228 THE NATURAL HISTORY RETIEW. 



in the fields or pastures near liis house, where many become per- 

 fectly naturalised, and reproduce themselves 3'ear after year. He has 

 thus (he thinks) excellent opportunities of observing the permanence 

 or variability of forms. He tells us that these experiments and 

 observations have been carried on for twenty five years. In one case 

 he takes us back to the year 1829, often for ten, twelve, or fifteen 

 years, but in some cases only for three, four, or five years. He was 

 led to this course of study by observing that, in a state of nature, 

 plants, comprehended imder the same name, presented differences 

 which, though not very conspicuous, seemed to him certainly not 

 merely individual. On inquiry, he found that the leading botanists 

 of the day explained these differences by variability of type, an 

 explanation which did not satisfy him, and which, therefore, made 

 him wish to observe for himself the existence of the variability thus 

 taken for granted. He therefore repeated his observations on more 

 numerous individuals in their native places of growth, selecting them 

 in all states and of all ages. As he still got the same result, he 

 proceeded to cultivate the several forms in his garden, or in some 

 other easily accessible place. Finding that they always remained con- 

 stant year after year, it became evident to him that they were more 

 than casual forms. He then raised each separate form repeatedly 

 from seed, and as they still retained all the characters of the parent 

 plant, any remaining doubt was changed to certainty. It became 

 evident that each form was a distinct species, and it was then a 

 matter of necessity to give it a name, so that it might be known 

 from other species, from which nature itself had distinguished it. 



In the broadest and most positive terms M. Jordan gives it as 

 the result of his experience, not only that wild plants when trans- 

 planted into his garden retain their characters unaltered throughout 

 a long succession of years, but that their seeds produce year after 

 year, invariably, the same form as the parent x^lant. Once only, so 

 far as we have observed, does he give a hint of slight variation in 

 the oftspring. In describing the species (ten in number) allied to 

 Erysimimi Bocconei, he tells us that they are all so closely related 

 to each other that, without good specimens and great attention, they 

 cannot be distinguished in the Herbarium. At first therefore, he 

 says, one would be inclined to regard them as modifications of one 

 common type ; but this, though s])ecious, would be wrong, as he 

 has raised them many times from seed since 1840 and 1841, in 

 which years he collected the greater number of them. Slight 



