614 



ECOLOGY AND EVOLUTION 



graphic area concurrently. On the other 

 hand, it cannot be presumed that diver- 

 gence of these two species would have 

 taken place without the habitat separation. 



If monophagous or ohgophagus groups 

 are richer in number of species than poly- 

 phagous groups, host-plant isolation may be 

 presumed. Petersen (1932) illustrates this 

 relationship among certain genera of mi- 

 crolepidoptera. 



Host isolation between closely related 

 phytophages on taxonomically unrelated 

 plants is illustrated by three species of 

 beetles of the genus Coenomjcha. All three 

 species are nocturnal, but each feeds and 

 mates on its host plant, C. bowlesi on Arte- 

 mesia tridentata, C. testacea on Eriogonum 

 fasciculattim, and C. ampla on Juniperus 

 californicus (Tilden and Mansfield, 1944). 



The two phases of heterecious species of 

 rusts* are usually found on quite unrelated 

 hosts (Arthur, 1929). Some species have a 

 wide range of hosts for the gametophytic 

 phase and a restricted range for the sporo- 

 phytic phase. Mobile others show the re- 

 verse relation. Pticcinia graminis in the 

 gametophytic phase occurs on a few spe- 

 cies of the barberry (Berberis) and Ma- 

 honia, while its sporophytic phase is found 

 on ninety-eight species of thirty-five genera 

 of Poaceae in North America alone. Ptic- 

 cinia subnitens in the sporophytic phase is 

 largely confined to one species of grass, 

 Distichlis spicata, while its gametophytic 

 phase occurs on ninety species of sixty-four 

 genera and twenty-four families. Restricted 

 hosts for both phases are found in P. ano- 

 mdla, with its gametophytic phase on Or- 

 nithogaliim and its sporophytic phase on 

 cultivated barley. An autecious species may 

 occur on a single genus of plants— for ex- 

 ample, P. cirsii occurs on forty-one species 

 of Cirsium—OT may be found on only one 

 species of host— for example, P. stiaveolens 

 on Cirsium arvense. Many so-called phys- 

 iologic races have developed within mor- 

 phologic species, and these may show 

 greater host restriction than the species as 

 a whole. Such forms, when reproductively 

 isolated and when genetically distinctive 

 through their phvsiologic reactions to their 

 hosts, are evidently species rather than 

 races (p. 625). It is well known, however, 



• The nomenclature of the complex life cycle 

 of rusts may be found in any general text- 

 book of botany. 



that in some instances the same race may 

 infect a number of host species and even 

 genera. 



Varieties of wheat, barley, and oats may 

 be detected by their reactions to certain 

 species of rusts* and conform to the classi- 

 fication based on serology. Immunity of a 

 host is more likely to occur in relation to 

 a rust specialized for its host than to rusts 

 with a wide variety of hosts. 



Parasites can be such sensitive biologic 

 testing agents as to diflFerentiate groups 

 otherwise indistinguishable. It is fairly ob- 

 vious that genetic changes involving subtle 

 physiologic relations on the part of either 

 the host or the parasite may be subject to 

 habitat isolation and selection, thus influ- 

 encing the evolution of both the hosts and 

 their parasites. 



In rusts, the basidiospore is generally 

 incapable of living on a host that bears 

 the diplont (2 N chromosomes) thallus of 

 a species infecting two host species (hetere- 

 cious) (Jackson, 1931). However, the hap- 

 loid basidiospores may sometimes, though 

 rarely, have acquired the ability to infect 

 the host normally only susceptible to in- 

 fection by diploid uredospores or aecidio- 

 spores, in this manner establishing the hap- 

 loid (1 N chromosomes) thallus on the 

 normal host species for the diplont. Me- 

 lampsora amijgdalinae on the willow 

 (Salix), Gymnosporangiiim bermudianum 

 on the juniper (Juniperus), Synomyces 

 reichei on Sterna, and Puccinia komarovi 

 on the jewel weed (Impatiens), are ex- 

 amples. The origins of these species of 

 rusts thus appear to be excellent examples 

 of habitat isolation. 



Ewing (1933, 1938) reports that four 

 species of lice, closely related to Pedicuhis 

 humaniis americanus found on the Ameri- 

 can Indian, are found respectively on three 

 species of spider monkeys (Ateles) and on 

 the saki (Pithecia monachus) from South 

 and Central America. The human louse dif- 

 fers so slightlv from these four species that 

 it is assumed that they did not evolve with 

 the evolution of the monkey hosts, but 

 rather that the lice transferred from man to 

 the monkeys when the American Indian in- 

 vaded South America. If this surmise is cor- 



* Festuca elatior has been distincuished from 

 F. pratensis on the basis of the differential re- 

 action of Puccinia phlei-pratensis, a physiologic 

 form of P. graminis. 



