BY ARTIFICIAL MEANS. 367 



rations, from those living ones that we do possess, if ever it be 

 possible, will enable us to reply in certain manner to the question 

 (possibly premature) of the editor of La Diatomiste. 



Cultures in Series.— Along with cultures with media differently 

 composed (termed abnormal) cultures may be attempted having 

 special objects in view. These have for their object the multipli- 

 cation of some forms of Diatoms by the addition to a pure state 

 of such and such algae ; by this means I obtain easily Schizonemes 

 in long branching thalli. These mixed cultures are very interesting 

 to carry on, but their description in the general work would carry 

 us too far. I prefer to speak on the mode of culture in series. 



A diatom cultivated in a normal medium increases according 

 to a certain law, easily verified by the living dissociate species. 

 The individual sown being of a known shape, you prove that the 

 number of the frustules of the same shape which proceed from it 

 are as the terms of the expression (i + i)'^ carried out.* 



When in this case the Diatoms of medium size are much the 

 more abundant, in transferring a small number of frustules into 

 new macerations, you ultimately ehminate the large-sized Diatoms, 

 and obtain at the end of a time more or less long, after the tenth, 

 twentieth, or thirtieth successive cultures. Diatoms of very 

 reduced size which are in this alternative, either many kinds dis- 

 appear if they continue the multiplication, or become the starting 

 point for the re-establishment of the size of the species. Diatoms 

 arrive at their minimum of smallness by way of cultures in series, 

 growing in a very great proportion, and furnishing numerous auxo- 

 spores, which you meet with by millions in a culture of a dozen 

 cubic centimetres of volume. 



In employing this method of culture, I have helped to re-estab- 

 lish the size among many Melosiras, the Nutnmidoides and the 

 variafts, the Cyclotella covita^ twO Biddulphias (the rhombus and 

 the aurita), and that whether by natural or artificial light. With 

 the Melosira varia?is I have lately succeeded, in mid-winter and 

 with a very low temperature ; the size in the proportion of one to 

 three, the intermediate and the diametrical division of the large 

 spheres, which are produced in the articulated chains of the 



* See Annales de Micrographie , Vol. iv., p. 532. 



