638 



ECOLOGY AND EVOLUTION 



The relationship between linear, surface, 

 and volume measurements of an organism 

 must be in a balanced functional propor- 

 tion (p. 131). D'Arcy Thompson points out 

 that a tenfold increase of man's linear di- 

 mensions would make it impossible for 

 him to support his own weight, since the 

 cross section of a thighbone would increase 

 a hundredfold and weight would increase 

 a thousandfold (Huxley, 1942, p. 494; see 

 also Hiestand, 1928). This functional as- 

 pect of proportions doubtless sets up selec- 

 tion pressures that guide the evolution of 

 balance and compromise (Thompson, 

 1917). 



Seemingly intrinsic nonadaptive evolu- 

 tionary tendencies sometimes leading to ex- 

 tinction have been termed orthogenetic 

 (see Simpson, 1944, p. 150, for summary 

 discussion). Orthogenesis involving an in- 

 nate tendency for genes to mutate in a 

 given direction may be possible in a few 

 sequences of alleles, but it is doubtful 

 whether such a series is to be found in 

 paleontological sequences (Simpson, 1944, 

 p. 156), which would be expected to re- 

 sult from much more complex genetic 

 changes. Orthogenesis should be distin- 

 guished from allometric constancy in phy- 

 logenetic series (pseudo-orthogenesis) in 

 which genetic change of an adaptive pri- 

 mary character is genetically and physiolog- 

 ically related to allometric secondary char- 

 acters; and it must also be distinguished 

 from evolution in a single adaptive direc- 

 tion, which may be explained by a constant 

 selective pressure in a single type of en- 

 vironment. 



Horses seem to have evolved increased 

 adaptation to life on grassy plains through 

 more efficient cursorial mechanisms and 

 better adjustment to feeding on siliceous 

 grasses. Directional trends guided by se- 

 lection may be termed orthoselection, and 

 do not imply any intrinsic tendency to 

 evolve in a straight line (Plate, 1913). As 

 might be expected, the evolution of horses 

 exhibits divergent and convergent trends 

 as well as different directional trends (Stir- 

 ton, 1940; Weigelt, 1942; Simpson, 1944, 

 p. 157). 



Through physiological relations of differ- 

 ential growth rates, selection for one char- 

 acter, such as size in deer, may carry with 

 it a secondary allometric character, such as 

 size and form of antlers. There may be 



Httle selection based on the functional dif- 

 ferences of antler size, and yet, with an in- 

 crease in general size, these structures 

 would be expected to evolve in a straight 

 Une from proportionatel)' small to propor- 

 tionately large antlers. An advantage in 

 size might thus be ultimately overcome b^ 

 a disadvantage of overdeveloped antlers, 

 and selection might remove extremely large 

 individuals. An optimum size during the 

 breeding period might have a selective ad- 

 vantage over size greater than the optimum 

 during the postreproductive period, when 

 disproportioned individuals might be elim- 

 inated with Uttle evolutionary effect upon 

 the species. In fact, the death of some in- 

 dividuals may be of advantage to the popu- 

 lation (p. 692). Such a pseudo-orthoge- 

 netic series may be adequately explained 

 within the framework of Darwinian natural 

 selection (Huxley, 1932; Rensch, 1939, 

 1943). 



Most genetic variance, aside from recom- 

 bination, is produced by rare and disadvan- 

 tageous genes the supply of which is kept 

 up by mutation alone, but only in so far 

 as it includes such genes does a popula- 

 tion usually have the genetic elasticity 

 that permits an evolutionary response to a 

 long-term change in environment. Haldane 

 (1932, p. 198) showed by mathematical 

 theory that if selection of a metrical char- 

 acter determined by many genes increases 

 the optimum stature of a species by a 

 certain quantity, then, when the mean 

 stature reaches the new optimum, the in- 

 cidence of some genes will increase even 

 though they are past their point of adaptive 

 balance. The stature will, so to speak, over- 

 shoot the mark, and will thus initiate a 

 selection pressure in a reversed direction. 

 Mutation pressure is itself hereditary and 

 subject to selection (Simpson, 1944, p. 

 175), so that a degree of evolutionary mo- 

 mentum may produce a fluctuation of a 

 character, first above and then below the 

 optimal value. This mechanism may give 

 us an explanation on Darwinian lines of 

 some unadaptive orthogenesis, although it 

 would apply only for a brief time with 

 slight selection pressure. Such a momen- 

 tum effect is probably of little importance 

 in known evolutionary sequences and is 

 unlikely to be the major factor in the ex- 

 tinction of any species. 



Orthogenesis has been assumed to lead 



