This report is dedicated to the spring Chinook of Butte Creek, and to the memory of Randy Brown. 9 Address to the American Fisheries Society. Copied from Joel Hedgpeth, The Passing of the Salmon, in Lufkin (1991). 22 streams (e.g., golden trout), and other salmon can mature in lakes. Masu occur only in Asia, and some Pacific "trout" (e.g., gila trout, O. gilae) occur only in North American, but other Pacific salmon occur on both sides of the North Pacific (Groot and Margolis 1991; Moyle 2002) .

more »

... Chinook and coho salmon have been successfully introduced to the Great Lakes (Carl 1982) , and Chinook populations are established in New Zealand as well; O. mykiss is now widely distributed around the world (Moyle 2002). The anadromous Pacific salmon are major pelagic predators in the North Pacific, feeding on crustaceans, mollusks, tunicates, and other fish (Groot and Margolis 1991; Moyle 2002) . Central Valley Chinook, however, forage primarily in coastal waters off California and Oregon, and Central Valley steelhead may do so as well (Ch. 11). All adult Chinook, coho (O. kisutch) , sockeye, chum, pink and anadromous masu salmon die shortly after spawning, as do most steelhead and cutthroat. Since these fish grow mainly in the ocean, they carry nutrients from the ocean to streams that benefit juvenile salmon and other aquatic organisms. Spawning salmon are also important food for many terrestrial animals, and by various routes marine-derived nutrients from salmon carcasses also move into riparian and adjacent terrestrial ecosystems (Stockner 2003) . Salmon are also important to people. In the Central Valley, as elsewhere, Chinook were an important food for Native Americans (Yoshiyama 1999, but see Gobalet et al. 2004 ), and a commercial fishery developed as soon as the gold rush created a market. Salmon canning began at Sacramento in 1864, and there was a gill net fishery in the Delta until 1957, although in California most salmon have been taken by trolling in the ocean since 1917 (Clark 1928; Fry and Hughes 1951) . Both Chinook and steelhead also support popular recreational fisheries, in both fresh and salt water for Chinook, and in fresh water for steelhead. However, salmon are important to people not only as something to catch and eat; there is something mythical about their spawning migrations that touches a deep chord in the human psyche. Most anadromous Pacific salmon have variable life history patterns, but pink salmon has such a rigid two-year life cycle that populations spawning in the same stream in even and odd years are genetically isolated, and some streams have runs only every other year (Heard 1991). Juvenile pinks, like juvenile chum, migrate to sea directly after emerging from the redds, but chum may spend two to five years at sea, most commonly four (Salo 1991). Coho salmon life cycles are also rather rigid, with almost all females spawning at age three, although males commonly mature at two or three, and at higher latitudes many coho spend two years in streams and spawn at three or four (Sandercock 1991) . Chinook salmon may rear in streams for a few days to two years, and spend a few months to seven years at sea, and some males mature precociously as parr (Rutter 1904; Healey 1991; Zimmerman 2003) . The life histories of anadromous steelhead are even more variable; they may spend a few months to three years in streams, and a few months to five years at sea (Shapovalov and Taft 1954; Withler 1966) . Some do not go past the estuary of their home stream, but others make far-ranging sea migrations (Pearcy et al. 1990 ). Not all die after spawning, and survivors may spawn again after another trip to sea. As one consequence, while simple statements can be made about pink salmon or even 14 Waples et al. ( 2004 ) note that the genetic differences between spring and fall-run populations in the lower Columbia River basin are small enough that they could have developed in 80-100 years. 34 spend a few months in freshwater before spawning, but since spawning occurs in the winter and early spring it is much harder to observe than spawning by spring, winter, and fall Chinook. Juvenile steelhead emerge from late winter to summer. Naturally produced steelhead from the upper Sacramento River and tributaries spend one to three, but usually two, years in fresh water before emigrating, usually in the spring; fish from lower tributaries such as the American River mainly emigrate after one year (Titus et al. 2004 ). Historically, steelhead spawned high enough in stream systems that the water remained tolerably cool for juveniles in the summer. Steelhead can ascend steeper streams and spawn in smaller tributaries than Chinook. Like spring Chinook, steelhead lost most of their natural spawning habitat in the Central Valley to dams. Many populations of O. mykiss, including existing Central Valley populations, consist of both anadromous and non-anadromous individuals (McEwan 2001; Moyle 2002) . It seems likely that dams that release cool water through the summer, such as Shasta on the Sacramento River, New Melones on the Stanislaus River, and New Bullard Bar on the Yuba River, have encouraged a shift toward the non-anadromous life history pattern. This is not a unique situation; dramatic changes in the proportions of anadromous and non-anadromous individuals have been reported in other populations of salmonids (Thorpe 1987; Morita et al. 2000; . Whether this entails genetic change or is simply a phenotypic response to the changed environment is not certain, although in the long term a genetic response to the changed environment seems inevitable. In any event, examples of anadromous progeny of nonanadromous females and vice versa have been documented (Titus et al. 2004 ). 15 Abundance data on anadromous O. mykiss are discouraging. Counts at the Red Bluff Diversion Dam until 1993 showed a rapidly declining population in the upper Sacramento River (Figure 2-9 ). More recent data are scant. The NOAA Fisheries updated status report report (Good et al. 2005) estimated the average number of naturally spawning female steelhead for 1998-2000 at 3,628, based on ratio of unclipped to clipped smolts captured in the USFWS at Chipps Island, an assumed average fecundity (5,000), and a 1% estimate of egg to smolt survival. This estimate may be low, because 3,000 seems a more reasonable estimate for the average fecundity of Central Valley steelhead (Ch. 5), but 1% egg to smolt survival may also be low (e.g., Kostow 2004). Central Valley steelhead were listed as threatened under the Federal ESA in 1998.