782 PART IV. THE PHYSIOLOGY OF PLANTS. 



acters of the individual will be those transmitted to it by the 

 chromosomes of the gametes. Consequently, if the gametes were 

 derived from two distinct individuals, the characters of the new 

 individual will be those of its two parents. In this way the more 

 obvious phenomena of heredity can be traced to a material basis, 

 and are thereby rendered more comprehensible. 



But what is true of the new individual, applies likewise to its 

 parents : the characters which each parent transmits to the off- 

 spring are those which it has itself received from its two parents, 

 and so on : hence the characters inherited by any individual are to 

 be regarded as belonging rather to the race than to its immediate 

 progenitors. This conception also can be traced to a material 

 basis. It has been suggested that the discoid segments (con- 

 sisting of linin and chromatin, p. 96) of which the nuclear chro- 

 mosomes consist, each represent, in a gamete, substance derived 

 from a number of ancestors, the whole chromosome representing 

 many ancestors, and the chromosomes together all the ancestors 

 whose substance still persists in the gamete and will be trans- 

 mitted by the gametes to the next generation. The term id is 

 used to designate one of these material units which seem to con- 

 stitute the physical basis of heredity. 



Applying these considerations to the elucidation of the repro- 

 ductive process in a plant, such as a Fern, whose life-history pre- 

 sents antithetic alternation of generations, it would appear that 

 the sudden reduction by half of the number of the nuclear chromo- 

 somes which attends the initiation of the gametophyte (see p. 771) 

 is to be attributed to the fusion of the ids in pairs : and further, 

 that it is not until this stage that a real fusion of the nuclear 

 substance of the gametes actually takes place. 



The phenomena of heredity as manifested in the products of 

 sexual reproduction may be accounted for in connexion with this 

 fusion of the parental ids. When, for instance, in hybridisation, all 

 the parental ids exert their full influence, the offspring is precisely 

 intermediate in character: but when, from some cause which 

 cannot now be explained, some of the ids are paralysed or neu- 

 tralized, the offspring resembles one parent more than the other. 

 The character of the hybrid too, whether blended or blotched, may 

 be referred to some such cause. 



Turning now to the phenomena of variation : this may be ac- 

 counted for, when it depends on reversion, on the assumption that 

 some of the ancestral ids which have remained neutralized and 



