296 AN INTRODUCTION TO MODERN GENETICS 



portance to random fluctuations in gene ratios. The whole set of 

 theoretical investigations provides a satisfactory picture of many of the 

 most important characteristics of evolution. We have an account of the 

 gradual transformxation of widespread species, by Haldane selection of 

 new mutations; of the production of many apparently non-adaptive 

 varieties in small groups or groups with localized inbreeding; and of 

 the comparatively rapid emergence of a new species, which may found 

 an important new genus or family, when a new type of adaptation be- 

 comes possible through the occurrence of a new gene or the production 

 of a new ecological niche. 



The main phenomenon for which some explanation in terms of 

 natural selection is required but is not yet provided is trend-evolution 

 of the type shown by the Gryphaeas (p. 242). The difficulty is not so 

 much the evolution in one direction over a long period of time. That 

 might be brought about by selection in a gradually changing environ- 

 ment or it might be due to a sort of inertia which we should expect to 

 find in evolution; granted that a character is dependent on the inter- 

 action of many genes, it will be easier to continue a line of evolutionary 

 change, for which many of the modifiers are already present, than to 

 start off on a completely new line. Thus the uni-directional nature of 

 trend evolution is not particularly surprising. What is remarkable is 

 that the trend continues so far as to lead to the extinction of the race. 



Fisher suggested that in certain circumstances the genotypic inertia 

 may be manifested in another way and if the gene ratios are changed 

 by selection, they may overshoot the equilibrium point at which the 

 population would be best adapted. It is conceivable that it was some- 

 thing of this kind which caused orthogenetic trends to go so far as to 

 lead to the extinction of the species. Haldane has suggested another 

 possibility, based on the supposition that there was stringent selection 

 in young stages ; selection for strength in the young might, for instance, 

 favour individuals with a heterogonic growth mechanism which could 

 not be controlled in later development and produced adults with some 

 parts excessively developed. But it is not clear whether this intra- 

 specific selection could lead to the extinction of a species in competition 

 with the other species in its habitat. 



It has sometimes been suggested that trends occur, not directiy under 

 the control of natural selection, but by progressive mutation in a single 

 direction. Such a directed series of mutations would be as inexplicable, 

 on our present ideas, as the trends themselves; and further, it is easy to 

 see, from the discussion on p. 290, that mutation pressure cannot be 



