Origin of Species in the Genus Hieracium. 351 
crossing new forms do actually arise ; and they are at once perfectly 
constant, and behave like ordinary species. My experiments have 
at present reached the fifth generation and in several cases I have 
had more than 100 individuals of a single generation, all the plants 
of which were alike, and like the F x -individual from which they 
were derived. I have commonly used the orange-red H. aurantiacum 
as male parent, because it is so easy to trace its influence in the 
colour of the corolla of the hybrids. The following figures will 
show the extent of some experiments:— 
Number of individuals of the different generations of two 
crosses between two H. excelleus and H. aurantiacum. 
Sister hybrids 
from the 
same cross. 
In all my experiments the rule of constancy of the later hybrid 
generations has held good with one exception: in the F 2 of No.46 3 
there appeared one very aberrant individual—a mutant—which 
unfortunately was sterile and the behaviour of which by isolation 
and by crossing consequently could not be investigated. 
As an example of the heterogeneity of the F x -generation I may 
mention that a single cross between H. auricula (a sexual species) 
and H. aurantiacum gave twenty-nine hybrid-individuals all of which 
differed from each other. Most of them were sterile and weak and 
have since died, but a few of them proved to be fertile and at 
present have reached the third generation. The individuals of the 
F 2 and F 3 generations remain like the individual of the F x generation 
from which they sprang. 
It is thus evident that hybridisation plays a role as a factor in 
the origin of new forms in the sub-genus Pilosella, and it is probable 
that many forms have originated in this way. On the other hand, 
F\ F 2 F 8 F 4 F 6 
No. A6 1 
1 
3 
12 
— 
— 
No. 46 g 
1 
110 
38 
— 
— 
No. 46 3 
1 
52 
3 
28 
— 
No. 46 4 
1 
38 
205 
17 
14 
No. 48a 
1 
14 
106 
30 
16 
