254 Numerical Aberrations [CH. 



that in some types of animals and plants the amplitude of 

 the fluctuations about normality is greater than in others. 

 This is of course a point well adapted to statistical study 

 which sooner or later must be undertaken. Similarly there 

 are indications that in cases where the point can be tested, 

 the seeds of individual pods may give strangely aberrant 

 numbers though the plants as a whole may give regular 

 results. Miss Saunders has often been struck with this 

 peculiarity in studying the genetics of Stocks, but as yet 

 the figures have not been examined by statistical methods. 

 Knowing what we do as to the capacity of individual flowers 

 to sport, the suggestion obviously arises that such indi- 

 viduality may be manifested by the collective offspring 

 of one ovary even though the flower in which it was 

 formed may show no peculiarity. 



The very wide departures from expectation shown in 

 many pedigrees of human diseases and defects are certainly 

 in part attributable to the imperfections of the records, but 

 I cannot doubt that the discrepancies are in part due to 

 genuine physiological causes. In regard to some of these 

 it is I think still open to question whether the transmission 

 is a process comparable with that which we ordinarily 

 designate as Heredity. Some element is obviously handed 

 on from individual to individual, but it seems to me possible 

 that this element or poison is distributed irregularly among 

 the germ-cells, spreading among them by a process which 

 is mechanical, like the spread of an oil-stain in a heap of 

 paper, or of a fungus in a heap of seeds. In the present 

 state of pathological knowledge it is premature to make any 

 suggestion as to the possible nature of such poisons. I am 

 told by competent authorities that in the cases, for example, 

 of the various polymorphic hereditary paralyses it is very 

 improbable that pathogenic organisms can be the exciting 

 cause ; nevertheless, from a study of the inheritance in an 

 ample series of families, I am inclined to suppose that the 

 element transmitted is something apart from the normal 

 organism, and that it is handed on by a process independent 

 of the gametic cell-divisions. In such cases I do not an- 

 ticipate that any "law" of inheritance can be discovered, for 

 if my view is correct, the process is not heredity in the 

 naturalist's sense at all. 



