464 ANIMAL NATURE OF DIATOMK.E. 



existing remains, and a new one appears in the in- 

 ternal side. Does the formation of the new vitta 

 precede or follow the median deduplication ? I 

 have not made such observations as enable me to 

 decide, but it seems to me that the former is the 

 case. 



And the distribution of the internal coloured substance 

 bears a great resemblance to that of Melosirce, especially 

 in a state of desiccation. 



As to the stipes, as well in this genus as the next one, 

 I have to repeat what has been said of Cocconema. 



47. LiCMOPHORA. — Bacilli flab ellati a latere pri- 

 mario aitguste cuneati, altera latere lineares basi et apice 

 rotundati. Stipes crassus rigidus. 



The resemblance of this to the preceding genus is only 

 apparent. But a true affinity connects Licmopliora to 

 Synedra, from which it only differs by that character 

 which, though regarded as essential to the Meridiese is 

 found very inconstant in Fragillariese and reappears in 

 many species oiSurirella. The vittse, in Licmopliora, are 

 not to be compared with those in Bhipidophora. They 

 are nothing more than the usual longitudinal canals 

 projecting into the cavity, by which the apparent per- 

 forations or sections of their cavities appear very near 

 the margin of the summit. The distribution of the in- 

 ternal coloured substance is different from that in the 

 two preceding genera, and greatly resembles that of 

 Synedra. 



If we compare these with the sections Tabularia 

 and Grallatoria, we cannot at all recognise the alleged 

 affinity. 



For this reason, I think we ought to exclude the 

 Licmopliora divisa from this genus, and place it in the 

 preceding one. 



The four species that remain, [fidyens, radians, flabel- 

 lata, Meneghiniana,) though described and figured by 

 Kiitzing with so much diligence, are very difficult to 



