HEREDITY BATESON. 369 



cultivated plants are the outcome of deliberate crossing. There is 

 generally no doubt in the matter. We have pretty full histories of 

 these crosses in gladiolus, orchids, cineraria, begonia, calceolaria, 

 pelargonium, etc. A very few certainly arise from a single origin. 

 The sweet pea is the clearest case, and there are others Avhich I should 

 name with hesitation. The c^-clamen is one of them, but we know 

 that efforts to cross cyclamens were made early in the cultural history 

 of the plant, and they may well have been successful. Several plants 

 for which single origins are alleged, such as the Chinese primrose, 

 the dahlia, and tobacco, came to us in an already domesticated state, 

 and their origins remain altogether mysterious. Formerly single 

 origins were generally presumed, but at the present time numbers of 

 the chief products of domestication, dogs, horses, cattle, sheep, poul- 

 try, wdieat, oats, rice, plums, cherries, have in turn been accepted as 

 " polyphyletic " or, in other words, derived from several distinct 

 forms. The reason that has led to these judgments is that the dis- 

 tinctions between the chief varieties can be traced as far back as the 

 evidence reaches, and that these distinctions are so great, so far 

 transcending anything that we actually know variation capable of 

 effecting, that it seems pleasanter to postpone the difficulty, relegat- 

 ing the critical differentiation to some misty antiquity into \Ahich we 

 shall not be asked to penetrate. For it need scarcely be said that 

 this is mere procrastination. If the origin of a form under domesti- 

 cation is hard to imagine, it becomes no easier to conceive of such 

 enormous deviations from type coming to pass in the wild state. 

 Examine any two thoroughly distinct species w^hich meet each other 

 in their distribution, as for instance. Lychnis dlurtm and vespertina 

 do. In areas of overlap are many intermediate forms. These used 

 to be taken to be transitional steps, and the specific distinctness of 

 vespertina and d'mrna was on that account questioned. Once it is 

 known that these supposed intergrades are merely mongrels between 

 the two species the transition from one to the other is practically be- 

 yond our powers of imagination to conceive. If both these can sur- 

 vive, why has their common parent perished? Why, when they 

 cross, do they not reconstruct it instead of producing partially sterile 

 hybrids? I take this example to show how entirely the facts were 

 formerly misinterpreted. 



When once the idea of a true-breeding — or, as we say, homo- 

 zygous — type is grasped, the problem of variation becomes an insist- 

 ent oppression. What can make such a type vary? We know^, of 

 course, one way by which novelty can be introduced — by crossing. 

 Cross two well-marked varieties — for instance, of Chinese Primula — 

 each breeding true, and in the second generation by mere recombi- 

 nation of the various factors which the two Darental types severally 

 18618°— SM 1915 24 



