152 DE VRIES— ELEMENTARY SPECIES IN AGRICULTURE. [April 1 8 



continuous selection of its pedigree. It is especially in Germany 

 that this method of slow amelioration is much beloved and has given 

 admirable results. One of the best known instances, and for which 

 the historical records are the most complete, is the renowned rye of 

 Schlanstedt, produced by Rimpau, which is now largely cultivated 

 all over the central parts of Germany and the northern districts of 

 France. In the year 1876 I had the privilege of visiting Mr. 

 Rimpau on his farm at Schlanstedt and of studying his cultures. 

 The choicest of his new rye occupied a small patch out on the 

 fields, but surrounded by cultures of vegetables and other plants not 

 belonging to the cereals. These minor cultures occupied a large 

 square, which in its turn was surrounded by a complete range of 

 shrubs. Thus the rye, standing in the midst of the square, was suffi- 

 possible contamination by pollen of other varieties. On the other 

 hand, it was given the same soil and exposure and almost the same 

 cultural treatment as the average cultures. 



This race had been started by Rimpau nine years before, in the 

 year 1867. At the time of the harvest of that year he had inspected, 

 as he told me, a large number of his rye fields and selected all the ears 

 which seemed to him to surpass the others quite strikingly. He 

 brought home a handful of them, repeated the trial and mixed their 

 seeds. This mixed condition of his seed in the beginning of his 

 race has now become the weak point, where the whole principle of 

 his method is open to criticism. 



The seeds were sown next year, and in the harvest the same 

 selection of the best ears was repeated. Care was taken to exclude 

 all those, which by some external condition could have profited from 

 more space or more manure than the remainder, and could have 

 grown large by such accidental means. No care, however, was 

 taken to isolate the individuals and to sow their seeds separately, the 

 principle being that all the plants belonged to one race, and that this 

 race had to be ameliorated. This principle of ameliorating a race 

 without isolating its possible constituents seemed at that period to 

 be the right one, though now it can hardly be considered as scien- 

 tifically correct. 



Each year in the same way the best ears were chosen from the 

 continuance of the choicest strain, and after the exclusion of all ears 



