Colorado River Ecology and Dam Management: Proceedings of a Symposium May 24-25, 1990 Santa Fe, New Mexico (1991)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

7 Native Fishes of the Grand Canyon Region: An Obituary? W. L. MINCKLEY, Arizona State University, Tempe, Arizona The highly endemic fish fauna of the Colorado River downstream from Glen Canyon Dam has been largely extirpated. Damming, diversion. and other human regulation stabilized the system, which enhanced introduced nonnative fishes and proved detrimental to native species. Historical ac- counts are used to (1) reconstruct the ecology of fishes in a pristine Colo- rado River, (2) document changes resulting in the decline of native species, and (3) indict the introduction and establishment of alien fishes as the ulti- mate factor in extirpation of the natives. Despite legislation and dedicated efforts by scientists and fisheries managers in the 1970s and early 1980s, trends toward extinction of native fishes were not reversed. Deemphasis of environmental issues because of economic and political pressures for devel- opment in the later 1980s makes survival of this unique fauna doubtful. INTRODUCTION The U.S. National Park Service was founded in 1916 with a mandate to preserve resources Placed under its care "unimpaired for the enjoyment of future generations." Shortly thereafter, in 1919, Grand Canyon National Park was established to protect the natural attributes of spectacular gorges cut by the Colorado River. These efforts failed for some of the native aquatic biota. Construction and operation of dams and reservoirs outside the park resulted in a chain reaction of environmental change that forced its 124

NATIVE FISlIES... 125 native fishes to local extinction (Dolan et al., 1974; Johnson and Carothers, 1987). This paper concerns the decline and disappearance of indigenous fishes from the lower Colorado River, including the Grand Canyon region. To document this, the fishes are first placed in the context of a pristine ecosys- tem, reconstructed logically and with a modicum of speculation from his- toric data. Demise of the fauna started just after 1900, as the river was progressively harnessed for water supply, flood control, and power produc- tion. The fish fauna collapsed from downstream to upstream, in the same sequence as the river was regulated. Reasons for this ecological catastrophe are discussed for individual species and for the fauna as a whole. Finally, the grim prognosis for native fishes is tempered by a brief review of legal statutes, scientific facts, and moral obligations, all of which exist to save them. But before this happens, it is clear that society must develop an ethic sufficient to realize this goal. THE PRISTINE HABITAT The Colorado River collects water, sediments, and dissolved solids from more than 600,000 km2 of mountains, plateaus, and basins (Graf, 1985~. From north to south, the river flows through the Rocky Mountain, Colorado plateau, and basin and range physiographic provinces. High elevations pro- vide spring and early summer runoff in the form of snowmelt, and most of the water yield is from that source. Historically, low flow predominated in summer through winter, with late summer spates in the south, where a bimodal pattern of winter rains and late summer monsoons prevails. After leaving the Rockies, the nver winds through desert, where signifi- cant amounts of water are lost to evaporation. Ions are concentrated and added by inflowing salt springs; as a result, total dissolved solids increase. Along with this material are sediments from the headwaters, joined by even more stripped from the sparsely vegetated deserts and plateaus by runoff of infrequent, often violent rains. The course of the river almost defies description. After the steep headwa- ters, it alternatively meanders through broad, aggraded valleys and then slices through bedrock uplifts. Most valley reaches are now modified for agriculture. The river is constrained and channelized from its original state of flowing through braided channels, over bars of shifting sand and gravel, and between alluvial banks. Runs and riffles were smooth and strong, and oxbow lakes, backwaters behind sandbars, and other lentic habitats were common. The canyons, if not used for a dam site or drowned by a reservoir, remain much the same as they were. They vary in morphology, with steep or sloping walls. Some are deeper than 1,500 m, others are less deep. A few are relatively straight, while most are sinuous. Rapids are common, with

126 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT high waves, whirlpools, and other turbulence creating "whitewater" (in the jargon of the river), which reflects underwater obstructions such as bedrock dikes, stony debris from side canyons, or rockfalls from the canyon walls (Leopold, 1969; Dolan et al., 1978~. Between rapids are "pools," where deep, strong currents, or flatwaters, flow through unobstructed channels. Most accounts of the river stress its variability in discharge and sediment load. In a -40-year, pre-dam period of record, discharge at Yuma, Arizona, varied over five orders of magnitude, from 0.5 to 7,100 m3 so (Dill, 1944~. Variations in transported sediment were also substantial. On an annual ba- sis, materials accumulated during low discharge, to again be moved by high water in spring. Longer-term cycles also existed. From 45.4 to 455 million metric tons years of silt was transported through Grand Canyon between 1922 and 1935 (Howard, 1947; Howard and Dolan, 1981), while records at Yuma varied from 36.3 to 326 million metric tons in the same period. Thus, a significant percentage was aggrading on the floodplain and along the channel. Bedload must have varied similarly, but measurements then, as now, were unreliable. Degradation of these deposits ocurred during episodic realignment and cutting of valley deposits by major floods, when virtual slurries of sand, gravel, and boulders must have ground inexorably through canyons. Such processes are graphically described in pre-dam accounts of building and slouching of banks and bars along the lower river (Sykes, 1937; Smith and Crampton, 1987; Beer, 1988~. Although impressive to humans, the dangers posed for desert fishes by upper limits of these variations are more imagined than real. The physical force of flooding rarely harms either individuals or populations (Meffe, 1984; Meffe and Minckley, 1987; Minckley and Meffe, 1987~. Fishes ap- parently move to clear the surface along shore even in canyons, thereby avoiding both the force of currents and molar action of transported bedload, as well as higher turbidities in deeper, turbulent channels. In wide places, they simply swim onto floodplains and then swim back as water recedes. Coarse, suspended sediments are similarly innocuous except, perhaps, through abrasion. High concentration of clays may suffocate fishes by clogging their branchial chambers (Wallen, 1951), but such has rarely been observed in western streams (Minckley, 1973~. A major seasonal problem may have been the food supply. Food may have been limiting during periods of high turbidity and bedload transport. As detailed in this chapter by Blinn and Cole, lower parts of large erosive streams are notoriously depauperate of fish foods. Algae and other primary producers are limited by turbidity and scour, and plankton is rare. Inverte- brates need primary producers, detrital materials, or other animals to eat, and most are excluded from unstable bottoms anyway. Conditions are quite different at low water. Coarse particles settle quickly, water clears to expose boulders and canyon walls to sunlight for consider-

lIATIVE FISlIES... 127 able depths, and bars are stabilized and armored with gravel as declining discharges sweep away the sand. Such places provide substrate for biologi- cal colonization. No studies are available for large unmodified rivers of the lower Colo- rado basin, but primary producers and stream invertebrates in small desert streams are remarkably resilient (Meffe and Minckley, 1987; Grimm and Fisher, 1989~. Diatoms reappear a day after flash flooding, and complex algal communities reestablish in weeks (Fisher et al., 1982; Fisher, 1986~. Insects with life cycles of days or weeks recolonize in a few days, and those with longer generation times reappear over a month or two (Bruns and Minckley, 1980; Gray, 1981; Gray and Fisher, 1981; Minckley, 1981~. Many feed on detritus, finely ground from leaves, branches, and trunks of trees being carried to the sea (Minckley and Rinne, 19851. Thus, considering the predictably long periods of low discharge in the river, algae and inverte- brates may have been in ample supply for much of the year. Added to this were terrestrial invertebrates (e.g., Tyus and Minckley, 1988) and diverse inputs from tributaries. Despite these benefits of low discharge, drought is the single most dan- gerous time for fishes. Many western fishes appear to require little more than water to survive, but they do need water. In valley reaches, infiltration into deep, coarse alluvial fill robs the channel of surface flow, and intermittence or desiccation may result. High temperatures and oxygen depletion in the remnant pools become critical and lethal. On the other hand, these massive alleviated valleys, many of which are actually structural intermontane ba- sins, also act as vast subterranean reservoirs, dammed by the bedrock through which the canyons are cut. These reservoirs were the key to fish survival. They leaked downslope into canyon-bound reaches, where pools held water in scoured holes and shaded undercuts along cliffs. Subterranean water percolating through coarse alluvium tends to be cool, mixed, and rich in nutrients (Grimm et al., 1981~; thus, canyons often provide salubrious, highly productive habitat if suffi- cient sunlight is available. Bedrock reaches are also rich in springs, adding security to the system. In the large and complex Colorado River basin, mainstream fishes had many canyon refugia and after a few weeks, at most, would have been saved from even the greatest regional drought by runoff from somewhere. Clearly it worked over the millennia, since some of the fishes have persisted since Miocene (Minckley, et al., 1986~. UNIQUENESS OF THE FISH FAUNA The special nature of the Colorado River fish fauna was recognized early. Evermann and Rutter (1895, p. 475) listed 5 families, 18 genera, and 32 indigenous species, and wrote:

128 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT Though the families and species . . . are very few, they are of unusual interest to the student of geographic distribution . . . over 78 percent of the species of fishes now known from the Colorado Basin are peculiar to it . . . a larger percentage of species peculiar to a single river basin than is found elsewhere In North America. Subsequent work has not altered these conclusions. Miller (1959) re- ported 31 freshwater species, while Minckley et al. (1986) documented 32. Recognized endemism at the species level still hovers near 75%. and Carlson and Muth (1989) compiled 44 taxa, 93% endemic when undescribed forms and subspecies were included. Clearly, as a result of long isolation and special conditions, the Colorado River basin supports the most distinctive ichthyofauna in North America. Larger fishes of the Colorado River mainstream share an intriguing, number of features (Hubbs, 1941; Miller and Webb, 1986~. Their adult body sizes are large, and some species live exceptionally long lives. (All fish measure- ments reported here are for total length, from the extreme tip of the snout to the distal end of the caudal fin.) Body shapes are more fusiform and stream- lined than most, with streamlining carried to extremes to include small depressed skulls, large predorsal humps or keels (or both), and elongated, pencil-thin caudal peduncles. Their eyes are small, the fins are expansive and often falcate (sickle shaped) with strong leading rays, and the skins are thick and leathery, especially on the head, anterior body, and leading edges of fins. Finally, the scales are tiny, deeply embedded in the skin, or some- times essentially absent. Most of these special features have been related to severity of habitat. Body and fin shape, structure, and size are construed as hydrodynamic adaptations for maintaining position and maneuvering in swift, turbulent currents. Large bodies and fins may, however, be as much a necessity to generate sufficient power to negotiate such places. Leathery skins with small, embedded, or reduced scales may minimize friction, counteract sediment abrasion, or provide a rigid external sheath to maximize muscle efficiency. Reduced eyes may be another way to avoid abrasion. Finally, long life seems adaptive to unpredictable environments, since the ability to repro- duce spans decades rather than only a few years. On the other hand, large size (correlated with long life) scarcely seems adaptive to seasonally low discharge. Despite alternative possibilities, common trends in morphology across a number of unrelated taxa speak for commonality in other than phylogeny, and the river's harsh selective arena seems a reasonable choice. HISTORIC REVIEW Early Records and Surveys Faunal remains in archaeological sites show that Colorado River fishes were caught and eaten by Indians (Miller, 1955; Euler, 1978; Miller and

NATIVE FISHES... 129 Smith, 1984~. Early canyon explorers ate them too, and valuable records appear in the accounts of John Wesley Powell (1875), Stegner, (1954), Robert Brewster Stanton's railway surveys (Smith and Crampton, 1987), the adventures of Ellsworth and Emery Kolb (Kolb and Kolb, 1914; Kolb, 1989), and others. The earliest scientific collectors sampled tributaries (Baird and Girard, 1853a-c; Girard, 1857, 1859; Cope and Yarrow, 1875; Kirsch, 1889), the Colorado mainstream near Yuma (Abbott, 1860; Evermann and Rutter, 1895; Gilbert and Scofield, 1898; Chamberlain, 1904; Snyder, 1915), and at up- stream crossings on the Grand (= Colorado3 and Green Jordan, 1891; Evermann and Rutter, 1895) rivers. Canyons remained inhospitable to humans until rubberized boats made whitewater rafting safe and reliable. Even with mod- ern equipment, sampling is difficult at best, and the faunas of canyon-bound reaches of the river remain the least understood of all. Most species had nonetheless been collected and described before 1900 (Minckley and Dou- glas, 1991~. The early surge of inquiry declined over almost four decades of complacency that followed Jordan and Evermann's (1896-1900) monograph "The Fishes of North and Middle America." EVIDENCES OF FAUNAL CHANGE Carl L. Hubbs and Robert R. Miller reinitiated studies of Colorado River fishes in the late 1930s. They began by producing revisionary works and compilations of records (Hubbs and Miller, 1941; Miller, 1943), descrip- tions of new taxa and life stages (Miller, 1946a; Winn and Miller, 1954; Miller and Hubbs, 1960), and records of hybrids (Hubbs et al., 1942; Hubbs and Miller, 1953), and they moved with time to authoritative biogeographic accounts (Hubbs and Miller, 1948; Miller, 1959~. Scattered throughout these works were notes on faunal declines, and Miller (1946b) published an early plea for study of native fishes before further alterations of the large western rivers were undertaken. Dill's (1944) survey of the lower Colorado River was the first to provide insight on both native and introduced fishes downstream from the new Boulder (= Hoover) and Parker dams. He noted reductions in native species attrib- uted to environmental changes associated with damming. Wallis (1951) considered the fauna "in urgent need for further research, because so many forces are fast exterminating it." Jonez et al. (1951) and Jonez and Sumner (1954) also expressed concern for native fishes in the face of rapid environmental change. Alien fishes attracted early attention (Dill, 1944; Beland, 1953b; Hubbs, 1954~. Reservoirs changed the river in ways that enhanced lentic-adapted, nonnative species aonez et al., 1951; Beland, 1953a; Kimsey, 1958; Nicola, 1979), and reservoir spor~sheries became important regional resources (Moffett, 1942, 1943; Wallis, 1951; Jonez and Sumner, 1954~. A remarkable array of both native and nonnative species were used as bait (Miller, 1952), and bait

130 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT and forage fishes escaped or were intentionally stocked to join and feed expanding game fish populations (Hubbs, 1954; Kimsey et al., 1957; LaRivers, 1962; U.S. Fish and Wildlife Service, 1980, 1981~. By the 1960s, although essentially invisible to the untrained eye, indig- enous fishes of the lower Colorado River had been largely replaced by exotic species. Miller (1961) wrote of dramatic changes in abundance and distribu- tion and of local extirpation of native forms. At this same time, Glen Canyon, Flaming Gorge, Navajo, and other major dams authorized by the Colorado River Storage Project Act of 1956 were also nearing completion. These were far from invisible. The magnitude of change was finally realized, and alarms began to sound in an emerging conservation community. The Colorado River had indeed been tamed, its wildness was lost except in the most isolated and inaccessible reaches, and the public was stirred to react. Resistance against other dams was led by the Sierra Club, and conservationists voiced their opposition to further modifications (Fradkin, 1981~. In addition to this controversy, almost 700 km of the Green River system above the new Flaming Gorge Dam was poisoned in 1962 to clear the way for an introduced trout fishery (Holder, 1991~. Some of the targets were native fishes. The operation went awry, and fishes were killed far down- stream through Dinosaur National Monument (Miller, 1963a, 1964~. This event, possibly more than anything else, solidified the resolve of those protective of native fishes. In addition to making available funds for reservoir planning and con- struction, the storage act provided for evaluation of-the archaeological sig- nificance of areas to be inundated. Preimpoundment surveys conducted along with archaeological salvage operations included biological collections for use in interpreting paleo-Indian ecology. Studies were made in the Flaming Gorge Reservoir basin (Dibble and Stout, 1960; Woodbury, 1963), Navajo basin (Pendergast and Stout, 1961), and the future basin of Lake Powell (Woodbury, 1959; McDonald and Dotson, 1960~. Marble and Grand can- yons were ignored since they were not to be directly affected. By the time Miller (1968) and Suttkus and colleagues (Suttkus et al., 1976; Suttkus and Clemmer, 1979) began sampling in 1968 and 1970, respectively, the down- stream impacts of Glen Canyon Dam were already evident. In 1963, partially in response to public outcry over the poisoning, agen- cies such as the National Park Service, Fish and Wildlife Service, and state conservation departments began studies of Green River fishes (Vanicek, 1967; Vanicek and Kramer, 1969; Vanicek et al., 1970~. Such research expanded elsewhere as new legislation came to bear in the late 1960s (Holder, 1973; Holden and Stalnaker, 1975~. Disappearing species became rallying points for the Endangered Species Protection Act of 1966, which evolved to the Endangered Species Act of 1973. Recognition that habitat was being lost at an unacceptable rate stimulated the National Environmental Protec-

NATIVE FISHES... 131 tion Act of 1969, which mandated assessment and disclosure of impacts of federal projects. After 1966, lists of species in jeopardy served to focus research and management efforts and agencies such as the Bureau of Recla- mation became involved to satisfy legal requirements for project construc- tion and operations. Ongoing research has since documented the status and ecology of fishes targeted by official listings. Some resulted in major reports (Minckley, 1979; Carothers and Minckley, 1981; Miller et al., 1982b-c; Maddux et al., 1987; Tyus et al., 1987; Ohmart et al., 1988), only excerpts of which have appeared in the open literature. Smaller, more specific projects dealt with species' biol- ogy or habitat; many of these are cited in recovery plans (U.S. Fish and Wildlife Service, 1989b-d) or status reports for rare species (Seethaler, 1978; McAda and Wydoski, 1980; Minckley, 1983; Kaeding and Osmundson, 1988; Minckley et al., 1991; Tyus, 1991~. Reviews varied from annotated bibliogra- phies (Wydoski et al., 1980), through edited symposia (Spofford et al., 1980; Miller et al., 1982a; Adams and Lamarra, 1983; Minckley and Deacon, 1991), to contributions dealing with the fauna (Deacon, 1968, 1979; Joseph et al., 1977; Holden, 1979; Behnke and Benson, 1983; Stanford and Ward, 1986c) and its ecosystem (Deacon and Minckley, 1974; Ono et al., 1983; Williams et al., 1985; Stanford and Ward, 1986a,b; Carlson and Muth, 1989~. In short, a wealth of information now exists on Colorado River fishes. NATIVE FISHES Six of eight fishes native to Grand Canyon National Park (Table 7-1) are endemic. Speckled dace and roundtail chub are known from adjacent rivers, and the latter, known only from a few specimens (C. O. Minckley, 1980), is excluded from further consideration. Four of the remainder, humpback chub, bonytail, Colorado squawfish, and razorback sucker, are listed or proposed as endangered by the Department of the Interior (U.S. Fish and Wildlife Service 1989a, 1990~. Of these, only the humpback chub (Figure 7-1) per- sists as a reproducing population. The other three are extirpated or exceed- ingly rare. Speckled dace, flannelmouth sucker, and bluehead sucker remain relatively common (Minckley, 1985~. Most early accounts of fishes in the Colorado River were for large spe- cies sought for food, e.g., "Colorado River salmon" (squawfish) was on the menu of Christmas dinner for the Stanton Party in 1889 (Measeles, 1981; Smith and Crampton, 1987~. Most records were more informal. When Kolb and Kolb (1914, p. 123) heard splashing near their camp at the Little Colo- rado River in 1911. they wrote: . . . the fins and tails of numerous fish could be seen above the water. The striking of their tails had caused the noise we had heard. The 'bony

132 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT TABLE 7-1 Common and scieniific names of naiive and introduced fishes recorded from Grand Canyon National Park, Arizona. Those taxa marked with an asterisk (*) are listed or proposed for listing as endangered by the U.S. Deparunent of the Interior. CLUPEIDAE, shads introduced Threadfin shad SALMONIDAE, salmons and trout all introduced Apaehe trout Cutthroat trout Silver Salmon Rainbow trout Brown trout Brook trout CYPRINIDAE, minnows Native species *Humpback ehub *Bonytail Roundtail ehub *Colorado squawfish Speekled daee Introdueed species Common earp Red shiner Golden shiner Fathead minnow Redside shiner CATOSTOMIDAE, suckers all native Flannelrnouth sucker Bluchead sucker *Razorback sucker ICTALURIDAE, bullhead catfishes all introduced Blaek bullhead Channel catfish FUNDULIDAE, killifishes introduced Plains killifish POECILIIDAE, livebearers- introdueed Mosquitofish CENTRARCHIDAE, sunfishes—all introduced Green sunfish Bluegill Largemouth bass PERCICHTHYIDAE, temperate basses introduced Striped bass Dorosoma petenense GHnther Oncorhynchus apache Miller O. clarki Riehardson O. kisutck Walbaum O. mykiss Walbaum Salmo trutta Linnaeus Salvelinus fontinalis Mitehill Gila cypha Miller G. elegans Baird and Girard G.r. robusta Baird and Girard Ptychocheilus lucius Girard Rhinichthys osculus Girard Cyprinus carpio Linnaeus Cyprinella lutrensis Baird and Girard Notemigonus crysoleucus Mitchill Pimephales promelas Rafinesque Richardsonius balteatus Richardson Catostomus latipinnis Baird and Girard Pantosteus discobolus Cope Xyrauchen texanus Abbott Ameiurus melas Rafinesque fctalurus punctatus Rafinesque Fundulus zebrinus Jordan and Gilbert Gambusia affinis Baird and Girard Lepomis cyanellus Rafinesque L. machrochirus Rafinesque Micropterus salmoides Lacepede Morone saxatilis Walbaum

NATIVE FISHES... a: 133 FIGURE 7-1 Male humpback chub, cat 38 cm long, captured from the Little Colorado River, Arizona; photograph by J. N. Rinne at Willow Beach National Fish Hatchery. tail' were spawning .... The Colorado is full of them; so are many other muddy streams of the Southwest. They seldom exceed 16 inches in length, and are silvery white in color. With a small flat head some- what like a pike, the body swells behind it to a large hump. Suttkus and Clemmer (1977) believed the fish to be humpback chubs rather than bonytail, which seems likely. True bonytail (Figure 7-2) were better described by Dellenbaugh (1984, p. 15) from the Green River: . . . a fish about ten to sixteen inches long, slim with fine scales and large fins. Their heads came down with a sudden curve to the mouth, and their bodies tapered off to a very small circumference just before the tail spread out. Another quotation from Dellenbaugh (1984, p. 98) introduces the Colo- rado squawfish (Figure 7-3 and 7-4~. The incident occulted near the present town of Green River, Utah, in 1871; Minckley (1973, p. 124) erred in attributing it to Grand Canyon in 1872: He [a member of the second Powell Expedition] thought his precious hook was caught on a snag. Pulling gently in order not to break his line the snag lifted with it and presently he was astounded to see, not the

134 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT FIGURE 7-2 Female bonytail, cat 50 cm long, trammel-netted from Lake Mohave, Arizona-Nevada; photograph by W. L. Minckley. branch of a tree or a waterlogged stick, but the head of an enormous fish appear above the surface .... Casting again another of the same kind came forth and then a third. The longest . . . was at least thirty or thirty-six inches with a circumference of fifteen inches. The others were considerably shorter but nevertheless very large fish. Colorado squawfish were familiar to all of the early canyon explorers. Stanton recorded "large" or"huge" fish caught in the Green and Colorado rivers (Smith and Crampton, 1987~. Dellenbaugh (1984, facing p. 102) pub- lished a photograph of two large ones taken by the Stanton party in 1889. In southern Wyoming in 1911, Kolb (1989, p. 15) . . . caught sight of fish gathered in a deep pool, under the foliage of a cottonwood tree which had fallen into the river. Our most tempting bait failed to interest them; so Emery, ever clever with hook and line, 'snagged' a catfish . . . for salmon bait, a fourteen-pound specimen rewarding our attempt .... They sometimes weigh twenty-five or thirty pounds, and are common at twenty pounds; being stockily built fish, with large, flat heads. I found no references to other species in this kind of older literature,

NATIVE FISHES... 135 FIGURE 7-3 Hatchery produced Colorado squawfish, cat 50 cm long (1974 year class), on display at the Arizona-Sonora Desert Museum, Arizona; photograph by J. N. Rinne. although hungry explorers undoubtedly ate any large fish. But suckers rarely take a hook, and speckled dace (Figure 7-5) are too small to be of much ~ . . culinary Interest. A few scientific collections were made specifically in Glen and Marble/ Grand canyons prior to the 1960s. The survey of Glen Canyon in 1958 included 27 sites, mostly in tributaries (Smith, 1959; Smith et al., 1959; McDonald and Dotson, 1960~. Seventeen species were taken, of which roundtail chub, Colorado squawfish, speckled dace, flannelmouth, bluehead, and ra- zorback suckers were native. Ten nonnatives were dominated by fathead minnow, carp, channel catfish, and green sunfish. One specimen caught on hook and line near Bright Angel Creek in Grand Canyon was used with two others from unknown sites to describe the hump- back chub (Miller, 1946a). Other species known from Marble/Grand can- yons at that time (Miller, 1946a), razorback and bluehead suckers, bonytail, and channel catfish, had also been caught by anglers. Anne and I. A. Rodeheffer seined speckled dace, squawfish, flannelmouth and bluchead suckers, and nonnative channel catfish in the Colorado and Paria rivers near Lees Ferry in 1934; C. H. Lowe caught dace from Big Nankoweap Creek in 1953; and O. L. Wallis secured a humpback chub near Spencer Creek in

136 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT :~ FIGURE 7-4 Hatchery produced Colorado squawfish, 42 cm long (1974 year class); photograph by J. E. Johnson. 1955 (University of Michigan Museum of Zoology catalog of specimens). Shortly after closure of Glen Canyon Dam, gill-net samples in and below the filling reservoir caught only large species, humpback and roundtail chubs, Ponytail, and apparent hybrid chubs (Holder, 1968; Stone, 1965-1969, 1971, 1972; Stone and Rathbun, 1967-1969; Holden and Stalnaker, 1970~. Bluehead and flannelmouth suckers were also there, as were introduced carp and channel catfish (unpublished data). THE FISH COMMUNITY As already noted, much is known of the ecology and distribution of Colorado River fishes. Unfortunately, however, many data are in state and federal agency reports, difficult to find and obtain, and mostly dealing with limited geographic areas. A review of the fish fauna of Marble/Grand can- yons was thus difficult to prepare without an overburdening volume of citations. I attempted to avoid this problem by presenting species accounts in the Appendix, which may be consulted for details. Under pristine conditions, most fishes of the lower Colorado River main- stream probably lived in valley reaches. Canyons would have been most

NATIVE FISHES... 137 FIGURE 7-5 Speckled dace, 62 mm long, from the Virgin River, Arizona; photo- graph by J. N. Rinne. densely populated in wide places near inflowing streams and in tributary mouths. During low flow, most species (or at least larger individuals) moved into canyons to avoid declining water levels or high summer water tempera- tures (Siebert, 1980~. Adults of each species lived in proximity to their preferred foods and among physical habitat features commensurate with their sizes and mor- phologies. Large squawfish occupy quiet places along shore, often near overhead cover such as logs or other debris or in "pockets" between boul- ders. Flannelmouth suckers (Figure 7-6) stay on the bottom in deep, quiet or slow-moving water, and they feed in eddies and along shorelines of pools and riffles. Large blueheads (Figure 7-7) also occupy deeper places except FIGURE 7-6 Flannelrnouth sucker, cat 40 cm long, from the Virgin River, Ari- zona; photograph by J. N. Rinne.

COLORADO RIVER ECOLOGY AND DAM MANAGEMENT ~ ?: A. i. i..~: FIGURE 7-7 Bluehead suckers, 38 to 45 cm long, from the Green River, Utah; photography by W. L. Minckley. These specimens, taken at the same locality, demonstrate the remarkable variation in caudal peduncle length and diameter seen in mainstream populations. when feeding, moving to hard bottoms in shallows to graze. Razorback suckers (Figures 7-8 and 7-9) occupied near-bottom space, most likely in open, flowing channels but also in backwaters and eddies when available. I suspect adult Ponytails to have stayed mostly in the water column in the channel with razorbacks. Deep pools and eddies of whitewater canyons support humpback chubs, alone except for transients of other species. Roundtail chubs are also common in such places in the upper basin: how the last two chubs partition habitat space is unknown. Speckled dace stay near the bot- tom in water a few centimeters deep along shorelines, on riffles, and in tributary mouths. Apparent trophic relations also seem fairly straightforward (Minckley, 1979), characterized by qualitative (kind) and quantitative (size) differences in foods and spatial segregation in feeding. Adult squawfish eat other fishes. Each of the three suckers is adapted differently: flannelmouth for feeding on small aquatic insects and other benthic animals; blueheads for scraping algae; and razorbacks with protrusible mouths and special gill rakers for sieving plankton or detritus. Studies of food habits tend to bear out the interpretations from morphology for all of these species. Humpback chubs

NATIVE FISHES... 139 FIGURE 7-8 Razorback sucker, cat 50 cm long, on display at the Arizona-Sonora Desert Museum, Arizona; photograph by J. N. Rinne. and bonytail both tend to be insectivores, eating both large and small terres- trial and aquatic invertebrates. The former must feed both in the water column and on bottom in deep eddies and zones of turbulence, and the latter feed in midwater and near surface in smooth-flowing channels. Speckled dace also are insectivores or facultative omnivores, delving over and into bottom interstices in search of invertebrates and other organic materials. By and large, reproductive sites seem more similar among species than are habitats and food habits. Perhaps the requirements for egg and larval development necessitate relatively swift current and clean, unconsolidated gravel substrates. The three suckers and speckled dace reproduce in spring on gravel or cobble riffles. Razorbacks spawn earlier than the others, and dace use shallower places and tributary mouths. Squawfish also spawn on gravel or cobble bottoms in current, in late spring or summer. We know nothing of bonytail spawning other than in artificial lakes or ponds, and humpbacks are even more of a mystery. Both breed in June-early July in the upper basin and in March-May downstream. Only the squawfish has evolved a life history that includes long spawn- ing migrations to precise sites that are presumably identified by olfaction (Tyus, 1985, 19861. Other species may select suitable spawning sites by cues such as substrate, depth, current, or proximity. In the upper basin,

140 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT :: ~ :: 'if _ £.? 0 - , 7. FIGURE 7-9 Male razorback sucker, cat 50 cm long, ~arnmel-netted from Lake Mohave, Ariona-Nevada; photograph by W. L. Minckley. radiotagged humpbacks rarely move more than 1 km from the point of tagging, and they presumably spawn there. However, tagged humpback chubs move upstream and downstream for more than 10.0 km in the Little Colo- rado River, and back and forth from there into the mainstream Colorado. Some humpbacks tagged in the Little Colorado returned as long as ~ years later (C. O. Minckley, 1989; Kubly, 1990~. The interrelations of movements and reproduction in Grand/Marble Canyon humpbacks are not yet under- stood. Speckled dace enter tributaries to spawn, both in Marble/Grand can- yons and elsewhere. Razorback suckers do not appear to make directed movements to specific sites for reproduction (Minckley et al., 1991), yet only a few spawning sites are known in the Green River (Tyus, 1987~. Razorbacks in Lake Mohave, Arizona-Nevada, spawn at various places in different years and perhaps in the same year (W. L. Minckley and P. C. Marsh, unpublished data). Different life stages of fishes have different ecologies, and perhaps espe- cially so in the Colorado River, where only a few species are present in a large and diverse environment. The pattern of downstream movement of young squawfish to nursery areas, growth, and then redispersal upstream as

NATIVE FISHES... 141 young adults exemplifies the use of almost all available habitat by a single species. Young of other species, including humpback chubs from canyons and dace from tributary mouths, also seem to travel downflow along shore- lines and into embayments and backwaters at first and then move with increased age and size to occupy their adult niches in the channel. REASONS FOR DISAPPEARANCE OF THE FAUNA Nonbiological Changes Human modification of the Colorado River basin is profound. The stream sustains more consumptive water use than any other river in the United States (Pillsbury, 1981), and less than 1% its of virgin flow now reaches the mouth (Petts, 1984~. At least 117 impoundments with individual capacities greater than 1.0 x 106 m3 have been built in the upper basin alone, and more than 40 diversions export water to adjacent basins (Bishop and Porcella, 1980~. The lower part of the basin is even more dramatically changed. Examination of the impacts of this development from other than the bio- logical perspective by Fradkin (1981), Worster (1985), Reisner (1986), and Weatherford and Brown (1986) are required reading for anyone trying to understand the magnitude of alteration to the river and its tributaries. In spite of this high level of exploitation. I am hard pressed to attribute the general disappearance of Colorado River fishes to physical changes in habitat. The basin remains large and diverse. Almost all of the native spe- cies were wide-ranging, including those now classed as endangered. Squaw- fish, Ponytails, and razorbacks all historically occupied ~10° of latitude, from southern Wyoming to Mexico, and humpback chubs lived from to Wyoming to below Lake Mead. The humpback chub may be specific in its habitat needs, but the Little Colorado and Yampa rivers are certainly very different in size and nature from the Green and Colorado mainstems, and the species persists in all four. Upstream from dams, discharge, sediment load, and other factors still vary over many orders of magnitude. I am also impressed by the persistence of this native fauna over geologic time (Minckley et al., 1986) and agree with M. L. Smith (1981) in considering all of the species as generalists. I seriously doubt that physicochemical changes wrought by humans in less than 100 years equal those occurring since Miocene, which began 20+ million years ago. The fact that these fishes adapt to, reproduce in, and mature in a spec- trum of artificial habitats also supports a hypothesis of broad ecological tolerances. They have proven relatively simple to maintain and rear in cap- tivity (Rinne et al., 1986; Johnson and Jensen, 19911. Although changes in a system may alter or invalidate cues required for reproduction or survival, these fishes succeed in places where habitat-mediated cues reinforced over

142 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT evolutionary time (other than seasonal temperatures, day lengths, etc.) can- not exist. Only two direct effects of dams blockage of spawning migrations and depression of summer water temperatures seem sure to extirpate most na- tive species, and I am not fully convinced these can be singled out. For example, squawfish move both upstream and downstream to reach their spawning sites (Tyus, 1986; Tyus and Karp, 1989~. Assuming that the site itself was not destroyed, a dam would only exclude the downstream part of a stock, and fish from upstream should persist. I argue later that loss of nursery areas was more important in the decline of this species than was interdiction of spawning migrations. Water temperature too low for reproduction or larval development clearly results in loss of populations and is the culprit excluding natives from Marble/ Grand canyons, yet less than half of the available mainstream is rendered too cold by hypolimnetic waters or physically drowned by reservoirs. Why then has the entire fish fauna collapsed? I forward an hypothesis, perhaps untestable in nature to answer this question: the introduction and enhance- ment of nonnative fishes as a result of river alterations forced the native species to extinction. To add salt to an open wound, this is a problem for which we were poorly prepared and one that we have not yet resolved to address. Biological Changes Remarkable numbers of nonnative fishes have been intentionally and inadvertently stocked into the Colorado River. Introductions and impacts of exotic fishes in North America were reviewed in Courtenay and Stauffer (1984), and many other works list and discuss such species in the Colorado River basin (e.g., LaRivers, 1962; Sigler and Miller, 1963; Minckley, 1973, 1982; Miller et al., 1982a; Moyle, 1978~. Early stockings were of preferred food fishes of the time, including carp, buffalofish, shads, various salmo- nids, and even eels. At least 20 species of nonnative fishes were planted in Utah prior to 1900 (Sigler and Sigler, 1987~; catfishes and carp were well established by then in the lower Colorado (Chamberlain, 1904~. As time went on, stocking shifted toward predatory species for reservoir sportf~sheries, additional bait species escaped, and forage fishes were introduced. Tropical species appeared mostly after World War II, planted in attempts at vegeta- tion control or by mistake as escapees from aquaria. Many aliens disappeared, but others established. The original fauna of 30 or so native species in the Colorado basin has increased to 80 or more, including salmonids from the Pacific Coast, Great Lakes, and Europe; min- nows from as far away as Europe and Asia and as near as the adjacent Great Basin; minnows, catfishes, sunfishes, and basses from the Mississippi-Rio

NATIVE FISHES... 143 Grande system; livebearers from Mexico; cichlids from Africa and Central America; anadromous sea bass from the Atlantic (in part via California); and others. Nonnative fishes have invaded essentially every habitat and continue to do so as species disperse from place to place and others appear. The Colo- rado River has become a crowded world. Larvae of squawfish and other natives share backwaters behind sandbars with many nonnative young (Kale and Tyus, 1990~. Species piscivorous when young (especially green sun- fish) or remaining small as adults (e.g., the voracious mosquitofish) are now there to meet and eat them. Juvenile squawfish may use abundant larvae, young, and adults of nonnative minnows as food, and adults seem to have little competitive difficulty, as yet, with comparable piscivores like north- ern pike and walleye (Tyus, 1991), but negative interactions seem inevi- table. Today, if larval native fishes move downstream to an ancestral nursery they find themselves instead in a reservoir. Predatory sunfish live along shorelines, and schools of planktivores such as threadfin shad serve as com- petitors (and perhaps predators) in open water. In addition, midwaters and shorelines in valley reaches are swarming with one or another kind of alien minnow. Sand, redside, and red shiners and fathead minnows are common, the first two upstream from Lake Powell and the last two throughout the river. Channel catfish are more prevalent in eddies of whitewater canyons than humpback chubs. Further examples seem unnecessary. The abundant presence of introduced fishes can only reduce space, food, and survivorship of the natives. We are unsure of the specific causes of native fish declines, for even the most apparent interactions. This is an area needing serious research. Spikedace and woundfin, both federally listed minnows of the lower Colorado basin, consistently disappear as red shiners invade and become abundant (Minckley and Deacon, 1968; Deacon, 1988; Marsh et al., 1989~. Attempts to quantify interactions for food and space have failed, except in pointing out some of the subtle ways the nonnatives may replace (or displace) the native forms (Abarca, 1989; Marsh et al., 1989~. Mosquitofish eat the young of native, endangered topminnows and remove the fins from adults so they succumb to secondary infections (Meffe, 1983, 1985~. Flathead catfish are invading the Salt River Canyon in Arizona, followed closely by disappearance of the native fauna (D. A. Hendrickson, personal communication). Indigenous fishes may simply be naive to alien predators (Meffe, 1983, 1985; Minckley, 1983), but we are not even sure of this. Few things seem to help native fishes survive the onslaught of alien forms, aside from strong evidence that flooding in canyons displaces nonnative fishes while native species are unaffected. In fact, native fishes are often enhanced by flood removal of predators and competitors (Minckley and

144 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT Meffe, 1987~. The effect is temporary, since the canyons are soon reinvaded by alien fishes from populations protected in reservoirs and ponds upstream. Part of the present ecology of the Colorado River includes fisheries for nonnative bouts. Fortunately, there seem to be few conflicts between bouts and indigenous, warmwater fishes that result in demise of the latter, mostly because of their separate temperature requirements. Few realize that trout were introduced into cool tributaries of the Marble/Grand canyons system in the 1920s, long before any dams (McKee, 1930; Brooks, 1931; Williamson and Tyler, 1932; U.S. Fish and Wildlife Service, 1980, 1981~. From there they spread to other tributaries through the Colorado River in winter; sum- mer temperatures in the river were too warm for them to survive. Today, trout permanently inhabit the mainstream because of cold, epilimnetic water from Lake Powell, the same reason the native, warmwater fishes disap- peared. Piscivorous brown trout may remove high-altitude minnows such as spikedace through predation, and nonnative bouts compete with, prey upon, and hybridize with native bouts, but these subjects do not pertain to the Colorado River mainstream. Tailwater trout fisheries in the desert southwest started with rainbows stocked immediately after Boulder (= Hoover) Dam was closed in 1935 (Moffett, 1942~. The fish grew rapidly to large size, and Eicher (1947) estimated that ~9,000 kg was harvested in 1946. Fish averaged ~40 cm long, varying upward to 60 cm. Problems were the same as perceived today for the fishery below Glen Canyon Dam an apparent scarcity of food (Moffett expressed amazement at the amount of filamentous algae consumed), ques- tions of instream reproduction versus stocking in maintaining the popula- tion, and possible impacts of river fluctuations. Hoover Dam has multiple penstocks, and warm downstream temperature was also a potential problem when higher-elevation water passed through the turbines. When Lake Mohave was closed in 1954, the riverine trout fishery below Lake Mead was reduced in size (Walks, 1951; Jonez and Sumner, 1954~. However, a"layered" fishery soon developed, with fast-growing rainbow trout living below the thermocline in summer (throughout the system in winter) and warmwater species in the epilimnion. Trout were planted later in Lake Mead to create another, highly successful layered fishery (Allen and Roden, 1978) that was destroyed by the proliferation of predatory striped bass stocked to form another new sportfishery in the late 1960s (Baker and Paulson, 1983~. No time was wasted in stocking trout in tailwaters of Flaming Gorge and Glen Canyon reservoirs. Flaming Gorge was further retrofitted with vari- able intake levels in the early 1980s, so that warm, near-surface water could be released in summer to enhance downstream trout (Johnson et al., 19879. The hypolimnion of Flaming Gorge Reservoir is 4.0 to 6.0°C year-around, too cold for even salmonids to achieve their growth potentials. Both the

NATIVE FISHES... 145 National Park Service and Arizona Game and Fish Department stocked trout as soon as Glen Canyon Dam was closed in 1963. Survival and growth were high, and the fishery expanded to `'world-class" status. It has declined a bit in quality but remains regionally substantial and important panisch, 1985; Taubert, 1985~. Because of persistence of cold water through most of Marble/ Grand canyons, trout remain common through more than 300 km of river (Maddux et al., 1987), although few are harvested downstream from Lee's Ferry because of limited access. Some of the economic (Richards and Wood, 1985), sociological (Caylor, 1986), and biological aspects (Persons et al., 1985; Montgomery et al., 1986; Maddux et al., 1987; Kondolf et al., 1989) of this coldwater fishery sandwiched between warmwater habitats have been studied. Unfortunately, most information remains in unpublished agency reports, yet to be reviewed and summarized. DISCUSSION Management Considerations Native fishes of the American West will not remain on earth without active management, and I argue forcefully that control of nonnative, warmwater species is the single most important requirement for achieving that goal. Despite appearances, no part of the river is natural so long as alien fishes predominate. The wishes of agencies and individuals to maintain native fishes in "natural" habitats cannot be realized, since none is natural so long as nonnative fishes are present. Removal of nonnative fishes from entire watersheds may be the only way to conserve some native populations, but there is no known way to remove established alien species from so large and complex a system as the Colorado River. On the other hand, as judged from research over the past 25 years (Minckley and Deacon, 1991), many habitats will continue to support natives if nonnative species are suppressed; management efforts must be tailored to minimize their impacts. This argument is not to belittle needs to preserve those parts of the system which retain their natural physicochemical and biological attributes. Upstream, instream, and downstream modifications that influence such reaches must be carefully screened for operational ways to maintain and increase naturalness. We must seize every opportunity, for example, to perpetuate and enhance flow through water conservation and careful scrutiny for con- firmation of need. When water is to be passed through a reach for down- stream use, delivery should be ire such a way as to mimic natural patterns and thus enhance habitats and native fishes. In modified reaches, new (and old) techniques of habitat reconstruction can be applied, varying from re- moval of regulation structures, a preferred alternative of many workers, to allowing them to fall into disrepair (Swales, 1989~. Building additional

146 COLORADO RIVER ECOLOGY AND DAhl MANAGEMENT structures that produce greater heterogeneity, a preferred alternative of de- velopment agencies, may or may not be advisable. Nonetheless, some de- velopments for aquaculture (Welcomme, 1989), e.g., excavation of flood- plain lakes and marshes, are applicable to recovery of Colorado River fishes. National wildlife refuges and other federal and state reserves already dedi- cated to waterfowl or other biological resources are already available (Minckley, 1985; Minckley et al., 1991), if and when approval of their use for native fishes is realized. In some cases, modifications of existing structures may not be advisable. For example, retrofitting Glen Canyon Dam with variable intake levels to allow manipulation of downstream water temperatures (U.S. National Park Service, 1977) might well open a Pandora's box. Warming the water in Marble/Grand canyons too much would result in replacement of trout by warmwater species, violating policies of agencies charged with managing the sportfishery (Arizona Game and Fish Department, 1978) and scarcely being appreciated by sportsmen. One of the last things needed is a pitched battle between sportsmen protecting trout and conservationists bent on re- covery of an endangered fauna. Warming would nonetheless allow rein~oduchon of squawfish and bonytail, with a chance for reestablishment, and might stimulate recruitment by the remnant populations of humpback chub and razorback suckers. Other native fishes could be benefited as well. However, since the cold water of today is as large a deterrent for nonnative warmwater species as for natives, the former would also be enhanced (U.S. Fish and Wildlife Service, 1978~. Miller (1968) caught more kinds and far more individuals of nonnative fishes in 1968, before extremely high, cold discharge, than were taken in later years. They had already invaded Grand/ Marble canyons from both upstream and downstream and were forced either out of the system or to lower population levels as mean temperatures dropped. New predators are also present since the 1960s striped bass both upstream and downstream and flathead catfish downstream of Lake Mohave, but moving inexorably upstream through unauthorized and illegal stocking. Warm water would benefit these fishes, which would in turn deter establishment of rein- troduced stocks of natives and contribute to extirpation of the natives that remain. Clearly, maintaining native fishes in the region is complex, and the effort requires careful foresight and planning. Politics of Conservation Advocates of natural habitats and native fishes must deal with conserva- tion at a number of levels, many of which are conflicting or otherwise difficult to resolve. First, aquatic habitats and fishes are not treated the same as other managed habitats and species. Agencies do not hesitate to exclude the public from an area near a bald eagle nest, remove feral burros

NATIVE FISlIES... 147 because they compete with native bighorn sheep, or kill coyotes to lessen predation on American antelope fawns. The public, with some notable ex- ceptions, goes along with these decisions and actions, partially because people sympathize with the species or life stage being protected. People do not identify with fish species or life stages, and they must be educated to do so. Thus, with few exceptions, agencies in the Southwest do not identify needs to protect spawning habitat or buffer spawning fishes from human harassment, nor do agencies regulate or prevent stocking of game or forage fishes that compete with or prey on natives or manipulate the take of large predators (by liberalizing size and creel limits, for example) to enhance a native form. Most fishes remain as commodities managed for catching and eating, especially if they are accessible, pretty, large, and tasty. Native species remain unknown despite legislation demanding their advertisement, management, and conservation. Recognition that resource development is often incompatible with native animals and plants resulted in legislation protecting them, thereby regulat- ing development. Thus, where researchers in the past sought answers to questions in pursuit of independent knowledge, today's workers are driven by mandate. Generalizations rarely come from such studies because of their short-term, problem-solving orientation. Research on perceived problems, directed and funded by agencies afflicted by the same problems, is always suspect and short term; answers may be obtained that further goals of projects that caused the problem, and frequently by personnel of the project itself. Only the most objective, impersonal, and highly trained investigators can do a creditable job under such circumstances. Recommendations of the Na- tional Research Council (1987) for participation by external, senior scien- tists in major environmental studies reflect a general concern for this situa- tion. A third problem centers on the information from agency-mediated stud- ies becoming short-circuited into a voluminous and difficult-to-access "gray" literature. Changes in career emphasis have created a cadre of nonacademic and nonindustrial professionals who communicate their research findings in unpublished reports. Such reports are preliminary in nature, or prepared under deadlines, and so information may be presented partially or prema- turely. Yet agencies demand such documents to account for expenditures, guide policies, and justify their actions. Most of these documents never reach the open, published literature. And only in the open literature are methods, data, and ideas subject to the benefits of peer review and criti- cism, interpretations sharpened, and gaps in information defined. Symposia like those of Miller et al. (1982a) on native fishes of the upper Colorado River basin and Minckley and Deacon (1991), reviewing management ef- forts for the western fish fauna, help in this regard. Fourth, I reemphasize that far more attention must be paid the role of

148 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT nonnative organisms in the displacement and replacement of native species. The Colorado River downstream from Lake Havasu has the dubious distinc- tion of being the world's only major stream populated by an entirely alien fish fauna, and we have only anecdotal knowledge of how its natives were destroyed! Sufficient data exist to indict nonnative fishes as a major part of the problem, yet stocking and spread of additional exotic species occurs and is proposed each year. As a case in point, the striped bass, stocked as a game fish in the late 1960s in Lake Mead, proved too voracious a piscivore even for other nonnative species. At first, they appeared a tremendous success. Anglers, stimulated by the presence of "trophy" fish that are readily caught, swarmed to catch them. Entrepreneurs developed facilities and products to catch the angler's money. Management agencies also reaped a bountiful harvest in funding for studies, public license fees, and goodwill so long as large fish were present, and strong criticism when the fishery collapsed. They were not expected to reproduce but did so prolifically, soon to overpopulate the res- ervoir. Available forage fishes, mostly threadfin shad, were unable to sus- tain the population. Large bass starved and died, and recruits persisted by shifting to smaller, inadequate food sources such as crayfish and zooplank- ton. This sequence of events was repeated in Lake Powell and now is hap- pening in Lake Mohave, so regional management agencies have had ample opportunity and time to experience the repercussions of this management catastrophe. Instead of absorbing their losses and seeking ways to reduce the striped bass population or use it most efficiently in its less than trophy state, a proposal now exists to stock rainbow smelt in Lake Powell, to augment food supplies (Gustaveson et al., 1990~. The smelt, another predatory spe- cies, is expected to spread downstream through Marble/Grand canyons to Lake Mead and below. Proponents consider the smelt a potential salvation for striped bass fisheries of lower Colorado River basin reservoirs. I am of the opinion that if stocked and established, the species will be detrimental to both native and nonnative fishes throughout the system. I seriously doubt that smelt can sustain themselves in the face of striped bass predation, and certainly not for long at a level required to maintain large individual bass. I predict that they will reproduce and persist in low numbers in reservoirs, as do threadfin shad. There also is a good chance they will become locally abundant where safe from bass predation, notably in creek mouths and other quiet places other than reservoirs for which they are intended. In such places, the smelt will compete with and eat other desirable fishes. Introductions such as this, which will further damage fishes already clearly in jeopardy, must be prohibited. Last, and most important in the real world, is the realization that agen- cies mandated to perpetuate endangered fishes may not act soon enough to

NATIVE FISHES... 149 save the native fauna of the Colorado River. Politics of the 1990s may not allow them to do so. The "good years" immediately following endangered species legislation in the 1960s and 1970s are over. In that period, agencies were responsive and even eager to participate in conservation of imperiled species and their habitats. Today, and for much of the 1980s, conservation issues have fallen by the wayside (Deacon and Minckley, 1991~. Intransi- gence has become a major problem, albeit often masked by the "need" for interpretation of legislation and decisions regarding agency policy. Intransi- gence is deadly serious and unacceptable when dealing with extinction. Agency administrators must further avoid discretionary decisions not to implement programs or otherwise support recovery of a species because of other considerations; to do so is an abrogation of responsibility (Williams and Deacon, 1991~. The public must demand through their governmental representatives that these issues be reemphasized. The task of conservation is primary, a responsibility to which other priorities should be subjugated. Species are not renewable resources. CONCLUSIONS The prognosis for native fishes in the Colorado River basin is poor. Water development is proceeding apace, and conservation efforts are caught in a whirlpool of development-related rationalizations. Even a complex Re- covery Implementation Program for fishes in the upper Colorado River ba- sin allows some water developments to proceed so long as a one-time fee is applied to recovery efforts (U.S. Fish and Wildlife Service, 1987; Wydoski and Hamill, 1991~. The program funds research, purchases water rights for assurance of instream flow, and even examines curtailment of stocking of nonnative fishes that may harm natives, but it neither addresses definition of recovery (a common failing of most programs) nor provides assurances of alternative actions if the fishes have not benefited at the end of the designated program period (15 Years). As pointed out by Deacon and Minckley (1991), developments will be in place and operating, but the fishes may be extinct. When two organisms compete for a scarce resource such as water, the disadvantaged species seldom prevails. The biologists have done their jobs. We know the life cycles and habitat requirements of endangered western fishes. Studies of their genetics, behav- ior, and reproduction are completed or in progress. In most instances, as problems arise, technology is such that solutions have been found prior to loss of species. Broodstocks of all of the larger fishes are already available, reintroduction programs are under way in parts of the Colorado River basin (Johnson and Jensen, 1991), and I am convinced that a successful manage- ment program could be devised and implemented for the Grand Canyon region. Current politics stand in the way, just as surely as politics of the

150 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT 1960s aided and abetted our efforts to learn enough to save this fauna. We must have a renewed commitment by agencies and individuals. It is time for administrators to use available knowledge and expertise to fulfill their obli- gations, both ethically and under the law. ACKNOWLEDGMENTS I thank M. E. Douglas, T. E. Dowling, P. C. Marsh, C. O. Minckley, and S. P. Vives for reading and commenting on the manuscript. C. O. Minckley further provided literature and advice based on his long experience in Grand Canyon. J. N. Rinne supplied photographs. G. C. Clemmer, J. E. Deacon, and R. R. Miller provided copies of their field notes, which were referred to extensively but not used as much as I originally anticipated. I also thank G. R. Marzolf, S. D. David, and others of the Water Science and Technology Board of the National Research Council for the opportunity to participate in the symposium and communicate some of my ideas on native fish conserva- tion. APPENDIX I Accounts of Species Humpback Chub Humpback chubs are restricted to the largest rivers and lower parts of major tributaries of the Colorado River system. In the mainstream, adults characterize whitewater reaches, where they occupy deep, swirling eddies along canyon walls or concentrate in zones of turbulence near boulders and other obstructions. In the Little Colorado River they live in deep pools near rapids and eddies, typically concentrating near cover such as overhanging ledges. Young have been taken in habitats similar to those of adults, from creek mouths, or from flatwater downstream from canyons. The special morphologies of Colorado River fishes are combined in adult humpback chubs to form the most unique physiognomy of any North American minnow. Its most bizarre feature is a predorsal hump, rising in extreme individuals from the nape to extend anteriorly over the back of the skull (Figure 7-1~. The hump is of muscle, connected through a wedge-shaped caudal peduncle to the large, stiff caudal fin. All fins are expansive and thickened anteriorly. The body appears almost naked, since the scales are small and deeply embedded. The eyes are small, the snout is bulbous and fleshy and overhangs the mouth. As with many Colorado River fishes, humpbacks can grow in 2 or 3 years to a relatively large size (~30 cm long) and then slowly over a life span of 20+ years (unpublished data) to a maximum

NATIVE FISHES... 151 length of more than 50 cm. Some tagged adults recaptured up to ~ years later had increased in length by only a few millimeters per year (Minckley, 1988~. Prior to formation of Lake Powell, humpbacks must have inhabited most of the river in Marble/Grand canyons and ranged upstream into turbulent parts of Glen Canyon. As determined from archaeological remains (Miller, 1955), they also lived in canyons downstream from Lake Mead. The only reproducing population in the lower basin is in the lowermost Little Colo- rado River (U.S. Fish and Wildlife Service, l989b; Kubly, 1990~. The few fish taken in Marble/Grand canyons presumably originate from the Little Colorado River stock (Kaeding and Zimmerman, 1982, 1983; Maddux et al., 1987~. Other populations persist in the Yampa and Green rivers, upper Colorado River, and Cataract Canyon below the confluence of the Green and Colorado rivers (Tyus and Karp, 1989; Valdez, 1990~. The absence of reproducing humpback chubs in the lower mainstream is a function of water temperature. Natural river temperatures varied freezing in winter to ~30°C in summer. Hypolimnetic water from Lake Powell now enters the reach at 7.0-10°C. Volume of flow is such that even in summer it rarely warms to 16°C by the time it enters Lake Mead, more than 400 km downstream (Cole and Kubly, 1976; Kubly and Cole, 1979~. The same pattern of avoidance of cold hypolimnetic water by native fishes was well documented below Flaming Gorge Dam (Vanicek et al., 1970; Holden and Stalnaker, 1975~. Only a few Nonh American cyprinids reproduce at such low temperatures (G. R. Smith, 1981), and no Colorado River species is known to do so, with the possible exception of the speckled dace (in part Marsh, 1985~. Successful development and hatching of humpback chub eggs require 15°C or higher, and natural reproduction most likely occurs between 19 and 22°C (Hamman, 1982a, b; Kaeding and Zimmerman, 1983~. The Little Colorado River maintains much of its natural character despite upstream development (Miller, 1961~. Snowmelt produces floods. The rest of the year, collective inputs of springs in its lower 21 km maintain a baseflow of ~5.6 m3 so, punctuated by infrequent runoff. Above its mouth, the stream flows in runs, rapids, and pools through a deep, narrow canyon. Travertine deposits from carbonate-rich springs are dominant features, forming scalloped cascades as well as a major waterfall 5.0 m high ~15 km up- stream. At baseflow, the stream runs clear but milky in appearance as a result of suspended carbonates. Spring waters are also high in sodium chlo- ride, with total salinities varying to more than 3.0 g liter~~ (Cooley et al., 1969; Kubly and Cole, 1979~. The mouth of the Little Colorado fluctuates in depth and character as water releases from Glen Canyon Dam pass through the Colorado River mainstream. At low release rates, the Little Colorado enters the mainstream through a swift chute ~40 m wide. At high releases in the Colorado River,

152 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT the mouth is ponded upstream for 0.5 km and varies from 1.0 to more than 3.0 m deep (Minckley, 1988~. Humpback chubs have consistently been the most abundant fish caught in the Little Colorado River over 14 years of sampling. When temperatures are similar in the two streams, chubs either move into the Colorado (Minckley et al., 1981) or swim back and forth between rivers. They move into the warmer Little Colorado in spring and summer, staging and spawning in the mouth, with some also moving upstream to spawn (Kaeding and Zimmerman, 1983; C. O. Minckley, 1988, 1989~. Population sizes vary over the repro- ductive season (Kubly, 1990), indicating considerable movement of fish both within the Little Colorado River and between it and the Colorado River. Quantitative estimates of adults in spring, generated from tag-recap- ture data in 1980-1981 (Kaeding and Zimmerman, 1982) and in the period 1987-1989 (C. O. Minckley, 198S, 1989), respectively indicated 7,000-8,000, 5,783, 7,060, and 10,120 fish. Other native species in the Little Colorado River include speckled dace and flannelmouth, bluehead, and razorback suckers, in decreasing order of abundance. Squawfish and bonytail were reported in the past below Grand Falls, 120 km upstream from the mouth (Miller, 1963b). Numerous alien fishes are found, but none abundantly. Fathead minnows, channel catfish, and carp are the most consistent in occurrence, and all three reproduce there. Perhaps salinity, carbonate deposition, floods, or some other special features precludes large populations. Food habits of humpback chubs have not been thoroughly studied. Kacding and Zimmerman (1983) reported mostly aquatic invertebrates in 27 fish that contained foods; one had eaten a fathead minnow. Three juveniles exam- ined by C. O. Minckley (1980; Minckley et al., 1981) had eaten aquatic invertebrates and algae. Some adults from the mouth of the Little Colorado defecated or had their guts filled with the macroalga Cladophora glomerata, which indicated recent ingress from the adjacent mainstream where the alga was common (Minckley, 1988~. Kubly (1990) also noted filamentous algae in 12 adults that contained foods; an unidentified fish was found, along with chironomid dipteran larvae and terrestrial insects. A few chubs caught be- low Glen Canyon Dam in the 1960s were filled with zooplankton (Minckley, 1973), and Tyus and Minckley (1988) recorded extensive feeding on crick- ets floating on the surface and used as bait in the Yampa River. Bonytail The ecology of the bonytail is poorly understood for a number of rea- sons. First, the vernacular "ponytail" was applied for years to any Colorado River basin chub of the genus Gila; therefore, older literature may or may not be usable. Second' it was long considered a subspecies of roundtail

NATIVE FISHES... 153 chub, and some workers combined the two kinds. Third, young chubs may be difficult to identify and are often lumped as "Gila spp.," or the rare species is simply ignored or missed in sorting. Next, apparent hybridization between chubs and questions of genetic integrity (Holder and Stalnaker, 1970; Valdez and Clemmer, 1982; Kaeding et al., 1986; Douglas et al., 1989; U.S. Fish and Wildlife Service, 1989b, c) have confused the issue even more. Last, bonytail are now too rare to study in nature (U.S. Fish and Wildlife Service, 1989c). While all this has gone on, the species declined in abundance to become the most endangered freshwater fish in North America. It now persists almost exclusively under artificial conditions (Minckley, 1985; Minckley et al., 1989~. Adult bonytails carry streamlining to different extremes than do hump- back chubs. They are slim and elongate, with small, depressed skulls that rise smoothly to low, predorsal humps. The dorsal profile descends posteri- orly to an attenuate, pencil-thin caudal peduncle, ending in a large, strong, deeply forked caudal fin (Figure 7-2~. Its skin is thick, with deeply embed- ded scales. The body and fins are reminiscent of a pelagic member of the mid-oceanic community, e.g., a mackerel or tuna characterized by powerful and sustained swimming yet capable of sudden bursts of speed to attack prey or avoid a predator. In Lake Mohave, the last place wild bonytails may be predictably collected, they are often netted in midwater over shoals 5.0 to 10 m deep and seem to seek areas a few tens of meters from wave- washed shorelines during storms. My best reconstruction of the natural habitat of bonytail is based on impressions from older literature, experience with formerly occupied habi- tats, and watching captives. The only direct observations known to me were those of Jonez and Sumner (1954), who "observed [it] in the sandy areas along the Colorado River." I suspect that the fish characterized valley reaches and perhaps flatwaters in canyons. Adult bonytail must have lived in mid- channel, maintaining position with low-amplitude beats of the caudal fin as they do in hatchery raceways, and moving from midwater to surface and bottom to inspect and feed on drifting particles in the water column. The potential power indicated by morphology would have been used to pass through zones of turbulence or escape predators. One adult caught and preserved near Bright Angel Creek in the 1940s (Miller, 1946a) and skeletal material from Stanton's Cave (Miller and Smith, 1984) document bonytail presence in Marble/Grand canyons. Adults were often taken along with humpbacks, roundtail chubs, and putative hybrids between Lees Ferry and Glen Canyon Dam, and in Lake Powell, for a few years following closure of the dam (Stone, 1965-1969, 1971, 1972; Stone and Rathbun, 1967-1969; Holden and Stalnaker, 1970~. Bonytalls were common in downstream reaches now inundated by reservoirs (Dill, 1944; Jonez et al., 1951; Wallis, 1951~. Jonez and Sumner (1954) observed small schools

154 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT (rarely more than 10 fish) scattered throughout Lake Mead in 1951-1954. All were longer than 25 cm, and no reproduction was seen although ripe females 30-36 cm long were observed along shorelines in March. They also saw breeding adults, apparent spawning aggregations, and actual spawning by at least 500 adults in Lake Mohave in May; 42 males and 21 females netted were 28-36 cm long. Spawning was observed with the aid of diving gear over a gravel shelf ~400 m long (Jonez and Sumner, 1954, p. 141- 142~: It appeared that each female had three to five male escorts. Neither males nor females dug nests, and the eggs were broadcast on the gravel shelf .... No effort was made to protect the eggs by covering them with gravel or by guarding them. However, the eggs adhered to the rocks, and that gave them some protection .... The bonytails found in the spawning area had bright reddish-gold sides and bellies. This color is not seen . . . at other times of the year; it is assumed that it is the spawning coloration. Bonytails from the swift-moving upper portion of Lake Mohave, although they were not observed spawning, occasionally were seen in like coloration in the spring of the year .... Large schools of adult carp were intermingling with the spawning bonytail. No young . . . were observed in the spawning area, and it is presumed that there will be enough survival . . . so that the bonytail will not become extinct in Lake Mohave. As already noted, the Lake Mohave population persists today, with only limited recruitment (Minckley et al., 1989~. The few available data indicate that bonytails feed on benthic and drift- ing aquatic invertebrates and terrestrial insects (Kirsch, 1889; Vanicek, 1967~. Jonez and Sumner (1954) reported plankton, algae, insects, and organic debris eaten in Lake Mead, and stomachs of adults from lakes Mohave and Havasu contained zooplankton (Minckley, 1973~. Like humpback chubs, bonytail grow quickly to large sizes. Some grow longer than 30 cm in their first year of life in artificial ponds (P. C. Marsh, unpublished data). Most wild fish captured from Lake Mohave vary be- tween 45 and 55 cm in total length, and their estimated ages average near 40 years (Minckley et al., 19891. The largest individual known to me was caught in 1969 from the Colorado River below Davis Dam (unpublished data). It was ~64 cm long, estimated from other objects of known size in a photograph obtained by Gary Edwards (then with Arizona Game and Fish Department). Colorado Squawlish This giant minnow once achieved lengths of 1.8 m and weights of 36 kg. It was most commonly caught during the spring-summer spawning migra-

NATIVE FISHES... 155 tion and staging of adults on and near the spawning grounds. Concentra- tions in canals and downstream from lower basin diversions before 1915 commonly included individuals up to 20 kg in weight; they and other native fishes were used as food for humans and domestic pigs and as fertilizer (Miller, 1961; Minckley, 1965, 1973~. Adult squawfish (Figure 7-3) are elongate and torpedo shaped, with long, flat heads, large mouths, and strong jaws. Large adults become markedly broadened across the back, reflecting their thick and powerful musculature. The caudal peduncle is stocky and short. The fins are relatively small, with the exception of a strong, expanded caudal, and the dorsal and pelvic fins are far back on the body. The scales are small and embedded, especially on the breast and belly, and the eyes are small. Young squawfish are more delicate in appearance, although clearly streamlined. There are far fewer squawfish today than in the past. Miller (1961y, Minckley (1965, 1973, 1985), and Minckley and Deacon (1968) documented their extirpation, mostly before 1970, from the Colorado River basin down- stream from Grand Canyon. One preserved specimen is known from near Lees Ferry in 1934 (University of Michigan number 117840), and another, from the mouth of Havasu Creek (Arizona State University number 7087), was caught by an angler in 1975 (Smith et al., 1979~. Both are juveniles less than 40 cm long. Skeletal remains from Stanton's Cave were of fish be- tween 1.2 and 1.5 m long (Miller and Smith, 1984~. A verbal report exists for squawfish in the Little Colorado River at Grand Falls (Miller, 1963b), and the species still lives in Lake Powell and in variable abundance up- stream (U.S. Fish and Wildlife Service, 1989d). Colorado squawfish have been studied in the upper Colorado basin for almost three decades. Using various methods, Tyus (1991) estimated ~8,000 adults in the upper Green and Yampa rivers in the period 1980-1989. Abso- lute numbers are about an order of magnitude lower in the Colorado River above and below its junction with the Green (Valdez et al. 1982a,b; Valdez, 1990), and the fish is even more scarce in the San Juan River (Platania et al., in press). Adult squawfish in the Green and Yampa rivers are solitary in relatively shallow, shoreline habitats, including shallow inlets and backwaters (Tyus and McAda, 1984~. They occupy the same short reaches of river (~5.0 km long or less) from autumn through spring (Tyus, 1991~. Spawning migra- tions begin with receding water levels in spring and early summer. Some fish move 200 km or more, passing through many kilometers of apparently suitabe habitat to arrive at one of two widely separated spawning grounds by late June to early August. Colorado squawfish have shown unerring fidelity for one or the other spawning reach. The same fish used the same reach over a number of different years, and no fish changed from one to the other; olfactory homing was proposed to explain this behavior (Tyus, 1985~.

156 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT On the spawning grounds, fish alternate between resting-staging areas in large pools or eddies and deposition-fertilization habitat on rapids. After spawning, adults return over a period of days or weeks to their previously occupied stream segments. Squawfish eggs adhere to cobble and gravel of clean-swept bottoms, and larvae hatch in 3.5-6.0 days at 20-22°C (Hamman, 1981; Marsh, 1985~. Temperatures at oviposition are typically higher than 18°C, and Marsh (1985) demonstrated dramatic reduction in hatching at temperatures less than 15°C. Under natural conditions, larvae emerge from the substrate in 3.0-15 days to move downstream. After a few weeks, the young concentrate 100 to 150 km below the spawning site in shallow, shoreline embayments formed as water levels recede through summer and autumn. Little is known of the ecology of juveniles from ~10 to 30 cm long. Field data indicate the greatest numbers in lower sections of the Green River and more adults upstream (Tyus, 1986; Tyus et al., 1987~. Partial separation of life stages may serve to reduce cannibalism (Tyus, 19911. Repopulation of upstream reaches must occur by subadults (Figure 7-4), 40-50 cm in total length. Colorado squawfish grow more slowly than other native species. Rinne et al. (1986) reported growth to ~50 cm long in 9 years under hatchery conditions and speculated that a 1.~-m individual would be 50 or more years old or older. I suspect that estimate is conservative and would not be surprised if such a fish were 75 years old or older. Huge fish no longer occur, perhaps because of a high susceptibility to fishermen (Tyus, 1991~. Growth rates of 59 tagged adults recaptured in the Green River averaged only 11.2 mm a year (Tyus, 1988~. In 1981-1988, 14 ripe females caught on Green River spawning grounds averaged 65 cm long and 194 males aver- aged 56 cm long (Tyus and Karp, 1989~. Examination of polished otoliths from wild Green River adults indicates ages of up to 30 years. Small squawfish eat aquatic insects and crustaceans until they are 50-100 mm long and then shift to fishes (Vanicek, 1967; Vanicek and Kramer, 1969; Holden and Wick, 1982~. Adults are piscivorous, lie-in-wait or slow- stalking predators, attacking in a rush that ends in strong suction created by opening the cavernous mouth. Under pristine conditions they must have eaten suckers most frequently. Today they feed on introduced fishes as well, sometimes choking on the erected fin spines of channel catfish that they attempt to swallow (Pimental et al., 1985~. In 1979, I theorized that adult squawfish in the lower Colorado River basin lived on the vast and complex Colorado delta (Minckley, 1979), mi- grating upstream to canyons of the Colorado and Gila rivers to spawn. Now we know more of their ecology, and if the pattern in the lower Colorado/ Gila was like that in the Green River, adults congregated from their indi- vidual home ranges along both watercourses and migrated to spawn in whitewate

NATIVE FISHES... 157 canyons. The broad, rich delta might well have been a major nursery area. However, striped mullet, entering the delta from the Gulf of California in remarkable numbers (Hendricks, 1961), may still have attracted adult squawfish. Size and diversity of the delta prior to the construction of dams (Sykes, 1937) could have provided ample space and food for adult and juvenile squawfish alike. Whatever the case, squawfish migration and reproduction in the lower basin were much earlier in the year than upstream. Young-of-the year seined in mid-May in southern Arizona were 32 mm long (Sigler and Miller, 1963) fully 2 months before spawning even typically occurs in the upper Green River. Newspaper accounts to be detailed elsewhere (W. L. Minckley and D. E. Brown, unpublished data) track angler catches of "white salmon" from the Gila River in the late 1800s at the Colorado-Gila River confluence near Yuma in March, at Gila Bend (~160 km upstream) in March-April, at Gillespie Dam (220 km upstream) in March-April, and at Tempe (~300 km upstream) in April-May. It seems likely that this sequence of catches fol- lows the upstream migration of fish to reproduce. Snowmelt often begins by early March in the Gila River basin and ends in May, and the spawning grounds were likely in the Salt River Canyon, another 50+ km upstream from Tempe, where `'commercial numbers" of squawfish were reported in those days (Chamberlain, 1904; Minckley, 1965, 1973~. Whitewater can- yons upstream from valley reaches that to serve as nurseries were also present on the Verde and upper Gila rivers, and adult squawfish were in both those systems (Minckley, 1973~. I have no comparable data for the lower Colorado mainstem but expect that runs were later there than on the Gila. Spring floods at Yuma extended well into summer because of the vast distances from sources of water; therefore, the fish might have moved any time from March through July. Adults aggregating from the delta and lower river may well have traveled more up the Colorado River than up the smaller Gila; distances from the delta to whitewater were a bit shorter in the former. Broad valleys of the lower river, including the basins now flooded by Lake Mead, would also have supported adults that could have spawned in upstream or downstream canyons. Speckled Dace This small minnow is the most widespread freshwater fish west of the Rocky Mountains, ranging from southern Arizona and California to Canada in desert springs, creeks, rivers, and high mountain brooks and lakes. The ecological breadth of this species is paralleled by its diverse morphology. Numerous subspecies are described, others are yet undescribed, and some workers (e.g., Minckley et al., 1986) suggest that the taxon actually repre-

158 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT sents a complex of species. In the Grand Canyon region, a mainstream form that also moves in and out of tributary mouths is referred to Rhinichtys osculus jarrowi (Figure 7-5~. Stocks in tributaries above barriers are mor- phologically differentiated but have not been identified to subspecies. The mainstream form has a terete body, with larger, longer, and more falcate fins, sharper snout, longer head, and body pigmentation consisting of dark, longitudinal stripes or a uniform brown dorsum contrasting with a white belly. Distinctive tributary populations tend to have robust bodies with small, rounded fins, short, blunt snouts and head, and strong black speckles over- lying an olive ground color on the upper body. Speckles or blotches some- times extend onto the venter. Breeding males of all the forms develop in- tense red pigmentation on the lower head, bases of fins, belly, and lateral body surfaces. Reproductive females remain drab or have yellow or orange pigmentation on the bases of ventral fins. Speckled dace remain locally common in the channel upstream from Lake Mead, especially in creek mouths (Minckley and Blinn, 1976; Minckley, 1978; Carothers and Minckley, 1981~. Large numbers can also be seined along sandbars in the river, but their actual abundance in that habitat has not been adequately assessed. They enter tributaries in March and spawn in April and May. Carothers and Minckley (1981) thought that reproduction was limited to tributaries at temperatures of 17 to 23°C. Young fish are common by late May and June. Both young and adults remain in lower parts of creeks throughout the summer and autumn, only to disappear into the mainstream in winter. Minckley (1978) attributed this winter exodus to interference by spawning aggregations of trout, but it also seems possible that winter river temperatures as warm as in tributaries make the main- stream more acceptable at that time of year. Speckled dace achieve adulthood in a year or less. They rarely exceed 80-100 mm in length, and they live 2 or 3 years (Minckley, 19739. Foods in both mainstream and tributary habitats are mostly benthic invertebrates and organic debris (Carothers and Minckley, 1981~. Ubiquity of the speckled dace spells danger, especially if it indeed repre- sents a complex of species rather than a single, variable, and widespread taxon. Commenness breeds complacency (or contempt), and most studies targeting a threatened or endangered species ignore other the other local fauna. At an extreme, significant numbers of streams have been poisoned for management toward recovery of endemic bouts (and even a few min- nows), but associated native species have suffered as a result (Rinne and Turner, 1991~. The trout is reintroduced after its habitat is renovated, but other warmwater, nontarget species are not. Speckled dace often bear the brunt of such a practice, and although we do not know what may currently exist with regard to this native species, we too often seem glibly prepared to allow the unexplored to vanish.

NATIVE FISHES... Flannelmouth Sucker 159 This is another species that persists in Marble/Grand canyons and throughout much of the upper Colorado River basin. It also occupies the Virgin River, Nevada-Anzona-Utah, which flows into Lake Mead. The flannelmouth sucker was described from the Gila River drainage of southern Arizona and also lived in the lower Colorado River near Yuma. It is now extirpated below Lake Mead (Minckley, 1985~. W. L. Minckley (1980) examined flannelmouth populations in the Grand Canyon region, concluding that morphological variation was greater than to be expected in a single species. A related, undescribed form lives in the Little Colorado River (Minckley, 1973, 1980~. This complex has only a few special features of body shape. Their bodies are thick anteriorly and slim posteriorly. The anterior scales are small and embedded in thick skin. The fins are large, and the caudal peduncle is relatively thin. Nonbreeding fish are light gray or tan dorsally and white on the belly (Figure 7-6~. Breeders are dark, with orange pigment on the fins and sometimes in a weak lateral band. Some adults exceed a length of 60 cm. Sexual maturity occurs at - 0 cm in the Yampa River (McAda and Wydoski, 1985~. Adult flannelmouths often lie in quiet tributary mouths, especially in the Paria River near Lees Ferry when the adjacent river is cold. Spawning was recorded by Carothers and Minckley (1981) in most larger, lower gradient stream mouths along Marble/Grand canyons. Ripe fish were caught in March to May in the Paria River at water temperatures of 17-23°C, and postreproductive fish remained there through the summer. As noted for other natives, adult flannelmouths left the Paria when its temperatures and that of the main- stream Colorado equilibrated in winter (Suttkus and Clemmer, 1979~. In the Yampa River, ripe adults appear from April through early June. They concentrate at the upstream ends of cobble bars in water 1.0 m deep and at velocities of ~1.0 m so (McAda and Wydoski, 1985~; abundant young, 30-40 mm long, are present by midsummer. Postreproductive adults remain in flatwater or eddies and margins of strong currents, generally in water 1 m or more deep. Young congregate on and downstream from riffles and along shorelines of flatwater reaches and in tributaries. Gut contents of adult flannelmouths were similar in mainstream and tributaries of Marble/Grand canyons and included aquatic invertebrates, organic de- bris, and sand (the last reflecting bottom-feeding habits). Most invertebrates were larvae of aquatic dipterans, along with an introduced amphipod crusta- cean now common in the channel. Mainstream flannelmouths also com- monly ate Cladophora glomerata. The alga was scarcely present in fish from tributaries (Carothers and Minckley, 1981~. The maximum estimated age based on growth rings on scales of upper basin fish was eight or nine years (Wiltzius, 1976; McAda, 1977; McAda

160 COLORADO RIVER ECOLOGY AlID DAM MANAGEMENT and Wydoski, 1985~. The scale method was rejected for large suckers in Marble/ Grand canyons by Usher et al. (1980) because a high percentage of the structures were regenerated and unreadable. The method has also proven unreliable for large razorback suckers in the lower basin. Minckley (1983) and McCarthy and Minckley (1987) reported that presumed annul) were obscure or too near one another to allow reliable separation after the first few years, and did not form at all in some years, all resulting in age under- estimation. Usher et al. (1980) and Carothers and Minckley (1981) con- cluded from opercular bones that the fish lived to a maximum of 10 years, which I also judge to be an underestimate. At the time of their study, neither the potential for long life nor the exceedingly slow growth now known for older Colorado River fishes was suspected. Thus, closely spaced annul) near the opercular margins may have been overlooked or misread. Scoppettone (1988) examined opercles from 30 Green River flannelmouths, the oldest of which was 30 years old and 53 cm long. Five sets of polished otoliths that I examined from Green River flannelmouths 50-59 cm in total length all were older than 17 years (unpublished data; the oldest fish was ~35 years old and 59 cm long). By any method of determination, it is clear that flannelmouths grow quickly in the first 3-5 years of life and then slow abruptly. Bluehead Sucker This is one of a group of specialized, mostly herbivorous fishes distrib- uted in high-gradient streams of western North America (Miller, 1959; Minckley et al., 19863. Smith (1966) placed them as the subgenus Pantosteus of Catostomus, while Minckley (1973) retained full generic status for the group. Their feeding adaptations include broad, disc-shaped lips and strong cartilaginous sheaths on the jaws, used for grazing films of algae and other organisms. There are few records of bluehead suckers downstream from Grand Can- yon. Miller (1952) reported them as bait along the lower river, and some perhaps from that source were netted from reservoirs by Jonez and Sumner (1954~. Perhaps the species was excluded downstream by the predominance of shifting sand bottoms. The Virgin River, tributary to the Colorado before being isolated by Lake Mead, is inhabited by a different species (Smith, 1966; Minckley et al., 1986~. Bluehead suckers in large rivers grow to ~50 cm long. They are bottom dwellers, with heads and bodies rounded above and flattened below, expan- sive fins, and thick skins. They have small, embedded scales on the anterior body and tiny eyes. The most remarkable feature of mainstream blueheads is their caudal peduncles, which vary among individuals from elongate and thin like that of a bonytail to short and thick like that of the Colorado squawfish (Figure 7-7~. Specimens with intermediate peduncles also occur. The significance of this remarkable polymorphism, and its maintenance where

NATIVE FISHES... 161 extremes co-occur, are yet to be explained. Populations in the Little Colo- rado River are morphologically distinct (Minckley, 1973), with thick, chubby bodies, small rounded fins, and body lengths rarely exceeding 20 cm. Nonbreeding, mainstream adults are bluish gray or olivaceous in color, with white bellies. The common name comes from breeding colors of adult males, which develop a blue patch on the top of the head that sometimes suffuses downward over the opercles. The lower fins become yellow or orange, and red or rosy lateral bands form along the sides. When the water is clear, adult blueheads stay in deep pools and eddies in daytime, moving to shallow riffles, along shore, or to other hard-bottomed places to feed at night. When turbidity is high, they occupy shallow places throughout the day. Bluehead suckers seem most common at the upper ends of valleys or where shallows are available in whitewater canyons, but they tend to be rare elsewhere. Young are on and downstream from riffles and along shore in flatwaters. Like other natives, blueheads entered tributaries of Marble/Grand canyons in spring through autumn but were rare in winter. Despite adaptations for scraping algae, foods of bluehead suckers in Marble/ Grand canyons were mostly immature dipterans and amphipods in the main- stream and dipterans in tributaries (Carothers and Minckley, 1981~. Dia- toms and organic debris were nonetheless abundant in guts at all seasons. The same comments as before apply to reported ages of this species; the 8.0-year maximum age for Marble/Grand canyons fish reported by Usher et al. (1980) (Carothers and Minckley, 1981) may have been underestimated. Scoppettone (1988) used opercles to determine the age of a 40-cm specimen from the Green River at 20 years, while I (unpublished data) used otoliths estimate an age of greater than 20 years for an unusually large bluehead sucker (47 cm long) from the Yampa River. Growth rates of blueheads in the upper basin are rapid in the first few years of life, only to slow abruptly when sexual maturity is attained at 30+ cm total length (Wiltzius, 1976; McAda, 1977; McAda and Wydoski, 1983~. Blueheads spawn over mixed gravel-sand or gravel-cobble bottoms in moving water of tributaries in the Grand Canyon region, typically in April and May. Water depths vary from a few centimeters to deeper than a meter, and temperatures are typically 16 and often 20°C. A single female is joined by two to five males, and gametes are deposited on and within the bottom (Suttkus and Clemmer, 1979; Maddux and Kepner, 1988~. Young appear in May and June. In the upper basin, they spawn over gravel and rubble, sometimes in very swift currents (faster than 1.0 m so) near the upper ends of riffles, and mostly in flatwater reaches. Razorback Sucker Minckley et al. (1991) reviewed the biology of the razorback sucker, presented details or~ its current status, documented its decline, and reported

162 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT on efforts for its recovery and management. Much of the following is ab- stracted from that work, which should be consulted for additional refer- ences. This species, at maxima of ~75 cm long and 4.0+ kg in weight, is clearly the largest catostomid in the Colorado basin and second in size among all fishes to the Colorado squawfish. The razorback is named for a laterally compressed, sharp-edged keel rising abruptly from the nape and extending to the dorsal fin. The anterior body is thick and triangular in cross section (Figure 7-8), narrowing abruptly to the caudal fin. The skull is broad and flat to concave on the top and flattened on the bottom. While the relatively small mouths of most suckers extend ventrally when opened, the large mouth of the razorback can be protruded anteriorly. Razorbacks have tiny eyes, thick and leathery skins, small and embedded scales, and large, strong fins. Their dorsal and lateral musculature is thick and well developed, extending back through a robust caudal peduncle. Sexual dimorphism is pronounced. Males tend to have higher, sharper keels and larger fins, while females develop large, edematous urogenital papillae when breeding. Razorback suckers are brown to brownish black on the back and yellow- ish on the belly (Figure 7-9), colors which darken and intensify during spawning. The dorsal body surfaces become suffused with dark red, concen- trated laterally into a distinct lateral band. Young razorbacks are olivaceous to tan in color, and they resemble juvenile flannelmouth suckers until the keel develops when the fish is 80-100 mm in total length (Minckley and Gustafson, 19821. This fish remains abundant only in Lake Mohave, Arizona-Nevada (~50,000 adults; Marsh and Minckley in press). It is rare or absent from the rest of the lower Colorado, and extirpated from the Gila River drainage prior to the 1960s. Razorbacks have essentially disappeared from Lake Mead, occur uncommonly in Lake Powell, and are also rare in the upper basin. An estimated 1000 razorback suckers persist in the upper Green River, the largest known concentration upstream from Lake Powell (Lanigan and Tyus, 1989~. One specimen from the Grand Canyon region was caught by an angler in 1944, Arizona Game and Fish Department personnel caught another near Lees Ferry in 1963, four were captured or seen in the Paria River in 1978- 1979 (Minckley and Carothers, 1980), one was taken in the mainstream in 1986, three were seen in 1987 in Bright Angel Creek, and two and three, respectively, were netted in the mouth of the Little Colorado River in 1989 and 1990. Putative hybrid razorback x flannelmouth suckers have also been recorded (Suttkus and Clemmer, 1979; Carothers and Minckley, 1981~. Con- sidering the size and complexity of the 400+ km of river in Marble/ Grand canyons, the number and diversity of tributaries, and the comparatively limited sampling for such a massive habitat, records of 13 razorback suck-

NATIVE FISHES... 163 ers since 1978 may indicate a significant remnant population. All were large adults, but all known populations consist of large, old fish. No recruit- ment is indicated for any wild population for more than 20 years. In the upper basin, razorback suckers spawn near the upstream ends of gravel-cobble riffles during May and early June, at water temperature higher than 16°C and velocities less than 1.0 m so, and generally in places less than 1 m deep (McAda and Wydoski, 1980; Tyus, 1987~. Turbidities are too high for direct spawning observations in the Green River, but in clear water in the short reach of river below Lake Mead, two or more males accosted each female entering a spawning Tea (Mueller, 1989~. Eggs were deposited in depressions formed by movements of the large fish against the bottom. The ecology of juveniles, only a few of which have ever been collected, is unknown. For years after impoundment, hundreds of breeding razorbacks aggre- gated over wave-washed, gravel-cobble bottoms along shorelines in lakes Mead, Mohave, and Havasu (Walks, 1951; Douglas, 1952; Jonez and Sumner, 1954~. There is evidence of the same phenomenon in reservoirs on the Salt River, Arizona, before the 1950s (Hubbs and Miller, 1953; Minckley, 1983~. These fish were either trapped by the dams or, most likely, originated from high production and survival of young before populations of exotic fishes exploded in the newly formed lakes. Based on paleo-Indian fishing wiers and other indirect evidence, razorbacks did the same thing in the Salton Sea when it was periodically filled from Colorado River floods in the distant past (Minckley et al., 1991~. Razorback suckers have consistently disappeared from lower basin reser- voirs 40-50 years after impoundment (Minckley, 1983~. As already noted, only Lake Mohave, formed by Davis Dam in 1954, now supports the spe- cies. If otolith aging is correct (McCarthy and Minckley, 1987), Lake Mohave fish averaged ~35 years old in 1980-1981. Most hatched about the time the lake was formed and thus may have only 5.0-15 years of longevity to go. It is notable in this regard that 10 Green River razorbacks taken downriver from Flaming Gorge Dam mostly after 1980, averaged ~29 years old (W. L. Minckley, 1989~; that reservoir was closed in 1963. Razorback spawning in Lake Mohave is most intense in late January through early April at temperatures higher than 15°C. Larvae emerge in a few days, move inshore, develop to ~12 mm long, and then disappear. Despite production of strong larval year classes, no recruits to the adult population have been discovered in sampling since 1967. Three hypotheses have been proposed to explain the larval disappear- ance: (1) starvation, (2) entrainment and transport from the reservoir, and (3) predation by nonnative fishes. The last seems most parsimonious. After removal of nonnative predators by rotenone, adults placed in a backwater isolated from the lake by a gravel spit reproduced naturally, and young fish

164 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT grew to 25 cm long or more in less than a year. The species also reproduces successfully, and young grow rapidly to large size in hatchery or other artificial ponds so long as predatory exotics are denied access (Minckley et al., 1991). Food habits of razorback suckers are poorly known. In reservoirs, both larvae and adults eat zooplankton, and the large, terminal mouths and spe- cialized gill rakers seem adaptations for seiving suspended materials from the water column (Hubbs and Miller, 1953; Marsh, 1987~. This characteris- tic fits if the species evolved in ancient lakes, which is possible although no fossils have been found in lacustrine sediments, or if they primarily inhab- ited backwaters and oxbows along the pristine river. Another alternative lies in the probable efficiency of such an anatomical system for seiving detritus from the stream itself. As noted before, finely ground detritus must have been far more abundant in the past than is com- monly realized. Razorbacks in the Green River channel apparently lie in depressions behind current-formed dunes on the bottom, where particulate organic material collects after elutnaiion from entrainment in the shifting, inorganic bottoms. This detritus is thus available as food, at least for a time until it is suspended and carried away or reinterred by ever-moving sand. Guts of razorbacks caught from rivers have usually contained large propor- tions of "organic matter," along with lesser amounts of aquatic inverte- brates and considerable sand (Marsh, 1987; Minckley et al., 1991~. Perhaps a large enough population to study, if ever found or reestablished, will provide tests for these speculations. REFERENCES Abarca, F. J. 1989. Potential food habit overlap between sp~kedace (Meda fulgida) and red shiner (Cyprinella lutrensis). Unpublished Master's Thesis, Arizona State University, Tempe. 66 pages. Abbott, C. C. 1860. Descriptions of four new species of North American Cyprinidae. Proceed- ings of the Philadelphia Academy of Natural Science 12:473-474. Adams, V. D., and V. A. Lamarra, eds. 1983. Aquatic Resources Management of the Colorado River Ecosystem. Ann Arbor Science Publisher, Ann Arbor, Mich. 697 pages. Allan, R. C., and D. L. Roden. 1978. Fish of Lake Mead and Lake Mohave. Nevada Depart- ment of Wildlife Biological Bulletin 7:1-105. Arizona Game and Fish Department. 1978. Letter to U.S. National Park Service, Grand Can- yon National Park, relative to U.S. National Park Service (1977) (see below). Office of the Director, Arizona Game and Fish Department, Phoenix. Baker, J. R., and L. J. Paulson. 1983. The effects of limited food availability on the striped bass fishery in Lake Mead. P. 551-561, in V.D. Adams and V.A. Lamarra, editors, Aquatic Resources Management of the Colorado River Ecosystem. Ann Arbor Science Publisher, Ann Arbor, Mich. Baird, S. F., and C. Girard. 1853a. Descnptions of some new fishes from the River Zu:i. Proceedings of the Philadelphia Academy of Natural Science 6:368-369. Baird, S. F., and C. Girard. 1 853b. Descriptions of some new fish collected by John M. Clark.

NATIVE FISlIES... 165 on the United States and Mexican Border survey, under Lt. Col. Jas. D. Graham. Proceed- ings of the Philadelphia Academy of Natural Science 6:387-390. Baird, S. F., and C. Girard. 1853c. Fishes. In Report of an expedition down the Zu:i and Colorado rivers by Capt. L. Sitgreaves. Thirty-second Congress, Second Session, Executive Report 59:48-59. Beer, B. 1988. We Swam the Grand Canyon: The True Story of a Cheap Vacation That Got a Little Out of Hand. The Mountaineers, Seattle. 171 pages. Behnke, R. J., and D. E. Benson. 1983. Endangered and threatened fishes of the upper Colo- rado River basin. Colorado State University Cooperative Extension Service Bulletin 503A: 1- 38. Beland, R. D. 1953a. The effect of channelizaiion on the fishery of the lower Colorado River. California Fish and Game 39:137-139. Beland, R. D. 1953b. The occurrence of two additional centrarchids in the lower Colorado River. California Fish and Game 39:149-151. Bishop, A. B., and D. P. Porcella. 1980. Physical and ecological aspects of the upper Colorado River basin. P. 17-56, in W. O. Spofford, A. L. Parker, and A. V. Kneese, eds, Energy Development in the Southwest~roblems of Water, Fish and Wildlife in the Upper Colo- rado River Basin, Volume 1. Resources for the Future, Washington, D.C. Brooks, J. P. 1931, Field observations. Grand Canyon Nature Notes 4(10):70. Bruns, D. A., and W. L. Minckley. 1980. Distribution and abundance of benthic invertebrates in a Sonoran Desert stream. Journal of Arid Environments 3:117-131. Carlson, C. A., and R. T. Muth. 1989. The Colorado River: Lifeline of the American South- west. P. 220-239, in D. P. Dodge, ea., Proceedings of the International Large River Sympo- sium. Canadian Journal of Fisheries and Aquatic Science, Special Publication 106. Carothers, S. W., and C. O. Minckley. 1981. A survey of the fishes, aquatic invertebrates, and aquatic plants of the Colorado River and selected tributaries from Lee's Ferry to Separation Rapid. Final Report, U.S. Bureau of Reclamation Contract 7-07-30-X0026, Lower Colo- rado River Region, Boulder City, Nev. Museum of Northern Arizona, Flagstaff. 401 p. Caylor, D. A. 1986. Characteristics and perceptions of sportfishing visitors to Lee's Ferry, Anzona: Implications for Management. Unpublished Master's Thesis, Northem Arizona University, Flagstaff. 126 p. Chamberlain, F. W. 1904. Unpublished Arizona field notes. On file at Smithsonian Institution, Washington, D.C. Cole, G. A., and D. M. Kubly. 1976. Limnological studies on the Colorado River and its main tributaries from Lee's Ferry to Diamond Creek, including its course in Grand Canyon National Park. Colorado River Research Series Contribution (Grand Canyon National Park, Grand Canyon, Ariz.) 37:1-88. Cooley, HI. E., J. W. Harshbarger, J. P. Akers, and W. F. Hardt. 1969. Regional hydrogeology of the Navajo and Hopi reservations, Arizona, New Mexico, and Utah. U.S. Geological Survey Professional Paper 521 -A:1-61. Cope, E. D., and H. C. Yarrow. 1875. Report upon the collections of fishes made in portions of Nevada, Utah, California, Colorado, New Mexico, and Arizona, during the years 1871, 1972, 1873, and 1874. Report on the Geographic and Geodetic Exploration and Survey West of the IOOtk Meridian (Wheeler Survey) 5(Zoology):635-703. Courtenay, W. R., Jr., and J. R. Stauffer, Jr. 1984. Distribution, Biology, and Management of Exotic Fishes. Johns Hopkins University Press, Baltimore. 430 pages. Deacon, J. E. 1968. Endangered non-game fishes of the west: Causes, prospects, and impor- tance. Proceedings of the Annual Conference, Western Association of Fish and Game Com- missioners 48 :534-549. Deacon, J. E. 1979. Endangered and threatened fishes of the west. Great Basin Naturalist Memoirs 3:41-64.

166 COLORADO RIVER ECOLOGY.^D DAM MANAGEMENT Deacon, J. E. 1988. The endangered woundfin and water management in the Virgin River, Utah, Arizona, Nevada. Fisheries (American Fisheries Society, Bethesda, Md.) 13:18-24. Deacon, J. E., and W. L. Minckley. 1974. Desert fishes. P. 385-488, in G. W. Brown, Jr., ea., Desert Biology, Volume II. Academic Press, New York. Deacon, J. E., and W. L. Minckley. 1991. Westem fishes and the real world: The enigma of "endangered species," revisited. In press, in W. L. Minckley and J. E. Deacon, eds., Battle Against Extinction: Native Fish Management in the American West. University of Arizona Press, Tucson. Dellenbaugh, F. S. 1984 (originally published 1908). A Canyon Voyage, the Narrative of the Second Powell Expedition down the Green-Colorado River from Wyoming, and the Explo- rations on Land, in the Years 1871 and 1872. University of Arizona Press, Tucson. 277 pages. Dill, W. A. 1944. The fishery of the lower Colorado River.:California Fish and Game 30:309- 401. Dibble, C. E., and C. C. Stout, eds. 1960. Ecological studies of the flora and fauna of Flaming Gorge Reservoir basin, Utah and Wyoming. University of Utah Anthropological Papers 48:1 -175. Dolan, R., A. Howard, and A. Gallenson. 1974. Man's impact on the Colorado River in Grand Canyon. American Scientist 62(4):392-401. Dolan, R., A. Howard, and D. Trimble. 1978. Structural control of the rapids and pools of the Colorado River in Grand Canyon. Science 202:629-631. Douglas, M. E., W. L. Minckley, and H. M. Tyus. 1989. Qualitative characters, identification of Colorado River chubs (Cyprinidae: genus Gila), and the "art of seeing well." Copeia 1989:653 -662. Douglas, P. A. 1952. Notes on the spawning of the humpback sucker, Xyrauchen texanus (Abbott). California Fish and Game 30:109-211. Eicher, G. J., Jr. 1947. Trout and the Colorado River. Arizona Wildlife Sportsman: 8(4):7-11. Euler, R. C. 1978. Archaeological and paleobiological studies at Stanton's Cave, Grand Can- yon National Park, Arizona A report of progress. P. 141-162, in National Geographic Society Research Report, 1978. Evermann, B. W., and C. Rutter. 1895. The fishes of the Colorado basin. Bulletin of the U.S. Fish Commissioner 14:473-486. Fisher, S. G. 1986, Structure and dynamics of desert streams. Pages 119-139, in W. G. Whitford, ea., Pattern and Process in Desert Ecosystems. University of New Mexico Press, Albu- querque. Fisher, S. G., L. J. Gray, N. B. Grimm, and D. E. Busch. 1982. Temporal succession in a desert stream. Ecological Monographs 52:93-118. Fradkin, P. L. 1981. A River No More: The Colorado River and the West. University of Arizona Press, Tucson. 360 p. Gilbert, C. H., and N. B. Scofield. 1898. Notes on a collection of fishes from the Colorado Basin in Arizona. Proceedings of the U.S. National Museum 20:487-499. Girard, C. 1857. Researches upon the cyprinid fishes inhabiting the fresh waters of the United States of America, west of the Mississippi Valley, from specimens in the Museum of the Smithsonian Institution. Proceedings of the Philadelphia Academy of Natural Science 8(1856):165- 213. Girard, C. 1859. Ichthyology of the boundary. P. 1-85, in Report of the United States and Mexican Boundary Survey, Made Under the Direction of the Secretary of the Interior, by W. H. Emory, Major, First Cavalry, and United States Commissioner 3. U.S. Govemment Printing Office, Washington, D.C. Graf, W. L. 1985. The Colorado River. Instability and Basin Management. Association of American Geographers, Research Publications in Geography 1985:1-86.

NATIVE FISHES... 167 Gray, L. J. 1981. Species composition and life histories of aquatic insects in a lowland Sonoran Desert stream. American Midland Naturalist 106:229-242. Gray, L. J., and S. G. Fisher. 1981. Postflood recolonization pathways of macroinvenebrates in a lowland Sonoran Desert stream. American Midland Naturalist 106:149-157. Grimm, N. B., and S. G. Fisher. 1989. Stability of periphyton and macroinvertebrates to disturbance by flash floods in a desert stream. Journal of the North American Benthological Society 8:293-307. Grimm, N. B., S. G. Fisher, and W. L. Minckley. 1981. Nitrogen and phosphorus dynamics in hot desert streams of southwestern U.S.A. Hydrobiologia 83:303-312. Gustaveson, A. W., H. R. Maddux, and B. L. Bonebrake. 1990. Assessment of a forage fish introduction into Lake Powell. Utah Department of Natural Resources, Division of Wildlife Resources, Salt Lake. 51 p. Hamman, R. L. 1981. Spawning and culture of Colorado squawfish in raceways. Progressive Fish-Culturist 43: 173-177. Hamman, R. L. 1982a. Culture of endangered Colorado River fishes. Section II. Induced spawning and culture of the humback chub. P. 158-167, in W. H. Miller, J. J. Valentine, D. L. Archer, H. M. Tyus, R. A. Valdez, and L. R. Kacding, eds., Colorado River Fisheries Project, Part 3. Contracted Studies. Final Report, U.S. Bureau of Reclamation Contract 9- 07-40-L-1016, and U.S. Bureau of Land Management Memorandum of Understanding CO- 910-MU9-933. U.S. Fish and Wildlife Service, Salt Lake City, Utah. Hamman, R. L. 1982b. Spawning and culture of humpback chub. Progressive Fish-Culturist 44:213-216. Hendricks, L. J. 1961. The striped mullet, Mugil cephalus Linnaeus. P. 93-94, in B. W. Walker, ea., The ecology of the Salton Sea, California, in relation to the sportf~shery. California Department of Fish and Game Fish Bulletin 113. Holden, P. B. 1968. Systematic studies of the genus Gila (Cyprinidae) of the Colorado River Basin. Unpublished Master's Thesis, Utah State University, Logan. 68 p. Holden, P. B. 1973. Distribution, abundance, and life history of the fishes of the upper Colorado River Basin. Unpublished Doctoral Dissertation, Utah State University, Logan. 59 p. Holden, P. B. 1979. Ecology of riverine fishes in regulated stream systems, with emphasis on the Colorado River. P. 57-74, in J. V. Ward and J. A. Stanford, eds., The Ecology of Regulated Streams. Plenum Press, New York. Holden, P. B. 1991. Ghosts of the Green River: Impacts of Green River poisoning on manage- ment of native Sshes. In press, in W. L. Minckley and J. E. Deacon, eds., Battle Against Extinction: Native Fish Management in the American West. University of Arizona Press, Tucson. Holden, P. B., and C. B. Stalnaker. 1970. Systematic studies of the cyprinid genus Gila in the upper Colorado River basin. Copeia 1970:409-429. Holden, P. B., and C. B. Stalnaker. 1975. Distribution and abundance of mainstream fishes of the middle and upper Colorado River basin, 1967-1973. Transactions of the American Fisheries Society 104:217-231. Holden, P. B., and E. J. Wick. 1982. Life history and prospects for recovery of Colorado squawfish. P. 98-108, in W. H. Miller, H. M. Tyus, and C. A. Carlson, eds., Fishes of the Upper Colorado River System: Present and Future. Westem Division, American Fisheries Society, Bethesda, Md. Howard, A. S., and R. Dolan. 1981. Geomorphology of the Colorado River in the Grand Canyon. Journal of Geology 89:269-298. Howard, C. S. 1947. Suspended sediment in the Colorado River. U.S. Geological Survey Water-Supply Paper 998:1-165. Hubbs, C. L. 1941. The relation of hydrological conditions to speciation in fishes. P. 182-195, in A Symposium on Hydrobiology. University of Wisconsin Press, Madison.

168 COLORADO RIVER ECOLOGY ACID DAM MANAGEMENT Hubbs, C. L. 1954. Establishment of a forage fish, the red shiner (Notropis lutrensis), in the lower Colorado River system. California Fish and Game 40:287-294. Hubbs, C. L., L. C. Hubbs, and R. C. Johnson. 1942. Hybridization in nature between species of catostomid fishes. Contributions of the University of Michigan Laboratory of Vertebrate Biology 22:1-76. Hubbs, C. L., and R.R. Miller. 1941. Studies of the fishes of the order Cyprinodontes. XVII. Genera and species of the Colorado River system. Occasional Papers of the University of Michigan Museum of Zoology 433:1-9. Hubbs, C. L., and R. R. Miller. 1948. The zoological evidence: Correlation between fish distribution and hydrographic history in the desert basins of western United States. Bulletin of the University of Utah 30:17-166. Hubbs, C. L., and R. R. Miller. 1953. Hybridization in nature between the fish genera Catostomus and Xyrauchen. Papers of the University of Michigan Museum of Zoology 38:207-233. Janisch, J. L. 1985. Evaluation of Lees Ferry fishery and future management. Arizona Game and Fish Department Publications 85-3:1-23. Johnson, J. E., and B. L. Jensen. 1991. Hatcheries for endangered freshwater fishes. In press, in W. L. Minckley and J. E. Deacon, eds., Battle Against Extinction: Native Fish Manage- ment in the American West. University of Arizona Press, Tucson. Johnson, J. E., R. R. Kramer, E. Larsen, and L. Bonebrake. 1987. Flaming Gorge fisheries investigations: Trout growth, survival, and microhabitat selection in the Green River, Utah. 1978-82. Utah Division Wildlife Publication 87-13:1-185. Johnson, R. R., and S. W. Carothers. 1987. External threats: The dilemma of resource manage- ment on the Colorado River in Grand Canyon National Park, USA. Environmental Manage- ment 11:99-107. Jonez, A., R. D. B eland, G. Duncan, and R. A. Wagner. 1951. Fisheries report of the lower Colorado River. Fisheries Task Group-Colorado River. Great Basin Field Commission. Arizona Game and Fish Commission., Phoenix. 89 p. Jonez, A., and R. C. Sumner. 1954. Lakes Mead and Mohave investigations, a comparative study of an established reservoir as related to a newly created impoundment. Final Report, Federal Aid Wildlife Restoration (Dingell-Johnson) Project F-1-R, Nevada Game and Fish Commission, Carson City. 186 p. Jordan, D. S. 1891. Repon of explorations in Utah and Colorado during the summer of 1889, with an account of the fishes found in each of the river basins examined. Bulletin of the U.S. Fish Commissioner 9:1-40. Jordan, D. S., and B. W. Evermann. 1896-1900. lathe fishes of North and Middle America. Bulletin of the U.S. National Museum 47(4 pans):1-3313. Joseph, T. W., J. A. Sinning, R. J. Behnke, and P. B. Holden. 1977. An evaluation of the status, life history, and habitat requirements of endangered and threatened fishes of the upper Colorado River system. U.S. Fish Wildlife Service FWS/OBS-77062:1-169. Kaeding, L. R., B. D. Burdick, P. A. Schrader, and W. R. Noonan. 1986. Recent capture of a bonytail (Gila elegans) and observations on this nearly extinct cyprinid from the Colorado River. Copeia 1986:1021-1023. Kaeding, L. R., and D. B. Osmundson. 1988. Interaction of slow growth and increased early- life mortality: A hypothesis on the decline of Colorado squawfish in the upstream regions of its historic range. Environmental Biology of Fishes 22:287-298. Kaeding, L. R., and M. A. Zimmerman. 1982. Life history and ecology of the humpback chub in the Little Colorado and Colorado rivers of the Grand Canyon, Arizona. P. 281-320, in W. H. Miller, J. J. Valentine, D. L. Archer, H. M. Tyus, R. A. Valdez, and L. R. Kaeding, eds., Colorado River Fishery Project, Part II. Field Investigations. Final Report, U.S. Bureau of Reclamation Contract 9-07-40-L-1016, and U.S. Bureau of Land Management

NATIVE FISHES... 169 Memorandum of Understanding CO-910-MU9-933. U.S. Fish and Wildlife Service, Salt Lake City, Utah. Kaeding, L. R., and M. A. Zimmerman. 1983. Life history and ecology of the humpback chub in the Little Colorado and Colorado rivers of Grand Canyon. Transactions of the American Fisheries Society 112:577-594. Karp, C.A., and H. M. Tyus. 1990. Behavioral interactions between young Colorado squawfish and six fish species. Cope u: 1990:25-34. Kimsey, J. B. 1958. Fisheries problems in impounded waters of Califomia and the lower Colorado River. Transactions of the American Fisheries Society 87:319-332. Kimsey, J. B., R. H. Hagy, and G. W. McCammon. 1957. Progress report on the Mississippi threadfin shad, Dorosorna petenense atchafaylae (sic.) in the Colorado River for 1956. California Department of Fish and Game Inland Fisheries Administrative Report 57-23:1- 48. Kirsch, P. H. 1889. Notes on a collection of fishes obtained in the Gila River at Fort Thomas, Arizona. Proceedings of the U.S. National Museum 11:555-558. Kolb, E. L. 1989 (originally published 1914). Through the Grand Canyon from Wyonung to Mexico. University of Arizona Press, Tucson. 344 p. Kolb, E. L., and E. Kolb. 1914. Experience in the Grand Canyon. National Geographic Maga- zine 26(2):99-184. Kondolf, G. M., S. S. Cook, H. R. Maddux, and W. R. Persons. 1989. Spawning gravels of rainbow trout in Glen and Grand Canyons, Arizona. Journal of the Arizona-Nevada Acad- emy of Science 23:19-28. Kubly, D. M. 1990. The endangered humpback chub (Gila cyp).a) in Arizona: A review of past studies and suggestions for future research. Repon, U.S. Bureau of Reclamation, Salt Lake City. Arizona Game and Fish Department, Phoenix. 138 p. (draft, February 1990). Kubly, D. M., and G. A. Cole. 1979. The chemistry of the Colorado River and its tributaries in Marble and Grand canyons. P. 565-572, in Proceedings of the First Annual Conference on Scientific Research in the National Parks. U.S. National Park Service Transactions and Proceedings Series 5. LaRivers, I. 1962. Fish and Fisheries of Nevada. Nevada State Printing Office, Carson City. 782 p. Leopold, L. B. 1969. The rapids and pools—Grand Canyon. U.S. Geological Survey Profes- sional Paper 669d:131-145. Maddux, H. R., and W. G. Kepner. 1988. Spawning of bluehead sucker in Kanab Creek, Arizona (Pisces: Catostomidae). The Southwestern Naturalist 33:364-365. Maddux, H. R., D. M. Kubly, J. C. DeVos, Jr., W. M. Persons, R. Stacdicke, and R. L. Wright. 1987. Effects of varied flow regimes on aquatic resource sof Glen and Grand canyons. Final Report, U.S. Bureau of Reclamation Contract 4-AG-40-01810. Arizona Game and Fish Department, Phoenix. 291 p. Marsh, P. C. 1985. Effects of incubation temperature on survival of embryos of native Colo- rado River fishes. The Southwestern Naturalist 30:129-140. Marsh, P. C. 1987. Food of adult razorback sucker in Lake Mohave, Arizona-Nevada. Transac- tions of the American Fisheries Society 116:1 17-119. Marsh, P. C., F. J. Abarca, M. E. Douglas, and W. L. Minckley. 1989. Spikedace (Meda fulgida) and loach minnow (Tiaroga cobitis) relative to introduced red shiner (Cyprinella lutrensis). Final Report Arizona Game and Fish Department Contract. Arizona State Uni- versity, Tempe. 116 p. McAda, C. W. 1977. Aspects of the life history of three catostomids native to the upper Colorado River basin. Unpublished Master's Thesis, Utah State University, Logan. 73 p. McAda, C. W., and R. W. Wydoski. 1980. The razorback sucker, Xyrauchen texanus, in the

170 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT upper Colorado River basin, 1974-1976. U.S. Fish Wildlife Service Technical Papers 99:1 -15. McAda, C. W., and R. W. Wydoski. 1983. Maturity and fecundity of the bluehead sucker, Catostomus discobolus (Catostomidae), in the upper Colorado River basin, 1975-76. The Southwestern Naturalist 28:120-123. McAda, C. W., and R. W. Wydoski. 1985. Growth and reproduction of the flannelmouth sucker, Catostomus latipinnis, in the upper Colorado River basin, 1975-76. Great Basin Natural ist 45 :281-286. McCarthy, M. S., and W. L. Minckley. 1987. Age estimation for razorback sucker (Pisces: Catostomidae) from Lake Mohave, Arizona and Nevada. Journal of the Arizona-Nevada Academy of Science 21:87-97. McDonald, D. B., and P. A. Dotson. 1960. Fishery investigations of the Glen Canyon and Flaming Gorge impoundment areas. Utah State Department of Fish and Game Information Bulletin 60-3: 1-70. McKee, E. D. 1930. Briefs. Grand Canyon Nature Notes 4(4):37. Mcaseles, E. B. 1981. A Crossing on the Colorado, Lees Ferry. Pruett Publishing Co., Boulder, Colorado. 13 8 p. Meffe, G. K. 1983. Ecology of species replacement in the Sonoran topminnow (Poeciliopsis occidentalis) and the mosquitofish (Gambusia affinis) in Arizona. Unpublished Doctoral Dissertation, Arizona State University, Tempe. 143 p. Mcffe, G. K. 1984. Effects of abiotic disturbance on coexistence of predatory-prey fish spe- cies. Ecology 65:1525-1534. Mcffe, G. K. 1985. Predation and species replacement in Southwestem fishes: A case study. The Southwestern Naturalist 30:173-187. Meffe, G. K., and W. L. Minckley. 1987. Persistence and stability of fish and invertebrate assemblages in a repeatedly disturbed Sonoran Desert stream. American Midland Naturalist 1 17:177-191. Miller, R. R. 1943. The status of Cyrinodon macularius and Cyprinodon nevadensis, two desert fishes of western Nonh America. Occasional Papers of the University of Michigan Museum of Zoology 473:1-25. Miller, R. R. 1946a. Gila cyp ha, a remarkable new species of fish from the Colorado River in Grand Canyon, Arizona. Journal of the Washington Academy of Science 35:403-415. Miller, R. R. 1946b. The need for ichthyological surveys of the major rivers of western North America. Science 104:517-519. Miller, R. R. 1952. Bait fishes of the lower Colorado River, from Lalce Mead, Nevada, to Yuma, Arizona, with a key for their identification. California Fish and Game 38:7-42. Miller, R. R. 1955. Fish remains from archaeological sites in the lower Colorado River basin, Arizona. Papers of the Michigan Academy Science, Arts, and Letters 40:125-136. Miller, R. R. 1959. Origin and affinities of the freshwater fish fauna of western North America. Pages 187-222, in C.L. Hubbs, ea., Zoogeography. American Association for the Advance- ment of Science Publication 51(1958), Washington, D.C. Miller, R. R. 1961. Man and the changing fish fauna of the American Southwest. Papers of the Michigan Academy of Science, Arts, and Letters 46:365-404. Miller, R. R. 1963a. Is our native underwater life worth saving? Nationa Parks Magazine 37 :4-9. Miller, R. R. 1963b. Distribution, variation and ecology of Lepidomeda vittata, a rare cyprinid fish endemic to eastem Arizona. Copeia 1963:1-5. Miller, R. R. 1964. Fishes of Dinosaur. Naturalist 15:24-29. Miller, R. R. 1968. Unpublished field notes: 1968 Arizona collecting expedition. On file in the Fish Division, University of Michigan Museum of Zoology, Ann Arbor, Mich. Miller, R. R., and C. L. Hubbs. 1960. The spiny-rayed cyprinid fishes (Plagopterini) of the

NATIVE FISIlES... 171 Colorado River system. Miscellaneous Publications of the University of Michigan Museum of Zoology 1 1 5:1 -39. Miller, R. R., and G. R. Smith. 1984. Fish remains from Stanton's Cave, Grand Canyon of the Colorado, Arizona, with notes on the taxonomy of Gila cypha. P. 61-65, in R. C. Euler, ea., The Archaeology, Geology, and Paleabiology of Stanton's Cave, Grand Canyon National Park, Arizona. Grand Canyon Natural History Association Monograph 6. Miller, R. R., and P. W. Webb. 1986. Tactics used by Rhinichthys and some other fresh water fishes in response to strong currents. Abstracts of the Sixty-fourth Annual Meeting, Amen- can Society of Ichthyologists and Herpetologists, Victoria, British Columbia. Miller, W. H., H. M. Tyus, and C. A. Carlson, eds. 1982a. Fishes of the Upper Colorado River System: Present and Future. Western Division, American Fisheries Society, Bethesda, Md. 131 p. Miller, W. H., J. J. Valentine, D. L. Archer, H. M. Tyus, R. A. Valdez, and L. R. Kacding, eds. 1982b. Colorado River Fisheries Project, Part 1. Summary Report. Final Report, U.S. Bu- reau of Reclamation Contract 9-070-L-1016, and U.S. Bureau of Land Management Memo- randum of Understanding CO-910-MU9-933. U.S. and Fish Wildlife Service, Salt Lake City. 42 p. Miller, W. H., J. J. Valentine, D. L. Archer, H. M. Tyus, R. A. Valdez, and L. R. Kaeding, eds. 1982c. Ibid. Part 2. Field investigations. Ibid. 365 p. Miller, W. H., J. J. Valentine, D. L. Archer, H. M. Tyus, R. A. Valdez, and L. R. Kacding, eds. 1982d. Ibid. Part 3. Contracted Studies. Ibid. 324 p. Minckley, C. O. 1978. A report on aquatic investigations conducted during 1976, 1977 on Bright Angel, Phantom and Pipe creeks, Grand Canyon National Park, Coconino County, Arizona. Annual Investigation Report, Grand Canyon National Park, Grand Canyon, Ariz. Northem Arizona University, Flagstaff. 112 p. Minckley, C. O. 1980. River resource monitoring project, Gila spp. studies, Grand Canyon National Park: Part 2. Repon National Park Service Contract CX-82-1070012. Museum of Northem Anzona, Flagstaff. 23 p. Minckley, C. O. 1988. Final report on research conducted on the Little Colorado River popula- tion of the humpback chub, during May 1987, 1988. Repon to Non-game Branch, Arizona Game and Fish Department, Phoenix. Nonhem Arizona University, Flagstaff. 131 p. Minckley, C. O. 1989. Final report on research conducted on the Little Colorado River popula- tion of the humpback chub during May 1989. Ibid. 36 p. Minckley, C. O., and D. W. Blinn, 1976. Summer distribution and reproductive status of fish of the Colorado River in Grand Canyon National Park and vicinity, 1975. Final Report National Park Service Contract CX-82-1060008. Northern Arizona University, Flagstaff. 17 p. Minckley, C. O., and S. W. Carothers. 1980. Recent collections of the Colorado squawfish and razorback sucker from the San Juan and Colorado rivers in New Mexico and Arizona. The Southwestern Naturalist 24:686-687. Minckley, C. O., S. W. Carothers, J. W. Jordan, and H. D. Usher. 1981. Observations on the humpback chub, Gila cyp ha, within the Colorado and Little Colorado rivers, Grand Canyon National Park, Arizona. P. 176-183, in Proceedings of the Second Annual Conference on Scientific Research in the National Parks. U.S. National Park Service Transactions and Proceedings Series 8. Minckley, W. L. 1965. Native fishes as natural resources. P. 48-60, in J. L. Gardner, ea., Native Plants and Animals as Resources in Arid Lands of the southwestern United States. Southwest-Rocky Mountain Division, American Association for the Advancement of Sci- ence, Contribution 8, Committee on Desert and Arid Lands Research. Minckley, W. L. 1973. Fishes of Arizona. Arizona Game and Fish Department, Phoenix. 293 p. Minckley, W. L. 1979. Aquatic habitats and fishes of the lower Colorado River, southwestern

172 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT United States. Final Report U.S. Bureau of Reclamation Contract 14-06-300-2529. Lower Colorado River Region, Boulder City, Nevada. Anzona State University, Tempe. 478 p. Minckley, W. L. 1980. Morphological variation in catostomid fishes of the Grand Canyon Region, middle Colorado River basin. Final Report U.S. National Park Service Contract, Grand Canyon National Park, Grand Canyon, Arizona. Arizona State University, Tempe. 27 p. Minckley, W. L. 1981. Ecological studies of Aravaipa Creek and environs, central Arizona, relative to past, present, and future uses. Final Report, U.S. Bureau of Land Management Contract YA-512<T6-98. Anzona State University, Tempe. 362 p. Minckley, W. L. 1982. Food relations of introduced fishes of the lower Colorado River, An- zona-Califomia. California Fish and Game 68:165-187. Minckley, W. L. 1983. Status of the razorback sucker, Xyrauchen texanus (Abbott) in the lower Colorado River basin. The Southwestern Naturalist 28:165-187. Minckley, W. L. 1985. Native fishes and natural aquatic habitats in U.S. Fish and Wildlife Service Region II, west of the Continental Divide. Final Report, Interagency Personnel Act Agreement, Region 2, U.S. Fish Wildlife Service, Albuquerque, N.M. Arizona State Uni- versity, Tempe. 158 p. Minckley, W. L. 1989. Aging of Colorado squawfish/razorback sucker by otolith examination. Final Report for U.S. Fish Wildlife Service Purchase Order 20181-87-0026, Albuquerque, N.M. Anzona State University, Tempe. 9 p. Minckley, W. L., D. G. Buth, and R. L. Mayden. 1989. Origin of brood stock and allozyme variation in hatchery-reared bonytail, an endangerd North American cyprinid fish. Transac- tions of the American Fisheries Society 118:139-145. Minckley, W. L., and J. E. Deacon. 1968. Southwestem fishes and the enigma of "endangered species." Science 159:1424-1432. Minckley, W. L., and J. E. Deacon, eds. 1991. In press, Battle Against Extinction: Native Fish Management in the American West. University of Anzona Press, Tucson. Minckley, W. L., and M. E. Douglas. 1991. Discovery and extinction of western fishes: A blink of the eye in geologic time. In press, in W. L. Minckley and J. E. Deacon, eds., Battle Against Extinction: Native Fish Management in the American West. University of Arizona Press, Tucson. Minckley, W. L., and E. S. Gustafson. 1982. Early development of the razorback sucker, Xyrauchen tenants (Abbott). Great Basin Naturalist 42:533-561. Minckley, W. L., D. A. Hendrickson, and C. E. Bond. 1986. Geography of western North American freshwater fishes: Descriptions and relations to intracontinental tectonism. P. 519-613, In C. H. Hocutt and E. O. Wiley, eds., Zoogeography of North American Fresh- water Fishes. John Wiley and Sons, New York. Minckley, W. L., P. C. Marsh, J. E. Brooks, J. E. Johnson, and B. L. Jensen. 1991. Manage- ment toward recovery of the razorback sucker. In press, in W. L. Minckley and J. E. Deacon, eds., Battle Against Extinction: Native Fish Management in the American West. University of Arizona Press, Tucson. Minckley, W. L., and G. K. Meffe. 1987. Differential selection for native fishes by flooding in streams of the arid American Southwest. P. 93-104, in W. J. Matthews and D. C. Heins, eds., Ecology and Evolution of North American Stream Fish Communities. University of Oklahoma Press, Norman. Minckley, W. L., and J. N. Rinne. 1985. Large organic debris in southwestern streams: An historical review. Desert Plants 7:142-153. Moffett, J. W. 1942. A fishery survey of the Colorado River below Boulder Dam. California Fish and Game 28:76-86. Moffett, J. W. 1943. A preliminary report on the fishery of Lake Mead. Transactions of the North American Wildlife Conference 8:179-186.

NATIVE FISHES... 173 Montgomery, W. L, W. Liebfried, and K. Gooby. 1986. Feeding by rainbow trout on Cladophora glomerata at Lee's Ferry, Colorado River, Arizona; the roles of Cladophora and epiphytic diatoms in trout nutrition. Final Report, U.S. Bureau of Reclamation Contract, Lower Colo- rado River Region, Boulder City, Nev. Northem Arizona University, Flagstaff, et al. 20 p. Moyle, P. B. 1976. Inland Fishes of California. University of California Press, Berkeley. 405 p. Mueller, G. 1989. Obsecvations of spawning razorback sucker (Xyrauchen texanus) utilizing river habitat in the lower Colorado River, Arizona-Nevada. The Southwestern Naturalist 34:147-149. National Research Council. 1987. River and Dam Management: A Review of the Bureau of Reclamation's Glen Canyon Environmental Studies. National Academy Press, Washington, D.C. 203 p. Nicola, S. J. 1979. Fisheries problems associated with the development of the lower Colorado River. California Department of Fish and Game, Inland Fisheries Endangered Species Program Special Publication 79-4:1-18. Ohmart, R. D., B. W. Anderson, and W. C. Hunter. 1988. The ecology of the lower Colorado River from Davis Dam to the Mexico-United States International Boundary: A community profile. U.S. Fish and Wildlife Service Biological Report 85(7.19):1-243, 2 appendices. Ono, R. D., J. D. Williams, and A. Wagner. 1983. Vanishing Fishes of North America. Stone Wall Press, Washington, D.C. 257 p. Pendergast, D. M., and C. C. Stout, editors. 1961. Ecological studies of the flora and fauna of Navajo Reservoir basin, Colorado and New Mexico, University of Utah Anthropological Papers 55:1-122. Persons, W. B., K. McCormack, and T. McCall. 1985. Fishery investigations of the Colorado River from Glen Canyon Dam to the confluence of the Pana River: Assessment of the impact of fluctuating flows on the Lee's Ferry fishery. Arizona Garne and Fish Department Publication 86-6:1-93. Pelts, G. E. 1984. Impounded Rivers Perspectives for Ecological Management. John Wiley and Sons, New York. 326 p. Pillsbury, A. F. 1981. The salinity of rivers. Scientific American 245:55-65. Pirnental, R., R. V. Bulkley, and H. M. Tyus. 1985. Choking of Colorado squawfish, Ptychocheilus lucius (Cyprinidae), on channel catfish, Ictalurus punctatus (Ictaluridae), as a cause of mortality. The Southwestern Naturalist 30:154-158. Platania, S. P., K. R. Bestgen, M. A. Morretti, D. L. Propst, and J. E. Brooks. In press. Status of Colorado squswfish and razorback sucker in the San Juan River, Colorado, New Mexico, and Utah, The Southwestern Naturalist 36. Powell, J. W. 1875, Exploration of the Colorado River of the West and Its Tributaries, Ex- plored in 1869, 1870, 1871, and 1872. U.S. Government Printing Office, Washington, D.C. Reisner, M. 1986. Cadillac Desert: The American West and Its Disappearing Water. Viking Press, New York. 582 p. Richards, M. T., and D. B. Wood. 1985. The economic value of sportfishing at Lees Ferry, Arizona. U.S. Department of Agriculture. Forest Service General Technical Report RM- 120:219-222. Rinne, J. N., J. E. Johnson, B. L. Jensen, A. W. Ruger, and R. Sorenson. 1986. The role of hatcheries in the management and recovery of threatened and endangered fishes. Pages 271-285, in R. H. Stroud, ea., Fish Culture in Fisheries Managment. American Fisheries Society, Bethesda, Md. Rinne, J. N., and P. R. Turner. 1991. Reclamation and alteration as management techniques, and a review of methodology in stream renovation. In press, in W.L. Minckley and J. E. Deacon, eds., Battle Against Extinction: Native Fish Management in the American West. University of Arizona Press, Tucson.

174 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT Scoppettone, G. G. 1988. Growth and longevity of the cui-ui (Chasmistes cujus) and suspected longevity in other catostomids and cyprinids. Transactions of the American Fisheries Soci- ety 117:301-307. Seethaler, K. H. 1978. Life history and ecology of the Colorado squawfish (Ptychocheilus lucius) in the Upper Colorado River Basin. Unpublished Master's Thesis, Utah State Uni- versity, Logan. 156 p. Siebert, D. J. 1980. Movements of fishes in Aravaipa Creek, Arizona. Unpublished NIaster's Thesis, Arizona State University, Tempe. 56 p. Sigler, W. F., and R. R. Miller. 1963. Fishes of Utah. Utah Department of Fish and Game, Salt Lake City. 203 p. Sigler, W. F., and J. W. Sigler. 1987. Fishes of the Great Basin: A Natural History. University of Nevada Press, Reno. 425 p. Smith, D. L., and C. G. Crampton, eds. 1987. The Colorado River Survey: Robert B. Stanton and the Denver, Colorado Canyon & Pacific Railroad. Howe Brothers Publishers, Salt Lake City. 305 p. Smith, G. R. 1959. Annotated checklist of fishes of Glen Canyon. P. 195-199, in A. M. Woodbury, ea., Ecological Studies of the Flora and Fauna in Glen Canyon. University of Utah Anthropological Papers 40. Smith, G. R. 1966. Distribution and evolution of the North American catostomid fishes of the subgenus Pantosteus, genus Catostomus. Miscellaneous Publications of the University of Michigan Museum of Zoology 29:1-132. Smith, G. R. 1981. Effects of habitat size on species richness and adult body size of desert fishes. P. 125-172, in R. J. Naiman and D. L. Soltz, eds., Fishes in North American Deserts. John Wiley and Sons, New York. Smith, G. R., G. G. Musser, and D. B. McDonald. 1959. Appendi~c A. Aquatic survey tabula- tion. P. 177-194, in A. hI. Woodbury, ea., Ecological studies of the flora and fauna in Glen Canyon. University of Utah Anthropological Papers 40. Smith, R., R. R. Miller, and W. D. Sable. 1979. Species relationships among fishes of the genus Gila of the upper Colorado River Basin. P. 613-623, in Proceedings of the First Annual Conference on Scientific Research in the National Parks. U.S. National Park Ser- vice Transactions and Proceedings Series 5. Smith, M. L 1981. Late Cenozoic fishes ~n the warm deserts of Nonh America: A reinterpretation of desen adaptations. P. 11-38, in R. J. Naiman and D. L. Soltz, eds., Fishes in Nor~h American Deserts. John Wiley and Sons, New York. Snyder, J. O. 1915. Notes on a collection of fishes made by Dr. Edgar A. Mearns from rivers tributa~y to the Gulf of Califomia. Proceedings of the U.S. National Museum 49: 573-586. Spofford, W. O., A. L. Parker, and A. V. Kneese, eds. 1980. Energy Development in the Southwest~'roblems of Water, Fish and Wildlife in the Upper Colorado River Basin, 2 Volumes. Resources for the Future, Washington, D.C. Stanford, J. A., and J. V. Ward. 1986a. The Colorado River system. P. 353-374, in B. R. Davies and K. F. Walker, eds., The Ecology of River Systems. Dr. SV. Junk, Dordrecht, The Netherlands. Stanford, J. A., and J. V. Ward. 1986b. Reservoirs of the Colorado system. P. 375-383, Ibid. Stanford, J. A., and J. V. Ward. 1986c. Fishes of the Colorado system. P. 385-402, Ibid. Stegner, W. 1954. Beyond the Hundredth Meridian~ohn Wesley Powell and the Second Opening of the West. Houghton-MifRin Company, Boston. 438 p. Stone, J. L. 1965. Tailwater fisheries investigations-creel census of the Colorado River below Glen Canyon Dam, July 1, 1964-June 30, 1965. Arizona Game and Fish Department, Colorado River Storage Project (Public Law 485, Section 8) Report. 13 p. Stone, J. L. 1966. Ibid. July 1, 1965-June 30, 1966. Ibid. 13 p. Stone, J. L. 1967. Ibid. July 1, 1966-June 30, 1967. Ibid. 33 p.

NATIVE FISIlES... 175 Stone, J. L. 1968, Ibid. July 1, 1967-June 30, 1968. Ibid. 35 p. Stone, J. L. 1969. Ibid. July 1, 1968-June 30, 1969. Ibid. 47 p. Stone, J. L. 1971. Ibid. July 1, 1970-June 30, 1971. Ibid. 47 p. Stone, J. L. 1972. Ibid. July 1, 1971 -June 30, 1972. Ibid. 23 p. Stone, J. L., and N. L. Rathbun. 1967. Tailwater fisheries investigations, creel census and limnological study of the Colorado River below Glen Canyon Dam, July 1, 1966-Junc 30, 1967. Arizona Game and Fish Department, Phoenix. 54 p. Stone, J. L., and N. L. Rathbun. 1968. Ibid. July 1, 1967-June 30, 1968. Ibid. 56 p. Stone, J. L., and N. L. Rathbun. 1969. Ibid. July 1, 1968-June 30, 1969. Ibid. 60 p. Suttkus, R. D., and G. H. Clemmer. 1977. The humpback chub, Cila cypha, in the Grand Canyon area of the Colorado River, Occasional Papers of the Tulane University Museum of Natural History 1: 1-30. Suttkus, R. D., and G. H. Clemmer. 1979. Fishes of the Colorado River in Grand Canyon National Park. Pages 599-604, in Proceedings of the First Annual Conference on Scientific Research in the National Parks. U.S. National Park Service Transactions and Proceedings Series 5. Suttkus, R. D., G. H. Clcmmer, C. Jones, and C. R. Shoop, 1976. Survey of fishes, mammals and herpetofauna of the Colorado River in Grand Canyon. Colorado River Research Series Contribution (Grand Canyon National Park, Grand Canyon, Ariz.) 34:1-48. Swales. S. 1989. The use of instream habitat improvement methodology in mitigating the adverse effects of river regulation on fisheries. P. 185-208, in J. A. Gore and G. E. I'ctts, eds., Alternatives in Regulated River Management. Chemical Rubber Company Press, Boca Raton, Fla. Sykes, G. 1937. The Colorado Delta. Publications of the Carnegie Institution of Washington 460. 193 p. Taubert, B. Do 1985. Changes in the trout fisheries of the lower Colorado Roar in Arizona 1'. 175-180, in F. Richardson and R.H. Hamre, tech. coords., Wild Trout Symposium, III. U.S. Government Printing Office, Washington, D.C. Tyus, H. M. 1985. Homing behavior noted for Colorado squawfish. Copeia 1985:213-215. Tyus, H. M. 1986. Life strategies in the evolution of the Colorado squawfish (Ptychocheil~s lucius). Great Basin Naturalist 46:656-661. Tyus, H. M. 1987. Distribution, reproduction and habitat use of the razorback sucker in the Green River, Utah, 1979-86. Transactions of the American Fisheries Society 116: 111-116. Tyus, H. M. 1988. Long-term retention of implanted transmitters in Colorado squawfish and razorback suckers. North American Journal of Fisheries Management 8:264-267. Tyus, H. M. 1991. Management of the Colorado squawfish. In press, in W. L. NIincklcy and J. E. Deacon, eds., Battle Against Extinction: Native Fish Management in the American West. University of Arizona Press, Tucson. Tyus, H. M., R. L. Jones, and L. M. Tnnca. 1987. Green River rare and cndangcrcd fish studies, 1982-1985. Final Report, Colorado River Fishes Monitoring Project. U.S. Fish and Wildlife Service, Vernal, Utah. 117 p. Tyus, H. M., and C. A. Karp. 1989. Habitat use and streamflow needs of rare and endangered fishes, Yampa River, Colorado. U.S. Fish Wildlife Service Biological Report 89(14):1-27. Tyus, H. M., and C. W. McAda. 1984. Migration, movements, and habitat prcfcrcnccs of Colorado squawfish, Ptychocheilus Lucius, in the Green, White, and Yampa rivers, Colo- rado and Utah. The Southwestern Naturalist 29:289-299. Tyus, H. M., and W. L. Minckley. 1988. Migrating Mormon crickets, Anabrus simplex (Or- thoptera: Tettigoniidae), as food for stream fishes. Great Basin Naturalist 48:24-30. ., . , .. r ~ ---a ~~ Glen U.S. U.S. Fish and Wildlife Service. 1978. Memorandum (Biological opinion of the effects of Canyon Dam on the Colorado River as it affects endangered species) to Regional Director, Bureau of Reclamation, Salt Lake City. U.S. Fish and Wildlife Service, Albuquerque. 7 p.

176 COLORADO RIVER ECOLOGY AND DAM MANAGEMENT U.S. Fish and Wildlife Service. 1980. Aquatic study Colorado River from Lee's Ferry to the southern International Boundary, and selected tributaries, Arizona, Califomia, Nevada. Special report on distribution and abundance of fishes of the lower Colorado River. Final Report, U.S. Water and Power Resources Service (Bureau of Reclamation) Contract 97-03-X0066, Lower Colorado River Region, Boulder City, Nevada. U.S. Fish Wildlife Service, Phoenix. 157p. U.S. Fish and Wildlife Service. 1981. Aquatic study of the lower Colorado River. Ibid. 278 p. U.S. Fish and Wildlife Service. 1987. Final recovery implementation program for endan- gered fish species in the upper Colorado River basin. U.S. Fish and Wildlife Service, Denver. 74 p. U.S. Fish and Wildlife Service. 1989a. Title Wildlife and Fisheries. Part 17 Endangered and threatened wildlife and plants. Subpart B lists. Endangered and threatened wildlife (CFR 17.11 and 17.12). U.S. Government Printing Office, Washington, D.C. U.S. Fish and Wildlife Service. 1989b. Humpback chub, Gila cypha, recovery plan (technical review draft). Region 6, U.S. Fish and Wildlife Service, Denver. 56 p. U.S. Fish and Wildlife Service. 1989c. Bonytall chub, Gila elegans, recovery plan (technical review draft). Ibid. 49 p. U.S. Fish and Wildlife Service. 1989d. Colorado squawfish Ptychocheilus lucius, recovery plan (technical review draft). Ibid. 78 p. U.S. Fish and Wildlife Service. 1990. Endangered and threatened wildlife and plants; Proposal to determine the razorback sucker ~yrauchen texanus) to be an endangered species. Fed- eral Register 55:21154-21161. Usher, H. D., S. W. Carothers, C. O. Minckley, and J. W. Jordan. 1980. Age and growth rate of the flannelmouth sucker, Catostomus latipinnis and bluehead mountain sucker, Pantosteus discobolus, in the Colorado River, Grand Canyon National Parlc. P. 163-175, in Proceed- ings of the Second Annual Conference on Scientific Research in the National Parks. U.S. National Park Service Transactions and Proceedings Series 8. U.S. National Park Service. 1977. Grand Canyon National Park, Arizona. Natural Resource Management Plan and Environmental Assessment. Grand Canyon National Park, Grand Canyon, Arizona 127 p. Valdez, R. A. 1990. The endangered fish of Cataract Canyon. Report U.S. Bureau of Reclama- tion Contract 6-CS-40-03980. BID/WEST, Incorporated, Logan, Utah. 94 p., 8 appendices. Valdez, R. A., and G. C. Clemmer. 1982. Life history and prospects for recovery of the humpback and bonytail chubs. P. 109-119, in W. H. Miller, H. M. Tyus, and C. A. Carlson, eds., Fishes of the upper Colorado River system: Present and future. Westen~ Division, American Fisheries Society, Bethesda, Md. Valdez, R. A., P. G. Magnan, R. Smith, and B. Nilson. 1982a. Upper Colorado River investiga- tions (Rifle, Colorado, to Lake Powell, Utah), P. 101-279, in W. H. Miller, J. J. Valentine, D. L. Archer, H. M. Tyus, R. A. Valdez, and L. R. Kaeding, eds., Colorado River Fisheries Project, Part 3. Contracted Studies. Final Report, U.S. Bureau of Reclamation Contract 9- 07-40-L-1016, and U.S. Bureau of Land Management Memorandum of Understanding CO- 910-MU9-933. U.S. Fish and Wildlife Service, Salt Lake City, Utah. Valdez, R. A., P. G. Maqnan, M. McInery, and R. P. Smith. 1982b. Tributary report: Fishery investigations of the Gunnison and Dolores rivers. P. 321-362, Ibid. Vanicek, C. D. 1967. Ecological studies of native Green River fishes below Flaming Gorge Dam, 1964-1966. Unpublished Doctoral Dissertation. Utah State University, Logan, Utah. 124p. Vanicek, C. D., and R. H. Kramer. 1969. Life history of the Colorado squawfish, Ptychocheilus lucius, and the Colorado chub, Gila robusta, in the Green River in Dinosaur National Monument. 1964-1966. Transactions of the American Fisheries Society 98:193-208. Vanicek, C. D., R. H. Kramer, and D. R. Franklin. 1970. Distribution of Green River fishes in

NATIVE FISHES... 177 Utah and Colorado following closure of Flaming Gorge Dam. The Southwestern Naturalist 14:297-315. Wallen, I. E. 1951. The direct effect of turbidity on fishes. Bulletin of the Oklahoma Agricul- tural and Mechanical College 48:1-27. Wallis, O. L. 1951. The status of the fish fauna of the Lake Mead National Recreation Area, Arizona-Nevada. Transactions of the American Fisheries Society 80:84-92. Welcomme, R. L. 1989. Floodplain fisheries management. P. 209-233, in J. A. Gore and G. E. Pelts, eds., Alternatives in Regulated River Management. Chemical Rubber Company Press, Boca Raton, Fla. Weatherford, G. D., and F. L. Brown, eds. 1986. New Courses for the Colorado River: Major Issues for the Next Century. University of New Mexico Press, Albuquerque. 253 p. Williams, C. D., and J. E. Deacon. 1991. Ethics, federal legislation,, and litigation in the battle against extinction. In press, in W. L. Minckley and J. E. Deacon, eds., Battle Against Extinction: Native Fish Management in the American West. University of Arizona Press, Tucson. Williams, J. E., D. B. Bowman, J. E. Brooks, A. A. Echelle, R. J. Edwards, D. A. Hendrickson, and J. J. Landye. 1985. Endangered aquatic ecosystems in North American deserts with a list of vanishing fishes of the region. Journal of the Arizona-Nevada Academy of Science 20:1 -62. Williamson, R. R., and C. F. Tyler. 1932. Trout propagation in Grand Canyon National Park. Grand Canyon Nature Notes 7(2):11-16. Wiltzius, W. J. 1976. Some historic influences of resewairs and irrigation diversions on flows, temperatures and fish distributions in the Gunnison River. Colorado Division Wildlife Resources, Fort Collins. 100 p. Winn, H. E., and R. R. Miller. 1954. Native postlarval fishes of the lower Colorado River basin, with a key to their identification. California Fish and Game 40:273-285. Woodbury, A. M., ed. 1959. Ecological studies of the flora and fauna in Glen Canyon. Univer- sity of Utah Anthropological Papers 40:1-229. Woodbury, A. M., ed. 1963. Studies of biota in Dinosaur National Monument, Utah and Colorado. University of Utah, Division of Biological Science Miscellaneous Papers 1: 1-77. Worster, D. 1985. Rivers of Empire: Water, Aridity, and Growth of the American West. Pan- theon Press, New York. 402 p. Wydoski, R. S., K. Gilbert, K. H. Seethaler, C. W. McAda, and J. A. Wydoski. 1980. Anno- tated bibliography for aquatic resource management of the upper Colorado River ecosys- tem. U.S. Fish and Wildlife Service Research Publication 135:1-186. Wydoski, R. S., and J. Hamill. 1991. Evolution of a cooperative recovery program for endan- gered fishes in the upper Colorado River basin. In press, in W. L. Minckley and J. E. Deacon, eds., Battle Against Extinction: Native Fish Management in the American West. University of Arizona Press, Tucson.