Marine Flankton from the East-Greenland Sea. 



267 



^*-^?B» 



Fig. 2. Coscinodisvus siibbuUens .lørg. An empty 

 and partl\- broken frustule. Height 85 /j.. 500 t. m. 



run parallel to its margins, one thin line near the margin and one 

 much coarser line situated more or less half way between the mar- 

 gin and the margin of the 

 valve. The first named 

 line makes a deviation 

 from the parallel in one 

 place, where it bends 

 rather abruptly towards 

 the coarse line and merges 

 into it on a very short 

 way, thus forming a V- 

 shaped figure with the 

 tip cut off; in this place 

 the line is coarser than 

 elsewhere and coarsest 

 where it disappears in the 



other line. The V-shaped places of the two connecting parts of a 

 cell never face each other, oftenest they are on the opposite halfs 

 of the girdle, but sometimes rather near each other, as e. g. shown 

 in fig. 3, to the right. The narrow part of the V is always directed 

 towards the corresponding valve. 



In two samples from August — September 1907 where Cose, siib- 

 buliens was dominant, I happened to find among the numerous nor- 

 mal cells some few auxospores or more correctly cells developed 

 from auxospores.. Figs. 3—4 represent such cases: A large cell (dia- 

 meter in a few measured spe- 

 cimens 280—320«) carries on 

 the one valvar side the folded 

 and crumpled rest of the pe- 

 rizonium; the cell itself is very 

 young which is seen from the 

 absence of a distinct girdle- 

 part; the nucleus is situated 

 close to the innerside ot that 

 valve, which turns away from the perizonium. I did not succeed 

 in finding other stages, but the knowledge at hand is sufficient to 

 show that the auxospore formation probably goes on in the same 

 manner as in Thalassiosira gravida (Gran, Norske Nordhavs-Exp., 

 Protophyta 1897) or in Melosira (see f i. G. Karsten, Wissensch. 

 Meeresunters., 1899, p. 183), with the difference that follows from the 

 fact that the cells of Cose, siibbiiliens occur solitary, not in chains. 

 Hence it results that the auxospores immediately become separated 

 from their mother-cells, which makes it difficult lo observe them. 



vizi 



Fig. 3. Coscinodisciis subbuliens Jørg. The 



two figures to the left represent the same 



cell. All 71 t. m. 



