Data & Analytics

Identification of salmon species from archaeological remains on the Northwest Coast

Description
Identification of salmon species from archaeological remains on the Northwest Coast
Published
of 14
6
Published
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Similar Documents
Share
Transcript
  17 Chapter 2 Identication of Salmon Speciesfrom Archaeological Remains on theNorthwest Coast Trevor J. Orchard, Department of Anthropology, McMaster University Paul Szpak, Department of Anthropology, The University of Western Ontario   Introduction Fish were universally an important resource to Northwest Coast First Nations ethnograph-ically (e.g., Donald 2003; Suttles 1990), and a wide variety o sh species are represented inarchaeological assemblages rom the culture area (e.g., Butler and Campbell 2004; Hanson1991; Orchard and Clark 2005). Te most commonly utilized taxa prehistorically and eth-nographically tend also to be those species that have been prominent in modern commer-cial sheries, namely salmon ( Oncorhynchus spp.), Pacic cod ( Gadus macrocephalus ) andother gadids (Gadidae), herring ( Clupea pallasii ), and halibut ( Hippoglossus stenolepis ).Salmon have been particularly prominent in discussions o Northwest Coast culture his-tory (Butler and Campbell 2006; Coupland, Stewart, and Patton 2010; Schalk 1977), andin some areas they have been a major component o traditional diets or as many as 9000years (Butler and O’Connor 2004; Cannon 1991; Cannon and Yang 2006). Although theNorthwest Coast is thus oen characterized as dominated by salmon shing (Butler andCampbell 2006; Donald 2003; but see Monks 1987), the region is perhaps better charac-terized by considerable variability in the local use o sh species (Cannon 2000; Orchard2007, 2009). Even in areas where salmon was a predominant resource ethnographically,it is increasingly evident that there is considerable variability in the timing o the adop-tion o salmon-ocused economies. Documenting and understanding this variability inboth the adoption o salmon-ocused economies and the use o other sh taxa is critical  18  the  archaeology  of  north  pacific  fisheries or understanding variability in Northwest Coast cultural developments, and more gener-ally or understanding the rise o the sedentary, socially complex cultures documentedethnographically (e.g., Ames 2003; Leposky et al. 2005; Prentiss et al. 2007). Insights intolong-term patterns o sh use are also becoming increasingly important as baselines orunderstanding more recent historic patterns in sheries and sh populations (Jackson etal. 2001; McKechnie 2007; Pauly et al. 2002).Salmon pose a particular problem or archaeological analysis. A thorough under-standing o the timing, technology, and logistics o prehistoric salmon shing dependslargely on the species o salmon involved (Kew 1992). Unortunately, as with other sets o closely related animal species (Bochenski 2008), salmon bones are notoriously dicult toidentiy to species, particularly when the bulk o salmon assemblages tends to consist o  vertebrae (Butler and Bowers 1998; Cannon 1988). Pacic Salmon Life Histories Te various species o Pacic salmon typically spawn at dierent ages and at dierent adultsizes (able 2.1; Hart 1973; Healey 1986; Quinn 2004). In coastal British Columbia, pinksalmon ( Oncorhynchus gorbuscha ), with only very rare exceptions, spawn at two years o age and at mean weights o roughly 2.2 kg. Coho salmon ( O. kisutch ) spawn at ages rang-ing rom two to six years, with the vast majority (95 percent) spawning during their third tbl 2.1. Summary data on the sizes and lie history characteristics o Pacifc salmon SpeciesLength atmaturityWeight at maturity*Age atspawningFreshwaterOceanic Pink( O. gorbuscha )48.5 to 57.2 cm ( ♂ )46.0 to 56.1 cm ( ♀ )Mean ( ♀ ): 52 cmTypical: 1.4–2.3 kg (3–5 lbs)Mean: 2.2 kg (4.8 lbs)Max: 5.5 kg (12 lbs)2 yearsRapid migrationto oceanollowingemergence o ryReside in ocean18 monthsChum( O. keta )51.6 to 88.9 cm ( ♂ )55.5 to 77.6 cm ( ♀ )Mean ( ♀ ): 68 cmTypical: 4.5–6.8 kg (10–15 lbs)Mean: 5.4 kg (11.7 lbs)Max: 15 kg (33 lbs)2–7 years;usually 3–5yearsRapid migrationto oceanollowingemergence o ryReside in ocean2–7 years(usually 3–5)Coho( O. kisutch )21.0 to 72.7 cm ( ♂ )27.8 to 70.0 cm ( ♀ )Mean ( ♀ ): 64 cmTypical: 2.7–5.4 kg (6–12 lbs)Mean: 4.1 kg (9 lbs)Max: 14 kg (31 lbs)2–6 years;usually 3 yearsTypically residein streams or 1or 2 yearsTypically spend1–2 years at seaSockeye( O. nerka )30.0 to 73.7 cm ( ♂ )46.6 to 72.4 cm ( ♀ )Mean ( ♀ ): 55 cmTypical: 2.3–3.6 kg (5–8 lbs)Mean: 2.7 kg (6.0 lbs)Max: 6.8 kg (15 lbs)3–8 years;usually 4 or 5yearsTypically residein lakes or riversor 1–3 yearsReside in ocean1–4 yearsChinook( O. tshawytscha )10.2 to 115.0 cmMean ( ♀ ): 87 cmTypical: 4.5–6.8 kg (10–15 lbs)Mean: 6.8 kg (15 lbs)Max: 57 kg (125 lbs)3–8 years;usually 4 or 5yearsRapid migrationor 1 year spentin riversReside in ocean1–5 years*With the exception o pink salmon, mean weights are rom Kew (1992) or Fraser River stocks specifcally. These provide some generalindications o average sizes. Source : Compiled rom Hart (1973), Healey (1986), Kew (1992), and Quinn (2004).  Identication of Salmon Species from Archaeological Remains on the Northwest Coast 19 year o lie, and are slightly larger than pinks, averaging 4.1 kg at maturity. Chum salmon( O. keta ) spawn at ages ranging rom two to seven years, with the vast majority spawningat three to ve years o age, and are larger still, averaging 5.4 kg at maturity. Finally, sock-eye salmon ( O. nerka ) and chinook salmon ( O. tshawytscha ) generally spawn at our orve years o age, though both species can spawn at ages ranging rom three to eight years.Sockeye are relatively small, spawning at a mean size o roughly 2.7 kg, while chinook arethe largest o the Pacic salmon, reaching a mean size o 6.8 kg at maturity. Importantly,the lie histories o these ve species o Pacic salmon consist o variable amounts o timespent in reshwater and saltwater environments. Pink and chum salmon spend the major-ity o their lives in saltwater, chinook may spend their rst year in reshwater, coho typi-cally spend one to two years in reshwater, and sockeye typically reside in reshwater orthe rst one to three years o lie. In act, some sockeye populations, known as “kokanee,”are permanently resident in reshwater environments (Quinn 2004:15).wo other species o salmon are also represented along the Pacic coast. Te steelhead( O. mykiss ) and the coastal cutthroat trout ( O. clarkii ) are somewhat less common thanthe ve species discussed above, and they do not orm the massive spawning runs typicalo the other Pacic salmon species. While specimens o steelhead and cutthroat may, andundoubtedly have, turned up in archaeological assemblages in low numbers, they will notbe considered urther here. Species Identication of Salmon Assemblages: Previous Approaches Te variable spawning ages o the dierent species o Pacic salmon led Cannon (1988) topropose a method o ageing salmon remains based on examination o growth lines or annuliin vertebrae. Cannon’s experiments indicated that these annuli were easily visible throughradiographic analysis, and relatively large numbers o vertebrae could be examined quickly and easily on X-ray plates. Given the tendency o the various salmon species to spawn atdierent ages, the resulting age proles could be used to iner the probable species composi-tions o the assemblages under consideration. Surprisingly, relatively ew researchers haveemployed this technique subsequent to Cannon’s (1988) initial article, though the appli-cation o this approach has increased slightly in recent years. Berry (2000; Hayden 1997)employed radiography to iner the species o salmon represented in assemblages rom theKeatley Creek site. Notably, Berry’s interpretations have been challenged by more recentancient DNA (aDNA) analysis o salmon remains rom that site (Speller, Yang, and Hayden2005). More recently, Coupland and colleagues (Coupland, Colten, and Conlogue 2002)and rost (2005) utilized the radiographic technique to examine the composition o salmonassemblages rom sites in the Prince Rupert Harbour area and in the Gul o Georgia, respec-tively. Most recently, Clark (2007) employed radiographic analysis o salmon vertebrae romsites in the Gul Islands o the southern Gul o Georgia to explore the rise o ree netting inthat area. Tese applications o Cannon’s (1988) radiographic technique highlight the poten-tial utility o this approach, and the relative ease with which it can be applied.In recent years, aDNA analysis has also been increasingly used to identiy salmonspecies rom archaeological vertebrae (Butler and Bowers 1998; Cannon and Yang 2006;  20  the  archaeology  of  north  pacific  fisheries Speller, Yang, and Hayden 2005; rost 2005; Yang, Cannon, and Saunders 2004). Tisapproach has proven very successul, with archaeological salmon vertebrae generally showing excellent DNA preservation and acilitating the secure species identication o small samples o vertebrae rom a number o sites. While aDNA analysis is the most accu-rate means o speciating salmon bones, it is prohibitively expensive in many cases. Teincreasing study o aDNA, however, has led to a renement in the more cost-eectivemethod o radiography, and it has also highlighted the potential o using metric charac-teristics o salmon vertebrae to provide additional inormation on the species compositiono archaeological assemblages. Specically, Cannon and colleagues have compared theresults o aDNA analysis o salmon vertebrae rom Namu to the results obtained throughradiographic analysis and to measurements o the transverse diameter o the vertebrae(Cannon and Yang 2006; Yang, Cannon, and Saunders 2004; also see rost 2005). woparticular observations arising rom these comparisons are o interest here. First, aDNAanalysis o Namu salmon vertebrae revealed that pink salmon was the most commonspecies represented in the assemblage, contributing roughly 42 percent o the analyzed vertebrae (Cannon and Yang 2006:128). In contrast, previous analyses based entirely onradiography had concluded that pink salmon contributed less than 5 percent o the salmon vertebrae at Namu (Cannon 1988:107). Pink salmon generally spawn during their secondyear, and thus their vertebrae should show only a single winter annulus. However, Cannonand Yang (2006:128) ound that both a modern comparative specimen o pink salmon andmany archaeological vertebrae identied by aDNA analysis as pink salmon showed twogrowth lines. Tis is in accordance with biological studies that indicate that pink salmonoen exhibit a supplementary growth check line on their scales that can be conused witha winter annulus (Bilton and Ricker 1965). Second, through the comparison o the metrictraits o vertebrae to the species determination o those vertebrae through aDNA analysis,Cannon and Yang (2006:132) ound that vertebrae with a transverse diameter o 10.5 mmor greater were all identied as either chum or chinook salmon, while vertebrae with atransverse diameter o 8.0 mm or less were all identied as either pink or sockeye salmon.Vertebrae alling between 8.0 and 10.5 mm derived rom the ull range o species exceptchinook (Cannon and Yang 2006:132). Tese results have been expanded by subsequentdata rom Namu that indicate that vertebrae less than or equal to 8.0 mm are overwhelm-ingly dominated by pink and sockeye salmon, but also contain small numbers o chumand coho (Aubrey Cannon, pers. comm., May 2010). Species Identication of Salmon Assemblages: Moving Forward Te recent applications o Cannon’s (1988) radiographic technique, combined with theinsights arising rom applications o aDNA analysis (Cannon and Yang 2006) outlinedabove, provide a basis or employing a more rened technique combining radiography andmetric analysis. While Cannon and Yang (2006), or example, suggest that the conusionarising rom the tendency o pink salmon vertebrae to display two radiographic annulireduces the utility o the radiographic method, we argue that the realization that pinksalmon can exhibit one or two apparent growth annuli under radiographic analysis can  Identication of Salmon Species from Archaeological Remains on the Northwest Coast 21 improve the radiographic technique and allow more accurate assessment o the possiblerange o species present in an archaeological assemblage. Similarly, the apparently consis-tent patterning in metric characteristics (i.e., transverse vertebral diameter) exhibited by the dierent salmon species identied by Cannon and Yang (2006) provides a means orproviding better estimates o the composition o large assemblages o salmon vertebrae(e.g., rost 2005).Consideration o a limited number o modern comparative specimens generally sup-ports the metric observations o Cannon and Yang (2006). ransverse vertebral width wasmeasured or all vertebrae rom twenty individual Pacic salmon rom the zooarchaeo-logical comparative collections o McMaster University, the University o oronto, and theUniversity o Victoria (able 2.2). Te sample sizes or some species, particularly chum(one individual, sixty-one total vertebrae), are small, and the comparative samples as awhole likely do not encompass the ull range o sizes o each species (able 2.1). Measure-ments o additional comparative specimens in the uture will help to rene this approach.Nevertheless, these data generally support the size categories suggested by Cannon andYang (2006). For example, only specimens o chum and chinook include vertebrae greaterthan 10.5 mm, while neither chum nor chinook include vertebrae that all below 8.0 mmin width. Only two aspects o these comparative data are dierent rom the results pre-sented by Cannon and Yang (2006). First, chinook salmon represent a relatively widerange o sizes, and need to be included in the 8.0 mm to 10.5 mm size category. Cannonand Yang’s (2006) ailure to include chinook in this category resulted rom the very limitednumber o chinook present in their sample (only a single vertebra measuring 18.0 mm).Similarly, coho are well represented in the 8.0 mm or less size category among the com-parative specimens measured or this paper, while the Namu coho sample was again toosmall to represent this diversity (Cannon and Yang 2006). As indicated above, subsequentanalysis o Namu samples has identied coho in the small size category (Aubrey Cannon,2010, pers. comm.). Notably, small numbers o chum vertebrae also all into the small sizecategory in these more recent Namu analyses, though the proportion o chum vertebraeless than or equal to 8.0 mm is very small (Aubrey Cannon, 2010, pers. comm.). Overall,the suggestion o a tripartite division o salmon vertebrae by size is supported.In Cannon’s (1988) srcinal application o the radiographic technique, he used a por-table dental X-ray machine to generate radiographs o salmon vertebrae on dental X-ray  tbl 2.2. Metric data rom modern comparative salmon specimens o spawning age SpeciesIndividuals(n)Vertebrae(n)Meantransversediameter (mm)StandarddeviationRange (mm) Chinook743712.461.129.82–15.37Chum16110.090.648.84–10.86Coho53038.300.846.08–9.78Pink31796.861.354.55–8.94Sockeye42427.890.546.55–8.96
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x