Dec, 1921.] ON THE FORMATION OF AUXOSPORES 227 



that figured by Gran, but it carries, when fully formed, a whirl of 

 fine hairs on the lower valve. 



Unfortunately I have been unable to see the original figures 

 given by Cleve and consequently have not been able to make a 

 comparative study. But I am strongly inclined to believe that the 

 full-grown resting spore has puncta on the epitheca and fine hairs on 

 the hypotheca. These hairs are visible in water mount, and are 

 nearh' straight, attaining about 40 iJ- in length (Fig. 7.). 



From the observation mentioned above, I may summerize the 

 result of my studies as follows : 



1). In the process of the auxospore-formation two pores are 

 formed directly opposite to each other on the middle portion 

 of the narrow girdle of the mother cell. Cell-content bulges 

 out through one of the two. 

 2). The secondarily formed valve of the auxospore has no seta 

 which may perhaps be formed subsequently after the chain 

 become detached from the mother frustule. 

 3j. As a specific character of the daughter-chain, the secondari- 

 ly formed frustule is much more vaulted than the primarily 

 formed one and is destitute of setae. 

 4j. The resting spore in its fully formed state, has both crowns 



of puncta and fine hairs on the valves of the same spore. 

 In concluding, I must not neglect to present mj' heartily thanks 

 to Professor Shigeo Sasa, Director of the Fisherj^ Institute of Sapporo, 

 and especially Dr. K. Okamura for so kindly taking the trouble of 

 looking over the manuscript and proofs. 



Imperial Fisheries Institute, 

 Fukagawa, Tokyo. 



Explanation of Plate Y. 



Fig. 1. A part of chain with frustules, the cell-content bulging out tlirongh a pore. 



Fig. 2. Spore-forming frustule has discharged about an half of tiie content. 



Fig. 3. An auxospore has just connpleted :ind adhering to an emptied frustule. Tlie 



nucleus takes its position at the remotest end of the spore, and the chro- 



moplasts are pnrietally disposed. 

 Fig. 4. A pair of the auxospores. The upper one has fixed the epitheca in size and 



liaving the setae on its corner. The lower one is still young. Fine pro- 



toplusmic thread are seen, wl*ich connect the spore and the mother frustule. 

 Fig. 5. A part of chain with one auxospore and two daughter-chains. 

 Fig. 6. A daughter-ch:iin with 7 cells, adhering to the mother frustule. The :ixis 



of the daughter-chain lies at right angle with that of tlie motiier one. 

 Fig. 7. Two resting spores in full development which bear both wreaths of puncta 



and fine hairs on tiie valves. All x335. 



