62 BRITISH DIATOMACEiE. 



conclude, as I have stated in the Introduction, that the lines arise 

 from internal structure ; that the appearance both of perforations and 

 elevations may be produced in the same object by a slight alteration 

 in the focus of the instrument ; and that such appearances are merely 

 optical illusions, produced by the reflection and refraction of the 

 rays passing through the minute cellular structure of the siliceous 

 epiderm. 



It is evident that no optical effect can confer upon circular eleva- 

 tions or perforations hexagonal outlines, while it is easy to see how 

 the latter, when out of focus, or viewed by an object-glass of imper- 

 fect definition, may assume the semblance of rounded beads or dots. 

 And, moreover, as the view which regards the striae of Pleurosigma as 

 arising from the cellular arrangement of the epiderm, harmonises the 

 structure of this extensive genus with that of the other members of 

 the family, it seems recommended to our acceptance by claims too 

 strong to be resisted. 



Nothing is known of the Reproduction, properly so called, either 

 of this or of the three preceding genera, multiplication by self-division 

 being the only mode of increase hitherto detected. While this pro- 

 cess is actively going forward, the mucus generated by the dividing 

 frustules is often so considerable as to produce the appearance and 

 effect of a distinct frond, which assumes the form of a thin pellicle of 

 some little tenacity. At other times, when the mucous secretion does 

 not assume the continuity of a pellicle, it invests the individual frus- 

 tule with a transparent envelope, which has the appearance of an ex- 

 terior membrane, and has been sometimes mistaken for such. On 

 one occasion I also met with the frustules of P. Hippocampus enclosed 

 in mucous or gelatinous tubes, precisely like those of a Colletonema ; 

 but these conditions must be regarded, for the present at least, as 

 temporary or accidental, and cannot be admitted into the specific or 

 generic descriptions. 



Owing to the compression of the valves, the scanty development 

 of the connecting-membrane, and the elongate form of the frustule in 

 the present genus, the front view is either of a linear or linear-lanceo- 

 late form ; two examples are given in PI. XXIII. 203 & 207 ; hut 

 I have omitted a special notice in the case of each species, as I have 

 not found this feature of much assistance in their discrimination. On 

 the other hand, as the colour of the dry valve is often conspicuous, 

 and seems to depend on the structural character of the silicified mem- 



