|
<br />170
<br />
<br />DOUGLAS ET AL.
<br />
<br />these, in turn, may change during ontogeny so as
<br />to become less discriminating. Thus, application
<br />of a few discriminating adult characteristics (as in
<br />Table 1) may not be reliable enough to segregate
<br />juveniles.
<br />With regard to our third hypothesis, putative hy-
<br />brids were generally assigned to either one or the
<br />other of their parental forms. Hybrids included
<br />specimens of G. seminuda (one of which was as-
<br />signed to G. cypha rather than G. elegans). If the
<br />premise that these are indeed "hybrids" is ac-
<br />ceptable, then our results appear intuitive and the
<br />hypothesis of hybrid intermediacy cannot be re-
<br />jected (in spite of the fact that DeMarais et al.
<br />[1992] clearly demonstrated that G. seminuda was
<br />phenotypically more like G. robusta than G. ele-
<br />gans).
<br />
<br />Choice of Characters and the "Art of Seeing
<br />Well"
<br />
<br />When meristic data were evaluated, dorsal or
<br />anal rays appeared diagnostic in virtually every
<br />species comparison (Table 1). Yet, surprisingly,
<br />these characters were of little assistance in field
<br />application, being discriminatory in only one com-
<br />parison. We also note in this context that vertebral
<br />counts and numbers of gill rakers were equally as
<br />discriminating as the former.
<br />However, morphological data provided greatest
<br />discriminating power in segregating species. This
<br />may be because morphometric characters best de-
<br />scribe shape and shape change, and these aspects
<br />are paramount in discriminating among big-river
<br />Gila. Morphological characters also formed 71 %
<br />of the total number of characters used in our anal-
<br />yses, so their ascendancy may also be related to
<br />their numerical superiority. Nevertheless, it is
<br />clear that the three Gila in this report can easily
<br />be assigned to species, based upon a variety of
<br />counts and measurements. So, in a sense, we are
<br />troubled that there is such a history of confusion
<br />among fishery managers with regard to this species
<br />complex, particularly given results in Table I and
<br />Figure 2. We can only conclude that this bewil-
<br />derment is a manifestation of those perspectives
<br />offered by Holden (1991, see above). Douglas et
<br />al. (1989) touched upon a similar point when they
<br />stated
<br />
<br />From our perspective, some confusion shrouding
<br />identification of problematic Gila hinges upon appar-
<br />ent misunderstanding of the populational nature of
<br />natural selection (a statistical concept). The conser-
<br />vative nature of fisheries science, with reliance on an
<br />idealized morphological archetype to represent a spe-
<br />
<br />cies, contributes substantially to this problem. Vari-
<br />ation within and among populations and species must
<br />be recognized as a natural phenomenon amenable to
<br />statistical analysis (as herein), rather than as an in-
<br />frequent occurrence to be either redefined or dis-
<br />missed.
<br />
<br />The failure of fishery managers to accept this
<br />premise on a broad basis, coupled with an histor-
<br />ical bias against native fishes in western North
<br />America, has both introduced and perpetuated the
<br />confusion surrounding this group of fishes. In our
<br />analyses, it appears as if answers to species dis-
<br />tinctness hinge upon selection of characters and
<br />adequacy of analyses (Douglas 1993), coupled
<br />with "the art of seeing well," which was defined
<br />by Rafinesque (1820) as "the art . . . of noticing
<br />and distinguishing with accuracy the objects which
<br />we perceive. . . [It] is a high faculty of the mind,
<br />unfolded in a few individuals, and despised by
<br />those who can neither acquire it nor appreciate its
<br />results. "
<br />
<br />Acknowledgments
<br />
<br />M. L. Smith and W. D. Sable collected primary
<br />data for this report. Their perseverance and atten-
<br />tion to detail are greatly appreciated. Manuscript
<br />review was provided by M. R. Douglas. This paper
<br />is dedicated to the memory of Frances Hubbs Mil-
<br />ler, whose able assistance and acumen for nearly
<br />half a century propelled many ofR.R.M. 's research
<br />projects to culmination.
<br />
<br />References
<br />
<br />Bosley, C. E. 1960. Pre-impoundment study of the
<br />Flaming Gorge Reservoir. Wyoming Game and Fish
<br />Commission Technical Report 9: 1-81.
<br />DeMarais, B. D., T. E. Dowling, M. E. Douglas, W. L.
<br />Minckley, and P. C. Marsh. 1992. Origin of Gila
<br />seminuda (Teleostei: Cyprinidae) through intro-
<br />gressive hybridization: implications for evolution
<br />and conservation. Proceedings of the National
<br />Academy of Sciences 89:2747-2751.
<br />Douglas, M. E. 1993. An analysis of sexual dimorphism
<br />in an endangered cyprinid fish, Gila cypha Miller,
<br />using video image analysis. Copeia 1993:334-343.
<br />Douglas, M. E., W. L. Minck1ey, and H. M. Tyus. 1989.
<br />Qualitative characters, identification of Colorado
<br />River chubs (Cyprinidae: genus Gila) and the "art
<br />of seeing wel1." Copeia 1989:653-662.
<br />Douglas, M. E., G. D. Schnell, and D. J. Hough. 1984.
<br />Differentiation between inshore and offshore spot-
<br />ted dolphins from the eastern tropical Pacific Ocean.
<br />Journal of Mammalogy 65:375-387.
<br />Douglas, M. E., G. D. Schnell, D. J. Hough, and W. E
<br />Perrin. 1992. Geographic variation in cranial mor-
<br />phology of spinner dolphins Stenella longirostris in
<br />the eastern tropical Pacific Ocean. Fishery Bulletin
<br />90:54-76.
<br />
|