Interpreting Early Species Range Descriptions for Pacific , Oncorhynchus spp., in Coastal Watersheds: The Historical Context

BRIAN C. SPENCE

Introduction tablished viability criteria for each list- range of species should be done with Scientists and fish and wildlife man- ed DPS or ESU, and development of considerable care. agers are often asked to determine the recovery plans for these listed units. In seeking to define the natural fresh- historical ranges of species in areas Likewise, historical information on the water ranges of Pacific salmon, scien- where populations have been extirpat- freshwater distribution of Pacific salm- tists and managers have often turned ed as a result of anthropogenic actions. on influenced delineation of essential to writings and collection records of For Pacific salmonids,Oncorhynchus fish habitat under the Magnuson Act pioneering ichthyologists to substanti- spp., freshwater or spawning range de- (PFMC, 1999). ate the historical occurrence in partic- terminations have been critical in sev- Misidentification of historical sal- ular regions or watersheds. These ear- eral phases of Endangered Species Act monid habitats can have important ly descriptions and accounts are gen- (ESA) implementation. These include ramifications. On one hand, failure to erally assumed to provide evidence of the delineation of distinct popula- accurately identify a portion of a spe- occurrence during periods when the tion segments (DPS’s) or evolutionari- cies’ natural range could result in un- impacts of harvest and habitat degra- ly significant units (ESU’s) considered derestimation of habitat loss and lead dation were less pervasive. For salm- for ESA listing, designation of criti- to inadequate protection of habitats on in California, the writings of Da- cal habitats, development of technical that may be important to the long- vid Starr Jordan, Charles Henry Gil- guidance documents that both posited term recovery, persistence, evolution, bert, John Otterbein Snyder, and other historical population structure and es- or sustained production of an ESU or faculty and staff at DPS. Conversely, incorrectly conclud- during the late 1800’s and early 1900’s ing that a species occupied a water- have been particularly influential (Fig. shed or region when it did not could 1). These scientists, as well as collab- Brian C. Spence is with the Fisheries Ecology Division, Southwest Fisheries Science Center, lead to costly recovery efforts and re- orators such as Barton Warren Ever- National Marine Fisheries Service, NOAA, 110 strictions on human activities in water- mann (a former student of Jordan’s Shaffer Road, Santa Cruz, CA 95060 (email: [email protected]). sheds where the likelihood of natural- at University), worked close- ly sustaining populations is negligible. ly with the Commission doi: https://doi.org/10.7755/MFR.81.1.1 Consequently, assessing the historical of Fish and Fisheries (U.S. Fish Com-

ABSTRACT—Scientists and manag- scientists and others have often taken these es. Further confounding interpretation of ers implementing endangered species laws early range descriptions at face value, early reports is that the first systematic ex- often face the task of defining the histori- without critically examining the underlying plorations of coastal watersheds took place cal geographic ranges for threatened and historical context. When Jordan and his well after significant anthropogenic damage endangered species. To do so, they com- contemporaries first began writing about to salmon habitats had already occurred; monly turn to the writings of early biolo- the ranges of Pacific salmon, scientific ex- thus, failure to detect species on these sur- gists seeking accounts of species in regions ploration of coastal watersheds of Cali- veys does not necessarily indicate a species where they may have been extirpated as a fornia was in its infancy. Additionally, the was absent, either at the time of the survey result of anthropogenic activities over the taxonomy and nomenclature of Pacific sal- or in the years prior to significant human last 150–175 years. In the case of Pacific monids were in states of extreme disarray, disturbances. As a result, any single writ- salmon, Oncorhynchus spp., the writings with numerous putative species described ing of Jordan’s and his colleagues between of , Charles Henry Gil- based on variations due to age, sex, and the late 1870’s and the early 1900’s is like- bert, John Otterbein Snyder, and other fac- reproductive condition. ly to contain species range information that ulty and staff at Stanford University during Even after Jordan and Gilbert began is equivocal, if not demonstrably inaccu- the late 1800’s and early 1900’s have been to resolve Pacific salmon taxonomy in the rate. This is not to disparage Jordan and his particularly influential, as these scientists 1880’s, confusion in nomenclature, exacer- contemporaries in any way or to diminish were widely recognized as the leading au- bated by a primitive understanding of Pacif- their extraordinary scientific achievements. thorities on west coast fishes and salmo- ic salmon life histories, contributed to fre- However, scientists and managers need to nids in particular. quent misidentification of west coast salmo- be cognizant of these limitations when using Because of the tremendous achieve- nids and hence inaccurate descriptions of historical writings to guide management of ments of these pioneering ichthyologists, their historical freshwater spawning rang- endangered species.

81(1) 1 cal context within which these writ- ings and observations were embedded. This context has many dimensions, the understanding of which is critical to proper interpretation (Fig. 2). First, the earliest descriptions of the distributions of California fishes, in- cluding the Pacific salmonids, were based on a very limited number of scientific surveys or collections, with data from coastal watersheds of Cali- fornia being particularly sparse. Second, prior to the 1900’s, the tax- onomy and nomenclature of Pacific salmonids were in complete disarray; hence, not all reports of occurrence of a particular species in either the scien- tific or popular literature are necessar- ily reliable. Third, even after more systemat- ic surveys of California’s freshwater fish faunas were initiated and some of the taxonomic ambiguity had been re- solved, misidentification of collected specimens remained commonplace, in part because of the poor understanding of salmonid life histories that existed at the time. Fourth, significant artificial culture and transplanting of certain Pacific salmonids in California began in the 1870’s; thus, the potential exists for some early records of salmonid occur- rence to be the product of hatchery re- leases. Finally, by the mid- to late-1800’s, significant alteration of freshwater habitats had already occurred as a re- sult of logging, mining, cattle graz- ing, dam building, agricultural land conversions, water diversions, release Figure 1.—Renowned ichthyologists of the late 1800s whose writings and explora- of pollutants, and other human activ- tions of California’s fish fauna contained or influenced early range descriptions for ities. Thus, in many areas local extir- Pacific salmon, including (clockwise from top left) David Starr Jordan (ca. 1879), Charles Henry Gilbert (ca. 1880), John Otterbein Snyder (ca. 1900), and Barton pations of salmon and steelhead, O. Warren Evermann (ca. 1899). (Photos: Jordan, Gilbert, and Snyder from Stanford mykiss, populations likely occurred Historical Photograph Collection (SC1071). Evermann from California Academy well before any formal scientific sur- of Sciences GSMAEL.) veys of native fish faunas had taken place. All of these issues potentially mission) and were widely recognized and Lindsey, 1970; Scott and Cross- confound the interpretation of histori- as the leading authorities on fishes man, 1973; Behnke, 2002; Wydoski cal evidence. of North America during this era. In- and Whitney, 2003; Quinn, 2005). In this paper, I endeavor to provide deed, the Pacific salmon range descrip- Owing to the lofty achievements and historical context for interpreting ear- tions of Jordan and his contemporaries status these ichthyological luminar- ly Pacific salmon1 species distribu- from the late 1800’s continue to be ref- ies attained, scientists have often tak- 1In this paper, I use the term “Pacific salmon” erenced, either directly or indirectly, in en these early accounts at face value when referring to the five semelparous species many books and reports (e.g., McPhail without fully considering the histori- of Pacific salmon O.( tshawytscha, O. kisutch,

2 Marine Fisheries Review tion information in the coastal regions of California, focusing on the writings and collection records of David Starr Jordan and his contemporaries during the late 1800’s and early 1900’s. Cal- ifornia marks the southernmost lim- its of the current distributions of both coho salmon, O. kisutch, and Chinook salmon, O. tshawytscha. Pink, O. gor- buscha, chum, O. keta, and sockeye salmon, O. nerka, have also been re- ported in the state, though it remains equivocal as to whether or not persis- tent populations of these latter three species have occurred in California under climatic conditions that have prevailed in the past 150–200 years. This paper does not seek to provide definitive answers regarding the histor- ical freshwater ranges of the different species of Pacific salmon in Califor- nia, or to address the fluctuating rang- es over geological time frames (for longer term perspectives on the evo- lutionary history of Pacific salmonids, see Minckley et al., 1986; Waples et al., 2008; Penaluna et al., 2016). I be- lieve the scarcity of definitive histori- cal records and ambiguity in the writ- ten accounts precludes making final judgments with high confidence. Rath- Figure 2.—Primary elements of historical context required for interpreting early descriptions of the ranges of Pacific salmon on the West Coast of North America. er, the paper is intended to raise a cau- tionary flag against over-interpreting these early writings, as there are many range” in the context of this paper. In may periodically be observed in habi- examples demonstrating that the scien- the simplest terms, a species’ range tats that cannot support persistent pop- tific foundation for accurately describ- may be defined as the set of geograph- ulations. ing the historical freshwater ranges of ic locations at which that species has Although salmon most often stray to these species simply did not exist in been observed (in any of its various streams within relatively close prox- the 1800’s. life stages), as is commonly depict- imity to natal streams, this is not al- ed in species range maps (Brown et ways the case. For example, adult pink The Nature of Pacific al., 1996). For Pacific salmonids (and salmon have periodically (including Salmon Range Limits many other species), this definition is recently) been observed in streams and Before evaluating early accounts unsatisfying because, by their nature, rivers of central California including and descriptions of Pacific salmon anadromous salmonids tend to contin- Pescadero Creek in San Mateo Coun- species in California, it is important ually test the limits of their distribu- ty2, the San Lorenzo River in Santa to define what I mean by a “species’ tion through the process of straying, Cruz County (Scofield, 1916), and the i.e., dispersal of individuals to non-na- Salinas River and Big Creek in Mon- O. nerka, O. gorbuscha, and O. keta) that oc- tal streams to reproduce. terey County (Skiles et al., 2013), cur on the west coast of North America and the Indeed, straying behavior has adap- sites that are 1,300–1,400 km south of term “Pacific salmonids” when including theO. mykiss and O. clarkii complexes and members of tive value, particularly in populations Puget Sound, where the southern-most the genus Salvelinus that enter Pacific coastal wa- that live in unstable environments persistent populations of pink salm- tersheds. The history of the taxonomy surround- where there is high variation in juve- ing the O. mykiss and O. clarkii species complex- es is far more complicated than for the five Pa- nile survival (Quinn, 1984); hence, we 2Jankovitz, Jon. 2017. Summary of annual wa- cific salmon species and is beyond the scope of might expect higher rates of straying ter quality monitoring, fish sampling, and active this paper, though some mention of these species management: Pescadero Creek Lagoon. Califor- is included to illustrate some of the difficulties in near the periphery of a species’ range. nia Department of Fish and Wildlife, Santa Cruz, establishing species ranges. As a result of this behavior, adult fish 39 p.

81(1) 3 on currently reside.3 Likewise, chum gration. This definition allows for the knowledge from the 1800’s to the ear- salmon have occasionally been ob- fact that sink populations can per- ly 1900’s. Critical to this evaluation is served in central California streams sist, sometimes indefinitely, in habi- an understanding of the spatial extent and rivers (Scofield, 1916; Ettlinger et tats where within-habitat reproduc- of scientific surveys conducted dur- al., 2015), even though the nearest per- tion is insufficient to balance mortal- ing this era. To our good fortune, Ev- sistent extant populations appear to be ity, provided that there is significant ermann and Clark (1931) published an 850–1,000 km to the north near Til- immigration (Pulliam, 1988; Holt, exhaustive review of scientific litera- lamook Bay, Oregon (Johnson et al., 1996; Boughton, 2000). This distinc- ture related to the occurrence of fish- 1997). tion is important, as some of the ear- es in the rivers and lakes of California Sexton et al. (2009) suggested that liest freshwater range descriptions ap- covering the period from 1850 to the the challenge in defining species range pear to consider the southernmost ob- mid-1920’s. limits is in identifying environments in servation of a species as indicative of The explicit goal of their review which birth rates exceed death rates, the range, absent any evidence of pop- was to examine “all literature per- how these environments are distribut- ulation persistence. taining to the freshwater fishes of the ed across the landscape, and how they Unless otherwise indicated in this state, as species or kinds, in order that are connected by dispersal. This con- paper, the terms “range” or “species we might know not only what spe- ceptualization implies that species range” refer to the freshwater spawn- cies are known to occur, in Califor- ranges are best defined by groups (or ing ranges of the Pacific salmon. In nia, but also the geographic distribu- populations) of animals, and hence not the marine environment, the south- tion of each of those species within the by the occasional and transient occur- ern extent of the distribution of Pacif- state.” From this treatise, it is possible rence of a species in a particular wa- ic salmon varies through time depend- to reconstruct to a substantial degree tershed. Yet even from this population ing on water temperature, food avail- the spatial extent of scientific explo- perspective, defining a species range ability, and ocean circulation patterns ration in California from 1850 to the for salmon remains problematic be- (Pearcy, 1992). Chinook salmon have late 1870’s, when Jordan and his col- cause environments are inherently dy- been observed as far south as coast- leagues first began writing about the namic. Expansion and contraction of a al waters off San Diego County (Mill- distribution of Pacific salmonids. species range in response to climatic er and Lea, 1972), and are still caught Evermann and Clarks’ analysis in- shifts is a natural process at temporal with some regularity in the Santa Bar- dicates that scientific investigation of scales ranging from years to millennia bara Channel, south of Point Concep- fishes in California began in the early or longer (Brown et al., 1996; Waples tion (Winans et al., 2001; Bellinger 1850’s. Prior accounts of early Span- et al., 2008). Consequently, any defini- et al., 2015). Coho salmon have been ish, Russian, and American explorers tion of a species range should include reported near the Coronado Islands, of California, while sometimes pro- a relevant temporal scale; depending south of the California–Mexico border viding indications of habitat condi- on the purpose, this definition can be (Scofield, 1937). Even pink salmon tions in selected rivers at the time of at geologic, evolutionary, or historical have been observed as far south as the exploration, yield little specific infor- time scales. coast of San Diego County (Hubbs, mation about the fishes encountered Such choices are inevitably some- 1946). Thus, while not discussed fur- (e.g., Fremont, 1856; Costanso, 1911; what arbitrary. However, for purposes ther in this paper, the marine distribu- Fages, 1937; Brewer, 1974; Jackson of implementing the Endangered Spe- tion of Pacific salmon and the regular- and Spence, 1970; Goetzmann, 1993; cies and Magnuson acts, perhaps the ity with which ocean conditions favor Crespi, 2001; Beidleman, 2006; Gib- most useful definition extends that of occurrence may play key roles in de- son, 2013). Occasionally, these narra- Sexton et al. (2009) to include individ- termining opportunities for coloniza- tives will make mention of “salmon” ual populations where, at time scales tion and, hence, changes in the fresh- or “trout,” but any descriptions lack of ~100 years (i.e., scales most rele- water spawning ranges of salmon over sufficient detail for species determina- vant to management4), births plus im- longer periods of time. tion. migration exceeds deaths plus emi- The interest in scientific descrip- Scientific Exploration of California’s tions of fishes and other biota in Cal- 3Small numbers of pink salmon are observed at Fish Fauna: 1850–1879 ifornia that emerged during the early dams along the Columbia River in most years, but it appears that these individuals are likely Interpreting the early scientific re- 1850’s was undoubtedly precipitated strays and not members of persistent spawning cord on the occurrence and distribu- by the influx of settlers to the state fol- populations. tion of Pacific salmon in California lowing the discovery of gold at Sutter’s 4The 100-year time frame is suggested by McEl- hany et al. (2000) as reasonable for assessing via- watersheds requires a fundamental un- Mill in 1848. Still, from 1850 to 1879, bility and extinction risk of Pacific salmonid pop- derstanding of the state of scientific most scientific collections of freshwa- ulations for Endangered Species Act purposes. ter fishes were isolated events in which The authors acknowledge that longer term pro- cesses are important to persistence of an ESU or credible quantitative predictions about viability individuals provided specimens to sci- DPS; however, they note the difficulty in making over longer time spans. entists who then published species de-

4 Marine Fisheries Review scriptions. This is true of both the first the coastal region between Los An- and golden trouts), O. clarkii (Lahon- freshwater fish described in California, geles and , as well as tan from Lake Tahoe), tule perch, Hysterocarpus traski, col- portions of Napa, Sonoma, and Lake or Salvelinus confluentus (bull trout). lected in the lower Sacramento River counties before turning east to the During this era, the only collection re- (Gibbons, 1854), and the first salmo- Sacramento River basin (including a cord for salmon reported by Evermann nid, Salmo irideus (=O. mykiss iride- short trip to Weaverville in the head- and Clark (1931) was of specimens us5), which was taken from San Le- waters of the Trinity River) and the of O. quinnat (=O. tshawytscha) tak- andro Creek in the east San Francisco Sierra Nevada. However, the expedi- en by Stone at the McCloud River near Bay region (Gibbons, 1855). Both of tion’s interest lay primarily in map- the site of the Baird Hatchery (Bean, these descriptions were read at meet- ping topography, collecting fossils, 1880). Equally important, scientific ef- ings of the newly formed California and documenting mineral resources. fort was focused almost exclusively in Academy of Sciences. Brewer’s journals occasionally men- the San Francisco Bay area, the Sac- Additional scientific collectionstion “trout” being caught or eaten by ramento Valley, and the Sierra Nevada. were made during the Pacific Railroad the expedition;6 however, the only ref- The coastal watersheds of California surveys conducted between 1853 and erences to “salmon” include observa- outside of the San Francisco Bay area 1855. These surveys sought to discov- tions of Native Americans catching were essentially unexplored, with only er the most practicable railroad routes and drying salmon near Battle Creek, isolated records of salmonids from from the Mississippi River basin to and the expedition trading for salm- Humboldt Bay, the Mendocino Coast, the Pacific Coast, as well as from San on near the Pit River (Shasta County) and the Ventura River in southern Cal- Diego, Calif., northward to Seattle, (Brewer, 1974). ifornia having been reported, with Wash. (Moore, 1986). At the behest of It was not until the 1870’s that fur- none of those collections involving Pa- Assistant Sec- ther scientific accounts of salmonids cific salmon (Fig. 3). The collection of retary , sur- in California were published. Livings- Chinook salmon from the McCloud geons and officers were assigned to ton Stone, who in 1872 established River stood as the only document- these expeditions as field naturalists the Baird Hatchery (USFC, 1874), ed record of any of the Pacific salmon and were provided manuals and ma- the first federally owned hatchery on species in California. Obviously, fish- terials for making collections (Dall, the west coast, provided information ermen were aware of the presence of 1915; Moore, 1986)(see also Rivinus on salmonid fishes from the McCloud salmon in California, as commercial and Youssef (1992) for an account of River in the northern Sacramento Riv- fishing on the Sacramento had begun Baird’s life and career at the Smithso- er basin. Additional information was in the 1850’s (Collins, 1892) and ac- nian). generated by Henry W. Henshaw’s ex- celerated with the advent of canning Specimens were returned to the plorations of the Sierra Nevada’s Kern technology pioneered by Hume and Smithsonian and findings were sum- River Basin and Lake Tahoe, though Company (McEvoy, 1986). Neverthe- marized in a series of zoological re- these accounts were exclusively of res- less, whatever local knowledge of dif- ports prepared by French biologist ident trout forms (Jordan and Hen- ferent salmon species may have exist- Charles Frédéric Girard, a former stu- shaw, 1878). ed at the time, it had not made its way dent of renowned Swiss geologist and Collectively, Evermann and Clark into the scientific literature before the zoologist Louis Agassiz of Harvard (1931) documents 23 unique fresh- 1880’s. University. Girard was recruited by water or estuarine collection localities7 Baird specifically to assist with ich- for salmonids in the state between the Salmonid Taxonomy and thyological and herpetological collec- years 1850 and 1879 (Fig. 3; Supple- Nomenclature: 1850–1879 tions made during the Pacific Railroad, mental Table S1). All but one of these In addition to being a period of lim- Northwest Boundary, and Mexican observations were of what are current- ited scientific exploration, the pre- Boundary surveys (Jackson and Kim- ly recognized as various subspecies 1880 era was also a period of tremen- ler, 1999). or forms of O. mykiss (i.e., rainbow/ dous confusion with respect to the Following the Pacific Railroad sur- steelhead trout and derivative redband taxonomy and nomenclature of the Pa- veys, there was little scientific atten- cific salmonids. The five species of Pa- tion paid to salmonids of California. 6Locations where Brewer mentioned trout include cific salmon were originally described Battle Creek and the Pit River (Shasta County); Between 1860 and 1864, William H. the South Fork Kings River (Fresno County); the by German naturalist Georg Wilhelm Brewer led an expedition of the Cali- Middle Fork San Joaquin River (near the border Steller based on observations made of Madera and Fresno counties) and Lake Pyr- 8 fornia Geologic Survey that covered amid and the Truckee River (likely in Nevada). in Kamchatka during the 1700’s. These lacked any detail that would allow identi- 5Throughout this paper, when citing a scientific fication to species, though in the latter case, these 8Jordan (1892b) implied that the Steller’s origi- name that is no longer considered valid, I adopt were almost certainly Lahontan cutthroat trout. nal descriptions of the five Pacific salmon species the convention of adding a parenthetical with the 7The list of Evermann and Clark is slightly lon- were based on observations made in cir- currently accepted name preceded by an equals ger than 23; however, several of these are simply ca 1731; however, this is incorrect. Steller was sign “=”. duplicates of previous records. in fact present on Russia’s American Expedition

81(1) 5 The Russian vernacular names giv- en by Steller were adopted by Wal- baum as scientific names in 1792 (Jor- dan, 1888). However, over the next 90 or so years, this taxonomic clarity dis- integrated, as numerous putative spe- cies were described in North America based on variations in morphology due to sex, life stage, and breeding condi- tion. Jordan (1888) described the re- sulting taxonomic chaos as follows:

“Since Steller’s time, writers of all degrees of incompetence, and writers with scanty material or with no material at all, have done their worst to confuse our knowl- edge of these salmon, until it be- came evident that no exact knowl- edge of any of the species re- mains. In the current system of a few years ago,9 the breeding males of the five species known by Stell- er constituted a separate genus of many species (Oncorhynchus Suckley); the females were placed in the genus Salmo, and the young formed still other species of a third genus, called Fario, supposed to Figure 3.—California locations where scientific collections of salmonids were made be a genus of trout. The young in the period from 1850 to 1879. Numbers correspond to observations in Supple- breeding males (grilse) of one of mental Table S1. the species (Oncorhynchus ner- ka) made still a fourth genus, des- ignated as Hypsifario. Not one of are really distinguished. Not less of Jordan’s scathing indictment was the writers on these fishes of twen- than thirty-five nominal species of George Suckley, a surgeon and natu- ty-five years ago knew a single Oncorhynchus have already been ralist appointed to the Pacific Railroad species definitely, at sight, or used described from the North Pacific, Survey of the 47th and 49th parallels. knowingly in their descriptions a although, so far as is now known, Suckley inherited the task of preparing single character by which species only the five originally noticed by a report (Suckley, 187410) on the salm- Steller really exist. The descrip- on and trout based on work of Caleb (June 1741–August 1742); however, the expedi- tive literature of the Pacific salm- Burwell Rowan Kennerly. Kennerly tion did not reach Alaska (Kayak Island) until July on is among the very worst ex- was likewise a physician and natural- 1741, and Steller spent just ten hurried hours on tant in science. This is not, how- ist who participated in both the Pacific the island, primarily collecting plant specimens (Ford, 1966; Littlepage, 2006). Although Steller ever, altogether the fault of the Railroad and Northwestern Boundary did report finding dried “red salmon” in a Native authors, but it is in great part due surveys but who died in 1861 before American “cellar” on the island and concluded to the extraordinary variability in completing his report. Suckley’s re- that they were the same species as the “nerka” of Kamchatka, his brief stay on the island was clear- appearance of the different species port described no fewer than 27 nomi- ly not the basis for his Pacific salmon descrip- of salmon. These variations are, as nal species of salmon and trout in riv- tions, which include not only the characteristics of various species, but aspects of their life histo- will be seen, due to several differ- ers that drain into the Pacific Ocean, ries as well. Rather, Steller’s descriptions likely ent causes, notably to differenc- including 17 he classified as anadro- resulted from observations that both he and his es in surroundings, in sex, and in mous “salmon” (Table 1).11 assistant, Stepan Krasheninnikov, made in Kam- chatka both before and after the American Expe- age, and in conditions connected dition. Steller died before he was able to publish with the process of reproduction.” 10Suckley’s report was written in 1861, but was these descriptions, but they were published sub- not formally published until 1874, five years af- sequently by Krasheninnikov (1764). ter his death. 9Jordan inserted a footnote referencing the report Although not explicitly acknowl- 11Jordan did call out Suckley by name in a sub- of the U.S. Pacific Railroad Explorations, 1858. edged in this paper, the primary target sequent paper (Jordan 1892b), saying that Suck-

6 Marine Fisheries Review Table 1.—Putative salmon and trout “species” identified in Suckley (1874)1 for waters draining to the Pacific Ocean.2

Scientific name Common names Location Modern identification I. Anadromous salmon: species running up into fresh water to spawn; the young remaining there for a greater or less time, then returning to the sea, in which they continue to abide, except during the period of reproduction (Salmon.)

a. Intermaxillaries of males long, decurved, projecting, and hooking downward considerably beyond the top or knob of the lower jaw (Subgenus Oncorhynchus) 1. Salmo scouleri, Richardson Hook-nosed salmon; fall salmon Pacific Coast O. gorbuscha, O. kisutch 2. Salmo proteus, Pallas Hump-backed salmon Alaska Coast O. gorbuscha 3. Salmo cooperi, Suckley Cooper’s salmon; little red salmon; [Ta-ah-nia] Columbia River O. tshawytscha, O. nerka3 4. Salmo dermatinus, Richardson [red-fish, tleukh-ko] Bering Sea O. keta 5. Salmo consuetus, Richardson Yukon River O. keta 6. Salmo canis, Suckley Dog salmon, spotted salmon, [Le Kai] Puget Sound O. keta

b. Jaws of adult males when fresh-run, symmetrical and either subequal or the point of the lower jaw received in a notch between premaxillaries. †Without red spots; not feeding in fresh water, except from caprice 8. Salmo quinnat, Richardson The California salmon West Coast of U.S. O. tshawytscha 9. Salmo confluentus, Suckley Towatl salmon Northwest Coast O. tshawytscha 10. Salmo aurora, Girard Red-char; salmon Columbia River O. clarkii 11. Salmo argyreus, Girard West Coast O. tshawytscha 12. Salmo paucidens, Richardson Weak-toothed salmon Fraser River O. nerka 13. Salmo tsuppitch, Richardson White salmon; silvery-white Salmon trout Columbia River O. kisutch, O. clarkii henshawi 14. Salmo clarkii, Richardson Clark’s salmon Columbia River O. clarkii 16. Salmo gairdneri, Richardson Gairdner’s salmon Columbia River O. mykiss 17. Salmo truncatus, Suckley The short-tailed salmon Columbia River O. mykiss 18. Salmo richardi, Suckley Richard’s salmon; Suk-kegh salmon Fraser River O. tshawytscha, O. nerka3 ††Spotted with red; feeding freely in fresh water 19. Salmo campbelli, Suckley Campbell’s salmon Columbia River Salvelinus malma or confluentus

II. Non-anadromous species (not running up from the sea, but living entirely in freshwater or only occasionally passing down to sea. [Trout]).

c. Spotted with red or black; found in flowing fresh water; feeding, spawning, and spending the greater part of the year in the same; retiring to deep, still water in the winter; access to salt water usually relishes, but not indispensable. ††Black-spotted 26. Salmo iridea, Gibbons Pacific brook-trout California streams O. mykiss irideus 27. Salmo masoni, Suckley Mason’s trout Columbia River O. mykiss irideus 29. Salmo lewisi, Girard Missouri trout Rocky Mtn slopes N. of South Pass O. clarkii 30. Salmo brevicauda, Suckley Short-tailed trout Puget Sound waters O. clarkii

d. Trout found in deep rivers or lakes, ascending shallow streams to spawn †Black-spotted 31. Salmo gibbsii, Suckley Columbia salmon trout Columbia River O. mykiss gairdneri4 33. Salmo kennerlyi, Suckley Kennerly’s trout; Chiloweyuk red salmon-trout Chiloweyuck Lake; Fraser’s River O. nerka 34. Salmo warreni, Suckley Warren’s trout Fraser’s River O. tshawytscha (immature) ††Red-spotted 35. Salmo bairdii, Suckley Baird’s trout; red-spotted Rocky Mtn. trout Clark’s Fork, Columbia River S. confluentus 36. Salmo parkii, Suckley Parker’s River trout Kootenay River; Rocky Mtns. S. confluentus

e. Lake trout: passing lives in deep freshwater lakes. 43. Salmo newberryi, Girard Newberry’s salmon Klamath River O. mykiss newberrii 1Numbers, as well as spellings of species and geographic locations, are as they appear in Suckley (1874:92–94); species not found on West Coast of North America are omit- ted. 2Redeterminations of species come primarily from Jordan and Evermann (1896), with some corrections based on modern knowledge of species distributions. 3Salmo cooperi and Salmo richardi have both been considered synonyms for O. tshawytscha and O. nerka. 4Salmo gibbsii is considered a synonym for O. mykiss gairdneri (Behnke, 2002).

Eventually, these so-called salmon work with, many of which were sim- Coast of North America alone that were determined to include a mixture ply dried skins or specimens damaged salmon occur in anything like the of the five species currently recog- by alcohol (Anonymous, 1880). numbers which formerly prevailed nized as Pacific salmon (O. gorbuscha, Resolving the taxonomy of the Pa- in the East, though the species are O. keta, O. kisutch, O. tshawytscha, cific salmon complex became a priori- entirely distinct and peculiar to and O. nerka), as well as anadromous ty of the newly formed U.S. Fish Com- the Pacific. The waters of Califor- forms of Pacific trout and char, in- mission (USFC), to which Baird was nia, Oregon, and cluding O. mykiss, O. clarkii, Salveli- appointed as the first Commission- boast of the possession of sever- nus malma, and S. confluentus. Addi- er (Rivinus and Yousef, 1992). Baird al kinds, how many of which has tionally, Suckley identified ten putative (1874) championed the need for im- not been ascertained, as the differ- non-anadromous “species,” several of proved understanding of the taxono- ent ages and sexes of one have in which were later determined to be im- my and distributions of Pacific salmon many instances been described as mature forms of salmon or steelhead. species as follows: two or more totally distinct spe- In their defense, Suckley and Kennerly cies. One of the objects of the Fish had a limited number of specimens to “The western salmon (Salmo Commission is to solve the prob- quinnat?12)—It is on the West lem in question, by securing spec- ley “succeeded in carrying the confusion to an extreme, making as many as three genera from a 12The question mark following the scientific single species of salmon, founded on differences name is present in Baird’s 1874 report and in- the numerous “varieties” of Pacific salmon were of age and sex.” dicates the prevailing uncertainty about whether all members of the same species.

81(1) 7 imens of all ages and both sexes Notably absent from Jordan’s 1878 undertook an extensive exploration of from all North American locali- list was O. kisutch. Interestingly, Jor- the fish fauna along the Pacific Coast ties, and, by critical investigation dan did assign the name Salar tsup- of North America from San Diego, Ca- and comparison, to determine pre- pitch to fish that were eventually deter- lif., to the mouth of the Fraser River, cisely the limitations and relation- mined to be O. kisutch. The genus Sa- B.C. Jordan was approaching his 29th ships of each kind.” lar included what Jordan referred to as birthday when the expedition depart- the “Salmon trout,” which contained ed; Gilbert had just turned 20. The ex- Baird asked Jordan to take on this representatives of both the modern O. pedition was initiated at the request of task in 1877 (Jordan, 1922). Jordan’s mykiss and O. clarkii complexes. In a the U.S. Census Bureau in cooperation initial work was based on specimens second paper (Jordan, 1878b), he re- with the U.S. Fish Commission under that were collected during the Pacif- ferred to Salar as a subgenus of the the direction of Baird. Including the re- ic Railway Survey of the 1850’s and genus Salmo, and called Salmo tsup- turn trip to San Francisco, the journey housed at the United States National pitch the “Black Trout of Lake Tahoe,” took roughly 7 months to complete and, Museum of Natural History (USNM) apparently believing it was the same according to Evermann (1930), result- at the Smithsonian Institution, as well species as both Lahontan cutthroat ed in “the discovery and describing of as subsequent collections, the most trout (O. clarkii henshawi) from the more new species than any fishery sur- noteworthy of which were those made Lake Tahoe Basin and O. mykiss from vey previously made.” It remains one at a new hatchery on the Clackamas the Kern river drainage. Shortly there- of the most remarkable achievements River in Oregon (described in Jordan, after, in a letter to Forest and Stream, in the history of North American ich- 1878b; Jordan, 1922). Jordan pub- Jordan (1880) added the name O. tsup- thyology. By Jordan’s (1922) account, lished what he called his “tentative pitch to the list of Oncorhynchus spe- roughly 400 different species were col- conclusions” about Pacific salmon tax- cies, noting that he and Theodore Gill lected, about 80 of which were not pre- onomy a year later in A Catalogue of had earlier assigned the name “tsup- viously known to science.14 Most of the Fishes of the Fresh Waters of North pitch” to a different fish, which turned these specimens were shipped to the America (Jordan, 1878a). In this book, out to be “the same as Salmo clarki.” USNM, where many still reside. Jordan recognized five putative spe- Curiously, O. keta is absent from Jor- Shortly after the expedition, Jordan cies of Oncorhynchus: Oncorhynchus dan’s 1880 list, despite its mention in and Gilbert (1881a) published Notes gorbuscha (Walbaum), O. keta (Wal- the 1878 paper. on the fishes of the Pacific Coast of the baum), O. quinnat (Richardson; =O. From the writings of Jordan be- United States, which provided brief tshawytscha), O. nerka (Walbaum), tween 1878 and 1880, we can see that descriptions of all known species (in- and O. kennerlyi (Suckley; =O. nerka). although all five species of semelpa- cluding the Pacific salmonids) and their The latter two “species” were eventu- rous Pacific salmon occurring in the geographic distributions. This marks ally resolved to be the same, O. ner- had at various times the first of Jordan and Gilbert’s writ- ka, with O. kennerlyi being the land- been identified, there remained linger- ings that recognized all five species of locked form of sockeye salmon known ing uncertainty about how many spe- Pacific salmon found in western North as “kokanee.” cies were truly present. Moreover, it is America and correctly assigned them The geographic ranges for O. gor- abundantly clear, both from the pauci- to the genus Oncorhynchus. buscha, O. keta, and O. nerka were ty of collection records in California It is tempting to assume that infor- listed simply as “North Pacific coasts and the lack of complete, unambigu- mation gleaned during the 1880 ex- of Asia and America,” with the distri- ous descriptions of the Pacific salmon pedition provided Jordan and Gilbert bution of “O. quinnat” described as species (and particularly coho salm- with a solid scientific basis for estab- the “Coasts of California to British on), that the scientific basis for defin- lishing the historical spawning dis- Columbia.” Curiously, the range of O. ing the ranges of Pacific salmon spe- tributions of Pacific salmonids along kennerlyi is described as “Sacramen- cies even in the most general terms the coast of California, Oregon, and to River to British Columbia.” It is un- was simply lacking at this time. . However, closer scrutiny clear what records of this species led of the details of the expedition reveals Jordan to conclude it occurred in Cal- The 1880 Pacific Coast Expedition that, for several reasons, this is not ifornia, though Jordan (1878b) did re- In 1880, Jordan, accompanied by his mark that the kennerlyi he examined assistant and graduate student Gilbert, 14It is not entirely clear how Jordan (1922) ar- bore resemblance to quinnat salmon; rived at these numbers. The list of Pacific Coast thus, this was perhaps a case of mis- fishes compiled by Jordan and Gilbert (1880) ly also reported O. kennerlyi under the heading contains 270 unique species, of which 262 were identification.13 “Species not anadromous, nor running up from observed on the 1880 expedition and 44 were the sea, but living entirely in fresh water or only identified as new species; however, it is clear that 13Jordan references Suckley in his description of occasionally passing down to the sea.” This sug- further splitting of taxa occurred following publi- O. kennerlyi; however, Suckley (1874) reports the gests these were kokanee salmon, though many cation of the 1880 list, which may account for the known localities of O. kennerlyi as being Chilow- of Suckley’s designations of anadromous versus difference. Regardless, the expedition was clearly eyuck Lake (likely Chilliwack Lake) and “Fra- non-anadromous later proved to be incorrect, so a watershed event in the history of North Ameri- ser’s River,” both in British Columbia. Sucke- we cannot be certain. can .

8 Marine Fisheries Review likely the case. Foremost, the expedi- tion was focused almost exclusively on marine fishes. Indeed, the specific charge given to Jordan was “… to visit or communicate with every post office within five miles of the coast of Cali- fornia, Oregon, and Washington, to list the various species of fishes and other marine animals [emphasis added] in- habiting adjacent waters, and to report on their habits, food, and value; also to describe in detail the past, present, and probable future of all industries related to the sea” (Jordan, 1922). The emphasis on marine species is clearly evident in the ~555 specimen jars attributable to the Pacific Coast Expedition that are currently housed at the USNM (Supplemental Table S2). Of these lots, only 36 were reported as collected from fresh waters, with all but two of these freshwater specimens taken from the lower Sacramento, Co- lumbia, and Fraser rivers.15 The USNM electronic database lists 25 collections of Pacific salmonids from the 1880 ex- pedition, and the original USNM led- ger identifies another 11 salmonid re- cords from the expedition that have since been lost or destroyed. Of the 36 total salmonids, 16 were identified as collected from fresh wa- ter16, including Chinook, coho, chum, and sockeye salmon from the Fras- er River; Chinook, sockeye, and pink salmon, as well as steelhead and cut- throat trout, from the Columbia Riv- er17; Chinook, coho, chum, and pink

15The two freshwater collections outside of the Fraser, Columbia, and Sacramento rivers include specimens originally identified asCypriniodon californiensis (=Cypriniodon macularius) from Figure 4.—Locations visited by David Starr Jordan and Charles Henry Gilbert dur- the desert east of San Diego, and Salmo iride- ing the 1880 Pacific Coast Expedition from San Diego to the Fraser River. Gray us (=O. mykiss irideus) collected from the San dots indicate ports and locations where samples were collected. *Jordan and Gil- Luis Rey River. The latter specimen has no col- lection date but is presumed to be the basis for bert sampled 7 locations within Puget Sound between Port Townshend and Olympia. Jordan and Gilbert’s (1881a) description that lists the species as far south as the San Luis Rey Riv- er. Another four freshwater collection records are salmon from the Sacramento River; the collection were apparently taken at listed in the USNM ledger, but are not in the elec- tronic database due to the fact that the specimens and O. mykiss from the San Luis Rey sea (Table 2; Supplemental Table S3). were lost or destroyed (USNM 27307, 27308, River.18 All remaining salmonids in Equally important from the perspec- 27352, and 27358); all of these were from the tive of establishing the southern lim- Columbia and Sacramento rivers. 18 16 USNM-26795 is undated but is listed as having its of the spawning distributions of Pa- Although listed as “fresh water” specimens, the been collected by David Starr Jordan in the San cific salmonids is that fact that Jordan collection localities of Astoria, and the lower Sac- Luis [Rey] River near San Diego. Given that Jor- ramento and Fraser rivers are within the zone of dan wrote about S. irideus in the San Luis Rey tidal influence. River in 1881, it seems likely that this specimen the expedition and their collection locations. Jor- 17Jordan and Gilbert’s (1880) table also lists O. was collected during the 1880 expedition, though dan does, however, mention stopping at Mission kisutch from the Columbia River, but no record curiously it is not listed in Jordan and Gilbert San Luis Rey while en route to San Diego (Jor- was found in the USNM database or ledger. (1880), which catalogs all species obtained on dan, 1922).

81(1) 9 Table 2.—Salmonids collected during Jordan and Gilbert’s 1880 Pacific Coast Expedition as listed in Jordan and Gilbert (1880), with notes from USNM ichthyology database and collection ledger. m = specimens from marine waters; f = specimens from fresh waters; u – uncertain if marine or freshwater collection. 2 3 1 4

No. Scientific name assigned by Jordan and Gilbert Accepted scientific name (AFS) Puget Sound Columbia River San Francisco Monterey Bay San Luis Obispo Santa Barbara San Pedro San Diego 210 Salvelinus malma (Walbaum) Jordan & Gilbert Salvelinus confluentus/Salvelinus malma5 m f 211 Salmo purpuratus Pallas Oncorhynchus clarkii/Oncorhynchus mykiss m f u6 u6 212 Salmo irideus Gibbons Oncorhynchus mykiss irideus m f7 213 Salmo gairdneri Richardson Oncorhynchus mykiss u f u u 214 Oncorhynchus kennerlyi (Suckley) Jordan Oncorhynchus nerka u f 215 Oncorhynchus gorbuscha (Walbaum) Gill & Jordan Oncorhynchus gorbuscha u f f 216 Oncorhynchus keta (Walbaum) Gilbert & Jordan Oncorhynchus keta mf f mf 217 Oncorhynchus kisutch (Walbaum) Jordan & Gilbert Oncorhynchus kisutch f f mf 218 Oncorhynchus tshawytscha (Walbaum) Jordan & Gilbert Oncorhynchus tshawytscha mf f f m 219 Oncorhynchus nerka (Walbaum) Gilbert & Jordan Oncorhynchus nerka f f 1Puget Sound includes multiple sites in the Sound, as well as Cape Flattery, Neah Bay, and Fraser River. 2San Francisco includes Point Reyes and Sacramento River. 3Monterey includes Santa Cruz and Soquel. 4San Pedro includes Wilmington and Santa Catalina Island. 5The distinction between S. confluentus and S. malma had not been recognized, and as the distribution of the two species overlaps in the Puget Sound Region, it is possible that Jordan’s samples included either or both species. The specimens from the Columbia River are most likely S. confluentus, as S. malma is not known to occur in the basin. 6The diagnostic characteristics (namely, presence of basibranchial teeth) used by Jordan to classify trout species do not reliably discriminate between O. mykiss and O. clarkii. The specimens from San Francisco and Monterey Bay reported as Salmo purpuratus were almost certainly O. mykiss, as the southern-most extent of O. clarkii appears to be the Eel River basin in northern California (Behnke, 2002). 7Specimen is in USNM collection (USNM 26795), but San Diego was not listed as collection locality in Jordan and Gilbert (1880) table. and Gilbert did not make any collec- This table identifies the primary sam- In addition to the limited sam- tions—either freshwater or marine—in pling localities as San Diego, San Pe- pling of fresh waters and critical spa- the region from Point Reyes, Calif., to dro/Wilmington (including the Chan- tial gaps in the sampling of the Pa- Astoria, Oreg., near the mouth of the nel Islands), Santa Barbara, San Luis cific Coast expedition, the timing of Columbia River, an expanse of nearly Obispo, Monterey, San Francisco (in- the Pacific Coast expedition was such 1,000 km of coastline (Fig. 4). Jordan cluding Pt. Reyes), the Columbia Riv- that it is unlikely that Jordan and Gil- (1922:222) wrote in his autobiogra- er, and several locations within the bert would have directly encountered phy that “After a fairly thorough inves- Puget Sound and Strait of Juan de any fall-spawning salmon species in tigation of the marine interests of San Fuca region. Consequently, while the streams and rivers. The expedition de- Francisco and neighboring waters19, expedition did yield some information parted from San Diego in January and we went in May directly to Astoria, the on the coastal distribution of Pacific did not arrive in Monterey until late great salmon center of our coast at the salmonids, the spatial extent of fresh- March (Jordan, 1922). The party then mouth of the Columbia.” The bypass- water collections by Jordan and Gil- traveled to San Francisco Bay, where ing of the northern California Coast bert was clearly insufficient to rigor- it remained until May. After exploring and virtually all of coastal Oregon is ously define the freshwater ranges of the area in and around San Francisco affirmed by the lack of any specimens any salmonid species on the west coast Bay, the expedition travelled direct- from this region in the USNM collec- of North America, contrary to what ly to Astoria, where it remained until tions associated with the expedition has been implied by others (Kaczynski early June, before heading north to the (Supplemental Table S3), as well as in and Alvarado, 2006).21 Puget Sound region (Jordan, 1922; Pi- Jordan and Gilbert’s detailed table list- etsch and Dunn, 1997). In California, ing of localities for all species collect- and Gilbert typically wrote “seen by us at [geo- the migration and spawning season for ed on the expedition (Jordan and Gil- graphic location]” when they themselves collect- fall-spawning Pacific salmon species bert, 1880; Supplemental Table S4).20 ed specimens on the 1880 expedition. The lack of is typically over by early winter, which any USNM collection records from these locali- ties supports this interpretation. suggests that whatever information 19 In addition to surveying the waters of San Fran- 21In addition to incorrectly concluding that Jor- they obtained about the spawning dis- cisco Bay and the lower Sacramento River, Jor- dan’s Pacific Coast Expedition involved signif- tributions of pink, chum, coho, sock- dan also made marine collections near Point icant surveys of freshwater systems, Kazcynski eye, or fall-run Chinook salmon in Reyes, about 50 km north of the Golden Gate. and Alvarado (2006) also imply that addition- 20Jordan and Gilbert (1881a) describes a few al surveys of stream systems were conducted in California was not based on first-hand non-salmonid species as occurring from Cape California in 1889, 1892, 1904, and 1908, citing Mendocino southward, or in Humboldt Bay; how- an article by Thompson (1922). Thompson’s ar- ever, these geographic references appear to be ob- ticle, however, makes no references to surveys in- to the development of California’s marine fisher- tained from fishermen or other sources. Jordan volving fresh waters. In fact the article is devoted ies and excludes Pacific salmon entirely.

10 Marine Fisheries Review observations. Jordan and Gilbert rou- Table 3.—Description of Pacific salmon species ranges given in Jordan and Gilbert (1881). Spellings and capital- ization are as found in Jordan and Gilbert. Modern scientific name in square brackets if different from histori- tinely spoke with local fishermen and cal name. cannery workers during the expedition Species (common names) Range description and, in fact, obtained many specimens Oncorhynchus gorbuscha either directly from fishermen or by (Humpback Salmon; Haddo; Hone; Holia) Sacramento River northward to the Arctic Sea; abundant in Puget purchasing them at local fish markets Sound on alternate years, 1880 being a year of scarcity. Occasionally seen in the Columbia and Sacramento, but not (Pietsch and Dunn, 1997). It is thus sufficiently abundant to constitute a distinct run. likely that their perceptions regard- Oncorhynchus keta San Francisco to Behring’s Straits; very abundant in fall when it runs ing species ranges were influenced by (Dog salmon; Quarlsch; Kayko; Le Kai) in all streams, but not to great distance. Not seen by us anywhere in these anecdotal accounts. the spring. In September of 1880, the expedi- Oncorhynchus kisutch Sacramento River to Puget Sound and northward; very abundant in (Coho salmon of Frazer’s River; summer and fall. It is rarely taken in the Columbia in the spring, but tion ended and Jordan returned to In- Silver Salmon; Kisutch; Bielaya Ryba) great numbers run up the river in the fall. diana (Jordan, 1922). Less than three Oncorhynchus chouicha [tshawytscha] From Ventura River northward to Behring’s Straits, ascending months later, he and Gilbert pub- (Quinnat Salmon; King Salmon; Chouicha; Sacramento, Rogue’s, Klamath, Columbia, and Frazer’s Rivers Chinook Salmon; Spring Salmon; in spring, as well as the streams of Alaska, Kamtschatka, Japan, lished their tabular list of species from Columbia River Salmon; Sacramento Salmon; and Northern China; in fall ascending these and probably all other the Pacific Coast (Jordan and Gilbert, Winter Salmon; White Salmon). rivers in greater or less abundance; the young taken in Monterey Bay, 1880), followed within a year by Notes Puget Sound, etc. in summer in considerable numbers. on the Fishes of the Pacific Coast of Oncorhynchus nerka From Columbia River to the Aleutian Islands; the principal salmon (Blue-back; Sukkeye; Red-Fish; Rascal; of Frazer’s River; unknown in Eel River, Rogue River, and in the the United States (Jordan and Gilbert, Frazer’s River Salmon; Krasnaya Ryba) Sacramento. In the Columbia River it is much less abundant than the 1881a), which included species range Quinnat salmon, and its flesh is less firm and paler. descriptions for the Pacific salmonids (Table 3). Clearly, information from the expedition shaped these descrip- pears traceable to the collection of bert 1880:457)(Table 2). Nor do there tions. specimens near San Francisco (USNM appear to be any other pre-1880 re- For four of the five Pacific salmon 27220) and in the Sacramento River cords of Chinook salmon from wa- species, the southern extent of the re- (USNM 27358). Neither of these latter ters south of San Francisco Bay in the ported distribution is directly traceable specimens is listed in the USNM elec- USNM collection. Hence, the basis for to observations made on the 1880 ex- tronic database; however, both appear Jordan and Gilbert’s assertion that the pedition. Jordan and Gilbert’s (1881a) in the original USNM ledger. Finally, range of Chinook salmon extended as statement that coho salmon occur Jordan and Gilbert’s (1881a) descrip- far south as the Ventura River remains from the “Sacramento River to Puget tion of the range of sockeye salmon a mystery. Sound and northward” likely ema- “from the Columbia River to the Aleu- From an ecological perspective, the nates from the fact that they reported collecting coho salmon in the marine tian Islands” aligns with an 1880 col- Ventura River and other coastal wa- waters near San Francisco (USNM lection record from the lower Colum- tersheds in southern California seem 27222), as well as in the Sacramen- bia River (USNM 27303). For all four unlikely habitats for Chinook salm- to River (USNM 2725022). Their de- of these species, the USNM museum on. At the southern end of the Santa scription of pink salmon as “occasion- records also match the tabular descrip- Cruz Mountains near Monterey Bay, ally seen in the Columbia and Sacra- tions provided in Jordan and Gilbert the coastal mountains transition from mento, but not sufficiently abundant (1880) (Table 2). coastal redwood forests to chapar- to constitute a distinct run” likely re- The lone instance in which Jor- ral and oak woodland dominated veg- flects the reported collection of an dan and Gilberts’ (1881a) range de- etation characteristic of Mediterra- adult pink salmon in the Sacramen- scription for a salmon species does nean climates. Watersheds of Ventura to River (USNM 27249).23 Likewise, not match evidence gathered during and neighboring Santa Barbara coun- the conclusion that chum salmon oc- the 1880 expedition is the assertion ties, some 240–420 km south of Mon- cur as far south as San Francisco ap- that the range of Chinook salmon ex- terey, experience short winters and re- tended “From Ventura River north- ceive substantially less precipitation 22This latter record comes from the USNM led- ward to Behring’s Straits.” This de- than those north of Monterey, usually ger; it is not present in the USNM electronic da- scription is puzzling for several rea- tabase. in infrequent but intense storms. In the sons. Foremost, Jordan and Gilbert’s 23This collection record is somewhat puzzling. larger watersheds of this region (e.g., The specimen is an adult male in reproductive detailed accounting of collections Santa Maria, Santa Ynez, Ventura, and condition, yet Jordan and Gilbert did not arrive made during the 1880 expedition re- Santa Clara river basins), rivers flow in San Francisco until sometime around April, an unlikely time to encounter a mature male in fresh- veals no evidence that Chinook salm- across broad coastal plains or through water. Most likely, the specimen was collected on (or any other Pacific salmon) were alluvial valleys before reaching the Pa- and preserved earlier in the year by a third par- observed or collected any farther south ty and presented to Jordan when he arrived in cific Ocean, and these reaches may San Francisco. than Monterey Bay (Jordan and Gil- go dry or have minimal surface flows

81(1) 11 during summer and fall.24 Further, It may be that local fishermen did oc- once spawned as far south as the Ven- for many of these systems, a sandbar casionally catch a stray Chinook salm- tura River has been repeated count- forms at the mouth of the river during on in the river, as their ocean distribu- less times in textbooks, journal arti- the dry season, which precludes fish tion can extend to southern California cles, agency reports, and elsewhere from migrating to and from the ocean. under certain conditions. Alternative- (see e.g., Myers et al., 1998; Behnke, In drought years, these sandbars may ly, fishermen may have reported catch- 200230; PFMC, 1999; Quinn, 2005; not breach until mid-winter or do so ing “salmon” in the river that were in- Gustafson et al., 2007). intermittently, posing a challenge for stead steelhead, which were (and still In total, it is evident that while the Chinook salmon, which must not only are) present in the river (Busby et al., Pacific Coast Expedition was a monu- enter streams to spawn in winter, but 1996; Boughton et al., 2006). Such mental achievement, the limited spa- emigrate to the ocean before the sand misidentifications were routine in the tial extent of freshwater sampling (i.e., bar closes again.25 These habitat dif- late 1800’s (see discussion below). largely confined to the Sacramento, ferences and intermittent ocean con- And lastly, for reasons that will be- Columbia, and Fraser rivers), includ- nectivity make the notion of persistent come apparent in the next section, it is ing the large gap in sampling between populations of Chinook salmon in the also plausible that fishermen reported San Francisco and the Columbia Riv- Ventura River and other Southern Cal- catching Chinook salmon in the ocean er, coupled with lack of temporal over- ifornia watersheds highly dubious.26 near Ventura, and that Jordan and Gil- lap between the expedition and the pri- Hubbs (1946) believed that Jor- bert assumed those fish were salm- mary period of migration and spawn- dan and Gilbert’s (1881a) inclusion of on and had spawned in nearby riv- ing by fall-run species greatly limited the Ventura River in the range of Chi- ers.28 Indeed, Hubbs (1946) noted that what was learned about the spawning nook salmon was based on an obser- all other records of Chinook salmon distributions of Pacific salmon. Equal- vation made by Evermann; howev- from southern California (e.g., Cro- ly important, it is evident that Jordan er, Hubbs provided no elaboration on ker, 1930, 1936, among others) were and Gilbert’s (1881a) fundamental why he thought that to be the case.27 based on fish caught at sea between conceptualization of a species’ range San Pedro and La Jolla. Other ear- was quite distinct from how modern 24These conditions have been exacerbated by hu- ly references that assert Chinook oc- scientists view this question. Whereas man activities since the 1800’s; however, Beller currence in the Ventura River appear current scientists typically define spe- et al. (2011) present evidence that these channels were intermittent even prior to such disturbanc- to be restating Jordan’s conclusions, cies ranges in terms of those portions es. (Beller, E. E., R. M. Gossinger, M. N. Sa- rather than indicative of new observa- of the landscape where biotic and abi- lomon, S. J. Dark, E. D. Stein, B. K. Orr, T. R. tions (e.g., Gill, 1883; Collins, 1892). otic conditions permit persistent popu- Longcore, G. C. Coffman, A. A. Whipple, R. A. Askevold, B. Stanford, and J. R. Beagle. 2011. Despite the lack of corroborating ev- lations of a given species (see above), Historical ecology of the lower Santa Clara Riv- idence of natural occurrence of per- Jordan and Gilbert’s range descriptions er, Ventura River, and Oxnard Plain: an analysis sistent Chinook salmon populations in simply reported what they believed to of terrestrial, riverine, and coastal habitats. Rep. prep. for State Coastal Conservancy. A report of fresh waters south of Monterey Bay, be the full geographical extent encom- SFEI’s Historical Ecology Program, SFEI Publi- both during the 1880 expedition and passing all observations of the organ- cation 461, San Francisco Estuary Inst., Oakland, in the 138 years since,29 the assertion ism. Moreover, as will be seen in the Calif. (Avail. at http://www.sfei.org/documents/ historical-ecology-lower-santa-clara-river-ventu- made by Jordan that Chinook salmon next section, the poor understanding ra-river-and-oxnard-plain-analysis-terrest). of Pacific salmon life histories at the 25Some Chinook salmon juveniles exhibit Ventura County from 1879 to 1881 (Jennings, time likely led Jordan and Gilbert to stream-type life histories where they over-sum- 1997). However, Evermann did not begin his for- mer in stream habitats and migrate either dur- mal education with Jordan until 1881, when he believe that the freshwater and marine ing the following fall (typically to estuaries) or returned to (http://researchar- distributions of salmon were essential- spring. True “stream-type” life history is typical- chive.calacademy.org/research/library/special/ ly identical. ly most common in systems with snowmelt-dom- findaids/evermann.html; http://socialarchive.iath. inated hydrology. Where observed in rain-domi- .edu/ark:/99166/w6n301tx). State of Salmon Life nated systems of California and Oregon, yearling 28In discussing the range of Chinook salmon, Jor- smolts generally comprise a small fraction of to- dan and Gilbert (1881b) comment that the Ven- History Knowledge tal smolts produced (e.g., Sparkman, M. D., R. tura River is the southernmost river in coastal Park, L. Osborn, S. Holt, and M. Griffin. 2016. California that is “not muddy and alkaline at its Given the taxonomic uncertain- Lower Redwood Creek juvenile salmonid (smolt) mouth,” which suggests they had some informa- abundance project, 2004–2015 seasons. CDFW tion about putative salmon specific to the Ven- ty and confusion surrounding Pacif- Proj. 2a7, Fish. Restoration Grant Program (Proj. tura River. ic salmon and the rudimentary under- No. P1210322). Calif. Dep. Fish Wildl, N. Reg., 29 85 p.). There were documented occurrences of Chi- standing of species distributions that 26 nook salmon in both the Ventura and Santa Ynez prevailed in the late 1800’s, it should Anadromous O. mykiss do inhabit these wa- rivers in southern California; however, these were tersheds, likely because of greater plasticity in the result of plantings of hatchery fish (CFGC, migration and spawning timing (November– 1916). In modern times, occasional stray adults 30Interestingly, although Behnke’s (2002) narra- April), coupled with greater diversity of life his- have been reported in Big Creek on the Big Sur tive makes mention of Chinook salmon extending tories (e.g., stream-residency, variability in age at Coast (T. Williams, NMFS, Southwest Fisheries as far south as the Ventura River, his map depict- smoltification, and iteroparity). Science Center, personal commun., Sep., 2017), ing the native range of Chinook salmon extends 27Evermann apparently met Jordan in 1877 at as well as the lower reaches of streams in Ven- only as far south as the San Francisco Bay area and subsequently moved to tura and Orange counties (Alagona et al., 2012) and San Joaquin River basin.

12 Marine Fisheries Review not be surprising that understanding of Of course, we now know that Pacif- pressing our ignorance. In gener- the life histories of the salmon species ic salmon in the ocean may travel hun- al, it seems to be true that in those was even worse. Particularly relevant dreds or even thousands of miles from rivers and in those years when the to the question of historical ranges is their natal streams. Jordan dogmatical- spring run is greatest, the fall run that fact that Jordan was a staunch op- ly adhered to his view from 1881 un- is least to be depended on.” (Jor- ponent of what he termed the “parent til the early 1920’s, repeating it in nu- dan and Gilbert, 1881b) stream theory,” the idea that salmon merous publications (Jordan and Ev- have an innate ability to home to their ermann, 1896, 1922; Jordan, 1888, Virtually identical statements were natal streams to spawn. Jordan and 1903, 1904b, 1905, 1907a). Ironically, published by Jordan into the 1890’s Gilbert (1881b) wrote, it was Gilbert, his protégé and co-au- (e.g., Jordan, 1892a, 1894a, 1896), in- thor on the 1881 paper dismissing the dicating that the question remained “It is the prevailing impression home stream theory, who eventually unresolved. that the salmon have some special provided the first compelling scientif- Also in dispute among scientists was instinct which leads them to re- ic evidence in favor of it, demonstrat- the question of whether all salmon die turn to spawn in the same spawn- ing that fish from different populations after spawning. Suckley (1874) spoke ing grounds where they were orig- exhibited unique early growth patterns of spawned out Salmo scouleri (=O. inally hatched. We fail to find any on their scales (Dunn, 1996). gorbuscha) lingering in streams until evidence of this in the case of Pa- Jordan was eventually persuaded by January, February, and March before cific coast salmon, and we do not Gilbert’s careful work, as acknowl- suddenly disappearing in April, “prob- believe it to be true. It seems more edged in Jordan and Evermann (1923). ably returning by the spring floods to probable that the young salmon, Nevertheless, for the majority of his salt water, although the Indians say hatched in any river, mostly re- career, Jordan would have assumed that but few return to sea.” Converse- main in the ocean within a radi- any ocean-caught salmon had near- ly, in reference to Salmo proteus (also us of twenty, thirty or forty miles by origins. He and his contemporaries =O. gorbuscha), he stated that “ac- of its mouth. These, in their move- thus would have considered the south- cording to the natives of our coast, the ments about in the ocean, may ern boundary of marine and freshwa- hunch-back never returns to the sea af- come into contact with the cold ter spawning ranges of Pacific salmon ter spawning, but dies in freshwater.” waters of their parent rivers, or species to be identical, plus or minus In his description of Salmo canis (=O. perhaps any other river, at a con- a few tens of miles. The salient lesson keta), Suckley (1874) noted that “They siderable distance from the shore. is that, without concrete corroborat- [local Native Americans] say that most In the case of the quinnat and the ing evidence of freshwater occurrence of the individuals return to the seas af- blue-back, their “instinct” leads of Pacific salmon species, we should ter spawning, many more comparative- them to ascend these fresh waters, view the early freshwater spawning ly than do of the S. scouleri.” and in a majority of cases these range descriptions of Jordan, Gilbert, The debate regarding semelpari- waters will be those in which the and others with appropriate caution. ty in Pacific salmon continued into fishes in question were originally Other aspects of salmon life histo- the 1870’s. Hallock (1877), a sports- spawned.” ries were also poorly understood. For man, popular writer, and founder of Jordan’s belief that Pacific salmon example, Jordan and Gilbert were un- Forest and Stream magazine, implied at sea remained in close proximity to der the impression that variation in that Livingston Stone had definitively their natal rivers and that they did not run timing of species such as Chinook resolved the question in studies con- possess a homing instinct potentially salmon was environmentally induced, ducted on the McCloud River, quoting has significant implications when in- rather than a trait of a particular life- an uncited report in which Stone “set- terpreting his early species range de- history type. In 1881, they wrote the tles the question finally, and proved scriptions. Namely, it raises the pos- following: beyond a shadow of a doubt, that all sibility that had Jordan encountered of the thousands of Sacramento salm- or heard reports of salmon captured “High waters on any of these riv- on that spawned in the McCloud, not at sea, he may have assumed that they ers in the spring is always followed one in a hundred returned to the sea originated from a nearby stream or riv- by an increased run of salmon. The alive.”31 However, Stone (1880) later er. Indeed, this provides one plausible canners think, and this is probably responded that Hallock (and others32) explanation for the purported occur- true, that salmon which would not have run till later, are brought up rence of Chinook salmon in the Ven- 31Although I was unable to find the report to tura River, though as noted earlier the by the contact with the cold wa- which Hallock referred, there is text in the U.S. presence of occasional straying indi- ter. The cause of this effect of cold Commission on Fish and Fisheries report for 1873–74 and 1874–1875 that is consistent with viduals or misidentification of a large fresh water is not understood. We Hallock’s assertion (USFC, 1876: p. xxiii). O. mykiss in fresh water are plausible may call it an instinct of the salm- 32Hallock responded specifically to a letter by explanations as well. on, which is another way of ex- Horace Dunn, a salmon fishing enthusiast who

81(1) 13 had misinterpreted his findings. Spe- older than a five-pounder; and it freshwater residence was dictated pri- cifically, Stone argued that, while he would not be presuming too much marily by the distance a juvenile fish stood behind his findings of semelpari- to assert that a Salmon of that size must travel to reach the ocean, rather ty for Chinook salmon in the McCloud must have made more than one trip than being a trait of a particular spe- River, he also believed that “the Sac- to the sea. While in the lake they cies or life history type: ramento salmon which spawn near the do not appear to eat anything, and sea are, many of them, able to return the stomachs of several which I ex- “As soon as the yolk-sac has been to salt water.” In short, Stone appar- amined were entirely empty. I can- absorbed, the fish is large enough ently believed that the ability of fish to not understand how they get rid of to swim and eat, and is known as survive spawning was a function of the their long hooked nose and hump.” a fry. When this stage is reached, distance they needed to travel to their the young of the Pacific spawning streams, presumably be- Four years later, Jordan (1888) re- begin their migration to salt water. cause fish with longer migrations did published his 1881 descriptions, again They float down-stream backward, not have sufficient energy reserves or asserting that large spring-run “quin- for ease in breathing and catching were too battered by their journeys to nat” salmon in the Columbia and Sac- food....The age at which young survive and successfully return to the ramento rivers likely survived spawn- Pacific salmon reach the ocean de- ocean. ing from the previous year. The belief pends, of course, very much on the Jordan and Gilbert (1881b) shared of Jordan and others of this era that distance they have to travel. Those Stone’s view that some salmon sur- Pacific salmon could be repeat spawn- that hatch in the head-waters of vived to spawn multiple times. In ref- ers may have contributed to pub- the Columbia are over fifteen hun- erence to spring-run Chinook salmon lished accounts of salmon “returning dred miles from the ocean, and, as in the Columbia and Sacramento riv- to the sea in spring” following spawn- they fight the current more strong- ers, they wrote the following: ing (e.g., CFC, 1872; Hallock, 1877), ly with age, are probably nearly though misidentification of steelhead a year en route; while those that “It is questioned whether these kelts as salmon was likely also a con- hatch in Olema creek, near San large fishes are: (a.) Those which, tributing factor. Francisco, may reach Tomales Bay of the same age, have grown more Even less well understood during in a day’s travel.” rapidly; (b.) Those which are older this era were the juvenile life stages but have, for some reason, failed of Pacific salmonids. In their writings Rutter’s description implies that to spawn; or (c.) Those which have from 1881 and 1894, Jordan and Gil- the downstream movement of salm- survived one or more spawning bert rarely discuss the early life history on smolts was primarily a passive pro- seasons. All of these origins may of salmon or steelhead, and what little cess, rather than active directional mi- be possible in individual cases; we is written tends to be vague. Regarding gration. That there were significant are, however, of the opinion that salmon, Jordan and Gilbert (1881b) differences in the typical age and tim- the majority of these large fish are wrote that “The manner of spawning ing of outmigration among species or those which have hitherto run in is probably similar for all the species, life history types went largely unrec- the fall and so may have survived but we have no data except for any ognized. In fact, it was not until the the spawning season previous.” but the quinnat….The young hatch in work of Chamberlain (1907) in Alaska about sixty days, and most of them re- that these life history differences be- Jordan (1884a) perpetuated the be- turn to the ocean during the high water gan to be fully appreciated. In sections lief that salmon could spawn more of spring,” the latter sentence being re- that follow, we will see some exam- than once, quoting the following de- peated in subsequent publications (Jor- ples of where this poor understanding scription of sockeye salmon written by dan, 1888, 1892a, 1894a, 1896). Al- of juvenile life histories contributed to Captain Charles Bendire: though one cannot be completely cer- misidentification of species and where tain, Jordan and Gilbert’s language those misidentifications contributed to “The only thing as yet which I suggests that they believed the major- species freshwater range descriptions can’t understand is, how do they ity young salmon went to sea in their in California that are either inaccurate get rid of the hooked nose and first year of life, regardless of species. or, at best, highly dubious. hump after going back to salt wa- Hence, they were apparently unaware ter? They surely can’t all die after of life history types where young What’s in a Name? spawning, as sometimes one that spend a year or more in fresh waters. Although questions about the fresh- weighs as much as ten pounds is Cloudsley Rutter (1903), a former water ranges of various Pacific salm- caught, and this fish is certainly student of Jordan’s who had, by the on were not definitively answered by early 1900’s, achieved considerable the 1880 Pacific Coast Expedition, (erroneously) argued that “a very large portion of stature as a fishery biologist (Jennings, Jordan and Gilbert did succeed in re- the [Chinook] run return to sea again.” 2014), believed that the period of solving the long-running confusion re-

14 Marine Fisheries Review garding the number of Pacific salm- head (e.g., Anonymous 1885a, 1885b, 1885d). Reports of hundreds of “trout” on species in western North America. 1885c, 1886). Even Jordan occasional- being caught in local streams during Jordan and Gilbert (1881a) contained ly slipped into this casual usage. When the summer almost certainly includ- descriptions of the five Pacific salm- describing the fisheries of Carmel, Ca- ed some juvenile salmon, and those on species known today, as well as a lif., Jordan (1887) wrote the following: of “salmon” being captured during table of meristic traits of each species the spring are almost assuredly steel- including average counts of gill rakers, “There is no regular fishing done head. Wrote one Santa Cruz newspa- branchiostegal rays, anal fin rays, py- at Carmelo (sic). In the river of per (Anonymous, 1890): loric caeca, and scales in longitudinal that name, a great many trout are series.33 With one exception, the av- taken and sold in Monterey at 12 ½ “The State Board of Fish Com- erage values of various meristic traits cents a pound. In the spring salm- missioners has decided to for- given by Jordan and Gilbert fall within on ascend the river and are taken ward to the authorities at Wash- the currently accepted ranges for each by the farmers. In the summer the ington one of the fish commonly species.34 Hence, Jordan and Gilbert water in the river is low and a bar known as “salmon trout” and find (1881a) clearly set the table for a bet- is formed across its mouth, caus- out whether it is a salmon or trout. ter understanding of the distributions ing many young salmon to be- This fish has been taken from the of these species along the California come landlocked. These are easily water in large quantities at all sea- Coast. caught by the farmers and whalers sons of the year and the law has Yet, despite this significant advance, at Carmelo (sic)” been evaded by the fishermen progress on this front was slow in swearing that the fish were young coming. Exploration of the fish faunas The springtime ascent of these sup- salmon in the closed trout season of California’s coastal watersheds re- posed “salmon” leaves little doubt that and full-grown trout in the closed mained limited for the next 15 years. the fish to which Jordan refers were salmon season.” Scientific publications from the 1880’s steelhead, as the rain-dominated hy- Hence anglers (wittingly or not) and early 1890’s added but few new drology of the Carmel River would not were routinely breaking the law, and collection localities for Pacific salmon have allowed access to spring-run Chi- the newspapers may have been loath to in California (Supplemental Table S1; nook salmon. Whether Jordan (1887) acknowledge wrongdoing by local cit- Evermann and Clark, 1931), and most intentionally used the term “salmon” izens. Jordan himself was on occasion descriptions of species ranges from to describe steelhead or was simply shipped specimens for species identi- this era (e.g., Eigenmann, 1890; Jor- adopting the terminology of local fish- fication in order to determine if they dan, 1894a) tend simply to repeat the ermen is unclear, but the latter expla- had been captured in violation of fish- range descriptions published in Jordan nation would seem the more likely, as ing regulations (Anonymous, 1885e). and Gilbert (1881a). I can find no other writings where he Perhaps no other common name has Further, while Jordan and Gilbert did not discriminate between salmon generated more confusion than the (1881a) clarified the formal taxono- and steelhead. term “dog salmon.” Although in mod- my of Pacific salmon, it is evident that The term “salmon trout” lends to ern times, “dog salmon” has generally confusion surrounding common names similar confusion. In general, salm- been reserved for O. keta, throughout of Pacific salmon and trout continued on trout was applied to anadromous the 1880’s and well into the 1900’s, the unabated for decades in both scientif- forms of O. mykiss, O.clarkii, and even term was routinely applied by scien- ic and popular literature. In writings of Salvelinus malma (Jordan and Gilbert, tists and fishermen to the males of all the 1800’s, the term “salmon” itself is 1881a; Jordan and Evermann, 1896), Pacific salmon species during the fall highly ambiguous, as it was commonly as well as lake-dwelling interior pop- spawning season (Jordan and Gilbert, used to describe any large, ocean-run ulations of cutthroat trout (O. clarkii 1881b), no doubt due to the elongation salmonid. Many newspaper accounts henshawi and O. clarkii utah) with no of the jaws and development of large from the 1800’s refer to “salmon” re- access to the sea (e.g., Fremont, 1856). canine-like front teeth that occurs as turning to coastal rivers when, based However, references to “salmon trout” males approach reproductive maturi- on the timing of these observations, in the popular literature and newspa- ty. In his description of O. keta, Jordan these fish were in all likelihood steel- pers of this era should be interpret- (1884a) wrote the following: ed with caution, as the term may have 33Gilbert’s journal from the expedition contains a been misapplied, both accidentally and “This species, during the period of handwritten table of meristic traits that matches purposefully. its run in the fall, generally goes by that published in Jordan and Gilbert (1881a). This table was apparently completed while the expedi- California fish and game laws were the name of “Dog salmon,” under tion was in Puget Sound, near its end. such that it was illegal to fish for which name the males of the Sil- 34 Jordan and Gilbert listed coho salmon as aver- “salmon” in streams during certain ver Salmon [O. kisutch], and even aging 10 gill rakers on the upper arch. The range is now given as 6 to 9 (Scott and Crossman, 1973; summer and winter months, and for the Quinnat [O. tshawytscha], are McPhail and Lindsey, 1970). “trout” during the winter (Anonymous, often confounded with it.”

81(1) 15 This same paper also lists “Dog tencies in use of the term “dog salm- a window that certainly includes steel- Salmon” as a common name for O. on” into the 1900’s. In fact, even 50 head. A San Francisco-based newspa- gorbuscha, and in a second paper pub- years later, Shapovalov and Taft (1954) per (Anonymous, 1889) reporting on lished the same year, Jordan (1884b) in their classic study of Waddell Creek a meeting of a sportsmen’s club stated lists “Dog Salmon” as a common still felt compelled to offer the follow- the following: name for O. nerka (but curiously, not ing clarification: for the other four Pacific salmon spe- “A member moved to invite the cies). Thus, between those two publi- “One popular misconception that Fish Commissioner to the club cations, Jordan noted that “dog salm- has existed among the various meeting. ‘My reason,’ he said, ‘is on” had been applied to all five spe- parts of the Pacific Coast is that that they can give us some valu- cies of Pacific salmon. the hook-nosed salmon, called able information on the different Other scientists of the era added to ‘dog salmon’ by local residents kinds of fish. Now, I have been the confusion. William N. Lockington, form a distinct species. Such fish fishing for years, but for the life a well-known zoologist and curator are simply males whose snouts of me I do not know what a salm- of the California Academy of Scienc- have become hooked and elon- on trout actually is. Some call it a es museum from 1875–1881, switched gated during the spawning sea- salmon trout, some call it a silver the common names of O. keta and O. son. This phenomenon takes place salmon, and some call it a steel- kisutch, attributing the names “Sil- to a greater or lesser extent in all headed salmon.’ ” verside Salmon” and “Cohoe salm- of the species of Pacific salmons on” to O. keta, and “Dog Salmon” to and to some extent in steelhead. The quote indicates that some lo- O. kisutch (Lockington, 1880). Two A distinct species of salmon, the cal fishermen referred to steelhead as years later, Lockington (1882) partial- chum salmon (Oncorhynchus “silver salmon.” The terms “silvers” or ly corrected his mistake, assigning the keta), is sometimes also known as “silver salmon” were also historical- common name “silver salmon” to O. dog salmon, but occurs compara- ly applied to chum salmon, O. keta, in kisutch, but listing O. gorbuscha as the tively infrequently in California. the Yukon River (Gilbert, 1922); sock- “dog salmon.” Common names applied to the sil- eye salmon, O. nerka, in the Columbia Rutter (1902) describes experiments ver salmon are jack salmon (ap- (Bendire, 1881); and spring-run Chi- in which Chinook salmon were intro- plied seasonally to young males), nook, O. tshawytscha, in the Colum- duced into Lagunitas Creek in Marin dog salmon, or hookbill (applied bia (Murphy, 1879), Klamath (Snyder, County (then known as Paper-mill to males with hooked snouts and 1931), and Battle Creek (Sacramen- Creek) to determine if fish returned to red sides), coho, and silversides.” to River) watersheds (Anonymous, their natal stream to spawn. Laguni- 1899). Jordan (1891) pointed out that tas Creek was selected by Rutter spe- Often times, common names were the term “silver salmon” was “care- cifically because it was believed not to the product of regional usage. Use of lessly” applied by fishermen to juve- support Chinook salmon. Wrote Rut- the term “silver salmon” is illustrative. niles of all salmon species. ter (1902), “Paper-mill Creek is not Although now most commonly ap- An excellent illustration of the con- suitable for Quinnat salmon, being en- plied to coho salmon, O. kisutch, silver fusion in salmonid nomenclature sur- tirely too small, but it is frequented by salmon was applied to other species as rounds the activities of a small coun- Dog salmon and Steelheads.” Although well. For example, Gibbons (1876) de- ty-owned hatchery in the Santa Cruz there have been occasional reports of scribed what he believed to be a new Mountains during the early 1900’s. chum salmon in Lagunitas Creek over species of salmonid he named Sal- In 1906, a local newspaper reported the years, the “Dog salmon” to which mo mendocinensis (=O. mykiss iride- that Hatchery Manager Frank Sheb- Rutter referred were almost certainly us), noting that “When first caught, ley planned to import “silver salmon” coho salmon, as this species has per- the females are of a bright silver col- from Baker Lake in the Washington sistently occupied the stream for more or; hence, some call them ‘silver salm- Cascades and that this would “give us than a century (Eigenmann, 1890).35 on.’” He goes on to state that the another valuable game fish,” implying Thus, even among leaders in the sci- spawning season for this species com- that the species was not found in local entific community, we find inconsis- menced “usually in the latter part of waters (Anonymous, 1905). Howev- March…”, consistent with Evermann er, just a few months later, another lo- 35Eigenmann (1890) reported that juvenile coho salmon were collected from Lagunitas Creek by and Clark’s (1931) interpretation that cal paper quotes Superintendent Sheb- Mr. Chas. Ohm in March 1890. Additionally, ju- these were steelhead. Likewise, Wake- ley saying it would be a “good idea to veniles collected from the stream by Snyder and man (1870: in CFC, 1872) reported get eggs from dog salmon which run his assistants in 1909 currently reside in the U.S. National Museum collection (USNM 75327). “salmon and silver salmon” from 2 to up Scotts Creek [a nearby stream] and These specimens, initially cataloged as Salmo iri- 30 pounds being taken by fishermen hatch out a quantity. The dog salm- dea, were only recently determined through ge- netic analysis to be coho salmon. See Spence et in Pescadero and San Gregorio creeks on come up the creek in the fall, af- al. (2011) for details. during the months October to March, ter waiting in the lagoon for the fall

16 Marine Fisheries Review rains to come down and raise the wa- quinnat salmon was used by Jordan River region (Jordan, 1884a). Simi- ter” (Anonymous, 1906). In all likeli- and others not only in specific refer- larly, Gilbert (1922) reported that the hood, the local “dog salmon” that She- ence to O. tshawytscha, but also as a name “chinook” was applied to O. bley referred to were O. kisutch and general name for all five of the Pacif- kisutch in the Yukon Basin. not O. keta, as evidenced by the fact ic salmon species. This is evidenced Lewis and Clark described a fish that the article comments on the high in the following quote from Jordan they called the “White Salmon Trout” quality of “dog salmon” as a food fish, (1892a): from the Columbia River and its trib- stating that “dog salmon are the rich- utaries, which subsequent authors in- est fish that swim the rivers. They have “They [the Pacific Coast salm- terpreted to be coho salmon (Coues, more oil, and more and tastier flesh on] have therefore been placed 1893; Burroughs, 1961; Cutright, than any other salmon.” in another genus known as On- 1969); however, more critical analy- O. keta have long been recognized corhynchus. For the lack of any sis of the timing of these observations as the least desirable food fish of the other common name they are al- (March and April), as well as meris- five Pacific salmon species, a fact rec- ways spoken of as and will always tic characteristics, indicates these were ognized even in Jordan’s earliest writ- be canned, as long as the canning most likely steelhead (Trotter and Bis- ings (see e.g., Jordan and Gilbert, industry lasts, under the name of son, 1988; Butler, 2004).37 1881a). I suspect that Shebley was Salmon. The Chinook name, Quin- Since the 1880’s, “blue-back” has simply unaware that the “silver salm- nat, was early applied to them, and been used to describe sockeye salm- on” he intended to bring down from if we feel the need of some oth- on (Jordan, 1880); however, the term Washington and the fish known local- er name to distinguish them from has also been applied to both steelhead ly as “dog salmon” were the same spe- real salmon [Atlantic salmon, Sal- and coastal cutthroat trout. This con- cies, O. kisutch.36 As will be seen mo- mo salar] we may call the Pacif- fusion may have led Jordan and Ev- mentarily, the ambiguous use of “dog ic Coast salmon Quinnat, or Quin- ermann (1896) to write that sockeye salmon” appears to have played a key nat Salmon. These species all live salmon once occurred in the Klam- role in some early species identifica- in the ocean, ascend the rivers ath River (Snyder, 1931). And finally, tions that tangibly influenced Jordan’s in spring and summer, spawn in the terms “hookbill” or “hook-nosed” range descriptions of both coho and fresh water in the fall, the young, salmon were used for pink salmon chum salmon. as soon as they are able to swim, (Suckley, 1874; Kerry, 1874), coho Another nomenclatural inconsisten- floating tail foremost down riv- salmon (Shapovalov and Taft, 1954), cy and potential source of ambiguity er and growing rapidly as soon as and Chinook salmon (Snyder, 1931). is the term “quinnat salmon.” Quinnat they reach the ocean and the pecu- In summary, the use of various non- salmon is generally taken by modern liar ocean food. There are five spe- definitive common names continued scientists to be a synonym for Chi- cies of these Quinnats, which will for decades, despite the fact that Jor- nook salmon, and, in fact, Jordan’s be described farther on [emphasis dan and Gilbert (1881a) had accurate- earliest writings on Pacific salm- added].” ly described the five Pacific salmon on assigned the scientific name “On- species and their diagnostic character- corhynchus quinnat” or “Salmo quin- In another paper, Jordan (1894a:129) istics.38 This often makes it difficult to nat” to what is now recognized as O. wrote “all the so-called salmon of the interpret writings from the late 1800’s. tshawytscha (Jordan, 1878a, 1878b, Pacific Coast, all the Quinnats or spe- Accounts from the popular literature 1884b). However, in the 19th century, cies of Oncorhynchus, have an in- and newspaper articles are particular- creased number of rays in the anal fin, ly dubious, but, as will be seen in the 36Other evidence supports the interpretation that the “dog salmon” Shebley referred to were not O. from fourteen to twenty, not counting next section, this confusion was often keta. An 1887 news article notes the discovery of the stubs or rudiments in front of the evident in the scientific literature as a “Piscatorial Curiosity” in the San Lorenzo Riv- first ray.” These examples illustrate well. er, which upon examination of an expert was de- termined to be O. keta (Anonymous, 1887). The that the potential exists to misinterpret expert noted that “This fish is a stranger to San- early references to “Quinnat salmon” ta Cruz waters, none of the kind to my knowl- as specific to Chinook salmon when, edge ever being seen here before.” Nearly thir- 37 ty years later, Scofield (1916) again document- in fact, it may be a generic reference The naming of the White Salmon River in ed occurrence of three O. keta adults in the San to Pacific salmon as a group. Washington is traceable to Lewis and Clark’s Lorenzo River, stating that “The dog salmon has “white salmon trout.” For a thorough discus- been reported from the Sacramento River also, A few additional common names sion, see McMillan, B. 2017. Lewis and Clark’s but never before from a point as far south as the have the potential to cause confusion white salmon trout: coho salmon or steelhead? 200 years of getting it wrong. Part III: Fort Clat- San Lorenzo River.” The fact that these obser- in the historical literature. The term vations of O. keta in the Santa Cruz Mountain sop. The Conservation Angler (avail. at https:// were viewed as rare, even unprecedented events, “white salmon” was used as a common theconservationangler.wordpress.com/category/ argues against Shebley’s Scott Creek “dog salm- name for O. chouicha (=O. tshawyts- essays-by-bill-mcmillan/). on” being O. keta, particularly since Shebley be- 38It was not until the late 20th century that the lieved them abundant enough to supply the hatch- cha; Jordan and Gilbert, 1881a), as taxonomic status of steelhead was settled (Stear- ery with a ready source of eggs. well as O. kisutch in the Columbia ley and Smith, 1993).

81(1) 17 Table 4.—Sampling localities and salmonid records from the 1895 Carmel River Expedition from the California Academy of Sciences collection database.

CAS Collection Stanford Revised cat. no. date Collection location1 Collectors ledger ID species ID Disposition SU 4679 not given Carmel River (at mouth) Scofield & Rutter Salmo irideus O. mykiss irideus discarded SU 4672 not given Carmel River Rutter & Pierson Salmo irideus O. mykiss irideus discarded SU 4675 not given Salinas2 River (Soledad) Rutter & Scofield Salmo irideus O. mykiss irideus extant SU 4680 not given Soquel Creek Scofield & Seale Salmo irideus O. mykiss irideus discarded SU 4670 not given San Lorenzo River3 Rutter & Pierson Salmo gairdneri irideus O. mykiss irideus extant SU 4673 not given Wilder Creek Scofield & Seale Salmo gairdneri irideus O. mykiss irideus extant SU 4802 not given Liddell Creek Rutter & Scofield Salmo irideus O. mykiss irideus transferred SU 4799 not given Liddell Creek Rutter & Scofield Salmo irideus O. mykiss irideus extant SU 4783 6/6/1895 Laguna Creek4 Rutter & Scofield Salmo irideus O. mykiss irideus extant SU 4798 not given San Vicente Creek5 Rutter & Scofield Salmo irideus O. mykiss irideus with 4665 SU 4665 6/6/1895 San Vicente Creek5 Rutter & Scofield Salmo irideus O. mykiss irideus extant SU 4685 not given San Vicente Creek6 Rutter & Scofield Oncorhynchus keta O. kisutch7 extant SU 4674 not given Scott Creek Rutter & Seale Salmo irideus O. mykiss irideus extant SU 4797 6/5/1895 Scott Creek Rutter & Seale Oncorhynchus tschawytscha O. kisutch extant SU 4756 not given Waddell Creek Salmo irideus O. mykiss irideus discarded SU 4671 not given Waddell Creek Rutter, Seale, Scofield Salmo irideus O. mykiss irideus transferred to 4796 SU 4796 not given Waddell Creek Rutter & Scofield Salmo irideus O. mykiss irideus extant SU 4667 6/5/1895 Waddell Creek Rutter & Scofield Oncorhynchus keta O. kisutch extant SU 4666 not given Gazos Creek8 Rutter & Pierson Salmo irideus O. mykiss irideus extant SU 4686 not given Gazos Creek8 Rutter & Pierson Oncorhynchus keta O. kisutch extant SU 4668 not given Pescadero Creek (Harrison) Rutter & Scofield Salmo irideus O. mykiss irideus extant SU 4669 not given San Gregorio Creek9 Scofield & party Salmo irideus O. mykiss irideus extant 1Modern spellings are used. Alternative spellings found in Stanford ledger are footnoted. 2Misspelled as Salinal River 3Listed as San Lorenzo Creek 4Misspelled as Lagoon Creek in ledger and Lagoona on label 5Misspelled as San Macento Creek 6Misspelled as San Vicinto Creek 7One specimen in the jar has been tentatively identified as O. tshawytscha. 8Misspelled as Gassus Creek 9No creek name given, but locality listed as San Gregorio, La Honda

1895–1910: Systematic of the Zoology Department (Hubbs, veys constituted the first systematic Exploration Begins 1964). Snyder, then an undergraduate, sampling of fish faunas in coastal wa- likewise followed Jordan from Indiana tersheds of California and Oregon, and More systematic exploration of Cal- to Stanford, completing his Bachelor as such, were the first scientific collec- ifornia’s coastal watersheds did not be- of Arts degree in 1897 and his Master tions with the potential to elucidate the gin until several years after Jordan ar- of Arts degree in 1899, with Gilbert freshwater spawning ranges of vari- rived at the newly founded Stanford serving as his major professor (Brit- ous salmonid species. However, read- University in 1891 and assumed the tain and Jennings, 2008). Both Gilbert ers should recognize that these sur- position of University President (Ever- and Snyder played key roles in the ex- veys were hardly exhaustive. Collec- mann and Clark, 1931). In his first five ploration of California’s coastal wa- tion records indicate that surveys most years at Stanford, Jordan spent much ters, as did others under their direc- often took place at one or two loca- of his time handling the administrative tion, including Cloudsley Rutter, Nor- tions within a watershed, and that sur- demands of running the university, in- man Scofield, Alvin Seale, and Charles vey sites on consecutive days were of- cluding averting financial collapse of Pierson. ten 10 miles or more apart. Given that the institution following the death of Jordan and others at Stanford re- the mode of transportation was prob- its founder in 1893 sumed field work in zoology in late ably horseback, the actual sampling (Jordan, 1922). Any free time Jordan 1894 (Jordan, 1922:526). Over the time at each location was likely fair- had for scientific work during the ear- next two decades, the Stanford faculty ly short. More importantly, despite ly 1890’s was devoted largely to prep- and staff led many influential investi- the credentials of the expedition lead- aration of the four-volume treatise The gations into the fish faunas of Mexico, ers, the misidentification of salmo- Fishes of North and Middle America South America, Hawaii, Japan, Russia, nid fishes on these surveys appears to with Barton Warren Evermann. This and other faraway destinations (Böhl- have occurred, and this likely contrib- book gave descriptions of some 3,127 ke, 1953); however, it is the local ex- uted substantially to confusion regard- species of fishes found north of the plorations of California and Oregon ing the freshwater distribution of cer- Isthmus of Panama. The first of four that are germane to the question of Pa- tain species of salmon. volumes was published in 1896, and it cific salmon distributions. was only during the latter stages of its Two expeditions in coastal Califor- The Carmel River preparation that Jordan resumed field nia are particularly noteworthy: the Expedition (1895) explorations (Jordan, 1922:524). Carmel River Expedition of 1895, and The 1895 Carmel River expedition Jordan brought Gilbert with him the Northern California/Oregon Expe- was conducted by Rutter, Scofield, from Indiana, appointing him Chair dition of 1897 and 1899. These sur- Seale, and Pierson, in June 1895. The

18 Marine Fisheries Review foursome sampled streams in water- was originally identified as O. keta, (1907) account of the expedition does sheds from the Carmel River in Mon- but may be O. tshawytscha.41 All of not provide details of the route, this terey County, north to San Gregorio these redeterminations were apparent- can be reconstructed to a large degree Creek in San Mateo County, roughly a ly made while the specimens were still from records in the CAS and USNM 140 km stretch of coastline (Table 4). in the possession of Stanford Univer- ichthyological collections. The 1897 The expedition involved the collection sity, as indicated by labels that are still expedition began in Napa County in of fishes from freshwater and possi- present in the sample jars housed at late May, moved coastward to the Rus- bly lagoon habitats in at least 15 coast- the CAS.42 sian and Gualala rivers (Sonoma Co.) al streams. Importantly, the region These Carmel Expedition collec- in June, and then proceeded northward covered by this expedition included tions are noteworthy in that they rep- as far as the Smith River near the Or- streams of the Santa Cruz Mountains, resent the first collection of juvenile egon border. which may very well have marked the coho salmon from the Santa Cruz The expedition then turned inland, southern limits of the historical spawn- Mountain region, which lies to the visiting Cottonwood Creek and Shas- ing distributions of both coho and Chi- south of San Francisco, previously ta River in the upper Klamath Basin in nook salmon. Specimens obtained dur- listed by Jordan and Gilbert (1881a) late July43 (Table 5). Overall, this first ing this expedition were incorporated as the southern limit of the species’ segment of the Northern California– into the Stanford ichthyological collec- known distribution. That juvenile coho Oregon Expedition covered approx- tion, and many still reside at the Cal- salmon were observed in four different imately 445 km of the northern Cali- ifornia Academy of Sciences (CAS), watersheds provides strong evidence fornia Coast, including all of the major which took possession of the Stanford that the species regularly occupied coastal basins (e.g., Russian, Guala- collection in 1969.39 These provide a streams in the region and that this was la, Garcia, Navarro, Russian, Albion, striking example of how the correct not an isolated case of straying. Yet be- Noyo, Ten Mile, Mattole, Bear, Eel, identification of salmon species, espe- cause the initial species identifications Mad, Klamath, and Smith rivers) and cially juveniles, continued to be elu- were incorrect, the observations were several smaller basins as well. The ex- sive even well after Jordan and Gilbert not immediately recognized as an ex- pedition returned to Stanford in Sep- (1881a) published descriptions of the tension of the published range of coho tember 1897, and the study resumed in diagnostic characteristics of these spe- salmon. Rather, the misidentification late July 1899, this time with Snyder cies. of coho salmon as chum salmon likely leading a group of three other Stanford Based on the Stanford museum led- explains changes in the range descrip- students.44 This extension of the sur- ger and collection database, the Car- tion of the latter species in the writings vey included interior basins of south- mel River expedition reported collect- of Jordan (see below). ern Oregon, the Willamette Basin, as ing Salmo irideus (=O. mykiss iride- well as coastal watersheds as far north us) from 13 of the streams and rivers The Oregon and as the Nehalem River (Supplemental sampled in the Central Coast region40 Northern California Coast Table S5). (Table 4). The original Stanford ledger Expedition (1897–1899) Snyder (1907) published an exten- entries indicate that, in addition to “S. The second important investigation sive table detailing the species of fish irideus,” two other salmonid species of coastal watersheds in the northern collected at each sampling location. were collected. These include spec- California and Oregon was led by Gil- Regrettably, Snyder purposely omitted imens initially identified as O. keta bert and Snyder during the summers salmon and trout species from this ta- from Waddell, Gazos, and San Vicente of 1897 and 1899. Although Snyder’s ble, though he did not offer a reason creeks and specimens identified as O. for doing so. However, in the body of tshawytscha from Scott Creek. How- 41The CAS database record for SU 4685 iden- the paper, he included narratives of all ever, all of these putative O. keta and tifies the specimens asO. kisutch; however, the species encountered, including three “Specimen remarks” column indicates that “one O. tshawytscha specimens have sub- spec. is a chinook.” This determination was made salmon species, as well as both steel- sequently been determined to be O. by CAS staff (D. Catania, CAS, personal com- head and cutthroat trout. In these nar- mun., Nov., 2004), though given the small size kisutch (Adams et al., 2007, and Spen- of San Vicente Creek, the redetermination may be ratives, we find strong evidence that ce et al., 2011 provide details), with questionable. Unfortunately, stomachs from all Gilbert and Snyder misidentified ju- the possible exception of one speci- of the 1895 specimens were removed for dietary venile salmonids during these surveys, analysis; thus, the most useful meristic charac- men from San Vicente Creek, which teristic for discriminating between coho and Chi- even though accurate descriptions of nook salmon—numbers of pyloric caeca—can- characters used to discriminate among not be determined. The redeterminations of most 39See http://www.calacademy.org/scientists/ich- specimens originally identified asO. keta as O. thyology-about. kisutch is consistent with the time of collection 43Collections attributed to the 1897 expedition 40Specimens from four of these localities have (June), which is after most southern chum popu- were also made in southern Oregon; however, since been discarded from the CAS collection; lations would have emigrated to the ocean. these collections were actually made by Frank however, I have personally examined specimens 42The SU 4667 and SU 4797 jars both have Stan- Cramer and Keinosuke Otaki (Snyder, 1907). from 7 of the remaining 9 localities and have ford University labels identifying Oncorhynchus 44Snyder was either finishing or had just finished confirmed that they areO. mykiss. kisutch as the species. his master’s degree.

81(1) 19 Table 5. —Sampling localities and salmonid records from the California portion of the 1897/1899 Northern California/Oregon Expedition led by Gilbert and Snyder, as deter- mined from the California Academy of Sciences (CAS) and U.S. National Museum (USNM) collection database records.

Date Location Watershed CAS(SU)1/USNM Collection records Salmonids reported Collector 5/18 1897 Walker Cr. Walker Cr. USNM 75406 Snyder 5/26/1897 Conn Cr./Napa R. Napa R. SU 60339, 15970/USNM 58318, 58314, 59827 O. mykiss gairdneri2 Gilbert 5/27/1897 Putah Cr. (near Monticello) Sacramento R. SU 37851, 23966 Gilbert et al. 5/28/1897 Berryessa Cr. Sacramento R. USNM 58145, 58123 Gilbert 5/29/1897 Aetna Springs (Putah Cr.) Sacramento R. SU 15975 Gilbert et al. 5/30/1897 Napa Cr. (Calistoga) Napa R. SU 24642 Gilbert 5/31/1897 Napa Cr. (Calistoga) Napa R. SU 24282, 54910 O. mykiss gairdneri Gilbert et al. 6/1/1897 Knights Valley Cr.3 (near Kellog) Russian R. USNM 62304 Gilbert 6/2/1897 Dry Cr. (near Healdsburg) Russian R. SU 37013, 66735, 66736, 22479/USNM 58139 Gilbert et al. 6/3/1897 Russian R. (near Healdsburg) Russian R. SU 28768, 37014 Gilbert 6/8/1897 Gualala R. (at jct. of Wheatfield Fork) Gualala R. SU 14903 Gilbert et al. 6/8/1897 Gualala R. (jc.t of Gualala and N. Fk) Gualala R. USNM 58328 Gilbert 6/10/1897 Garcia R. (5 mi. above mouth) Garcia R. SU 37977/USNM 62441 O. mykiss iridea Gilbert 6/12/1897 Garcia R. (1.5 mi. above mouth) Garcia R. SU 24111/USNM 58379 Gilbert et al. 6/13/1897 Navarro R. (4 mi. above mouth) Navarro R. SU 14902, 59766, 54835/USNM 62298 O. mykiss gairdneri Gilbert et al. 6/14/1897 Navarro R. (near Philo) Navarro R. SU 22488/USNM 58238, 74477 Gilbert et al./Snyder 6/17/1897 Russian R. (at Ukiah) Russian R. SU 15976 Gilbert et al. 6/22/1897 Albion R. (2 mi. below Comptsche) Albion R. SU 58573, 38016 Gilbert et al. 6/24/1897 Noyo R. Noyo R. SU 38011, 54863, 548704 O. tshawytscha, O. kisutch Gilbert et al. 6/25/1897 Ten Mile R. Ten Mile R. SU 37745, 20138 Gilbert et al. 6/26/1897 Cottoneva Cr.5 (2 mi. above Rockport) Cottaneva Cr. SU 54805 O. mykiss gairdneri Gilbert 6/27/1897 Usal Cr. Usal Cr. SU 62102, 59858, 59830, 59855, 60049/USNM 62299 Oncorhynchus spp., Salmo spp.6 C.H. Gilbert 6/30/1897 South Fk. Eel R. (at Garberville) Eel R. SU 24661/USNM 58277, 58163 Gilbert 7/1/1897 South Fk. Eel R. (at Myers Flat) Eel R. SU 54909/USNM 75333 O. mykiss gairdneri Gilbert 7/3/1897 Van Duzen R.7 Eel R. SU 60045 O. mykiss gairdneri Gilbert 7/5/1897 Mattole R. (at Petrolia) Mattole R. SU 37993, 60046 Salmo spp.8 Gilbert 7/6/1897 Bear R. (at Capetown) Bear R. SU 54810/USNM 58188 O. mykiss gairdneri Gilbert 7/7/1897 Mad R. (4 mi. above mouth) Mad R. SU 9861, 37047, 54811/USNM 126870, 61579 O. mykiss gairdneri Gilbert et al. 7/8/1897 Maple Cr. Maple Cr. SU 24114 Gilbert 7/8/1897 Little R. Little R. SU 5980 Gilbert et al. 7/9/1897 Redwood Cr. (at Orick) Redwood Cr. SU 67289, 37988, 26012 O. mykiss iridea Snyder & Gilbert et al. 7/10/1897 Klamath R. (at mouth) Klamath R. USNM 58081 Gilbert 7/12/1897 Klamath R. (at mouth) Klamath R. SU 37021, 37799, 600739 Salmo spp.8 Gilbert et al. 7/12/1897 Hunter Cr. Klamath R. SU 26011, CAS 225431 O. mykiss iridea Snyder 7/14/1897 Smith R. (near mouth) Smith R. SU 60379/USNM 62301 Oncorhynchus spp. C.H. Gilbert 7/16/1897 Smith R. (at Gasquet) Smith R. SU 41882 O. mykiss iridea C.H. Gilbert et al. 7/23/1897 Cottonwood Cr. Klamath R. SU 59847, 37798, 54907 O. mykiss gairdneri C.H. Gilbert et al. 7/24/1897 Shasta R. Klamath R. SU 60191, 31852, 59691, 54809, 54844 O. mykiss gairdneri, O. tshawytscha C.H. Gilbert 1Collections from the CAS database that were acquired from Stanford University retain their SU record numbers. Specimen numbers in bold are salmonids reported for the locality. 2All specimens identified in CAS and USNM database as O. mykiss gairdneri are likely O. mykiss irideus, as defined by Behnke, 2002. 3Currently known as Redwood Creek, tributary to Maacama Creek. 4CAS database lists collection date for SU 54863 and 54870 as 6/24/1900; however, as Cottus aleuticus were collected at this locality on 6/24/1897 and there are no records of additional collections by Gilbert in 1900 anywhere in this region, I suspect that the 1900 date in the Stanford ledger is in error. 5Locality listed as Cottonwood Creek; however, Cottoneva Creek flows through Rockport and lies between Ten Mile River and Usal Creek, which were sampled on the preceding and subsequent days. 6Two salmonids were apparently collected from Usal Creek, but no species’ determinations were made. 7Locality listed in Stanford ledger as Van Duzin Creek. 8No species determination made. 9Collection date listed in ledger as 7/12/1899; however, given other collections at this locality on this date in 1897 and lack of any 1899 specimens from the region, I assume the 1899 date is in error.

species had been published nearly two or early spring—with peak emigration salmon reach their peak abundance in decades earlier (Jordan and Gilbert, typically occurring in March and April the state—would have failed to pro- 1881a). Specifically, in his description and only rarely extending into June duce coho salmon juveniles. of Oncorhynchus keta, Snyder (1907) (Johnson et al., 1997). Hence, it is im- The clinching evidence in support of wrote the following: plausible that juvenile chum salmon this interpretation of these records can were not only present but more abun- be found in two subsequent publica- “Occurs in all except the small- dant than any other salmon during the tions. Five years after Snyder’s (1907) est streams between the Sacra- expedition. description of the expedition was pub- mento and Columbia rivers. The Further support for this interpre- lished, Gilbert (1912) openly acknowl- young of this salmon were appar- tation is found in Snyder’s narrative edged that he and his contemporaries ently more abundant than those of for O. kisutch, in which he lists Red- had commonly misidentified juvenile any other.” wood Creek (near Orick) as the only coho salmon as chum salmon in the California locality where this species past. This is evident in the following The timing of the surveys makes was found. Here again, it seems high- description of “dog salmon” (O. keta): this claim highly dubious, since they ly improbable that surveys conducted took place from mid-June to Septem- at more than two dozen sites in coastal “Less is known of the life histo- ber. Chum salmon migrate to sea al- watersheds of Mendocino, Humboldt, ry of the dog salmon than any of most immediately after emergence and Del Norte counties—the heart of the species thus far considered. from the gravel—typically late winter the coastal redwood zone where coho Our knowledge of the young is en-

20 Marine Fisheries Review tirely due to Chamberlain [1907], One perplexing aspect of the 1897 believe that many juvenile salmon who secured them on their sea- expedition is scarcity of specimens of samples, including the coho samples ward migration as fry, some with O. kisutch in the CAS museum collec- Snyder believed to be O. keta, were remnants of the yolk still attached. tion, especially given Snyder’s state- never cataloged in the Stanford ledger They were not associated with ment that (misidentified) O. keta were and have since been lost. larger individuals which could be the most abundant salmonid in coastal In summary, the Northern Califor- considered yearlings. As stated by streams. Only one coho salmon spec- nia/Oregon expedition led by Snyder Chamberlain, ‘records of the oc- imen (SU 54870) from this expedi- and Gilbert should have yielded great- currence of large individuals in tion currently resides in the CAS and er specificity regarding the distribu- streams have not been authenti- USNM collections, and its date of col- tion of various salmon species in Cali- cated, and, so far as is known, all lection is equivocal (see Table 5 foot- fornia and Oregon. However, the mis- leave fresh water as soon as they note). One can only speculate as to identification of coho salmon as chum are able to swim.’ Records of year- the fate of other coho salmon speci- salmon, as well as the apparent failure ling dog salmon have been made mens and why they were never record- to catalog collection localities for coho by the writer [Gilbert] and by oth- ed in the Stanford ledger. Other spec- and Chinook salmon, greatly dimin- ers in the streams of Washing- imens from this expedition were not ished the ability of these surveys to in- ton, Oregon, and California, but catalogued in the Stanford ledger un- form species freshwater range deter- all such have been founded on in- til 1961–62, so perhaps these jars were minations. correct identification of the coho lost or transferred elsewhere. yearlings.” Interestingly, in 1919 Snyder and Other Surveys two students (Bonnot and Hays) rep- In addition to the above surveys, Here, Gilbert explicitly credits the licated the 1897 survey of north- the Stanford Zoology Department was misidentification of juvenile coho ern California coast streams, visit- also involved with other more local- salmon as chum salmon to the poor ing many of survey locations on near- ized surveys in coastal watersheds of understanding of juvenile life histories ly the exact same day of year. CAS central and south-central California that existed during that time. Twen- specimens from the 1919 expedi- that are germane to the discussion of ty-four years later, Snyder (1931) tac- tion include coho salmon from Ten Pacific salmon freshwater ranges. Jor- itly admitted his 1907 error, writing Mile River (SU 54775), South Fork dan himself, apparently with assis- that “Humpback and dog salmon are Eel River at Garberville and Myers tance from a local resident, surveyed not common enough anywhere in the Ranch (SU 54845, 54864), Mad Riv- three streams in coastal San Luis Obis- State [California] to be of commercial er (SU 60068), Hunter Creek (SU po County, including San Luis Creek importance; in fact, they are so rarely 23613), Klamath River (SU 54783), (=San Luis Obispo Creek) near the seen as to be unknown to any but the Turwar Creek (tributary to Klamath town of Avila, Corral de Piedra Creek most observant fisherman.” River; SU54771), and Rowdy Creek (a tributary of Pismo Creek), and Ar- These admissions reinforce the need (tributary to Smith River; SU 54776). royo Grande Creek (Jordan, 1894b). to exercise considerable caution when Likewise, they include specimens of Interestingly, Jordan’s surveys pro- interpreting collection records and Chinook salmon from South Fork Eel duced no salmonids of any kind in any writings related to the distributions River (SU 54854, 23614), Van Duzen of these three streams.46 of both coho and chum salmon prior River (SU 59806, 54855), Mad Riv- However, with regard to Arroyo to 1912. Yet, despite the fact that both er (SU 59660), Klamath River (SU Grande Creek, Jordan stated the fol- Snyder and Gilbert both eventually ac- 54917; CAS 210057), and Rowdy lowing: knowledged their errors, the misidenti- Creek (SU 54774). Given that 1) both fication influenced Jordan’s subsequent coho and Chinook were common in “In this stream and in the oth- writings (see below) and has contrib- these rivers during the 1919 collec- ers, trout are occasionally tak- uted to a perception that chum salmon tions, and 2) Snyder (1907) reported en and sometimes salmon enter were once substantially more abundant Chinook salmon as being present dur- them from the sea. Lopez Creek, in California that has been perpetuat- ing the 1897 surveys in the Eel, Mad, a mountain tributary of Arroyo ed in the modern literature (see e.g., and Klamath rivers, I am inclined to Grande, is the best-known trout Johnson et al., 2012).45 stream in San Luis Obispo County. cho Cordova, Calif., 272 p.); however, in a sub- sequent report, Moyle et al. (2015) acknowledge It is said by anglers that the brook 45Moyle et al. (1995) cite Jordan and Gilbert’s that the current status of chum salmon in Cali- trout exist in the mountains and (1881a) statement that chum salmon were report- fornia is equivocal, in part because of the poor the salmon trout come up from the ed to occur in “all streams from San Francisco to understanding of historical occurrence (Moyle, [the] Bering Straits” (Moyle, P. B., R. M. Yoshi- P. B., R. M. Quinones, J. V. Katz, and J. Weav- yama, J. E. Williams, and E. D. Wikramanayake. er. 2015. Fish species of special concern in Cal- 46Unfortunately, Jordan’s paper does not indicate 1995. Fish species of special concern in Califor- ifornia, third ed. Rep. prep. for Calif. Dep. Fish the time of year that the surveys were conducted, nia, second ed. Final rep. for contr. 2128IF prep. Wild., Sacramento, Calif., 842 p. (avail. at www. which would influence the likelihood of encoun- for Calif. Dep. Fish Game, Inland Fish. Div., Ran- wildlife.ca.gov/Conservation/SSC/Fishes)). tering certain species.

81(1) 21 on the lower mainstem of the Pajaro River (Santa Cruz/Monterey counties). O. mykiss was the only salmonid spe- cies collected on this survey (USNM 75320). The following October, a 4-day survey sampled the Salinas Riv- er and one tributary (Arroyo Seco) in Monterey County, as well as a site in the San Lorenzo River basin near the town of Felton. Again, O. mykiss is the only salmonid species represent- ed in the collections, being found in both the San Lorenzo River and Ar- royo Seco (USNM 75331, 75323), al- though as will be discussed in a subse- quent section, given the land-use dis- turbances that had occurred prior to 1906, this should not be construed as evidence that other salmonid species were never present.47 Early Artificial Propagation and Outplanting of Salmon in California Figure 5.—Baird Hatchery along the McCloud River taken between 1890 and 1897. Figure from Stone (1897). In evaluating early historical records of occurrence of various Pacific salm- on in California, it is important to ac- sea and ‘promiscuously mix with to both the north and south (Boughton knowledge the possibility that some it.’ This seems another way of say- et al., 2006). Yet the assertion of Chi- early observations might be the result ing that the brook trout (irideus) nook salmon seems considerably more of plantings of hatchery-reared fish and the salmon trout (gairdne- dubious, for reasons similar to those into waters outside of their historical ri) are but forms of the same fish. outlined earlier regarding the Ventura range. Artificial propagation of sal- The individuals which run to the River—namely, these are watersheds monids by private interests in the state sea grow larger and are more sil- dominated by grassland, chaparral, pre-dates 1870 (Leitritz, 1970), and very in color than those which re- and oak woodlands, with relatively dry the first government-operated salm- main in the brooks.” climates and intermittent connectivity on hatchery was built by the U.S. Fish between the freshwater and marine en- Commission on the McCloud River in Jordan concludes his narrative with vironments. 1872 (Fig. 5). One might thus presume a “list of the fishes of the streams of Swift et al. (1993) expressed similar that stocking of salmon species out- San Luis Obispo County so far as re- doubt about the veracity of this report. side of their accepted historical range corded,” which includes three salmo- Despite the apparent lack of direct ev- in California occurred early on. How- nids: Salmo mykiss gairdneri, Sal- idence of occurrence, Evermann and ever, the distribution of salmon from mo mykiss irideus, and Oncorhynchus Clark (1931) and others that have fol- these early hatchery activities is well tschawytscha. Given that no salmo- lowed (e.g., PFMC, 1999:A-51; Gus- documented and indicates that, prior nids were actually observed or collect- tafson et al., 2007) have cited Jordan’s to 1905, releases of hatchery salmon ed during his surveys, it appears Jor- (1894b) paper as evidence of Chinook likely had a negligible influence on the dan relied on other information—quite salmon in creeks of San Luis Obispo perceived freshwater ranges of salmon possibly accounts from the local resi- County. species within California. dent that assisted him—to conclude In addition to the San Luis Obis- that Chinook salmon and the resident po area surveys, Stanford faculty and 47Snyder also conducted surveys of Monterey Bay tributaries during 1909, reporting the occur- and anadromous forms of O. mykiss staff also conducted limited surveys in rence of Chinook and coho salmon in the San Lo- occurred in San Luis Obispo County coastal watersheds of the Central Cali- renzo River, and Chinook salmon in the Pajaro streams. fornia Coast. In October of 1896, Gil- River (Snyder, 1913). However, these observa- tions occurred during a period when both species The observation of resident and bert and Snyder led a three-day sur- were being reared and released from Brookdale anadromous forms of O. mykiss steel- vey of streams in the Pajaro River sys- Hatchery into Monterey Bay tributaries (Cobb, 1911). Thus, the usefulness of these observa- head raises no eyebrows, as steelhead tem, sampling Uvas and Llagas creeks tions in establishing natural historical presence did and still do occupy coastal streams (Santa Clara County), along with sites is limited.

22 Marine Fisheries Review Figure 6.—Distribution of Chinook salmon eggs and fry from Baird Hatchery during the periods (a) 1872–1883 and (b) 1888–1905.

In this section, I review the pub- any of these early private facilities at- Fish Commission for release into the lished information on the propaga- tempted to propagate Pacific salmon. upper Sacramento River and its tribu- tion and distribution of the five Pacif- The Baird Hatchery, built and oper- taries or retained at the Baird Hatch- ic salmon species in California in the ated under the direction of Livingston ery for release back into the McCloud late 1800’s and early 1900’s, which Stone, was located on the McCloud River. Hence, within-state releases of demonstrates both that early propaga- River a few kilometers upstream of its Chinook salmon in California during tion efforts focused almost exclusively confluence with the Pit River, a trib- this early period were largely confined on Chinook salmon, and that the vast utary of the upper Sacramento Riv- to the Sacramento River basin, not far majority of salmon eggs produced by er in Shasta County. In its first years from the Baird facility. early California hatcheries were either of operation, the primary purpose The only documented e­xceptions shipped out of state (or country) or re- of the hatchery was to produce Chi- were the transfer of a combined leased back into local waters. Only af- nook salmon eggs to be shipped to 550,000 eggs to the San Francisco ter 1905, with the establishment of the eastern United States and beyond Sports Club in 1876 and 1877 (pre- more state-operated salmon hatching to Europe and New Zealand (Cobb, sumably for release in or near San Fran- and rearing facilities, as well as im- 1911). Indeed, Baird (1874) and others cisco Bay; Clark, 1929), and the 1881 proved transportation, did more wide- brimmed with optimism that, because release of about 150,500 fry into var- spread out-of-basin plantings within of their ability to tolerate comparative- ious waters including the Russian Riv- California begin to occur. ly warm waters, salmon from the west er basin (30,000); Donner Lake and Establishment of the Baird Hatchery coast would thrive in waters along the nearby streams (20,000)49; Pescadero on the McCloud River in 1872 was a east coast well south of the range of and San Gregorio creeks in San Ma- landmark event, representing the first other eastern salmonids.48 teo County (15,000 each); Santa Cruz salmon hatchery in the western Unit- Between 1872 and 1883, the Baird County (15,000); the Salinas and Paja- ed States (USFC, 1874; Lichatowich, Hatchery produced roughly 64 mil- ro rivers in northern Monterey Coun- 1999). Prior to 1872, a few small pri- lion Chinook salmon eggs. About 62% ty (15,000), and the San Francisco Bay vately owned hatcheries had been op- of these were shipped out of Califor- area (40,50050) (CFC, 1882). Collec- erated in California; however, these nia to other states or countries (Fig. tively, releases within California but hatcheries primarily reared either fish 6a; Clark, 1929). Of the remaining outside of the Sacramento River Basin imported from the U.S. east coast 38%, that vast majority (98%) were ei- and San Francisco Bay made up a tiny (e.g., eastern brook trout, Salvelinus ther turned over to the California State fontinalis; landlocked Atlantic salmon, Salmo salar; and lake whitefish,Core - 49 48Livingston Stone, in a letter to the California The Truckee River watershed in which Donner gonus clupeaformis), or local cutthroat Fish Commission (Stone, 1875), struck a slightly Lake resides drains to the east side of the Sierra trout (O. clarkii henshawi) from Lake more cautious tone, implying that even if the ma- and not to the Pacific Ocean. jority of experimental introductions failed, a sin- 50An unknown number of fish from the Bay Area Tahoe and Truckee River basin (Lei- gle successful introduction would “pay over and release total was released in Grass Valley, Calif., tritz, 1970). There is no evidence that over again for all that the failures cost.” part of the Sacramento River watershed.

81(1) 23 fraction (<0.1%) of the total Chinook Battle Creek, and Mill Creek water- like the San Lorenzo River, Pescadero salmon hatchery production. sheds. Creek, and perhaps the Salinas ­River In late 1883, activities at the Baird Nearly 50 million fish (about 10% in years when rainfall was early and Hatchery were suspended when the of the total) were transferred into the sufficient to breach sand bars allow- number of returning adults declined Eel River basin between 1897 and ing fish to enter. Thus, the numerical- precipitously. This decline was attrib- 1905, representing the first signifi- ly minor plantings of Chinook salmon uted to construction of the railroad up cant out-of-basin, within-state trans- fry into streams between the Golden the Sacramento Valley from Redding fers of Chinook salmon in Califor- Gate and Monterey Bay overlaps ar- northward. Heavy blasting by the rail- nia. Additionally, between 1896 and eas where there was likely at least spo- road crews apparently deterred salm- 1898, about 1.75 million fry were re- radic natural occurrence of Chinook on from entering the upper Sacra- leased into Lagunitas Creek as part salmon in coastal watersheds. mento River. Additionally, thousands of experiments to introduce Chinook The early history of coho salmon more salmon were intentionally de- salmon into the watershed and deter- propagation in California is consid- stroyed by the 3,000–6,000 railroad la- mine if fish returned to their “home erably simpler. Prior to 1889, there is borers51 who used blasting powder to stream” to spawn52 (Rutter, 1902). no evidence that artificial propagation kill fish for food (Stone, 1885). While Only 5% of the salmon were shipped of coho salmon took place in Califor- many fish were eaten by the work- outside of California. nia. In that year, the U.S. Fish Com- ers, many more likely died due to the In summary, although the numbers mission completed construction of a highly destructive method of capture of Chinook salmon produced at Baird, new hatchery facility at Fort Gaston (Shebley, 1922; Leitritz, 1970). The Sisson, Mill, and Battle Creek hatch- in the Hoopa Valley along the Trinity Baird Hatchery resumed operation in eries were substantial, the overwhelm- River (Humboldt Co.). In its first year 1888 shortly after the northward rail ing majority of hatchery fish were ei- of operation, the hatchery hatched ap- line from Redding, Calif., to Ashland, ther transferred out of state or released proximately 100,000 Chinook salm- Oreg., was completed. back into the Sacramento River and on eggs that had been transferred from By this time, it was clear that at- its tributaries. The only major out-of- the Baird Hatchery and released the tempts to establish Chinook salm- basin transfers of Sacramento River resulting fry (~30,000) into the Trinity on runs in eastern waters flowing into salmon to coastal watersheds were to River (USFC, 1893). the Atlantic Ocean and Gulf of Mex- the Eel River, which already supported The following year, traps were con- ico had failed entirely (McDonald, native populations of Chinook salmon. structed on both the Trinity River and 1892:xxxv). As a consequence, opti- I have found no evidence that fish Redwood Creek, a coastal watershed mism about spreading Chinook salm- were released in California south of only 13 miles overland from Fort Gas- on to other parts of the world faded, Monterey Bay prior to 1905. Con- ton where an auxiliary station was es- and the primary purpose of the feder- sequently, the likelihood that hatch- tablished. U.S. Fish Commission re- ally owned facility shifted toward sup- ery plantings played a significant role cords indicate that “collecting of the plying eggs for the newly established in early descriptions of the freshwater eggs of the quinnat and steelhead and state-operated Sisson Hatchery range of Chinook salmon seems low. salmon was begun in November, 1889, (now known as Mt. Shasta Hatchery). One might argue that the plantings to and continued three months,” the fry From here, Chinook salmon began to Pescadero, San Gregorio, and Santa of which were released locally into be distributed more widely within Cal- Cruz and northern Monterey Counties nearby Supply Creek, which provided ifornia. may have been outside of the histori- water for the Fort Gaston hatchery. At Two new federal facilities were es- cal spawning distribution of Chinook Redwood Creek, about 30,000 “salm- tablished on Battle Creek (1895) and salmon. However, the number of fry on” eggs were collected, and the re- Mill Creek (1903), which substantial- released (45,000 total) was small, and sulting 25,000 fry were released one ly increased egg production. Between survival of fry is generally very low month later into Minor Creek, a trib- 1888 and 1905, almost 480 million (Flagg et al., 1995). Furthermore, giv- utary of the creek (USFC, 1893). Giv- Chinook salmon eggs and fry were en that hundreds of thousands (if not en that the term “quinnat” salmon was produced at these three facilities, an more) of Chinook salmon from Cen- sometimes applied generally to west average of more than 26 million per tral Valley rivers reared in Monterey coast salmon species, and that the fish year (Clark, 1929). The shift in pri- Bay each year, it seems highly likely collected in Redwood Creek were re- orities is evident in the fish distribu- that strays routinely entered systems ferred to only as “salmon,” one can- tion records (Fig. 6b). Approximately not be 100% certain that coho salm- 84% of these fish were released back on were not part of these collections. 52Although adult Chinook salmon did return to into the upper Sacramento, McCloud, the stream, Rutter still rejected the idea that this But all progeny were listed as “quinnat reflected a homing tendency, instead conclud- salmon” in the summary of distribu- ing that these fish never left Tomales Bay and re- 51Shebley (1922) reported the number of labor- turned to Lagunitas Creek simply because it was tion tables, and as all hatchery-reared ers as 9,000. the nearest freshwater source. fish were released locally, this distinc-

24 Marine Fisheries Review tion is immaterial to the question of ing “silver salmon” eggs and trans- cal streams (Clark, 1929). Additional- whether these outplantings influenced porting them to the Sisson Hatchery ly, records indicate that between 1915 the historical distributions of Califor- in the Sacramento River basin. About and 1919, 150,000 Chinook salmon nia’s salmon. 1.4 million fry were produced, half of fry were released into the Santa Ynez Over the next several years, salm- which were released into the Klam- River and other Santa Barbara Coun- on continued to be collected from both ath River and half into the Sacramen- ty streams, and another 125,000 were the Trinity River at Fort Gaston and to River. According to the California released into the Ventura River (Cobb, Redwood Creek, with apparent ex- Fish and Game Commission (CFGC, 1921). However, these plantings oc- change of individuals among these ba- 1913), the collections at Klamathon curred well after Jordan’s initial de- sins. In 1893, a new station was estab- marked “the first effort made in this scriptions of species ranges were writ- lished near Korbel along the Mad Riv- state to increase the run of silver salm- ten in the 1880’s. er. The annual U.S. Fish Commission on.” More accurately, this marked the Report for 1894 (Bean, 1896), con- first state-run hatchery operations, The State of California’s tains the first explicit mention of “Sil- as coho salmon had been propagated Rivers and Streams in ver salmon” being captured and prop- at the federally owned facilities dis- the Late 19th Century agated in California, with all 280,000 cussed above from 1894 to 1897, and When reading the historical salmon fry being released into the Mad Riv- the County of Santa Cruz’s Brookdale literature, one may be inclined to as- er.53 Over the next three years, an ad- Hatchery had reared coho salmon im- sume that habitat conditions for salm- ditional 1.77 million coho fry and ported from Baker Lake, Wash., and on during the late 1800’s were far bet- yearlings were reared and released, released them into the San Lorenzo ter than at present and that, therefore, but none of these releases occurred River between 1906 and 1910 (Spen- these early accounts provide a bet- outside of the Trinity River, Redwood ce et al., 2011). ter indication of historical freshwa- Creek, or Mad River basins (Ravenel, Records of propagation or trans- ter distribution than more recent infor- 1896, 1898). planting of pink, chum, and sockeye mation. While this generalization may In 1898, few salmon and no steel- salmon between 1870 and 1905 are hold true in some locations, it would head reached the traps on the Trini- rare to nonexistent. In fall of 1891, be far from correct to assume that Cal- ty River, so no collection of eggs was two female and one male pink salm- ifornia’s watersheds within the accept- made at Fort Gaston. Returning Chi- on were collected and propagated at ed historical range of anadromous sal- nook and “nerka” salmon (but curi- Baird Station, but all fry were released monids were in pristine condition. ously no coho salmon) were appar- back into the McCloud River (USFC, The discovery of gold at Sutter’s Mill ently collected in Redwood Creek and 1894). Similarly, the record of “ner- in 1848 triggered a massive influx of the resulting fry released locally. Kor- ka” salmon being collected, propagat- settlers to California. Between 1850 bel Station was not operated that year. ed, and released in Redwood Creek in and 1900, California’s population in- Due to the difficult access to the Fort 1898 appears to be the only record of creased more than fifteen-fold, from Gaston, Redwood Creek, and Korbel sockeye salmon being propagated in under 100,000 to nearly 1.5 million stations, which had to be supplied and California in the 1800’s. I have found people, nearly half which took up resi- serviced entirely with pack mules, the no evidence to indicate that chum dence in the counties immediately sur- stations were abandoned at the close salmon were ever bred or transplant- rounding San Francisco Bay (CSDC, of the year (Ravenel, 1899). In sum, ed into California waters in the 1800’s. 2012). This tremendous increase in although these three federal facilities In summary, there is little evidence population prompted many human in- were the first in California to propa- to suggest that, in the period 1870– dustries that dramatically affected sal- gate coho salmon, there are no records 1905, observations of salmon resulting monid habitats in many coastal Cali- to indicate that any of these fish were from planting of fish outside of their fornia watersheds well before the first ever transplanted outside of the imme- historical range would have apprecia- scientific exploration of these water- diate vicinity of these stations. bly influenced the range descriptions sheds had occurred and the taxonomy Following closure of those sta- published by Jordan and others. After of the various salmon species was re- tions, there were no federal or state- this period, the situation changed sig- solved. run hatcheries in California that pro- nificantly, as out-of-basin transfers be- The principal land-use activities that duced coho salmon until the 1910–11, came more commonplace. For exam- degraded salmonid habitats in Califor- season, when the state opened Klam- ple, between 1905 and 1911, approx- nia during the 19th century varied with athon Egg Collecting Station, collect- imately 4.6 million Chinook salmon region. Logging unquestionably had eggs were transferred to Brookdale severe and widespread impacts on Cal- 53The origin of these fish is unclear. The text Hatchery in the Santa Cruz Moun- ifornia’s coastal populations of Pacific reported capturing “silverside salmon” on the tains (Santa Cruz Co.), with the re- salmon, particularly in coastal moun- Trinity River, but did not indicate if any coho salmon were collected at the Redwood Creek or sulting fry likely released into the San tains from the Oregon border south to Korbel facilities.º Lorenzo River and perhaps other lo- northern Monterey County with exten-

81(1) 25 nate one standing in the pit below (Stanger, 1963). The pace of logging was slow and the spatial extent lim- ited. In 1842, the first water-powered sawmill in the Santa Cruz Mountains was built along the San Lorenzo Riv- er near its confluence with Zayante Creek (Poli, 1956).54 Following the discovery of gold in 1848 at Sutter’s Mill, California’s population expand- ed rapidly and with it the demand for lumber, shingles, fence posts, rail- road ties, and a variety of other wood products, all of which were produced from redwoods (Carranco, 1982). Wa- ter-powered mills thus began to pro- liferate throughout the Santa Cruz Mountain region. By 1860, more than 10 million board feet of timber were being cut annually in Santa Cruz Figure 7.—Oxen hauling logs on a “corduroy road” in Mendocino County, ca. 1851. County (Payne, 1978). Photo courtesy of the Held-Poage Memorial Home and Mendocino Historical Soci- The typical practice was for fallers ety, Ukiah, California, from the Robert Lee Collection. to clear cut an area within a mile or two of an existing mill site. Frequent- sive stands of such commercially har- it is instructive to consider a few spe- ly, the area was burned to remove un- vestable conifers as coast redwoods, cific examples in detail to illustrate the derstory vegetation, which facilitated Sequoia sempervirens; Douglas fir, enormous effects that logging, mining, skidding of logs to the mill site. Trees Pseudotsuga menzeisii; Sitka spruce, grazing, and other land and water uses were bucked into shorter lengths and Picea sitchensis; and pine, Pinus spp. likely had on salmon populations well often dragged over so-called “cordu- Hydraulic mining for gold had its before the 1900’s. These examples roy roads” by teams of oxen or hors- greatest impact on rivers and streams demonstrate that the baseline of salm- es to the mill site (Fig. 7). Cordu- draining the west slope of the Sier- on productivity and possibly distribu- roy roads were skid trails constructed ra Nevada from the Feather River in tion had very likely already undergone by placing small diameter logs (typi- Plumas County south to the Merced a substantial shift before Jordan and cally 8–10 inches) at close intervals River in Mariposa County, though others began writing about salmon dis- perpendicular to the direction of the large-scale mining also occurred in the tributions in California. roads. These roads were greased with upper Klamath and Trinity River ba- beef tallow to facilitate the sliding of sins. The chaparral and grassland dom- Impacts of Early Logging logs to the mill (Payne, 1978). Often inated regions of central and southern Although logging took place in the times, corduroy roads were built di- coastal California were subjected to 1800’s up and down the northern and rectly atop a stream channel. Skid- grazing by domestic cattle and sheep central California coasts, the history ding logs was among the more ar- even before the gold rush, but the in- of logging in the Santa Cruz Moun- duous and labor-intensive practices. tensity of grazing increased many-fold tains is especially germane since it Thus, when the local supply of logs in the 1850’s in response to the gold- encompasses the region that likely ran out, it was more cost-effective to fueled increase in California’s human represents the southern extent of the dismantle and relocate the mill than population. Accompanying these ac- historical distribution of persistent re- to skid logs greater distances. Conse- tivities were dam building, water di- producing populations of coho salm- quently, new mill sites were continu- versions, agricultural land conversion, on (Spence et al., 2011) and possibly ally being established penetrating fur- release of pollutants, and many other Chinook salmon as well. Prior to the ther into the more remote regions of activities that further affected salmon 1840’s, logging in Santa Cruz Coun- a basin (Stanger, 1967). and their habitats. ty occurred on a limited basis using An exhaustive treatment of the many “pit and whipsaw” methods in which 54The sawmill, owned by Isaac Graham, is often human alterations to the habitats of a large pit was dug beneath a fallen purported to be the first power-operated sawmill anadromous fishes that took place in tree, and planks were sawn off by a in California; however, apparently Cooper’s saw- mill along Mark West Creek in Sonoma Coun- California during the 1800’s is beyond team of two sawyers, one standing ty predates the Graham mill by 8 years, having the scope of this paper. Nevertheless, atop the log and another less fortu- been built in 1834 (California State Parks, 2017).

26 Marine Fisheries Review The pace of logging accelerated in the 1870’s with the advent of steam- powered mills equipped with circular saws, coupled with the steam donkey, a mechanical steam-powered winch that was capable of dragging logs thousands of feet to mill sites (Payne, 1978). The reduced dependence on wa- ter for power meant that milling could occur year-round, not just seasonal- ly as had been the case in years prior. Additionally, transportation of logs to mill sites and lumber to ports was fa- cilitated by the construction of narrow- gauge rail lines, sometimes on impres- sively steep terrain. With these new tools available, logging in the Santa Cruz Mountains reached its peak, and by 1884, annual harvest had increased to approximately ~34 million board feet in the Big Basin and San Loren- zo Valley areas alone (California State Parks, 2011). This rapid rate of harvest quickly depleted the timber supply in the Santa Cruz Mountains, and by the early 1900’s many mills had closed. Over the 55-yr period from 1850 to 1905, no fewer than 340 different mill sites had been used in the Santa Cruz Mountains, and no sizeable watershed had been spared (Fig. 8). Figure 8.—Locations of sawmills and shingle mills in the Santa Cruz Mountains The impacts of these intensive log- between 1840 and 1905. Circles adjacent to stream mouths indicate additional mills known to have been present in the watershed but that could not be reliably placed ging activities on salmon habitats in on the map. Based on data from Rood (1975) and Stanger (1967). the region were enormous. The long- lasting impacts of logging practices on salmonid habitats are well documented gen levels, to the detriment of salm- sun, with concentrated mephitic (Meehan, 1991; Spence et al., 1995). on and other aquatic life. In 1871, one gases emitting all the stinks and Loss of large wood and sources of local Santa Cruz newspaper indicates smells so famed, in Cologne, be- wood recruitment, increases in stream that dumping sawdust into streams had sides the innumerable stenches lo- temperatures from canopy remov- “become universal…until our pure cally incident to decayed vegeta- al, and accelerated erosion and sedi- limpid streams were discolored, and tion and rotten saw-dust, so that mentation in streams were undoubted- the water became in some instances the workmen could hardly endure ly pervasive. In addition, early logging black as tar—a moving mass of turgid it….We are thus particular in de- practices included construction of mill filth.” The article elaborates on how scribing what we saw, to give the ponds, often with dams that preclud- mill ponds downstream of dumping exact truth of the matter, as it ap- ed fish passage (CFC, 1892). These sites frequently filled up with sawdust, peared, and to correct…a former sometimes stood in place for many describing one situation on Soquel expressed impression that placing years at some of the larger, more per- Creek where an impoundment 175 saw-dust in running brooks was manent mill sites, blocking access to yards long by 20 yards wide was filled only injurious to the fish—brook spawning areas. to a depth of two to four feet. They de- trout and salmon—so plentiful in Furthermore, up until the late scribed the process of cleaning out the all the streams in this country.” 1880’s, it was standard practice for dam as follows (Anonymous, 1871): mill operators to dump sawdust and Several years later, the Santa Cruz refuse wood directly into the streams. “…a more intolerable nuisance we Sentinel (Anonymous, 1877) again de- Decomposition of this material un- have seldom seen anywhere; the cried the dumping of sawdust into lo- doubtedly affected dissolved oxy- saw-dust was steaming in the hot cal streams as follows:

81(1) 27 actment of a state law that banned the A practice of dumping sawdust in streams and required mill owners to burn or otherwise discard their refuse away from water (California State Legisla- ture, 1889). In Mendocino, Humboldt, and Del Norte counties, the logging history parallels that of the Santa Cruz Moun- tains, both in time and with respect to logging methods. The first saw- mills were built in the early 1850’s. In these counties, however, there were fewer roads for hauling wood; thus, the timber companies relied primari- ly on the rivers themselves to transport logs from remote areas down to river mouths, where they could be milled and the lumber taken to markets in San Francisco and elsewhere. Riverine B transport was accomplished through the use of splash and crib dams (Fig. 9a), particularly destructive practices that were largely avoided in the San- ta Cruz Mountains because of the ear- ly availability of rail transport. Splash and crib damming involved constructing a dam across the stream for the purpose raising water levels, splash dams being temporary and crib dams being larger and semi-perma- nent (Jackson, 1991). Logs were rolled into the stream either above or below the dam (Fig. 9b). Then, during storm events, either flood gates were opened or the dam was demolished with ex- plosives, releasing the water, logs, and remnants of the dam (Carranco, 1982). Figure 9.—(a) Crib dam under construction on the Navarro River, Mendocino Coun- As described by Jackson (1991), when ty, date unknown, and (b) log jam on Big River, ca. 1924. Photos provided courtesy the dam broke, “all Hell broke loose. of the Held-Poage Memorial Home and Mendocino Historical Society, Ukiah, Cali- The water, the former jammed logs, fornia, from the Robert Lee Collection. and last but not least the dam, all went down the river. When the dam went, “We have always maintained that vers that within the last six months all evidence of its existence was re- lumberman dumping their refuse have filled so many new-made moved.” Another witness described timber and saw-dust into the San graves. Poisoned water is slow poi- a “wall of logs, trash, rocks and oth- Lorenzo River and its tributaries son to the drinker thereof….” er debris that towered nearly 20 feet were committing a nuisance. They high,” noting that it was odd to see are destroying the fish and poi- The newspaper’s unfounded claim of the “dust flying as the logs in the drive soning the water for human use. a possible linkage between sawdust and sheared off sharp points on the stream A dumb brute has too much sense diphtheria notwithstanding, it is clear banks.” When the torrent of logs even- to drink from the water fresh and that the local citizens were well aware tually came to a halt, a new dam was inky black from a saw-dust pile. of and concerned about the impacts of built and the process repeated until We fear that this fact has some- sawdust dumping on salmon and trout logs reached the river mouths, where thing to do with the prolonged ex- in the region’s streams. Finally, a doz- they were collected on large booms. istence of diptheria and other fe- en years later, public outcry led to en- Needless these to say, these so-

28 Marine Fisheries Review called “log drives” had immediate and devastating impacts on any fish rearing in the stream or eggs deposited with- in the gravel. Even more importantly, the long-term result of these splash- dam activities was that river chan- nels were often widened and heavily scoured, leaving little of the structur- al complexity and other critical habitat elements (e.g., spawning gravels, cov- er, velocity refuge) needed by salmon (Sedell et al., 1991). Impacts of Hydraulic Mining In terms of habitat damage, few human activities could rival the dev- astating impacts of hydraulic min- ing for gold. At hydraulic mine sites, high-pressure jets of water were di- rected at stream banks, terraces, and hillsides to dislodge rock and move sediment through sluice box- es, from which gold was retrieved (Fig. 10). Through this process, mas- sive amounts of sediment were there- by flushed into streams and rivers. Gilbert (1917) estimated that be- tween 1849 and 1909, more than 684 million cubic yards of gravel, sand, silt, and other mining debris moved through the Yuba River system alone. When mining was complete, a bar- ren wasteland of tailings was left within and along the stream channel. Downstream of the mining sites, riv- er channels filled with massive depos- its of debris, sometimes many tens of feet in depth (Gilbert, 1917). To sup- ply the hydraulic mines with water, numerous dams were built higher in the mountains and water was diverted through extensive networks of ditches and flumes, sometimes totaling hun- Figure 10.—Hydraulic mining at the La Grange Mine (1851–1918) near Junction dreds of miles (Kelley, 1954, 1989). City in the Trinity River basin. Photo taken ca. 1914. From Clark 1970. Salmon populations that suffered the greatest damage were found in the foothill region of the western Sier- ra Nevada from Merced County in the can, Mokelumne, and Tuolumne to which salmon formerly resort- south to Plumas County in the north. Rivers for the purpose of spawn- ed for spawning, have, for this Six years after its founding, the Cali- ing, and many places, such as purpose, been destroyed by min- fornia Fish Commission (CFC 1877:5) Salmon Falls, on the American, ing. As mining is the more impor- wrote the following: were named from the abundance tant industry, of course, for this of these fish….At the present time evil there is no remedy, other than “In eighteen hundred and fifty no salmon enter these streams. by artificial means to increase the the salmon resorted in vast num- It would be safe to estimate that supply in those tributaries that are bers to the Feather, Yuba, Ameri- one-half the streams in this State still the resort of these fish.”

81(1) 29 A few years later, the CFC (1880:3) netti Act of 1893, affirmed the right teration of physical habitats, but chem- added further detail: to use hydraulic techniques; however, ical contamination as well. the Act likewise prohibited release of “The numbers of salmon that mining debris and established a “Cal- Impacts of Grazing could have been taken in this riv- ifornia Debris Commission” to over- Livestock were first brought into er [Sacramento], before the great- see dam structures designed to contain California in 1769 by Spanish mis- er part of their spawning beds mining wastes. Ultimately, the costs of sionaries (Burcham, 1957). Between had been destroyed by sediment building debris dams and other con- 1769 and 1833, the Spanish estab- from the gold mines, can nev- tainment structures proved prohibitive, lished 21 missions in the coastal re- er be known. It is the testimo- and most hydraulic mining operations gion from San Diego to Sonoma along ny of all the pioneer miners that ceased by 1900 (Mitchell, 1994; Kel- the famed El Camino Real in what was every tributary of the Sacramen- ley, 1989; Krause et al., 2010).58 then called Alta California. To encour- to, at the commencement of min- Although hydraulic mining activity age settlement around the missions ing, was, in the season, filled with was centered in the foothills of west- and associated pueblos and presidios, this fish, hurrying and struggling ern Sierra Nevada, it was not confined the Spanish government issued land as if to reach the very sources of solely to this region. Hydraulic mining grants that permitted settlement and these rivers. A few salmon con- and associated channel dredging was granted grazing rights on surrounding tinued to enter the Feather, Yuba, particularly intense in tributaries to the lands. The crown retained ownership Bear, and American Rivers until Klamath River, including the Trinity, of the lands; thus, the grants effec- the floods of the Winter 1860-1, Salmon, Scott, and Shasta rivers (Alp- tively amounted to free leases. How- which covered the gravel bottoms ers et al., 2005). Because the Klam- ever, when Mexico gained its indepen- of those streams with mining sed- ath-Trinity mountains were outside of dence from Spain in 1821, it moved to iment, and thereby destroyed their the Central Valley, they were exempt establish rules for petitioning for land spawning grounds.” from the Caminetti Act; thus, hydrau- grants in an effort to break the land lic mining continued into the 1950’s in monopoly of the missions and give ti- Although the damage to fisher- this region (Krause et al., 2010). tle of the lands to settlers (Burcham, ies was extreme, it was the aggrada- According to state records for the 1957; Cleland, 1964). tion of sediments and debris on farm- year 1898, 307 hydraulic mines were In the years of Spanish and Mex- lands that eventually led to the end of operated in Trinity County alone, in- ican rule, livestock were reared pri- hydraulic mining in the Mother Lode cluding the two largest hydraulic marily for local subsistence, as well Region (Sumner and Smith, 1940). mines in the world at that time, the as for the production of hides, tallow, The accumulated debris buried valu- La Grange mine near Junction City and wool, all non-perishable products able agricultural lands, clogged irri- and the Union Hill mine near Douglas that could be exported. However, with gation ditches and shipping channels, City (Bailey, 2008). The Salmon Riv- the massive influx of settlers to Cali- and contributed to flooding of towns er in the middle portion of the Klam- fornia following the end of the Mex- downstream of the mines. Rulings ath basin was similarly hard hit, with ican-American war and the discovery handed down in both state (People v an estimated 15 million cubic yards of of gold, the demand for food products Gold Run Ditch and Mining Co.55) material washed off of riparian hill- (beef and mutton) skyrocketed, as did and federal courts (Woodruff v North sides between 1870 and 1950 (Elder the value of livestock (Cleland, 1964). Bloomfield56 in 1884 effectively halted et al., 2002). Many large tailing piles Cattle that sold for $4 per head in large-scale hydraulic mining (Sumner still exist today that continue to limit the late 1840’s could fetch $75 in the and Smith, 1940; Kelley, 1989).57 The riparian function. Liquid mercury was growing urban centers by the 1850’s Sawyer Decision in the North Bloom- added to troughs in the sluice in or- (Larson-Praplan, 2016). Though com- field case, though it did not prohibit der to enhance recovery of gold, which prehensive estimates of the total num- hydraulic mining outright, did prohib- bonded to the heavy mercury to form bers of livestock are not available for it the discharge of mining debris into gold-mercury amalgam (Alpers et al., this era, estimates from the U.S. Cen- the Yuba River (Mitchell, 1994). Fed- 2005). Thus, the impacts of hydraulic sus Bureau indicate that between 1850 eral legislation, known as the Cami- mining included not only massive al- and 1860, the number of beef cattle in California increased nearly four-fold 55People vs. The Gold Run Ditch and Mining Co., and the number of sheep increased 62- et al., 66 Cal. 138, 151, 4 P. 1152 (1884) 58Hydraulic mining saw a brief resurgence af- fold (Burcham, 1957:143). The Cen- 56Woodruff vs. North Bloomfield Mining Co., 18 ter passage of the Caminetti Act; however, a ma- sus Bureau counts underestimate the F 753 (1884). jor storm in January of 1896 extensively dam- 57Illegal hydraulic mining continued to occur for aged water supply systems and caused the fail- total number of livestock on Califor- some time after the injunction, as ordinary miners ure of many debris dams, sending a new wave of nia lands during this era as they count- continued to use mining facilities abandoned by sediments down the Sierra rivers (Kelley, 1989). the larger corporations (Kelley, 1989), but gener- Many mine owners subsequently abandoned or ed only livestock on farms and exclud- ally at a much smaller scale. sold their water rights. ed those animals grazed on the open

30 Marine Fisheries Review range. Regardless, the numbers of occupancy by Pacific salmon, they more northerly counties for tanbark cattle, sheep, and feral horses clearly were likely marginal to begin with giv- supplies (Bowcutt, 2015). Although numbered in the millions. en the Mediterranean climate, inter- the impacts of tree removal were rel- Impacts from this extensive graz- mittent connectivity with the ocean, atively minor compared to logging of ing on fish habitats during the 1800’s and temperatures near the upper lim- the redwoods, they nevertheless added are poorly documented, though mod- its of tolerability for species like coho to the cumulative impact of land-clear- ern understanding of grazing impacts salmon that spend a year or more in ing activities. Perhaps the greatest im- indicates that alteration of streams and fresh water. Consequently, even mod- pact of tanneries to aquatic ecosystems rivers was likely substantial (Platts, est changes in hydrology (and hence was the discharge of tannery wastes 1991, and Spence et al., 1995, provide ocean connectivity) and temperature into local streams, which was com- reviews of grazing impacts). The ran- regimes would likely have rendered mon practice. In 1885, one of Santa chos of the Spanish and Mexican era them unsuitable for salmon. Though Cruz’s most prominent businessmen, were invariably situated around per- populations of O. mykiss have been Frederick A. Hihn, wrote an editorial manent water sources. Livestock often able to persist in these watersheds, it is in a local newspaper lobbying for de- congregate in riparian areas as these almost certainly because of their great- velopment of a sewage system through provide water, shade, and more suc- er life history diversity, including the downtown Santa Cruz that would emp- culent vegetation; thus, riparian veg- ability to complete their life cycle en- ty into the bay. Responding to claims etation is frequently reduced or elim- tirely in fresh water. For this species, from opponents that the sewer line inated by grazing and trampling. The headwater tributaries with perennial would pollute local bathing beaches, loss of riparian canopy increases the flow offer refugia where populations he wrote the following (Hihn, 1885): amount of solar radiation reaching the can persist during prolonged droughts stream channel, resulting in warmer when access to and from the sea is in- “Compare this with the foul water stream temperatures. Further, the loss termittent. of the lagoon which empties into of root structure from riparian vegeta- the bay near the railroad wharf, tion reduces the frequency of undercut Other Developmental Activities and into which the filth of a tan- banks and decreases channel complex- In addition to logging, mining, and nery is being discharged, or com- ity, often resulting in widening and grazing, all of which had broad re- pare it with the water of the San shallowing of the channel. gional impacts, other industries likely Lorenzo River, into which a tan- In upland areas, removal of vegeta- created additional impacts to salmon nery and other factories drain, tion and soil compaction by livestock habitats in coastal and inland water- and which for years has actual- leads to decreased infiltration of rain- sheds prior to the turn of the century ly received, and is now receiving, fall into soils and more rapid rout- though on a more local basis. Again a large part of the sewage of our ing of surface runoff to stream chan- activities in the Santa Cruz Moun- city, either directly or by perco- nels, leading to flashier stream flows. tain region are illustrative. Beginning lation, and still the lower part of When coupled with loss of riparian in the 1840’s, a number of tanneries this same river is a favorite bath- vegetation, the roots of which stabilize were established in the area, making ing place.” streambanks, this increased stream en- use of the bark of tanoak trees, Noth- ergy typically leads to increased ero- olithocarpus densiflorus, that grow in In addition to the tanning indus- sion and downcutting of stream chan- these mountains. Three of these were try, the Santa Cruz Mountains were nels, which can separate them from located in the lower San Lorenzo Riv- also home to the California Powder the adjacent floodplain. Collectively, er, a fourth in Scotts Valley a few kilo- Works, established in 1861 as the first these processes can lead to substan- meters inland, and a fifth in the near- and largest black powder manufactur- tially altered hydrologic regimes. Re- by town of Soquel (Lehmann, 2000). ing company west of the Mississip- duced subsurface and floodplain stor- By the late 1870’s, there were ten tan- pi River (Brown, 2008). Operated un- age of water during the winter months neries in operation, which collective- til 1914, the plant was located on the means that once perennial streams ly provided roughly half of the saddle San Lorenzo River about 6 km up- may become intermittent or ephemeral leather produced in California (Gen- stream of its mouth. To provide power during the dry months (Elmore, 1992). dron and Domhoff, 2009). to the plant, a diversion dam was con- The impacts of grazing on riverine Impacts of the tanning industry on structed 4 river km upstream, which habitats along California’s central and local streams and rivers are poorly routed water through a flume and tun- southern coasts are important to rec- documented but were likely substan- nel cut through a steep mountain spur ognize, particularly in the absence of tial. Large quantities of bark were re- to the Powder Works facilities. Dur- credible evidence suggesting that any quired for tanning, and by late in the ing periods of low flow, the diversion populations of any Pacific salmon spe- century, tanoak trees had been severe- dam had the capacity to divert all of cies persisted in these watersheds. If ly depleted (Jepsen, 1911; Lehmann, the San Lorenzo River water into the such habitats were ever conducive to 2000), requiring tanneries to look to flume, resulting in dewatering of sev-

81(1) 31 eral kilometers of stream. In response, directly into the ocean. Water quality ic fauna. In addition to the widespread the State Fish Commission eventual- problems persisted, however, as claims physical impacts of logging, mining, ly negotiated for the owner to install were later brought against the mill and grazing, streams and rivers were three fish ladders in the flume to facil- owners for allowing “lime water, coc- commonly dammed or straightened itate passage of fish around the dewa- culus indicus, factory refuse, and sub- for a variety of purposes and served as tered reach (CFC, 1892). This did not stances deleterious to fish” to run into dumping grounds for all manner of re- occur until 1891, thirty years after the Soquel Creek (Anonymous, 1897). fuse and waste, from sawdust to tan- flume was built. The Corralitos Paper and Board nery and paper mill effluent, not to Finally, a number of paper mills Mill (1880–1902) on Corralitos Creek mention raw sewage. were in operation in the Santa Cruz (Koch, 1973) likewise polluted the Except for the occasional mention of Mountain region prior to the turn of stream with its effluent. Local resi- impacts from these industries on fish- the century. The San Lorenzo Paper dents battled the mill owner for sev- eries in local newspapers and reports Mill, which operated intermittently be- eral years, eventually resulting in an from local game wardens published in tween 1861 and 1872, was located on indictment of the owner for “fouling the State Fish Commission’s biennial the San Lorenzo River not far down- the creek” with waste from his pa- reports, the documentation of damage stream from the Powder Works site. per mill. The Santa Cruz Surf (Anon- is limited. Nevertheless, all of these To provide water to power the mill, ymous, 1888) reported that “The wit- impacts had been ongoing for several the owners constructed a 5.2 m high nesses all agreed that there were fish decades prior to Jordan and Gilbert’s by 54.9 m wide channel-spanning in the stream before the mill was built (1881a) resolution of Pacific salmon dam across the San Lorenzo River, and that there are none there now.” taxonomy or the first attempts to char- as well as a 1 km flume to carry wa- Such cases were not unique to the acterize the fish faunas of these coast- ter to the mill site (Koch, 1973). The Santa Cruz region. For example, the al watersheds. Knowledge of these im- dam formed a lake that backed water San Francisco Call newspaper (Anon- pacts should serve as a sharp remind- up onto the Powder Works property, to ymous, 1892) decried the “flagrant vi- er of the aphorism “the absence of the dismay of the Powder Works own- olation of the fish laws” by Taylor Pa- evidence is not evidence of absence” er. Eventually, the two owners resolved per Mill on the eponymously named when evaluating historical evidence the issue by constructing a flume be- Papermill Creek (now known as La- for Pacific salmon occurrence in spe- tween the Powder Works and the pa- gunitas Creek) in Marin County, stat- cific regions or watersheds. per mill, which allowed the paper mill ing that the “slaughter of fish from the owner to tear down the dam. Nev- poisoned refuse of the mill was enor- Evolution of the Writings of ertheless, for the years it stood, it al- mous.” The paper described the scene Jordan, Gilbert, and Evermann most certainly blocked passage to the as follows: In light of the historical context pro- majority of spawning habitat for what vided above, it is instructive to re- were likely the largest populations of “Every season and almost every view what Jordan and his collabora- coho salmon and steelhead in the San- month, the trout seek to pass the tors (chiefly Gilbert and Evermann) ta Cruz Mountain region. mill on their way up the stream. wrote about the distribution of Pacific Effluent from the paper mill was al- And every season and almost ev- salmon species during the period be- most certainly dumped into the river, ery month hundreds and thousands tween 1881 and the early 1900’s. Dur- as was common practice. The South of the trout die in the attempt. ing this time, Jordan and colleagues Coast Paper Mill (1879–1924), locat- They are killed by the lime and published many papers and books that ed on Soquel Creek near the town of acid cast into the crystal stream included descriptions of the five west Soquel, discharged highly acidic efflu- with the refuse from Taylor’s Pa- coast salmon species and their putative ent that killed fish59 and created what a per-mill, and their dead carcasses freshwater ranges (e.g., Jordan and local newspaper described as an “un- strew the shores for miles below Gilbert, 1881a,b; Jordan and Gilbert, bearable nuisance to residents of the the mill. Since 1850 this thing has 1882; Jordan, 1888; Jordan, 1892a; neighborhood” (Anonymous, 1885f). been going on.” Jordan, 1894a; Jordan and Evermann, To remedy the situation, the mill own- 1896; Jordan and Evermann, 1902; ers financed construction of a 2.4 km This collection of examples, though Jordan, 1904a; Jordan and Evermann, flume that carried mill waste around it represents only a small fraction of 1904; Jordan, 1905; Jordan, 1907a; the town of Capitola and dumped it the myriad human-caused changes that and Jordan, 1907b). were underway in coastal watersheds In the latter publications, the spe- in California during the 1800’s, high- cies descriptions are taken nearly ver- 59Johnston, P. D. 1973. Aptos and the mid-San- ta Cruz County area from the 1890s through lights the fact that coastal watersheds batim from Jordan’s previous publi- World War II. Interviewed and edited by Eliza- in California were far from pristine cations, as many were updates to pre- beth Spedding Calciano. University of California when scientists first began undertaking viously published books or reports Santa Cruz Library. (avail. at http://library.ucsc. edu/reg-hist/johnston). studies of salmonids and other aquat- prepared for the California State Fish

32 Marine Fisheries Review and Game Commission. Nevertheless, China north to Bering Straits.” This small streams along the coast.” This following these writings through time, new language had not appeared in any description is repeated nearly verbatim one can observe subtle but significant previous Jordan publications and begs in subsequent writings through 1896 changes in species range descriptions the question: if Jordan and Evermann (e.g., Jordan, 1888, 1892a, 1894a; and that reflect both new information that had definitive evidence of the occur- Jordan and Evermann, 1896). Jordan had gathered—not all of which rence of Chinook salmon in the Ven- Then, quite abruptly, Jordan and Ev- was necessarily accurate—as well as tura River from the 1880 expedition, ermann (1902) in American Food and his evolving perceptions of the ranges as well as in San Luis Obispo Coun- Game Fishes revised their description, of these species. ty, as asserted in Jordan (1894b), then writing instead that it is the dog salm- I begin with Chinook salmon. As why were the authors now describing on, not the silver salmon, that predom- mentioned earlier, following their the southern end of the range as Mon- inates “in most of the streams along 1880 Pacific Coast expedition, Jordan terey Bay, which lies nearly 425 km to the coast.” Recall that during Gil- and Gilbert (1881a) wrote that Chi- the north? Bear in mind that, at the time bert and Snyder’s 1897 and 1899 sur- nook salmon were found “From Ven- this book was written, Jordan still ac- vey of Northern California and South- tura River northward to Behring’s tively opposed the “home stream the- ern Oregon coastal streams, they mis- Straits…” despite the fact that they ory” and assumed that salmon did not took yearling coho salmon for chum did not actually record collecting Chi- migrate more than a few tens of miles salmon. I suspect that Jordan and Ev- nook salmon anywhere south of Mon- from their natal streams (Jordan, 1903). ermann accepted Gilbert and Sny- terey Bay. References to Chinook Thus, if he and Evermann had first- der’s initial and erroneous conclusion salmon in the Ventura River remained hand knowledge of occurrence south that chum salmon were the most abun- in Jordan’s writings through publica- of Monterey Bay (i.e., in the Ventura dant salmon captured during these ex- tion of Jordan and Evermann’s (1896) River or in San Luis Obispo County), peditions (Snyder, 1907) and so re- monumental work The Fishes of North then it seems peculiar that they would vised their narrative. Curiously, subse- and Middle America.60 have changed their range description. quent editions of American Food and However, in Jordan and Evermann’s Jordan and Evermann continued Game Fishes (Jordan and Evermann, (1902) American Food and Game to publish these seemingly disparate 1904, 1908, 1922) retained this lan- Fishes, we find evidence that Jordan range descriptions in subsequent edi- guage, while Jordan’s (1904a) report had perhaps begun to doubt his pre- tions of American Food and Game to the California Fish and Game Com- vious assertion that Chinook salm- Fishes (Jordan and Evermann, 1904, mission reverted back to silver salm- on ranged as far south as the Ventu- 1908, 1922). But in the final version on as being “in most streams along the ra River. In their general description of Check List of the Fishes and Fish- coast.” In the latter case, this may have of Pacific salmon, Jordan and Ever- like Vertebrates of North and Middle resulted from Jordan simply recycling mann note that “Concerning the hab- America North of the Northern Bound- language from his previous reports to its and distribution of the salmon we ary of Venezuela and Colombia (Jor- the California Fish and Game Com- quote (with such modification as lat- dan et al., 1930), the range of Chi- mission (Jordan, 1888, 1892a, 1894a). er observations necessitate) the fol- nook salmon is listed simply as “Alas- A second example of where a spe- lowing, based upon investigation made ka, Oregon, and California, southward cies misidentification appears to have by Jordan, Evermann, and Gilbert.” to Monterey, and to northern China.” influenced Jordan’s range description This acknowledgment is followed by We cannot know with complete cer- emanates from the Carmel River Ex- a paragraph taken more or less ver- tainty if Jordan had come to disbelieve pedition. As noted earlier, fish collect- batim from previous papers (Jordan, whatever information led to his origi- ed from four Santa Cruz Mountain 1888, 1892a, 1894a), which includes nal Ventura River and San Luis Obis- streams in 1895 on the Carmel Riv- the statement that “Only the quinnat po descriptions, but it would certain- er Expedition were initially identified or king salmon has been noticed south ly seem so, especially in the absence as chum salmon but were later deter- of San Francisco. Its range has been of any subsequent information to con- mined to be coho salmon. In most of traced as far as Ventura River.” firm spawning populations of Chinook his earliest writings, Jordan stated that Yet later, in a section of the book de- salmon in these regions. “Only the quinnat [or king salmon] voted solely to Chinook salmon, the The writings of Jordan and his col- has been noticed south of San Francis- species’ range is described as follows: laborators related to coho and chum co” (e.g., Jordan and Gilbert, 1881b; “It is found on both coasts of the Pacif- salmon contain examples where new Jordan, 1888, 1892a, 1894a; Jordan ic, from Monterey Bay, California and but incorrect information appears to and Evermann, 1896, 1902). have influenced their range descrip- Yet, in 1904 and afterwards, Jordan 60The full title of this book was The Fishes of tions. In the earliest of these writ- changed this language to “Only the North and Middle America: A Descriptive Cat- ings, Jordan and Gilbert (1881b) note quinnat and dog salmon have been no- alogue of the Species of Fish-like Vertebrates that “the silver salmon [predominates] ticed south of San Francisco” (Jordan, Found in the Waters of North America, North of the Isthmus of Panama. in Puget Sound” and “in most of the 1904a, 1905, 1907a), though other

81(1) 33 Jordan publications retained the orig- remarked that “All the species have Commission written during this same inal language (e.g., Jordan and Ever- been seen by us in the Columbia and period state that the “blueback is not mann, 1908, 1914, 1922). Here, I sus- in Frazer’s River; all but the blue-back found in the Sacramento” and do not pect that Jordan modified this verbiage [sockeye] in waters tributary to Puget mention the occasional appearances to accommodate the apparent new, but Sound,” and in subsequent papers (Jor- (Jordan, 1904a, 1907b). incorrect, identification of chum salm- dan, 1892a) he expanded the state- Only Jordan’s descriptions of pink on from four Santa Cruz Mountain ment to “all but the blue-back in wa- salmon distribution remained fairly streams collected during the Carmel ters in the Sacramento and in waters consistent through time. Jordan and River Expedition. Had the coho salm- tributary to Puget Sound.” In each of Gilbert apparently acquired a pink on from these streams been correct- these descriptions he indicated that it salmon collected from the Sacramen- ly identified from the outset, it seems is the silver salmon that predominates to River while on their 1880 Pacif- likely that subsequent confusion about in Puget Sound. Ten years later, how- ic Coast Expedition (Jordan and Gil- whether coho salmon were native to ever, Jordan and Evermann (1902) in- bert, 1880)(Table 2), so from the be- Santa Cruz Mountain streams would dicated that all five species, including ginning Jordan describes their range never have occurred (see Spence et al., sockeye, have been seen in waters trib- as from the “Sacramento northward 2011). utary to Puget Sound, and in all sub- to the Arctic Sea,” though they quali- Additional uncertainty in the dis- sequent writings he indicated that both fy this by stating that it is “Occasion- tribution of coho salmon is evident “red and silver salmon” predominate ally seen in the Columbia and Sacra- through time in Jordan’s writings. In in Puget Sound (Jordan, 1904a, 1905, mento, but not sufficiently abundant to Jordan (1884a), he wrote that “The 1907a). Through time, it is apparent constitute a distinct run” (Jordan and Silver Salmon enters all the rivers that Jordan received new information Gilbert, 1881a). In later writings, Jor- from the Sacramento to Kamtchat- on both the distribution and abundance dan notes that pink salmon in the Sac- ka [sic].” Similarly, in two subsequent of sockeye salmon in the Puget Sound ramento River are referred to as “Lost reports prepared for the State Fish region. salmon” (Jordan and Evermann, 1896) Commission (Jordan, 1888, 1892a) Jordan and Evermann (1896) de- and that it is “rarely seen in Califor- he stated that they are found “in most scribed the range of sockeye salmon nia” (Jordan, 1907b), further suggest- streams along the coast.” Yet in Jordan as from “Klamath and Rogue rivers ing he did not believe them to occur in (1894a:131), following a description northward to Kamchatka and Japan,” the state on a regular basis. of coho salmon identical to the pre- representing Jordan’s first mention of vious two publications, one finds the sockeye salmon in either California or Conclusions puzzling statement that “This species Oregon waters. As noted earlier, Sny- The evolution in the writings of [Silver salmon] is not common south der (1931) found “nothing to substan- Jordan and colleagues regarding the of the Columbia, but is sometimes tak- tiate” the assertion of sockeye in the freshwater distribution of Pacific salm- en in California.” Klamath River and questioned whether on serves as an important reminder The implication that coho salmon these observations might be based on that, when seeking to determine histor- were not abundant in Oregon or north- collection of sea-run O. mykiss and O. ical ranges of species, holding up any ern California is surprising, consider- clarkii, which local fishermen referred individual reference as definitive evi- ing that by the late 1880’s canneries to as “blue-backs.” On the other hand, dence of occurrence is both risky and on the Nehalem, Nestucca, Siuslaw, Cobb (1921) reported that 20 sock- inappropriate. The fluidity—and in Umpqua, and Rogue Rivers, as well as eye salmon were taken from the Klam- some cases, apparent inaccuracy or in- Tillamook, Yaquina, Alsea, and Coos ath River in the autumn of 1915. All consistency—of Jordan’s Pacific salm- Bays in coastal Oregon were process- editions of American Food and Game on range descriptions over the years is ing between 1.5 and 2.5 million coho Fishes contain a general description of not surprising given the limited extent salmon annually (Meengs and Lack- Pacific salmon ranges that states “all of exploration and the primitive state ey, 2005). Jordan returned to his state- but the blueback” occurs in the Sacra- of knowledge related to Pacific salmon ment that coho salmon were “in most mento (Jordan and Evermann, 1902– that prevailed in the 1880’s. Thus, this streams along the coast” in subsequent 1923); however, in species-specific de- conclusion should in no way be con- writings. Nevertheless, this momen- scriptions within these books, they ac- strued as diminishing the credibility or tary vacillation suggests lingering un- knowledge that the sockeye salmon accomplishments of some of our most certainty about the distribution, identi- “has occasionally been reported from revered ichthyologists. The inaccura- ty, and abundance of various species. the Sacramento and Klamath rivers, cies and uncertainties merely reflect The progression of writings on but is not at all common south of the the natural and sometimes non-linear sockeye salmon demonstrates shifts in Columbia.” It appears by this time that progression of understanding in a sci- understanding through time regarding they viewed sockeye salmon as occa- entific discipline that was in its earli- occurrence. In Jordan’s earliest writ- sional visitors to California. Jordan’s est stages. ings (Jordan and Gilbert, 1881b) he periodic reports to the California Fish Jordan, Gilbert, Snyder, and other

34 Marine Fisheries Review 19th century scientists were pioneers ed difficulties discriminating among compounding questions of the repro- in every sense of the word. Conse- salmonid species using fish bones, ductive range of these species. quently, criticism should not be direct- as well as the fact that middens often Perhaps the last word on the chal- ed toward the fact that they occasion- contain little evidence of salmonids, lenges of defining species ranges ally got some things wrong. Rather, if even in areas where they were known should go to David Starr Jordan, who criticism is to be levied, it is with those to be abundant. In short, credible as- wrote in A Classification of Fishes of us who, in standing on the broad sessment of historical occurrence in a (Jordan, 1923), “I may repeat a warn- shoulders of these ichthyological lumi- particular watershed requires integrat- ing as old as science itself: that we naries, have relied wholly on their con- ing many different sources of informa- must not expect a degree of accuracy clusions without stopping to fully con- tion, rather than reliance on any single which the subject in question does not sider the nature of the scientific foun- information type (see Yoshiyama et al., permit.” dation upon which they stood. 2001; Alagona et al., 2012; and Ham- Resolving the historical rang- ilton et al., 2017 for excellent exam- Acknowledgments es of the five Pacific salmon species ples of integrated analysis of salmon I would like to express my sincere in western North America remains a and steelhead occurrence in California gratitude to Kit Johnston, librarian at challenging endeavor, and the writings river systems). the NMFS Southwest Fisheries Sci- of pioneering ichthyologists constitute An unfortunate reality is that, even if ence Center, Fisheries Ecology Di- only one source of information. Muse- all relevant evidence is gathered, it sim- vision, who acquired dozens of ob- um records may provide definitive ev- ply may not be possible to reconstruct scure historical references in support idence supporting occurrence in par- the historical distributions of certain of this research. Additionally, I thank ticular locations; however, even these salmonid species in some areas. This is Dave Catania, curator of the Califor- records must be used with caution, as in part because species were likely ex- nia Academy of Sciences Ichthyolog- specimens contained in museum col- tirpated from many watersheds by hu- ical Collection, who provided much lections may be misidentified and thus man activities before anyone had re- valuable information and granted ac- species determinations should be vali- corded their presence. For example, cess to museum specimens collect- dated through genetic or morphomet- although pink and chum salmon have ed by Jordan, Gilbert, and Snyder. I ric/meristic analysis (see footnote 35 been periodically observed in Califor- gratefully acknowledge Dave Rundio for example). nia streams throughout the last 150 and Tommy Williams, NMFS South- In some cases, it may be possi- years, there is little compelling evi- west Fisheries Science Center, with ble to extract important information dence to suggest that persistent repro- both of whom I had numerous help- from 19th century popular literature or ducing populations of either species ful discussions regarding the organi- newspaper accounts, though my expe- occurred in the state. This does not zation and content of this manuscript. rience has been that this information mean that they did not, as many of the Finally, I thank Mark Capelli, NMFS is seldom definitive due to vagaries in low-gradient habitats that would have West Coast Region, and two anony- taxonomy and nomenclature that ex- been the most likely spawning areas mous reviewers for their thorough and isted at the time. Despite this caveat, for these fish would have been severe- thoughtful reviews of earlier drafts of such writings may contain clues in an- ly impacted by logging, splash dam- this manuscript. cillary information such as the tim- ming, mining, and other human de- ing of migration or size of individu- velopments in the 1800’s. But build- Literature Cited als that may be useful in discriminat- ing the case for their occurrence based Adams, P. B., L. W. Botsford, K. W. Gobalet, R. A. Leidy, D. R. McEwan, P. B. Moyle, J. ing among species. Overlap in these on traceable historical evidence simply J. Smith, J. G. Williams, and R. M. Yoshiya- characteristics often precludes defini- may not be possible. ma. 2007. Coho salmon are native south of tive conclusions regarding species, but In these instances, we must be will- San Francisco Bay: a reexamination of North American coho salmon’s southern range lim- it may allow exclusion of some spe- ing to accept that the question is un- it. Fisheries 32(9):441–451 (doi: https://doi. cies and help establish the likelihood answerable and resist the temptation org/10.1577/1548-8446(2007)32[441:CSAN of others. to place undue weight on the accounts SO]2.0.CO;2). Alagona, P. S., S. D. Cooper, M. Capelli, M. Use of ethnographic and archaeo- of our scientific predecessors, howev- Stoecker, and P. Beedle. 2012. A history of logical information may also prove er esteemed they may be, simply be- steelhead and rainbow trout (Oncorhynchus informative, though both have their cause it is the only information avail- mykiss) in the Santa Ynez River watershed, Santa Barbara County, California. Bull. So. limitations. As noted by Alagona et able to us. Further, given the highly Calif. Acad. Sci. 111(3):163–222. al. (2012), Native American cultures migratory nature of the Pacific salm- Alpers, C. N., M. P. Hunerlach, J. T. May, and R. L. Hothem. 2005. Mercury contamination sometimes had different names for on, and salmonids generally, a fixed from historical gold mining in California. “trout” and “salmon,” but it is unclear freshwater spawning range for the spe- Publ. U.S. Geol. Surv. 61, 7 p. if salmon referred to actual salmon or cies over geologic periods of time, or Anonymous. 1871. Saw dust in running streams. Santa Cruz Sentinel 6(49):2 [20 May]. simply sea-run O. mykiss. Likewise, in response to shorter climatic chang- ______. 1877. The saw-dust nuisance. San- Gobalet et al. (2004) have highlight- es, should not be expected, further ta Cruz Sentinel 23(22):2 [3 Nov.].

81(1) 35 ______. 1880. The revision of the Salmon- ica. PLoS ONE 10(7):e0131276 (doi: https:// parks.ca.gov/pages/540/files/BigBasinRed- idae. Forest Stream 15(7):130. (Editor’s note doi.org/10.1371/journal.pone.0131276). woodsFinalWebLayout101816.pdf). preceding an article by D. S. Jordan.). Bendire, C. 1881. Notes on Salmonidae of ______. 2017. Cooper’s sawmill: historical ______. 1885a. “James Bolton caught for- the upper Columbia. Proc. U.S. Natl. Mus. landmark. Calif. State Parks Off. Hist. Pres- ty-eight salmon…” Santa Cruz Sentinel 4:81–87 (doi: https://doi.org/10.5479/ ervation (accessed 2 Nov. 2017 at http://ohp. 1(298):3 [3 Apr.]. si.00963801.4-196.81). parks.ca.gov/ListedResources/Detail/835). ______. 1885b. “Salmon have been quite Böhlke, J. 1953. A catalogue of the type spec- Carranco, L. 1982. Redwood lumber industry. plentiful in the Pescadero creek…” Santa imens of recent fishes in the Natural Histo- Golden West Books, San Marino, Calif., 218 Cruz Sentinel 1(300):1 [5 Apr.]. ry Museum of Stanford University. Stanford p. ______. 1885c. “F. Voron and J. Notley re- Ichthyol. Bull. 5:1–168. CFC. 1872. Report of the Commissioners turned Wednesday from a fishing expedition Boughton, D. A. 2000. The dispersal system of of Fisheries of the State of California for to Waddell’s creek. They caught seven salm- a butterfly: a test of source-sink theory sug- the years 1870 and 1871, 24 p. Calif. Fish on and one hundred and fifty trout….” Santa gests the intermediate-scale hypothesis. Comm., Sacramento. Cruz Sentinel 1(304):3 [10 Apr.]. Am. Nat. 156(2):131–144 (doi: https://doi. ______. 1877. Report of the Commission- ______. 1885d. The game law. Santa Cruz org/10.1086/303380). ers of Fisheries of the State of California for Sentinel 2(105):7 [19 Aug.]. ______, P. B. Adams, E. Anderson, C. Fu- the years 1876 and 1877, 30 p. Calif. Fish ______. 1885e. A high authority says trout saro, E. Keller, E. Kelley, L. Lentsch, J. Comm., Sacramento. and salmon trout are the same thing. Daily Nielsen, K. Perry, H. Regan, J. Smith, C. ______. 1880. Report of the Commission- Morning Astorian 23(69):3 [21 Mar.]. (With Swift, L. Thompson, and F. Watson. 2006. ers of Fisheries of the State of California for excerpts of letter from D. S. Jordan.). Steelhead of the south-central/southern Cal- the year 1880, 70 p. Calif. Fish Comm., Sac- ______. 1885f. Soquel siftings. Santa Cruz ifornia coast: population characterization ramento. Daily Surf 3(133):3 [12 Oct.]. for recovery planning. U.S. Dep. Commer., ______. 1882. Report of the Commission- ______. 1886. “Salmon are beginning to NOAA Tech. Memo. NMFS-SWFSC-394, ers of Fisheries of the State of California for make their appearance…” Santa Cruz Senti- 116 p. the years 1881 and 1882, 23 p. Calif. Fish nel 4(147):5 [6 Apr.]. Bowcutt, F. 2015. The tanoak tree: an environ- Comm., Sacramento. ______. 1887. A piscatorial curiosity. Santa mental history of a Pacific Coast hardwood. ______. 1892. Biennial report of the State Cruz Sentinel 8(10):3 [26 Oct.]. Univ. Wash. Press, Seattle, 240 p. Board of Fish Commissioners of the State of ______. 1888. The paper mill case. Santa Brewer, W. H. 1974. Up and down California in California for the years 1891 and 1892, 65 p. Cruz Surf 10(24):5 [29 May]. 1860–1864: the journal of William H. Brew- Calif. Fish Comm., Sacramento. ______. 1889. Anti-preserve men. Daily er, 3rd ed. (F. P. Farquhar, Editor.) Univ. Calif. CFGC. 1913. Fish and Game Commission twen- Alta Calif. 81(94):8 [2 Oct.]. Press, Berkeley, 583 p. ty-second biennial report for the years 1910– ______. 1890. “The State Board of Fish Brittain, M. R., and M. R. Jennings. 2008. Stan- 1912. Calif. Fish Game Comm. 80 p. Commissioners…” Santa Cruz Sentinel ford University’s John Otterbein Snyder: stu- ______. 1916. Fish and Game Commission 12(69):2 [7 Jan.]. dent, collaborator, and colleague of David twenty-fourth biennial report for the years ______. 1892. Fish killed by poison. San Starr Jordan and Charles Henry Gilbert. Mar. 1914–1916. Calif. Fish Game Comm. 248 p. Francisco Morning Call 72(86):7 [25 Aug.]. Fish. Rev. 70(1):24–29. Chamberlain, F. M. 1907. Some observations on ______. 1897. Defiling the stream. Santa Brown, B. 2008. The California Powder Works salmon and trout in Alaska. In Rep. Comm. Cruz Surf 28(145):4 [22 Dec.]. and San Lorenzo Paper Mill: self-guided tour. Fish. for the fiscal year 1906 and special pa- ______. 1899. “It is said that the upper wa- Paradise Park Masonic Club, Santa Cruz, Ca- pers. U.S. Bur. Fish. Doc. 627, 112 p, Wash., D.C. ters of Battle Creek…are teeming with silver lif. (excerpts avail. at http://www.santacruzpl. salmon…” Red Bluff News 39:3 [21 July]. Clark, G. H. 1929. Sacramento-San Joaquin org/history/articles/511/). salmon (Oncorhynchus tschawytscha) fish- ______. 1905. Our county fish hatchery. Brown, J. H., G. C. Stevens, and D. M. ery of California. Calif. Div. Fish Game, Fish Mountain Echo (Santa Cruz) 10(8):2 [16 Kaufman. 1996. The geographic range: size, Bull. 17, 73 p. Dec.]. shape, boundaries, and internal structure. Clark, W. B. 1970. Gold districts of California. ______. 1906. New additions to Supt. She- Annu. Rev. Ecol. Systematics 27(1996):597– Calif. Div. Mines Geol., Bull. 193, 186 p. bley’s family: dog salmon may be propagat- 623 (doi: https://doi.org/10.1146/annurev. Cleland, R. G. 1964. The cattle on a thousand ed at the hatchery—half-million king and ecolsys.27.1.597). hills. Southern California 1850–1880. Second quinaiutt [sic] salmon fry ready. Evening Burcham, L. T. 1957. California range land: an ed., second print. Huntington Lib., San Mari- Sentinel (Santa Cruz) 10(234):1 [7 Mar.]. historico-ecological study of the range re- no, Calif., 365 p. Bailey, J. 2008. The other California gold: Trin- source of California. Div. Forestry, Calif. Cobb, J. N. 1911. The salmon fisheries of the ity County placer mining, 1848–1962. U.S. Dep. Nat. Resour., Sacramento, Calif., 261 p. Pacific Coast. In Rep. Comm. Fish. for the Bur. Reclam., Denver, Colo., 103 p. (online at Burroughs, R. D. 1961. The natural history of fiscal year 1910 and special papers. U.S. Bur. http://odp.trrp.net/DataPort/doc.php?id=387). the Lewis and Clark Expedition. Mich. State Fish. Doc. 751, 179 p., Wash., D.C. Baird, S. F. 1874. Part II. C. Multiplication of Univ. Press, East Lansing, 340 p. ______. 1921. Pacific salmon fisheries. Ap- fish in general. In Rep. Comm. 1872–1873, p. Busby, P. J., T. C. Wainwright, G. J. Bryant, L. pendix I to Rep. U.S. Comm. Fish. for 1921. xxxi–xcli. U.S. Comm. Fish Fish. Wash, D.C. J. Lierheimer, R. S. Waples, F. W. Waknitz, U.S. Bur. Fish. Doc. 902., 268 p., Wash., D.C. Bean, T. H. 1880. Check-list of duplicates of and I. V. Lagomarsino. 1996. Status review of Collins, J. W. 1892. Report on the fisheries of North American fishes distributed by the west coast steelhead from Washington, Idaho, the Pacific Coast of the United States. In Rep. Smithsonian Institution in behalf of the Unit- Oregon, and California. U.S. Dep. Commer., Comm. Fish Fish. for 1888, p. 3–269. U.S. ed States National Museum, 1877–1880. NOAA Tech. Memo. NMFS-NWFSC-27, 261 Comm. Fish Fish. Wash., D.C. Proc. U.S. Natl. Mus. 3:75–116. p. Costanso, M. 1911. The Portola expedition of ______. 1896. Report on the propaga- Butler, V. L. 2004. Where have all the native fish 1769–1770, diary of Miguel Costanso (transl. tion and distribution of food fishes. In Rep. gone? The fate of fish that Lewis and Clark and edited by M. B. Carpio and F. J. Teggart). Comm. for the year ending June 30, 1894, p. encountered on the lower Columbia River. Univ. Calif. Press, Berkeley, 167 p. 20–80. U.S. Comm. Fish Fish., Wash., D.C. Oreg. Hist. Quar. 105(3):438–463. Coues, E. (Editor). History of the expedition un- Behnke, R. J. 2002. Trout and salmon of North California State Legislature. 1889. Chapter der the command of Lewis and Clark, to the America. Free Press, N.Y., 359 p. LXV. An Act to amend an Act entitled “An sources of the Missouri River, thence across Beidleman, R. G. 2006. California’s frontier nat- Act to establish a Penal Code,” approved Feb- the Rocky Mountains and down the Colum- uralists. Univ. Calif. Press, Berkeley, 484 p. ruary 14, 1872, relating to the depositing of bia River to the Pacific Ocean performed dur- Bellinger, M. R., M. A. Banks, S. J. Bates, E. sawdust in the waters of this state [approved ing the years 1804-5-6 by order of the gov- D. Crandall, J. C. Garza, G. Sylvia, and P. March 4, 1889], p. 61–62. Twenty-eighth ses- ernment of the United States, Vol. 3, p. 821- W. Lawson. 2015. Geo-references, abun- sion of the State Legislature, Sacramento. 1298. Francis P. Harper, N.Y. dance calibrated ocean distribution of Chi- California State Parks. 2011. Big Basin Red- Crespi, J. 2001. A description of distant roads: nook salmon (Oncorhynchus tshawytscha) woods State Park. Calif. State Parks, Sacra- original journals of the first expedition into stocks across the west coast of North Amer- mento (accessed 2 Nov. 2017 at https://www. California, 1769–1770 (transl. and edited by

36 Marine Fisheries Review A. K. Brown). San Diego State Univ. Press., tics in Santa Cruz. Westview Press, Boulder, Hist. Biol. 32(3):509–555 (doi: https://doi. 848 p. Colo., 256 p. org/10.1023/A:1004784904703). Croker, R. S. 1930. Unusual fish at San Pedro, Gibbons, W. P. 1854. Description of four new Jennings, M. R. 1997. Barton Warren Ever- California. Calif. Fish Game 16(1):77. species of viviparous fish, read before the mann (1853–1932) and his contribution to ______. 1936. King salmon in southern California Academy of Natural Sciences, North American ichthyology. In T. W. Pietsch California. Calif. Fish Game 22(4):323. Monday evening, 15 May 1854. Daily Plac- and W. D. Anderson, Jr. (Editors). Collection CSDC. 2012. Historical census populations of er Times and Transcript (San Franc.), 18 May, building in ichthyology and herpetology, p. California, counties, and incorporated Cit- p. 2. 291–310. Am. Soc. Ichthyol. Herpetol., Spec. ies, 1850–2010. Demographic Res. Unit, Ca- ______. 1855. Untitled description of a Publ. 3, Lawrence, Kan. lif. Dep. Finance, Calif. State Data Cent. (ac- new species of trout (Salmo iridea), in record ______. 2014. Cloudsley Louis Rutter cessed 2 Nov. 2017 at http://www.dof.ca.gov/ of 19 March 1855 meeting. Proc. Calif. Acad. (1867–1903): pioneer salmon biologist and Reports/Demographic_Reports/). Nat. Sci. 1:36–37. resident naturalist, fisheries steamer Alba- Cutright, P. R. 1969. Lewis and Clark: pioneer- ______. 1876. Description of a new species tross. Mar. Fish. Rev. 76(4):1–27 (doi: https:// ing naturalists. Univ. Nebraska Press, Lin- of trout from Mendocino County. Proc. Calif. doi.org/10.7755/MFR.76.4.1). coln, 506 p. Acad. Sci. Ser. 1(VI):142–144. Jepsen, W. L. 1911. Tanbark oak and the tanning Dall, W. H. 1915. Spencer Fullerton Baird: a bi- Gibson, J. R. 2013. California through Russian industry. In California tanbark oak, p. 5–23. ography including selections from his corre- eyes 1806–1848. Early California Commen- U.S. For. Serv. Bull. 75. spondence with Audubon, Agassiz, Dana, and taries, vol. 2. Arthur H. Clark Co., Norman, Johnson, O., W. S. Grant, R. G. Kope, K. Neely, others. J. P. Lippincott Co., Phila. Lond., 462 492 p. F. W. Waknitz, and R. S. Waples. 1997. Sta- p. Gilbert, C. H. 1912. Age at maturity of the Pa- tus review of chum salmon from Washington, Dunn, J. R. 1996. Charles Henry Gilbert (1859– cific coast salmon of the genusOncorhyn - Oregon, and California. U.S. Dep. Commer., 1928): an early fishery biologist and his con- chus. U.S. Bur. Fish. Bull. 32:1–22. NOAA Tech. Memo. NMFS-NWFSC-32, 280 tributions to knowledge of Pacific salm- ______. 1922. The salmon of the Yukon p. on (Oncorhynchus spp.). Rev. Fish. Sci. River. U.S. Bur. Fish. Bull. 38:317–332. ______, A. Elz, J. Hard, and D. Stewart. 4(2):133–184 (doi: https://doi.org/10.1080/10 Gilbert, K. G. 1917. Hydraulic mining debris in 2012. Why did the chum cross the road? Ge- 641269609388583). the Sierra Nevada. U.S. Geol. Surv. Prof. Pap. netics and life history of chum salmon in the Eigenmann, C. H. 1890. The food fishes of the 105, 154 p. southern portion of their range. N. Pac. Anad. California fresh waters. In Biennial report of Gill, T. 1883. An account of progress in zoology Fish Comm. Tech. Rep. 8:135–137. the State Board of Fish Commissioners of the in the year 1881. Smithsonian Rep. for 1881, Jordan, D. S. 1878a. A catalogue of the fishes of State of California for the years 1888–1890, Wash., D.C., 90 p. the fresh waters of North America. Extracted p. 53–65. Calif. Fish Comm., Sacramento. Gobalet, K. W., P. D. Schulz, P. D. Wake, and N. from U.S. Geol. Geogr. Surv. Bull. 4(2):407– Elder, D., B. Olson, A. Olson, J. Villeponteaux, Siefkin. 2004. Archaeological perspectives on 442. and P. Brucker. 2002. Salmon River subba- Native American fisheries of California, with ______. 1878b. Notes on a collection of sin restoration strategy: steps to recovery and emphasis on steelhead and salmon. Trans. fishes from Clackamas River, Oregon. Proc. conservation of aquatic resources. Rep. prep. Am. Fish. Soc. 133(4):801–833. U.S. Natl. Mus. 1:69–85 (doi: https://doi. for Klamath River Basin Fisheries Restora- Goetzmann, W. H. 1993. Exploration and em- org/10.5479/si.00963801.20.69). tion Task Force (Interagency Agreement 14- pire. The explorer and the scientist in the ______. 1880. The revision of the Salmoni- 48-11333-98-H019). U.S. Fish Wildlife Serv. winning of the West. Hist. Book Club, N.Y., dae. Forest Stream 15(7):130. Office, Yreka, Calif., 52 p. 656 p. ______. 1884a. The salmons of the Pacif- ic. In G. B. Goode (Editor), The fisheries and Elmore, W. 1992. Riparian responses to graz- Gustafson, R. G., R. S. Waples, J. M. Myers, L. ing practices. In R. J. Naiman (Editor), Wa- fishery industries of the United States; Sec- A. Weitkamp, G. J. Bryant, O. W. Johnson, tion I, Natural history of useful aquatic ani- tershed management: balancing sustainabil- and J. J. Hard. 2007. Pacific salmon extinc- mals, p. 474–479. U.S. Comm. Fish Fish., ity and environmental change, p. 442–457. tions: quantifying lost and remaining diversi- Wash., D.C. Springer-Verlag, N.Y. ty. Cons. Biol. 21(4):1009–1020 (doi: https:// ______. 1884b. Manual of the vertebrates Ettlinger, E., G. Andrew, P. Doughty, and V. doi.org/10.1111/j.1523-1739.2007.00693.x). of the northern United States including the Rogers. 2015. Adult salmonid monitoring in Hallock, C. 1877. The sportsman’s gazetteer and district east of the Mississippi River, and the Lagunitas Creek watershed—2014–2015. general guide. Forest Stream Publ. Co., N.Y., north of North Carolina and Tennessee, ex- Rep. to Calif. Dep. Fish Wildl., Fish. Restora- 688 p. clusive of marine species. Fourth ed. Jansen, tion Grant Program, Grant #P1130417, 23 p. Hamilton, J. B., D. W. Rondorf, W. R. Tinnis- McClurg & Co., Chicago, 406 p. Marin Municipal Water District, Corte Made- wood, R. J. Leary, T. Mayer, C. Gavette, and ______. 1887. The fisheries of the Pacific ra, Calif. L. A. Casal. 2017. The persistence and char- Coast. In G. B. Goode (Editor), The fisher- Evermann, B. W. 1930. David Starr Jordan, the acteristics of Chinook salmon migrations to ies and fishing industries of the United States. man. Copeia 1930(4):93–106 (doi: https:// the upper Klamath River prior to exclusion Section II, Geographic review for the year doi.org/10.2307/1436463). by dams. Oreg. Hist. Q. 117(3):327–377. 1880, p. 591–630. U.S. Comm. Fish Fish., ______and H. W. Clark. 1931. A distribu- Hihn, F.A. 1885. Pacific Avenue sewer. Santa Wash., D.C. tional list of the species of freshwater fishes Cruz Sentinel 4(8): 5 [22 Oct.]. ______. 1888. The salmon family. In Sci- known to occur in California. Calif. Div. Fish Holt, R. D. 1996. Adaptive evolution in source– ence sketches, p. 35–82. McClurg & Co., Game, Fish Bull. 35, 67 p. sink environments: direct and indirect ef- Chicago. Fages, P. 1937. A historical, political, and natu- fects of density-dependence on niche evolu- ______. 1891. Steelhead or salmon trout: ral description of California (transl. by H. I. tion. Oikos 75(2):182–192 (doi: https://doi. a lean, spent fish, that should be reject- Priestly). Univ. Calif. Press, Berkeley, 83 p. org/10.2307/3546242). ed by canners. San Francisco Morning Call Flagg, T. A., F. W. Waknitz, D. J. Maynard, G. B. Hubbs, C. L. 1946. Wandering of pink salmon 70(161): 12 [excerpts from a communication Milner, and C. V. W. Mahnken. 1995. Impacts and other salmonid fishes into southern Cali- from David Starr Jordan to the Call; 15 Nov.]. of hatcheries on native coho salmon popula- fornia. Calif. Fish Game 32(2):81–86. ______. 1892a. Salmon and trout of the tions in the lower Columbia River. Am. Fish. ______. 1964. History of ichthyology in the Pacific Coast. In Biennial report of the State Soc. Symposium 15:366–357. United States after 1850. Copeia 1964(1):42– Board of Fish Commissioners of the State of Ford, C. 1966. Where the sea breaks its back: 60 (doi: https:doi.org/10.2307/1440831). California for the years 1891–1892, p. 44–58. the epic story of a pioneer naturalist and the Jackson, D. D., and M. L. Spence (Editors). Calif. Board Fish. Comm., Sacramento. discovery of Alaska. Little, Brown and Com- 1970. The expeditions of John Charles Fre- ______. 1892b. History of zoological ex- pany, Boston, 206 p. mont, vol. 1. Univ. Illi. Press, Urbana, 854 p. plorations of the Pacific Coast. Calif. Fish Fremont, J. C. 1856. The exploring expedition to Jackson, W. F. 1991. Big River was dammed. Game 17(2):156–158. the Rocky Mountains, Oregon and California. FMMC Books, Mendocino, Calif., 165 p. ______. 1894a. Salmon and trout of the Pa- Miller, Ortin & Mulligan, N.Y., 456 p. Jackson, J. R., and W. C. Kimler. 1999. Taxon- cific Coast. In Thirteenth biennial report of Gendron, R., and G. W. Domhoff. 2009. The omy and the personal equation: the historical the State Board of Fish Commissioners of the leftmost city: power and progressive poli- fates of Charles Girard and Louis Agassiz. J. State of California, for the years 1893–1894,

81(1) 37 p. 125–141. Calif. Board Fish. Comm., Sac- ______and ______. 1881b. Ob- McDonald, M. 1892. Report of the Commis- ramento. servations on the salmon of the Pacif- sioner. In Rep. Comm. Fish Fish. for 1888, p. ______. 1894b. Notes on the fresh-water ic. Am. Nat. 15:177–186 (doi: https://doi. ix–xxxix. U.S. Comm. Fish Fish. Wash., D.C. fishes of San Luis Obispo, County. Bull. U.S. org/10.1086/272773). McElhany, P., M. H. Ruckelshaus, M. J. Ford, Fish Comm. 14:141–142. ______and ______. 1882. Synop- T. C. Wainwright, and E. P. Bjorkstedt. 2000. ______. 1896. The salmon family. In Sci- sis of the fishes of North America. Bull. Viable salmonid populations and the recovery ence sketches (new, enl. ed.), p. 35–82. A.C. U.S. Natl. Mus. 16, 1018 p. (doi: https://doi. of evolutionarily significant units. U.S. Dep. McClurg & Co., Chicago. org/10.5479/si.03629236.16.i). Commer., NOAA Tech. Memo. NMFS-NWF- ______. 1903. The parent-stream theory of ______and H. W. Henshaw. 1878. Report SC-42, 156 p. the return of salmon. Pop. Sci. Mon. 64:48– upon the fishes collected during the years McEvoy, A. F. 1986. The fisherman’s problem: 52. 1875, 1876, and 1877, in California and Ne- ecology and law in the California fisheries, ______. 1904a. Pacific species of salm- vada, appendix K. In Annu. Rep. U.S. Chief 1850–1980. Camb. Univ. Press, Camb., 369 on and trout. In Eighteenth biennial report of of Engineers for 1878, part III, p. 1609–1622, p. the State Board of Fish Commissioners of the Wash., D.C. McPhail, J. D., and C. C. Lindsey. 1970. Fresh- State of California for the years 1903–1904, Kaczynski, V. W., and F. Alvarado. 2006. As- water fishes of northwestern Canada and p. 75–97. Calif. Board Fish Comm., Sacra- sessment of the southern range limit of Alaska. Bull. Fish. Res. Board Can. 173, 381 mento. North American coho salmon: difficulties p. ______. 1904b. The parent-stream theory in establishing natural range boundaries. Meehan, W. R. (Editor). 1991. Influences of for- of the return of salmon. In Eighteenth bien- est and rangeland management on salmo- Fisheries 31(8):374–391 (doi: https://doi. nid fishes and their habitats. Am. Fish. Soc., nial report of the State Board of Fish Com- org/10.1577/1548-8446(2006)31[374:AOTS missioners of the State of California for the Bethesda, Md., Spec. Publ. 19, 751 p. RL]2.0.CO;2). Meengs, C.C., and Lackey, R.T. 2005. Estimat- years 1903–1904, p. 98–102. Calif. Board Kelley, R. L. 1954. Forgotten giant: the hydrau- Fish Comm., Sacramento. ing the size of historical Oregon salmon runs. lic gold mining industry in California. Pac. Rev. Fish. Sci. 13(1):51–66 (doi: https://doi. ______. 1905. A guide to the study of fish- Hist. Rev. 23(4):343–356. es. Vol II. Henry Holt and Co., N.Y., 599 p. org/10.1080/10641260590921509). ______. 1989. Battling the inland sea: Miller, D. J., and R. N. Lea. 1972. Guide to the ______. 1907a. Fishes. Henry Holt and Co, floods, public policy, and the Sacramento Val- N.Y., 789 p. coastal marine fishes of California. Calif. ley. Univ. Calif. Press, Berkeley, 395 p. Dep. Fish Game, Fish Bull. 157, 249 p. ______. 1907b. The trout and salmon of the Kerry, M. 1874. The Salmonidae of the Pacific. Pacific Coast. In Nineteenth biennial report Minckley, W. L., D. A. Hendrickson, and C. E. Forest Stream 2(24):369–370. Bong. 1986. Geography of western North of the State Board of Fish Commissioners of Koch, M. 1973. Santa Cruz County, parade of the State of California for the years 1905– American fishes: description and relation- the past. Valley Publ., Fresno, Calif., 254 p. ships to intracontinental tectonism. In C. H. 1906, p. 77–92. Calif. Board Fish Comm., Krasheninnikov, S. P. 1764. The history of Ka- Sacramento. Hocutt and E. O. Wiley (Editors), The zooge- mtschatka, and the Kurilski Islands, with the ography of North American freshwater fishes, ______. 1922. The days of a man, volume countries adjacent (transl. from Russ. by J. one: being memories of a naturalist, teacher, p. 519–613. John Wiley & Sons, N.Y. Grieves), R. Raikes, Gloucester, Scotland. and minor prophet of democracy, 1851–1899, Mitchell, M. D. 1994. Land and water poli- 280 p. vol. I. World Book Co., Yonkers-on-Hudson, cies in the Sacramento-San Joaquin Delta. Geogr. Rev. 84(4):411–423 (doi: https://doi. N.Y., 710 p. Krause, A., P. Wilcock, and D. Gaeuman. 2010. One hundred and fifty years of sediment ma- org/10.2307/215756). ______. 1923. A classification of fishes in- Moore, J. A. 1986. Zoology of the Pacific Rail- cluding families and genera as far as known. nipulation on the Trinity River, CA. Proceed- ings of the 2nd Joint Federal Interagency Con- road Surveys. Am. Zool. 26(2):331–341. Stanford Univ. Publ., Univ. Ser., Biol. Sci. Murphy, J. M. 1879. Rambles in north-western 3(2): 77–243. ference on Sedimentation and Hydrologic Modeling, June 27–July10, Las Vegas, NV America from the Pacific Ocean to the Rocky ______and B. W. Evermann. 1896. The Mountains. Chapman and Hall, Lond., 364 p. fishes of North and Middle America. A de- (avail. at https://acwi.gov/sos/pubs/2ndJFIC/ Contents/3D_Krause_3_1_10.pdf). Myers, J. M., R. G. Kope, G. J. Bryant, D. Teel, scriptive catalogue of the species of fish- L. J. Lierheimer, T. C. Wainwright, W. S. like vertebrates found in the waters of North Larson-Praplan, S. 2016. History of rangeland livestock production. In M. R. George (Ed- Grant, F. W. Waknitz, K. Neely, S. T. Lind- America, north of the Isthmus of Panama. ley, and R. S. Waples. 1998. Status review Part I. Bull. U.S. Natl. Mus., No. 47, 954 p. itor), Ecology and management of annual rangelands, p. 60–72. Dep. Plant Sci., Univ. of Chinook salmon from Washington, Idaho, Gov. Print. Off., Wash., D.C. Oregon, and California. U.S. Dep. Commer. ______and ______. 1904–1923. Calif., Davis (avail. at http://rangelandar- chive.ucdavis.edu/files/252894.pdf). NOAA Tech. Memo. NMFS-NWFSC-35, 443 American food and game fishes. A popular p. account of all the species found in Ameri- Lehmann, S. 2000. Economic development of the city of Santa Cruz 1850–1950. In Fully Payne, S. M. 1978. A howling wilderness: a his- ca north of the equator, with keys for ready tory of the Summit Road area of the San- identification, life histories and methods of developed context statement for the City of Santa Cruz, p. 10–14. Prep. for City of Santa ta Cruz Mountains 1850–1906. Loma Prieta capture. Doubleday, Page & Co., N.Y., 572 Publ. Co., Santa Cruz, Calif., 159 p. p. (Note: book was reissued many times be- Cruz Plan. Develop. Dep., Santa Cruz, Calif. (excerpts avail. at http://scplweb.santacruzpl. Pearcy, W. G. 1992. Ocean ecology of North Pa- tween 1902 and 1923). cific salmonids. Univ. Wash. Press, Seattle, ______and ______and H. W. Clark. org/history/work/edindtan.shtml). 179 p. 1930. Check list of the fishes and fishlike ver- Leitritz, E. 1970. A history of California’s fish Penaluna, B. E., A. Abadia-Cardoso, J. B. Dun- tebrates of North and Middle America north hatcheries 1870–1960. Calif. Dep. Fish ham, F. J. Garcia De Leon, R. E. Gress- of the northern boundary of Venezuela and Game, Fish Bull. 150, 125 p. well, A. R. Luna, E. B. Taylor, B. B. Shepa- Columbia. In Rep. U.S. Comm. Fish. for the Lichatowich, J. A. 1999. Salmon without rivers. rd, R. Al-Chokhachy, C. C. Muhlfeld, K. R. fiscal year 1928, Part II, p. 1–670. U.S. Bur. Island Press, Wash., D.C., 317 p. Bestgen, K. Rogers, M. A. Escalante, E. R. Fish. Wash., D.C. Littlepage, D. 2006. Steller’s Island: adventures Keeley, G. M. Temple, J. E. Williams, K. R. ______and C. H. Gilbert. 1880. List of of a pioneer naturalist in Alaska. Mountain- Matthews, R. Pierce, R. L. Mayden, R. P. Ko- the fishes of the Pacific Coast of the Unit- eers Books, Seattle, 237 p. vach, J. C. Garza, and K. D. Fausch. 2016. ed States, with a table showing the distri- Lockington, W. N. 1880. Report on the edi- Conservation of native Pacific trout in west- bution of the species. Proc. U.S. Natl. Mus. ble fishes of the Pacific Coast, U.S.A. In Re- ern North America. Fisheries 41(16):286–300 3:452–458 (doi: https://doi.org/10.5479/ port of the Commissioners of Fisheries of the (doi: https://doi.org/10.1080/03632415.2016. si.00963801.3-173.452). State of California for the year 1880, p. 16– 1175888). ______and ______. 1881a. Notes 66. Calif. Fish. Comm., Sacramento. PFMC. 1999. Amendment 14 to the Pacific on the fishes of the Pacific coast of the Unit- ______. 1882. Sketch of the progress of Coast salmon plan, appendix A: identification ed States. Proc. U.S. Natl. Mus. 4:29–70 North American ichthyology in the years and description of essential fish habitat, ad- (doi: https://doi.org/10.5479/si.00963801.4- 1880–1881. Am. Nat. 16(10):765–772 (doi: verse impacts, and recommended conserva- 191.29). https://doi.org/10.1086/273178). tion measures for salmon. Pac. Fish. Manage.

38 Marine Fisheries Review Counc., Portland, Oreg. (https://www.pcoun- Sexton, J. P., P. J. McIntyre, A. L. Angert, and ______. 1885. Report of operations at the cil.org/wp-content/uploads/99efh1.pdf). K. J. Rice. 2009. Evolution and ecology United States salmon-breeding station on the Pietsch, T. W., and J. R. Dunn. 1997. Early col- of species range limits. Annu. Rev. Ecol., McCloud River, California, during the year lection building in Puget Sound and adjacent Evolution, and Systematics 40:415–436 1883. In Report of the Commissioner for waters: the 1880 expedition of David Starr (doi: https://doi.org/10.1146/annurev.ecol- 1883, p. 989–1000. U.S. Comm. Fish Fish., Jordan (1851–1931) and Charles Henry Gil- sys.110308.120317). Wash., D.C. bert (1859–1928). In T. W. Pietsch and W. D. Shapovalov, L., and A. C. Taft. 1954. The life ______. 1897. The artificial propagation Anderson, Jr. (Editors), Collection building histories of the steelhead rainbow trout (Sal- of salmon on the Pacific Coast of the Unit- in ichthyology and herpetology, p. 279–290. mo gairdneri gairdneri) and silver salmon ed States with notes on the natural history of Am. Soc. Ichthyol. Herpetol., Spec. Publ. 3, (Oncorhynchus kisutch) with special refer- the quinnat salmon. U.S. Fish Comm. Bull. Lawrence, Kan. ence to Waddell Creek, California, and rec- 16:205–235. Platts, W. S. 1991. Livestock grazing. In W. R. ommendations regarding their management. Suckley, G. 1874. On the North American spe- Meehan (Editor), Influences of forest and Calif. Dep. Fish Game, Fish Bull. 98, 303 p. cies of salmon and trout. In Rep. Comm. rangeland management on salmonid fishes Shebley, W. H. 1922. A history of fishcultur- Fish Fish. for 1872 and 1873, p. 91–160. U.S. and their habitats, p. 389-423. Am. Fish. Soc. al operations in California. Calif. Fish Game Comm. Fish Fish.,Wash., D.C. (Note: report Spec. Publ.19. 8(2):62–99. was originally written in 1861.) Poli, A. 1956. Ownership and use of for- Skiles, T. D., R. M. Yoshiyama, and P. B. Moyle Sumner, F. H., and O. R. Smith. 1940. Hydraulic est land in northwestern California. Land 2013. Pink salmon (Oncorhynchus gorbus- mining and debris dams in relation to fish life Econ. 32(2):144–151 (doi: https://doi. cha) in the Salinas River, California: new re- org/10.2307/3159757). cord and historical perspectives. Calif. Fish in the American and Yuba rivers of Califor- Pulliam, H. R. 1988. Sources, sinks, and popu- Game 99(1):55–59. nia. Calif. Fish Game 26(2):2–22. lation regulation. Am. Nat. 132(5):652–661 Snyder, J. O. 1907. The fishes of the coastal Swift, C., T. R. Haglund, M. Ruiz, and R. N. (doi: https://doi.org/10.1086/284880). streams of Oregon and northern California. Fisher. 1993. The status and distribution of Quinn, T. P. 1984. Homing and straying in Pacif- U.S. Bur. Fish. Bull. 28:153–189. the freshwater fishes of southern California. ic salmon. In J. D. McCleave (Editor), Mech- ______. 1913.The fishes of the streams Bull. S. Calif. Acad. Sci. 92(3):101–167. anisms of migration in fishes, p. 357–362. tributary to Monterey Bay, California. U.S. Thompson, W. F. 1922. The fisheries of Cal- Plenum Press, N.Y. Bur. Fish. Bull. 32:47–72. ifornia and their care. Calif. Fish Game ______. 2005. The behavior and ecology of ______. 1931. The salmon and the fishery 8(3):165–177. Pacific salmon and trout. Univ. Wash. Press, of Klamath River. Calif. Div. Fish Game, Fish Trotter, P. C., and P. A. Bisson. 1988. The histo- Seattle, 378 p. Bull. 34, 122 p. ry of the discovery of the cutthroat trout. Am. Ravenel, W. de C. 1896. Report on the propaga- Spence, B. C., G. A. Lomnicky, R. M. Hughes, Fish Soc. Symposium 4:8–12. tion and distribution of food-fishes. In Rep. and R. P. Novitzki. 1995. An ecosystem ap- USFC. 1874. Report of the Commissioner for Comm. for the year ending June 30, 1895, p. proach to salmonid conservation. ManTech 1872 and 1873. In Rep. Comm. Fish Fish., 6–72. U.S. Comm. Fish Fish., Wash., D.C. Environ. Res. Serv. Corp., Rep. TR-4501- p. i–xcii. U.S. Comm. Fish Fish., Wash., D.C. ______. 1898. Report on the propaga- 96-60577, Corvallis, Oreg., 356 p. (avail. at ______. 1876. Report of the Commission- tion and distribution of food-fishes. In Rep. http://www.westcoast.fisheries.noaa.gov/pub- er for 1873–4 and 1874–5. In Rep. Comm. Comm. for the year ending June 30, 1897, p. lications/reference_documents/the_mantech_ Fish Fish., p. vii–xlvi. U.S. Comm. Fish Fish., xviii–xc. U.S. Comm. Fish Fish., Wash., D.C. report.html). Wash., D.C. ______. 1899. Report on the propaga- Spence, B. C., W. G. Duffy, J. C. Garza, B. C. ______. 1893. Report of the United States tion and distribution of food-fishes. In Rep. Harvey, S. M. Sogard, L. A. Weitkamp, T. H. Commissioner of Fish and Fisheries for the Comm. for the year ending June 30, 1898, p. Williams, and D. A. Boughton. 2011. Histor- fiscal years 1890–90 and 1890–91. In Rep. xxxi–cxxii. U.S. Comm. Fish Fish., Wash., ical occurrence of coho salmon (Oncorhyn- Comm. for 1889 to 1891, p. 1–96. U.S. D.C. chus kisutch) in streams of the Santa Cruz Comm. Fish Fish., Wash., D.C. Rivinus, E. F., and E. M. Youssef. 1992. Spen- Mountain region of California: response to ______. 1894. Report of the United States cer Baird of the Smithsonian. Smithson. Inst. an Endangered Species Act petition to del- Commissioner of Fish and Fisheries for the Press., Wash., D.C., 228 p. ist coho salmon south of San Francisco Bay. fiscal year ending June 30, 1892. In Rep. Rood, R.C. 1975. The historical geography and U.S. Dep. Commer., NOAA Tech. Memo. Comm. for the year ending June 30, 1892, p. environmental impact of the lumber industry NMFS-SWFSC-472, 113 p. vii–cxci. U.S. Comm. Fish Fish., Wash., D.C. of the Santa Cruz Mountains. Senior thesis, Stanger, F. 1963. South from San Francisco. Waples, R. S., G. R. Pess, and T. Beechie. Univ. Calif. Santa Cruz, 83 p. San Mateo Co. Hist. Asso., San Mateo, Ca- 2008. Evolutionary history of Pacific salm- Rutter, C. 1902. A natural history of the quinnat lif., 214 p. on in dynamic environments. Evolutionary salmon: a report of investigations in the Sac- ______. 1967. Sawmills in the redwoods. Applications 1(2):189–206 (doi: https://doi. ramento River, 1896–1901. Bull. U.S. Fish San Mateo Co. Hist. Asso., San Mateo, Ca- org/10.1111/j.1752-4571.2008.00023.x). Comm. 22:65–141. lif., 160 p. Winans, G. A., D. Viele, A. Grover, M. Palm- ______. 1903. The Pacific salmons. Coun- Stearley, R. F. and G. R. Smith 1993. Phylog- try Life in Am. 4(2):124–127. eny of the Pacific trouts and salmons (On- er-Zwahlen, D. Teel, and D. Van Doornik. Scofield, N. B. 1916. The humpback and dog corhynchus) and genera of the family Sal- 2001. An update of genetic stock identi- salmon taken in the San Lorenzo River. Calif. monidae. Tran. Am. Fish. Soc. 122(1):1– fication of Chinook salmon in the Pacific Fish Game 2(1):41. 33 (doi: https://doi.org/10.1577/1548-8659 Northwest: test fisheries in California. Rev. Scofield, W. L. 1937. A silver salmon at (1993)122<0001:POTPTA>2.3.CO;2). Fish. Sci. 9(4):213–237 (doi: https://doi. Los Coronados Islands. Calif. Fish Game Stone. L. 1875. Sacramento salmon. In Report org/10.1080/20016491101753). 23(3):245. of the Commissioners of Fish and Fisheries Wydoski, R. S., and R. R. Whitney 2003. Inland Scott, W. B., and E. J. Crossman. 1973. Fresh- of the State of California for the years 1874 fishes of Washington, Second ed. Univ. Wash. water fishes of Canada. Bull. Fish. Res. and 1875, p. 16–21. (Letter sent to the Chair- Press, Seattle, 322 p. Board Can. 184, 966 p. man of the California Fish Commission, San Yoshiyama, R. M., E. R. Gerstung, F. W. Fisher, Sedell, J. R., F. N. Leone, and W. S. Duval. Francisco). and P. B. Moyle. 2001. Historical and present 1991. Water transportation and storage. In W. ______. 1880. Notes on letter of Mr. Hor- distribution of Chinook salmon in the Central R. Meehan (Editor), Influences of forest and ace Dunn to Prof. Spencer F. Baird, dated San Valley drainage of California. In R. L. Brown rangeland management on salmonid fishes Francisco, September 26, 1876. In Report of (Editor), Contributions to the biology of Cen- and their habitats, p. 325–368. Am. Fish. Soc. the Commissioner for 1878, p. 817–818. U.S. tral Valley salmonids, vol.1., p. 71–176. Ca- Spec. Publ. 19. Comm. Fish Fish., Wash., D.C. lif. Dep. Fish Game, Fish Bull. 179.

81(1) 39