686 Transaetions of the Society. 



with Stephanodiscus Niagavfe Ehr. The latter is distinguished 

 from C. Carconensis by the greater number of smooth rays 

 separating the radiating puncta, part only of which proceed from 

 the marginal spines. The lines of puncta in 8. Niagara are 

 simple in the centre, but consist of two, or at most three, contiguous 

 puncta as they approach the periphery, whereas the lines of puncta 

 in C Carconensis occur in groups. 



Dr. Arnott, in the paper previously quoted (p. 246), says: 

 " Smith unfortunately referred a species (0. minutula) obtained in 

 Lough Mourne deposit to C. antiqua, a species which does not 

 occur in any of the Irish deposits which I have examined." It is 

 somewhat difficult to imagine how Professor Smith could have 

 mistaken C. mimdula for G. antiqua, the difference being so well 

 marked. I have never been able to find the latter in the Lough 

 Mourne deposit or in the marl Co. Down, but in a deposit from 

 Strangford, Co. Down, it is not uncommon, and I think finer than 

 in the Peterhead deposit. 



]\rELOSIRE^. 



Melosira Borreri Greville. Sm. S. B. D. tab. 50, fig. 330. 

 Var. moniliformis = M. moniliformis Ag. Kiitzing, Bacil. 



tab. 3, fig. 2. 

 Var. suhglobosa Grun. — Generally somewhat smaller. Frus- 

 tules nearly spherical or elongated, the ends slightly 

 flattened. 

 Var. odogona Grun. — Frustules nearly cylindrical, with flat 

 ends and broadly oblique corners, so that the frustules in 

 f. V. appear octagonal. 

 The three varieties are not uncommon in Baku Harbour. 

 The filaments are attached by a short stipes to other algae, 

 particularly Cladophora. The frustules are sometimes attached 

 to each other and sometimes connected by a short gelatinous 

 isthmus. 



The interesting var. octogona, which I for a long time thought 

 a very different species, and which I first found on Vaucheria 

 javanica Kg. from Java, had such oblique corners to the cylindrical 

 frustules that I could not recognize its relationship to M. Borreri. 

 I afterwards found examples which more nearly approached M. 

 Borreri on algae from Upolo, Austraha, and Kamtschatka, and 

 which stood between it and the typical M. Bor7'eri, from the 

 Lagunes of Venice. The vars. octogona and suhorhicularis are 

 usually more finely punctate than M. Borreri; the latter has 

 frequently the central part of the valve smooth, whilst the former 

 almost always have it punctate. The puncta in all the forms of 



