82 Mr Bateson & Miss D. F. M. Pertz, Notes on the inheritance 



By these successive countings the relative numbers of each 

 kind of flower borne by each plant during its flowering period 

 were approximately obtained. In a good many cases days were 

 missed, and in 1894 and 1895 the plants had not altogether 

 finished flowering when the counting stopped. We have not 

 attempted to make any correction for this, but there is no reason 

 to suppose that the results are thereby materially affected. 



In the Tables the number of each kind of flower is given as a 

 percentage of the total number of flowers borne by the plant. This 

 total is given in the right-hand column of each Table. In the 

 bottom line of each Table is given the average percentage of each 

 kind of flower borne by the whole family sprung from one capsule. 

 This average is calculated on the grand total of flowers borne by 

 the whole family. 



It is noticeable that while there is very great diversity among 

 the offspring of the same self-fertilised capsule, yet the average 

 degree of abnormality in the family generally remains fairly 

 constant for the descendants of each original plant. 



The extraordinary irregularity in the percentages amongst 

 these offspring of self- fertilisation is not a little remarkable. 

 It might perhaps have been expected that as the result of 

 successive self-fertilisation some diminution in variability would 

 occur. So far as our figures go, there is no indication of such a 

 result. 



An attempt was made to discover whether the abnormal 

 flowers were produced with any greater frequency at one part of 

 the flowering period than at another. Decided evidence was 

 found that this is so. Though there are not a few exceptions, the 

 countings shew in a large majority of cases that the chief output 

 of abnormal flowers takes place in the earlier part of the flower- 

 ing season and especially just before the greatest output of 

 flowers, after which time the percentage of abnormality declines. 

 It was not thought worth while to print the detailed records on 

 which this statement is based. 



The plants studied, 135 in all, were severally descended from 

 three original plants, and the descendants of each of these three 

 are referred to as Race I., Race II. and Race III. respectively. 



Race I. The original plant from which Race I. descended was 

 growing wild in a plot of waste garden-land near Cambridge. On 

 it 5 flowers were chosen, being respectively A, normal ; B, 2 an- 

 terior petals ; C, D and E, each having 2 posterior petals. As to 

 the fertilisation of these flowers nothing was known. 



From these flowers the first generation was raised and the 

 succeeding generations were produced by self-fertilisation, the 

 detailed pedigrees being given in the Tables. 



