7022 - Laserfiche WebLink

Home Browse Search

MAYDEN ET AL.-SYSTEMATICS OF WESTERN CYPRINIDS 821 Colorado river basin (Holden, 1980). One fossil species, Ptychocheilus arciferus (Cope), is from Miocene-Pliocene lake beds of the Snake River Plain in southwestern Idaho (Kimmel, 1975; Smith, 1975, 1981), and another, Ptychocheilus prelucius Uyeno and Miller, is from the Mio- cene-Pliocene Lake Bidahochi Formation in Ar- izona (Uyeno and Miller, 1965; Baskin, 1978; Smith, 1981; Minckley et al., 1986). Ptychocheilus lucius is recognized as an endan- gered species at both the federal (United States Fish and Wildlife Service, 1987) and interna- tional (International Union for Conservation of Nature [IUCN], 1986) levels. A hatchery prop- agation and reintroduction program has been developed for this species (Johnson, 1987). Several relationships among species of Ptycho- cheilus have been proposed. Ptychocheilus orego- nensis and P. umpquae are quite similar and were believed by Bond (1980), Wallace (1980), and Carney and Page (1990) to be closely related. Perceptions of the relationship of P. grandis to the other species of the genus have been prob- lematic; Moyle (1980a) considered it closely re- lated to P. oregonensis and P. umpquae, whereas Holden (1980) considered P. grandis and P. lu- cius to be closest relatives. Recently, Carney and Page (1990) presented alternative hypotheses of relationships based on meristic characters. These authors presented two perspectives on intrageneric relationships. One pattern includ- ed P. oregonensis and P. umpquae as ancestral, and P. lucius and P. grandis as derived sister species. The other pattern indicated a sister group relationship between P. oregonensis and P. umpquae, with P. lucius as the most primitive. The latter resolution was chosen as more likely, based on dates and presumed relationships of fossil Ptychocheilus. Ptychocheilus prelucius is con- sidered to be closely related to P. lucius, and from the Miocene (Smith, 1975; Carney and Page, 1990) although Smith (1981) listed P. prel- ucius as from the Pliocene. Ptychocheilus arctiferus is considered to be closely related to P. grandis and P. oregonensis, and from the Pliocene (Smith, 1975; Carney and Page, 1990) although Smith (1981) listed P. arctiferus from the Miocene as well as the Pliocene. In a study of relationships among western North American cyprinid genera, Avise and Ayala (1976) demonstrated a sister group re- lationship between Ptychocheilus (represented by P. grandis) and the monotypic Mylopharodon. The latter has a distribution sympatric with P. gran- dis throughout most of the Sacramento-San Joa- quin drainage and includes the Pit and Russian rivers (Moyle, 1976, 1980b). The degree of morphological similarity between Ptychocheilus and Mylopharodon prompted Hopkirk (1973) to suggest congeneric status for these taxa. Avise and Ayala (1976) interpreted the degree of allo- zyme similarity between P. grandis and M. cono- cephalus (Nei's genetic similarity coefficient of I = 0.88) as support for Hopkirk's generic syn- onymy. Our study employed a phylogenetic treat- ment of morphometric and meristic traits to assist identification and ascertain relationships among extant species of the genera Ptychocheilus and Mylopharodon. The latter could serve as a "taxonomic out-group" (Watrous and Wheeler, 1981) for the study of intrageneric relationships within Ptychocheilus. However, we chose to in- clude M. conocephalus in our in-group to further clarify its relationship to species of Ptychocheilus. Avise and Ayala (1976) indicated that a pair of monotypic genera, Lavinia exilicauda and Hes- peroleucus symmetricus, are likely to be the sister group to the lineage of Ptychocheilus plus Mylo- pharodon. We chose H. symmetricus as an appro- priate outgroup to provide a perspective on the relationships among the ingroup taxa. MATERIALS AND METHODS Forty-six measurements and six counts were taken on all specimens (Table 1). The morpho- metric measures consisted of a few standard fin- length measurements (sensu Hubbs and L.agler, 1958) and truss measurements between body landmarks (sensu Strauss and Bookstein, 1982). Inasmuch as they were possible, bone-to-bone measurements were taken. Prior to any use of multivariate discrimination procedure, errors were searched for using percentages of stan- dard length and prepectoral length (Table 2). We have presented the means and standard de- viations so that other researchers can study and compare the proposed phylogeny of the species in this dataset with phylogenies produced when different coding methods are used. Although it would have been possible to per- form Student's t-tests to determine significance of differences in means, species recognition and identification is an issue only with respect to the differences between P. umpquae and P. oregonen- sis. Further, such tests would only be useful when all potential groups have similar size distribu- tions. That was not the case with the individuals we had, and so multivariate methods were used.