Obsidian and Ancient Manufactured Glasses

10: Polynesian Volcanic Glass: Uses, Sourcing, and Distribution

130 Abstract Volcanic glass has been used the world over for manufacturing a range of cutting tools used for utilitarian as well as ceremonial purposes. In Polynesia, aside from the mata’a of Easter Island, few formal tools are made and routine artifact forms are simple flakes and cores approximately 20 mm in size. Flakes rarely exhibit retouch or use-wear and few flakes in a typical assemblage appear used. However, volcanic glass flakes and debitage are common artifacts in early sites in Samoa and Tonga, often associated with the first pottery-bearing deposits. Hawaiian volcanic glass artifacts are found throughout the sequence. Obsidian is well attested in New Zealand archaeological sites, where it is found as flakes throughout all time periods and across all regions (Sheppard 2004; Sheppard et al. 2011). Despite the presence of a substantial source of large boulders and cobbles on Mayor Island, New Zealand, formal tools were not made, while large and small flakes, some with retouch, were common. This paper summarizes the distribution of volcanic glass throughout Polynesia except Easter Island and New Zealand and describes the prehistoric uses as inferred from use-wear studies and recent residue analyses. Volcanic glass artifacts and source material have also been used effectively in documenting prehistoric interaction, and these initial pursuits are outlined. 1. Definitions In this paper the term volcanic glass is a general label that indicates noncrystalline glasses where obsidian is a subclass of high-silica rhyolitic volcanic glass (WeislerandClague1998:104),andbasalticglassrefers to Hawaiian glasses that are more or less basaltic in composition and are correctly called trachylite, which is commonly found as chilled margins of dikes (Gary et al. 1972:721). Pitchstone, where the only Polynesian occurrence is on Pitcairn Island, forms by the deposition and consolidation of ash flows and has a waxy, dull, resinous luster (Bates and Jackson 1984:254, 386). 2. Geologic settings of volcanic glass Volcanic glass is found in a range of geological contexts, both primary (in situ) and secondary or detrital accumulations. Dikes are the most numerous sources of volcanic glass in Polynesia (Weisler 1990:fig. 4). These are intrusive bodies, 1 m to more than 10 m in width, that follow lines of weakness through existing strata. The margins of the dikes cool rapidly and solidify or quench before much crystal formation, thus forming a glassy groundmass . The thin selvages or borders are characteristically about 20 mm thick (figure 10.1). At present, there has been no recorded evidence of “quarrying” at these sources. However, the volcanic glass from these selvages is cubelike in shape and was commonly reduced by bipolar reduction. Volcanic glass can also form as a thin “crust” (<30 mm thick) on lava flows over an extensive area. These deposits are relatively uncommon and are only found on the leeward slopes of Hawaii Island where the lavas are geologically young and relatively unweathered (Williams 2004). Chapter 10 Polynesian Volcanic Glass Uses, Sourcing, and Distribution Marshall Weisler Polynesian Volcanic Glass | 131 Volcanic cones form a third geologic context in which rare deposits of volcanic glass have been found. The best-known source of trachytic volcanic glass is the 1.6-km-diameter pumice and volcanic glass cone at Pu‘u Wa‘awa‘a on Hawaii Island (Stearns and Macdonald 1946:205). Detrital or secondary sources of volcanic glass are characterized by pebbles and cobbles that accumulate in drainages downslope of in situ geologic sources such as dikes, plugs, or other outcrops. The cortex of detrital material is often pitted, rounded, and smoothed during transport. 3. Uses Since the raw material is quite small, there is no tradition of formal tool manufacture as in other parts of the world where obsidian is a preferred resource for manufacturing flake blades and a range of bifaces (e.g., arrow and spear points). Based on retouch, microscopic edge damage, step flake scars, and crushing , most researchers have assumed that small sharp flakes were used for light or fine cutting and scraping tasks (Kirch 1979; Kirch and Kelly 1975; Rosendahl 1972), although after a comprehensive study of 980 specimens, Schousboe et al. were “not prepared to make any definitive statements on the functions of the Waimea-Kawaihae [Hawaii Island] artifacts” (1983:368). Some have speculated that volcanic glass flakes may have been used for circumcision (Orliac 1997:206). Morwood (1974) analyzed obsidian flakes from several North Island, New Zealand, sites and suggested uses including woodworking, scraping, and cutting tasks. Microscopic edge damage, usually small flake scars along only one edge of the flake, was thought to result from cutting, scraping, or boring tasks when scaling and gutting fish, butchering dog and pig, scraping vegetables, and preparing fiber or bark. None of these explanations has been tied directly to empirical evidence of these inferred tasks. However, residue studies are a significant avenue for unraveling the uses of volcanic glass in prehistory. A recent study examining a total of 29 Hawaiian basaltic glass artifacts from Moloka‘i and pitchstone flakes from Henderson Island (Pitcairn Group) documented residues suggestive of plant processing and Figure 10.1. A volcanic glass dike at Waiahewahewa, west Moloka‘i, Hawaiian Islands. The thin ~20 mm thick selvedges of volcanic glass are located just above the lines of white dots added to the figure. The scale is 50 cm long. 132 | OBSIDIAN AND GLASS PROVENANCE shell incising (Weisler and Haslam 2005). Further analysis on an expanded sample from a broader range of sites should elucidate further the functions of this nondescript artifact class. Volcanic glass has also been used by archaeologists in Polynesia to determine stratigraphic integrity . Green (1974:148) assumed that volcanic glass artifacts found in the European levels of a Western Samoan house site were displaced from the older, pottery-bearing layer below. 4. Sourcing or Interaction Studies In briefly reviewing interaction studies, or the geochemical characterization of artifacts and geological source material, it is necessary to separate obsidian from volcanic glass since much work has been done with the former in New Zealand (e.g., Green 1962; Ward 1974a; Leach 1978; Leach et al. 1986; Leach and Warren 1981; Sheppard 2004; Sheppard et al. 2011), and the demonstrated efficacy of geochemical analysis with volcanic glass in Tonga and Samoa (Clark and Michlovic 1996; Clark and Wright 1995; Sheppard et al. 1989; Ward 1974b), the Pitcairn Group (Weisler 1995; Weisler and Clague 1998), and Hawai‘i (Weisler 1990; Weisler and Clague 1998) has only recently been pursued further by others (Burley et al. ms; McCoy et al. 2011). Ward (1974b) used X-ray fluorescence (XRF) analysisof23volcanicglassartifacts—14fromSamoa excavated by Green (1974) and 9 from Tonga—to determineifthegeologicsourcewasonTafahiIsland, Tonga. Only 4 elements were analyzed in this early study: zirconium (Zr), magnesium (Mn), rubidium (Rb), and strontium (Sr), and only Tongan artifacts were assigned to the Tafahi source. Sheppard et al. (1989) analyzed 7 volcanic glass artifacts from the same assemblage with a non-destructive Instrumental Neutron Activation Analysis (INAA) using 20 elements and XRF (analyzing 10 elements) to determine the source. Samples from a suspected archaeological source on Goat Island, Pago Pago harbor, Tutuila, were analyzed and compared to the artifacts suggesting a local Samoan provenance, but not Goat Island. Petrographic descriptions of the artifacts and geological material also demonstrated that Goat Island was not the source of the artifacts. These and other studies show that volcanic glass is often obtained from local sources. From a collection of 51 volcanic glass artifacts from 12 archaeological sites spanning the culturehistorical sequence from Tonga, Burley et al. (ms) used the new portable X-ray fluorescence technique (p-XRF) (Sheppard et al. 2010) to characterize sources and artifacts for examining inter-island voyaging and interaction between Tonga and Samoa. Using a suite of 18 elements, rubidium (Rb) and strontium (Sr) were effective in separating sources. Tonga and Samoa were considered an integrated homeland region, yet the sourcing results document little extra-archipelago interaction. The transfer of volcanic glass within Tonga occurred throughout the sequence along a 600-km length of islands. Despite its proven efficacy more than 20 years ago (Weisler 1990; Weisler and Clague 1998), geochemical characterization studies in Hawai‘i have been slow to take hold (McCoy et al. 2011). From the electron microprobe analysis of volcanic glass artifactsfromseveralhabitationsitesonMoloka ‘i,2were assigned to a source on O‘ahu and 1 to the Mauna Kea adze quarry area on Hawaii Island (Weisler and Clague 1998:122), documenting the first inter-island transport of this material. McCoy et al. (2011) analyzed 3329 volcanic glass artifacts from 87 archaeological sites throughout the Hawaiian Islands and the Pu‘u Wa‘awa‘a source using non-destructive XRF and focusing on selected mid-Z elements (Sr, Zr, and Y). The Pu‘u Wa‘awa‘a trachyte appears to be the only Hawaiian volcanic glass that is distinctive macroscopically (Olson 1983). Using primarily artifact size and distance from Pu‘u Wa‘awa‘a, McCoy et al. (2011) developed a cost-surface model in which it was possible to infer down-the-line exchange and remote direct access to the source (see also Weisler and Green 2001:436–440 for defining interaction spheres in East Polynesia based on travel time). The distinctive Pitcairn pitchstone can be sourced in hand specimen due to its color, luster , translucency, and opaqueness (Weisler and Clague 1998:105–109). However, energy-dispersive X-ray fluorescence was also used to characterize the Pitcairn pitchstone (Weisler 1994:94, 1997:165). From habitation contexts dating as early as AD 900 Polynesian Volcanic Glass | 133 on Henderson Island (located approximately 100 km north of Pitcairn Island), some 132 pieces of volcanic glass were recovered from five sites. This material was part of a comprehensive suite of imported materials used to reconstruct an interaction sphere operating for about six centuries that was ultimately tied to the Mangareva group, about 400 km west (Weisler 1997). One of the most thoughtful and interdisciplinary studies of volcanic glass underscores the benefit of archaeologists working with geochemists (see also Weisler 2008; Weisler and Clague 1998). Geochemist Wright analyzed 14 volcanic glass artifacts from the ‘Aoa site, Tutuila, using the scanning electron microscope energy-dispersive (EDAX) attachment for the standard suite of 10 elements from SiO2 to P2 O5 (Clark and Wright 1995). Rather than using statistical clustering programs to assign these artifacts to a source, a multistage, geochemically-informed procedure was used. Tutuila and ‘Upolu rocks can be separated from other Pacific volcanic glasses as these Samoan rocks have higher aluminum/calcium (Al2 O3 /CaO) ratios. Alkalies (Na2 O + K2 O) were plotted against silica (SiO2 ) to clearly demonstrate that Tutuila basalts are easily differentiated from those of ‘Upolu. Furthermore, low values for titanium (TiO2 ) are diagnostic for Tutuila (Clark and Wright 1995:256) because TiO2 was removed from the magmas by early crystallization of titanium magnetite (Natland 1980). In other words, there are clear petrogenetic reasons why Tutuila volcanic glass is geochemically distinctive. 5. Distribution From west to east, the archaeological evidence for volcanic glass in each of the Polynesian archipelagoes is briefly summarized (figure 10.2). Where little evidence is available, it is presented in more detail. Conversely, the data for Hawai‘i is summarized and ample references provided for additional details. It will be apparent from this review that actual “sources” are rare aside from the few basaltic glass dike sources on Moloka‘i and the Pu‘u Wa‘awa‘a trachyte on west Hawaii Island, the undescribed source on Tafahi (Tonga) and Futuna (Wallis and Futuna group), and secondary alluvial deposits in Samoa. 5.1 West Polynesia Wallis and Futuna. Situated between Fiji and Samoa in the extreme west of Polynesia, the two smaller of the three islands of the Wallis and Futuna group, Futuna (80 km2 ) and Alofi (35 km2 ), are deeply weathered volcanic islands. Kirch recovered 59 volcanic Figure 10.2. A map of West and East Polynesia with the locations of archipelagoes and islands mentioned in the text. The Andesite Line separates Oceanic Island Basalt (OIB) islands from those of continental origin, which have a greater diversity of rock types. 134 | OBSIDIAN AND GLASS PROVENANCE glass cores and flakes from excavations at site FU-19, about 5 km south from the source on Futuna (1981: fig. 1, 140). Similar to sites in Tonga and Samoa, many of the artifacts were from the pottery-bearing earliest layers. The Futuna volcanic glass artifacts were quite opaque with numerous microfractures filled with an orange-colored material (Kirch 1981:140). These artifacts probably occurred as small nodules with the average length of cores 14.2 mm. There was no evidence of retouch, and the bipolar technique was probably the reduction strategy. No details were provided on the source. Samoa. Savai‘i, ‘Upolu, and Tutuila, ranging in size from 1820 km2 to 124 km2 , are the three major islands of Samoa and probably reflect a hotspot origin (Duncan 1985). Dikes with volcanic glass borders and plugs with breccias containing volcanic glass exposed by erosion in steep drainages produced nodules that subsequently developed water-rounded cortex that may be the source of artifacts found in several habitation sites. Terrell (1969:168–170) identified one of the first occurrences of Samoan “obsidian ” restricted primarily to pottery-bearing contexts at a habitation mound (SU-VA-4), Western Samoa. Like many researchers who first encounter volcanic glass artifacts in Polynesia, Terrell was surprised by the small size of the artifacts. Of 74 specimens, 23 cores were 10–23 mm long (mean = 15 mm), and 50 flakes ranged from 5 to 17 mm (mean = 10 mm). Only one flake had obvious retouching and was probably used for “fine cutting” (Terrell 1969:169). Because at least one of the cores had water-rounded cortex, the source of the volcanic glass was assumed to be the nearby Falefa Valley where natural siliceous pebbles were found in the alluvium. The artifacts had a dull black luster with fresh surfaces that were glassy and a few had reddish streaks (Terrell 1969:169). Green recovered 22 flakes and cores, primarily from pottery-bearing layers, at a Western Samoan house site; he commented that more artifacts could have been recovered if the matrix was not muddy (1974:146–149). The average size of the flakes and cores was 6–23 mm (mean = 12 mm). The flakes were produced by the bipolar reduction of cores. A single, detrital local source of the volcanic glass was suggested by the water-rounded cortex and small artifact size. A small core (13.3 mm maximum length) was reported from the To‘aga site, Manu‘a Islands, American Samoa, that was “a reddish brown color, with black spots and banding” (Kirch 1993:165) and was likely to be an import. Excavations at a 3000-year-old habitation site at the mouth of ‘Aoa Valley, Tutuila Island, recovered 177 flakes (24 of which were used) and 98 cores from the deepest layers of the site (Clark and Michlovic 1996:table 2). Most of the artifacts were <10 mm in diameter, 31.5% had cortex, all were reddish brown to black in color, and many exhibited the characteristics of the bipolar reduction strategy. Secondary deposits, such as streambeds, were probable sources due to the partially weathered cortex (Clark and Michlovic 1996:161), or material may have originated from quenched magma exposed near the base of plugs as chunks in breccias on Tutuila (Clark and Wright 1995:256). Only 14 volcanic glass artifacts (5 cores and 9 flakes) have been reported from seven other Samoan sites (see Clark and Wright 1995:table 5). Tonga. Consisting of about 150 islands (the largest , Tongatapu, is 257 km2 ), this archipelago consists of three main subgroups with younger volcanic arc islands west of the older limestone-covered islands. Although volcanic “country” rock is available on many of the islands, only one fine-grained basalt source on the southernmost island of ‘Ata (Weisler 2004) and one volcanic glass source on Tafahi are currently known. Ten samples from Tafahi were analyzed by X-ray fluorescence to characterize the source and were compared to nine artifacts collected by Rogers (1974) on Niuatoputapu 7 km away. Ward reported (1974c:345) that the artifacts from Lapita contexts were most likely from Tafahi. The Tafahi source is at Tefitomaka and consists of nodules of volcanic glass, up to the size of a “chicken egg,” embedded in tuff (Dye 1988). Dye mentioned similar tuff outcrops behind Viapoa village, but the Tefitomaka source is more easily accessible. Dye (1988:table 3) collected 97 pieces of volcanic glass from five surface and excavated Tafahi sites; no further data were provided. Only two pieces of unworked volcanic glass were recovered from Poulson’s extensive excavations on Tongatapu, Tonga, both coming from pottery-bearing Polynesian Volcanic Glass | 135 (i.e., early) layers (Poulson 1987). This small amount is likely a reflection of the collection techniques used during the 1960s when the excavations were conducted. From excavations on Niuatoputapu, Kirch recovered 11,457 volcanic glass flakes and cores primarily associated with 3000 BP ceramic sites (Kirch 1988: 214–217). Less than 3 percent of the assemblage was analyzed for metrical attributes (length, width, thickness , weight, striking platform thickness and width) and morphological attributes (presence/absence of cortex, dorsal ridges, bulb of percussion, and termination type). The typical flake was <23 mm in length, generally lacking cortex, with one or two dorsal ridges; percussion bulbs were lacking or diffuse, and a range of flake terminations were present. Hard hammer and anvil were used for flake production. Only occasional flakes had secondary retouch (Kirch 1988:215). Burley recovered a varied assemblage of volcanic glassartifactsfromrecentexcavationsonVava‘u(personal communication 2004). Connaughton (2007: 209) recovered 4 “obsidian” flakes from later prehistoric contexts on Falevai, suggesting a local source (probably Tafahi). No further data were presented in his preliminary report. Niue. Situated approximately 400 km east of Tonga, 700 km south of Samoa, and 500 km west of the Cook Islands, isolated Niue (260 km2 ) is strategically situated for addressing issues of interaction and isolation. As a raised limestone (makatea) island, there are no naturally occurring volcanics. Although one polished basalt adze flake was recovered from recent archaeological excavations that might be from Samoa (Walter and Anderson 2002:77–78), no volcanic glass was recovered. 5.2 East Polynesia Cook Islands. Only 6 of the 15 Cook Islands are volcanic (Weisler and Sinton 1997:table 10.6), and geologists Wood and Hay lament that “volcanic rocks are almost all weathered to a considerable degree . . . and rock specimens can be obtained in a few places” (Wood and Hay 1970:27). Dikes have been identified on four islands (Walter and Sheppard 2001:table 29.1) without any mention of glassy selvages . Because of the eroded topography and highly weathered nature of most of the volcanic islands, no locally available volcanic glass artifacts have been reported from Cook Island archaeological sites despite extensive research on several of the islands (Allen 1992; Kirch et al. 1995; Walter 1998; Weisler et al. 1994). It is likely that any volcanic glass artifacts would be imported from outside the archipelago. This may be the case for an approximately 35-mmlong “obsidian” retouched flake reported from a surface context from site AIT.3, ‘Are Karioi, Aitutaki (Bellwood 1978:fig. 46c, 132). Society Islands. The Society Islands are a timeprogressive hotspot-evolved archipelago much like Hawai‘i. The large island of Tahiti has fine-grained basalt sources in Papeno‘o Valley, but no volcanic glass has been reported. Indeed, dikes are considered rare on Tahiti and have been observed at only four localities on Mo‘orea (Williams 1933:29, 73). The artifact class is rare in Society Islands sites with only 6 pieces recovered from recent excavations in ‘Opunohu Valley, Mo‘orea; these specimens ranged from about 11 mm to 28 mm long and all had cortex suggesting they were reduced from small nodules (Kahn 2005:366–367). Rappaport et al. (1967:199) reported 3 “obsidian” flakes from their coastal excavations at the Ta‘auroa site on Mo‘orea. Leach geochemically analyzed 2 of these reported flakes, and 1 of them (catalog number 85/2218) turned out to be “English” chert, perhaps a gun flint. Another specimen (85/2232) reported as “obsidian” had only 12.8% silica, which is extremely low for any volcanic glass and it was assumed to be basaltic glass (B.F. Leach, letter to the American Museum of Natural History, 22 July 1981), although this assumption seems unlikely as Hawaiian basaltic glass is approximately 48–52% silica (Weisler and Clague 1998: table 5.2). Historic artifacts were found throughout the unstratified deposits (Rappaport et al. 1967:200) and it may be that all of the “obsidian” specimens were historic chert artifacts. However, this does not discount the use of volcanic glass prior to European contact in the Society Islands, for Orliac reports 2 flakes of volcanic glass from the Putoa rockshelter in the Papeno‘o Valley, Tahiti (Orliac 1997:206; see also Kahn 2005) from presumably prehistoric contexts . From work thus far, prehistoric volcanic glass artifacts are extremely rare in the Society Islands. 136 | OBSIDIAN AND GLASS PROVENANCE Australs. Geologically, the five Austral Islands form the southern extent of the Southern Cook Islands–Australs volcanic chain that shares geochemical and geochronological affinities (Woodhead 1996:2, fig. 1). Archaeological field research in this area of French Polynesia has been limited to late prehistoric settlement pattern surveys on Ra‘ivavae where no suitable fine-grained basalt dikes were located (Edwards 2003:161) that could have contained volcanic glass selvages. Test excavations by the Norwegian Archaeological Expedition on Rapa did not report any volcanic glass artifacts (Smith 1965), nor did recent excavations there by Anderson recover any. Anderson did not locate any geological sources on the island (personal communication 2004). No volcanic glass artifacts were reported by Vérin (1969) from Rurutu. Recent excavations on Tubuai (Bollt, personal communication 2004) and on Rurutu (Bollt 2008) did not recover any volcanic glass artifacts, but fine-grained basalt debitage was common. Tuamotus and other atoll archipelagoes. The 75 low coral atolls and 1 raised makatea island of the Tuamotus are dispersed over approximately 1000 km between the Society Islands to the northwest and southeast to Mangareva. There are no naturally occurring volcanic rocks and no volcanic glass has been reported to date (Emory 1975). The same is true of the atoll archipelagoes of Tuvalu and Tokelau (West Polynesia) and the Line Islands of East Polynesia, although imported basalt adze material is known from many atolls throughout these groups (e.g., Best et al. 1984; Collerson and Weisler 2007). Mangareva. Mangareva is situated between the Tuamotus to the northwest and the Pitcairn Group to the southeast that together form the PitcairnMangareva lineament that is time-progressive with increasing island age toward the northwest (Duncan and McDougall 1976). Mangareva consists of 4 main volcanic islands and 22 smaller islands, coral islets, and sand cays. Dikes are found on all the volcanic islands except Aukena (Brousse and Guille 1974:fig. 1), but only a few personally observed have very thin (<5 mm) selvages of volcanic glass, unusable for flake production (Weisler and Green 2001:429). Island-wide surveys (Weisler 1996) have failed to locate any useable volcanic glass sources. With more than 100 m2 excavated throughout the group (Conte and Kirch 2004; Green and Weisler 2000), no volcanic glass artifacts have been recovered. Pitcairn Group. Made famous by the Bounty mutineers, the Pitcairn Group consists of four islands: Pitcairn, a Pleistocene volcano about 4.5 km2 in size and the only volcanic island of the group; Henderson, a raised limestone makatea island some 37 km2 in area; and the diminutive atolls of Oeno and Ducie. Hence, only Pitcairn has naturally occurring sources of fine-grained basalt and a highly siliceous volcanic glass or ignimbrite (also known as pitchstone). The main source of volcanic glass is located at the cliffs above Down Rope where cobbles and pebbles of angular black, glassy rock have eroded out of the tuff and accumulated at the beach below (Carter 1967:28, fig. 1; Weisler 1995:fig. 2). In thin section the glass is pale brown with rare crystals of anorthoclase, scattered microphenocrysts of fayalitic olivine, and augite together with numerous microlites of sodic feldspar (Carter 1967:28). It has characteristic black to gray patches in a glassy groundmass (Weisler and Clague 1998: 106, fig. 5.3). This volcanic glass is easily identifiable in hand specimen when compared to all the known glass sources in Polynesia. Flakes of this ignimbrite are found in archaeological sites across Pitcairn (Gathercole 1964), and flakes and cores were also transferred prehistorically to Henderson (see above) approximately 100 km distant (Weisler 1994:94; 1997). Beginning as early as AD 900, Pitcairn volcanic glass appears in Henderson habitation sites, where 132 artifacts weighing 110 g were found mostly at north coast sites (Weisler 1997:165, table 9.1, figs. 9.41 and 9.4m). Marquesas. Thus far only one volcanic glass flake has been recovered from a Marquesan archaeological site. No volcanic glass artifacts were reported from extensive excavations at Hanamiai, Tahuata (Rolett 1998) or other sites on Nuku Hiva (Allen and McAlister 2010; Rolett and Conte 1995; Suggs 1961). This is especially curious since the Marquesas are oceanic basalt islands with intrusive dikes. It is likely that the glassy selvages are too thin for stone-tool production, similar to the situation in Mangareva. The only known volcanic glass flake is approximately 15 mm long, shiny, dark brownish-black, and opaque (M.S. Allen, personal Polynesian Volcanic Glass | 137 communication, 2011). It was recovered from layer IV at Teavau‘ua, Anaho, Nuku Hiva and dated to AD 1200–1400 (Allen 2009:fig. 5). Hawai‘i. Unlike at other Polynesian islands, volcanic glass artifacts are very common in Hawaiian prehistoric habitation sites. The Hawaiian Islands are the longest linear, time-progressive chain in the Pacific, trending nearly 2500 km northwest from Hawaii Island positioned above the active hotspot (Clague and Hazlett 1989; Macdonald and Abbott 1970:fig. 1). The marked age gradient is reflected in the relatively pristine lava flows of west and southeast Hawaii Island to the widespread, deeply dissected valleys and gulches toward the north of the main islands as exemplified by O‘ahu and Kaua‘i. Hawaiian volcanic glass occurs in three geologic formations: rare trachyte associated with volcanic cones (Cross 1904), most commonly as a glassy border or selvage along dikes (Weisler 1990), and occasionally as surface chills on lava flows (Williams 2004). Because of the erosional regime of the archipelago , more dikes are found on the older islands of Kaua‘i and O‘ahu (Weisler 1990:table 1), but more useable, unweathered volcanic glass appears to be found on the younger islands. In this latter regard, only surface chill glass, of stone-tool quality, has been found on the youngest Hawaii Island (Williams 2004). The biggest source of trachytic volcanic glass is at Pu‘u Wa‘awa‘a, a pumice and volcanic glass cone some 1.6 km in diameter, located on the north slope of Hualalai volcano (Stearns and Macdonald 1946:205). The largest nodules of Hawaiian volcanic glass are known from here (Macdonald and Abbott 1970:109), yet only recently have volcanic glass artifacts in distant habitation sites been geochemically identified as coming from this source (McCoy et al. 2011). Clearly, more geochemical sourcing work in Hawai‘i is warranted from successful early studies (Weisler 1990; Weisler and Clague 1998). Like non-obsidian volcanic glass artifacts found throughout Polynesia, flakes and cores are usually no larger than 20 mm and few flakes exhibit any evidence of use-wear and/or retouch. Technological studies document a bipolar technique for reducing cores to flakes (Schousboe et al. 1983), and use-wear studies have not been very illuminating in determining solid evidence of specific uses, although residue studies are beginning to provide empirical evidence suggesting plant processing and shell working (Weisler and Haslam 2005). 6. Conclusions Volcanic glass has a limited distribution in Polynesia as summarized in table 10.1. Outside of Easter Island, the only obsidian deposits are located on the North Island of New Zealand, where four major source regions are known (Sheppard et al. 2011). Due to the quantity of volcanic glass-bearing dikes found across the Hawaiian Islands (Weisler 1990:fig. 1) and widespread chill glass sources associated with flows on Hawaii Island (Williams 2004), flakes and cores are very common in prehistoric archaeological sites from all time periods. Indeed, the Hawaiian Islands may have more volcanic glass artifacts than any Polynesian archipelago. It is interesting to note that volcanic glass artifacts in West Polynesian sites, however, are most often associated with the oldest pottery-bearing layers in Futuna, Alofi, Samoa, and Tonga. Perhaps these colonists, coming from Lapita settlements to the west, were trying to replicate their long tradition of obsidian use and trade so common in Near Oceania (see Kirch 1997). The inferior sources of volcanic glass in West Polynesia proved to be of limited utility and its use eventually declined or ended. Ironically, the diminutive island of Pitcairn (4.5 km2 ) has one of the largest volcanic glass sources in Polynesia, where nodules can exceed 100 mm, yet artifacts are not abundant in habitation sites. Due to the differing age and weathering regimes of East Polynesian archipelagoes, volcanic glass has a patchy distribution. Of the volcanic islands, no indigenous volcanic glass artifacts are known from the Cook Islands, Australs, Mangareva, and the Marquesas, while they are extremely rare in the Society Islands, where the few reported flakes may well be imports. In this regard, it will be interesting to see if the erosional regime of the main Hawaiian Islands is mirrored by the average amount of volcanic glass artifacts in archaeological sites; for example, we may expect to find more volcanic glass artifacts in Hawaii Island (0.6 million years old) sites as opposed to sites on Kaua‘i (5.25 million years old). 138 | OBSIDIAN AND GLASS PROVENANCE Sourcing studies are the most developed in New Zealand, with generally smaller successful applications in Hawai‘i, Pitcairn, Tonga, and Samoa. Technological studies are well developed, but have not been applied across Polynesia, where the most detailed studies have been in Hawai‘i (Schousboe et al. 1983), in Tonga (Kirch 1988), and most recently in New Zealand (Holdaway 2004). Of primary concern for future studies should be (1) the discovery, description , and geochemical characterization of new in situ and secondary sources of volcanic glass; (2) increased emphasis on geochemical analysis of artifacts from dated contexts for reconstructing patterns of interaction ; and (3) the more widespread documentation of edge damage and use-wear to infer tool function, including residue analysis for determining how small sharp flakes were utilized (Weisler and Haslam 2005). Although Polynesia does not garner worldwide attention for its volcanic glass artifacts, the subject merits continued study for enhancing our understanding of this nondescript material and its place in ancient Polynesian societies. Acknowledgments Aversionofthispaperwasdeliveredatthe“Obsidian Summit” conference at Rikkyo University, Tokyo, in 2004, and I thank the organizers for their efforts and memorable hospitality. Dave Burley provided a manuscript on his Tongan work, Melinda Suzy Allen provided unpublished data on a volcanic glass flake from Nuku Hiva (Marquesas), and Mark McCoy made an “in press” version of his 2011 paper available . The Australian Research Council is acknowledged for supporting this research. Thanks to all. References Allen, M.S., 1992. Dynamic Landscapes and Human Sub­ sistence: Archaeological Investigations on Aitutaki Island, Southern Cook Islands. University Microfilms (UMI), Ann Arbor, MI. Allen, M.S. 2009. Morphological variability and temporal patterning in Marquesan domestic architecture: Anaho Valley in regional context. Asian Perspectives 48, 342–382. Allen, M.S., McAlister, A.J., 2010. The Hakaea beach site, Marquesan colonisation, and models of East Polynesian settlement. Archaeology in Oceania 45, 54–65. Bates, R.L., Jackson, J.A. (Eds.), 1984. Dictionary of Geo­ logical Terms. Anchor Press, Garden City, NY. Bellwood, P.S., 1978. Archaeological Research in the Cook Islands. Pacific Anthropological Records 27. DepartmentofAnthropology ,B.P.BishopMuseum,Honolulu. Best, S., Sheppard, P., Green, R., Parker, R., 1984. Necromancing the stone: Archaeologists and adzes in Samoa. Journal of the Polynesian Society 101, 45–85. Bollt, R., 2008. Peva: The Archaeology of an Austral Island Settlement. Bishop Museum Bulletin in Anthropology 12. Bishop Museum Press, Honolulu. Brousse, R., Guille, G., 1974. Réseau de dykes aux iles Gambier. Cahiers du Pacifique 18, 253–264. Table 10.1 Volcanic glass artifacts in Polynesia Wallis & Cook Society Futuna Samoa Tonga Islands Islands Australs Cores x x x - - metric attributes x x x - - discrete attributes - - x - - Flakes x x x x x metric attributes x x x - x discrete attributes x x x - x Material Description hand specimen x x - - - petrographic - - - - - geochemical - x x - - Relative Quantity common v. common v. common rare rare absent Polynesian Volcanic Glass | 139 Burley, D.V., Sheppard, P.J., Simonin, M., ms. Tongan and Samoan volcanic glass: pXRF analysis and implications for constructs of ancestral Polynesian society. Carter, R.M., 1967. The Geology of Pitcairn Island, South Pacific Ocean. Bulletin 231. B.P. Bishop Museum, Honolulu. Clague, D.A., Hazlett, R.W., 1989. Geologic Field Guide to the Hawaiian Islands. 28th International Geologic Congress field trip guidebook T188/304. American Geophysical Union, Washington, D.C. Clark, J.T., Michlovic, M.G., 1996. An early settlement in the Polynesian homeland: Excavations at ‘Aoa Valley, Tutuila Island, American Samoa. Journal of Field Archaeology 23, 151–167. Clark, J.T., Wright, E., 1995. Volcanic glass in Samoa: A technological and geochemical study. Journal of the Polynesian Society 104, 239–266. Collerson, K.D., Weisler, M.I., 2007. Stone adze compositions and the extent of ancient Polynesian voyaging and trade. Science 317, 1907–1911. Connaughton, S.P., 2007. Can we dig it? Archaeology of Ancestral Polynesian society in Tonga: A first look from Falevai. In: Bedford, S., Sand, C., Connaughton, S.P. (Eds.), Oceanic Explorations: Lapita and Western Pacific Settlement. Terra Australis 26. Australian National University, Canberra, pp. 199–212. Conte, E., Kirch, P.V. (Eds.), 2004. Archaeological Investi­ gations in the Mangareva Islands (Gambier Archipel­ ago), French Polynesia. Contribution 62, Archaeological Research Facility, University of California, Berkeley. Cross, W., 1904. An occurrence of trachyte on the island of Hawaii. Journal of Geology 12, 510–523. Duncan, R.A., 1985. Radiometric ages from volcanic rocks along the New Hebrides–Samoa lineament. In: Brocher, T.M. (Ed.), Geological Investigations of the Northern Melanesian Borderland. CircumPacific Council for Energy and Mineral Resources, Houston, pp. 67–76. Duncan, R.A., McDougall, I., 1976. Linear volcanism in French Polynesia. J. Volcanol. Geothermal Research 1, 197–227. Dye, T.S., 1988. Appendix C. Archaeological investigations on Tafahi Island. In: Kirch, P.V. (Ed.), Niuato­ putapu: The Prehistory of a Polynesian Chiefdom. Thomas Burke Memorial, Washington State Museum Monograph no. 5. Burke Museum, Seattle, pp. 278–287. Edwards, E., 2003. Ra‘ivavae, Archaeological Survey of Ra‘ivavae, French Polynesia. Bearsville Press, Easter Island Foundation, Los Osos, CA. Emory, K.P.E., 1975. Material Culture of the Tuamotu Archipelago. Pacific Anthropological Records 22. Department of Anthropology, B.P. Bishop Museum, Honolulu. Gary, M., McAfee Jr., R., Wolf, C.L., (Eds.), 1972. Gloss­ ary of Geology. American Geological Institute, Washington , D.C. Gathercole, P., 1964. Preliminary Report on Archaeological Fieldwork on Pitcairn Island, January–March 1964. Report on file, Library, University of Otago, Dunedin, New Zealand. Green, R.C., 1962. Obsidian: Its application to archaeology . New Zealand Archaeological Association News­ letter 5, 8–16. Green, R.C., 1974. Excavation of the prehistoric occupations of SU-SA-3. In: Green R.C., Davidson, J.M. (Eds.), Archaeology in Western Samoa. Vol. 2. Bulletin of the Auckland Institute and Museum no. 6. Auckland Institute and Museum, Auckland, pp. 108–154. Green, R.C., Weisler, M.I., 2000. Mangarevan Archa­ eology: Interpretations Using New Data and 40 Year Old Excavations to Establish a Sequence from 1200 to 1900 AD. University of Otago Studies in Prehistoric Archaeology no. 19. Department of Anthropology, University of Otago, Dunedin, New Zealand. Holdaway, S.J., 2004. The Kohika obsidian artifacts: Technology and distribution. In: Irwin, G. (Ed.), Kohika, The Archaeology of a Late Maori Lake Village in the Ngati Awa Rohe, Bay of Plenty, New Zealand. Auckland University Press, Auckland, pp. 177–197. Kahn, J.G., 2005. Household and community organization in the Late Prehistoric Society Island chiefdoms Pitcairn Marquesas Hawaiian Mangareva Group Islands Islands - x - x - x - x - x - x - x x x - x x x - x - x - x x x - x - x - x - x absent v. common rare v. common 140 | OBSIDIAN AND GLASS PROVENANCE (French Polynesia). Ph.D. dissertation, University of California, Berkeley. Kirch,P.V.,1979.MarineExploitationinPrehistoricHawaii. Pacific Anthropological Records 29. Department of Anthropology, B.P. Bishop Museum, Honolulu. Kirch, P.V., 1981. Lapitoid settlements of Futuna and Alofi, Western Polynesia. Archaeology in Oceania 16, 127–143. Kirch, P.V., 1988. Niuatoputapu: The Prehistory of a Poly­ nesian Chiefdom. Thomas Burke Memorial Washington State Museum Monograph no. 5. Burke Museum, Seattle. Kirch, P.V., 1993. Non-ceramic portable artifacts from To’aga. In: Kirch, P.V., Hunt, T.L. (Eds.), The To’aga Site: Three Millennia of Polynesian Occupation in the Manu’a Islands, American Samoa. Contributions of the University of California Archaeological Research Facility no. 51. University of California, Berkeley, pp. 157–165. Kirch, P.V., 1997. The Lapita Peoples: Ancestors of the Oceanic World. Blackwell, Oxford, UK. Kirch, P.V., Kelly, M. (Eds.), 1975. Prehistory and Ecol­ ogy in a Windward Hawaiian Valley. Pacific Anthropological Records 24. Department of Anthropology, B.P. Bishop Museum, Honolulu. Kirch, P.V., Steadman, D.W., Butler, V.L., Hather, J., Weisler, M.I., 1995. Prehistory and human ecology in Eastern Polynesia: Excavations at the Tangatatau rockshelter, Mangaia, Cook Islands. Archaeology in Oceania 30, 47–65. Leach, B.F., 1978. Four centuries of community interaction and trade in Cook Strait, New Zealand. Man­ kind 11, 391–405. Leach, B.F., Warren, S., 1981. Neutron activation analysis of New Zealand and Oceanic obsidians: Towards a simple screening technique. In: Leach, B., Davidson, J. (Eds.), Archaeological Studies of Pacific Stone Resources. B.A.R. International Series 104. Archaeopress , Oxford, UK, pp. 151–166. Leach, B.F., Anderson, A., Sutton, D., Bird, R., Duerden, P., Clayton, E., 1986. Origin of prehistoric obsidian artefacts from the Chatham and Kermadec Islands. New Zealand Journal of Archaeology 8, 143–170. Macdonald, G.A., Abbott, A.T., 1970. Volcanoes in the Sea: The Geology of Hawaii. University of Hawaii Press, Honolulu. McCoy, M.D., Mills, P.R., Lundblad, S., Rieth, T., Kahn, J.G., Gard, R., 2011. A cost surface model of volcanic glass quarrying and exchange in Hawai‘i. Journal of Archaeological Science 38, 2547–2560. Morwood, M., 1974. A functional analysis of obsidian flakes from three archaeological sites on Great Barrier Island and one at Tokoroa. Records of the Auckland Institute Museum 11, 77–99. Natland, J.H., 1980. The progression of volcanism in the Samoan linear volcanic chain. American Journal of Science 280-A, 709–735. Olson, L., 1983. Hawaiian volcanic glass applied to dating and “sourcing”: Archaeological context. In: Clark, J.T., Kirch, P.V. (Eds.), Archaeological Investigations of the Mudland­Waimea­Kawaihae Road Corridor, Island of Hawai‘i: An Interdisciplinary Study of an Environmental Transect. Departmental Report Series 83-1. B.P. Bishop Museum, Honolulu, pp. 325–340. Orliac, M., 1997. Human occupation and environmental modifications in the Papeno’o Valley, Tahiti. In: Kirch, P.V., Hunt, T.L. (Eds.), Historical Ecology in the Pacific Islands, Prehistoric Environmental and Landscape Change. Yale University Press, New Haven, CT, pp. 200–229. Poulson, J., 1987. Early Tongan Prehistory. Terra Australis 12. Department of Prehistory, Research School of Pacific Studies, Australian National University, Canberra. Rappaport, R.A., Rappaport, A., Green, R.C., 1967. Descriptive and comparative analysis of portable artifacts recovered from coastal excavations. In: Green, R.C., Green, K., Rappaport, R.A., Rappaport, A., Davidson, J. (Eds.), Archaeology on the Island of Mo’orea, French Polynesia. Anthropological Papers of the American Museum of Natural History 51 (2). New York, pp. 184–200. Rogers, G., 1974. Archaeological discoveries on Niuatoputapu Island, Tonga. Journal of the Polynesian Soc­ iety 83, 308–348. Rolett, B.V., 1998. Hanamiai: Prehistoric Colonization and Culture Change in the Marquesas Islands (East Polynesia). Yale University Publications in Anthropology 81. Yale University, New Haven, CT. Rolett, B.V., Conte, E., 1995. Renewed investigation of the Ha’atuatua dune (Nukuhiva, Marquesas Islands): A key site in Polynesian prehistory. Journal of the Polynesian Society 104, 195–228. Rosendahl, P.H., 1972. Aboriginal agriculture and residence patterns in Upland Lapakahi, Island of Hawaii. Ph.D. dissertation, University of Hawai‘i, Honolulu. Schousboe, R., Riford, M., Kirch, P.V., 1983. Volcanic glass flaked stone artifacts. In: Clark, J.T., Kirch, P.V. (Eds.), Archaeological Investigations of the Mudlane­ Waimea­Kawaihae Road Corridor, Island of Hawai‘i: Polynesian Volcanic Glass | 141 An Interdisciplinary Study of an Environmental Tran­ sect. Departmental Report Series 83-1. B.P. Bishop Museum, Honolulu, pp. 348–370. Sheppard, P.J., 2004. Moving stones: Comments on the archaeology of spatial interaction in New Zealand. In: Furey, L., Holdaway, S. (Eds.), Change Through Time: 50 Years of New Zealand Archaeology. New Zealand Archaeological Association Monograph 26, pp. 147–168. Sheppard, P.J., Hancock, R.G.V., Pavlish, L.A., Parker, R., 1989. Samoan volcanic glass. Archaeology in Oceania 24, 70–74. Sheppard, P.J., Irwin, G.J., Lin, S.C., McCaffrey, C.P., 2011. Characterization of New Zealand obsidian using pXRF. Journal of Archaeological Science 38, 45–56. Sheppard, P.J., Trichereau, B., Milicich, C., 2010. Pacific obsidian sourcing by portable XRF. Archaeology in Oceania 45, 21–30. Smith, C.S., 1965. Report 5. Test excavations and surveys of miscellaneous sites on the island of Rapa Iti. In: Heyerdahl, T., Ferdon Jr., E.N. (Eds.), Reports of the Norwegian Archaeological Expedition to Easter Island and the East Pacific. Vol. 2: Miscellaneous Papers. Monographs of the School of American Research and the Kon-Tiki Museum no. 24, part 2. Rand McNally and Co., Chicago, pp. 77–87. Stearns, H.T., Macdonald, G.A., 1946. Geology and Ground­Water Resources of the Island of Hawaii. Bulletin 9. Division of Hydrography, Territory of Hawaii. Suggs, R.C., 1961. The Archaeology of Nuku Hiva, Mar­ quesas Islands, French Polynesia. Anthropological Papers of the American Museum of Natural History 49 (1). New York. Terrell, J., 1969. Excavations at SU-VA-4. In: Green, R.C., Davidson, J.M. (Eds.), Archaeology in Western Samoa. Vol. 1. Bulletin of the Auckland Institute and Museum no. 6. Auckland Institute and Museum, Auckland, pp. 158–176. Vérin, P., 1969. L’Ancienne Civilisation de Rurutu (îles Australes­Polynésie Française), La Periode Classique. Office de la Researche Scientifique et Technique Outre-Mer, Orstom, Paris. Walter, R., 1998. Anai’o: The Archaeology of a Fourteenth Century Polynesian Community in the Cook Islands. New Zealand Archaeological Association Monograph 22. Walter, R., Anderson, A., 2002. The Archaeology of Niue Island, West Polynesia. Bishop Museum Bulletin in Anthropology 10. Bishop Museum Press, Honolulu. Walter, R., Sheppard, P., 2001. Cook Island basalt sourcing : Current issues and directions. In: Jones, M., Sheppard, P. (Eds.), Australasian Connections and New Directions: Proceedings of the 7th Australasian ArchaeometryConference.ResearchinAnthropology and Linguistics no. 5. Department of Anthropology, University of Auckland, Auckland, pp. 383–399. Ward, G.K., 1974a. A systematic approach to the definition of sources of raw material. Archaeometry 16, 41–53. Ward, G.K., 1974b. An investigation of the source of the obsidian flakes from two Samoan sites. In: Green, R.C., Davidson, J.M. (Eds.), Archaeology in Western Samoa. Vol. 2. Bulletin of the Auckland Institute and Museum no. 6. Auckland Institute and Museum, Auckland, pp. 167–169. Ward, G.K., 1974c. Appendix 3: Source of obsidian from Niuatoputapu sites. Journal of the Polynesian Society 83, 345. Weisler, M.I., 1990. Sources and sourcing of volcanic glass in Hawai’i: Implications for exchange studies. Archaeology in Oceania 25, 16–23. Weisler, M.I., 1994. The settlement of marginal Polynesia: New evidence from Henderson Island. Journal of Field Archaeology 21, 83–102. Weisler, M.I., 1995. Henderson Island prehistory: ColonizationandextinctiononaremotePolynesianisland . Biology Journal of the Linnean Society 56, 377–404. Weisler, M.I., 1996. An archaeological survey of Mangareva : Implications for regional settlement models and interaction studies. Man and Culture in Oceania 12, 61–85. Weisler, M.I., 1997. Prehistoric long-distance interaction at the margins of Oceania. In: Weisler, M.I. (Ed.), Prehistoric Long­Distance Interaction in Oceania: An Interdisciplinary Approach. New Zealand Archaeological Association Monograph 21. New Zealand Archaeological Association, Dunedin, New Zealand, pp. 149–172. Weisler, M.I., 2004. A stone tool basalt source on ‘Ata, southern Tonga. New Zealand Journal of Archaeology 25 (2003), 113–120. Weisler, M.I., 2008. Sourcing studies are best done in collaboration with geochemists. Comment on Atholl Anderson’s “Traditionalism, Interaction and LongDistance Seafaring in Polynesia.” Journal of Island Coastal Archaeology 3, 265–267. Weisler, M.I., Clague, D.A., 1998. Characterization of archaeological volcanic glass from Oceania: The utility of three techniques. In: Shackley, M.S. (Ed.), 142 | OBSIDIAN AND GLASS PROVENANCE Archaeological Obsidian Studies, Method and Theory. Advances in Archaeological and Museum Science 3. Plenum Press, New York, pp. 103–128. Weisler, M.I., Green, R.C., 2001. Holistic approaches to interaction studies: A Polynesian example. In: Jones, M., Sheppard, P. (Eds.) Australasian Connections and New Directions: Proceedings of the 7th Australasian ArchaeometryConference.ResearchinAnthropology and Linguistics no. 5. Department of Anthropology, University of Auckland, pp. 413–453. Weisler, M.I., Haslam, M., 2005. Determining the function of Polynesian volcanic glass artifacts: Preliminary results of a residue study. Hawaiian Archaeology 10, 1–17. Weisler, M.I., Kirch, P.V., Endicott, J.M., 1994. The Mata’are basalt source: Implications for prehistoric interactions studies in the Cook Islands. Journal of the Polynesian Society 103, 203–216. Williams, H., 1933. Geology of Tahiti, Moorea, and Maiao. B.P. Bishop Museum Bulletin 105. Honolulu. Williams, S., 2004. The Pōhakuloa chill glass quarry complex, U.S. Army Pohakuloa training area, Hawai‘i Island. Hawaiian Archaeology 9, 105–118. Wood, B.L., Hay, R.F., 1970. Geology of the Cook Islands. New Zealand Geological Survey Bulletin n.s. 82. New Zealand Department of Scientific and Industrial Research, Wellington. Woodhead, J.D., 1996. Extreme HIMU in an oceanic setting: The geochemistry of Mangaia Island (Polynesia ), and temporal evolution of the Cook-Austral hotspot. Journal of Volcanology and Geothermal Research 72, 1–19. ...