CULTURE OF PLANKTON DIATOM THALASSWSIRA GRAVIDA CLEVE. 421 



chains is very great, especially at first, the colour is a deep rich brown, 

 and healthy growth in a flask will go on for months. In a contaminated 

 culture, on the other hand, growth is slower and only quite short chains 

 are seen, the colour is a much hghter brown, and the culture does not 

 continue to grow in a healthy way, generally forming auxospores and often 

 dvino; off altogether in the course of two or three weeks. 



All the main conclusions detailed in this paper have been confirmed 

 with clean and healthy cultures. Experiments with contaminated cul- 

 tures are not, however, without value, since they sometimes emphasize 

 the differences between culture media that it is desired to compare, a 

 contaminated culture often faihng to grow at all in an unfavourable 

 medium, whereas a clean culture might give a growth, less in amount, it 

 is true, but not much different in character from the growth in the control 

 culture in a favourable medium. 



THE ARTIFICIAL WATER. 



The artificial sea- water used in the experiments was made by dissolving 

 Kahlbaum"s pure chemicals in ordinary distilled water made in a copper 

 still which had been redistilled in all-glass apparatus after being treated 

 with bichromate of potash and sulphuric acid, to destroy volatile organic 

 matter. This double distilled water contained at most 0-01 mg. of am- 

 monia per Htre.* 



The composition of the water was based on the analysis of sea-water 

 published by Dittmar in the " Challenger " Reports. f The figures given 

 by Dittmar are : — 



Per 100 parts halogen. 



CI 99-848 



Br -3402 



SO3 11-576 



CO2 -2742 



CaO 3-026 



MgO 11-212 



K2O 2-405 



Na^O 74-462 



Dividing these figures by the respective molecular or atomic weights, 

 and treating those for CI and Br together as chlorine, we get after 



* In connection with the preparation of the artificial sea-water I received constant help 

 and advice from my colleague, Mr. D. J. Matthews. Without his ready assistance in con- 

 nection with all chemical questions tliis investigation could hardly have been carried out. 



t "Challenger" Report, Chemistry, Vol. I, p. 203. 



