I)ec,i!)-2i.] ON THE FORMATION OF AUXOSPOnES 223 



mode of auxospore-formation of Chaetoceras teres. But, in details of 

 that of this species I have found some points differing from Ch. debile 

 and others. So here I shall give a short description of what I have 

 observed. 



Chaetoceras teres Cleve is found whole year round in the vicinity 

 of Oshoro. Its maximum occurrence thereabout is to be found from 

 April to October. In the season we frequently meet with either forma- 

 tion of auxospores or resting spores ; the former is more frequent 

 during March and April, and the latter during September and October. 

 Cells producing auxospores measure 18-25 iJ. in breadth (measured in 

 broader girdle) and have no marked morphological differences with 

 the exception of highly refractive granules and tall frustules, 2^l:i-^ 

 times as high as the normal vegetative cell. (Figs. 1-5). 



The fertile cells are often found in 2 or 4? consecutive numbers in 

 one row, each having either auxospores or resting spores. The 

 case those having 4 fertile cells in continuation, is of rather occur- 

 rence ; in some cases only one fertile cells either in an interrupted or 

 continued row is, in my observation, due to the fact that the fertile 

 ones have originated from the same mother cell, and as the con- 

 sequence, pair of the fertile cells are of the same age and most 

 probably may have the same physiological significance (Fig. 4, 5, 

 and 7). 



In the cells now going to form auxospores, highly refractive 

 granules, which were already present, become very much increased 

 and aggregate themselves around the nucleus which is located on the 

 wall of the cell. The nucleus gradually changes its position and 

 moves towards the narrow girdle. Soon after the removal of the 

 nucleus, a small pore makes its appearance at the middle of the 

 narrow girdle. Its formation does not interfere with the new posi- 

 tion of the nucleus. The pore is slightly elliptical in shape, measur- 

 ing 5 H in the shorter axis and can readily be perceptible in the 

 living state when a few drops of some staining reagent are added. 



A sea-water solution of aniline blue is recommendable for this 

 purpose. A small portion of the cell-contents now bulges out through 

 the pore and the remaining portion within the cell, contracts receding 

 from the lining of the frustule. (Fig. 1 and 2). 



In the mean while, another opening is formed in the middle of 

 the side directly opposite the first pore. (Figs. 2-5). The case is 

 different from Ch. dehile, Ch. scolopcndra, Ch. compressum, Ch. con- 

 strictum, and Ch. didymum, so far as I have observed, in having 



