132 REDUCTION OF VARIABILITY. 



boundaries of the species, there will be descendants of hybrids, 

 and that some individuals will be able to thrive there, their 

 genotype permitting it. Thus we may picture the establishing 

 of new sub-species in new regions, new for the group. In reality, 

 such cases are not too rarely met with. A very typical instance 

 is the finding of a very highly specialized Kangaroo mouse, by 

 Grinnell, in a Califomian desert. 



If we are familiar with two distinct species of plants, such as 

 Lychnis vespertina and Lychnis diurna, and we see that each 

 prefers surroimdings, which are just a Uttle different from the 

 optimum environment of the other, we can see that this sUght 

 difference serves to keep them distinct as species. In one type 

 of environment, in open, dryer, grassy spots, there is a great 

 majority of Lychnis diurna, and occasional hybrids with ves- 

 pertina have a poor chance of estabhshing themselves as a 

 group apart in this same environment. They are too far in the 

 minority. 



If in Santpoort (Holland), a new field gets into a condition 

 such as favours the growth of Lychnis diurna, the seeds of this 

 species, which have been drifting into the field continually for 

 years, will finally have a chance of developing into flowering 

 plants. The field will be stocked from seeds of neighbouring 

 plants, and later on, a few individuals of this field may furnish 

 seed which succeeds in establishing itself in another field, when 

 it has come in the right condition. In such a group of plants 

 where cross-breeding is the rule, new types of hybrid origin will 

 stand a poor chance of establishing themselves, unless either of 

 two things happen, namely, unless a group of them is effective- 

 ly isolated, or unless some new character confers a very con- 

 siderable advantage upon the individuals presenting it. 



Eventually the hybrids between vespertina and diurna will 

 get lost into the multitude of diurnas, or into the multitude of 

 vespertinas. One of us has observed small colonies of Lychnis in 

 Santpoort, which were evidently of hybrid origin, such as a 

 field in which aU the plants were typical Lychnis diurna with 

 the exception that they had the teeth of the capsule erect as in 



