518 PKESIDENTS ADDRESS. SECTION Ql. 



and Tschermak. His theory may be thus stated : — In any cross or 

 hybrid, one of the contrasted characters of the parents will in certain 

 cases appear to the apparent exclusion of the other. This prevailing 

 character may be called the dominant character, and the apparently 

 suppressed character, the recessive. If reproduction be from the 

 dominant forms alone it is found that in the offspring there are 

 apparently three dominant forms to one recessive form.* These 

 recessives breed true. The dominants, on the other hand, consist of 

 one-third of true-breeding dominants and two-thirds with suppressed 

 recessive characters, since in the next generation the latter again 

 produce the mixture of dominants and recessives in the proportion 

 three to one. The pure dominants and pure recessives are virtually 

 pure stock again, breeding true for an unlimited number of genera- 

 tions. 



From these iiiustrations it is easy to see how essential is the 

 statistical method for the investigation of all important human 

 phenomena, and it is also evident how hopelessly invalid is what may 

 be called the popular dilution conception of such phenomena as we 

 are considering, f 



7. Unit Characters. — The facts are often not so simple as is 

 implied in the last illustration. Some characteristics or " characters" 

 of an organism behave in inheritance as if they were transmissible 

 only en bloc, and like a radicle in chemistry, can be completely re- 

 placed by another radicle, but cannot be invaded as regards their 

 individual integrity. This notion of unit characters is quite con- 

 sistent with Weismann's theory of sets of determinants or primary 

 constituents, each corresponding to an independently heritable struc- 

 ture. It goes far to explain what the close examination of Mehdelian 

 phenomena has coniurmed — viz., that the percentage of relationship 

 may be far more elaborate than as instanced above. The subject 

 has been mathematically developed by Professor Karl Pearson, who 

 shoAvs that while the Mendelian and Galtonian doctrines are recon- 

 cileable, the full statement of the nature of the phenomena is veiy 

 complex. J This complexity may be in part due to the fact that some 

 characters are analogous to .atoms or simple radicles (allelomorphs), 

 while others may consist of several components (compo\md allelo- 

 morphs), and may enter into new combinations like the more complex 

 molecules of organic chemistry. 



8. Theory of Disease. — ^Among what are called transmissible 

 characters is predisposition to certain diseases. Persons who are 

 unsatisfied until the economic significance of any natural truth is disr- 

 closed, will at once recognise the value of statistical deductions as to 

 the frequency of transmission of such adverse predispositions. It is 

 hopeless to attempt, in any adequate way, to refer to this subject, 



* The actual iiumljers, in the case of 1,064 hvbrid.« of the tall and dwarf pea, 

 were 787 tall, 277 dwarf, instead of the theoretical 798 tall and 266 dwarf, or 7^*0 and 

 26'0 per cent., instead of 75"0 and 25'0 ptir cent. Toyama with Siamese silk moths got 

 74'96 and 25'04. With the Lina lapponica, a Californian beetle. Miss McCraeken got 

 747.5 and 25 •2.5 per cent. With rabbits, Hurst got 70 and 24 per cent., and in another 

 case 77 and 23 per cent. 



+ Recent embryological investigations go far to explain the physical mechanism 

 of this — viz., the chromosomes and their elimination. 



t See, " On a generalised theory of Alternative inheritance, with special reference 

 to Mendel's laws," by Karl Pearson, F.R.8., Phil. Trans. A., vol. 203, pp. 82-8(1. 



