THE PURE LINE HYPOTHESIS. 



545 



nasturtium plant (A) ciurino- December, 191 5, and January, 1916. 

 Thirty seeds were obtained. These were sown on 31st January, 

 1 916, and 21 seeds germinated. The seedlings were planted out, 

 and by 23rd March two plants were in flower. 



4. Offspring of Hybrids. 



The 21 hybrids were self-fertilised, and seeds were obtained 

 from each plant during' May- July, 1916. These seeds were sown 

 in pans on 30th September, 1916, and on the 17th October the 

 seedling's were planted out in fully exposed, long plots adjacent 

 to the Natal Museum. The plants flowered in November and 

 December, and the observations were made at that time. 



5. The Effect of Self -fertilisation on Fertility. 



During the Natal winter (May-July) the hybrids were self- 

 fertilised, and seeds were collected from every plant. Many of 

 the pollinations were quite inefi'ectual. and it was clear that self- 

 ing was prejudicial, since foreign pollen acted promptly and 

 'fertilisation almost invariably occurred. The aim was to obtain 

 about 25 seeds from each plant, and since equal attention was 

 paid to all the plants, the number of seeds actually obtained from 

 a plant gives a very fair comparative measure of its fertility with 

 self-fertilisation. The number of seeds which germinated in 

 every set compared with the numl^er sown gives another com- 

 parative measure of fertility, and it will be seen that these two 

 measures have a distinct tendency to agree with each other. 

 The figures are given in the following table : — 



Thus in the first seven sets of seed, with an average of 12 seeds 

 in the set, the percentage number which germinated was 40% ; 

 in the second seven sets of seed, with an average of 19 seeds in 

 the set, the percentage number which germinated was 58% ; while 



