PRESIDENTIAL ADDRESS SECTION D. 91 



•same hereditary substance is perpetuating itself through a series 

 of generations. To such a group of individuals, produced in the 

 manner indicated and roughly similar to one another, without 

 tendency to break into types, the term "pure line" has been 

 -applied. If the pure-line hypothesis is absolutely correct, then 

 selection within a self-fertilised family has no result. Johannsen's 

 ideas favour the view of the constancy of factors, as held by Men- 

 delians. Karl Pearson, however, has shown that Johannsen's 

 results were not conclusive. 



Dr. Warren, in South Africa, experimenting with foxgloves 

 .and nasturtiums, obtained results which do not favour the pure- 

 .line hypothesis. He found that the fluctuating variations are 

 inheritable, not as stated by the Mendelians, and that the factors 

 handed down to the fertilised germ-cell are not constant, but are 

 variable in nature. He concludes that "the smallest variation 

 may be inheritable, and can be utilised in the course of evolution." 



Although the researches of Morgan and his collaborators in 

 America on chromosomes as bearers of the hereditary material 

 have been interpreted in Mendelian terms, yet recently these 

 interpretations have been challenged, notably by W. E. Castle and 

 by H. S. Jennings. We must therefore endeavour briefly to con- 

 sider the brilliant researches of T. H. Morgan, which were largely 

 lone on Drosophila ampelophila, the fruit fly or pomace fly. This 

 fly breeds quickly. Much consideration was given to eye colour. 

 The eye is normally red, but various mutations appeared during 

 the investigation, thus one mutation showed white eye colour, 

 another eosin eye colour, and there were found ultimately seven 

 grades of colour due to changes in the X chromosome. This is an 

 example of a single unit factor with many grades. Morgan and 

 his collaborators belong to the Mendelian school. The seven 

 grades of eye colour on the mutation hypothesis are explained by 

 "multiple modifying factors." The grades of colour are thought 

 by them to be essentially discontinuous, but the steps in the series 

 become minute, and in the end barely detectable. Bridges, con- 

 tinuing the work, found seven secondary grades within one of the 

 primary ones. We must also mention the important researches 

 of W. E. Castle, especially those on hooded rats, showing graded 

 results of the amount of colour in the rat's coat in biparental 

 inheritance. Castle explains the graded coat colour in rats and the 

 graded eye colour in Drosophila not as the working out of Men- 

 delian recombinations of mosaic-like parts, but he believes in 

 actual alterations of the hereditary constitution, indeed, an actual 

 change in a single-unit factor, and that in rats he can by selection 

 gradually increase or decrease the amount of colour in the coat, 

 passing by continuous stages from one extreme to another. 



H. S. Jennings holds much the same views as Castle. Jennings 

 worked on Paramoecium and on Difflugia corona, where he found 

 that a particular stock or strain resulting bv fission from a single 

 parent, does differentiate gradually, with the passage of genera- 

 tions into many hereditarily diverse strains. Most of these heredi- 

 tary variations were minute gradations, and "variation was as 



9 



