8238 - Laserfiche WebLink

Home Browse Search

VICARIANCE AND WESTERN NORTH AMERICAN FISHES 1 sumably respond with change in phenotype as ecological con- ditions change. An ecophenotypic individual or population differs from an ecotype in that a genetic component is not involved (Mayr 1942, p. 28). However, we define ecotypy as an accumulation of genetic differences between populations of the same species living at different points along an ecological gradient, such that they differ from one another in aspects of appearance, life history, or another feature. An ecotype will thus retain its phenotype when placed under new conditions. For this model, body shapes of the various Gila are interpreted as ecotypic re- sponses to local environments (such as conditions in small tributaries as compared with those in larger rivers). Local conditions are presumably identical or similar enough to elicit common ecotypic responses within corresponding habitat patches across a drainage basin and differ from other kinds of patches. Botanists have long recognized that a single plant species may exhibit a variety of forms when grown in different hab- itats. The classic case of ecotypy from which the term was coined was Turresson's (1922) description of four ecotypes occurring throughout the geographic range of hawkweed (Hieracium umbellatum). These forms retained their distinct phenotypes even when cultivated under identical conditions, which lead Turresson to argue that they represented within- species evolutionary differentiation according to habitat. Other demonstrations were provided by Clausen et al. (1948) for yarrow (Achillea millefolium) and Antonovics and Brad- shaw (1970) for several other plants. Is there a difference then between geographic race and ecotype? Dobzhansky (1970, p. 295) considered the two terms largely synonymous, but zoologists rarely speak of ecotypes. He further stated, "it may be advisable to restrict the term ecotype to races that occur mosaic-fashion in a quasi- continuously inhabited territory, whenever a certain type of environment appears." Mayr (1957, p. 378) observed that "purely ecotypic local adaptations are not necessarily good isolating mechanisms because they tend to disappear as soon as the environmental differences disappear." Quinn (1978) reviewed history of the plant ecotype and concluded that it was neither a good evolutionary nor ecological unit. How- ever, West-Eberhard (1989) argued for the evolutionary im- portance of polyphenism (i.e., environmentally cued alter- native phenotypes in a population) as a mechanism of spe- ciation. Ecotypy was first mentioned in the ichthyological literature by Hubbs (1940, 1941) and subsequently referred to by a variety of authors (Beckman 1953; LaRivers 1962; Sigler and Miller 1963). Miller (1945, 1946) applied the term "ecological subspecies" specifically to chubs of the genus Gila in the Colorado River Basin, and Rinne (1976) and DeMarais (1986) discussed and rejected ecotypy relative to the same group of fishes. Model II: Hybridization or lntergradation Discordant morphological variation within a species' range may also stem from hybridization, which is not uncommon among fishes and other externally fertilizing organisms (Hubbs 1955; Grant 1971; Smith 1992) and might have great- er evolutionary importance than previously thought (Arnold 241 1992; Dowling and DeMarais 1993; Dowling and Secor 1997). Our model hypothesizes that nigra, which exhibits discordant (intermediate) morphological variation, originated as a hybrid resulting from contact between robusta and in- termedia (see also Smith 1992, table 1; Minckley, unpubl.). These populations do not currently occupy areas in which parental taxa are sympatric and thus cannot hybridize today, but did so sometime in the past. Introduction of congeneric exotic species and human-in- duced habitat modifications are often provided as causal fac- tors in modern examples of hybridization (Hubbs 1955, 1961; Miller and Smith 1981). In the current situation, human-in- duced perturbations are not involved (unless at the native American level, which is unlikely) because fish of all body shapes were present when ichthyologists first sampled the region (see Rinne 1976). Natural phenomena must have gen- erated similar consequences (Anderson 1949; Hubbs 1955), particularly when drainage transfers through geologic and geomorphic events or habitat alterations resulting from cli- mate change placed allopatric populations in contact. Zones of hybridization are difficult to differentiate from zones of primary intergradation, a problem stemming in part from the definition of hybrid versus intergrade. Mayr (1982, p. 263) stated, "most taxonomists speak of intergradation when two subspecies gradually merge with each other. There is usually found between the two a series of intermediate populations which connect the two extremes perfectly and which are no more variable than other neighboring popula- tions. In a hybrid zone, however, the change from one. . . to another is very abrupt, and in the break area, populations occur which are exceedingly variable. In such hybrid pop- ulations, some individuals may be indistinguishable from ei- ther . . . , but the majority of individuals are intermediate in various character combinations." Intraspecific variability resulting from introgression is widely recognized in Western American fishes (Smith 1966; Smith et al. 1979, 1983; Miller and Smith 1981; Dowling and Secor 1997). Hybridization among and between popu- lations and species may, in fact, have played important roles in their evolutionary histories (Minckley et al. 1986; De- Marais et al. 1992; Dowling and DeMarais 1993). In this study we do not test the hypothesis of relatively ancient origin of a separate taxon through hybridization, as recently dem- onstrated by DeMarais et al. (1992) for the related Gila sem- inuda Cope and Yarrow and proposed for G. nigra by Minck- ley (unpubl.). Nor do we attempt to differentiate between hybridization and intergradation, although the latter is cer- tainly not indicated by the distributional mosaic of parapatric populations of robusta, intermedia, and the morphologically intermediate nigra. Model Ill: Vicariance A vicariant event is the cgeographic separation of a once- continuous biota such that it becomes two or more subunits (Wiley 1981). In. the model applied here, the three forms of Gila are viewed as independent lineages (regardless of ori- gins) whose limits of natural variation overlap. They are hy- pothesized as previously sympatric. Their geographic ranges were, however, fragmented then intermingled as a result of