1897] NOTES AND COMMENTS 11 



division, and (2) by the formation of so-called auxospores, which 

 are usually the result of conjugation. The former method of 

 division is of unusual interest in the diatoms, since, owing to the 

 peculiar arrangement of their valves by which one fits inside the 

 other, a formation of progressively smaller and smaller individuals 

 is a necessary result of simple division. The growth in size of the 

 valves is usually supposed to be impossible on account of their 

 silicified nature. It is imagined, however, that the production of 

 infinitely small individuals, which would obviously result from the 

 continuance of such a method of division, is prevented by the in- 

 terpolation of auxospore formation ; by the latter means the diatom 

 is given an opportunity of regaining its maximum size. 



Mr George Murray, in a paper entitled " On the reproduction of 

 some marine Diatoms" (Proc. Roy. Soc, Eclin., Dec. 1896), has, 

 however, thrown new light on reproduction in this group. The 

 author has observed in some marine forms a very interesting and 

 totally new method of division. This form of reproduction was 

 observed in Coscinodiscus, Biddulphia, Choeboceros, &c, but was 

 followed out most fully in the first-mentioned genus. In Coscino- 

 discus the cell-contents divide by successive division into eight or 

 sixteen portions, and these become rounded off and lie free in the 

 mother cell like spores in a sporangium. Each of these portions 

 becomes invested with valves showing the characteristic markings, 

 and in fact becomes a young Coscinodiscus. These young forms ulti- 

 mately escape from the parent cell, and are found floating free in the 

 water as packets of eight or sixteen small individuals enclosed in a 

 delicate membrane ; later on the several individuals themselves 

 become completely free. It is by this method of division that the 

 enormous quantities of marine diatoms, occurring in many waters at 

 certain seasons, are doubtless produced. 



As mentioned above, diatoms are usually supposed to be incapable 

 of superficial growth, owing to the silicified nature of their membrane ; 

 but, as Mr Murray points out, this view is hardly well founded. 

 At all events, it raises the question of the nature of the membranes 

 produced by the young diatoms inside the old valves. These diatoms 

 must obviously be much smaller than the parent, and so, in the 

 absence of growth, continued reproduction by this method would have 

 as disastrous an effect on the size of the species as the method of 

 simple division, and would similarly necessitate the production of 

 auxospores. With this question in view Mr Murray investigated 

 the nature of the membrane of the daughter forms, and his con- 

 clusion is that the valves are either not silicified or only incompletely 

 so. There is thus nothing to prevent the further growth of these 

 young diatoms to the full size of the species. That this later growth 

 takes place is supported by observation, for the very small forms of 



