je 1 Arend. L. Hagedoorn and A. C. Hagedoorn. 



become intermediately coloured, those parts of the body, on which 

 the moult proceeded when the wheather was still warm, being coloured, 

 and those parts, on which the new hair began to grow when it was 

 colder becoming white. These animals are only therefore either white 

 or coloured, because their moult sets in at two widely different 

 periods of the year, with a marked difference of temperature. 



Some plants, such as the lilac and Dahlia arborea bear their 

 leaves either in pairs, one opposite the other, or in whorls of three, 

 and the better-nourished buds only develop into shoots with whorls 

 of three leaves. Here no intermediate stages occur. The difference 

 between such a case, and the production of fasciations, which is 

 favoured by similar conditions, lies probably in some relatively simple, 

 but not as yet understood physiological- morphological mechanism. 



Another good example of discontinuous variation induced by 

 non-genetic factors, is the case of Polygonum aiuphibhim. In this 

 plant, shoots growing from the water differ markedly from such as 

 grow from the land, and this difference exists even between shoots 

 coming from the same rhizome. 



The most beautiful example of discontinuous variation caused 

 by non-genetic factors, is one given by de Vries. He grew a strain 

 of teazels (Dipsacus), which had the peculiarity of producing normal 

 as well as peculiarly twisted individuals. No intermediates occured. 

 The better nourished a young plant of this strain, the more chance 

 it has to grow up twisted, and by growing the young rozettes of 

 this strain under very favourable circumstances, it is possible to get 

 a very large proportion of twisted individuals. 



The ordinary teazel presents nothing of the kind. No matter 

 how well nourished the plants are, they never grow up twisted. 



This case beautifully illustrates the meaning of "genotype" and 

 "phenotype". There are two phenotypes among teazels, the normal 

 and the twisted. And there are also two genotvpes, the normal and 

 the other. One group of teazels, being of one genotype, always 

 consists of some plants of one phenotype and some of an other, and 

 reversely, if we find a group of normal teazels, about the origin of 

 which we are not informed, they may, for as far as we know, belong 

 all to one genotype or all to the other, or some to one and some to 

 the other. 



All those cases, in which genotypically homogeneous groups of 

 organisms fall into more than one phenotype are instances of dis- 

 continuous non-inheritable variation. Most cases of de Vries's "ever- 



