INTRODUCED FISHES OF THE LOWER COLORADO 87
<br />.had, but black crappie ate a1.1981 ),with fewer animals and biomass where currents were present CMinck-
<br />+te threadfin shad, rainbow ley 1979, Cowell and Hudson 1967), presumably in response to changes in
<br />gmbystoma tigrinum, and sediments (Schulback and Sandholm 1962). Shoreline populations of benthic
<br />'amander is a common bait invertebrates were locally dominated by Asiatic clams, reflecting high plankton
<br />~~nd a single specimen alto populations in the lake. These were eaten by carp, channel catfish, and redear
<br /> sunfish, at frequencies comparable to those in the river channel (based on
<br />,wer Colorado River were qualitative examination of stomachs).
<br />'~s were found among de- Epilimnetic penstock intakes in Parker Dam allow warm water to flow down-
<br />~iepended upon a broader stream, thus enhancing habitat for warmwater fishes in the .river belovti~ Lake
<br />'?ic invertebrates, also had Havasu. Particulate materials, including plankton, in turn enhance filter-feeding
<br />piscivores tended to have benthic animals, as do hard bottoms and an abundant micro- and macroflora.
<br />catfish, 39.1%). Edwards Macrophytes, benthic algae, and phytoplankton made up a significant part of the
<br />~~Pd bass as empty. Small- diet of all but centrarachids near Parker, Arizona, with phytoplankton being
<br />~i an incidence of empty derived in part from the pseudofeces of Asiatic clams. Detritus, both from auto-
<br />r .played a wider food base and allochthonous sources, also was present in stomachs of many species at high
<br /> frequencies, especially threadfin shad, carp, and channel catfish. Benthic insects,
<br /> consisting mostly of chironomid dipteran larvae, comprised major parts of the
<br /> diet of all species present. Other invertebrates, excepting clams and crayfish,
<br />the lower Colorado River were broadly represented, but far less significant than chironomids. Clams ~~ere
<br />,~~ar Davis Dam, waters of eaten by specialists (carp, channel catfish, and redear sunfish), forming major
<br />major proportion of basic parts of their diets. Crayfish were generally taken by piscivores, with ttie excQp-
<br />i!us pass. through the dam lion of smallmouth bass, who appeared to feed selectively upon them. Other
<br />eding invertebrates (e.g. ~ fishes were important in the diets of six species, and especially so for channel
<br />hreadfin shad also drawn and flathead catfishes, largemouth bass, and black crappie. Fishes lowest in the
<br />d supply for striped bass food web of the system, threadfin shad and red shiner, were eaten by other
<br />Cold water resulting from fishes most frequently, and were the most abundant species in the river (Minck-
<br />temperate (and obviously ley 1979). Other prey species were mostly juvenile centrarchids, for the most
<br />-eduction is relatively high, part secondary consumers in the system.
<br />data available), but essen- In the lowermost reaches, detritivory became a major mode of life for sailfin
<br />food web. molly, striped mullet, and mouthbrooder. Accumulation of organic materials
<br />,retinue to bP influenced by from upstream, resulting from high rates of production, lack of flooding, and in
<br />~ckwaters provide habitat part from diminution in discharge as a function of progressive water use, allows
<br />relatively devoid of foods these fishes to maintain and expand their populations. However, constraints of
<br />n shad almost non-existent temperature upstream (too low in winter or near Davis Dam) for the molly and
<br /> mouthbrooders, and distance from the sea plus intervening barriers for the
<br />~rvoir that differs greatly mullet, undoubtedly limit their over-all distribution more than food.
<br />Few stomachs of fishes
<br />
<br />~'ed in Table 1 ), but food DISCUSSION AND CONCLUSIONS
<br />.e, Minckley, and Bernell Although the Colorado River provides a relatively low food diversity, food
<br />re distributed relative to habits of introduced fishes of the Colorado River do not differ substantially from
<br />post abundant near nutri- _
<br />those of the same species within their native ranges (see reviews in Calhoun
<br />}-at~ons, and piscivorous 1966). Many abundant forage species are introduced (probably not so For most
<br />~~~~ic predators were more insects and oligochaetes) and are characterized by high reproductive rates and
<br />• ith a more general distri- high-density, monospecific populations. These features are also evident in the
<br />t studied in Lake Havasu, new fish fauna, with many species now populating transitory habitats where
<br />:trient inputs at the upper- populations may explode, stunt, then eventually stabilize at low levels or disap-
<br />m {Portz 1973, Minckley . + pear.
<br />.c.> similar in diversity and The food web of the Colorado River is based upon autochthonous materials.
<br />impoundments (Rinse et Fishes in other large rivers often depend upon allochthonous inputs. In the
<br />
|