Two Lata-Faniilies. 279 



I made the most extensive sowing, in these famihes, 

 and the richest in mutations, in 1900; I shah deal with it 

 first. In August 1899 I artificially fertilized 18 plants 

 of O. lata in parchment bags, all with pollen of 0. La- 

 marckiana which had either grown from /a/a-seed or 

 were from pure race. In the spring of 1900 I sowed 

 the seed, of each plant separately, in pans^ and ])ricked 

 out all the young seedlings, irrespective of their char- 

 acters, into wooden boxes filled with manured soil This 

 was done as soon as the first leaves had completely un- 

 folded and before the characters of any mutants that 

 there might be were discernible. The seedlings remained 

 in the wooden boxes until they had become strong young 

 rosettes ; and it was in these boxes that the mutants could 

 first be recognized. Plate IV and figure 48 give an idea 

 of the appearance of the boxes at this stage. 



Plate IV is reproduced from a photograph which I 

 took on the 18th of May 1900. The camera was so 

 placed that its optical axis was vertical ; all that then 

 remained to be done was to push the box under the cam- 

 era at the proper distance from it. 



This plan obviated the necessity of tilting the box 

 which might easily have fatal results for the young 

 plants ; and made it possible to grow them subsequently. 



I dealt with over 2000 seedlings in this experiment. 

 So that there was great opportunity for choosing good 

 groups for photographing. I either chose groups in- 

 cluding many examples of mutants of the same species 

 (Plate IV) or groups comprising many different kinds of 

 mutants (Fig. 48). I mention this because the figures 



*My plan was to sterilize the soil in these pans by heating it. 

 before sowing the seed, up to 90-100° C. This killed the seeds of 

 weeds which the soil invariably contained. The soil was not man- 

 ured. My seedlings did splendidly in soil sterilized in this way. 



