244 Inheritance in the Groundsel 



could produce the hybrid in each case and from its progeny select the 

 corresponding radiate form, one would be able to demonstrate two facts 

 of importance: (1) the general transmissibility of the RR character, and 

 (2) the realitj' and permanence of the characters which serve to define 

 the four types. 



This transference has been successfully made and the new types 

 thus produced have been maintained as pure cultures for various 

 periods. 



a. Praccox, radiatus was produced in 1908 and has been kept pure for 3 i,'enerations. 



(PI. XVI, fig. 7.) 



b. Latifolius, radiutus ,, 1908 „ ,, 3 ,, 



c. Multicaulis, radiatus „ 1909 ,, ,, 2 „ 



(PI. XVI, fig. 4.) 



d. Genevensis, radiatus ,, 1911. (PI. XVI, fig. 9.) 



Methods of crossing adopted to secure the F^ generation in each case. 

 It is of course possible to cross radiate groundsels by removing the 

 disc florets while they are still young and crossing the ray florets with 

 foreign pollen. Seeds have been actually obtained from 8. viscosus 

 and fi-om S. squaUdus when endeavouring to obtain hybrids in this 

 way, but the plants raised from them were in each case identical 

 in character with the mother plants. There are two possible inter- 

 pretations of such results : — either (1) the ray florets are apogamous 

 and are thus independent of fertilization, or (2) pollen of the same 

 type as that of the mother plant accidentally reached the stigmas. It 

 is quite likely that the ray florets of one of these species are apogamous. 

 Though then it may be possible to cross the individual flowers of the 

 common groundsel, it is not practicable in the open air, and in the green- 

 house it is quite unnecessary to do this in order to secure the best result. 

 The two plants selected as parents are isolated as much as possible, and 

 the flowers protected from insect visits. If the capitula are then rubbed 

 together cross-fertilization is efifected in both of the possible way.s, seeds 

 are set freely, and the parentage of the offspring is readily determined. 

 All the seeds of a crossed head are used to produce a fresh colony of the 

 next generation. Two kinds of plants are to be expected in this : — 

 (a) plants like the mother produced by self-pollination, and (b) hybrids 

 produced by the crossing. The hybrids when praecox is the mother 

 plant are much larger and later plants than their pure-bred neighbours. 

 In the large number of crosses so far made, there has never been any 

 difficulty in distinguishing the hybrids soon after the plants are bedded 

 out. When praecox is cue of the parents, the hybrids may often be 



