DISCONTINUOUS VARIATION. 69 



into Paraguay likewise prefer associating together." 

 Again, " It has been observed, in a district stocked with 

 heavy Lincolnshire and light Norfolk sheep, that both 

 kinds, though bred together, when turned out, in a short 

 time separate to a sheep." Still again, with respect to 

 fallow-deer, " Mr. Bennett states that the dark and pale 

 coloured herds, which have long been kept together in 

 the Forest of Dean, in High Meadow Woods, and in the 

 New Forest, have never been known to mingle." Dar- 

 win adduces other similar instances, in the case of the 

 dog, horse, sheep, rabbit, and pigeon* hence there can 

 be little doubt of the genuineness of the phenomenon, 

 even though it is not based on very exact observation. 



Supposing that the above view is correct, it follows 

 that between at least some varieties there must exist a 

 greater or less degree of sterility. Of course it is not 

 necessary that divergence of morphological character 

 should always be accompanied by corresponding diver- 

 gence of physiological character; but merely that this 

 is sometimes the case. Upon this point Darwin has col- 

 lected a considerable amount of evidence in his " Ani- 

 mals and Plants." * One or two of the cases there 

 cited may be quoted here. Gartner found that a 

 variety of dwarf maize, bearing yellow seed, showed a 

 considerably diminished fertility with a tall maize bear- 

 ing red seed, though both varieties were perfectly fer- 

 tile when crossed inter se. Again, in the genus Ver- 

 bascum, numerous experiments were made by Gartner 

 with the white and yellow varieties of V. lychnitis and 

 V. blatteria, when he found that crosses between simi- 

 larly coloured flowers yielded more seed than those be- 



*Vol. ii. p. 82. 



