NATURAL SELECTION 



657 



with, on the whole, progressively more 

 born dead with inciease in size of litter. 

 Other inbred famihes showed an interme- 

 diate range of mortality at birth. Control 

 stocks not inbred showed relatively Uttle ef- 

 fect of litter size, the percentage born alive 

 averaging between 80 and 90 up to litters 

 of five. 



It might be expected that large litters 

 would regularly be at a disadxantage com- 

 pared with small ones (p. 396). But a 

 small litter in guinea pigs may be an indi- 

 cation of unfavorable conditions that may 

 more than offset the advantage derived 

 from small numbers. The relation of size 

 of litter to chances of death at birth is 

 probably complex. Not only is there a dif- 

 ferential in inbred families of guinea pigs, 

 but a differential occurs between different 

 groups of animals. Mortality at birth is 

 greater in man, the larger the number of 

 young born at one time. 



During mammalian embryonic life 

 within the uterus or marsupial pouch, 

 and during the nesting period in birds, 

 competition between siblings may be acute 

 (Haldane, 1932, p. 124). At the same 

 time, the number of young is in general 

 inversely correlated with the amount of 

 postembryonic parental care and protection 

 (p. 701). 



It may here be mentioned that the nest- 

 ling cuckoo, parasitizing the family in- 

 stincts of its foster-parents, ehminates the 

 eggs or young that share its nest by using 

 its hollowed back to shove its competitors 

 out of the nest (Beebe, 1944, p. 16; 

 Baker, 1942). No sibling nestlings have 

 evolved such a mechanism for doing away 

 with conspecific nest competitors— an adap- 

 tation that would be advantageous for the 

 individual, but harmful for the species. 

 This instance indicates that, in some cases 

 at least, the relation of competition and 

 cooperation within a species and between 

 species is at least quantitatively different, 

 and this differential sets up selection pres- 

 sures that affect the evolution of intraspe- 

 cific adaptations, as contrasted with inter- 

 specific adaptations (see p. 683). 



In pure stands of plants, aggregated ani- 

 mals, and in social groups, one might imag- 

 ine an increase in intraspecies competition, 

 while in solitary and scattered forms other 

 species might exert a stronger competitive 

 pressure. Complete elimination is more fre- 



quent in mixed cultures of plant species 

 than in cultures of different strains of the 

 same species (Salisbury, 1936). In trees 

 there seems to be a tendency toward stands 

 of a few or single species in temperate 

 forests under more severe conditions, while 

 tropical forests have a much larger num- 

 ber of species, with fewer individuals of 

 any given species per unit of area. A gra- 

 dation in numbers of species also occurs 

 between continental and island biotas. Dar- 

 lington (1943) stated that large areas ap- 

 pear to be inhabited by many species of 

 carabid beetles with sparse, unstable popu- 

 lations, while small areas have fewer species 

 in denser, more stable populations. See 

 Elton (1946), Williams (1947), and Crom- 

 bie (1947), for discussions of interspecies 

 competition in communities compared to 

 faunas. 



Recently experiments have demonstrated 

 some of the simpler aspects of competitive 

 relations. Cause (1934a) observed unicellu- 

 lar organisms under controlled conditions in 

 order to estabHsh types of competitive rela- 

 tionships. He says: "The competition be- 

 tween two species for a common place in 

 the microcosm may be either (1) a com- 

 petition for a certain fixed and Umited 

 amount of energy, or (2) a competition for 

 a source of energy kept continually at a 

 certain level." To these may be added 

 competition for other niche factors (p. 

 271). 



In order to investigate the first of these 

 problems. Cause experimented with two 

 species of yeast cells producing alcoholic 

 fermentation ( Saccharomyces cerevisiae and 

 Schizosaccharomyces kephir). He calcu- 

 lated the coeflBcients of multipUcation in 

 these species and the factor that hmited 

 their growth (alcohol production). He then 

 evaluated the coefficients of the struggle 

 for existence (alcohol production per unit 

 of yeast volume), and correlated these 

 parameters in the form of an equation of 

 the struggle for existence (Volterra, 1926; 

 Cause, 1932a), and obtained an agreement 

 in general features with the observed 

 growth of a mixed population. Further ex- 

 periments showed that under slightly dif- 

 ferent conditions (a greater content of oxy- 

 gen in the nutritive medium) the com- 

 plicating effect of the by-products of fer- 

 mentation decreased. The forecasts of the 



