— 236 - 



This experiment was made in 1903 and was carried out in 

 the following way: 



Some newly opened flower-heads of the same clump of H. 

 pilosella which was used in castration experiment Ser. 1 4, and 

 Ser. Hi, were selected. Flower-heads of H. aurantiacum taken 

 from the Danish Quarter of the Botanical Garden (Ser. Ii and 2), 

 were passed accross the heads of II. pilosella in such a way that 

 the aurantiacum-pollen, which was present in large quantity, 

 could be cought by the papillæ of the stigmas. Afterwards the 

 heads (with small labels) were left to ripen. The crossing was 

 made in July and the ripe seeds were collected in August; next 

 year (1904) they were sown in baked soil the 1 st of May, germinated 

 the 25 th of May and were planted out in September; there were 

 then 19 plants, which flowered for the first time, very sparingly, 

 in October. It then appeared that, while the 18 specimens were 

 true H. pilosella, the 19 th was a hybrid, as will be seen from the 

 fig. 7 on the Plate, which is a painting of the first flowering- 

 scape. In June of 1905 the plant flowered again and continued so 

 most of the summer. On June 7 th three heads were castrated, but 

 they did not develop any ripe fruit; the same negative result came 

 from a new castration of 4 heads in September. The intact heads 

 gave mostly empty fruits, but also a few ones, which have germi- 

 nated in September of 1905 and which (summer of 1906) are now 

 planted out. Five specimens of this offspring have flowered now 

 and are very remarkable being not like each other, some nearer 

 Pilosella than the parent plant. There is the chance that the 

 flowers of the hybrid may have been fertilisated by another plant, 

 because the fruits have been taken from heads which not have 

 been isolated under glass. 



The primary hybrid is still alive and flowers in this summer; 

 of the pure ^/ose^a-offspring one plant is kept back, while the 

 others have been thrown away. 



The most important characters of the hybrid compared with 

 those of the parents are to been seen from the scheme on p. 237. 



The explanation of this experiment is not easy to give; the 

 fact is, that the same plants of H. aurantiacum and H. pilosella 

 which, as proved by the castration experiments, produce apogamic 

 seeds, are able to cross among themselves, forming a hybrid of 

 ordinary, nearly intermediate aspect; the hybrid has a very reduced 

 power of setting fruit and has hitherto not given any apogamic seed. 



