ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 855 



He also points out the following objections to tlie theory of the 

 formation of coral atolls in deep waters out of the calcareous secre- 

 tions of deep-water life : (1) It is very improbable that submarine 

 eruptions ever make the large and well-defined craters, like those of 

 subaerial action, which are appealed to in order to explain the lagoon 

 feature of atolls. (2) Many coral atolls are twenty miles or more in 

 diameter, which is vastly larger than the largest of craters. (3) The 

 atolls are never circular, and the larger have the irregularities of out- 

 line or diversities of form characterizing other large islands of the 

 ocean. (4) In the actual reefs and islands of the Feejee group, all 

 the conditions, from the first stage to that of the almost completed 

 atoll, are well illustrated, one island having only a single peak of 

 rock within the lagoon not 1-lOOth of the whole area, which a little 

 more of subsidence would put beneath the waters and leave the lagoon 

 wholly free. 



Polymorphism of Alcyonaria.* — Prof. M. Marshall directs atten- 

 tion to the occurrence of tentaculato-zooids in two members of the group 

 Pennatulidfe — the first the variety of Pennatula phospJiorea, known 

 as aculeata, and the second a new species of Umbellula, U. gracilis, 

 obtained in the Faroe Channel during the ' Triton ' dredging expedition 

 in 1882. In the first case the tentacles, which vary from one to five 

 in number, are fused together to form a conical spine, strengthened 

 by very stout calcareous spicules, and projecting a considerable dis- 

 tance beyond the mouth. In the case of U. gracilis the tentacle is 

 single, and differs from that of all other pennatulid zooids in pre- 

 senting a fringe of pinnules along each side identical with those of 

 the typical polyps. The morphological importance of this unitenta- 

 cular condition is discussed at some length, the single tentacle being 

 shown to have constant anatomical relations and to correspond to 

 the single tentacle present in the young embryos of Actinia mesem- 

 hryanthemum. In conclusion, arguments are adduced against Prof. 

 Kolliker's statement that Umbellula is one of the more primitive 

 genera of Pennatulidae. 



Ciliated Groove (Siphonoglyphe) in the Stomodseum of the 

 Alcyonariaas.t — S. J. Hickson finds that in Alcyonium there is a 

 groove lined by remarkably long cilia, situated on the ventral side of 

 the stomodaeum. This groove, which has been already referred to by 

 O. and E. Hertwig, has important morphological relations in the 

 group Alcyonaria which have not been previously referred to. He 

 proposes to call it the siphonoglyphe. 



The cilia of the siphonoglyphe, as seen in a living Alcyonium, 

 moving in unison, produce a current from without inwards which 

 brings particles of food and fresh streams of water into the canal- 

 system of the colony. The cilia lining the rest of the stomodseum 

 produce currents in an opposite direction, from within outwards. 



A siphonoglyphe, varying in size and in length of the cilia, is 

 present in the same position in all the non-dimorphic Alcyonarians 



* Proc. Brit. Assoc. Adv. Sci. 1883. Cf. Nature, xxix. (1883) p. 580. 

 t Proc. Eoy. Soc, xxxv. (1883) pp. 280-1. 



