RECENT RESEARCHES IN THE DIATOMACE^. 83 



Cocconeis frequently. Carter first found that two cells 

 secrete a gelatinous envelope, open, and by a true act of 

 copulation construct a single spore, which is first globular, 

 then becomes ellipsoid, and finally separates into two longi- 

 tudinal 25ortions, each of which is an auxospore. On the 

 contrary, Smith maintained that a single cell pours out its 

 contents, and therefrom developes a single spore ; but the 

 author adds that while Carter's observations refer to Cocconeis 

 Pediculus, Smith refers to what he calls the nearly allied 

 species, Cocconeis Placentula. LiUlers agrees in this point 

 with Carter, and Dr. Pfitzer confirms their position in regard 

 to C Pediculus. The following differences are noticeable. 

 One supposes the separation has been completed within the 

 envelope, the other not till it has been thrown off". According 

 to Carter the firstling-cells turn the concave sides one to 

 the other ; according to Liiders they are parallel. Dr. Pfitzer 

 observes that the material at his command was too young to 

 enable him to decide this question. 



Gomphonemeee. 



In this are included three genera — 1. Sphenella ; 2, Gom- 

 phonema ; 3. Rhoicosphenia. They are distinguishable by 

 the fact that in the general structure of the valves they re- 

 semble the Naviculece. Like the latter, the frustules possess 

 three nodules on each side, and two median-lines divided into 

 two parts by the central nodule. Still they are unsymmetri- 

 cally constructed, the upper end being broader than the lower. 

 As the Achnanthea exhibit a want of symmetry in the axis 

 of separation, and the Cymbellece in the longitudinal axis, 

 the Gomphonetneae are unsymmetrical in the transverse axis* 

 The two last-named groups are more closely related than has 

 been hitherto believed both in respect to the structure of their 

 valves and also of their cell-contents. On the one side 

 the Cymbellece so far as they are stipitate show a distinction 

 between the upper and under ends, which is not noticeable in 

 the case of valves exposed to the action of heat or in the free- 

 living forms ; and on the other side the Gomphonemece are 

 unsymmetrical not only in the transverse axis, as was before 

 stated, but also in the longitudinal axis. In all, this feature 

 manifests itself in the structure of the primordial cell ; in 

 some, in the structure of the valve itself. In Sphenella 

 vulgaris, Kiitx., the valves are noticeably more decidedly con- 

 vex on one side than on the other ; and in other cases in 

 which the margin of the valve appears symmetrical, the 

 sculpture on the two sides of the longitudinal line is different. 

 Tuffen West's delineation of Gomphonema geminatum in 



