426 E. J. ALLEN. 



liquid from the precipitate which is formed on the addition of sokition B. 

 The clear liquid is referred to in what follows under the name " artificial 

 miqueled water." 



It was found, however, that even after the addition of these two 

 solutions very slight growth, if any, took place on inoculating with a small 

 quantity (say one drop) of healthy Thalassiosira gravida culture. This was 

 the case even after the addition of potassium bromide and iodide, or of 

 Miquel's own solution A, which contains these two salts. If, on the other 

 hand, the artificial miqueled water was inoculated with a cotisiderable 

 quantity of a culture of Thalassiosira in which natural sea-water had 

 formed the basis of the culture medium, so that a considerable quantity 

 of this natural water was transferred to the artificial medium, then the 

 latter would develop a fine healthy growth. Experiences of this kind led 

 me to suspect that the irregularities which had previously been met with 

 in trying to make cultures in artificial media* might be due to varying 

 amounts of natural sea-water introduced when inoculating. Definite 

 experiments were therefore undertaken in order to ascertain whether the 

 addition of natural sea-water to the artificial miqueled water would make 

 it effective as a culture medium, and if so what proportion of natural 

 sea-water was essential. In all cases the culture medium was boiled after 

 the addition of the natural sea-water and then alloived to cool before inocula- 

 tion. 



As a result of these experiments it was shown that an addition of even 

 1 per cent of natural sea- water to the artificial miqueled water was suffi- 

 cient to give very heavy cultures after inoculation with only one drop of 

 Thalassiosira culture, and that with an addition of 4 per cent of natural 

 sea- water better cultures were obtained than in any other culture medium 

 known to me. This result has now been obtained so many times that it 

 is in my opinion quite definitely established. If the proportion of natural 

 sea-water added is reduced below 1 per cent smaller growths are 

 obtained, and it is somewhat difficult to decide whether there is a definite 

 minimum below which no growth takes place. An addition of 0-3 per 

 cent of natural sea-water in one satisfactory experiment produced quite 

 a heavy growth, whereas without this addition only a small growth was 

 obtained. It has often been observed that whilst flasks containing 75 c.c. 

 of artificial miqueled water show distinct signs of diatom gro^^i;h when 

 inoculated with one or two drops of a culture of Thalassiosira in natural 

 sea- water, such flasks inoculated with one or two drops of a culture which 

 has artificial sea-water as the basis of the culture medium show practi- 



* Allen and Xelson, loc. ciL, p. 4-17 [Q.J. M.S., p. 394]. 



