334 THE VARIATION OF ANIMALS IN NATURE 



gratuitously large or complex as to embarrass and be a positive 

 hindrance or danger to the owner ; but we cannot always 

 affirm that there are no compensating adjustments. Thus in 

 many species of the African Land Snail Ennea the aperture 

 of the shell is filled up with such a dense palisade of denticles 

 that it seems that the owner can hardly emerge. The difficulty 

 of emergence past this palisade must be very great in any case 

 and can be overcome only by movements that call for peculiar 

 modifications. 



It seems that for the cases of extravagant growth we have 

 at least four explanations, viz. : (i) The direct adaptive value 

 of the excessive growth, (2) Huxley's theory based on the facts 

 of heterogony, (3) Fisher and Haldane's theory of the effect of 

 selection on a metrical character determined by many genes, 

 and (4) the theory of an internal impulse. 



(1) Haldane's theory of accelerated development (p. 328) 

 during inter-uterine competition was not specifically framed to 

 include rapid growth as distinct from rapid differentiation. It 

 has, however, been adopted in this sense by Castle (1932), 

 who has produced some evidence in its favour. It is possible 

 that some increase of total body-size may be due to selection 

 favouring larger and more powerful embryos and also young 

 in the post-natal stage. But the theory can scarcely be used 

 by itself to explain (a) the exaggerated size of the adult seen in 

 some species, and (b) the size of individual parts used in adult 

 life (e.g. the canine teeth of Machaerodus) . 



We may next consider from the adaptive point of view 

 some individual instances of the excessive growth of parts in 

 the adult phase. 



(a) Matthew (1901, 1910), in his study of the excessive 

 growth of the canines in the Machaerodont Tigers, objected 

 to the theory of an internal momentum. From a study of the 

 associated parts he affirms that these large teeth were made for 

 a stabbing or gashing stroke and suggests that in the absence of 

 the lighter, thinner-skinned animals that provide the prey of 

 the modern Felidae the mid-Tertiary Machaerodonts preyed on 

 the heavy, thick-skinned Pachyderms of various groups which 

 could be attacked only in this way, and that their extinction 

 was not due to the excessive growth of the canines, as has been 

 suggested, but to the extinction or localisation of their normal 

 prey. But quite apart from the difficulty of ascertaining 



