250 MR. C. F. ROUSSELET ON THE POLYZOA [Mar. 5, 



for giving me every facility for working in his department of the 

 British Museum (Nat. Hist.). 



EXPLANATION OP^ THE PLATES. 

 Plate XII. 



'Pi"-. 1. Iinier side of the posterior part of the second and anterior part of the third 

 right lower molars of Elephas creticus. (M. 9383 a.) f iiat. size. 



2. Unworn plates of a molar of E. creticus. (M. 9379.) Nat. size. 



3. Crown view of right lower third molar of E. creticus. (M. 9381.) J nat. 



size. 



Plate XIII. 

 Fio-. 1. Crown and (1 a) side views of second lower molar of E. creticus. (M. 9378.) 

 J nat. size. 



2. Crown view of second upper molar of E. creticus. One of the plates has 



been restored. (M. 9377.) f nat. size. 



3. Crown view of the tirst and second right lower molars (or last milk-molar and 



first molar) of Elephas antiquus Falconer. (M. 9384.) f nat. size. 

 (The numbers of the specimens are those in the British Museum register.) 



3. Zoological Results o£ the Third Tanganyika Expedition, 

 conducted by Dr. W. A. Cunnington, 1904-1905.— 

 Report on the Polyzoa. By Charles F. Rotjsselet, 

 F.R.M.S. 



[Received February 2, 1907.] 

 (Plates XIV. & XV.*) 



The freshwater Polyzoa collected in Lake Tanganyika by 

 Dr. W. A. Cunnington are attached to stones and shells which 

 were partly obtained in shallow water and partly dredged from 

 20 to 40 fathoms. 



Altogether the collection contains five species, three of which 

 belono-'to the Phylactolsemata and two to the Gymnolaemata. 

 Amono-st the latter is Moove' s Arachnoidia r a?/ -la?ikester i {iOf), 

 which was found in some abundance on shells of Paramelania 

 dredged from deep water. 



Two species of the Phylactolsemata are of the Plumatella type, 

 Avith horseshoe-shaped lophophore. One of these appears to be a 

 new species, very closely adherent to stones, with half-formed 

 ea -shaped tubes, which I have named Plumatella tanganyikce. 



To the second species I have given no name, as the few frag- 

 ments of tubes and the total absence of statoblasts ofier no 

 characters that w^ould distinguish it from Plumatella repens. 



The third Phylactolajmatous specimen is an interesting new 

 species of the genus Fredericella, which I have named Fredericella 

 cunnhigtoni in honour of its discoverer, who dredged it from 

 25 fathoms near Mshale. 



The second of the Gymnolfematous species is of special interest, 

 as beinf always found associated with, and imbedded in, a fresh- 



* For explanation of the Plates, see p. 257. 



t The numbers refer to the Bibliography, pp. 256, 257. 



