244 Transactions of tJic [Sess. 



Scliizonemfe and otlier frondose forms was extravasated. Sclileiden, 

 on the other hand, regarded the longitudinal band as a cleft, but 

 the median and terminal nodules as thickened areas of siliceous 

 matter. This conception has also been held by Siebold and Nilgeli, 

 who regarded the nodules as minute monticules or eminences ; 

 but it was rejected by Professor Bailey, who found by dissolving 

 the siliceous substance of the valves with hydrofluoric acid — the 

 objective of his microscope being protected by a slip of mica fixed 

 by means of Canada balsam — that the nodules and longitudinal 

 band were the last to disappear, and that they must accordingly be 

 looked upon as tlie thickest regions. 



The true nature of the strife, as well as of the more clearly 

 defined bands and valve nodules, was disputed by the earlier 

 observers, as well as by those of the present day. By some they 

 were regarded as dejsressions, by others as elevations — various 

 arguments having been adduced in support of both hypotheses. 



Structures in the Diatom valve called " vittte " were believed 

 by Kiitzing to possess a very sj)ecial function, and were employed 

 by him in naming a subsection of the group — the " Vittat^." 

 Smith, however, held that these appearances were not special 

 organs, but mere modifications in the outline of the valve, which 

 was simply inflected at the positions in question. The " canali- 

 culi," so often referred to by Professor W. Smith, were, on the 

 other hand, regarded by him as inter - lamellar channels or 

 hollows intervening between the siliceous coat and the internal 

 membrane, and formed by undulating flexures of the epidermal 

 envelope, their function being supposed to be the conveyance of 

 the nutriment which was believed to enter from without — through 

 pores existing along the line of suture — to the surface of that 

 membrane. 



A considerable amount of discussion has recently taken place 

 on the question of the value of the external markings of Diatom 

 valves, from a systematic point of view. Ehrenberg regarded the 

 number of striae or costte as constant in a given space on the 

 valves of members of the same species ; while Smith maintained 

 that the strias were constant in frustules originating from tlie same 

 embryo, though they might vary in those from different embryos. 

 Castracane,^ by the ajjplication of photo-micrography to this ques- 

 tion, concludes generally (1) that the number of strise is constant 

 in frustules of the same species, though these are of different 

 size and outline ; (2) that the striae are not always of absolute 

 specific value, being in a few cases irregular, but that they are 

 so when regularly disposed ; (3) that in a series of valves origin- 

 ating from different sporangial frustules belonging to the same 



' ' Atti. Accad. Pontif. Nuovi Lincei,' vol. xxxi., ser. vi. : May 26, 1878. 



