous Asconidae 

 (Caleolynthus, Leucosolenia, Auloplegma), the calcareous spicules are replaced by 

 xenophya (or by foreign skeletal bodies taken from the sea-bottom), we should have 

 the Ammoconidae figured in PI. VIII. — Ammolynthus (figs. 1, 2), Ammosolenia (fig. 3), 

 Ammoconia (figs. 4, 5). 



Ammoconidae and Physemaria (Ammolynthus and Haliphysema). — A new light is 

 thrown by the Ammoconidae upon those interesting primitive Metazoa which I described 

 in 1876 as Physemaria (Haliphysema and Gastrophysema). I had observed two of 

 these organisms in the Mediterranean in the living condition, and bearing eggs 

 (Haliphysema primordiale in Corsica. 1875, and Gastrophysema dithalamium in 

 Smyrna, 1873). The structures which I found in the walls of these remarkable animals 

 are essentially the same as in the Ammoconidae collected by the Challenger. The only 

 important difference is that the thin wall of the tubular body is apparently solid and 

 imperforate in the Physemaria, porous and perforate in the Ammoconidae. 



This difference may be explained in two ways. The body-wall of the Physemaria 

 may be indeed imperforate, and in this case they retain the primordial position on the 

 lowermost step of the Metazoa, which I had assigned to them, as " Gastraeadae of the 

 present time." On the other hand, it may be that the body-wall is perforated by numerous 

 microscopical pores, and that these were closed temporarily and accidentally during the 



(zool. CHALL. EXP. — paet lxxxii. — 1889.) Nnnn 4 



