BULLETIN FISHING ON THE FRONTIER: VERTEBRATE REMAINS FROM AMOTAPE, SICHES, AND HONDA PHASE OCCUPATIONS AT SITIO SICHES (PV 7-19), PERÚ Elizabeth J. Reitz, Daniel H. Sandweiss, and Nicole R. Cannarozzi Vol. 56, No. 4, pp. 109–181 January 30, 2019 ISSN 2373-9991 UNIVERSITY OF FLORIDA GAINESVILLE The FLORIDA MUSEUM OF NATURAL HISTORY is Florida’s state museum of natural history, dedicated to understanding, preserving, and interpreting biological diversity and cultural heritage. The BULLETIN OF THE FLORIDA MUSEUM OF NATURAL HISTORY is an on-line, open-ac- cess, peer-reviewed journal that publishes results of original research in zoology, botany, paleontology, archaeology, and museum science. New issues of the Bulletin are published at irregular intervals, and volumes are not necessarily completed in any one year. Volumes contain between 150 and 300 pages, sometimes more. The number of papers contained in each volume varies, depending upon the number of pages in each paper, but four numbers is the current standard. Multi-author issues of related papers have been published together, and inquiries about putting together such isues are welcomed. Address all inqui- ries to the Editor of the Bulletin. Richard C. Hulbert Jr., Editor Bulletin Committee Richard C. Hulbert Jr. Jacqueline Miller Larry M. Page David W. Steadman Roger W. Portell, Treasurer Jonathon I. Bloch, Ex officio Member ISSN: 2373-9991 Copyright © 2019 by the Florida Museum of Natural History, University of Florida. All rights reserved. Text, images and other media are for nonprofit, educational, and personal use of students, scholars, and the public. Any commercial use or republication by printed or electronic media is strictly prohibited without written permission of the museum. Publication Date: January 30, 2019 This and all other issues of the Bulletin can be freely downloaded at: http://www.flmnh.ufl.edu/bulletin/publications/ Send communications about this publication to: Editor of the Bulletin; Florida Museum of Natural History; University of Florida; P.O. Box 117800; Gainesville, FL 32611-7800 USA FAX: 352-846-0287; Email: bulletin@flmnh.ufl.edu URL: https://www.flmnh.ufl.edu/bulletin/home/ mailto:bulletin@flmnh.ufl.edu FISHING ON THE FRONTIER: VERTEBRATE REMAINS FROM AMOTAPE, SICHES, AND HONDA PHASE OCCUPATIONS AT SITIO SICHES (PV 7-19), PERÚ Elizabeth J. Reitz,1 Daniel H. Sandweiss2, and Nicole R. Cannarozzi3 1Georgia Museum of Natural History, University of Georgia, Athens, GA 30602-1882, USA. 2Department of Anthropology and Climate Change Institute, South Stevens Hall, University of Maine, Orono, ME 04469, USA. 3Division of Environmental Archaeology, Dickinson Hall, Florida Museum of Natural History, University of Florida, Gainesville, FL 32611-7800, USA. Reitz, E. J., D. H. Sandweiss, and N. R. Cannarozzi. 2019. Fishing on the frontier: vertebrate remains from Amotape, Siches, and Honda Phase occupations at Sitio Siches (PV 7-19), Perú. USA. Bulletin of the Florida Museum of Natural History 56(4):109–181. ABSTRACT This study provides zooarchaeological data and interpretations of cultural and environmental changes between ~10,700 and 5200 cal B.P. by comparing diachronic trends within a single coastal site (Sitio Siches) with synchronic trends across multiple coastal sites in different, adjacent ecological regions. Sitio Siches (PV 7-19), located at 4° 24.7’ S on the Pacific coast of Perú, is a stratified site with an occupational sequence extending from the Early to the Later Preceramic period. Radiocarbon dates divide the occupation into three cultural phases: Amotape (9500–9000 14C B.P./~10,700–10,100 cal B.P.), Siches (7100–6000 14C B.P./~7900–6800 cal B.P.), and Honda (5150–4550 14C B.P./~5800–5200 cal B.P.). The total Sitio Siches assemblage contains 27,180 vertebrate specimens (NISP) and the remains of an estimated 2,759 minimum number of non-human vertebrate individuals (MNI). Marine vertebrate individuals dominate the assemblage. A temporal trend within the assemblage suggests changes in fishing strategies that may reflect changes in marine conditions. These measures are variable within the Siches-phase occupation, likely reflecting responses to environmental and cultural transitions between 7100 and 6000 14C B.P. Overall the results show that people living at the site during the Amotape phase experienced warm marine waters. During the Siches phase, coastal waters were even warmer, an interpretation consistent with isotopic values in sea catfish otoliths. Variations in fishing strategies during the Siches phase may be responses to sharper annual or seasonal fluctuations in Sea Surface Temperature than were experienced during the Amotape or Honda occupations. The Amotape collection from Sitio Siches offers a glimpse of an early fishing tradition on the northern coast of Perú. The Honda collection reflects modern conditions at the frontier between the Peru-Chilean and Panamanian provinces. A comparison of Sitio Siches with other vertebrate assemblages from within the Peru-Chilean province finds broad similarities among assemblages at similar latitudes throughout the province. This suggests that the patterns observed in these assemblages are not due exclusively to cultural dynamics, but also reflect larger environmental patterns influencing the economies of all human communities in the region. Key words: Sitio Siches, Peruvian coast, climate change, vertebrate zooarchaeology, Preceramic period, fishing strategies. 110 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) INTRODUCTION Archaeological studies of early- and mid-Holocene coastal environments of Perú shows that this period was more dynamic than once thought (e.g., Sand- weiss et al., 1996, 2007; Fontugne et al., 1999; Reitz, 2001; Reitz and Sandweiss, 2001; Andrus et al., 2002; Richardson and Sandweiss, 2008; Carré et al., 2014). This is particularly evident in archae- ological materials excavated from sites south of 5° S latitude, in what is today a warm-temperate water regime dominated by the Peru Current and subject to occasional El Niño/Southern Oscilla- tions (ENSO; Fig. 1; Tables 1–2). These studies also demonstrate that a regional economy based on marine resources flourished along the coast at least from the Late Pleistocene, that subsequent societal developments reflect ecological as well as cultural influences, and that the modern resource base pre- vailing today was established ca. 5000 14C B.P. (5700 cal B.P.).1 Although evidence for a regional economy and a dynamic early- and mid-Holocene environ- ment is strong, zooarchaeological data are limited for stratified sites bridging the transition from the Late Pleistocene ecological regime to the modern one (ca. 10,000–5000 14C B.P. [11,400–5700 cal B.P.]). Sitio Siches (PV 7-19), located at 4° 24.7ʹ S on the Pacific coast of northern Perú, contains a stratigraphic sequence of archaeological mate- rials deposited between 9500 and 4550 14C B.P. (~10,700–5200 cal B.P.). These materials offer the opportunity to: (1) study fishing strategies on the north coast during the early- and mid-Holocene TABLE OF CONTENTS Introduction ........................................................................................................... 110 Resource Area ....................................................................................................... 115 Defining Ecological Groups .................................................................................. 115 Summary of Other Zooarchaeological Assemblages ............................................121 Sitio Siches ...........................................................................................................127 Site Description ................................................................................................127 Field Methods ..................................................................................................128 Structures, Plant Remains, Geochemistry, and Molluscs ................................130 Zooarchaeological Methods .............................................................................131 A Preview of the Sitio Siches Results ...................................................................135 Results, Amotape Phase (Analytical Unit 4) ........................................................136 Results, Siches Phase (Analytical Units 3 and 2) .................................................136 Results, Honda Phase ............................................................................................145 Summary of Sitio Siches Results ..........................................................................152 Fishing on the Frontier: Sitio Siches .....................................................................153 Sitio Siches in the Broader Context ......................................................................163 Conclusions ...........................................................................................................166 Acknowledgements ...............................................................................................167 Footnotes ...............................................................................................................167 Literature Cited .....................................................................................................168 Appendix 1. Vertebrate Ecological Groups ..........................................................175 REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 111 using vertebrate remains recovered from strati- fied deposits and (2) correlate these strategies with environmental and cultural transitions. We elabo- rate here on our previous assessments of fishing strategies and other aspects of cultural life in a changing environment using additional and revised data from Sitio Siches which strengthen our argu- ments for cultural and environmental dynamism Figure 1. Map of study region. Stars denote modern cities, dots denote archaeological sites. 112 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) Ta bl e 1. In fo rm at io n ab ou t s ite s r ef er en ce d in th e te xt . N o 14 C d at es a re a va ila bl e fo r A lto S al av er ry , w hi ch is d at ed b y as so ci at io n w ith si m ila r s ite s a nd a rti fa ct s. Th e da te fo r L as V eg as d oe s n ot in cl ud e e ar lie r, Pr e- Ve ga s d ep os its , w hi ch ar e n ot in cl ud ed in th es e f au na l s am pl es . A ge ra ng es ar e f ro m th e m ea ns o f o ld es t a nd m os t r ec en t 14 C d at es w he re a va ila bl e, d is co un tin g ou tli er s r ej ec te d by th e ex ca va to rs . C al ib ra te d da te s w er e ru n us in g C al ib 7 .1 (S tu iv er a nd R ei m er , 1 99 3) , S H ca l1 3 cu rv e (H og g et a l., 2 01 3) , a nd ro un de d to n ea re st 1 00 y ea rs . T he 1 0, 57 5 da te fo r t he R in g Si te is a sh el l d at e ca lib ra te d us in g ht tp :// ca lib .o rg /m ar in e/ . Pr ov in ce a nd S ite D ep os it da te s,14 C B .P . D ep os it da te s, ca l B .P . R ef er en ce s So ut he rn P er u- C hi le an P ro vi nc e (1 7° –1 2° S ) Q ue br ad a Ja gu ay , S ec to r I I ( 16 ° 3 0ʹ S ) 10 ,8 67 –1 0, 25 0 12 ,7 00 –1 1, 90 0 (1 2, 00 0 fr om sh or t-l iv ed sa m pl es ) Sa nd w ei ss e t a l., 1 99 8; M cI nn is , 1 99 9; S an dw ei ss , 2 01 4; Jo ne s et a l., 2 01 7; R ei tz e t a l., 2 01 7 R in g Si te (1 7° 4 0ʹ S ) 10 ,5 75 (m ar in e) ; 87 55 –5 06 0 98 00 –5 80 0 Sa nd w ei ss e t a l., 1 98 9; R ei tz e t a l., 2 01 5; R ei tz e t a l., 2 01 6; R ei tz e t a l., 2 01 7 Pa lo m a (P ro ba bi lit y sa m pl es ) ( 12 ° 2 5ʹ S ) 77 35 –4 78 0 85 00 –5 50 0 B en fe r, 19 84 ; R ei tz , 1 98 8a , 1 98 8b , 2 00 3 N or th er n Pe ru -C hi le an P ro vi nc e (1 0° –8 ° S ) Pa ijá n (1 0° S ) 10 ,3 80 –8 26 0 12 ,2 00 –9 20 0 Lu m br er as , 1 97 4: 28 ; C ha uc ha t, 19 76 , 1 97 8, 1 98 8, 1 99 2; W in g, 19 86 A lm ej as (6 .3 5 m m fr ac tio n) (9 ° 4 0ʹ S ) 70 00 79 00 T. P oz or sk i a nd S . P oz or sk i, 19 84 , R ei tz , 1 99 5; S . P oz or sk i a nd T. P oz or sk i, 20 03 O st ra B as e C am p (8 ° 5 5ʹ S ) 62 50 –5 45 0 71 00 –6 20 0 Sa nd w ei ss e t a l., 1 98 3; S an dw ei ss e t a l., 1 99 6; R ei tz , 2 00 1; R ei tz a nd S an dw ei ss , 2 00 1 A lto S al av er ry (8 ° 1 0' S ) 44 50 –3 75 0 50 00 –4 00 0 Po zo rs ki , 1 97 6; S . P oz or sk i a nd T . P oz or sk i, 19 79 Pa na m an ia n Pr ov in ce (4 °3 0ʹ –2 ° S ) La s V eg as (O G SE 80 ) ( 2° 1 4ʹ S ) 10 ,0 00 –6 60 0 11 ,4 00 –7 50 0 B yr d, 1 97 6, 1 99 6; C ha se , 1 98 8; S to th er t, 19 88 S. S ic he s, A m ot ap e Ph as e (4 ° 2 5ʹ S ) 95 00 –9 00 0 10 ,7 00 –1 0, 10 0 R ic ha rd so n, 1 97 3, 1 97 8; R ic ha rd so n et a l., 1 99 6; S an dw ei ss , 20 03 S. S ic he s, Si ch es P ha se (4 ° 2 5ʹ S ) 71 00 –6 00 0 79 00 –6 80 0 R ic ha rd so n, 1 97 3, 1 97 8; R ic ha rd so n et a l., 1 99 6; S an dw ei ss , 20 03 S. S ic he s, H on da P ha se (4 ° 2 5ʹ S ) 51 50 –4 55 0 58 00 –5 20 0 R ic ha rd so n, 1 97 3, 1 97 8; R ic ha rd so n et a l., 1 99 6; S an dw ei ss , 20 03 R ea l A lto (2 ° 2 3 ʹ S ) M id dl e Va ld iv ia B yr d, 1 97 6, 1 99 6 REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 113 (Sandweiss et al., 1996, 1997; Andrus et al., 2002). A dynamic relationship exists among humans as biological and cultural organisms, the environments in which they live, and archaeologi- cal deposits. Human economies reflect the environ- ments in which they occur and alterations in long- established fishing habits at coastal sites may be evidence of cultural responses to changes in those environments. Cultures and environments change for many interrelated and complex reasons. Some of these changes may be stimulated primarily by human behavior. For example, a new technology might influence which resources are used or which resources are preferred or shunned. Likewise, non- anthropogenic factors, such as a change in coastal topography after a storm, may alter the availability and behaviors of economically important plants and animals. Evidence for these changes are recorded in archaeological assemblages, though it may be difficult to determine which is cause and which is a consequence. The case for environmental change is strengthened if a corpus of changes is observed in the biological remains recovered from a site over time or from several sites within a region. In the case of Sitio Siches, changes in local use of fishes preferring warmer or cooler waters could indicate a cultural change or a change in fish availability due to altered coastal water con- ditions. A patterned shift among groups of indica- tor animals, instead of in just one or two species, is more likely to represent changes in availability than changes in human tastes or choices. A review of published data from the region shows changes in ecological groups of indicator vertebrates recov- ered from Peruvian and Ecuadorian coastal sites located between 17° S and 2° S latitude and occu- pied between 10,867 and 3750 14C B.P. (~12,700– 4000 cal B.P., or 12,000–4000 cal B.P. using the most accurate Quebrada Jaguay dates; Figs. 1–2; Tables 1–2; Sandweiss et al., 1996; Reitz, 2001; Andrus et al., 2002; Reitz et al., 2015; Jones et al., 2017).2 Faunal assemblages from sites located between 17° S and 12° S latitude show a persis- tent focus on cool-water vertebrates. Vertebrate assemblages from sites located between 10° S and Table 2. Percentages of the Minimum Number of Individuals (MNI) from warm-water and cool-water ecological groups as a percentage of MNI classified to these two groups; organized chronologically within each province. Province and Site % Warm-water % Cool-water MNI Southern Peru-Chilean Province (17°–12° S) Quebrada Jaguay, Sector II (16° 30ʹ S) 0.5 99.5 210 Ring Site (17° 40ʹ S) 3.6 96.4 308 Paloma (Probability samples) (12° 25ʹ S) 1.8 98.2 112 Northern Peru-Chilean Province (10°–8° S) Paiján (10° S) 96.5 3.5 113 Almejas (6.35 mm fraction) (9° 40ʹ S) 53.5 46.5 86 Ostra Base Camp (8° 55ʹ S) 64.2 35.8 120 Alto Salaverry (8° 10ʹ S) 6.8 93.2 59 Panamanian Province (4°30ʹ–2° S) Las Vegas (OGSE80) (2° 14ʹ S) 93.9 6.1 49 S. Siches, Amotape Phase (4° 25ʹ S) 83.3 16.7 12 S. Siches, Siches Phase (4° 25ʹ S) 87.7 12.3 701 S. Siches, Honda Phase (4° 25ʹ S) 82.6 17.4 258 Real Alto (2° 23ʹ S) 98.3 1.7 175 Note: See Table 1 for references and Figure 1 for site locations. MNI in this table does not include mixed-water or terrestrial individuals. For the total MNI in each assemblage see Table 4. Sitio Siches data exclude anchovies and Pozo II. 114 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) Figure 2. Percentages of vertebrate individuals (MNI) in warm-water and cool-water ecological groups in some coastal Peruvian and Ecuadorian zooarchaeological assemblages. Sites arranged chronologically within the southern Peru-Chilean Province, the northern Peru-Chilean Province, and the Panamanian Province reported in Table 1. See Figure 1 for locations. 8° S latitude show a shift from animals associated with warm waters to those preferring cool waters (Fig. 2). Vertebrates from Ecuadorian sites between 4°30ʹ and 2° S latitude suggest an early emphasis on warm-water animals persisted, perhaps inten- sifying over time. Based on the Sitio Siches data reported here, we propose a climate history for the Sitio Siches with two climatic transitions: an early transition between cooler and warmer conditions ca. 10,000–9,000 14C B.P. (~11,400–10,100 cal B.P.) and a later transition to mixed water condi- tions, trending to the cool side, occurred ca. 6000– 5000 14C B.P. (~6800–5700 cal B.P.). The latter transition, at least, also seems to characterize sites between 10° S and 8° S latitude. Several lines of evidence contribute to this study. The first line of evidence lies in the spe- cific characteristics of the resource area, particu- larly modern biogeographical differences between a warm-temperate Peru-Chilean province and a warm-tropical Panamanian province (e.g., Briggs, 1974:42; Fig. 1). These differences are used to classify vertebrates found in archaeological con- texts into terrestrial, warm-water, cool-water, and mixed-water ecological groups. These four ecolog- ical groups are described broadly in the text. Addi- tional biological, ecological, and, when available, cultural aspects of these animals are summarized in Appendix 1. These ecological groups are used to summarize data from nine published, quantified zooarchaeological studies of coastal assemblages representing over 30 years of research in which zooarchaeological materials were compared to ref- erence specimens of known identity (Tables 1–2). These studies constitute a large sample with good taxonomic resolution, allowing our study of Sitio Siches to sit within a regional corpus of similar studies. The review of these legacy zooarchaeo- logical studies is followed by an introduction to Sitio Siches, the field methods used to excavate the site, the site’s occupational sequence, and the zooarchaeological methods used in the Sitio Siches study. Finally, the results for each cultural phase at Sitio Siches are placed within the broader cultural and environmental context of life on the frontier between the warm-temperate Peru-Chilean and warm-tropical Panamanian provinces. REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 115 RESOURCE AREA Today the Peruvian coast is a cool, barren desert as far north as 5° S latitude (Guillén, 1980; Santander, 1980; Handler, 1984). It seldom rains on the coast except during southern migrations of the warm Equatorial Counter Current associated with ENSO. The desert is virtually devoid of plants and animals except where seasonal fog oases (lomas) form during the austral winter and in coastal river val- leys. Most coastal rivers flow only when it rains in the highland headwaters of the valleys, though some rivers are partially fed by glacial meltwa- ter. Freshwater fishes and terrestrial vertebrates are scarce (Eigenmann, 1921; Grimwood, 1969; Eisenberg and Redford, 1999). Shallow warm bays with brackish waters are rare, but not unknown. In contrast, an abundant and characteristic group of marine organisms is supported by nutrients brought to the ocean surface by cold benthic waters ris- ing from the ocean floor in response to prevailing winds and currents, the upwelling associated with the Peru Current (Briggs, 1974; Santander, 1980). Biogeographers divide modern coastal waters into a warm-temperate Peru-Chilean province and a warm-tropical Panamanian province at 3° or 4° S latitude, just south of the Gulf of Guayaquil (e.g., Briggs, 1974:42; Fig. 1). Others place this bound- ary at about 6° S latitude, near Punta Aguja (e.g., Ekman, 1953:38). The boundary is formed where the cold, north-flowing Peru Current meets south- flowing, equatorial waters, deflecting the Peru Cur- rent westward away from the South American con- tinent. Although technically referred to as “warm- temperate,” the waters of the Peru-Chilean province are cold by comparison to the warmer waters of the Panamanian province. The Peru-Chilean province is characterized by localized nutrient upwellings that contrast with the mangrove swamps and exten- sive estuaries of the Panamanian province. As with most biogeographical boundaries, this one is not impermeable and animals typical of one province are found in others, though often in lower numbers or only periodically (Chirichigno, 1982). Transition zones are frontiers between bio- geographic provinces and characterized by gradi- ents in environmental parameters such as tempera- ture, dissolved oxygen, salinity, turbidity, and pro- ductivity (e.g., Scartascini, 2017). DEFINING ECOLOGICAL GROUPS Animal use by people is associated with many environmental and cultural factors operating on different scales (e.g., topography, seasonality, ecosystem productivity, social affiliations, trade, scheduling conflicts, technology), but one of these is local availability. A preference for using local resources facilitates interpreting temporal and spa- tial variability in the archaeological record from the perspective of environmental change at a spe- cific site. Analysis of animal remains from Sitio Siches relies on the modern distributions of warm-tropical (henceforth referred to as warm) and warm-tem- perate (cool) vertebrates derived from the biologi- cal literature (Table 3; Appendix 1; Fowler, 1945; Hildebrand, 1946; Schweigger, 1964; Chirichigno, 1974, 1982; Froese and Pauly, 2004). These verte- brates form one terrestrial and three marine groups that we use to summarize vertebrate data from nine coastal zooarchaeological assemblages and to establish a zooarchaeological context within which to consider these same ecological groups in the Sitio Siches assemblage. The prevalence of habitat and niche prefer- ences among animals represented in archaeologi- cal assemblages permits us to assign vertebrates recovered from archaeological sites to these groups (Table 3; Appendix 1; Kenward and Hall, 1997). Kenward and Hall (1997:665, italics in the original) define an indicator taxon as “...one which reliably carries the implication of the occurrence of some event, activity, or ecological condition in the past” and an indicator group as “...a natural grouping of organisms selected because it includes a range of stenotopic species which together encompass a wide spectrum of ecological conditions or human activities relevant to the aims of the study being carried out.” In the following pages four indicator groups, here referred to as ecological groups, are used: warm-water, cool-water, mixed-water, and terrestrial. Except for terrestrial animals, member- ship in each group is largely defined by preferences 116 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) Vernacular name Taxonomic name Terrestrial taxa Opossum Didelphis sp. Anteater Tamandua tetradactyla Spider monkey Ateles spp. Rabbit Sylvilagus cf. brasiliensis New World rats, mice UID Rodent, Sigmodontinae, Akodon sp., Oryzomys spp., Phyllotis sp., Sigmodon sp. Viscacha Lagidium peruanum Fox and other canids Canidae, Lycalopex cf. sechurae Raccoon Procyonidae Weasel Mustela sp. Puma Puma concolor (formerly Felis concolor) New World camelid Camelidae, Lamini, guanaco Deer Cervidae, Odocoileus virginianus, Mazama sp. Pigeon, dove Columbidae Macaw, parrot Psittacidae Passerine bird Passeriformes, Corvidae, Mimidae Freshwater turtle Emydidae Lizard Iguanidae, Tropidurus sp., Teiidae, Dicrodon sp., Callopistes flavipunctatus, Gekkonidae Snake Colubridae, Boa constrictor (formerly Constrictor constrictor), Drymarchon corais, Viperidae, Bothrops sp. Frog, toad Anura, Bufonidae, Ranidae Warm-water taxa Ladyfish Elops affinis Bonefish Albula vulpes Sea catfish Ariidae, Bagre spp., Galeichthys peruvianus Toadfish Batracoididae, Batrachoides sp. Mullet Mugil spp. Snook Centropomus spp. Grouper, sea perch Epinephelus spp., Hemanthias peruanus Jack Caranx spp., Oligoplites spp., Selene spp. Snapper Lutjanidae, Lutjanus spp. Mojarra Gerreidae, Eucinostomus spp. Grunt Conodon spp., Haemulon spp., Orthopristis spp. Porgy Sparidae, Calamus brachysomus Drum Bairdiella spp., Cynoscion cf. analis, Cynoscion cf. stolzmanni, Larimus cf. acclivis, L. cf. effulgens, L. cf. pacificus, Micropogonias spp., Odontoscion spp., Ophioscion spp., Umbrina cf. xanti Pococho Nicholsina denticulata (formerly Xenoscarus denticulata) Sleeper Eleotridae Puffer Spheroides annulatus Table 3. Classification of animals to ecological groups. REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 117 Table 3. Continued. Vernacular name Taxonomic name Cool-water/Peru Current taxa Marine otter Lontra felina Great grebe Podiceps major Inca tern Larosterna inca Humboldt penguin Spheniscus humboldti Shearwater Procellariidae, Macronectes giganteus Diving petrel Pelecanoididae, Pelecanoides garnotii Booby Sula spp. Black-created buzzard eagle Geranoaetus melanoleucas Eagle ray Myliobatidae, Myliobatis spp. Anchovy Engraulidae, Anchoa spp. Herring Clupeidae Merluza Merluccius spp. Cusk eel Genypterus maculatus Cabrilla Paralabrax spp. Jurel Trachurus murphyi Sargo Anisotremus scapularis Cabinza Isacia conceptionis Drum Corvina deliciosa (formerly Sciaena deliciosa), Cilus gilberti (formerly Sciaena gilberti), Paralonchurus spp. Morwong Cheilodactylidae, Cheilodactylus spp. Wrasse Labridae, Bodianus spp., Pimelometopon spp. Rollizo Pinguipes chilensis (occasionally Mugiloides chilensis) Chalapo Labrisomus philippii Mackerel Scombridae, Euthynnus spp., Sarda spp., Scomber spp., Scomberomorus spp., Thunnus spp. Palmera Seriolella violacea (formerly Neptomenus violacea) Mixed-Water Taxa Sealion Pinnipedia, Otariidae, Otaria flavescens (formerly O. byronia) Whale, dolphin Mysticeti, Delphinidae Duck Anatidae Rail Rallidae Sandpiper Scolopacidae Gull Laridae, Larus spp. Sooty shearwater Puffinus cf. griseus Cormorant Phalacrocorax spp. Pelican Pelecanus spp. Heron Ardea cocoi Sea turtle Cheloniidae Shark, ray Rhiniodon typus, Carcharhinidae, Mustelus sp., Rajiformes, Rhinobatos spp., Dasyatidae Eel Anguilliformes, Muraenidae Sea bass Serranidae 118 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) for conditions prevalent in the cool Peru-Chilean province or the warm Panamanian province, or the ability to inhabit either or both provinces. Biogeographical inferences are based on the premise that relationships exist among the identi- fied organisms, their environmental preferences, environmental conditions at the site, and human behavior. These inferences rely on contemporary observations to infer historical relationships not only among plants, animals, and environments, but also among cultures. This allows archaeolo- gists to evaluate evidence for both environmental and cultural change or stasis. Biogeographic infer- ences assume that behaviors of organisms observed today, such as feeding, reproduction, and habitat preferences, have not changed considerably within the last few decades or millennia. Furthermore, these inferences predict that changes in one of these variables should be associated with changes in others. These premises enable us to establish a standard against which to evaluate vertebrate evi- dence for environmental change and continuity in terms of the relative abundance of warm-water and cool-water vertebrates in zooarchaeological assem- blages when geochemical or other environmental data are unavailable. This approach is not without its problems, one of which is that people are not typical preda- tors and foragers. Explanations of cultural change and continuity in an environmental context must take fundamental aspects of human decision-mak- ing into account. The choices made by people are conditioned by a wide range of cultural perceptions about what to use, when to use it, how to use it, and other criteria beyond simple availability or ener- getics (Reitz, 2002). This is sometimes referred to as “cultural imprinting” (Rollins et al., 1990:471) or the “cultural filter” and is a fundamental aspect of all zooarchaeological interpretations (Reitz and Wing, 2008:5–7; Reitz and Shackley, 2012:5). Interpretations based on biogeography can- not be accepted uncritically as evidence of former environments and cultures because some relation- ships have changed over the centuries, particularly in places subject to human modifications of long duration (e.g., Carrott and Kenward, 2001). The premise that organisms have not changed their habitats and niches over time is unlikely to be true because many phenomena (e.g., competition, predation, community transformations, climate change) alter such associations, with or without human intervention (e.g., O’Connor and Evans, 2005; Benítez-López, 2018; Gaynor et al., 2018). Organisms previously common in a region may now be rare and those once rare or absent from the area may now be abundant. This is particularly the case for small organisms because many are highly sensitive to subtle environmental stresses and opportunities (e.g., Kenward, 1975, 2006; Webb et al., 1998; Plunkett et al., 2009). Thus, the suite Vernacular name Taxonomic name Jack Carangidae, Seriola spp., Trachinotus paitensis Grunt Haemulidae Drum Sciaenidae, Cynoscion spp., Menticirrhus spp., Stellifer spp. Sea chub Kyphosidae, Kyphosus spp. Spadefish Chaetodipterus zonatus Flounder Pleuronectiformes, Bothidae, Paralichthys spp. Table 3. Continued. Note: Taxonomy reflects attributions made during the original zooarchaeological study except when the currently valid spelling or epithet is clear. Ecological groups are based on information in Chirichigno (1974, 1982), Fowler (1945), Froese and Pauly (2004), Hildebrand (1946), and Schweigger (1964). See Appendix 1 and Reitz et al. (2015, 2017) for further discussion of the taxonomy used and the ecological classifications. REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 119 of organisms and/or their relative abundance in a catchment area may change considerably over time. Furthermore, a species may display one suite of characteristics in the middle part of its range and a different suite at the edge of its range (O’Connor and Evans, 2005:24). Environmental changes, such as those occurring during and after the Pleisto- cene/Holocene transition influence the distribution and abundance of plants and animals found in the archaeological record. These factors complicate efforts to interpret biogeographical inferences from animals represented in the archaeological record. Perhaps more problematic for assigning ani- mals to ecological groups is that many of the ani- mals prominent in zooarchaeological assemblages are notable for their flexibility and the wide range of habitats and niches they occupy in response to both anthropogenic and non-anthropogenic stim- uli. Animals tolerant of a wide range of environ- mental conditions commonly are fundamental to human economies and dominate zooarchaeological assemblages (e.g., Reitz, 2001, 2014; Reitz et al., 2015). Many of these are eurytopic, tolerant of a broad range of environmental conditions, though some may be less tolerant of environmental condi- tions and display other differences in growth, size, shape, and behavior at the edges of their preferred range than they do in optimal locations (O’Connor and Evans, 2005:24). Animals dominating zooar- chaeological assemblages are likely to be catholic ones with broad feeding, reproductive, and habitat characteristics making them accessible over a large catchment area and used by human communities that might otherwise share few other attributes. Often these same animals are relatively reliable in terms of accessibility, predictability, availabil- ity, and vulnerability to specific capture methods, making them even more attractive to fishing com- munities. Stenotopic species, those with rigid or limited habits and habitats and otherwise narrow environmental tolerances, often are rare or absent in zooarchaeological assemblages, particularly if these species are available irregularly. For these reasons, few animals consistently found in coastal zooarchaeological assemblages are confined to a single habitat or niche. Many occupy several different habitats or migrate from one habitat type to another during their life cycles. For example, mullets (Mugil spp.) move in and out of estuaries as part of their life cycles and might be found in both cooler offshore waters as well as warmer inshore locations depending on age and reproductive status. Mullets even are present in assemblages from the semiarid northern coast of Chile (Béarez et al., 2015). Other fishes com- mon in warm waters, such as bonefishes (Albula vulpes) and sea catfishes (Ariidae), also are found as far south as Chile, where the coastal waters are cool but bays offer warmer, though more variable, conditions. Bonefishes, sea catfishes, and mullets, however, are far more common in warm, shallow bays; and much more common in zooarchaeologi- cal assemblages associated with those conditions. A more practical limitation to assigning organisms to ecological groups is that many zoo- archaeological attributions are too broad to be assigned to a specific group. Most vertebrate attri- butions for Peruvian archaeological assemblages are to the level of genus, at best, though most organisms display habitat and niche preferences only at lower taxonomic levels. Ecological clas- sifications could be refined if attributions were at lower taxonomic levels. In addition, the limited life history data available for many animals in Peru- vian assemblages precludes strict, inflexible habi- tat assignments, even if that were appropriate. Despite problems associated with biogeo- graphical classifications of animals, ecological groups of vertebrates closely track the location of archaeological sites within the Peru-Chilean or Panamanian provinces (Table 4). This suggests that a cautious interpretation of these associations can offer insights into former environmental conditions along the Peruvian coast. The marine ecological groups used here rely heavily on Chirichigno’s (1982:xvi) catalogue of marine species of economic potential for the cen- tral and southwestern Pacific coast. This catalogue summarizes information about geographical distri- bution, habitats or zones of major concentration, average size taken by commercial fishing methods, methods of capture, and economic importance for 120 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) all marine species for which fisheries literature is available, excluding marine mammals. Chirichigno (1982) divided the central and southern Pacific coast into two fishing areas and several subareas. Area 77 extends from northern Mexico to southern Ecuador (32° 43ʹ N to 5° 00ʹ S) and Area 87 consists of Perú and Chile (5° 01ʹ S to 55° 00ʹ S). The Panamanian province falls within Area 77 and the Peru-Chilean province falls within Area 87 (Fig. 1). Of specific interest are Chirich- igno’s Subareas 77C, 87A, and 87B. Subarea 77C is primarily the Ecuadorian coast extending down to Paita, Perú (1° 39ʹ N to 5° 00ʹ S). Subarea 87A extends from Paita to Punta Aguja at 6° 00ʹ S. Sub- area 87B includes the rest of the Peruvian coast to 18° 20ʹ S. Subarea 87C extends from 18° 20ʹ S to 37° 00ʹ S, which consists largely of the north- ern half of the Chilean coast. Sitio Siches is at the southern end of Subarea 77C (Fig. 1; Table 1). Biogeographical characteristics of animals falling within Subareas 77C, 87A, and 87B are summa- rized in Appendix 1. This approach has the advantage of permit- ting us to compare the stratified Sitio Siches assem- blage to assemblages from nine other coastal sites using all vertebrate attributions in those assem- blages instead of relying on a few specimens from one or two animals from a limited number of con- texts. Many paleoenvironmental studies necessar- ily rely on limited samples due to cost and access to suitable archaeological and reference materials. This is particularly common in applications rely- ing on geochemical studies, which provide essen- tial information about the past but invariably are restricted to a small number of cases. The ecologi- cal groups defined here permit us to use much more of each zooarchaeological assemblage. Multi-proxy studies are important tests of the Table 4. Percentage of Minimum Number of Individuals (MNI) in each ecological group, organized chronologically within each province. Province and site % Terrestrial % Warm- water % Cool- water % Mixed- water MNI Southern Peru-Chilean Province (17°–12° S) Quebrada Jaguay, Sector II 15.8 0.4 83.0 0.8 253 Ring Site 1.4 2.5 68.1 28.0 436 Paloma Probability samples 2.3 1.5 83.3 12.9 132 Grab samples 3.6 5.1 79.8 11.6 277 Northern Peru-Chilean Province (10°–8° S) Paiján 83.8 14.3 0.5 1.3 761 Almejas 6.35 mm fraction 7.0 40.3 35.1 17.5 114 Column fraction 0.8 12.2 84.0 3.1 131 Ostra Base Camp - 53.5 30.0 16.7 144 Alto Salaverry - 6.3 85.9 7.8 64 Panamanian Province (4°30ʹ–2° S) Las Vegas (OGSE80) 57.2 31.7 2.1 9.0 145 Sitio Siches, Amotape Phase 11.8 29.4 5.9 52.9 34 Sitio Siches, Siches Phase 0.3 50.2 7.0 42.4 1225 Sitio Siches, Honda Phase 0.4 41.9 8.9 48.8 508 Real Alto 5.1 72.9 1.3 20.8 236 Note: Faunal data from Byrd (1976), Chase (1988), Pozorski & Pozorski (1979), Reitz (1988a, 1988b, 1995, 2003); Reitz et al. (2017); Reitz & Sandweiss (2001); Sandweiss et al. (1989), and Wing (1986). See Figure 1 for site locations and Table 1 for dates. REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 121 hypothesis that present environmental and ecologi- cal attributes of these animals reflect earlier ones. Nonetheless, these classifications and the interpre- tations derived from them should be tested using other lines of evidence, particularly geochemical data, whenever possible. We have pursued other research avenues to directly test the environmental conditions prevailing at Sitio Siches and will con- tinue to do so.3 SUMMARY OF OTHER ZOOARCHAEOLOGICAL ASSEMBLAGES Our comparative study of other coastal zooar- chaeological assemblages using ecological groups demonstrates that the early- and mid-Holocene coastal environment of Perú was dynamic and that a regional economic pattern based on marine resources flourished from the Late Pleistocene into the mid-Holocene. Zooarchaeological assemblages from sites located in the Peru-Chilean and Panama- nian provinces show a strong association between cool- or warm-water vertebrates and either the location of the archaeological site or when it was occupied (Figs. 1–3; Tables 2, 4; Sandweiss et al., 1996, 2007). Cool-water vertebrates dominate some assemblages from the southern Peru-Chilean province. We contrast this affiliation by compari- son with assemblages from two Panamanian prov- ince sites in which warm-water vertebrates are much more abundant than cool-water ones. Pervasive biases associated with meta-data analyses are well known (e.g., Reitz and Wing, 2008; Scartascini, 2017; Nims and Butler, 2018). Cross-site summaries such as this rely upon access to data quantified and presented in a similar fashion. Some of the inherent biases in such studies are con- trolled here by primarily using published data for assemblages studied using the comparative skel- etal collection and methods followed in the Envi- ronmental Archaeology Laboratory of the Florida Museum of Natural History under the supervision of Elizabeth S. Wing, curator of zooarchaeology, or of Elizabeth J. Reitz, who studied under Wing. This summary focuses on aspects of these other zooarchaeological assemblages that are considered most relevant to establishing a regional context for Figure 3. Percentages of vertebrate individuals (MNI) in ecological groups in some coastal Peruvian and Ecuadorian zooarchaeological assemblages. Sites arranged chronologically within the southern Peru- Chilean Province, the northern Peru-Chilean Province, and the Panamanian Province reported in Table 1. See Figure 1 for locations. 122 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) Sitio Siches. Readers should consult site-specific publications referenced below for additional infor- mation about field methods, analytical procedures, and results for each of these other assemblages. Distances from the coast reported for each of these sites are estimates informed by modern condi- tions and geological evidence. This dynamic coast- line has experienced and continues to experience tectonic and eustatic changes in sea level, shore- line erosion, and seismic events. Such phenomena impact the meaning of terms such as “coastal” and “inland” (e.g., Erlandson, 2016). The economic orientation of all of the sites summarized here was coastal regardless of the precise distance of each site from the ocean, with the possible exception of sites in the Paiján complex, as described below. A common attribute of the zooarchaeo- logical assemblages summarized here is that their study included an estimate of the Minimum Number of Individuals (MNI, in Spanish NMI). This is not the only way to quantify zooarchaeo- logical assemblages and is associated with a num- ber of well-known biases (e.g., Reitz and Wing, 2008:205–210). MNI, however, is one of the few forms of quantified data included in most, though not all, species lists available to us at this time. We assign vertebrate MNI estimates made by the origi- nal analysts of each assemblage to one of the four ecological groups developed using present-day biogeography (Table 3; Appendix 1; Fowler, 1945; Hildebrand, 1946; Schweigger, 1964; Chirichigno, 1974, 1982; Froese and Pauly, 1998, 2004; Reitz, 2001; Reitz et al., 2015). MNI published in the spe- cies list for each assemblage is used, regardless of the taxonomic level for which that estimate was made by the original analyst. This means that in some cases MNI estimates are for taxonomic levels as high as class (e.g., Mammal). More frequently MNI estimates are made for family- or genus-level attributions (e.g., Ariidae or Mugil spp.). Rarely are MNI estimates available at the level of species (e.g., Albula vulpes). This is the same approach used during the analysis of Sitio Siches. The taxonomic attributions made during the original study of each vertebrate assemblage are used in this summary whenever there is doubt about the current taxonomic affiliation of the ani- mal. Many of these data are from legacy studies conducted decades ago. The attributions reported by the original analyst reflect the taxonomic attri- butions of reference skeletons considered valid at the time of the original study. In some cases, subsequent changes in scientific names were rela- tively simple to implement (e.g., Sciaena deliciosa to Corvina deliciosa) and the current valid name is used here. In other instances, the name change leaves the original attribution ambiguous, as is the case for foxes (see Appendix 1 for discussion). In such cases, the name reported in the original study is used because to do otherwise would require re- analysis to verify the attribution. The term “fishes” refers to both cartilaginous (Chondrichthyes and Rajiformes) and bony fishes (Actinopterygii) unless otherwise stated. Other biases are not as readily controlled, particularly those associated with taphonomy, cultural choices manifested in these assemblages, recovery methods, and limited information about plant assemblages at each site. We include this information when we know it but the reader is referred to site-specific references for details of the methods used to excavate these sites and to ana- lyze each faunal assemblage. Sites for which the published data cannot be converted readily into the systematic, quantified format used in this study are not summarized here (e.g., Wing and Reitz, 1982a; Dillehay et al., 2012, 2017). The comparative summary begins with the southernmost site (Fig. 1; Tables 1, 4). The Ring Site is a stratified shell midden in southern Perú dating to the Early and Middle Preceramic Period (ca. 9800–5800 cal B.P.; Sandweiss et al., 1989; Reitz et al., 2015; Reitz et al., 2016; Reitz et al., 2017). It lies in the Osmore drainage about 7.5 km southeast of Ilo and 10 km south of the Ilo River, a stream with intermittent flow today. The site was less than 5 km from the shoreline when it was first occupied and is less than a kilometer from the present shoreline. The Ring Site is at the southern boundary between Subareas 87B and 87C. Daniel H. Sandweiss and James B. Richardson III con- ducted excavations at the site over several years, REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 123 but vertebrate data summarized here are from Unit C-1, which was excavated in 1985 using a 6.35- mm mesh screen. Fishes dominate the Ring Site vertebrate assemblage (Reitz et al., 2017). Terrestrial mam- mals (exclusively mice) constitute 0.9% of the Minimum Number of Individuals (MNI, or indi- viduals) and marine mammals contribute 4% of the individuals. Mammals include New World mice (Sigmodontinae), marine otters (Lontra felina), and seals or sealions (Pinnipedia). Birds are common in the assemblage (123 individuals; 28% of the MNI). All but one of these is a marine or coastal species. Marine fishes are the most common group of ver- tebrates (67% of the MNI). The most common of these are drums: corvina (Cilus gilberti) and lorna (Corvina deliciosa). Warm-water vertebrates con- stitute 2% of the MNI in the assemblage (Fig. 3; Table 4). The only warm-water species are sea cat- fishes (Ariidae), groupers (cf. Epinephelus spp.), jacks (Caranx spp.), and silver perches (Bairdiella spp.). In contrast, vertebrates typical of cool waters or the Peru Current constitute 68% of the MNI. The remaining individuals (30% of the MNI) are mixed-water or terrestrial forms and include ani- mals such as seals or sealions, pelicans, and cormo- rants. These may be found in either ecological area, though they are characteristic of the Peru Current. Quebrada Jaguay is a Terminal Pleistocene and Early Holocene site located on an alluvial terrace overlooking a seasonal stream in Subarea 87B (McInnis, 1999; Sandweiss et al., 1998; Sand- weiss, 2014; Jones et al., 2017; Reitz et al., 2017). The site is approximately 2 km from the modern shoreline; though at the end of the Pleistocene it probably was ca. 6–7 km from the coast. Heather McInnis and Sandweiss conducted excavations here in 1996 and 1999. The materials summarized here were excavated in 1999 from Sector II using a 1.7-mm screen and date around 12,700–11,900 cal B.P. based on bulk samples (Sandweiss, 2009; Jones et al., 2017; AMS dates on short-lived sam- ples suggest a start date of 12,000 cal B.P. [Jones et al., 2017 and recent work by Kurt Rademaker]). Obsidian recovered from the site indicates resi- dents of Jaguay had contact with the Andean high- lands some 130–150 km to the west (Sandweiss et al., 1998; Rademaker et al., 2013; Sandweiss and Rademaker, 2013). It is not known if people at Jag- uay obtained this obsidian through down-the-line trade or direct contact (Sandweiss and Rademaker, 2013). Bottle gourds (Lagenaria siceraria) and fragments of small, knotted cordage suitable for use in netting are present in Early Holocene con- texts (Sandweiss et al., 1998; Erickson et al., 2005; Sandweiss, 2009; Piperno, 2011). Some Terminal Pleistocene deposits at Quebrada Jaguay appear to be midden with no direct evidence of a house or hearth; but other Terminal Pleistocene deposits and some Early Holocene deposits are associated with domestic structures and activities (Sandweiss, 2009, 2014). Fishes dominate the Quebrada Jaguay Sec- tor II assemblage (Reitz et al., 2017). Indetermi- nate rodents and an indeterminate bird constitute 16% of the MNI. No marine mammals are pres- ent. Fishes contribute 84% of the MNI. The most common of these are the two drums: corvina (Cilus gilberti) and lorna (Corvina deliciosa). These two cool-water drums contribute 82% of the MNI. The only warm-water species are sea catfishes (Ariidae, 0.4% of the individuals; Fig. 3; Table 4). Mixed- water fishes also are rare. Paloma is an Archaic (Middle Preceramic; 8500–5500 cal B.P.) village site in the Chilca River Valley (Benfer, 1984; Reitz, 1988a, 1988b, 2003). The village lies about 65 km south of Lima, 3.5 km from the present-day shoreline, and 7.5 km north of the Chilca River, an intermittent stream (Ben- fer, 1984). Probability Samples from the site were recovered by Robert Benfer using a 1.5-mm geo- logical screen (Reitz, 1988a, 2003). Grab Samples were selected randomly by field personnel without the use of a screen (Reitz, 1988b, 2003). Only large specimens with an interesting shape were collected following the Grab Sample protocol; remains of small fishes, such as anchovies, would not be rep- resented in the Grab Samples. Although the Grab Samples are unreliable as indicators of environ- mental conditions or the economy at Paloma, they are summarized here because they contain verte- brates not present in the Probability Samples. 124 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) Mammals are rare in both Grab and Prob- ability Samples (Reitz, 1988a, 2003). In the Prob- ability Samples, mammals and birds contribute 2% of the vertebrate MNI. The remaining vertebrate individuals are fishes. The Probability Samples document heavy use of cool-water fauna typical of the Peru Current, 83% of the MNI (Fig. 3; Table 4). Only 1% are individuals from warm waters. Mixed-water or terrestrial animals constitute 15% of the MNI. In the Grab Samples, terrestrial mam- mals contribute 2% of the vertebrate MNI (Reitz, 1988b). These include spider monkey (Ateles spp.), mice (Sigmodontinae), fox (Dusicyon cf. sechu- rae), puma (Puma concolor), guanaco (Cameli- dae), and deer (Cervidae). Marine mammals con- tribute 4% of the MNI in the Grab Samples; birds 8%; and fishes 85%. Almejas is a Middle Preceramic (ca. 7900 cal B.P.) shell midden in the Casma River Valley (Pozorski and Pozorski, 1984, 2003; Reitz, 1995). The Casma River today is an intermittent stream. The site is 5.5 km from present shoreline and less than 5 m above sea level. Field work was conducted by Thomas and Shelia Pozorski in 1980. During excavation, faunal materials were recovered as one of two types of samples. Excavated soil was screened primarily through 6.35-mm mesh screen in the field. In addition, the residue from a 25-cm square in each level was screened first through a 6.35-mm mesh and the specimens captured in the 6.35-mm screen were combined with the remains recovered from the rest of the 6.35-mm fraction from the associated level. Then, the material from the 25-cm square that had passed through the 6.35- mm screen was re-screened through 2 mm (#10) and 710 μm (#25) geological soil screens. These fine-screened samples are referred to as column samples by the excavators. Column samples are fine-screened samples from the 25-cm columns from which the 6.35-mm fraction was removed and represent only that portion of the faunal assemblage that passed through the 6.35-mm mesh screen. By and large, evidence for use of larger animals, those with larger skeletal remains such as would be captured in the 6.35-mm mesh, is absent in the column samples. Although samples collected with fine-meshed screens normally would be considered more reliable as indicators of former environments and economies, in the case of Almejas, the 6.35- mm portion is more complete and representative than are the column samples. Terrestrial mammals constitute 7% of the individuals and marine mammals contribute 0.9% of the individuals in the 6.35-mm Almejas frac- tion (Reitz, 1995). Terrestrial vertebrates consist of seven mice (Sigmodontinae) and a deer (Cervidae). The most common group of vertebrates are marine fishes, which contribute 90% of the individuals. The most common of these are sea catfishes (Ariidae) and mullets (Mugil spp.). Both are typical of warm waters, such as lagoons, though their range can extend further south (Chirichigno, 1982:41, 242). Warm-water individuals constitute 40% of the ver- tebrates and cool-water vertebrates constitute 35% of the individuals in the 6.35-mm fraction (Fig. 3; Table 4). In the column samples, 12% of the verte- brates are members of the warm-water ecological group and 84% are typical of cool-water regimes or the Peru Current. This is due to the dominance in the column samples of anchovies (Engrauli- dae), herrings (Clupeidae), and very small mullets. These fishes are typical of shallow inshore waters and often are captured using seine nets or scoops in near-shore grassy locations (see Appendix 1). This may be evidence that such conditions existed near Almejas and that people at Almejas frequently used such methods to capture these fishes. Ostra Base Camp is located 6.5 km north of the mouth of the Santa River Valley and was exca- vated by Sandweiss (Sandweiss et al., 1983; Sand- weiss et al., 1996; Reitz, 2001; Reitz and Sand- weiss, 2001). This Middle Preceramic site dates between 7100 and 6200 cal B.P. and is interpreted as one of a series of collecting stations associated with what was, according to geoarchaeological work at the site, probably a relatively shallow bay. The Santa River is the largest of the permanent streams on the Peruvian coast and extends into the upland as part of the Callejón de Huaylas drainage system. Both 6.35-mm and 1.7-mm mesh screens were used during excavation to recover all exca- vated material (Reitz and Sandweiss, 2001). The Ostra assemblage contains no terrestrial animals (Reitz and Sandweiss, 2001). Single indi- REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 125 viduals identified as seal or sealion (Pinnipedia) and dolphin (Delphinidae) are the only mammals present. Birds contribute 7% of the individuals; cormorants (Phalacrocorax spp.) are particularly common. A sea turtle (Cheloniidae) also is present in the assemblage. Marine fishes constitute 91% of the vertebrate individuals. Members of the warm- water ecological group constitute 53% of the ver- tebrates (Fig. 3; Table 4). Among the warm-water species are bonefishes (Albula vulpes), sea cat- fishes (Ariidae), mullets (Mugil spp.), and puffers (Spheroides annulatus). These four warm-water fishes alone contribute 40% of the vertebrate indi- viduals in the Ostra Base Camp assemblage. Cool- water individuals contribute 30% of the vertebrates. Alto Salaverry is a shell midden in the Moche Valley excavated by Shelia and Thomas Pozorski (Pozorski, 1976; Pozorski and Pozorski, 1979). This Late Preceramic site is located ca. 1.5 km inland from a sand beach and 6 km south of the Moche River, an intermittent stream. No radiocarbon dates are available for the Alto Salaverry materials summarized here, but the Late Preceramic attribu- tion places it at between 5000 and 4000 cal B.P. A 6.35-mm mesh screen was used by Shelia Pozorski to recover these materials. Pozorski did the iden- tifications with the assistance of faculty from the University of Trujillo, the Lima Natural History Museum, Melody Shimada, and Wing (Pozorski, 1976:42, 58–59). The only mammals in the assem- blage are two sealion (Otaria flavescens [formerly O. byronia]) individuals. No birds are represented and 97% of the vertebrate individuals are marine fishes. This is a remarkably high percentage con- sidering that fishes are usually better represented when finer gauge screens are used. Members of the warm-water ecological group constitute 6% of the individuals and members of the cool-water group contribute 86% of the individuals (Fig. 3; Table 4). The warm-water animals are mullets and pocho- cos (Nicholsina denticulatus [formerly Xenoscarus denticulata]). The cool-water animals are primarily corvinas (Corvina deliciosa; 69% of the MNI). The term “Paiján complex” refers to numer- ous small sites associated with Early Preceramic people living in the Chicama Valley on the north coast of Perú during the Terminal Pleistocene and Early Holocene between about 12,200 and 9200 cal B.P. (Lumbreras, 1974:28, 30). The sites are between 15 and 36 km from the present-day coast- line, though, due to glacially lowered sea levels, the shoreline was as much as 15 km further west when these sites were occupied. Vertebrate remains were identified by Arlene Fradkin, Sylvia Scud- der, Erika Simons, and Gary Shapiro under the supervison of Wing (1986). They studied materials from 11 Paiján-complex sites, including Pampa de los Fósiles, Quebrada de Cupisnique, and Ascope excavated under the direction of Claude Chauchat (1976, 1978, 1988, 1992). A fine-gauge screen was used to recover the materials, though the dimen- sions of the screen are unknown. The sample size from these sites is highly variable and the data from all of them are merged in this summary. Of the 761 vertebrate individuals identified from this cluster of sites, 7% are terres- trial mammals, 74% are lizards and snakes, 3% are birds, and 16% are fishes (Wing, 1986). The terres- trial mammals are primarily rodents (Sigmodon- tinae, 5% of the MNI) but also include fox (Dusi- cyon cf. sechurae) and viscacha (Lagidium peru- anum). Less than 1% of the individuals are deer (Cervidae). The lizards are primarily cañan lizards (Dicrodon spp.; 70% of the MNI). Members of the warm-water ecological group constitute 14% of the individuals and cool-water individuals constitute 0.5% (Fig. 3; Table 4). Sea catfishes (Ariidae) and mullets (Mugil spp.) are the most abundant fishes, though bonefishes (Albula vulpes) and tallfin croak- ers (Micropogonias spp.) are common in some Pai- ján contexts. Small numbers of mojarras (Gerridae, Eucinostomus spp.) and porgies (Sparidae) also are present in the assemblage. The members of the cool-water group are anchovies (Engraulidae), her- rings (Clupeidae), and cocos (Paralonchurus spp.). Given the distance of Paiján-complex sites from the coast, the high percentage of marine ver- tebrates is noteworthy, demonstrating that fish were sufficiently important during the Paiján period that people at sites some distance from the coast made use of this important source of animal protein. The recent work by Dillehay and colleagues (2012, 2017) at Huaca Prieta shows that use of marine resources preceded the Paiján period in the Chi- 126 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) cama Valley. Today Huaca Prieta is located on the present-day shoreline in the Chicama Valley, but it was many kilometers inland during Paiján times (Dillehay et al., 2012, 2017). Vertebrate assemblages from two clearly tropical sites highlight differences that might exist between sites associated with cool or warm waters. The Middle Valdivia period faunal remains from Real Alto were excavated by Donald Lathrap and identified by Kathleen M. Byrd at the Florida Museum of Natural History (Byrd, 1976:113–117, 1996). Dolores Piperno and Deborah Pearsall (1998:247) refer to “Middle Valdivia” as Valdivia period III–V and date the occupation to ca. 4400– 3800 B.P., following Frederick Schwarz and J. Scott Raymond (1996). The site lies 5 km inland from the present-day coastline in Area 77C of the Panamanian province. Screen size used to recover these materials is not reported by Byrd. Fishes in the Middle Valdivia samples constitute 92% of the vertebrate individuals (Byrd, 1996:309). Ter- restrial mammals constitute 5% of the individu- als; birds 1% of the MNI; reptiles and amphib- ians 2%; and fishes 92%. The terrestrial animals at Real Alto are primarily deer (Cervidae, Mazama sp., Odocoileus sp.; 4% of the MNI). The assem- blage is dominated by sea catfishes (Ariidae; 49% of the MNI) and drums (Sciaenidae; 13% of the MNI). Although mullets (Mugil spp.) are present in the Real Alto assemblage, bonefishes (Albula vulpes) and anchovies (Engraulidae) are absent. Members of the warm-water ecological group con- tribute 73% of the vertebrate individuals and mem- bers of the cool-water group 1% (Fig. 3; Table 4). The mixed-water ecological group contains ducks (Anatidae) and sea turtles (Cheloniidae), in addi- tion to sharks and rays, grunts (Haemulidae), and seatrouts (Cynoscion spp.). Las Vegas (OGSE80) also is located on the Gulf of Guayaquil, about 3.5 km from the pres- ent coastline (Stothert, 1988; Piperno and Pearsall, 1998:244–245). Vertebrate remains were exca- vated by Karen E. Stothert and identified by Byrd (1976:99–100, 1996) and Thomas Chase (1988) under the supervision of Wing at the Florida Museum of Natural History. The screen size used to recover vertebrate remains is not reported by Byrd or Chase. This is one of the few sites for which botanical information is available. Deposits from ca. 9060 B.P. contain phytoliths from bottle gourds (Lagenaria siceraria) and lerén (Calathea allouia), both of which were introduced from elsewhere (Piperno and Pearsall, 1998:190–198, 244–245; Piperno et al., 2000; Piperno and Stothert, 2003). Beginning in the Terminal Pleistocene, deposits from the site also contain squash/gourd (Cucurbita spp.) phytoliths with sizes suggesting a temporal trajectory of intensifying human selective pressure (Piperno et al., 2000). By ca. 7170 B.P., domestic squash/gourds were present at the site. Las Vegas faunal samples studied by Byrd and by Chase differ in terms of the percentages of vertebrate individuals from terrestrial habitats, but warm-water individuals are more abundant than cool-water and mixed-water individuals in both studies. Terrestrial mammals constitute 46% of the individuals in the combined assemblage; birds 1%; reptiles and amphibians 12%; and fishes 41%. The terrestrial animals are primarily rab- bits (Sylvilagus cf. brasiliensis, 4% of the MNI), mice (26%), foxes (Dusicyon cf. sechurae, 10%), snakes and cañan lizards (6%), and frogs (5%). Brocket deer (Mazama sp.) constitute less than 2% of the individuals. Two fish families dominate the marine component: sea catfishes (Ariidae, 10% of the MNI) and two warm-water drums (Sciaenidae; 11% of the MNI). Although mullets (Mugil spp.) are present in the Las Vegas assemblage, bone- fishes (Albula vulpes) and anchovies (Engraulidae) are absent. Warm-water vertebrates contribute 32% of the MNI and members of the cool-water ecolog- ical group contribute 2% of the individuals (Fig. 3; Table 4). The cool-water fishes are three mackerels (Scombridae), which might be considered tropical forms in other locations and are very common in Ecuadorian zooarchaeological assemblages north of the Gulf of Guayaquil (Lippi, 1983; Sánchez, 1993; Reitz and Masucci, 2004:3.3–3.4). These three mackerels are quantified here in the cool- water group because cool-water members of this family tend to be more frequent in coastal assem- blages from southern Perú (Reitz et al., 2015). This summary shows that a distinct and characteristic group of animals is found in Peru- REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 127 vian assemblages and correlates with either time or location despite problems with ecological analo- gies and the ecological groups used here. To clar- ify the distinction between warm-water and cool- water vertebrates, Figure 2 and Table 2 summarize the relative abundance of individuals in the warm- water or cool-water ecological groups; excluding mixed-water and terrestrial individuals. South of 12° S latitude, members of the cool-water/Peru Current ecological group comprise at least 96% of the warm- or cool-water individuals. Between 10° S and 8° S latitude, however, warm-water indi- viduals are more abundant than cool-water ones in the Paiján, Almejas (6.35-mm fraction), and Ostra Base Camp assemblages even though these sites are in what is now a cool-water setting. These three assemblages are from deposits older than ca. 5000 14C B.P. (~5700 cal B.P.). The percentage of cool- water animals in the Alto Salaverry assemblage is more typical of present-day conditions, and this site dates more recently than 5700 cal B.P. In the Panamanian province, marine animals recovered from Las Vegas and Real Alto are almost exclu- sively members of the warm-water ecological group, though these sites date to opposite sides of the 5000 14C B.P. climatic transition.. The Paiján complex is noteworthy because it is the earliest assemblage located between 10° and 8° S latitude reported here (Fig. 1; Table 1). Warm- water animals constitute 96% of the marine individ- uals and cool-water individuals are rare, a relation- ship made clear when terrestrial and mixed-water animals are excluded from the calculation (Fig. 2; Table 2). In this respect, the Paiján assemblage is more like Panamanian-province assemblages than it is to others from the Peru-Chilean province. The similarity to Las Vegas is particularly interesting as some sites in the Paiján complex probably were contemporaneous with early Las Vegas. The suite of warm-water fishes in the Paiján assemblage (bonefishes [Albula vulpes], sea catfishes [Ariidae], and mullets [Mugil spp.]), however, is more like that in the Almejas and Ostra Base Camp assem- blages than like the suite of fishes from Las Vegas, which is dominated by sea catfishes accompanied by warm-water drums. The location of some Paiján sites further inland may explain the faunal patterns at sites in this complex. The cost of transporting fish from the coast to these sites might mean that this was done seasonally or using fishing techniques designed to maximize capture of specific fishes, producing a different assemblage of taxa. It also is possible that people reduced the costs of transportation and spoilage by processing fish on the coast instead of transporting inedible portions inland; leaving very little evidence of this activity at inland locations. More importantly to this discussion, conditions clearly existed in the waters near Paiján for fishes from the warm-water ecological group to be taken in large numbers, sufficient to make a substantial contribution to the economy of people living 15–50 km inland. The Alto Salaverry assemblage suggests a very different strategy prevailed in the northern Peru-Chilean province approximately 4,000 years after the Paiján sites were abandoned. Even though Alto Salaverry is less than 100 km south of Paiján, warm-water vertebrates are minor members of the Alto Salaverry assemblage. The difference in fishes reported from Paiján and from Alto Salaverry sug- gests both environmental and cultural changes by ca. 5000 B.P. These changes may have begun dur- ing the period represented by the Almejas and Ostra assemblages. It is this possibility which is explored further with stratified data from Sitio Siches. SITIO SICHES Sitio Siches was selected for archaeological study specifically because Richardson (1973) reported a transition from warmer to cooler water molluscs at sites located south of Sitio Siches and at Sitio Siches (see also Rollins et al., 1986, 1987). This phenomenon also was observed in molluscs from Almejas (Pozorski and Pozorski, 1984, 2003) and Ostra Base Camp (Sandweiss et al., 1983, 1996; Reitz and Sandweiss, 2001). The objective of the Sitio Siches excavation was to test specifically whether the ecological transition observed in the Sitio Siches molluscs was evident in vertebrates from that site. Site Description Sitio Siches (PV 7-19) is one of the few strat- 128 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) ified preceramic sites on the Peruvian coast bridg- ing the early- and mid-Holocene period for which quantified zooarchaeological data are available (Richardson et al., 1996). The site occupies ca. 20 ha and lies at the present-day southern edge of the Panamanian province about 450 km south of Las Vegas and 600 km north of Paiján (Fig. 1). Radio- carbon dates indicate the site was occupied for much of the early- and mid-Holocene. This occu- pation is subdivided into three phases: Amotape (9500–9000 14C B.P./~10,700–10,100 cal B.P.), Siches (7100–6000 14C B.P./~7900–6800 cal B.P.), and Honda (5150–4550 14C B.P./~5800–5200 cal B.P.). The Amotape phase is roughly contempo- raneous with the Paiján complex to the south and the early Las Vegas site to the north. The Honda phase is contemporary with the Valdivia period in southern Ecuador, represented in this study by Real Alto, and with the Late Preceramic period on the Peruvian coast, represented here by Alto Salaverry. Sitio Siches is located at the head of a que- brada (or ravine) of the same name overlooking what is today a permanent water seep (Richardson, 1973; Richardson et al., 1996). The seep is ca. 5 km inland from the coast following a direct line, though the distance doubles if one walks down the quebrada to reach the beach. Water in the seep is ground water emerging where the eroding head of the quebrada intersects the water table, well above present-day sea level. At the present time the seep contains water during the austral winter, which is the dry season. The seep has not been visited dur- ing today’s austral summer, but it likely has more flow during that season. If the area experienced greater precipitation during the mid-Holocene, the seep was a better source of water in the past than it is today. The nearest major permanent freshwater source is the Chira River, ca. 50 km away. Sea level rose throughout the occupational sequence of Sitio Siches. During the Amotape phase, the shoreline was further from the site than it is today, though the continental shelf is so narrow in this region that the Early Holocene shoreline was not far from the modern beach. As sea level rose during the Siches phase, the distance between the site and the shoreline diminished somewhat. Sea level had reached its modern position by the Honda phase. Field Methods Field work at the site was conducted by Rich- ardson, Sandweiss, and Rolando Paredes E. in 1995 (Richardson et al., 1996) and in 2001 by Richard- son and Sandweiss. Different portions of the site were excavated during the two field seasons (Fig. 4). The 1995 materials were excavated from two areas of the site designated Area I and Area II. Five 1-x-1 m pits were excavated in 1995, two in Area I (IA and IB), and three in Area II (IIA, IIB, IIC). Honda-phase materials are from Areas IA and IB excavated in 1995. Some Siches-phase materials also are from Areas IIA, IIB, and IIC excavated in 1995, referred to as Pozo II in 1995. In 2001, exca- vations focused on Area II, with 19 adjacent 1-x-1 m units excavated. The term “feature” (elemento) refers to any clearly distinguished stratigraphic unit that interrupted the general horizontal stra- tigraphy and was limited in extent. Thus, pits and hearths are called features as are very small units embedded in more extensive levels. All Amotape materials are from pit features excavated during 2001 in Area II. Most of the Siches-phase materials are from Area II units excavated in 2001. Natural stratigraphy defined levels within each unit. Gaps in the stratigraphic record suggest Sitio Siches was abandoned for up to 2,200 years between the Amotape and Siches occupations and again for a shorter period between the Siches and Honda occupations. It is likely that some unknown quantity of the material that accumulated between 9000 and 7100 14C B.P. (~10,100 and 7900 cal B.P.) was lost due to an erosional event. Amotape remains were found only in midden-filled pit fea- tures intruding into sterile deposits; the erosional event(s) removed overlying Amotape sheet mid- den that might have provided evidence for a more continuous occupation (Sandweiss, 2014:1061). Thus, the apparent Amotape-Siches occupational gap may be due in part to erosion of intervening deposits, probably as the result of flooding during El Niño events. Deposits began to re-accumulate at about 7100 14C B.P. (~7900 cal B.P.) during the Siches phase. An additional hiatus of about a mil- lennium in the stratigraphic record is interpreted REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 129 as an occupational hiatus between the Siches and Honda occupations. Other smaller gaps also may exist in the occupational sequence. Occupational gaps likely reflect social or environmental events that made the site unattract- ive to a community intent on capturing a specific suite of marine fishes. Although it is likely that the location was unoccupied for portions of the sequence, we do not interpret this as evidence that the site was but one stage in a migratory or seasonal round. The geological, architectural, and biologi- cal characteristics of the site suggest that breaks in Figure 4. Excavations at Sitio Siches. 130 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) the record were prolonged rather than short-term. When the site was occupied, the occupation was temporally continuous rather than intermittent. Two sample types were collected in the field: Muestra and General. Muestra samples are from a standard volume (1 bucket, approximately 12 liters) of the stratigraphically most secure material in each context. The 6.35-mm mesh Muestra frac- tion was sorted in the field and the 1.7-mm fraction was sorted in the laboratory. In 1995, no Muestra samples were taken from Pit IA, thus only the 6.35- mm Muestra fractions from Pit IB and Pit IIA-B-C are included in the vertebrate study reported here. General samples comprise the rest of the material from each pit and level and feature. The 6.35-mm General samples were collected from the screen. The 1.7-mm General fraction was sorted in the field for artifacts and interesting organic materials such as otoliths. The remainder of the 1.7-mm General fraction was discarded in the field without further study. The consequence of this recovery method is that anchovy remains probably are under-repre- sented in the Honda- and Siches-phase vertebrate samples excavated in 1995, which influenced the organization of the subsequent zooarchaeological study (as discussed below). Structures, Plant Remains, Geochemistry, and Molluscs Hearths and postholes suggest structures were present in all levels of the site, though they are most common in the early Siches-phase levels. The presence of structures supports our interpreta- tion that Siches was more than an occasional fish- ing camp visited during a seasonal round. The 1995 excavations recovered broken beads in the process of being manufactured from fossil scallop (Argo- pecten spp.) and ark (Anadara tuberculosa) shells. These were associated with microlith drills in the Honda deposits. Neither beads nor drills were pres- ent in the excavated Amotape or Siches deposits, though the Honda deposits containing beads and drills were in another portion of the site. The presence of cultivated plants elaborates upon the broad interaction sphere within which the residents of Sitio Siches participated, a sphere that extended into Ecuador, the hypothesized source of cultigens recovered from Sitio Siches. Dolores Piperno identified squash (Cucurbita spp.) phyto- liths in Amotape deposits and Asunción Cano iden- tified bottle gourd (Lagenaria siceraria) pericarp fragments in Siches and Honda deposits (Cano, 1995; Piperno et al., 2000; Piperno, 2005, 2011). In addition, Cano identified seeds of a grass (cadillo, Cenchrus spp.), algarrobo (Prosopis sp.), and mesquite or huarango (P. pallida [formerly Aca- cia pallida]). Although both algarrobo and mes- quite are used as firewood, human consumption of both legumes has a long history in South America (Giovannetti et al., 2008). Oxygen isotope profiles of archaeological sea catfish (Galeichthys peruvianus) from Ostra Base Camp and Sitio Siches indicate changes in sea surface temperatures (SST) when these profiles are compared to those from modern catfish otoliths (Andrus et al., 2002, 2003). At Ostra (8° 55ʹ S lati- tude), the oxygen isotope profile in the archaeo- logical sea catfish otoliths indicates that summer temperatures were about 3° C warmer than today but winters were like those prevailing today, sug- gesting a more strongly seasonal climate that may have included seasonal rainfall. The mid-Holocene annual temperature range at Siches (4° 25ʹ S lati- tude) reconstructed from Siches-phase otoliths was like that experienced today but offset about 3°–4° C warmer. These data also suggest that nearly trop- ical water conditions prevailed at Sitio Siches by the end of the Siches 3b sub-phase.4 Molluscs from Sitio Siches document changes in marine invertebrates over time at the site (Richardson, 1973, 1978; Richardson et al., 1996). Late Pleistocene deposits within ca. 30 km of Sitio Siches contain two mollusc species associated with red mangroves (Rhizophora mangle): Anadara tuberculosa and Ostrea colombiensis. Amotape deposits at Sitio Siches also contain these molluscs (Richardson, 1973, 1978; Richardson et al., 1996). Red mangroves develop in protected bays, estuar- ies, and areas where continental alluvium trans- ported by run-off accumulates. Both mangroves and estuaries require inputs of fresh water as rain- fall and run-off from streams, at least seasonally. This suggests warmer, wetter conditions of the REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 131 Late Pleistocene extended into the Amotape phase (Sandweiss, 2003). Other warm-water molluscs in Siches-phase deposits include almejas (Proto- thaca ecuatoriana) and navajas (Tagelus dombeii). Although very early Siches-phase deposits (analyt- ical unit 3b) contain these warm-water molluscs, deposits from the subsequent Honda phase do not (Sandweiss, 2003). Honda deposits contain a mix- ture of warm-water and cool-water molluscs. Man- groves are not present in the area today and local streams are intermittent. Zooarchaeological Methods Vertebrate remains from both the 1995 and the 2001 excavations are reported here.5 The Sitio Siches vertebrate assemblage was recovered in 1995 from five units and in 2001 from 19 units using 6.35-mm and 1.7-mm mesh screens. Frac- tions from the two mesh sizes were kept separate during analysis. The General samples were distin- guished from the Muestra samples during identifi- cation but are combined here. All 6.35-mm Muestra samples and the 6.35 + 1.7-mm General samples excavated in 1995 were studied. The 1995 Mues- tra samples were reported previously (Reitz, 1999) but are merged with the 1995 General samples here because the small quantity of Muestra mate- rial (number of identified specimens [NISP] = 369) precludes relevant analysis. Due to time constraints and the very large sample recovered in 2001, only a portion of the 2001 material was studied. Based on the results of analysis of the 1995 materials (Reitz, 1999), all the 2001 sample types (Muestra, Gen- eral, 6.35 mm, 1.7 mm) are combined for analysis. Surficial levels were omitted from analysis because of possible intrusions from more recent activities. Five major analytical units structure this analysis, with one of these subdivided into three subcomponents, for a total of seven analytical units. The major analytical units are defined by the area, pit, and temporal period for each sam- ple: Amotape, Siches, Pozo II, Lower Honda, and Upper Honda. Amotape constitutes analytical unit 4 (AU 4), consisting of the oldest deposits. Early Siches-phase materials (AU 3) are divided into two subcomponents designated analytical unit 3b (AU 3b; 7100–6400 14C B.P./~7900–7300 cal B.P.), and analytical unit 3a (AU 3a; 6300–6100 14C B.P./~7200–6900 cal B.P.). Later Siches-phase materials form analytical unit 2 (AU 2; ca. 6000 14C B.P./~6800 cal B.P.). These Siches-phase divisions are defined by stratigraphy and radiocarbon dates. Pozo II likely dates to 6590–6450 14C B.P. (~7500– 7300 cal B.P.). Analytical unit 1 (AU 1) refers to the mixed surface material in Sector II that over- lies the Siches deposits; AU 1 lacks stratigraphic integrity and was not analyzed for this study. The Siches-phase materials excavated from Pozo II in 1995 are contemporaneous with analytical unit 3 (early Siches) but are not merged with that level for reasons explained in more detail below. To test the possibility that the lower and upper levels of Area I, Pits A and B might record environmental change during the Honda phase, the Honda levels are sepa- rated into two analytical units, Lower Honda (LH) and Upper Honda (UH). The different unit naming conventions in Sectors I and II resulted from the need to accommodate the more complex stratigra- phy in Sector II encountered in the 2001 season. The division of the Honda materials into lower and upper components is based on field observations of a transition from warm- to cool- water invertebrates previously observed at other sites in the region as well as at Sitio Siches. This division is based on this knowledge instead of on radiocarbon dates or obvious changes in stratigra- phy. Only warm-water and mixed-water molluscs are present in Lower Honda samples, as well as in the Siches-phase deposits in Pozo II (1995 exca- vations; Richardson et al., 1996). Cool-water and mixed-water invertebrates are more common in the Upper Honda samples. Radiocarbon evidence is ambiguous about a significant chronological dif- ference between the lower and upper levels, and materials undoubtedly moved among adjacent levels (Richardson et al., 1996:9). Except for two outliers (one older, from the basal level but con- tradicted by three other dates in the same level, and one younger), the other 10 Honda dates range between 5150 ± 110 14C B.P. (~5800 cal B.P.) and 4705 ± 90 14C B.P. (~5400 cal B.P.), placing the Honda deposits at a climatic transition ca. 5000 14C B.P. (~5700 cal B.P.) that is most tightly dated in 132 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) the non-archaeological molluscan fauna from the paleoembayment at the Salinas de Chao (8° 40ʹ S latitude, 27–33 km NNE of the Ostra Base Camp; Perrier et al., 1994; Andrus et al., 2003:Figure 3). The vertebrate materials in these seven ana- lytical units were examined using standard zoo- archaeological methods (Reitz and Wing, 2008). Identifications were made by Reitz using the com- parative skeletal collection of the Environmental Archaeology Laboratory, Florida Museum of Nat- ural History, University of Florida. Alana Lynch assisted with the materials excavated in 1995 and Nicole R. Cannarozzi assisted with the 2001 mate- rials. During identification specimens are described by element, portion represented, and symmetry. Evidence for age at death, sex, and modifications is noted when observed. The Number of Identified Specimens (NISP) refers to the number/count of all specimens examined from each context with cross- mended specimens counted as single specimens. The only exceptions to this procedure are unmodi- fied fragments in the Unidentified (UID) Vertebrate category, which are not counted out of concern that the small fragments in this category would crumble further while being counted and were unlikely to be counted accurately. Modified UID Vertebrate specimens are counted so they may be included in tables summarizing modifications. All speci- mens are weighed to provide additional informa- tion about the relative abundance of the animals identified, but neither the weights nor the biomass estimates derived from those weights are included in this study because almost a quarter of the Sitio Siches assemblage consists of otoliths. Assem- blages dominated or consisting largely of otoliths, as is the case for Sitio Siches, yield less reliable biomass estimates than do assemblages consisting of a greater mix of other skeletal specimens. The analysis reported here largely relies on estimates of the Minimum Number of Individuals (MNI or NMI; sensu White [1953]). MNI is used because it is one of the few measures of abun- dance that is available both for Sitio Siches and several other coastal sites, permitting us to do a regional study. MNI facilitates comparing relative abundance across classes with different anatomi- cal structures and is relevant to the larger goal of assessing ecological transitions from warmer to cooler water vertebrates using ecological groups, taxonomic abundance, richness, diversity, equita- bility, and trophic levels on a regional scale. MNI estimates are based on the symmetry, portion, size, and age of skeletal remains (Reitz and Wing, 2008:205–210). MNI is estimated for the lowest taxonomic level consistent with the level of attribution in each analytical unit (compare the list of attributions in Table 5 with those in Table 6). MNI is influenced by the way data from archaeo- logical proveniences are aggregated during analy- sis. The aggregation of separate samples into one analytical unit allows for a conservative estimate of MNI but the “maximum distinction” method, which estimates MNI for separate samples and then combines those estimates, may be more appropri- ately applied when analysis discerns discrete sam- ple units that are definitively separated by time and space (Grayson, 1973). The maximum distinction method yields a much larger MNI. MNI for Sitio Siches is estimated for each of the seven analyti- cal units, with no distinction made for recovery method or subdivisions within each analytical unit. MNI estimates for Pozo II are presented in some of the accompanying tables but are not included in the analysis. Ubiquity is defined as the number of analyti- cal units in which a taxon for which MNI is esti- mated is present. Taxa present in every analytical unit are highly ubiquitous and those present in only one have a very low ubiquity. The presence or absence of elements in an archaeological sample provides data on site forma- tion processes, such as the preservation potential, butchering patterns, or transportation and disposal decisions. The skeletal element categories used here are: Otolith, Tooth, Other cranial specimens, Axial, Appendicular, Foot, and Other specimens. The Other cranial category includes only cranial bones. Teeth are quantified separately in the cat- egory “Tooth.” The Axial category includes either mammal vertebrae and ribs or fish vertebrae. Appendicular includes all bird specimens from the sternum through the phalanges of the manus and REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 133 from the synsacrum through the phalanges of the pes. For mammals and lizards, the Appendicular category includes only forequarter and hindquar- ter specimens other than phalanges, which are included in the Foot category. Mammalian Appen- dicular specimens in the Sitio Siches assemblage consist only of humerii, radii, and tibiae. Mamma- lian Foot elements are carpals, tarsals, metapodials, and phalanges. Specimens quantified as “Other” are too fragmentary to be assigned to one of these more informative categories. The relative age of the mammals, thought to be either seals or sealions, is estimated from the degree of epiphyseal fusion for diagnostic ele- ments (Reitz and Wing, 2008:70–73). When mam- mals are young their elements are not fully formed. Along the area of growth, the shaft, known as the diaphysis, and the ends of the elements, the epiphy- ses, are not fused. When growth is complete the diaphysis and epiphysis fuse. Although environ- mental factors influence the actual age when the element reaches adult dimensions and is fully fused (Watson, 1978), the process follows a regular temporal sequence (Silver, 1963; Schmid, 1972). During analysis, marine mammal specimens are recorded in one of three general categories based on when the maturation process generally is com- pleted and complete fusion occurs for that por- tion of the element. This is more informative for unfused specimens that fuse in the first year or so of life (juveniles) and for fused specimens that are the last to fuse (adults) than for other specimens. An element that fuses before or at ca. 18 months of age and is found fused archaeologically could be from an animal that died immediately after fusion was complete or many years later. The ambiguity inherent in age grouping is controlled by recording each element under the oldest category possible. Modifications to specimens can indicate weathering, human activity such as butchery, or other site formation processes (Reitz and Wing, 2008:123–143). Except for a single modified speci- men from a Lower Honda context, all of the modi- fications observed in the Sitio Siches assemblage can be attributed to burning, probably intentional. These burned specimens may be the result of expo- sure to fire during food preparation; but more likely they were burned intentionally or unintentionally to reduce waste or as additional fuel for fires. Although NISP for specimens attributed to UID Vertebrate are not included in the list of taxonomic attributions, modified UID Vertebrate specimens are quantified in the modification table below. The variety and degree of specialization reflected in the assemblage is measured by the rich- ness, diversity, and equitability of animals in the assemblage. Richness is defined as the number of taxa for which MNI is estimated in each analytical unit. Diversity and equitability permit economic strategies to be assessed in terms of the variety or heterogeneity (diversity) of animals used at the site and the evenness with which taxa were used (equitability). Equitability measures the degree of dependence on specific taxa and the effective vari- ety of species used at the site based on the even, or uneven, use of individual taxa. Biases associ- ated with these indices are discussed elsewhere (Hardesty 1975; Grayson, 1981; Reitz and Wing, 2008:235–246). Diversity and equitability are estimated using the MNI values for each analytical unit. To mea- sure diversity, the Shannon-Weaver Index is used. The formula is: H’ = -Σpilogepi where pi is the number of the ith species, divided by the sample size (Shannon and Weaver, 1949:14). Pi is the evenness component because the Shannon- Weaver Index measures both how many species were used and how much each was used. Equita- bility is estimated using the formula: V’ = H’/Loge S where H’ is the Diversity Index and Loge S is the natural log of the number of observed species (Sheldon, 1969). These indices are interconnected. Diversity increases as both the number of species and the equitability of species increases. A diversity value of 4.99 is very high; a sample with many species identified and in which the number of individuals in the sample slowly declines from most abun- dant to least abundant will be high in diversity. 134 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) Diversity increases when a new taxon is added to the species list, but if another individual of a taxon that is already present in the assemblages is added, diversity decreases. A low diversity can be obtained either by having a few species or by hav- ing a low equitability, where one species is con- siderably more abundant than others. A low equi- tability value indicates that one species was used more heavily than other species in the sample. A high equitability index, approaching 1.0, indicates an even distribution of individuals among the spe- cies in the sample following a normal pattern in which there are a few abundant species, a moderate number of common ones, and several rare ones. Trophic level is based primarily on a spe- cies’ dietary composition, habits, and, to a lesser degree, habitat and reflects the health of the com- munity (Reitz, 2004; Quitmyer and Reitz, 2006). Mean trophic level (TL) is estimated by combining modern trophic level assignments derived from the biological literature with estimates of MNI in the Sitio Siches assemblage for these same taxa (Pauly and Christensen, 1995; Pauly et al., 1998; Pauly et al., 2000). The trophic level assignments used are those published in FishBase 1998 (Froese and Pauly, 1998; Pauly et al., 2000; Reitz, 2003, 2004). Trophic level is estimated using the formula: TLi = Σ (TLij) (MNIij) / Σ MNIi solving for the mean trophic level for the time period (TLi). The trophic level (TLij) of each taxon (j) for the time period (i) is multiplied by the esti- mated MNI of the taxon (j) for the time period (i). TLij is divided by the summed MNI for the time period (MNIi). This formula estimates the mean trophic level for each analytical unit. Sometimes it is necessary to use other taxonomic levels when the identifications in the archaeological data, the mod- ern fishery data, and/or FishBase 1998 are insuffi- ciently precise. In these cases, the trophic level for the closest taxonomic category is used. This same formula is used to estimate the relative emphasis placed on specific trophic levels instead of the mean trophic level by dividing the MNI in each trophic level by the total MNI for the phase. A comprehen- sive review of the strengths and weakness of this approach is published elsewhere (Reitz, 2004). Vertebrate data are summarized in several ways to facilitate comparison. One comparison is based on the number of individuals in biological categories such as Mammals, Birds, Reptiles, and Fishes. Another comparison classifies individuals in terms of the ecological groups used in the review of other coastal zooarchaeological assemblages above (e.g., Table 3). In addition, data are summa- rized in terms of the richness in each analytical unit and the percentage of individuals from each trophic level. As will be seen, the role of anchovies (Engraulidae) is one of the chief differences between Siches- and Honda-phase analytical units. Anchovy remains are very rare in the samples recovered in 1995 (NISP = 122; MNI = 4) and abundant in the samples recovered in 2001 (NISP = 16,091; MNI = 881; Tables 5–6). This difference may reflect environmental or cultural transitions, or both; but probably can be attributed more directly to the different procedures used for 1.7-mm frac- tions in 1995 and 2001. All Honda-phase materials and those from Siches-phase Pozo II are from the 1995 excavations. At that time, excavators did not recognize anchovy otoliths and did not extract them from the 1.7-mm General fractions. For this reason, the summary tables upon which the presentation of results is based in this report exclude all anchovies from consideration to facilitate inter-phase com- parisons. Anchovies, however, are retained in the accompanying lists of vertebrate attributions for each analytical unit. Pozo II is not quantified in the Siches-phase summaries for this same reason, but those data also are retained in the accompanying lists of vertebrate attributions. Unless otherwise stated, anchovies are not included in the following discussion and Pozo II is not discussed at all. Anchovies (Engraulidae) and seatrouts (Cynoscion spp.) pose additional obstacles to this study. Both taxa include some members specific to warm waters and other members characteristic of cool waters. Classifications of anchovies and sea- trouts to an ecological group requires an attribution that includes both a genus and species. None of the Sitio Siches anchovies could be identified beyond family and only a few of the Cynoscion could be REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 135 attributed to a level below genus (e.g., C. cf. ana- lis). Advances in archaeogenetics, geochemistry, and other methods that do not rely on morphology eventually may clarify the taxonomic and ecologi- cal affiliation of these two groups.3 Engraulids are found in warm-tropical/estua- rine waters, sometimes abundantly. Those found in Ecuadorian archaeological assemblages prob- ably are species commonly present in lagoons and warm waters, such as Anchoa spp. Engraulids south of Ecuador are likely cool-water forms, such as Engraulis ringens, which are strongly associ- ated with the Peru-Chilean province and abundant in fine-screened zooarchaeological assemblages from sites south of Lima, Perú. No engraulids are reported for either Las Vegas or Real Alto, the two Ecuadorian sites referenced in this study. Thus, engraulids in the Sitio Siches assemblage are attrib- uted only to family (Engraulidae) but classified as cool-water forms. Seatrouts of the genus Cynoscion also are wide-ranging. Some seatrouts inhabit Peruvian waters but the principal fishing zone for most spe- cies lies in Ecuador. One member of this genus (weakfish, C. analis), for example, is found in both Ecuadorian and Peruvian waters, but the principal fishing zone is Ecuador and northern Perú down to Punta Aguja (Fig. 1; Schweigger, 1964:259; Chirichigno, 1982:198–202). Another member of this genus (C. stolzmanni) prefers brackish lagoons and coastal waters; the principal fishing zone also is Ecuador (Froese and Pauly, 2004). In this study, animals attributed only to Cynoscion spp. are clas- sified as mixed-water forms. All of the primary and secondary data used in this study are subject to several common biases reviewed at length elsewhere (e.g., Grayson, 1984; Reitz and Wing, 2008). In general, small samples frequently yield short lists of taxa with undue emphasis on one taxon in relation to others. It is not possible to determine the nature or the extent of the bias, or correct for it, until more materials are studied. These data also reflect the fact that ele- ments of some animals are more readily identified than others and the animals represented by such elements may appear more significant in terms of specimen count than they were in the economy. If animals are identified largely by unpaired ele- ments, such as vertebrae, the estimated MNI for that animal will be low. High specimen count and low MNI for some animals are artifacts of analy- sis. This source of bias is particularly critical when interpreting assemblages dominated by fishes. In a highly stratified site such as Sitio Siches, vari- ous forms of turbation (e.g., bioturbation, flood- ing) as well as time averaging within and among stratigraphic levels may be two of the most com- mon analytical impediments (Reitz and Shackley, 2012:46), though the full range of taphonomic phe- nomena likely influenced these materials (Reitz and Shackley, 2012:41–68). A PREVIEW OF THE SITIO SICHES RESULTS Federico Scartascini (2017) reported a latitudinal environmental gradient for the desert coast of Pata- gonia, Argentina, despite the apparent homogene- ity of the Patagonian coast. As with the southern Atlantic sites studied by Scartascini (2017), a dom- inant aspect of Sitio Siches is its location in a tran- sition zone between two biogeographical provinces supporting a high-diversity, multi-species complex of marine organisms. The Sitio Siches study shows that people living at this early coastal site were able fishers, as were people at other early- and mid-Holocene coastal sites (Fig. 3; Table 4). Ecological groups, taxonomic richness, diversity, equitability, and mean trophic levels for each phase demonstrate the wide range of marine animals used for five millen- nia at Sitio Siches. These animals have such varied habits and habitat preferences that they required a wide range of fishing methods to acquire (Tables 3, 5–6; Appendix 1). Differences among the phases enhance our ability to see when and where ecologi- cal transitions among warm and cool marine con- ditions occurred (Tables 7–9). After the Amotape phase, half of the vertebrates are characteristic of warm-water conditions, though the dominant ani- mal (Sciaenidae: Cynoscion spp.) is a mixed-water, flexible genus able to live in both warm and cool waters. Over time temperatures trended from reli- ably warm to warmer waters with more sharply sea- 136 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) sonal fluctuations in water temperatures; eventually achieving cool conditions similar to those prevail- ing today at the frontier between the Panamanian and Peru-Chilean provinces. Vertebrates recovered from Sitio Siches suggest when these shifts in the warm-water/cool-water frontier occurred during a span of 5,000 years. RESULTS, AMOTAPE PHASE (ANALYTI- CAL UNIT 4) The earliest Sitio Siches deposits are those in the Amotape analytical unit (AU 4). The Amotape col- lection contains 335 specimens and the remains of an estimated 50 vertebrate individuals, includ- ing anchovies (Tables 5–6). During this phase, the site was further from the shoreline than during the subsequent Siches- and Honda-phase occupa- tions. This may explain why terrestrial individu- als constitute 12% of the MNI, excluding ancho- vies (Engraulidae; Tables 7–8). Despite the small sample size, this analytical unit contains one rodent individual (UID Rodent; Table 6), two cañan liz- ards (Dicrodon sp.), and a non-poisonous snake (Colubridae), in addition to fishes. Fish individu- als, however, dominate the collection (88% of the MNI; Table 7). Half of the individuals are mem- bers of the mixed-water ecological group (53% of the MNI) though members of the warm-water group comprise 29% of the individuals (Table 8). Members of the drum family (Sciaenidae) contrib- ute 62% of the individuals. Mixed-water drums contribute 81% of the drum individuals and warm- water drums constitute 19% of the drum individu- als (Table 9). The most abundant drums are mixed- water seatrouts (Cynoscion spp.). The most com- mon warm-water fishes are mullets (Mugil spp.). Other warm-water vertebrates include bonefishes (Albula vulpes) and sea catfishes (Ariidae). Cool- water individuals are rare (6% of the MNI; Table 8) and cool-water drums are absent (Table 9). Most of the fish specimens are vertebrae (37% of the 295 fish NISP) and otoliths (38%; Table 10). Anchovies (Engraulidae), contribute 40% of the fish vertebrae and 41% of the otoliths. Evidence of burning is found on 17% of the Amotape specimens (exclud- ing UID Vertebrate; Table 11). Diversity of all vertebrates in the collection (Fig. 5) and of the fish component (Fig. 6) is lower than in the subsequent Siches phase, but higher than in the Honda phase at the end of the sequence. This is the most equitable of the seven analytical units, but richness is very low, perhaps due to the small sample size. Low richness also suggests a fishing strategy focused on seatrouts but which also retained non-targeted fish captured as by-catch. The mean marine trophic level exploited during the Amotape phase is 3.2, reflecting the wide range of trophic levels used in addition to a focus on sea- trouts from trophic level 3.4 (Fig. 6; Table 12). RESULTS, SICHES PHASE (ANALYTICAL UNITS 3 AND 2) Although anchovy recovery from Siches-phase materials in 2001 was highly successful, unless otherwise specified, anchovies are not included in the analysis to maintain comparability with the Honda-phase materials in which anchovies are undoubtedly under-represented. Likewise, Pozo II, a Siches-phase unit, is not considered here due to the incomplete recovery of anchovies when that unit was excavated. Results for the three Siches- phase collections are summarized in stratigraphic order, beginning with the oldest deposits, those in analytical unit 3b. Siches Phase, Analytical Unit 3b The AU 3b collection contains 7,072 speci- mens and the remains of an estimated 442 verte- brate individuals, including anchovies (Engrauli- dae; Tables 5–6). Vertebrates include cormorants (Phalacrocorax spp.), cañan lizards (Dicrodon sp.), and fishes (Table 6). Excluding anchovies, fish individuals dominate the collection (99% of the MNI; Table 7). Half of the individuals are mem- bers of the warm-water ecological group (46% of the MNI) and the other half (43%) are members of the mixed-water group (Table 8). Members of the drum family constitute 78% of the individuals in AU 3b and warm-water drums constitute 44% of the drum individuals (Table 9). The most abun- dant fishes are mixed-water seatrouts (Cynoscion spp.), though six other fishes contribute between 5% and 9% of the fish individuals, resulting in the REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 137 highest overall diversity and the highest fish diver- sity of the seven analytical units (Table 12). Nota- ble warm-water vertebrates include bonefishes (Albula vulpes), mullets (Mugil spp.), weakfishes (Cynoscion cf. analis), Pacific drums (Larimus cf. pacificus), and tallfin croakers (Micropogonias spp.). The most common cool-water vertebrates are cabrillas (Paralabrax spp.). Cool-water individu- als are rare (10% of the MNI; Table 8). A limited variety of elements are represented in the Siches AU 3b (Table 13). Sixty-seven percent of the fish specimens (fish NISP = 6,996) are vertebrae and 15% are otoliths. Anchovies contribute 85% of the fish vertebrae and 52% of the otoliths. Evidence of burning is found on 3% of the Siches AU 3b speci- mens (excluding UID Vertebrate; Table 11). Siches Phase, Analytical Unit 3a The AU 3a collection contains 13,556 speci- mens and the remains of an estimated 1,364 ver- tebrate individuals, including anchovies (Engrauli- dae; Tables 5–6). Among these individuals are two terrestrial bird individuals (Corvidae, Mimidae) and one cañan lizard (Dicrodon sp.; Table 6). Less than 1% of the individuals are terrestrial birds and lizards (Table 7). Excluding anchovies, most of the individuals are fishes (99.6% of the MNI). Half of the fish individuals are members of the warm-water ecological group (50% of the MNI) and slightly fewer of the individuals (43%) are from the mixed- water group (Table 8). Members of the drum fam- ily constitute 84% of the individuals in AU 3a and warm-water drums constitute 49% of the drum individuals (Table 9). The most common fishes are warm-water weakfishes (Cynoscion cf. analis) and Pacific drums (Larimus cf. pacificus), in addition to mixed-water seatrouts (Cynoscion spp.) and star drums (Stellifer spp.). Other notable warm-water vertebrates include bonefishes (Albula vulpes), sea catfishes (Ariidae), mullets (Mugil spp.), and tallfin croakers (Micropogonias spp.). The most common cool-water animals are cabrillas (Paralabrax spp.). Cool-water individuals are rare (6% of the MNI). A limited variety of elements are represented in AU 3a (Table 14). Sixty percent of the fish specimens (fish NISP = 13,402) are vertebrae and 25% are otoliths. Anchovies contribute 90% of the fish ver- tebrae and 44% of the otoliths. Evidence of burn- ing is found on 4% of the Siches AU 3a specimens (excluding UID Vertebrate; Table 11). Siches Phase, Analytical Unit 2 Siches-phase AU 2 contains 997 specimens and the remains of an estimated 284 vertebrate individuals, including anchovies (Engraulidae; Tables 5–6). It contains two UID Bird specimens, for which MNI is not estimated, and otherwise con- sists entirely of fishes (Table 7). Excluding ancho- vies, half of the fish individuals are members of the warm-water ecological group (54% of the MNI) and 40% are members of the mixed-water group (Table 8). Drums are the most common vertebrates (82% of the MNI) and half of the drums are mem- bers of the warm-water group (51% of the MNI; Table 9). Mixed-water individuals constitute a high percentage of the specimens in this analytical unit primarily because of the prominence of mixed- water seatrouts (Cynoscion spp., 33% of the MNI). Weakfishes (Cynoscion cf. analis), Pacific drums (Larimus cf. pacificus), tallfin croakers (Micropo- gonias spp.), and Polla drums (Umbrina cf. xanti) contribute 37% of the individuals (Table 6). Other notable warm-water vertebrates include bonefishes (Albula vulpes), sea catfishes (Ariidae), mullets (Mugil spp.), grunts (Orthopristis spp.), and yel- low-eyed croakers (Odontoscion spp.). Cool-water individuals are rare (6% of the MNI). Most of the elements represented in AU 2 are from fishes (Table 15). Twelve percent of the fish specimens (fish NISP = 995) are vertebrae and 67% are oto- liths. Anchovies (Engraulidae) contribute 50% of the fish vertebrae and 28% of the otoliths. Evidence of burning is found on 11% of the AU 2 specimens (excluding UID Vertebrate; Table 11). Summary, Siches Phase The Siches-phase analytical units are simi- lar in most respects (Tables 7–9, 12). All three of the Siches-phase analytical units are dominated by fishes (Tables 6–7) and most of the vertebrates are from warm or mixed waters (Table 8). Seatrouts (Cynoscion spp. and Cynoscion cf. analis) are the dominant vertebrates in all three Siches-phase analytical units, contributing 43% of the Siches- 138 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) Table 5. Number of Identified Specimens (NISP) for each analytical unit. See Table 3 for vernacular names. Attribution AU 4 AU 3b AU 3a AU 2 Pozo II L. Honda U. Honda Total NISP UID Mammal 14 14 74 1152 485 1739 UID Rodent 1 1 Lycalopex cf. sechurae 1 1 Pinnipedia 8 1 9 Otariidae 45 9 54 UID Bird 4 23 43 2 223 96 391 Laridae 2 2 Larosterna inca 10 1 11 Puffinus cf. griseus 70 5 75 Sula spp. 6 6 Phalacrocorax spp. 1 2 1 4 Passeriformes 1 1 Corvidae 1 1 Mimidae 1 1 UID Squamata 2 10 8 20 UID Lizard 1 16 8 5 30 Teiidae 1 1 Dicrodon sp. 17 11 18 2 48 Colubridae 1 1 Chondrichthyes 2 5 4 14 3 28 Carcharhinidae 1 126 20 147 Rajiformes 1 5 1 7 Rhinobatos spp. 6 1 7 Dasyatidae 1 2 11 1 15 UID Actinopterygii 114 844 1471 328 16 1257 310 4340 Albula vulpes 1 34 42 7 3 2 89 Engraulidae 123 5598 10077 293 1 116 5 16213 Clupeidae 3 68 189 5 9 24 298 Ariidae 3 18 37 8 2 12 7 87 Mugil spp. 2 14 29 5 3 24 1 78 Serranidae 7 16 3 1 2 29 Epinephelus spp. 3 1 4 Hemanthias peruanus 1 1 3 3 8 Paralabrax spp. 2 21 73 11 9 16 3 135 Gerreidae 2 7 10 1 20 Haemulidae 1 7 8 Anisotremus spp. 1 1 16 10 28 Haemulon spp. 1 1 1 3 Isacia conceptionis 3 2 5 Orthopristis spp. 2 6 17 10 2 1 1 39 Calamus sp. 1 1 Sciaenidae 13 41 10 5 69 Bairdiella cf. ensifera 1 3 4 Corvina deliciosa 1 1 1 1 4 Cynoscion spp. 28 159 495 115 87 285 174 1343 Cynoscion cf. analis 37 159 54 250 REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 139 Table 5. Continued. Attribution AU 4 AU 3b AU 3a AU 2 Pozo II L. Honda U. Honda Total NISP Cynoscion cf. stolzmanni 1 3 4 Larimus spp. 2 2 Larimus cf. acclivis 4 17 3 24 Larimus cf. effulgens 3 7 2 2 1 15 Larimus cf. pacificus 2 38 186 31 11 190 48 506 Menticirrhus spp. 1 1 2 Micropogonias spp. 1 32 81 22 23 5 3 167 Odontoscion spp. 1 13 42 11 1 5 1 74 Ophioscion spp. 3 4 7 Paralonchurus spp. 5 11 3 4 2 5 30 Stellifer spp. 7 37 213 32 19 1 309 Umbrina cf. xanti 1 16 92 34 5 64 31 243 Labridae 9 58 1 70 4 142 UID Vertebrate 0 Total NISP 335 7072 13556 997 192 3766 1262 27180 Figure 5. Diversity (H') of all vertebrates in each Sitio Siches collection, excluding Engraulidae and Pozo II. See Table 12 for details. 140 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) Table 6. Ecological groups, trophic level, and Minimum Number of Individuals (MNI) for each analytical unit. See Table 3 for vernacular names. Attribution Ecological Group Trophic Level AU 4 AU 3b AU 3a AU 2 Pozo II L. Honda U. Honda Total UID Rodent Terrestrial 1 1 Lycalopex cf. sechurae Terrestrial 1 1 Otariidae Mixed 2 2 4 Larosterna inca Cool 2 1 3 Puffinus cf. griseus Mixed 6 1 7 Sula spp. Cool 1 1 Phalacrocorax spp. Mixed 1 1 1 3 Corvidae Terrestrial 1 1 Mimidae Terrestrial 1 1 Dicrodon sp. Terrestrial 2 1 1 1 5 Colubridae Terrestrial 1 1 Carcharhinidae Mixed 4.0 1 1 1 3 Rajiformes Mixed 3.5 1 1 Rhinobatos spp. Mixed 3.5 1 1 2 Dasyatidae Mixed 3.5 1 1 1 1 4 Albula vulpes Warm 3.0 1 13 21 6 2 2 45 Engraulidae Cool 2.2 16 205 572 88 1 2 1 885 Clupeidae Cool 2.6 1 5 5 1 5 1 18 Ariidae Warm 3.2 1 5 17 5 2 6 4 40 Mugil spp. Warm 2.0 2 2 13 3 2 4 1 27 Epinephelus spp. Warm 3.8 3 1 4 Hemanthias peruanus Warm 3.5 1 1 2 3 7 Paralabrax spp. Cool 3.5 1 14 35 8 6 10 2 76 Gerreidae Warm 3.3 1 4 6 1 12 Anisotremus spp. Cool 3.5 1 1 5 6 13 Haemulon spp. Warm 3.5 1 1 1 3 Isacia conceptionis Cool 3.5 2 2 4 Orthopristis spp. Warm 3.1 1 4 12 6 1 1 1 26 Calamus sp. Warm 3.4 1 1 Bairdiella cf. ensifera Warm 3.3 1 1 2 Corvina deliciosa Cool 3.5 1 1 1 1 4 Cynoscion spp. Mixed 3.4 14 86 257 64 49 133 96 699 Cynoscion cf. analis Warm 3.4 15 86 24 125 Cynoscion cf. stolzmanni Warm 4.0 1 2 3 Larimus cf. acclivis Warm 3.3 3 11 2 16 Larimus cf. effulgens Warm 3.3 2 4 1 2 1 10 Larimus cf. pacificus Warm 3.3 1 22 97 18 8 96 30 272 Menticirrhus spp. Mixed 3.4 1 1 2 Micropogonias spp. Warm 3.3 1 17 48 13 13 2 2 96 Odontoscion spp. Warm 3.3 1 8 25 6 1 3 1 45 Ophioscion spp. Warm 3.3 2 2 4 Paralonchurus spp. Cool 3.4 3 6 2 3 1 1 16 Stellifer spp. Mixed 3.3 3 14 80 13 12 1 123 Umbrina cf. xanti Warm 3.4 1 10 49 18 3 33 20 134 Labridae Cool 3.6 1 1 1 5 1 9 Total MNI 50 442 1364 284 108 331 180 2759 REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 141 Table 7. Summary of taxonomic class MNI for each analytical unit. This table does not include Engraulidae or Pozo II (see Tables 5 and 6 for those data). Class Mammals Birds Reptiles Fishes Total MNIAnalytical Unit MNI MNI % MNI MNI % MNI MNI % MNI MNI % AU 4, Amotape 1 2.9 3 8.8 30 88.2 34 AU 3b, Siches 1 0.4 1 0.4 235 99.2 237 AU 3a, Siches 2 0.3 1 0.1 789 99.6 792 AU 2, Siches 196 100.0 196 Siches Phase Totals 3 0.2 2 0.2 1220 99.6 1225 Lower Honda 3 0.9 10 3.0 1 0.3 315 95.7 329 Upper Honda 2 1.1 3 1.7 174 97.2 179 Table 8. Summary of ecological group MNI for each analytical unit. This table does not include Engraulidae or Pozo II (see Tables 5 and 6 for those data). See Table 3 and Appendix 1 for animals classified in each ecological group. Ecological Group Terrestrial Warm-water Cool-water Mixed-water Total MNIAnalytical Unit MNI MNI % MNI MNI % MNI MNI % MNI MNI % AU 4, Amotape 4 11.8 10 29.4 2 5.9 18 52.9 34 AU 3b, Siches 1 0.4 110 46.4 24 10.1 102 43.0 237 AU 3a, Siches 3 0.4 399 50.4 51 6.4 339 42.8 792 AU 2, Siches 106 54.1 11 5.6 79 40.3 196 Siches Phase Totals 4 0.3 615 50.2 86 7.0 520 42.4 1225 Lower Honda 2 0.6 150 45.6 32 9.7 145 44.1 329 Upper Honda 63 35.2 13 7.3 103 57.5 179 Table 9. Summary of drum MNI for analytical units. This table does not include Engraulidae or Pozo II (see Tables 5 and 6 for those data). See Table 3 and Appendix 1 for animals classified in each ecological group. Ecological Group Warm-water Cool-water Mixed-water Total drum MNIAnalytical Unit MNI MNI % MNI MNI % MNI MNI % AU 4, Amotape 4 19.0 17 81.0 21 AU 3b, Siches 81 43.8 4 2.2 100 54.1 185 AU 3a, Siches 325 48.6 7 1.0 337 50.4 669 AU 2, Siches 82 50.9 2 1.2 77 47.8 161 Siches Phase Totals 488 48.1 13 1.3 514 50.6 1015 Lower Honda 134 49.6 2 0.7 134 49.6 270 Upper Honda 54 35.3 2 1.3 97 63.4 153 142 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) phase individuals. The mean marine trophic level exploited during the Siches phase is 3.3, reflecting the observation that most of the fish individuals are drums from trophic level 3.4, but that 8–12% of the fish individuals in each analytical unit are from trophic levels below 3.3. Except for herrings (Clu- peidae), these low-trophic-level fishes are warm- water bonefishes (Albula vulpes), sea catfishes (Ariidae), mullets (Mugil spp.), and grunts (Ortho- pristis spp.). Cool-water individuals are rare (7% of the MNI). Subtle differences exist among the three Siches-phase analytical units (Figs. 5–8). Materials in the earliest of the Siches-phase analytical units (AU 3b) are slightly more diverse than those in the other two analytical units (Table 12). This applies to both the overall vertebrate portion of each col- lection (Fig. 5) and to the fish portion (Fig. 6). Although the mean trophic level is very similar in all three analytical units (Fig. 6), the percentage of fishes from trophic level 3.4 increases from 48% in the AU 3b collection to 55% in the AU 2 col- lection (Table 12). The higher emphasis on trophic level 3.4 observed in AU 2 can be attributed to an increase in the percentage of warm-water weakfish (Cynoscion cf. analis) and Polla drum (Umbrina cf. xanti) individuals (Tables 6, 9). Terrestrial indi- viduals are present in small numbers only in AU Table 10. Elements represented (NISP) in analytical unit 4. Axial includes either mammal vertebrae and ribs or fish vertebrae. Appendicular includes all bird specimens from the sternum through the phalanges of the manus and from the synsacrum through the phalanges of the pes. For mammals and lizards, this includes forequarter and hindquarter specimens other than phalanges, which are included in the foot category. NISP includes specimens from both the Muestra and General samples. Attribution Otolith Tooth Other Cranial Axial Appendicular Foot Other Total UID Mammal 14 14 UID Rodent 1 1 UID Bird 4 4 UID Squamata 1 1 2 UID Lizard 1 1 Dicrodon sp. 4 9 1 3 17 Colubridae 1 1 Chondrichthyes 2 2 Carcharhinidae 1 1 UID Actinopterygii 15 27 63 9 114 Albula vulpes 1 1 Engraulidae 46 33 44 123 Clupeidae 1 1 1 3 Ariidae 2 1 3 Mugil spp. 2 2 Paralabrax spp. 2 2 Gerreidae 2 2 Orthopristis spp. 2 2 Cynoscion spp. 28 28 Larimus cf. pacificus 2 2 Micropogonias spp. 1 1 Odontoscion spp. 1 1 Stellifer spp. 7 7 Umbrina cf. xanti 1 1 Total NISP 113 28 39 122 6 3 24 335 REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 143 Table 11. Modifications (NISP) in each analytical unit. No modifications other than burning were observed with the exception of a single worked UID vertebrate specimen in IB9a General, Lower Honda (FLMNH 01900358). NISP includes specimens from both the Muestra and General samples. Attribution AU 4 AU 3b AU 3a AU 2 L. Honda U. Honda Total UID Mammal 2 7 14 823 417 1263 UID Rodent 1 1 Lycalopex cf. sechurae 1 1 Pinnipedia 4 1 5 Otariidae 13 7 20 UID Bird 4 6 10 4 3 27 Puffinus cf. griseus 1 1 Mimidae 1 1 UID Squamata 1 4 2 7 UID Lizard 1 6 3 10 Dicrodon sp. 5 2 6 13 Chondrichthyes 1 3 3 7 Carcharhinidae 1 29 7 37 Rajiformes 2 2 Rhinobatos spp. 3 1 4 Dasyatidae 2 3 5 UID Actinopterygii 28 70 149 36 169 120 572 Albula vulpes 4 6 10 Engraulidae 1 17 48 2 13 81 Clupeidae 1 9 5 1 16 Ariidae 2 2 2 1 7 Mugil spp. 6 6 Serranidae 1 3 1 1 6 Paralabrax spp. 3 9 2 1 15 Haemulidae 1 1 Anisotremus spp. 3 3 Orthopristis spp. 1 1 2 Sciaenidae 1 5 6 Bairdiella cf. ensifera 1 1 Corvina deliciosa 1 1 Cynoscion spp. 8 40 138 30 5 221 Cynoscion cf. analis 9 28 16 9 62 Larimus cf. acclivis 3 3 Larimus cf. pacificus 9 35 4 13 2 63 Micropogonias spp. 7 29 8 1 45 Odontoscion spp. 1 4 5 Paralonchurus spp. 2 2 1 5 Stellifer spp. 6 28 5 39 Umbrina cf. xanti 8 2 3 13 Labridae 1 3 6 10 UID Vertebrate 378 438 1019 23 3 30 1891 Total Modifications 434 646 1573 132 1106 597 4488 144 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) Table 12. Summary of richness, diversity (H'), equitability (V'), and trophic levels (TL) for each analytical unit using MNI. Trophic level data from Froese and Pauly (1998, 2004). Data do not include Engraulidae or Pozo II. Category AU 4 AU 3b AU 3a AU 2 L. Honda U. Honda Total MNI 34 237 792 196 329 179 Total Richness 17 26 32 22 28 23 MNI Diversity 2.2612 2.4103 2.3449 2.3384 1.8757 1.7020 MNI Equitability 0.7981 0.7398 0.6766 0.7565 0.5629 0.5428 Fish MNI 30 235 789 196 315 174 Fish Richness 14 24 29 22 21 19 Fish Diversity 2.0136 2.3758 2.3246 2.3384 1.7009 1.5815 Fish Equitability 0.7630 0.7476 0.6904 0.7565 0.5587 0.5371 Mean Fish TL 3.250 3.319 3.326 3.329 3.342 3.373 % fish MNI in each trophic level: Trophic Level: % MNI % MNI % MNI % MNI % MNI % MNI 2.0 6.7 0.9 1.6 1.5 1.3 0.6 2.6 3.3 2.1 0.6 0.5 1.6 0.6 3.0 3.3 5.5 2.7 3.1 0.6 3.1 3.3 1.7 1.5 3.1 0.3 0.6 3.2 3.3 2.1 2.2 2.6 1.9 2.3 3.3 23.3 31.1 34.7 27.6 32.1 20.1 3.4 50.0 48.5 50.4 55.1 53.7 67.2 3.5 3.3 7.2 5.4 6.1 6.7 7.5 3.6-4.0 3.3 0.9 0.8 0.5 1.9 1.1 Figure 6. Mean trophic level (TL) and diversity (H') of fishes in each Sitio Siches collection, excluding Engraulidae and Pozo II. See Table 12 for details. REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 145 3b and AU 3a (Tables 6, 8). In both cases these are birds and cañan lizards, animals absent in AU 2 (Fig. 7; Tables 6–7). During the Siches Phase, warm-water ver- tebrates increased in abundance from 46% to 54% of the individuals, cool-water vertebrates decreased from a high of 10% of the individuals to a low of 6%, diversity declined slightly, equi- tability increased slightly, and the use of higher trophic levels increased between 7100 14C B.P. and 6000 14C B.P. (~7900 and 6800 cal B.P.; Figs. 5–8; Tables 7–9, 12). Individuals other than fishes are slightly more common in the earlier part of the Siches sequence (AU 3a and 3b) than in the latter part (AU 2; Table 7). RESULTS, HONDA PHASE Lower Honda Levels, LH The oldest of the Honda-phase collections, represented by the Lower Honda (LH) analytical unit, contains 3,766 specimens, the remains of an estimated 331 vertebrate individuals, including anchovies (Engraulidae; Tables 5–6). This analyti- cal unit contains the remains of a fox (Dusicyon cf. sechurae), marine mammals (Pinnipedia, Otari- idae), sea birds, and a cañan lizard (Dicrodon sp.), in addition to fishes. Most of the individuals are fishes (96% of the MNI, anchovies excluded; Table 7). Most individuals are members of the mixed- water (44% of the MNI) and warm-water ecologi- cal groups (46% of the MNI; Table 8). Members of the drum family constitute 82% of the individu- als (Table 6). Mixed-water drums constitute 49% of the drum individuals and warm-water drums contribute another 49% of the drum individuals (Table 9). The most abundant fish individuals are seatrouts (Cynoscion spp.), Pacific drums (Larimus cf. pacificus), and Polla drums (Umbrina cf. xanti). Other notable warm-water vertebrates are bone- fishes (Albula vulpes), sea catfishes (Ariidae), mul- lets (Mugil spp.), and tallfin croakers (Micropogo- nias spp.). The most common cool-water fishes are herrings (Clupeidae), cabrillas (Paralabrax spp.), sargos (Anisotremus spp.), and wrasses (Labridae). Cool-water individuals are rare (10% of the MNI; Table 8). Elements represented in the Lower Honda analytical unit are summarized in Table 16. The fox is represented by a distal fused metapodial in IA7b. Elements from the Foot constitute 70% of the marine mammal (Pinnipedia, Otariidae) speci- mens, though other portions of the skeleton are represented. The two sealion individuals include one juvenile and one individual which was prob- Figure 7. Percentages of vertebrate individuals (MNI) in ecological groups in each Sitio Siches collection, excluding Engraulidae and Pozo II. 146 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) ably an adult (Table 17). A proximal unfused tibia epiphysis and diaphysis (which cross-mend) are from a very young juvenile sealion. Eight of the 53 marine mammal specimens are from Pit IA and 45 are from Pit IB. Half (51%) of the 2,242 fish speci- mens are vertebrae and 26% are otoliths. Evidence of burning is found on 29% of the Lower Honda specimens (excluding UID vertebrate; Table 11). This collection contains the only modified speci- men observed, a small worked, burned UID Verte- brate specimen from IB9a General. Upper Honda Levels, UH The Upper Honda (UH) analytical unit rep- resents the youngest, and last, of the analytical units in the Sitio Siches sequence. The collection Table 13. Elements represented (NISP) in Siches-phase AU 3b. Format as in Table 10. Attribution Otolith Tooth Other Cranial Axial Appendicular Foot Other Total UID Mammal 2 12 14 UID Bird 1 2 15 5 23 Phalacrocorax spp. 1 1 UID Squamata 6 4 10 UID Lizard 3 7 6 16 Teiidae 1 1 Dicrodon sp. 4 6 1 11 Chondrichthyes 5 5 Rajiformes 1 1 UID Actinopterygii 52 104 81 591 16 844 Albula vulpes 21 13 34 Engraulidae 534 1080 3984 5598 Clupeidae 7 2 59 68 Ariidae 11 2 5 18 Mugil spp. 3 11 14 Serranidae 7 7 Paralabrax spp. 21 21 Gerreidae 7 7 Haemulon spp. 1 1 Orthopristis spp. 6 6 Sciaenidae 13 13 Bairdiella cf. ensifera 1 1 Corvina deliciosa 1 1 Cynoscion spp. 159 159 Cynoscion cf. analis 37 37 Cynoscion cf. stolzmanni 1 1 Larimus cf. acclivis 4 4 Larimus cf. effulgens 3 3 Larimus cf. pacificus 38 38 Micropogonias spp. 32 32 Odontoscion spp. 13 13 Ophioscion spp. 3 3 Paralonchurus spp. 5 5 Stellifer spp. 36 1 37 Umbrina cf. xanti 16 16 Labridae 9 9 Total NISP 1031 104 1183 4692 16 13 33 7072 REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 147 Table 14. Elements represented (NISP) in Siches-phase AU 3a. Format as in Table 10. Attribution Otolith Tooth Other Cranial Axial Appendicular Foot Other Total UID Mammal 1 2 71 74 UID Bird 1 19 23 43 Passeriformes 1 1 Corvidae 1 1 Mimidae 1 1 UID Squamata 1 7 8 UID Lizard 1 7 8 Dicrodon sp. 5 11 2 18 Chondrichthyes 1 3 4 Rajiformes 5 5 Rhinobatos spp. 6 6 Dasyatidae 1 1 UID Actinopterygii 284 530 28 604 25 1471 Albula vulpes 38 4 42 Engraulidae 1467 1334 7276 10077 Clupeidae 8 5 176 189 Ariidae 32 1 4 37 Mugil spp. 27 2 29 Serranidae 16 16 Epinephelus spp. 3 3 Hemanthias peruanus 1 1 Paralabrax spp. 73 73 Gerreidae 10 10 Haemulidae 1 1 Anisotremus spp. 1 1 Haemulon spp. 1 1 Isacia conceptionis 3 3 Orthopristis spp. 17 17 Sciaenidae 40 1 41 Bairdiella cf. ensifera 3 3 Corvina deliciosa 1 1 Cynoscion spp. 495 495 Cynoscion cf. analis 158 1 159 Cynoscion cf. stolzmanni 3 3 Larimus spp. 2 2 Larimus cf. acclivis 16 1 17 Larimus cf. effulgens 7 7 Larimus cf. pacificus 186 186 Micropogonias spp. 81 81 Odontoscion spp. 41 1 42 Ophioscion spp. 4 4 Paralonchurus spp. 11 11 Stellifer spp. 211 2 213 Umbrina cf. xanti 91 1 92 Labridae 51 7 58 Total NISP 3332 582 1384 8113 22 4 119 13556 148 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) contains 1,262 specimens and the remains of an estimated 180 vertebrate individuals, including anchovies (Tables 5–6). This collection contains the remains of marine mammals (Pinnipedia, Otariidae) and several species of sea birds, but no exclusively terrestrial vertebrates. Fishes dominate the collection even with anchovies excluded (97% of the MNI; Table 7). Most of the individuals either are members of the mixed-water ecological group (57% of the MNI) or of the warm-water group (35% of the MNI; Table 8). Members of the drum family constitute 85% of the individuals. Most of the drums are members of the mixed-water (63% of the MNI) or warm-water ecological group (35% of the MNI; Table 9). The most common fishes are seatrouts (Cynoscion spp.), Pacific drums (Larimus cf. pacificus), and Polla drums (Umbrina cf. xanti). Other notable vertebrates are warm-water sea cat- fishes (Ariidae) in addition to cool-water cabrillas (Paralabrax spp.) and sargos (Anisotremus spp.). Cool-water individuals are rare (7% of the MNI). Elements represented in the Upper Honda analytical unit are summarized in Table 18. The marine mammal specimens (Pinnipedia, Otariidae) are all from Pit IA; nine of the 10 specimens are from the foot. The sealion individuals include at least one juvenile and one individual whose age cannot be determined but which was at least a subadult if not an adult at death (Table 19). Most of the fish specimens are otoliths (49% of the 662 Table 15. Elements represented (NISP) in Siches-phase AU 2. Format as in Table 10. Attribution Otolith Tooth Other Cranial Axial Appendicular Foot Other Total UID Bird 2 2 Rajiformes 1 1 Rhinobatos spp. 1 1 Dasyatidae 2 2 UID Actinopterygii 118 151 1 50 8 328 Albula vulpes 7 7 Engraulidae 187 46 60 293 Clupeidae 5 5 Ariidae 8 8 Mugil spp. 5 5 Serranidae 3 3 Epinephelus spp. 1 1 Hemanthias peruanus 1 1 Paralabrax spp. 11 11 Gerreidae 1 1 Haemulon spp. 1 1 Orthopristis spp. 10 10 Sciaenidae 10 10 Cynoscion spp. 114 1 115 Cynoscion cf. analis 54 54 Larimus cf. acclivis 3 3 Larimus cf. effulgens 2 2 Larimus cf. pacificus 31 31 Micropogonias spp. 22 22 Odontoscion spp. 11 11 Paralonchurus spp. 3 3 Stellifer spp. 32 32 Umbrina cf. xanti 34 34 Total NISP 669 151 47 120 2 8 997 REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 149 fish NISP) and vertebrae (41% of the fish NISP). Evidence of burning is found on 45% of the Upper Honda specimens (excluding UID vertebrate), including eight marine mammal specimens (Table 11). Summary, Honda Phase Vertebrate remains in the Lower and Upper Honda analytical units are very similar in most respects (Tables 7–9, 12). Both analytical units are dominated by fishes (96–97% of the MNI) and most of the vertebrates are mixed-water or warm- water animals. Fishes in both analytical units are dominated by the same three drums: seatrouts (Cynoscion spp.), Pacific drums (Larimus cf. pacificus), and Polla drums (Umbrina cf. xanti). Table 16. Elements represented (NISP) in Lower Honda Analytical Unit. Format as in Table 10. Attribution Otolith Tooth Other Cranial Axial Appendicular Foot Other Total UID Mammal 2 5 9 1 1135 1152 Lycalopex cf. sechurae 1 1 Pinnipedia 2 6 8 Otariidae 11 2 1 31 45 UID Bird 42 6 175 223 Larosterna inca 10 10 Puffinis cf. griseus 7 3 60 70 Sula spp. 6 6 Phalacrocorax spp. 1 1 2 UID Lizard 5 5 Dicrodon sp. 2 2 Chondrichthyes 14 14 Carcharhinidae 4 122 126 Dasyatidae 11 11 UID Actinopterygii 35 12 18 814 378 1257 Albula vulpes 2 2 Engraulidae 116 116 Clupeidae 7 2 9 Ariidae 9 2 1 12 Mugil spp. 4 4 16 24 Serranidae 2 2 Hemanthias peruanus 3 3 Paralabrax spp. 14 2 16 Haemulidae 5 2 7 Anisotremus spp. 10 5 1 16 Isacia conceptionis 2 2 Orthopristis spp. 1 1 Calamus sp. 1 1 Corvina deliciosa 1 1 Cynoscion spp. 237 2 46 285 Larimus cf. pacificus 190 190 Menticirrhus spp. 1 1 Micropogonias spp. 3 2 5 Odontoscion spp. 5 5 Paralonchurus spp. 1 1 2 Umbrina cf. xanti 64 64 Labridae 49 20 1 70 Total NISP 593 67 82 1210 86 39 1689 3766 150 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) Table 17. Epiphyseal fusion for Pinnipedia and Otariidae specimens in Lower Honda analytical unit. Category Unfused Fused Total Early Fusing: Humerus, distal 1 1 Scapula, distal Radius, proximal 1 1 Acetabulum Metapodials, proximal 1st/2nd phalanx, proximal 1 5 6 Middle Fusing: Tibia, distal Calcaneus, proximal 1 1 2 Metapodials, distal 1 2 3 Late Fusing: Humerus, proximal Radius, distal Ulna, proximal Ulna, distal Femur, proximal Femur, distal Tibia, proximal 1 1 Total observations 5 9 14 Figure 8. Percentages of vertebrate individuals (MNI) in warm-water and cool-water ecological groups in Sitio Siches collections, excluding Engraulidae and Pozo II. REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 151 The dominance of mixed-water fishes can be traced to the attribution of Cynoscion specimens. Cool-water individuals are rare (9% of the MNI). The mean marine Honda-phase trophic level is 3.3, reflecting the observation that most of the fish individuals are drums and most of these are from trophic levels 3.3 and, especially, 3.4 (Table 12). Subtle differences exist between the Lower and Upper Honda analytical units (Figs. 5–8; Tables 7–9, 12). The contents of the Lower Honda analytical unit are slightly more diverse and more equitable than the Upper Honda contents. This applies to both the overall vertebrate component (Fig. 5) and to the fish portion of the two analytical units (Fig. 6). Although the mean trophic level is very similar for both analytical units, 54% of the Lower Honda individuals are from trophic level 3.4 compared to 67% of the Upper Honda individuals. The greater emphasis on trophic level 3.4 in the Upper Honda levels is attributable to a marked decline in the percentage of warm-water Larimus cf. pacificus individuals (from 30% of the fish MNI Table 18. Elements represented (NISP) in Upper Honda analytical unit. Format as in Table 10. Attribution Otolith Tooth Other Cranial Axial Appendicular Foot Other Total UID Mammal 1 484 485 Pinnipedia 1 1 Otariidae 1 8 9 UID Bird 1 4 1 90 96 Laridae 2 2 Larosterna inca 1 1 Puffinis cf. griseus 5 5 Phalacrocorax spp. 1 1 Chondrichthyes 3 3 Carcharhinidae 1 19 20 Dasyatidae 1 1 UID Actinopterygii 38 5 1 214 52 310 Engraulidae 5 5 Clupeidae 2 3 19 24 Ariidae 7 7 Mugil spp. 1 1 Hemanthias peruanus 3 3 Paralabrax spp. 3 3 Anisotremus spp. 10 10 Orthopristis spp. 1 1 Sciaenidae 5 5 Corvina deliciosa 1 1 Cynoscion spp. 169 5 174 Larimus cf. effulgens 1 1 Larimus cf. pacificus 48 48 Micropogonias spp. 3 3 Odontoscion spp. 1 1 Paralonchurus spp. 5 5 Stellifer spp. 1 1 Umbrina cf. xanti 31 31 Labridae 4 4 Total NISP 325 6 9 276 11 9 626 1262 152 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) to 17%) and an increase in the percentage of mixed- water Cynoscion spp. individuals (from 42% of the fish MNI to 55%; Table 6). Marine mammals and sea birds are present in both Honda analytical units but terrestrial individuals are present only in the Lower Honda analytical unit (Tables 6–8). The lower diversity and equitability, the slight increase in mean trophic level, and the absence of terrestrial vertebrates in the Upper Honda analytical unit may be due to the difference in sample size; but it also may be part of a broader temporal trend. This possibility is considered in more detail below. SUMMARY OF SITIO SICHES RESULTS The Sitio Siches results are summarized as follows: 1. Fishes typical of shallow, warm waters contrib- ute the majority of the taxa and individuals. 2. Rarely were terrestrial animals used at any time, but they are more abundant in the Amotape collec- tion than in the Siches or Honda collections. 3. Terrestrial animals may reflect early, slightly more humid conditions on the coast during the Amotape phase. 4. Large terrestrial game, such as deer, were not used by people living at Sitio Siches at any time. 5. Some marine mammals and marine birds were used, particularly during the Honda occupation. 6. Warm-water fishes constitute 29% of the indi- viduals in the Amotape deposits, 50% of the indi- vidual in Siches deposits; and 42% of the Honda individuals. 7. The mean trophic level exploited increases very slightly over time; from 3.2 in the Amotape occu- pation to 3.3 in the Honda occupation. 8. Seatrouts (Cynoscion spp.) from trophic level 3.4 constitute 47% of the Amotape fish individuals and 47% of the Honda fish individuals, after declin- ing to 33% in the Siches-phase collection. Warm- water fishes (Larimus cf. pacificus and Umbrina cf. xanti) increase from 7% of the fish individuals in the Amotape phase to 37% of the fish individuals in the Honda-phase collection. 9. The focus on mixed-water seatrouts and the two Table 19. Epiphyseal fusion for Pinnipedia and Otariidae specimens in Upper Honda analytical unit. Category Unfused Fused Total Early Fusing: Humerus, distal Scapula, distal Radius, proximal 1 1 Acetabulum Metapodials, proximal 1st/2nd phalanx, proximal 1 1 2 Middle Fusing: Tibia, distal Calcaneus, proximal Metapodials, distal Late Fusing: Humerus, proximal Radius, distal Ulna, proximal Ulna, distal Femur, proximal Femur, distal Tibia, proximal Total observations 1 2 3 REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 153 warm-water fishes in the Honda phase is associated with low diversity and equitability. 10. These differences likely reflect cultural responses to mid-Holocene environmental changes between 6000 and 5000 14C B.P. (~6800 and 5700 cal B.P.), with some evidence of this change observed in the late Siches-phase occupation. 11. People living at the site during the Amotape phase fished in warm marine waters. 12. People living at the site during the Siches phase fished in even warmer waters. Variations in fishing strategies during the Siches phase may be responses to sharper annual or seasonal deviations than were experienced during the Amotape or Honda occupa- tions. 13. The Honda collection reflects modern condi- tions of mixed waters trending to the cooler side at the frontier between the Peru-Chilean and Panama- nian provinces. FISHING ON THE FRONTIER: SITIO SICHES The following review is based on the premise that the people who lived at Sitio Siches did so year- round for generations and fished for a living, in addition to using terrestrial resources such as plants. Marine invertebrates and marine vegeta- tion were additional attractions accessible from the Sitio Siches location. Fishing was clearly promi- nent in the local economy (Fig. 9; Table 20). Access to fish and other marine resources was undoubtedly the reason people lived on the coast; the perma- nent water seep at the head of the Siches quebrada explains the site’s specific location. Sitio Siches was not unique in this. All zooarchaeological assemblages reviewed in this report clearly dem- onstrate that from the very beginning of our record, people skillfully and consistently used marine ver- tebrates (Table 4). Although quantified invertebrate data are not available for most of the assemblages reviewed here, the abundance of invertebrates at the sites themselves strengthens this assessment (see references in Table 1). Fishing requires a great deal of skill to be successful, measured in terms of the size of the catch, maintaining equipment in good condition, and the survival of the fishing party in an unforgiving sea. People at Sitio Siches were not novices at fishing; they pursued a variety of sophisticated methods to reduce risks and man- age their time, labor, and other resources. It is not uncommon for interpretations of coastal sites to assume explicitly or implicitly that fishing was but one part of seasonal round of hunt- ing, gathering, or farming. We observe that peo- ple who fish are highly protective of good fishing grounds and unlikely to leave them, or their fish- ing gear, unprotected to avoid being usurped by someone else. Nor is it likely that early Americans were unskilled at fishing. If the growing body of evidence for people entering this hemisphere along a Pacific coast route during the terminal Pleisto- cene (e.g., Dillehay et al., 2008; Braje et al., 2017; Scheib et al., 2018) is accurate (and we agree that it is), then people had thousands of years of coastal Table 20. Summary of MNI for each phase by class and for Sitio Siches. This table does not include Engraulidae or Pozo II (see Tables 5 and 6 for those data). Mammals Birds Reptiles Fishes Total MNI Phase MNI MNI % MNI MNI % MNI MNI % MNI MNI % Amotape 1 2.9 3 8.8 30 88.2 34 Siches 3 0.2 2 0.2 1220 99.6 1225 Honda 5 1.0 13 2.6 1 0.2 489 96.3 508 Sitio Siches total 6 0.3 16 0.9 6 0.3 1739 98.4 1767 154 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) experience upon which to build. By the time peo- ple reached Perú they possessed a deep tradition of accumulated knowledge, technologies, and social institutions upon which to draw. This was a resource base that the ancestors of people living at Sitio Siches had relied upon for thousands of years. The focus on marine resources by people liv- ing at Sitio Siches is highlighted by the observation that only six of the 11 mammal, bird, and reptile taxa (i.e., richness) in the Sitio Siches assemblage are from terrestrial habitats (Tables 6, 21). The only exclusively terrestrial animals in the assemblage are rodents, desert foxes, song birds, cañan lizards, and non-poisonous snakes. Only cañan lizards are present in deposits from all three phases. In terms of the dominant vertebrate class, 75% of the taxa are fishes (Table 22). Use of terrestrial resources probably was far more common in the Amotape phase than during the Siches or Honda phases and associated with cañan lizards. One generally would expect a col- lection as small as that representing the Amotape phase to contain fewer taxa compared to larger col- lections and it would not be expected to contain novel taxa. Both richness and diversity are often depressed in such small collections. The higher diversity of the Amotape collection compared to the lower diversity of the much larger Honda-phase collections (Figs. 5–6; Table 12) lends some cred- ibility to the conclusion that sample size is unlikely to be the only explanation for the higher use of ter- restrial resources during the Amotape phase com- pared to the later Siches or Honda phases. Nonetheless, cañan lizards could be intru- sive into the Amotape levels given their habit of burrowing into sandy soils. Although it is not pos- sible to eliminate bioturbation as an explanation for the presence of lizards in the Amotape depos- its, all 17 of the Amotape cañan specimens are from pit features (FN 18, 40, 89, 91, 136, 143). By contrast, only three specimens in Siches analyti- cal unit 3b (FN 32, 50, 104) and eight specimens in Siches analytical unit 3a (FN 87, 102, 104) are from features. Neither of the Lower Honda speci- mens is from a feature (I-A-7B; I-B-7). Although the Siches and Honda specimens may represent lizards that burrowed into the archaeological levels either in the past or recently; it seems less likely that the Amotape lizards are intrusive from later occupations. The post-Amotape decline in lizards may indicate that lizards were less attractive as food or medicinal sources after the Amotape phase, but later depictions on Moche jars and their role in post-Columbian dietary and medicinal practices Figure 9. Percentages of vertebrate individuals (MNI) in ecological groups in Sitio Siches compared to these groups in coastal Peruvian and Ecuadorian zooarchaeological assemblages from the northern Peru- Chilean Province and the Panamanian Province. Sites and collections arranged chronologically within each section. Sitio Siches data do not include Engraulidae or Pozo II. See Figure 1 for locations. REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 155 suggest that lizards continued to be used elsewhere (Holmberg, 1957). Other explanations for the lizard decline include changing availability due to decreasing precipitation and/or cultural preference. Lizards were the primary terrestrial prey for the Paiján people contemporary with the Amotape occupa- tion at Sitio Siches and much less common subse- quently. Analysis of the avifauna (Campbell, 1982) and insect fauna (Churcher, 1966) from the Pleis- tocene-age Talara Tar Seeps, about 30 km south of Sitio Siches, indicates that conditions were wetter at that location at the end of the Pleistocene than they are at present. Richardson (1973; Richard- son et al., 1996; Sandweiss, 2003) concludes that the Late Pleistocene climate near Sitio Siches was more humid than it is today. If cañan lizards are not intrusive into Amotape deposits, they and the other terrestrial animals in the Amotape collection may be evidence that conditions continued to be wetter than present into the early Holocene, between 9500 and 9000 14C B.P. (~10,700 and 10,100 cal B.P.). This is consistent with Chauchat’s (Chauchat, 1992; Pelegrin and Chauchat, 1993) interpretation of the Paiján environment as less arid than at pres- ent, in part because lizards were such common prey for Paiján people. The percentage of mammal and reptile individuals in the Sitio Siches assemblage Table 21. Richness of ecological groups for each phase and for Sitio Siches. Richness is the number of taxa for which MNI is estimated in each category. This table does not include Engraulidae or Pozo II (see Tables 5 and 6 for those data). Ecological Group Terrestrial Warm-water Cool-water Mixed-water Total RichnessPhase # taxa % # taxa % # taxa % # taxa % Amotape 3 17.6 9 52.9 2 11.8 3 17.6 17 Siches 3 8.8 18 52.9 7 20.6 6 17.6 34 Honda 2 6.7 11 36.7 9 30.0 8 26.7 30 Sitio Siches total 6 13.6 19 43.2 9 20.5 10 22.7 44 Table 22. Richness of classes for each phase and for Sitio Siches. This table does not include Engraulidae or Pozo II (see Tables 5 and 6 for those data). Phase Amotape Siches Honda Sitio Siches Class # taxa % # taxa % # taxa % # taxa % Mammals 1 5.9 2 6.7 3 6.8 Birds 3 8.8 4 13.3 6 13.6 Reptiles 2 11.8 1 2.9 1 3.3 2 4.5 Fishes 14 82.4 30 88.2 23 76.7 33 75.0 Total richness 17 34 30 44 156 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) (i.e., MNI; Table 20) clearly declines after the Amotape occupation, as does the number of taxa in the terrestrial ecological group (i.e., richness; Table 21). This suggests that the Sitio Siches landscape was less favorable for terrestrial animals, includ- ing lizards, after 9000 14C B.P. (~10,100 cal B.P.), particularly during the Honda phase. If lizards were deliberate prey during the Amotape phase at Sitio Siches and at sites in the Paiján complex, as seems likely, the relative abun- dance of fishes compared to lizards in the two assemblages may reflect the distance to the sea when the Paiján and Sitio Siches sites were active. Paiján fishing sites are unknown but must have existed given the presence of marine fauna in that assemblage. It seems likely that Paiján-era sites occupied by fishing communities are now sub- merged under the rising sea. Paiján sites are associ- ated with a wide continental shelf and coastal plain; even today these sites generally are located some distance from the sea. In contrast, the continental shelf and coastal plain is very narrow near Sitio Siches, which remains near the shoreline. Due to this coastal topography, the Amotape-phase mate- rials have survived as a record of a north-coast, Paiján-era fishing economy. The richness of the Sitio Siches marine assemblage (number of marine taxa = 38; Tables 3, 21–22) might suggest that people caught what- ever they could, but further consideration indicates that conclusion is incorrect. Given the wealth of the sea (e.g., Chirichigno, 1982; Appendix 1), the Sitio Siches assemblage reflects a strategy intentionally designed to catch drums, specifically seatrouts, with other marine taxa taken as by-catch. This catch was supplemented by the occasional terrestrial mammal, bird, or reptile. In all three phases, mem- bers of the genus Cynoscion contributed at least a quarter of the individuals (Table 6). This does not suggest occasional visits by landlubbers or adven- turous foragers taking advantage of fishes blown onto the beach by a storm, but a skilled fishing community whose members knew which animals were desirable as well as how, when, and where to acquire them; guided by cultural institutions supporting this risky endeavor. A similar focus on drums characterizes the Late Pleistocene deposits from Quebrada Jaguay as well as those from the Ring Site contemporary with Siches (Sandweiss et al., 1996; Reitz et al., 2017). Both sites are in south- ern Perú (Fig. 1), suggesting a widespread, targeted interest in this specific coastal resource during the Late Pleistocene, an interest that continued into the Holocene. Warm-water individuals, members of the ecological group characteristic of the Panamanian province, are abundant in the Sitio Siches assem- blage and individuals in the cool-water ecological group typical of the Peru-Chilean province are not, particularly in the Amotape collection (Figs. 9–10; Tables 23–24). The percentage of warm-water individuals increases from the Amotape collection (29% of the individuals) to 50% of the individu- als in the Siches collection and declines to 42% of the individuals in the Honda collection (Table 23). Less than 9% of the individuals are members of the Table 23. Summary of ecological group MNI for each phase and for Sitio Siches.This table does not include Engraulidae or Pozo II (see Tables 5 and 6 for those data). Ecological Group Terrestrial Warm-water Cool-water Mixed-water Total MNIPhase MNI MNI % MNI MNI % MNI MNI % MNI MNI % Amotape 4 11.8 10 29.4 2 5.9 18 52.9 34 Siches 4 0.3 615 50.2 86 7.0 520 42.4 1225 Honda 2 0.4 213 41.9 45 8.9 248 48.8 508 Sitio Siches total 10 0.6 838 47.4 133 7.5 786 44.5 1767 REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 157 cool-water group during any of the three phases. It seems very likely that most of the Cynoscion in the Sitio Siches assemblage either were animals typi- cal of warm waters, such as C. analis, or were other members of this genus tolerant of variable water conditions that are primarily warm, particularly during the Amotape and Siches phases. The domi- nance of a genus tolerant of a wide range of envi- ronmental conditions suggests waters adjacent to the site were subject to periodic variations in envi- ronmental conditions, as does the presence of high percentages of mixed-water individuals. As part of this trend, warm-water drum individuals increase from 19% of the drum indi- viduals in the Amotape collection to 48% in the Siches collection, and then decline slightly to 44% in the Honda collection (Fig. 11; Table 25). Some of the Siches-phase increase reflects the attribu- Table 24. Summary of warm-water and cool-water MNI for each phase and for Sitio Siches. This table does not include Engraulidae or Pozo II (see Tables 5 and 6 for those data). Phase Amotape Siches Honda Sitio Siches Ecological groups MNI MNI % MNI MNI % MNI MNI % MNI MNI % Warm-water 10 83.3 615 87.7 213 82.6 838 86.3 Cool-water 2 16.7 86 12.3 45 17.4 133 13.7 Total MNI 12 701 258 971 Table 25. Ecological groups of drum (Sciaenidae) MNI in each phase. This table does not include Pozo II. See Tables 5 and 6 for those data. Phase Amotape Siches Honda Ecological Group MNI MNI% MNI MNI% MNI MNI% Warm-water Bairdiella cf. ensifera 2 0.2 Cynoscion cf. analis 125 12.3 Cynoscion cf. stolzmanni 3 0.3 Larimus cf. acclivis 16 1.6 Larimus cf. effulgens 7 0.7 1 0.2 Larimus cf. pacificus 1 4.8 137 13.5 126 29.8 Micropogonias spp. 1 4.8 78 7.7 4 0.9 Odontoscion spp. 1 4.8 39 3.8 4 0.9 Ophioscion spp. 4 0.4 Umbrina cf. xanti 1 4.8 77 7.6 53 12.5 Cool-water Corvina deliciosa 2 0.2 2 0.5 Paralonchurus spp. 11 1.1 2 0.5 Mixed-water Cynoscion spp. 14 66.7 407 40.1 229 54.1 Menticirrhus spp. 0.0 1 0.2 Stellifer spp. 3 14.3 107 10.5 1 0.2 Total drum MNI 21 1015 423 158 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) tion of many fish specimens to a lower-level taxon (Cynoscion cf. analis), which is not present in the Amotape collection, constitutes 10% of the Siches- phase vertebrate individuals, and is absent in the Honda collection (Table 6). The warm-water Pacific drums (Larimus cf. pacificus) and Polla drums (Umbrina cf. xanti) increase over time. Much of this variability can be attributed to the greater rich- ness of the Siches-phase collection, which contains 14 drum taxa compared to six in the Amotape col- lection and 10 in the Honda collection (Table 25). With only a few exceptions, this suggests a progression in the use of warm-water resources from the Amotape phase through the Siches phase and increased use of animals in the mixed-water ecological groups during the Honda phase. This also suggests variable marine conditions best tol- erated by animals characterized by flexible niche and habitat preferences. If this interpretation is cor- rect, waters were primarily warm throughout the sequence, but were warmer, or more frequently warmer, during the Siches phase. This condition persisted for most of the Siches phase but was fol- lowed by cooler or more unstable conditions in the Honda phase. Anchovies (Engraulidae) are excluded from this analysis because of differences in the 1995 and 2001 field methods. The difference between the number of anchovies in the Honda-phase collec- tion excavated in 1995 (NISP = 121) and the Amo- tape/Siches collections (NISP = 16,091) excavated in 2001 almost certainly is because anchovy oto- liths were not recognized in 1995 when the Honda deposits were excavated (Tables 16, 18). This pre- cludes studying the role of these small fishes for the full occupational sequence. The role of anchovies (Engraulidae) in the Amotape/Siches sequence can be considered because the field crew recognized anchovy otoliths and vertebrae in the fine-screen samples during the 2001 field season when these materials were col- lected. Anchovies constitute 32% of the 2001 Amo- tape-phase individuals, peak at a high of 46% of the individuals in the earliest Siches-phase depos- its (AU 3b) and decline to 31% of the individuals in the last Siches-phase collection (AU 2; Table 26). Given the virtual absence of other cool-water members of the Peru-Chilean ecological group and the abundance of vertebrates more typical of the warm-water Panamanian province, it seems likely that the anchovies in the Sitio Siches assemblage are warm-water species though this cannot be deter- mined from the identifications. If that is true, then the trend for increased use of warm-water habitats described above suggests that water conditions would be described as persistently, but increas- Figure 10. Percentages of vertebrate individuals (MNI) in warm-water and cool-water ecological groups in Sitio Siches compared to these groups in coastal Peruvian and Ecuadorian zooarchaeological assemblages from the northern Peru-Chilean Province and the Panamanian Province. Sites and collections arranged chronologically within each section. Sitio Siches data do not include Engraulidae or Pozo II. See Figure 1 for locations. REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 159 ingly, warm until the last centuries of the Siches- phase sequence, when they began to become more frequently or persistently cooler. These proposed changes in water conditions are reflected in the mean trophic levels estimated for each phase and the trophic levels from which most of the fishes were taken (Fig. 12; Table 27). With anchovies (Engraulidae) excluded in order to examine the full stratigraphic sequence, the mean trophic level of the catch increased slightly from the Amotape phase (TL = 3.25) into the Honda phase (TL = 3.53), reflecting an increase in the use of drums from trophic level 3.4. This increase in higher-trophic-level drums persists throughout the occupational sequence from the Amotape phase through the subsequent Siches and Honda analyti- cal units (Table 12). Furthermore, the practice of using fish from several trophic levels during the Amotape phase contrasts with the emphasis on trophic levels 3.3 and 3.4 during the Honda phase. Approximately 20% of the Amotape fish individu- als are from trophic levels below 3.3, compared to 10% in the Siches-phase collection and 5% in the Honda-phase collection. Use of fishes from trophic levels above 3.4 was similar in the Amotape and Siches occupations (7% of the MNI); but use of higher-trophic-level fishes increased in the Honda phase (9% of the MNI). As the use of animals from higher trophic levels came to dominate the fishing effort at Sitio Siches, diversity and equitability declined (Fig. 13; Table 27). The decline in fish diversity is particu- larly interesting when sample size is considered because often diversity increases as sample size and richness increase. In this case, however, fish diversity in the small Amotape collection is more similar to that in the much larger Siches-phase col- lection and is much higher than in the larger Honda collection. Low diversity and equitability in a larger collection is further evidence that the Honda fishing effort focused on a single taxon, probably fishes common in a specific habitat and susceptible to a specific fishing technique (Tables 12, 27). In contrast, the strategy during the Amotape phase was more general, using few fish taxa but more evenly. The shift from this generalist strategy to a focused one can be seen in the Siches-phase col- lection and is pronounced in the Honda collection. This applies both to fishing effort as well as to the use of other vertebrates. Figure 11. Variation in ecological groups for drums (Sciaenidae) in collections from Sitio Siches. 160 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) Table 26. Summary of anchovy (Engraulidae) and drum (Sciaenidae) MNI for ecological groups in Amotape- and Siches-phase analytical units. This table does not include data for the Honda phase or Pozo II (see Tables 5 and 6 for those data). Phase Amotape Siches AU 3b Siches AU 3a Siches AU 2 Dates, B.P. 9500-9000 7100-6400 6300-6100 ca. 6000 Ecological Group MNI MNI % MNI MNI % MNI MNI % MNI MNI % Rodents, birds, reptiles 4 8.0 2 0.5 3 0.2 Engraulidae 16 32.0 205 46.4 572 41.9 88 31.0 Warm-water drums 4 8.0 81 18.3 325 23.8 82 28.9 Cool-water drums 0.0 4 0.9 7 0.5 2 0.7 Mixed-water drums 17 34.0 100 22.6 337 24.7 77 27.1 Other fishes 9 18.0 50 11.3 120 8.8 35 12.3 Total MNI 50 442 1364 284 Table 27. Summary of richness, diversity (H'), equitability (V'), and trophic levels (TL) for each phase using MNI. Trophic level data from Froese and Pauly (1998, 2004). Data do not include Engraulidae or Pozo II. Category Amotape Siches Honda MNI Total 34 1225 508 Total Richness 17 34 30 MNI Diversity 2.2612 2.3815 1.8494 MNI Equitability 0.7981 0.6753 0.5438 Fish MNI 30 1220 489 Fish Richness 14 30 23 Fish Diversity 2.0136 2.3592 1.6892 Fish Equitability 0.7630 0.6936 0.5387 Mean Fish TL 3.250 3.325 3.353 % fish MNI in each trophic level: Trophic Level: % MNI % MNI % MNI 2.0 6.7 1.5 1.0 2.6 3.3 0.9 1.2 3.0 3.3 3.3 0.4 3.1 3.3 1.8 0.4 3.2 3.3 2.2 2.0 3.3 23.3 32.9 27.8 3.4 50.0 50.8 58.5 3.5 3.3 5.9 7.0 3.6-4.0 3.3 0.7 1.6 REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 161 The lower diversity and equitability of the Honda collection, combined with an emphasis on high-trophic-level fishes, may reflect a change in marine conditions, particularly in light of contem- poraneous changes in the marine ecological groups and trophic levels exploited. Although molluscs are not examined as part of the formal zooarchaeologi- cal study, it is important to note that only warm- Figure 12. Fish mean trophic level (TL) and diversity (H') for the three phases at Sitio Siches, excluding Engraulidae and Pozo II. See Table 27 for details. Figure 13. Fish and total diversity (H') for the three phases at Sitio Siches, excluding Engraulidae and Pozo II. See Table 27 for details. 162 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) water and mixed-water molluscs were encountered in the Lower Honda deposits and in the Siches- phase deposits of Pozo II (Richardson, 1973; Rich- ardson et al., 1996; Sandweiss, 2003). Cooler-water molluscs were more common in the Upper Honda levels. We see a similar decline in the warm-water vertebrates present in the Lower and Upper Honda collections, though cool-water vertebrates also decline. This decline occurs in the context of an increase in mixed-water vertebrates (Fig. 7; Table 8). Populations of sessile molluscs respond to unfa- vorable conditions by not reproducing and larvae will not settle in unfavorable areas. As recruitment becomes limited or even unsuccessful, local mol- lusc populations may decline. The fishes used at Sitio Siches, however, are widespread, eurytopic animals, ones that tolerate a broad range of envi- ronmental conditions. They are unlikely to be as sensitive to changing water conditions as sessile organisms might be and the responses of fishes to gradual or modest changes in water conditions may be slower and less obvious. Thus, the response of molluscs to cooling waters might be more immedi- ate than would be the response of mobile fishes; but the increase in mixed-water vertebrates could be another aspect of such a phenomenon. The preference of the dominant drums (Cynoscion cf. analis; Larimus cf. pacificus; and Umbrina cf. xanti) for relatively shallow waters of the warm-water Panamanian province and the very northern portion of the Peru-Chilean province sug- gests that a similar habitat existed when Sitio Siches was occupied and that conditions were very favor- able for these three drums between roughly 7100 and 6000 14C B.P. (~7900 and 6800 cal B.P.). As acknowledged earlier, the classification of animals into ecological groups using modern biogeography must be viewed with caution because all of these animals can be found in other locations today and also may have changed their habitat preferences and niches over the intervening millennia, at least slightly. This is particularly the case for members of the genus Cynoscion. Cynoscion spp. are present in Peruvian zooarchaeological assemblages as far south as Quebrada de los Burros (Fig. 1; Béarez, 2000, 2012; Lavallée et al., 2011); but the species C. analis seldom is reported in southern assem- blages. When it is present, it is a minor component (e.g., Lo Demas [Sandweiss, 1992:115–116], Que- brada de los Burros [Béarez, 2000, 2012]). Lari- mus cf. pacificus and Umbrina cf. xanti are stron- ger markers for warm-water conditions than is Cynoscion cf. analis. The steady increase of these two, small, clearly warm-water drums and the decline in terrestrial vertebrates suggest warmer coastal waters and slightly more humid conditions prevailed during the Amotape occupation; warmer, perhaps drier, conditions during the Siches occu- pation, and cooler, perhaps even drier, conditions prevailed during the Honda occupation. Through- out the occupation of Sitio Siches, however, con- ditions were more similar to those in the southern Panamanian province (to which the site currently belongs) than those in the Peru-Chilean province. Fisheries research shows that temperature changes of only a few degrees can fundamentally alter a marine ecosystem (Chavez et al., 1999, 2003; Attrill and Power, 2002; Greene, 2002; Sandweiss et al., 2004). Changes in sea surface temperature (SST) such as those suggested by the Siches otolith geochemistry (Andrus et al., 2002) might alter the productivity of coastal waters, the reproductive and growth habits of traditional prey vertebrates, the distribution of these animals, and the trophic struc- ture of near-shore waters. The decline in diver- sity evident in the Honda-phase collections and the increased use of a higher-trophic-level drum from the mixed-water ecological group suggests an altered marine ecosystem reflecting temperature changes. It also seems likely that a wider annual varia- tion in water temperatures would increase local diversity as a greater seasonal replacement in ver- tebrate composition might occur. This seems to be the case at the mid-Holocene Ostra Base Camp (Reitz and Sandweiss, 2001), where otolith geo- chemistry indicates enhanced seasonality (Andrus et al., 2002). Thus far, the discussion has focused on environmental changes and presumed cultural responses. The possibility that the fishing strategy itself changed over time should not be discounted, REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 163 however. The trends observed suggest a change in technology, particularly between the Amotape and Siches phases. If, as seems possible, the engrau- lids in the Siches- and Amotape-phase collections are warm-water, Panamanian province anchovies rather than members of the cool-water Peru-Chil- ean province ecological group, they probably were caught in shallow, near-shore waters with nets, scoops, and similar devices. Other smaller warm- water fishes (e.g., sea catfishes [Ariidae], mullets [Mugil spp.], and small drums [Sciaenidae]) could be taken with these same hand-held devices or could be captured in shallow waters with mass- capture techniques such as trotlines, seine nets of various designs, weirs, and traps. Larger/older drum individuals were more likely taken in deeper waters with more substantial gear such as weirs in tidal areas, setlines with gorges or hooks in deeper waters, or leisters in shallow waters. The higher fish diversity in the Siches-phase collections may reflect a fishing technology that included tools and strategies designed to capture both large seatrouts and small anchovies. The richness of the by-catch is a strong argument for mass-capture devices. Much of the evidence for these devices would be invis- ible archaeologically to the extent they were lost or destroyed during use at the coast or manufactured from fugitive organic materials. This interpretation does not preclude the likelihood that technological changes took place within a changing marine envi- ronment. SITIO SICHES IN THE BROADER CONTEXT The goal of the Sitio Siches vertebrate analysis is to discover whether ecological shifts observed in molluscs from sites on the northern Peruvian coast are evident in vertebrates from Sitio Siches and, if so, whether Sitio Siches contributes to understand- ing the location and timing of ecological shifts along the Peruvian coast. We approached this goal by defining ecological groups of animals charac- teristic of distinct ecological zones and observing changes in the relative abundance of these groups in preceramic Peruvian vertebrate assemblages. We conclude that preceramic Peruvian assemblages reflect location, time period, and ecological shifts. South of 12° S latitude, animals characterized as members of the cool-water/Peru Current ecological group comprise over 96% of the warm- or cool- water individuals (Fig. 2; Table 2). Assemblages from sites between 10° S and 8° S latitude (Paiján, Almejas, and Ostra Base Camp) deposited prior to ca. 5000 14C B.P. (~5700 cal B.P.) contain fewer cool-water animals than do more recent assem- blages in the same area (e.g., Alto Salaverry, Los Morteros). In the Panamanian province, marine animals recovered from Real Alto and Las Vegas are primarily warm-water taxa. Sitio Siches lies within a transition zone between these two prov- inces and the animals used at the site reflect this frontier setting. A common feature shared by the early- and mid-Holocene archaeological assemblages from the northern Peru-Chilean province and the south- ern Panamanian province is the prominence of animals from the warm-water ecological group in assemblages from sites above 10° S latitude in the early part of the sequence (Figs. 9–10; Tables 2, 4). The identity of the taxa dominant in assemblages from these sites, however, suggests environmental conditions between 10°–8° S latitude and 4°30ʹ–2° S latitude were distinct. The occupation at Ostra Base Camp occurred shortly before the regional transition to cooler waters ca. 5000 14C B.P. (~5700 cal B.P.). Otolith geochemistry for the site suggests a higher seasonal SST amplitude prevailed during the mid-Holocene occupation compared to conditions prevailing today (Andrus et al., 2002). Bonefishes (Albula vulpes), sea catfishes (Ariidae), and mullets (Mugil spp.) contribute 35% of vertebrate individuals in the Ostra assemblage. No Pacific drums (Larimus cf. pacificus) or Polla drums (Umbrina cf. xanti) are present and seatrouts (Cynoscion spp.) are minor components. Another third of the vertebrate indi- viduals prefer cool waters (30%; Fig. 9; Table 4). Although it is likely that warm-water habitats, such as a shallow bay, were present near Ostra when it was occupied, it is unlikely that marine waters out- side the bay were as persistently warm or as stable as those prevailing at Sitio Siches. The youngest Siches-phase collection (AU 164 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) 2) is roughly contemporaneous with the Ostra assemblage and contains the highest percentage of warm-water individuals in the Sitio Siches assem- blage (Fig. 9; Table 8). This percentage (54%) is remarkably similar to the percentage of warm- water individuals in the Ostra assemblage (53%). Most of the AU 2 warm-water fishes, however, are drums, particularly seatrouts (Cynoscion spp.), Pacific drums (Larimus cf. pacificus), Polla drums (Umbrina cf. xanti), and tallfin croakers (Micropo- gonias spp.). The bonefishes (Albula vulpes), sea catfishes (Ariidae), and mullets (Mugil spp.) that dominate the Ostra assemblage contribute only 7% of the individuals in the AU 2 collection. Further, the percentage of mixed-water individuals (40%) in AU 2 is higher than that of cool-water individu- als (6% of the MNI), indicative of a frontier zone between the northern Peru-Chilean Province and the Panamanian Province. Both the Ostra assemblage and Siches-phase AU 2 collection stand in stark contrast to assem- blages from Las Vegas and Real Alto (a contempo- rary and a later site further north) and Alto Salaverry (a later site located further south). They also are dis- tinct from Los Morteros (Mauricio Llonto, 2015). Los Morteros was not reviewed earlier because it was studied under different protocols; but it is rel- evant here. The site is associated with the Pampa de las Salinas complex in the lower Chao Valley, ca. 5 km from the modern coastline and ca. 28 km north-northeast of Ostra (Fig.1; Sandweiss et al., 1983). Most of the faunal remains were in deposits that two radiocarbon dates suggest formed between 5570–5041 cal B.P. and 5469–5050 cal B.P. (Mau- ricio Llonto, 2015:411–416). These date about 700 years after Ostra (ca. 7100–6200 cal B.P.), and shortly after the climatic transition at about 5700 cal B.P. (see also Andrus et al., 2003:Figure 3 for dates on warm-, cool-, and mixed-water molluscs from non-archaeological deposits at the Salinas de Chao). The Morteros fish assemblage is dominated by cool-water forms (Mauricio Llonto, 2015:274). In many respects, the Sitio Siches assem- blage is more similar to the early Holocene Las Vegas or Middle Valdivia Real Alto assemblages in the Panamanian province. Even so, differences in the types of warm-water vertebrates exist. In the Ecuadorian assemblages, sea catfishes (Ari- idae) and drums (Sciaenidae) are equally common (Las Vegas; 10% and 11% of the Las Vegas MNI respectively) or sea catfishes are much more com- mon than drums (Real Alto; 49% and 13% of the Real Alto MNI respectively). Bonefishes (Albula vulpes) and anchovies (Engraulidae) are not pres- ent in either assemblage and mullets (Mugil spp.) are rare. Many of the warm-water fishes present in both assemblages are absent or rare in assemblages from further south (e.g., snooks [Centropomus spp.], snappers [Lutjanidae], yellow-eyed croak- ers [Odontoscion spp.]). No Umbrina are pres- ent in either Panamanian-province assemblage; Cynoscion and Larimus are minor members. A feature shared by all of the coastal ver- tebrate assemblages, regardless of the province in which they are classified here, is the decline in terrestrial animals between early assemblages (Quebrada Jaguay, Paiján, Las Vegas) and later ones (Fig. 3; Table 4). Terrestrial animals in the Paiján complex are primarily lizards (70% of the individuals) whereas the terrestrial animals in Las Vegas are rodents and foxes (36% of the individu- als). Deer are present in both the Paiján and Las Vegas assemblages but contribute less than 2% of the individuals. The Paiján complex lies much fur- ther from the shoreline than do either Las Vegas or Sitio Siches, so the dominance of terrestrial forms in Paiján samples is not surprising. Probably more interesting is the prominence of lizards in the Pai- ján complex and the virtual absence of terrestrial animals in the other assemblages reviewed here. This same decline in members of the terrestrial ecological group also is found in the stratified Sitio Siches deposits. After 5000 14C B.P. (~5700 cal B.P.), marine resource use in the northern Peru-Chilean province emphasized members of the cool-water ecological group (e.g., Alto Salaverry); but members of the warm-water group continued to dominate assem- blages in the Panamanian province. Although the decline in warm-water individuals between 6000 and 5000 14C B.P. (~6800 and 5700 cal B.P.) is less dramatic at Sitio Siches than it is after 5000 REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 165 14C B.P. among assemblages from sites located between 10° S and 8° S latitude, the percentage of warm-water individuals is nonetheless lower in the Honda-phase collection than in the preceding Siches-phase collection (Table 23). This suggests warmer waters or more sharply seasonal fluctua- tions in water temperatures prevailed during the Siches phase than during the Honda phase when water conditions were similar to their current sta- tus. The Honda conditions are likely typical of sites at the southern end of the Panamanian province after 5000 14C B.P. (~5700 cal B.P.). Warm-water individuals are even more abundant in the Real Alto assemblage than in the Honda-phase collec- tion when mixed-water animals are removed from consideration (Table 2). The abundance of mixed- water animals, however, may be more informative than appears at first glance. Members of this group are ones that are most likely to be cosmopolitan and tolerant of a wide range of water conditions. Some, such as marine mammals and birds, also are highly mobile and able to take advantage of short- term as well as long-term changes in the marine environment. The use of eurytopic animals toler- ant of a broad range of environmental conditions at Sitio Siches and Real Alto probably reflects on- going changes in the coastal environment. Several other explanations offer themselves for this pattern. One of these is recovery technique. It is highly probable that the low numbers of ancho- vies and herrings in some assemblages, such as in the Almejas 6.35-mm fraction and the 1995 sam- ples from Sitio Siches, are due to field methods. Much of this pattern, however, reflects changes in the types of larger fishes represented (bonefishes [Albula vulpes], sea catfishes [Ariidae], drums [Sciaenidae], and mullets [Mugil spp.]), which are unlikely to be as biased by a 6.35-mm screen as anchovies would be. Nonetheless, tests of the hypothesis proposed here should be constructed using large data sets rigorously and consistently collected with fine-screen recovery methods. Among the potential cultural explanations is the possibility that people along this stretch of the southeastern Pacific coast chose to increase their use of animals in the mixed-water ecological group, targeting seatrouts. Changes in residential patterns, such as a shift from year-round to occasional use of the site, might yield a similar increase in the use of mixed-water fishes, but structures and geo- chemical data suggest these sites were used more permanently than if they were simply seasonal fish camps. It seems particularly likely that technologi- cal changes occurred at Sitio Siches over time, par- ticularly in net and other fishing technologies, as well as in labor management. By themselves, these changes may or may not indicate changes in water conditions. If a change in technology occurred, as well as companion changes in social organization and the division of labor, this would explain the shift from members of the warm-water ecological group to those in the mixed-water group. Available geochemical data from Siches otoliths do indicate that marine waters during the Siches phase were 3° C warmer than present, so at least some of the changes are probably the result of climatic change. Additional environmental data are needed to test this. The broad similarities among assemblages at similar latitudes, however, suggest the patterns observed in the Sitio Siches assemblage are not exclusively due to cultural dynamics, but reflect larger environmental patterns influencing the econ- omies of all human communities in the region. By the end of the sequence domestic plants were part of the economies of coastal communities in north- ern Perú. Some communities may have specialized in fishing in order to trade coastal commodities with inland communities in exchange for cultivated plant products (e.g., Pozorski and Pozorski, 1979), perhaps trading with communities in what is now Ecuador (e.g., Piperno and Pearsall, 1998; Piperno et al., 2000; Piperno, 2005, 2011). An equally valid possibility is that warm- water animals increasingly became rare in the Peru- Chilean province after ca. 5000 14C B.P. (~5700 cal B.P.), or more difficult to capture, so that exploiting them was not energetically worthwhile. It is clear that conditions once existed in northern Perú for a typical estuarine complex of animals to be reliably taken in large numbers, a condition that did not prevail by the time Alto Salaverry was occupied. 166 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) Although this could indicate over-exploita- tion of the resource base impacting not just a single animal but an ecologically similar suite of indicator animals over a wide area, it seems more likely that coastal waters in northern Perú gradually cooled and those of the Panamanian province became persistently warm as the Peru Current assumed its present course as part of an overall mid-Holocene climatic trend. This change was less pronounced in the Panamanian province and in zooarchaeological assemblages from that region; but it is evident at the frontier between the two biogeographical prov- inces. Alto Salaverry and the Honda-phase occu- pation of Sitio Siches approximate modern water conditions at those two locations, with cool-water animals dominant to the south and slightly later at Alto Salaverry and warm-water animals dominant to the north and slightly earlier at Sitio Siches. This change in coastal conditions appears to be consistent with similar changes documented for other areas around the globe (Sandweiss et al., 1999). It also is consistent with the evidence from debris flows recovered from an alpine lake 75 km east of the Pacific Ocean in southwestern Ecuador (Rodbell et al., 1999), as well as the sites of Las Vegas and Sitio Siches. These suggest that El Niño cycles involving aperiodic oscillations in water temperature became more frequent over the last 5,700 years, replacing earlier conditions where waters were more consistently warm (Rodbell et al., 1999; Sandweiss et al., 1996, 2007). This tran- sition from a regime of persistent warm waters to one of cooler waters with increasingly frequent warmer oscillations may have begun around 11,400 cal B.P. and intensified after 7900–5700 cal B.P. The archaeological data reviewed here sup- port this possibility. The warmer-water fauna of the early Holocene was replaced by animals prefer- ring cooler waters at about this same time, around 5700 cal B.P.; variations among the Siches-phase collections may reflect transitional oscillations in water temperature during that phase leading up to the transition. Some of these may reflect coastal El Niños in addition to basin-scale events (e.g., Hu et al., 2018). On a broader scale, Sitio Siches demon- strates the benefits of comparing diachronic trends within a single site and synchronic trends across multiple regions and coastal biomes (e.g., Moss, 2012). It can no longer be considered appropriate to extrapolate from a single coastal site to all coastal sites, even when those sites are roughly contem- poraneous. Although shell, lithic, and bone tools are recovered from these sites, our ability to dem- onstrate cultural affiliation within Sitio Siches or among local and regional groups is limited because ceramic evidence for cultural affiliations is absent. Nonetheless, lithic evidence from Quebrada Jag- uay documents either down-the-line or direct trade occurred between coastal communities and inland locations (Sandweiss et al., 1998; Rademaker et al., 2013; Sandweiss and Rademaker, 2013). In the case of the Paiján complex, this trade, either direct or indirect through intermediaries, included fish products. Lithic evidence from Ostra Base Camp indicates a relationship existed between that community and coastal Ecuador (Sandweiss, 1996), 700 km to the north. The artifacts are part of a northwestern Andean pebble-figurine tradi- tion associated with the tropical coast. Botanical evidence from Sitio Siches also supports interac- tion with coastal Ecuador (Cano, 1995; Piperno and Pearsall, 1998; Piperno et al., 2000; Piperno, 2005, 2011). Thus, coastal communities did not exist in isolation, even early ones; though we argue against presuming all coastal sites were seasonal camps. It does, however, demonstrate the impor- tance of studying interactions among coastal and non-coastal communities on regional and interre- gional scales (e.g., Erlandson, 2016). CONCLUSIONS Sitio Siches data contribute to our knowledge of coastal life during the early- and mid-Holocene by documenting changes in resource use at the same time coastal environments responded to larger global climate phenomena. First, Sitio Siches con- tributes to regional data showing that early- and mid-Holocene coastal peoples were able fishers. Ecological groups, taxonomic richness, diversity, equitability, and mean trophic levels demonstrate that a wide range of marine animals were taken REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 167 from diverse habitats using several different tech- nologies and these dominate the Sitio Siches assem- blage. The strategies used to access this resource base were neither simple nor experimental. They clearly targeted specific fish but also used some of the by-catch. Second, these data enhance our abil- ity to see when and where ecological transitions among warm and cool marine conditions occurred. After the Amotape phase, half of the vertebrates are characteristic of warm-water conditions, though the dominant fish is sufficiently flexible to be found in both warm and cool waters. Over time water tem- peratures trended from reliably warm to warmer waters with more sharply seasonal fluctuations in water temperatures; eventually reaching cool con- ditions similar to those prevailing today at the fron- tier between these two provinces. This description is consistent with a transitional boundary between the Panamanian and Peru-Chilean provinces, an area in which the warm/cold water frontier shifted several times over a 5,000-year period, setting the stage for later cultural developments in this region. ACKNOWLEDGEMENTS We are grateful to James B. Richardson III for the opportunity to examine faunal materials excavated under his co-direction. We also appreciate access to the comparative skeletal collection at the Florida Museum of Natural History permitted by Elizabeth S. Wing and Kitty Emery and the assistance of Irvy R. Quitmyer and Alana Lynch with the Sitio Siches assemblage. We are grateful for the thought- ful comments on the manuscript provided by Kitty Emery, Jean Hudson, and an anonymous reviewer, and for the efforts of Ani St. Amand, University of Maine, for her assistance with the figures. Figures 1 and 4 were prepared by Susan Duser. This work was funded in part by the H. John Heinz III Chari- table Trust. FOOTNOTES 1. We include calibrated median probability dates using Calib 7.1 (Reimer et al., 2013) and the SHCal databases (Hogg et al., 2013). Age ranges for sites used in this report are based on the uncalibrated means of earliest and latest radiocarbon dates for each site, excluding outliers rejected by the excava- tors, or on excavator’s estimates when no 14C dates are available. The sources cited in Table 1 provide lists of dates and standard deviations. Throughout the text we refer to fish and fishing. Only vertebrate data are presented here because comparable inver- tebrate data are not available for all of the assem- blages upon which this study is based. Despite this, our concept of “fishing” includes crustaceans, mol- luscs, and echinoderms in addition to sharks, rays, and bony fishes. We consider fishing for inverte- brates to be among the important activities at Sitio Siches and other coastal sites, though in need of further study. 2. Gaps from 12°–10° and 8°–4° are due to lack of quantified zooarchaeological data meeting the critieria required to be included in this study, not because we choose to exclude quantified data at variance with our analysis. 3. Climatic interpretations based on zooarchaeo- logical attributions and modern habitat prefer- ences are subject to a wide range of biases that lend caution to our interpretations. To correct for this, we have submitted samples of Engraulidae oto- liths and vertebrae and Cynoscion otoliths to Dr. Mike Buckley (University of Manchester, UK) for Matrix Assisted Laser Desorption Ionization Time of Flight Mass Spectrometry (ZooMS) to obtain an additional proxy assessment of the climate signal by using proteins to distinguish between warm- water and cool-water members of these two groups. These two taxa were selected because they include species with divergent preferences for warmer or cooler waters. Distinguishing between species of these two fishes could substantially influence future studies. Oxygen isotopes from the final pre- mortem increment of Cynoscion otoliths also are being assessed. This multi-proxy study is on-going and results are not yet available. 4. Direct exposure to heat alters the biochemistry of otoliths (Andrus and Crowe, 2002), but none of the Ostra Base Camp or Sitio Siches otoliths subjected to biochemical analysis showed signs of being heat-altered (Andrus et al., 2002, 2003). 168 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) 5. Sitio Siches data reported here replace those in all previous publications. LITERATURE CITED Andrus, C. F. T., and D. E. Crowe. 2002. Altera- tion of otolith aragonite: effects of prehistoric cooking methods on otolith chemistry. Jour- nal of Archaeological Science 29:291–299. Andrus, C. F. T., D. E. Crowe, D. H. Sandweiss, E. J. Reitz, and C. S. Romanek. 2002. Otolith δ18O record of mid-Holocene sea surface tem- peratures in Peru. Science 295:1508–1511. Andrus, C. F. T., D. E. Crowe, D. H. Sandweiss, E. J. Reitz, and C. S. Romanek. 2003. Response to comments on “Otolith δ18O record of mid- Holocene sea surface temperatures in Peru”. Science 299:203b. Attrill, M. J., and M. Power. 2002. Climate influ- ence on a marine fish assemblage. Nature 417:275–278. Béarez, P. 2000. Archaic fishing at Quebrada De Los Burros, southern coast of Peru. Recon- struction of fish size using otoliths. Archaeo- fauna 9:29–34. Béarez, P. 2012. Los peces y la pesca. Pp. 99–123 in D. Lavallée and M. Julien, eds. Prehistoria de la Costa Extremo-Sur del Perú. Los Pesca- dores Arcaicos de la Quebrada de los Burros (10000–7000 A.P.). IFEA-PUCP, Lima, Peru. Béarez, P., D. Jackson, and N. Mollare. 2015. Early Archaic fishing (12,600–9,200 cal yr BP) in the semiarid north coast of Chile. The Journal of Island and Coastal Archaeology 10:133– 148. Benfer, R. A. 1984. The challenges and rewards of sedentism: the preceramic village of Paloma, Peru. Pp. 531–558 in M. N. Cohen and G. J. Armelagos, eds. Paleopathology and the Origin of Agriculture. Academic Press, New York. Benítez-López, A. 2018. Animals feel safer from humans in the dark. Science 360:1185–1186. Braje, T. J., T, D. Dillehay, J. M. Erlandson, R. G. Klein, and T. C. Rick. 2017. Finding the first Americans. Science 358:592–594. Briggs, J. 1974. Marine Zoogeography. McGraw- Hill, New York, 475 p. Byrd, K. M. 1976. Changing animal utilization pat- terns and their implications: southwest Ecua- dor (6500 B. C.–A. D. 1400). Ph.D. disserta- tion. University of Florida, Gainesville, 165 p. Byrd, K. M. 1996. Subsistence strategies in coastal Ecuador. Pp. 87–101 in E. J. Reitz, L. A. New- som, and S. J. Scudder, eds. Case Studies in Environmental Archaeology. Plenum Press, New York. Campbell, K. E. 1982. Late Pleistocene events along the coastal plain of northwestern Peru. Pp. 423–440 in G. Prance, ed. Biological Diversification in the Tropics. Columbia Uni- versity Press, New York. Cano Echevarria, A. A. 1995. Proyecto arqueo- logico “Siches:” Estudio de los restos veg- etales. Ms. on file, Anthropology Department and Climate Change Institute, University of Maine, Orono, 3 p. Carré, M., J. P. Sachs, S. Purca, A. J. Schauer, P. Braconnot, R. Angeles Falcón, M. Julien, and D. Lavallée. 2014. Holocene history of ENSO variance and asymmetry in the eastern tropi- cal Pacific. Science 345:1045–1048. Carrillo de Espinoza, N. 1970. Contribución al conocimiento de los reptiles del Peru. Pub- licaciones del Museo de Historia Natural “Javier Prado”, Serie A: Zoología 22, Lima, Peru, 64 p. Carrott, J., and H. Kenward. 2001. Species asso- ciations among insect remains from urban archaeological deposits and their significance in reconstructing past human environments. Journal of Archaeological Science 28:887– 905. Chase, T. 1988. Restos fáunicos. Pp. 171–178 in K. E. Stothert, ed. La Prehistoria Temprana de la Península de Santa Elena, Ecuador: Cultura Las Vegas. Miscelánea Antropológica Ecu- atoriana Serie Monográfica 10. Museos del Banco Central del Ecuador, Guayaquil. Chauchat, C. 1976. The Paiján Complex, Pampa de Cupisnique, Peru. Ñawpa Pacha 13:85–96. Chauchat, C. 1978. Additional observations on the Paiján Complex. Ñawpa Pacha 16:51–65. Chauchat, C. 1988. Early hunter-gatherers on the Peruvian coast. Pp. 41–66 in R. Keatinge, ed. REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 169 Peruvian Prehistory. University of Cambridge Press, Cambridge. Chauchat, C. 1992. Préhistoire de la Côte Nord du Pérou Le Paijanien de Cupisnique. Cahiers du Quaternaire 18, CNRS-Éditions, Paris, 391 p. Chavez, F. P., J. Ryan, S. E. Lluch-Cota, and C. M. Ñiquen. 2003. From anchovies to sardines and back: multidecadal change in the Pacific Ocean. Science 299:217–221. Chavez, F. P., P. G. Strutton, G. E. Friederich, R. A. Feely, G. C. Feldman, D. G. Foley, and M. J. McPhaden. 1999. Biological and chemical response of the equatorial Pacific Ocean to the 1997-1998 El Niño. Science 286:2126–2131. Chirichigno F., N. 1974. Clave para identificar los peces marinos del Perú. Instituto del Mar del Perú Informe No. 44, Callao, Perú, 387 p. Chirichigno F., N. 1982. Catálogo de especies mari- nas de interés económico actual o potencial para America Latina: Parte II, Pacífico Centro y Suroriental. FAO, Rome, Italy, 588 p. Churcher, C. S. 1966. The insect fauna from the Talara Tar-Seeps, Peru. Canadian Journal of Zoology 44:985–993. Cooke, R. 1992. Prehistoric nearshore and littoral fishing in the eastern tropical Pacific: an ich- thyological evaluation. Journal of World Pre- history 6:1–49. DEIS. 1978. Draft Environmental Impact State- ment for Preferred Alternative Location for a Fleet Ballistic Missile (FBM) Submarine Support Base: Appendix 2. Department of the Navy, Washington, D.C., 613 p. Dillehay T. D., C. Ramírez, M. Pino, M. B. Col- lins, J. Rossen, and J. D. Pino-Navarro. 2008. Monte Verde: Seaweed, food, medicine, and the peopling of South America. Science 320:784–786. Dillehay, T. D., D. Bonavia, S. L. Goodbred, Jr., M. Pino, V. Vásquez, and T.R. Tham. 2012. A late Pleistocene human presence at Huaca Prieta, Peru, and early Pacific coastal adapta- tions. Quaternary Research 77:418–423. Dillehay, T. D., S. Goodbred, M. Pino, V. F. Vásquez Sánchez, T. Rosales Tham, J. Adovasio, M. B. Collins, P. J. Netherly, C. A. Hastorf, K. L. Chiou, D. Piperno, I. Rey, and N. Velc. 2017. Simple technologies and diverse food strategies of the Late Pleistocene and Early Holocene at Huaca Prieta, coastal Peru. Sci- ence Advances 3:e1602778. DOI: 10.1126/ sciadv.1602778. Eigenmann, C. H. 1921. The nature and organi- zation of the fishes of the Pacific slope of Ecuador, Peru, and Chile. Proceedings of the American Philosophical Society 60:503–523. Ekman, S. 1953. Zoogeography of the Sea. Sidg- wick and Jackson, London, 417 p. Eisenberg, J. F., and K. H. Redford. 1999. Mam- mals of the Neotropics: The Central Neo- tropics, vol. 3, Ecuador, Peru, Bolivia, Brazil. University of Chicago Press, Chicago, 644 p. Erickson, D. L., B. D. Smith, A. C. Clark, D. H. Sandweiss, and N. Tuross. 2005. An Asian ori- gin for a 10,000-year-old domesticated plant in the Americas. Proceedings of the National Academy of Sciences 102:18,315–18,320. Erlandson, J. M. 2016. Coastal versus interior: some thoughts on the archaeology of Califor- nia’s Channel Islands. The Journal of Island and Coastal Archaeology 11:443–446. Eschmeyer, W. N., E. S. Herald, H. Hammann, and K. P. Smith. 1984. A Field Guide to Pacific Coast Fishes of North America. Houghton Mifflin, Boston, 336 p. Fontugne, M., P. Usselmann, D. Lavallée, M. Julien, and C. Hatté. 1999. El Niño variabil- ity in the coastal desert of southern Peru dur- ing the mid-Holocene. Quaternary Research 52:171–179. Fowler, H. W. 1945. Los peces del Perú: Catálogo sistemático de los peces que habitan en aguas peruanas. Publicaciones del Museo de Historia Natural “Javier Prado”, Universidad Nacional Mayor de San Marcos, Lima, Perú, 298 p. Froese, R., and D. Pauly, editors. 1998. FishBase 98: Concepts, Design and Data Sources. The International Center for Living Resources Management, Kakati City, Philippines, 293 p. Froese, R., and D. Pauly, editors. 2004. FishBase. World Wide Web electronic publication. , version (05/2004). 170 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) Gaynor, K. M., C. E. Hojnowski, N. H. Carter, and J. S. Brashares. 2018. The influence of human disturbance on wildlife nocturnality. Science 360:1232–1235. Giovannetti, M. A., V. S. Lerna, C. G. Bartoli, and A. Capparelli. 2008. Starch grain character- ization of Prosopis chilensis (Mol.) Stuntz and P. flexuosa DC, and the analysis of their archaeological remains in Andean South America. Journal of Archaeological Science 35:2973–2985. Grayson, D. K. 1973. On the methodology of fau- nal analysis. American Antiquity 38:432–439. Grayson, D. K. 1981. A critical view of the use of archaeological vertebrates in paleoenviron- mental reconstruction. Journal of Ethnobiol- ogy 1:28–38. Grayson, D. K. 1984. Quantitative Zooarchaeol- ogy: Topics in the Analysis of Archaeological Faunas. Academic Press, Orlando, 202 p. Grimwood, I. R. 1969. Notes on the distribution and status of some Peruvian mammals. Amer- ican Committee for International Wildlife Protection Special Publication No. 21, 86 p. Greene, K. 2002. Coastal cool down. Science 295:1823. Guillén G., O. 1980. The Peru Current system. 1: physical aspects. Pp. 185–216 in Proceedings of the Workshop on the Phenomenon Known as ‘El Niño’. UNESCO, Paris. Hardesty, D. L. 1975. The niche concept: sugges- tions for its use in human ecology. Human Ecology 3:71–85. Handler, P. 1984. Possible association of strato- spheric aerosols and El Niño type events. Geophysical Research Letters 11:1121–1124. Hildebrand, S. F. 1946. A Descriptive Catalog of the Shore Fishes of Peru. U.S. National Museum Bulletin 189, 530 p. Hoese, H. D., and R. H. Moore. 1977. Fishes of the Gulf of Mexico: Texas, Louisiana, and Adja- cent Waters. Texas A & M University Press, College Station, 327 p. Hogg, A. G., Q. Hua, P. G. Blackwell, C. E. Buck, T. P. Guilderson, T. J. Heaton, M. Niu, J. G. Palmer, P. J. Reimer, R. W. Reimer, C. S. M. Turney, and S. R. H. Zimmerman. 2013. SHCal13 Southern Hemisphere Calibra- tion, 0–50,000 Years cal BP. Radiocarbon 55:1889–1903. Hu, Z.-A., B. Huang, J. Zhu, A. Kumar, and M. J. McPhaden. 2018. On the variety of coastal El Niño events. Climate Dynamics Vol.: 0112 33456789. https://doi.org/10.1007/s00382- 018-4290-4. Holmberg, A. R. 1957. Lizard hunts on the north coast of Peru. Fieldiana: Anthropology 36:203–220. Jones, K. B., G. W. L. Hodgins, and D. H. Sand- weiss. 2017. Radiocarbon chronometry of site QJ-280, Quebrada Jaguay, a Ter- minal Pleistocene to Early Holocene fish- ing site in southern Peru. The Journal of Island and Coastal Archaeology. DOI: 10.1080/15564894.2017.1338316. Kenward, H. K. 1975. Pitfalls in the environmen- tal interpretation of insect death assemblages. Journal of Archaeological Science 2:85–94. Kenward, H. K. 2006. The visibility of past trees and woodlands: testing the value of insect remains. Journal of Archaeological Science 33:1368–1380. Kenward, H. K., and A. Hall. 1997. Enhancing bio- archaeological interpretation using indicator groups: stable manure as a paradigm. Journal of Archaeological Science 24:663–673. Koepcke, M. 1970. The Birds of the Department of Lima, Peru. Livingston Publishing, Wyn- newood, PA, 144 p. Lavallée, D., M. Julien, P. Béarez, A. Bolaños, M. Carré, A. Chevalier, T. Delabarde, M. Fon- tugne, C. Rodríguez-Loredo, L. Klaric, P. Usselmann, and M. Vanhaeren. 2011. Que- brada de los Burros. Los primeros pescadores del litoral Pacífico en el extremo sur peruano. Chungara, Revista de Antropología Chilena 43:333–351. Lippi, R. D. 1983. La Ponga and the Machalilla phase of coastal Ecuador. Ph.D. dissertation. University of Wisconsin, Madison, 534 p. Lumbreras, L. G. 1974. The Peoples and Cultures of Ancient Peru, translated by B. J. Meggars. REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 171 Smithsonian Institution Press, Washington, D.C., 248 p. McInnis, H. E. 1999. Subsistence and maritime adaptations at Quebrada Jaguay, Camaná, Peru: a faunal analysis. M.A. thesis. Univer- sity of Maine, Orono, 146 p. Mauricio Llonto, A. C. 2015. Los Morteros: early monumentality and environmental change in the lower Chao Valley, northern Peru- vian coast. Ph.D. dissertation. University of Maine, Orono, 559 p. Mann F., G. 1950. Peces de Chile: Clave de Deter- minación. Ministerio de Agricultura, Santiago de Chile, Chile, 48 p. Moss, M. 2012. Understanding variability in North- west coast faunal assemblages: beyond eco- nomic intensification and cultural complexity. The Journal of Island and Coastal Archaeol- ogy 7:1–22. Myers, R. A., and B. Worm. 2003. Rapid world- wide depletion of predatory fish communities. Nature 423:280–283. Nims, R., and V. L. Butler. 2018. Increasing the robustness of meta-analysis through life his- tory and middle-range models: an example from the Northeast Pacific. Journal of Archae- ological Method and Theory. DOI: 10.1007/ s10816–018-9383-1. Nowak, R. M., and J. L. Paradiso. 1983. Walker’s Mammals of the World. The Johns Hopkins University Press, Baltimore, 1362 p. O’Connor, T., and J. Evans. 2005. Environmental Archaeology: Principles and Methods, Sec- ond Edition. Sutton Publishing, Stroud, Eng- land, 280 p. Pauly, D., and V. Christensen. 1995. Primary pro- duction required to sustain global fisheries. Nature 374:255–257. Pauly, D., V. Christensen, J. Dalsgaard, R. Froese, and F. Torres, Jr. 1998. Fishing down marine food webs. Science 279:860–863. Pauly, D., V. Christensen, R. Froese, and M. L. Palomares. 2000. Fishing down aquatic food webs. American Scientist 88(1):46–51. Pelegrin, J., and C. Chauchat. 1993. Tecnología y función de las puntas de Paiján: el aporte de la experimentación. Latin American Antiquity 4:367–382. Perrier, C., C. Hillaire-Marcel, and L. Ortlieb. 1994. Paléogéographie littorale et enregistre- ment isotopique (13C, 18O) d’événements de type El Niño par les mollusques Holocènes et récents du nordouest Péruvien. Géographie Physique et Quaternaire 48:23–38. Piperno, D. R. 2005. Untitled report on Cucurbita spp. from Sitio Siches. Ms. on file, Anthropol- ogy Department and Climate Change Insti- tute, University of Maine, Orono, 6 p. Piperno, D. R. 2011. The origins of plant culti- vation and domestication in the New World tropics. Current Anthropology 52, Supple- ment 4:S453–S470. Piperno, D. R., T. C. Andres, and K. E. Stothert. 2000. Phytoliths in Cucurbita and other Neo- tropical Cucurbitaceae and their occurrence in early archaeological sites from the lowland American tropics. Journal of Archaeological Science 27:193–208. Piperno, D. R., and D. M. Pearsall. 1998. The Origins of Agriculture in the Lowland Neo- tropics. Academic Press, San Diego, 400 p. Piperno, D. R., and K. E. Stothert. 2003. Phyto- lith evidence for early Holocene Cucurbita domestication in southwest Ecuador. Science 299:1054–1057. Plunkett, G., N. J. Whitehouse, V. A. Hall, D. J. Charman, M, Blaauw, E. Kelly, and I. Mul- hall. 2009. A multi-proxy palaeoenvironmen- tal investigation of the findspot of an Iron Age bog body from Oldcroghan, Co. Offaly, Ireland. Journal of Archaeological Science 36:265–277. Pozorski, S. G. 1976. Prehistoric subsistence pat- terns and site economics in the Moche Valley, Peru. Ph.D. dissertation. University of Texas, Austin, 492 p. Pozorski, S., and T. Pozorski. 1979. Alto Salaverry: a Peruvian coastal preceramic site. Annals of Carnegie Museum 48:337–375. Pozorski, S., and T. Pozorski. 2003. Paleoenvi- ronment at Almejas: early exploitation of estuarine fauna on the north coast of Peru. 172 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) Pp. 52–70 in J. Haas and M. O. Dillon, eds., El Niño in Peru: Biology and Culture over 10,000 years: Papers from the VIII Annual A. Watson Armour III Spring Symposium, May 28-29, 1999, Chicago. Fieldiana, Botany, new series 43. Field Museum of Natural History, Chicago. Pozorski, T., and S. Pozorski. 1984. Almejas: early exploitation of warm-water shellfish on the north central coast of Peru. Final report sub- mitted to the National Science Foundation, grant BNS 82-03452. University of Texas, Pan-American, Edinburg, Texas. Quitmyer, I. R., and E. J. Reitz. 2006. Marine tro- phic levels targeted between AD 300 and 1500 on the Georgia coast, USA. Journal of Archaeological Science 33:806–822. Rademaker, K., G. R. M. Bromley, and D. H. Sand- weiss. 2013. Peru archaeological radiocarbon database, 13,000–7000 14C B.P. Quaternary International 301:34–45. Randall, J. 1968. Caribbean Reef Fishes. THF Pub- lications, Jersey City, 318 p. Reimer, P. J. E. Bard, A. Bayliss, J. W. Beck, P. G Blackwell, C. Bronk Ramsey, C. E Buck, H. Cheng, R. L. Edwards, M. Friedrich, P. M Grootes, T. P Guilderson, H. Haflidason, I. Hajdas, C. Hatté, T. J. Heaton, D. L. Hoff- mann, A. G. Hogg, K. A. Hughen, K. F. Kai- ser, B. Kromer, S. W Manning, M. Niu, R. W Reimer, D. A. Richards, E. M. Scott, J. R. Southon, R. A. Staff, C. S. M. Turney, and J. van der Plicht. 2013. IntCal13 and Marine13 radiocarbon age calibration curves 0–50,000 years cal BP. Radiocarbon 55:1869–1887. Reitz, E. J. 1988a. Faunal remains from Paloma, an Archaic site in Peru. American Anthropolo- gist 90:310–322. Reitz, E. J. 1988b. Preceramic animal use on the central coast. Pp. 31–55 in E. S. Wing and J. C. Wheeler, eds. Economic Prehistory of the Central Andes. BAR International Series 427, Oxford. Reitz, E. J. 1995. Environmental change at Alme- jas, Peru. Ms. on file, Zooarchaeology Labo- ratory, Georgia Museum of Natural History, University of Georgia, Athens. Reitz, E. J. 1999. Vertebrate remains from Sitio Siches (PV 7-19), Peru. Ms. on file, Zooarchae- ology Laboratory, Georgia Museum of Natural History, University of Georgia, Athens. Reitz, E. J. 2001. Fishing in Peru between 10000 and 3750 BP. International Journal of Osteo- archaeology 11:163–171. Reitz, E. J. 2002. The use of archaeofaunal data in fish management. Pp. 19–33 in R. C. G. M. Lauwerier and I. Plug, eds. The Future from the Past. Oxbow Books, Oxford. Reitz, E. J. 2003. Resource use through time at Paloma, Peru. Bulletin of the Florida Museum of Natural History 44(1):65–80. Reitz, E. J. 2004. “Fishing down the food web”: a case study from St. Augustine, Florida, USA. American Antiquity 69:63–83. Reitz, E. J. 2014. Continuity and resilience in the central Georgia Bight (USA) fishery between 2760 BC and AD 1580. Journal of Archaeo- logical Science 41:716–731. Reitz, E. J., S. D. deFrance, D. H. Sandweiss, and H. E. McInnis. 2015. Flexibility in southern Peru coastal economies: a vertebrate perspec- tive on the Terminal Pleistocene/Holocene transition. The Journal of Island and Coastal Archaeology 10:155–183. Reitz, E. J., H. E. McInnis, D. H. Sandweiss, and S. D. deFrance. 2016. Terminal Pleistocene and Early Holocene fishing strategies at Quebrada Jaguay and the Ring Site, southern Perú. Journal of Archaeological Science: Reports 8:447–453. Reitz, E. J., H. E. McInnis, D. H. Sandweiss, and S. D. deFrance. 2017. Variations in human adaptations during the Terminal Pleistocene and Early Holocene at Quebrada Jaguay (QJ- 280) and the Ring Site, southern Perú. The Journal of Island and Coastal Archaeology 12:224–254. Reitz, E. J., and M. A. Masucci. 2004. Guangala Fishers and Farmers: A Case Study of Ani- mal Use at El Azúcar, Southwestern Ecua- dor; Pescadores y Agricultores Guangala: Un Estudio de Caso de Uso Animal en El Azúcar, Suroeste de Ecuador. University of Pittsburgh Memoirs in Latin American Archaeology 14, REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 173 Pittsburgh, PA, 184 p. Reitz, E. J., and D. H. Sandweiss. 2001. Environ- mental change at Ostra Base Camp, a Peru- vian Pre-ceramic site. Journal of Archaeolog- ical Science 28:1085–1100. Reitz, E. J., and M. Shackley. 2012. Environmen- tal Archaeology. Springer Science+Business Media, New York, 516 p. Reitz, E. J., and E. S. Wing. 2008. Zooarchaeology, Second Edition. Cambridge University Press, Cambridge, 533 p. Richardson, J. B. III. 1973. The Preceramic sequence and the Pleistocene and Post- Pleistocene climate of northwest Peru. Pp. 199–211 in D. Lathrap and J. Douglas, eds. Variations in Anthropology: Essays in Honor of John C. McGregor. Illinois Archaeological Survey, Urbana. Richardson, J. B. III. 1978. Early man on the Peru- vian north coast, early maritime exploitation and Pleistocene and Holocene environment. Pp. 274–289 in A. L. Bryan, ed. Early Man in America from a Circum-Pacific Perspec- tive. Occasional Papers No. 1, Department of Anthropology, University of Alberta, Edmon- ton, Canada. Richardson, J. B. III, and D. H. Sandweiss. 2008. Climate change, El Niño, and the rise of com- plex society on the Peruvian coast during the Middle Holocene. Pp. 59–75 in D. H. Sand- weiss and J. Quilter, eds. El Niño, Catastroph- ism, and Culture Change in Ancient America. Harvard University Press, Cambridge, MA. Richardson, J. B. III, D. H. Sandweiss, and E. Pare- des. 1996. Proyecto Arqueológico Siches, Informe, Mayo de 1996. Ms. on file, Division of Archaeology, The Carnegie Museum of Natural History, Pittsburgh, PA. Rodbell, D. T., G. O. Seltzer, D. M. Anderson, M. B. Abbott, D. B. Enfield, and J. H. Newman. 1999. An ~15,000-year record of El Niño- driven alluviation in southwestern Ecuador. Science 283:516–520. Rollins, H. B., J. B. Richardson III, and D. H. Sandweiss. 1986. The birth of El Niño: Geo- archaeological evidence and implications. Geoarchaeology 1:3–16. Rollins, H. B., D. H. Sandweiss, U. Brand, and J. C. Rollins. 1987. Growth increment and stable isotope analysis of marine bivalves: implica- tions for the geoarchaeological record of El Niño: Geoarchaeology 2:181–197. Rollins, H. B., D. H. Sandweiss, and J. C. Roll- ins. 1990. Mollusks and coastal archaeology; A review. Pp. 467–478 in N.P. Lasca and J.D. Donahue, eds. Decade of North Ameri- can Geology, Centennial Special Volume 4: Archaeological Geology of North America. Geological Society of America, Boulder. Sánchez Mosquera, A. 1993. Primer gran paso de nuestra historia marítíma. Revista del Insti- tuto de Historia Maritima VIII (15):9–42. Sandweiss, D. H. 1992. The archaeology of Chin- cha fishermen: specialization and status in Inka Peru. Bulletin of the Carnegie Museum of Natural History 29:1–162. Sandweiss, D. H. 1996. Mid-Holocene cultural interaction on the north coast of Peru and Ecuador. Latin American Antiquity 7:41–50. Sandweiss, D. H. 2003. Terminal Pleistocene through Mid-Holocene archaeological sites as paleoclimatic archives for the Peruvian coast. Palaeogeography, Palaeoclimatology Palaeo- ecology 194:23–40. Sandweiss, D. H. 2009. Early fishing and inland monuments: challenging the maritime foun- dations of Andean civilization? Pp. 39–54 in J. Marcus, C. Stanish, and R. Williams, eds. Andean Civilization: Papers in Honor of Michael E. Moseley. Cotsen Institute of Archaeology, Los Angeles. Sandweiss, D. H. 2014. Early coastal South Amer- ica. Pp. 1048–1064 in C. Renfrew and P. Bahn, eds. The Cambridge World Prehistory. University of Cambridge Press, Cambridge. Sandweiss, D. H., K. A. Maasch, C. F. T. Andrus. E. J. Reitz, J. B. Richardson III, M. Riedinger- Whitmore, and H. B. Rollins. 2007. Mid- Holocene climate and culture change in coastal Peru. Pp. 25–50 in D. G. Anderson, K. A. Maasch, and D. H. Sandweiss, eds. Climate Change and Cultural Dynamics: A Global Perspective on Mid-Holocene Transi- tions. Academic Press, New York. 174 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) Sandweiss, D. H., K. A. Maasch, and D. G. Ander- son. 1999. Transitions in the Mid-Holocene. Science 283:499–500. Sandweiss, D. H., K. A. Maasch, F. Chai, C. F. T. Andrus, and E. J. Reitz. 2004. Geoarchaeo- logical evidence for multidecadal natural cli- matic variability and ancient Peruvian fisher- ies. Quaternary Research 61:330–334. Sandweiss, D. H., H. McInnis, R. L. Burger, A. Cano, B. Ojeda, R. Paredes, M. Sandweiss, and M. Glascock. 1998. Quebrada Jaguay: Early maritime adaptations in South America. Science 281:1830–1832. Sandweiss, D. H., and K. M. Rademaker. 2013. El poblamiento del sur peruano: costa y sierra. Boletín de Arqueología PUCP 15:275–293. Sandweiss, D. H., J. B. Richardson III, E. J. Reitz, J. T. Hsu, and R. A. Feldman. 1989. Early maritime adaptations in the Andes: prelimi- nary studies at the Ring Site, Peru. Pp. 5–84 in D. S. Rice, C. Stanish and P. R. Scarr, eds. Ecology, Settlement and History in the Osmore Drainage, Peru. BAR International Series 545i, Oxford. Sandweiss, D. H., J. B. Richardson III, E. J. Reitz, H. B. Rollins, and K. A. Maasch. 1996. Geo- archaeological evidence from Peru for a 5000 Years B.P. onset of El Niño. Science 273:1531–1533. Sandweiss, D. H., J. B. Richardson III, E. J. Reitz, H. B. Rollins, and K. A. Maasch. 1997. Deter- mining the early history of El Niño, response. Science 276:965. Sandweiss, D. H., H. B. Rollins, and J. B. Rich- ardson III. 1983. Landscape alteration and prehistoric human occupation on the north coast of Peru. Annals of Carnegie Museum 52:277–298. Santander, H. 1980. The Peru Current system. 2: biological aspects. Pp. 217–227 in Proceed- ings of the Workshop on the Phenomenon Known as ‘El Niño’. UNESCO, Paris. Scartascini, F. L. 2017. The role of ancient fishing on the desert coast of Patagonia, Argentina. The Journal of Island and Coastal Archaeol- ogy 12:115–132. Schwarz, F. A., and J. S. Raymond. 1996. Forma- tive settlement patterns in the Valdivia Valley, SW coastal Ecuador. Journal of Field Archae- ology 23:205-224. Scheib, C.L., et al. 2018. Ancient human parallel lineages within North America contributed to a coastal expansion. Science 360:1024–1027. Schmid, E. 1972. Atlas of Animal Bones for Pre- historians, Archaeologists, and Quaternary Geologists. Elsevier, Amsterdam, 159 p. Schweigger, E. 1964. El Litoral Peruano. Univer- sidad Nacional “Frederico Villarreal”, Lima, Peru, 414 p. Shannon, C. E., and W. Weaver. 1949. The Math- ematical Theory of Communication. Univer- sity of Illinois Press, Urbana, 117 p. Sheldon, A. L. 1969. Equitability indices: depen- dence on the species count. Ecology 50:466– 467. Silver, I. A. 1963. The ageing of domestic animals. Pp. 250–268 in D. Brothwell and E. Higgs, eds. Science in Archaeology. Praeger, New York. Stothert, K. E. 1988. La Prehistoria Temprana de la Peninsula de Santa Elena, Ecuador: Cul- tura Las Vegas. Museos del Banco Central del Ecuador, Guayaquil: Miscelánea Antrop- ológica Ecuatoriana Serie Monográfica 10, 271 p. Stuiver, M., and P. J. Reimer, 1993. Extended 14C data base and revised CALIB 3.0 14C age cal- ibration program. Pp. 215–230 in M. Stuiver, M., A. Long, A. and R. S. Kra, eds. Calibra- tion 1993. Radiocarbon 35: 215–230. Watson, J. P. N. 1978. The interpretation of epiphy- seal fusion data. Pp. 97–102 in D. R. Broth- well, J. D. Thomas, and J. Clutton-Brock, eds. Research Problems in Zooarchaeology. Uni- versity of London Institute of Archaeology Occasional Publications No. 3. Webb, S. C., R. E. M., Hedges, and R. Robinson. 1998. The seaweed fly Thoracochaeta zos- terae (Hal.) (Diptera: Sphaerocidae) in inland archaeological contexts: δ13C and δ15N solves the puzzle. Journal of Archaeological Science 25:1253–1257. White, T. E. 1953. Method of calculating the dietary percentages of various food animals utilized REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 175 by aboriginal peoples. American Antiquity 19:396–398. Wing, E. S. 1986. Methods employed in the identi- fication and analysis of the vertebrate remains associated with sites of the Paiján Culture. Ms. on file, Environmental Archaeology Lab- oratory, Florida Museum of Natural History, University of Florida, Gainesville. Wing, E. S., and E. J. Reitz. 1982a. Pisces, Rep- tilia, Aves, Mammalia. Pp. 191–200 in D. Bonavia, ed. Precerámico Peruano: Los Gavilanes, Mar, Desierto y Oásis en la His- toria del Hombre. Editorial Ausonia-Talleres Gráficos, Lima, Peru. Wing, E. S., and E. J. Reitz. 1982b. Prehistoric fishing economies of the Caribbean. Journal of New World Archaeology 5(2):13–33. APPENDIX 1 VERTEBRATE ECOLOGICAL GROUPS This review is restricted to vertebrates found in archaeological assemblages pertinent to the Sitio Siches study (Table 3). Scientific names assigned during the original study are used whenever there is uncertainty about the current valid taxonomic affiliation of the animal. When a name which is no longer valid is used here, it is because that was the attribution conferred at the time of the original zoo- archaeological study. Ecological group classifications rely heavily on Chirichigno’s (1982:xvi) division of the central and southeastern Pacific coast of the Americas into two extensive fishing areas and several subareas. She designates the Panamanian province as Area 77 and the Peru-Chilean province as Area 87 (Fig. 1). Of specific interest are Subareas 77C, 87A, and 87B. Subarea 77C is primarily the Ecuador- ian coast extending down to Paita, Perú (1° 38ʹ N to 5° S). Subarea 87A extends from Paita to Punta Aguja at 6° S. Subarea 87B includes the rest of the Peruvian coast down to 18° 20ʹ S and Subarea 87C includes most of the Chilean coast. MAMMALS, BIRDS, AND REPTILES Foxes (Lycalopex cf. sechurae), passerine birds (Corvidae, Mimidae), cañan lizards (Dicro- don sp.), and non-poisonous snakes (Colubridae) are among the terrestrial forms. The range of desert foxes includes southwestern Ecuador and coastal northwestern Perú (Eisenberg and Redford, 1999:284–285). They are found in areas with high daytime temperatures (28° C) and annual rainfall below 300 mm. Desert foxes are common compo- nents of many archaeological sites in this region. Foxes in the zooarchaeological studies reviewed here originally were attributed to Dusicyon sp. or D. cf. sechurae by several different researchers, all of whom used the reference materials in the FLMNH Environmental Archaeology Laboratory. Dusicyon was subsequently changed to Pseudalo- pex and recent work distinguishes two species in the genus now known as Lycalopex (L. sechurae and L. culpaeus). This leaves the original attribu- tion uncertain, though the specimens in every case are clearly foxes and their presence in assemblages from the coastal zone of northwestern Perú and southwestern Ecuador suggests they are L. sechu- rae (Eisenberg and Redford, 1999:285). Cañan lizards (Dicrodon holmbergi) are com- mon along the dry coast of Ecuador and northern Perú (Carrillo de Espinoza, 1970:29). Males may reach 51 cm in length, with females being smaller. These lizards are active during the day when they dig extensive burrows (Holmberg, 1957; Carrillo de Espinoza, 1970:29). These herbivorous lizards prefer algarroba fruits. They are highly prized for their nutritive qualities and as an aphrodisiac, a potential thought to derive from the algarroba (Holmberg, 1957). Holmberg (1957) describes the capture of Dicrodon holmbergi by Peruvians of the north coast using small nets and corrals. Marine mammals (Pinnipedia, Otariidae, Otaria flavescens [formerly O. byronia], Mysticeti, Delphinidae, Lontra felina) are largely pelagic and usually are classified as members of the mixed- water ecological group. Eared (Otariidae) and earless (Phocidae) seals were classified as Pinni- pedia at one time and this is the terminology used during the original study of the zooarchaeological assemblages when it was not possible to distin- guish between the two families. All of the marine carnivores are commonly associated with cool 176 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) waters, they also frequent coasts with warm waters and estuaries depending upon the seasonal habits of each species. Southern sealions (Otaria flaves- cens) usually are found from Perú south, but a dead individual was found in the Galapagos Islands, suggesting they enter warmer waters occasionally (Nowak and Paradiso, 1983:1100). Whales and dolphins generally are associated with a variety of water temperatures. For example, all five species of the spinner or striped dolphin genus Stenella are found in deep, clear, offshore waters in primarily, but not exclusively, tropical and subtropical seas (Nowak and Paradiso, 1983:878). Marine otters (Lontra felina) are found along the Pacific coast from northern Perú to Tierra del Fuego (Nowak and Paradiso, 1983:1015) and are considered cool- water animals. Most oceanic and shore birds are associ- ated with a variety of coastal and oceanic settings, though they may be taken from terrestrial settings, too (Koepcke, 1970). Some birds are typical of the Peruvian coast or the open ocean. These include Inca terns (Larosterna inca), penguins (Sphenis- cus humboldti), giant fulmars (Macronectes gigan- teus), diving petrels (Pelecanoides garnotii), boo- bies (Sula spp.), and buzzard eagles (Geranoaetus melanoleucas). These are considered members of the cool-water group. Cormorants (Phalacroco- rax spp.) and pelicans (Pelecanus spp.) are com- mon coastal Peruvian birds and members of the guano bird complex. They, as well as rails (Ral- lidae), sandpipers (Scolopacidae), gulls (Laridae), and herons (Ardea cocoi), also are common in tropical settings, however. Shearwaters (Puffinus cf. griseus) are seasonally common in Perú. These birds are considered members of the mixed-water ecological group. All of these birds could be taken from either aquatic or terrestrial locations. Cheloniidae are known from the Pacific coast. Green sea turtles (Chelonia mydas) are pelagic in Areas 77 and 87, though they are infrequent in Sub- area 87C (Chirichigno, 1982:493–494). Hawks- bills (Eretmochelys imbricata) and Pacific ridleys (Lepidochelys olivacea) are found only in Area 77 (Chirichigno, 1982:493–494). According to Sch- weigger (1964:304), C. mydas is perhaps the most frequent of the sea turtles along Perú’s coast, where it is fished in June and July. Schweigger reports L. olivacea only from the Gulf of Guayaquil. Sea turtles are classified as mixed-water vertebrates and might be considered terrestrial if they primar- ily were targeted on beaches, though Schweigger (1964:304) was not sure that sea turtles did nest on Peruvian beaches. FISHES Sharks (Carcharhinidae, Galeocerdo spp., Mustelus spp., Rhizoprionodon spp., Sphyrna spp.) and rays (Rajiformes, Dasyatidae, Myliobatidae, Rhinobatos spp.) inhabit a variety of marine habi- tats. Most are classified in the mixed-water ecologi- cal group. Although the eagle ray family (Mylioba- tidae) includes three species found exclusively in Area 77, two species are found in Area 87. The two species in Peruvian waters are the Chilean eagle ray (Myliobatis chilensis, Subareas 87B, C) and the Peruvian eagle ray (M. peruvianus, Subareas 87A, B, C). The Peruvian eagle ray prefers Subarea 87B. Eagle rays are pelagic and are found near the ocean floor (Chirichigno, 1982:354–355). One of the two eagle ray attributions in the assemblages reviewed was of the Peruvian eagle ray, which is classified as a cool-water form. Ladyfishes (Elops affinis) are found in Areas 77–87A, B over benthic platforms with mud or sand bottoms. They are particularly characteristic of estuaries (Chirichigno, 1982:2) and are classi- fied as members of the warm-water group. Although bonefishes (Albula vulpes) are found in Areas 77–87A, B over benthic platforms with mud or sand bottoms, they are particularly characteristic of shallow coastal waters, brackish estuaries, and warm bays (Chirichigno, 1982:1; Froese and Pauly, 2004). Bonefishes are more com- mon in Area 77 than in Area 87 and are not men- tioned by Schweigger (1964). They are classified as members of the warm-water ecological group. Eels (Anguilliformes, Muraenidae) are found in Areas 77–87, depending upon the specific attri- bution (Chirichigno, 1982:20–32). The most com- mon moray eel in Peruvian waters is the red moray (Gymnothorax wieneri); the white spotted moray REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 177 (Muraena albigutta or argus) prefers warmer waters (Schweigger, 1964:239). Members of both the order and the family are considered part of the mixed-water group. Anchovies (Engraulidae) are pelagic in coastal waters near the surface (Mann, 1950). As with herrings, members of this family are found in both tropical and temperate waters, including estuaries. Most anchovies are more typical of Area 77 than of Area 87, but Engraulis ringens (Sub- area 87B) is a classic member of the Peru Current community on which it depends (Chirichigno, 1982:10–20; Froese and Pauly, 2004). Schweigger (1964:238) argues E. ringens is not found north of Punta Aguja (ca. 6° S; see also, Froese and Pauly, 2004). It is not tolerant of warm or brackish waters. Elsewhere, however, members of this family may be the most common estuarine species. In one Geor- gia (USA) estuary, engraulids are the most abun- dant fish family (DEIS, 1978:D-452) and the bay anchovy (Anchoa mitchilli) is extremely common in bays and close inshore waters in the northern Gulf of Mexico (Hoese and Moore, 1977:137). A. curta, A. naso, and A. panamensis are found more commonly north of 6° S (Schweigger, 1964:238) in Subareas 77A, B (Chirichigno, 1982:12–15). For example, the longnose anchovy (A. nasus) is a wide-ranging, schooling species found in coastal waters and brackish bays in Subareas 77B, C–87A, B. Although anchovies may be found in warm-trop- ical/estuarine waters, sometimes abundantly, they are classified here as cool-water forms because of their traditional association with Peruvian zooar- chaeological assemblages and their abundance in fine-screened Peru-Chilean province faunal assem- blages recovered from sites south of Lima, Perú. In contrast, engraulids in Subarea 77C probably are not species associated with the Peru Current but are more likely to be species commonly found in lagoons and warm waters.3 Herrings (Clupeidae) are pelagic and may be found in coastal waters near the surface (Mann, 1950). Depending on the species, they are found in both warm and cool waters, including estuar- ies. Peruvian members of this family are highly characteristic of the upwellings of the Peru Cur- rent where they form an important part of the com- mercial fishery and food chain. Many are pelagic (Ethmidium maculatum, Etrumeus teres, Sardinops sagax) and are not common in shallow waters such as estuaries. Clupeids identified in sites located in Subarea 77C are more likely to be species com- monly found in estuaries and warm waters with a principal fishing zone of 77. This includes her- rings such as Harengula peruana, Ilisha furthii, Neoopisthopterus tropicus, and Opisthonema spp. (Chirichigno, 1982:2–10). One species in particu- lar, Sardinops sagax, follows warm El Niño waters southward (Chavez et al., 2003; Sandweiss et al., 2004). Although some herring species may be found in tropical or estuarine waters, they are clas- sified here as members of the cool-water ecological group because of the close association of Peruvian herring species with the Peru Current and their abundance in fine-screened Peruvian archaeologi- cal assemblages associated with the cooler waters of Subarea 87B. Sea catfishes (Ariidae) are found in shal- low, near-shore waters, often over muddy bottoms (Eschmeyer et al., 1984:96). Most are found in Area 77 (Chirichigno, 1982:34–41). Arius kes- sleri, A. multiradiatus, A. seemanni (A. jordani), A. troscheli, Bagre panamensis, B. pinnimacula- tus, and Galeichthys peruvianus are reported as far south as Area 87, though only the Peruvian catfish (G. peruvianus) is principally from Subarea 87A. According to Schweigger (1964:238–239), the Peruvian sea catfish is the most frequent of the sea catfishes in Perú and is found all along the coast of Perú where bottoms are muddy. It is found in coastal waters, but not brackish ones, south of 3° S (Froese and Pauly, 2004). Although sea catfishes are found in open waters, they are far more char- acteristic of lagoons and even are found in upper reaches of streams flowing into lagoons or in fresh water. They are classified here as members of the warm-water group, acknowledging that they are found elsewhere. Andrus et al. (2002) note that G. peruvianus tolerate a wide range of temperatures and do not migrate even in response to seasonal or ENSO-related sea surface temperature (SST) changes. 178 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) Although two species of hakes (Merluccius spp.) are found in Area 77, three are typical of Sub- areas 87A, B, C (Chirichigno, 1982:64–66). None of the hakes are common in lagoons and they are classified here as part of the cool-water fauna. Cusk-eels (Ophidiidae) range throughout Areas 77–87 (Chirichigno, 1982:56–63). Depend- ing upon the species, their primary fishing zone may be Area 77 through Subarea 87D. The pri- mary fishing zone for black cusk-eels (Genypterus maculatus) is Area 87B, C (Chirichigno, 1982:60). Black cusk-eels are classified in the cool-water group. Although mullets (Mugil spp.) are found from Area 77 through Area 87C, this is a cosmo- politan family with members ranging from coastal settings into brackish lagoons and fresh waters (Chirichigno, 1982:242–244). They are found in warm or temperate continental waters and where sandy bottoms are close to the shoreline (Sch- weigger, 1964:290). Roe mullets may be quite common in the surf and juveniles in lagoons. Mul- lets are reported from locations near Chiclayo and the River Zana (Schweigger, 1964:290). They are best caught with nets because their small mouths make taking a hook difficult. Fingerling mullets are taken in near-shore grassy locations with nets and scoops. They are considered warm-water fishes. Snooks (Centropomus spp.) range through- out Areas 77–87A and are quite common in lagoons (Chirichigno, 1982:133). They are considered warm-water fishes. Members of the sea bass (Serranidae) and grouper (Epinephelidae) families once were con- sidered members of the same large and com- plex family (Serranidae) with members found in many habitats, though most commonly in Area 77 (Chirichigno, 1982:137–152). The family has undergone considerable taxonomic revision, but the older taxonomy is maintained here to be con- sistent with the original zooarchaeological stud- ies. Groupers (Epinephelus spp.) are found over hard bottoms in Areas 77–87A, B (Chirichigno, 1982:142–144). Six species of Epinephelus range into Subareas 87A, B, but none have either Sub- areas 87A or 87B as their primary fishing zone. The range of sea perches (Hemanthias peruanus) extends from 77–87A, B, C with the principal fish- ing zone in Area 77. The sea bass Paralabrax spp. also is found over hard bottoms. Most are present only in Area 77, but P. callaensis (77B, C–87A), P. humeralis (77–87A, B, C), and P. furcifer (77– 87A) are found in Peruvian waters (Chirichigno, 1982:147–149). Cabrillas (P. humeralis) have Sub- areas 87A, B as a principal habitat and are not typi- cal of warm waters. Cabrilla typically are bottom feeders, especially in rocky locations (Schweigger, 1964:245, 247). Epinephelus and Hemanthias are classified as part of the warm-water ecological group and Paralabrax is classified as a member of the cool-water group. Jacks (Carangidae) are fast-swimming carni- vores found in shallow, near-shore waters, usually in schools. Very few carangids are common in Peru- vian waters, though Selene peruvianus, Trachinotus paitensis, and Trachurus murphyi are found there. The genus Caranx is limited to warm waters and several species are found in Area 77 (Chirichigno, 1982:159–172; Schweigger, 1964:250). Although some leatherjackets (Oligoplites spp.) and look- downs (Selene spp.) are found from Area 77 into Subarea 87A, their primary fishing zone is Area 77 (Chirichigno, 1982:166–168). In contrast, the pri- mary fishing zones for jurels (Trachurus murphyi) are Subareas 87B, C (Chirichigno, 1982:171). Caranx, Oligoplites, and Selene are classified as warm-water animals and Trachurus murphyi is classified as a cool-water form. The other jacks are classified as mixed-water because their primary fishing zones are so broad. Area 77 is the principal fishing zone for snappers (Lutjanidae, Lutjanus spp.), though five species range into Subareas 87A, B (Chirichigno, 1982:178–182). Because this particular attribu- tion is confined to an Ecuadorian zooarchaeologi- cal assemblage, the taxon is classified here in the warm-water ecological group. Mojarras (Gerreidae, Eucinostomus spp.) are found in Areas 77–87A. Their primary fishing zone is in Area 77 and they are classified here as mem- bers of the warm-water ecological group. Most grunts (Haemulidae) are found in Area REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 179 77 rather than in Area 87 (Chirichigno, 1982:183– 194). Sargos (Anisotremus spp.) are found over rocky or rock-sand bottoms in shallow waters where they eat invertebrates (Eschmeyer et al., 1984:217). A. scapularis is principally caught in Subareas 87B, C, though it is found from Areas 77C–87A, B, C. A. scapularis lives all along the coast from Paita in Perú to Taltal, Chile (Sch- weigger, 1964:257). Conodon, on the other hand, is typical of Area 77, though C. macrops may be found in Subarea 87A. Conodon is typically found in lagoons. According to Schweigger (1964:256), this fish has no commercial value in Perú. Haemu- lon also is more typical of Area 77, though H. stein- dachneri may be found in lagoons of Subarea 87A. Cabinza (Isacia conceptionis) is an omnivorous, schooling haemulid found in coastal waters near the surface very close to the shoreline in Subareas 87B, C. It is found all along the Peruvian coast and into Chile (Schweigger, 1964:256). Orthopristis also is found primarily in Area 77, though Ortho- pristis chalceus may be found in Subareas 87A, B. The primary fishing zone for Orthopristis is Area 77. Conodon, Haemulon, and Orthopristis, there- fore, are classified as warm-water forms. Anisotre- mus scapularis and Isacia conceptionis are consid- ered cool-water animals. Porgies (Calamus spp.) are members of the Sparidae family. They are found over benthic plat- forms in Areas 77–87A, B with a preference for Area 77 (Chirichigno, 1982:195). They are not mentioned by Schweigger (1964) and are consid- ered members of the warm-water ecological group. Drums (Sciaenidae) are members of the shal- low water, near-shore community (Eschmeyer et al., 1984:218) both in estuarine settings and as members of the Peru Current community. Many members of this family are particularly promi- nent in some early Peruvian zooarchaeological assemblages. The principal fishing zone for silver perch (Bairdiella ensifera) includes Areas 77–87A, though the fish is found in Subarea 87B (Chirich- igno, 1982:196–197). Schweigger (1964) does not mention Bairdiella in his survey of the Peruvian coast but it is listed in FishBase (Froese and Pauly, 2004) as a species that inhabits tropical lagoons and other brackish settings in Perú. Two mem- bers of the genus once known as Sciaena (now classified as Cilus gilberti and Corvina deliciosa) range from Area 77 through Area 87 but are more typical of Area 87 than of Area 77 (Chirichigno, 1982:216–218). They prefer coastal waters over sandy bottoms (Schweigger, 1964:261). It may be that corvinas (Cilus gilberti) prefer cooler waters and lornas (Corvina deliciosa) tolerate slightly warmer conditions (Chirichigno, 1982:217). Sea- trouts (Cynoscion) are wide-ranging and may be found from Area 77 through 87C, though they are more typical of Area 77. The range for weakfishes (Cynoscion analis) is 77C through 87A, B, C, but their principal fishing zone is Subarea 77C–87A (Schweigger, 1964:259; Chirichigno, 1982:198– 202). Cynoscion analis is common in coastal waters and young will enter brackish waters in lagoons and shallow bays (Froese and Pauly, 2004). Another species of weakfish (Cynoscion stolzmanni) is par- ticularly interesting because of its large size, reach- ing a maximum Total Length of 115 cm (Froese and Pauly, 2004). They are found in tropical brack- ish lagoons, as well as coastal waters (Froese and Pauly, 2004). The range of steeplined drums (Lari- mus acclivis) and shining drums (Larimus efful- gens) is 77–87A, with a principal fishing zone in 77C–87A (L. acclivis) and 77B (L. effulgens). Their range overlaps with that of Pacific drums (Larimus pacificus [gulosus]), which are found from 77B, C through 87A, B with a principal fishing zone in Area 77–87A. All three species are found over soft bottoms in coastal waters and brackish waters of lagoons and bays (Froese and Pauly, 2004). Lari- mus pacificus is considered an uncommon spe- cies today (Froese and Pauly, 2004). Members of the kingfish genus Menticirrhus are wide-ranging and may be found from 77A, B, C and 87A, B, C depending on the species. Only M. ophicephalus and M. paitensis have a primary fishing range as far south as 87A but the principal fishing zone for M. ophicephalus extends to 87B. The range of tallfin croakers (Micropogonias altipinnis) is from Area 77 into Subareas 87A, B, C, but the principal fish- ing zone is Subarea 77C (Chirichigno, 1982:198– 202). Two of the four Micropogonias species are 180 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 56(4) typical of lagoons in Area 77 and this genus is not mentioned by Schweigger (1964). FishBase (Fro- ese and Pauly, 2004) indicates that Micropogonias altipinnis is found in Perú along coastal sandy shores and in brackish lagoons. Two other genera of croaker also are found in Area 77. One of these, Odontoscion xanthops, is found only in Area 77. Members of the other genus, Ophioscion spp., are found primarily in Zone 77, though one species (O. sciera) is found in lagoons as far south as 87B (Chirichigno, 1982:211). Cocos (Paralonchurus spp.) are found along sandy coasts and in bays from Area 77 through Area 87, but most are typi- cal of Subareas 77C thru 87. The principal catch zone for P. peruanus is Subarea 87A (Schweigger, 1964:258; Chirichigno, 1982:213–214) and Fish- Base (Froese and Pauly, 2004) lists its range as 3° to 13° S. Star drums (Stellifer spp.) are wide-rang- ing drums found in Areas 77–87A, B, C. It is typi- cal, however, of lagoons and the primary fishing zone in Area 77, though S. minor is fished princi- pally in Subareas 87A, B (Chirichigno, 1982:221). Polla drums (Umbrina xanti) are found over sandy bottoms in Areas 77–87A, B and are fished primar- ily in 77–87A (Chirichigno, 1982:225). Although these drums are wide-ranging, some are more characteristic of warmer waters. Bairdiella spp., Cynoscion analis, Larimus spp., Micropogonias spp., Odontoscion spp., Ophi- oscion spp., and Umbrina xanti are classified here as warm-water forms. Cilus gilberti (formerly Sciaena gilberti), Corvina deliciosa (formerly S. deliciosa), and Paralonchurus spp. are considered cool-water animals. Cynoscion spp., Menticirrhus spp., and Stellifer spp. are considered mixed-water fishes. Morwongs (Cheilodactylus spp.) are consid- ered members of the cool-water community. They are found only in Area 87, with the primary fishing zone in Subareas 87B, C (Chirichigno, 1982:238– 239). Two other fishes are classified as members of the mixed-water ecological group, though a case could be made that they should be considered warm-water animals. Three species of sea chub (Kyphosus spp.) are mostly found in Area 77; how- ever, K. analogus ranges from Area 77 into Subarea 87B (Chirichigno, 1982:228–229). Spadefishes (Chaetodipterus zonatus) also range from Area 77 through Subarea 87B (Chirichigno, 1982:231). The primary fishing zone for both is Area 77 and they have only been identified in zooarchaeological assemblages from sites north of Lima. Members of the wrasse family (Labridae) commonly are found in warm, tropical waters asso- ciated with coral reefs (Randall, 1968); however, two genera are found in cool Peruvian waters. The genus Bodianus ranges from Areas 77–87 and is commonly fished throughout the area (Chirichigno, 1982:247–253). The genus Pimelometopon is most commonly fished in Subareas 87B, C (Chirich- igno, 1982:247–253). According to Schweigger (1964:265), Bodianus prefers the warmer northern waters and Pimelometopon is more wide ranging. Many members of this family are found in cool waters. Labridae is classified as a cool-water taxon. Parrotfishes (Scaridae) also are typical mem- bers of tropical coral reef communities (Randall, 1968). The species identified, pocochos (Nichol- sina denticulatus [formerly Xenoscarus denticu- lata]), range from Areas 77–87A, B, but are pri- marily fished in Area 77 (Chirichigno, 1982:254). Scarids are considered warm-water fishes. Several other fishes are classified as cool- water fishes. Rollizos (Pinguipes chilensis [occa- sionally Mugiloides chilensis]) and viejas (Labri- somus philippi) range through Subareas 87B, C (Chirichigno, 1982:256, 267). Mackerels (Euthyn- nus spp., Sarda spp., Scomber spp., Scomberomorus spp., Thunnus spp.) are pelagic fishes found in a variety of temperature conditions (Chirichigno, 1982:274–278). Some members of this family are uncommon in lagoons and are typical members of the Peru Current food chain. Other members char- acterize warm-water locations and are common in sites from later sites on the northern coast of Ecua- dor (Byrd, 1976:84; Lippi, 1983; Sánchez, 1993). Palmeras (Seriolella violacea [formerly Nepto- menus violacea]) range from Areas 77C through 87A, B, C (Chirichigno, 1982:290). Their primary fishing zone is Subarea 87B. Flounders (Bothidae) are carnivorous bottom REITZ ET AL.: Vertebrate remains from Sitio Siches, Perú 181 fishes common in shallow, near-shore waters. Most are found in Area 77, though several species are found in Area 87 as well (Schweigger, 1964:293; Chirichigno, 1982:294–306). The only flounders with Subareas 87A, B as a principal fishing zone are Etropus ectenes and Paralichthys adspersus, but because most identifications are to family, flounders are considered members of the mixed- water group. Two other fishes are typical of lagoons and are classified as warm-water species. Sleepers (Ele- otridae) are found only in Area 77, where they may be found in lagoons and even fresh water (Chirich- igno, 1982:270). Puffers (Spheroides annulatus) are found in Areas 77–87A, B but Area 77 is the principal fishing zone (Chirichigno, 1982:294– 306). According to Schweigger (1964:298), they only are found north of Paita. SUMMARY OF ECOLOGICAL GROUPS Certain vertebrates are considered typical of warm waters (Table 3). Members of the warm- water ecological group include bonefishes (Albula vulpes); sea catfishes (Ariidae); some groupers (e.g., Epinephelus spp.); grunts (e.g., Conodon spp., Haemulon spp., Orthopristis spp.); drums (e.g., Larimus spp., Umbrina cf. xanti); mullets (Mugil spp.); and puffers (Spheroides sp.). These animals are usually minor components of zooar- chaeological assemblages from sites within the Peru-Chilean province, though members of these families often are abundant elsewhere in the hemi- sphere in assemblages where tropical waters or lagoons are common (e.g., Wing and Reitz, 1982b; Cooke, 1992). Other vertebrates are typical of the cool waters associated with the Peru Current and often are very common in zooarchaeological assem- blages within the Peru-Chilean province. At many coastal sites the most abundant members of the cool-water ecological group are penguins (Sphe- niscus humboldti); boobies (Sula spp.); herrings (Clupeidae); anchovies (Engraulidae); some grou- pers (e.g., Paralabrax spp.), grunts (e.g., Anisotre- mus scapularis, Isacia conceptionis), drums (e.g., Paralonchurus peruanus, Corvina deliciosa, Cilus gilberti); and mackerels (Scombridae). In the pro- tocol used here, the dominance of some of these animals in zooarchaeological assemblages from sites located south of Lima and absence from sites north of Lima is considered a valid reason for including an animal in the cool-water group (Reitz, 2001; Reitz et al., 2015; Reitz et al., 2016; Reitz et al., 2017). Several marine vertebrates are typical of cool waters but also are found in warm settings, or the reverse. These variable species are classified as members of the mixed-water ecological group. These animals are tolerant of a wide range of envi- ronmental conditions and are widely distributed throughout the study area.