EVOLUTION 419 



The problem of the relative proportions of the selective and non-selective 

 death-rates is seen to be a difficult one, at any rate when we deal with wild 

 life and not life in a state of captivity. It is equally difficult to specify, 

 owing to the principle of correlation, the particular organ or organs for which 

 selection is taking place. We can, however, usually find the relative effect 

 of the selective death-rate on organs of different deviations from the mode ; 

 we can also determine the organ fittest to survive, i.e. the organ for which 

 the selective death-rate vanishes ; it will not necessarily coincide with the 

 modal value before or after selection. The fittest to survive is not identical 

 with the surviving type. 



While we have seen that selection can account for progressive changes in 

 living forms, we need for the origin of species first differentiation into local 

 races, and then provision against intercrossing until the principle of correla- 

 tion has rendered mutual fertility mechanically or physiologically impossible. 

 The correlation between fertility and other characters ought to be relatively 

 high, but it has been hitherto little studied ; the relative fertility of different 

 members of the same race in sc and inter se has also been so far insufficiently 

 investigated. These may be said to be unsolved problems in the theory of 

 evolution. Various hypotheses — isolation, recognition marks, physiological 

 selection — have been propounded with a view to providing the necessary 

 barrier to intercrossing. Possibly asexual propagation and self-fertilisation at 

 an early stage of organic development, and endogamy and homogamy at a 

 later, may have had as much influence as any of the above factors in keeping 

 divergent types true, until by the principle of correlation differentiation in 

 reproductive organs, or functions, following the differentiation of the selected 

 organs, had brought about relative infertility. It is not absence of explana- 

 tion, but rather of the quantitative testing of explanations, which hinders at 

 this point the development of the Darwinian theory. 



LITERATURE 



Darwin, Charles. — The Origin of Species, London, 1859. The Variation 

 of Animals and Plants under Domestication, London, 1868. The 

 Descent of Man, and Selection in Relation to Sex-, London, 1871. The 

 Effects of Cross and Self- fertilisation in the Vegetable Kingdom, 

 London, 1S76. The Different Forms of Flowers on Plants of the same 

 Species, London, 1877. 

 These hardly required citing, but the reminder may be given that Darwin 



always writes so that he is intelligible to the reader of average thoughtful- 



ness ; the study of his works does not need a specialist's training. 



Romanes, G. J. — Darwin and the Postdarwinians, London, vol. i. 1892, 

 vol. ii. 1895, vol. iii. 1 897. 



Wallace, A. R.— Darwinism, London, 1889. 



Bateson, W. — Materials for the Study of Variation, London, 1S94. 



The whole of these three works deserve study, but it must be in the 



