CAUSES OP THE EVOLUTION AND EXTINCTION OF THE TITANOTHERES 



845 



directly produces any modification of head propor- 

 tion. He implies that such modification, if natural, 

 would show only after thousands of years of residence 

 in a single locality, during which the indirect influ- 

 ence of change of feeding habits might play an impor- 

 tant part. He continues: 



I do not believe for a moment in Boas's observations [of 1911] 

 on the direct and rapid influence of environment in modifying 

 ontogenic head form. In my opinion the conditions of doli- 

 chocephaly and braohycephaly in man must have developed 

 very slowly through exceedingly long periods of time and in 

 widely separated areas, amidst widelj' different environments. 

 Brachycephaly has thus become especially distinctive of the 

 Central Asian high plateau populations, doUchocephaly of the 

 littoral and plains-dwelling peoples. But these unit characters 

 are now so fixed that environment is powerless to modify them 

 in a thousand 3'ears or so. 



The observations of Hrdlicka (1910.1, p. 214) on 

 the forms of the cranial vault and the face of the 

 Eskimo lead us to believe that they are attributable 

 rather to feeding habits than to the direct action of 

 environment. 



Apart from the difTerences of opinion as to the 

 direct influence of environment on head form as dis- 

 tinguished from the modifications of habit and the 

 influence of human selection, there is absolute una- 

 nimity both of evidence and of opinion as to the con- 

 tinuity of allometric evolution in the human species 

 which establishes different extremes of head form 

 under conditions of geographic and social isolation. 



GERMINAL AllOMETEOFS APPARENTLY INFLUENCED BY DIRECT ACTION 

 OF ENVIRONMENT 



The dwarfing as well as the gigantism that arise 

 continuously or suddenly are usually harmonic — that 

 is, all parts are contracted or enlarged harmoniously 

 owing to generally unfavorable or highly favorable 

 environmental conditions. This is generally true of 

 dwarfed breeds that arise through long-continued 

 exposure to a dwarfing environment. For example, 

 the Celtic pony {E. caballus celticus Ewart) appears 

 to be a harmonic dwarf of its remote Arab-like ances- 

 tor. Doubtless disharmonic dwarfing and gigantism 

 arise under the prolonged direct action of the environ- 

 ment, but at present no examples can be cited. 



It would be interesting to ascertain through experi- 

 ment whether there is any relation between the har- 

 monic dwarfing attributable to a prolonged unfavor- 

 able environment on successive generations and that 

 induced experimentally in a few months by removal 

 of the thyroid or parathyroid glands, as observed in 

 dogs and sheep. It does not appear probable that any 

 relation will be found, because thyroid and parathy- 

 roid dwarfing is apt to be disharmonic, causing the 

 acceleration of some parts and the general retardation 

 of others. 



GERMINAL ALLOMETEONS, FLUCTUATING AND NONFLUCTUATING 



The Darwinian theory, as we have shown above, 

 depends upon the continuous natural selection of 

 "fluctuating variations" that are continous in a given 

 direction. It appears to be highly probable as to 

 fluctuations that have high survival selection value, 

 such as fluctuations in the length in the entire series 

 of cervical vertebrae in the neck of the giraffe, an 

 evolution which is generally harmonic. Harmonic 

 allometrons generally, such as length of neck, length 

 oi limb, length of feet, may possess high survival 

 selection value and favor the Darwinian theory. 



The causation of disharmonic allometrons, which in 

 the titanothere skull are far more numerous than the 

 harmonic, presents, on the contrary, great theoretic 

 difficulties. The abbreviation of the face coordinately 

 with the elongation of the cranium is an example of 

 disharmony. The separate velocity of ontogenetic 

 and phylogenetic movement of a very large number of 

 biocharacters is so perfectly coordinated and adjusted 

 in every stage of development and evolution as to 

 make the Darwinian "fortuitous fluctuation" hypothe- 

 sis in a high degree untenable. 



CERTAIN GERMINAL ALLOMETEONS OF HIGH SURVIVAL SELECTION VALUE 



AU parts of the skeleton change in proportion during 

 the entire period of growth, prenatal and postnatal. 

 These changes of proportion are partly paleotelic — 

 that is, they are reversive to ancestral stages — partly 

 cenotelic — that is, they are adapted to existing con- 

 ditions. 



Certain prenatal germinal proportions exhibit espe- 

 cially high survival value; they are examples of 

 perfect mechanical adjustments of form that rise 

 from the germ before they have been developed by 

 individual modification. Apparently fatal to the 

 Lamarckian theory that all aUometrons are due to 

 the inheritance of modifications acquu-ed postnatally 

 are the many well-lvnown instances of prenatal allo- 

 metrons, such as those of certain precocious birds — 

 the Praecoces, for example — which are adapted to 

 run immediately on leaving the shell. Certain pre- 

 cocious desert-living cursorial mammals are also 

 capable of rapid movement immediately after birth. 

 The limbs of such animals have before birth the 

 proportions of those of adults. Conspicuous among 

 these are the horses. In the feral state the foals 

 must be able immediately after birth to keep up 

 with the herd, and in adaptation to this necessity 

 the members that contribute most to speed are pre- 

 natally so much accelerated iu development that 

 they attain almost their full adult length before birth. 

 Thus in thoroughbred horses the postnatal increase 

 in the length of the cannon bones is only 3 centimeters 



