NATURAL SELECTION 



693 



Proof that such a selection of postrepro- 

 ductive or sterile individuals influences evo- 

 lution is amply demonstrated in the phy- 

 logeny of the social insects. The most con- 

 clusive case is in the adaptive evolution of 

 the nasute soldier of the termites (Figs. 148 

 and 149). This soldier caste is completely 

 sterile (the rare abnormal alate-soldier in- 

 tercaste described by Adamson, 1940, does 

 not aflFect this conclusion). The nasute sol- 

 dier has evolved from mandibulate soldier 

 types (Fig. 263), and the evidence points 

 to the conclusion that this evolution is to- 

 ward increased defensive adaptation with 

 associated regressive evolution of the man- 

 dibles. The eflFective function of the soldier 

 in the defense of the colony helps select 

 the queen capable of producing such sol- 

 diers. The genetic characteristics of the re- 

 productives may thus be sorted through 

 selective survival as efiFectively as the 

 genetic characters of the gametes are 

 selected through the function of the sterile 

 somatic cells and organs incorporated with 

 the gametes within the multicellular in- 

 dividual. Consequently, it is necessary to 

 recognize the selection of whole integrated 

 population units in order to understand 

 adequately well-known facts of evolution. 



We have already discussed (p. 676) the 

 regression of supraorganismic adaptations, 

 a parallel to regressive evolution in individ- 

 ual organisms that emphasizes the analogy 

 of organism and supraorganism. Some 

 authors think that it is philosophically un- 

 sound to derive any valid scientific mean- 

 ing from the analogies between two dif- 

 ferent integrative levels. According to this 

 viewpoint, integrative mechanisms within a 

 unicellular organism yield no knowledge 

 applicable to integrative mechanisms be- 

 tween cells of a multicellular organism; and 

 comparisons of multicellular coordination 

 with social coordination (either insect or 

 human) is deemed fallacious. Similarities 

 between two levels are regarded as "purely 

 formal and therefore meaningless" (Novi- 

 koff, 1945, 1945a). Simpson (1941, p. 18) 

 says: "The biologist who elevates the or- 

 ganism-epiorganism metaphor into a stand- 

 ard for social interpretation and recom- 

 mendation is guilty of the most reckless, 

 unjustified, and nonscientific extrapolation." 



It is true that in comparing distinct in- 

 tegrative levels, homologies (resemblances 

 with a common genetic basis) cannot be 



assumed, although we have seen (pp. 689, 

 690) that the identical mechanisms of one 

 level may sometimes integrate a higher 

 level. Similarities may, however, be con- 

 vergent and therefore analogous, and a 

 comparison of similar pressures guiding 

 unlike organisms toward analogous func- 

 tions is significant. The principles that 

 order similarities and difiFerences may be 

 formulated by study, comparison, and eval- 

 uation of the data. Much scientific re- 

 search is based upon this observational 

 and analytic method. Care must be taken 

 to study truly functional similarities and 

 diS^erences, and not to be led astray by ver- 

 balisms, euphonious metaphors, or purely 

 chance resemblances (Gerard and Emer- 

 son, 1945; Schneirla, 1946). 



The theory of emergent evolution has 

 been applied to the concept of organismic 

 levels. This theory recognizes that new or 

 novel properties and characteristics emerge 

 from new combinations. Complex associa- 

 tions have properties that are not merely 

 the sum of the properties of the constituent 

 parts (Jennings, 1927; Wheeler, 1928a; 

 Nabours, 1930, 1930a; Morgan, 1933; 

 Wright, 1935; Needham, 1943). Wheeler 

 emphasizes the fact that emergent char- 

 acteristics may be losses as well as gains, 

 so that the whole is not necessarily more 

 than the sum of its parts, but may be less. 



Some proponents of the theory of emer- 

 gent evolution state that the novel prop- 

 erties arising from interaction are funda- 

 mentally unpredictable from a knowledge 

 of the unassociated parts. This philosophi- 

 cal aspect of the theory is beyond our field 

 of enquiry. In essence, emergent evolution 

 emphasizes the basic necessity for the 

 study of wholes, as contrasted to the study 

 of parts, and adds a certain dignity to 

 synthetic sciences. Biology is the study of 

 the properties of whole systems as well as 

 of parts, and ecology, among the various 

 subsciences of biology, tends to be holistic 

 in its approach. 



Human social evolution is beyond the 

 scope of this book. Biological evolution in- 

 volves germinal changes. Social evolution 

 of man involves cultural changes. We hold, 

 however, that the social scientist may find 

 many significant parallels in biological and 

 social evolutionary mechanisms (Brodv, 

 1944; Emerson, 1947; see pp. 630, 632, 

 639, 686, 691). We also think that human 



