670 



ECOLOGY AND EVOLUTION 



geographic species and forms of Bematistes 

 (Acraeinae) and forms of Pseiidacraea 

 eurytus (Nymphalinae). Carpenter regards 

 Pseudacraea as a transition from warning 

 to mimetic coloration, the mimics being 

 distasteful, though less so than are the 

 models they resemble (pseudaposematic). 

 When models are relatively scarce in cer- 

 tain areas, no one form of Pseudacraea has 

 any particular survival value over another, 

 and nonmimetic intermediates are more 

 common in the population. Tliis relative 

 abundance of nonmimetic forms in relation 

 to populations of models is particularly 

 evident in collections from the Kome Island 

 group made in 1914 and 1918-1919 (see 

 Table 54), during which time there was a 

 great fluctuation in the abundance of the 

 model species. When models are abundant, 

 proportionately fewer nonmimetic interme- 

 diates are to be found, and the forms that 

 match the models are proportionately more 

 common. Greater variation in the mimick- 

 ing species is to be expected in niches in 

 which models are absent (Ford, 1936). If 

 models became more abundant in a given 

 region, the mimics would be expected to 

 increase in numbers in direct proportion 

 (Carpenter, 1936; Goldschmidt, 1945). 



Carpenter (1936a) supports the theory 

 of natural selection, as applied to mimetic 

 resemblance, by a number of propositions 

 that expand and augment Wallace's rules. 

 These propositions may be summarized as 

 follows : 



1. Warningly colored models are well de- 

 fended against predators. 



2. Education of predators to avoid the warn- 

 ingly colored models is facilitated when a 

 number of unrelated species resemble each 

 other and all have protective adaptations 

 ( MuUerian mimicry ) . 



3. Deceptive mimicry (Batesian mimicry) 

 overextended in numbers of individuals tends 

 to defeat itself, so that Miillerian mimicry, by 

 reenforcement of selection through greater 

 numbers, is more common. 



4. Inedibility is not absolute, but relative. 



5. Mimicry is best exemplified when there is 

 an abundance of other food for the predators, 

 since scarcity of food tends to reduce relative 

 immunity. 



6. Resemblance of a mimicking species is 

 superficial and visual, not necessarily funda- 

 mental and anatomical, and diverse methods 

 ot mimetic resemblance are known. 



7. Mimicry in polymorphic species is always 

 resemblance to warningly colored models. 



allows a beneficial versatility, and may be 

 controlled in its expression by single genes 

 (p. 688). 



8. Chance cannot explain the nicety of 

 gcograpliic correlation of models and mimics. 



9. Incidence of nonmimetic forms of a 

 normally mimetic species rises as the number of 

 models decreases. 



10. Mimicry is a phenomenon too complex to 

 be explained by parallel mutation. 



11. Enviromnental influences other than pre- 

 dation do not explain mimicry— the larvae of 

 mimic and model usually develop in different 

 habitats, and certain models (Lycidae) are 

 mimicked by diverse insect orders the world 

 over. 



As noted previously, probably the weak- 

 est part of the theory of mimicry is the lack 

 of statistically significant survival experi- 

 ments, with appropriate predators, on pairs 

 or series of mimics. 



Goldschmidt (1945) has reviewed many 

 problems of the evolution of mimicry in a 

 scholarly manner. He favors the idea that 

 mimetic resemblance is produced by a 

 single or simple gene mutation rather than 

 by gradual development of a complex gene 

 pattern with potentialities that might be 

 initiated or inhibited during development 

 by a single gene. The complexity of the 

 adaptation involved in most mimicry indi- 

 cates that there would be httle chance of 

 such detailed adjustment except through 

 a polygenic pattern (E. B. Ford, 1937; see 

 also pp. 632, 648, 688). 



A phenomenon rather closely related to 

 that of mimicry involving predatory selec- 

 tion is the egg-mimicry found among the 

 various species and genera of parasitic 

 cuckoos in the Old World. Although not 

 always exact, a large proportion of cuckoo 

 eggs match the eggs of the normal host 

 birds remarkably in color, markings, and to 

 some extent in size. Baker (1942) explains 

 this mimicry as the result of natural selec- 

 tion and provides reasonable proof for the 

 following assertions: the eggs of any indi- 

 vidual cuckoo are all closely ahke; species 

 of foster parents regularly victimized are 

 few in number in all areas; the eggs laid by 

 cuckoos often show close resemblance to 

 the eggs of the foster parent; the foster 

 parents can and often do discriminate be- 

 tween the eggs of the cuckoo and their 

 own; the greater the diflFerence between the 

 eggs, the more easily are they recognized 

 and the greater is the discrimination exer- 



