1 Samples analyzed by WXRF as polished nodules. 

Collection Localities

The collection localities discussed here are not the result of a systematic survey to collect and record all the potential sources in the Jemez Mountains, but the result of an attempt to understand the secondary depositional regime of the sources flowing out from the Jemez Mountains into the surrounding stream systems, as noted above. The emphasis here was on understanding the secondary distribution of the major sources that appear in the archaeological record in the northern Southwest; El Rechuelos, Cerro Toledo Rhyolite, and Valle Grande.

El Rechuelos

El Rechuelos is mistakenly called "Polvadera Peak" obsidian in the archaeological vernacular (see also Glascock et al. 1999). Polvadera Peak, while a rhyolite dome, did not produce artifact quality obsidian. The obsidian artifacts that appear in the regional archaeological record are from El Rechuelos Rhyolite as properly noted by Baugh and Nelson (1987). Indeed, El Rechuelos obsidian is derived from a number of small domes north, west, and south of Polvadera Peak as noted by Baugh and Nelson (1987) and Wolfman (1994; see also Figures here). Collections here were made at two to three small coalesced domes near the head of Cañada del Ojitos and as secondary deposits in Cañada del Ojitos (collection locality 080999). The center of the domes is located at UTM 13S 0371131/3993999 north of Polavadera Peak on the Polvadera Peak quadrangle. The three domes are approximately 50 meters in diameter each and exhibit an ashy lava with rhyolite and aphyric obsidian nodules up to 15 cm in diameter, but dominated by nodules between 1 cm and 5 cm. Core fragments and primary and secondary flakes are common in the area.

Polvadera Peak in background with glass producing rhyolite domes on the left and small domes in foreground.

Small nodules under 10-15 mm are common in the alluvium throughout the area near Polvadera Peak. It is impossible to determine the precise origin of these nodules. Presumably they are remnants of various eruptive events associated with El Rechuelos Rhyolite. The samples analyzed, the results of which are presented in the table below are identical to the data presented in Baugh and Nelson (1987) and Glascock et al. (1999).

El Rechuelos obsidian is generally very prominent in northern New Mexico archaeological collections. Although it is not distributed geologically over a large area, it is one of the finest raw materials for tool production in the Jemez Mountains. Its high quality as a toolstone probably explains its desirability in prehistory. Cerro Toledo Rhyolite and Valle Grande Rhyolite, while present in large nodule sizes, often have devitrified spherulites in the glass, so more careful selection had to be made in prehistory. In nearly 500 nodules collected from the El Rechuelos area, few of the nodules exhibited spherulites or phenocrysts in the fabric. Additionally, El Rechuelos glass is megascopically distinctive from the other two major sources in the Jemez Mountains. It is uniformly granular in character, apparently from ash in the matrix. Cerro Toledo and Valle Grande glass is generally not granular and more vitreous.

Rabbit Mountain Ash Flow Tuffs and Cerro Toledo Rhyolite

Known in the vernacular as "Obsidian Ridge" obsidian, are derived from the Cerro Toledo Rhyolite eruptions, and following Baugh and Nelson (1987) and the geological literature are all classified as Cerro Toledo Rhyolite (Bailey et al. 1969; Gardner et al. 1986; Heiken et al. 1986; Self et al. 1986; Smith et al. 1970; Figure 3.13 here).

There were six pyroclastic eruptive events associated with the Cerro Toledo Rhyolite:

All tuff sequences from Toledo intracaldera activity are separated by epiclastic sedimentary rocks that represent periods of erosion and deposition in channels. All consist of rhyolitic tephra and most contain Plinian pumice falls and thin beds of very fine grained ash of phreatomagmatic origin. Most Toledo deposits are thickest in paleocanyons cut into lower Bandelier Tuff and older rocks [as with the Rabbit Mountain ash flow]. Some of the phreatomagmatic tephra flowed down canyons from the caldera as base surges (Heiken et al. 1986:1802).

Two major ash flows are relevant here. One derived from the Toledo embayment on the northeast side of the caldera is a 20 km wide band that trends to the northeast and is now highly eroded and interbedded in places with the earlier Puye Formation from around Guaje Mountain north to Santa Fe Forest Road 144. This area has eroded rapidly and obsidian from this tuff is now an integral part of the Rio Grande alluvium north of Santa Fe. The other major ash flow is derived from the Rabbit Mountain eruption and is comprised of a southeast trending 4 km wide and 7 km long "tuff blanket" interbedded with a rhyolite breccia three to six meters thick that contains abundant obsidian erupted as lapilli during the Rabbit Mountain ash flow (Heiken et al. 1986). All of this is still eroding into the southeast trending canyons toward the Rio Grande. The surge deposits immediately south of Rabbit Mountain contain abundant obsidian chemically identical to the samples from the ridges farther south and in the Rio Grande alluvium, as well as in sediments above the Puye Formation between Española and the Toledo Embayment. Heiken et al. NAA analysis of Rabbit Mountain lavas is very similar to those from this study (1986:1810; see data here).

Lower Cochiti Canyon from Forest Road 289 looking south.  Bandelier Tuff exposed on east canyon walls with Rabbit Mountain tuffs above eroding into Rio Grande.  Sandia Mountains in background.

While Obsidian Ridge has received all the "press" as the source of obsidian from Cerro Toledo Rhyolite on the southern edge of the caldera, the density of nodules and nodule sizes on ridges to the west is greater by a factor of two or more. All these ridges, of course, are remnants of the Rabbit Mountain ash flow and base surge, and the depth of canyons like Cochiti Canyon is a result of the loosely compacted tephra that comprises this plateau. At Locality 081199-1 (UTM 13S 0371337/3962354), nodules on the ridge top are up to 200 per m² with over half that number of cores and flakes. This density of nodules and artifacts forms a discontinuous distribution all the way to Rabbit Mountain. The discontinuity is probably due to cooling dynamics and/or subsequent colluviation. Where high density obsidian is exposed, prehistoric production and procurement is evident. At the base of Rabbit Mountain the density is about 1/8 that of Locality 081199-1, and south of this locality the density falls off rapidly. At Locality 081199-1 nodules range from pea gravel to 16 cm in diameter (Figures 3.14 and 3.15). Flake sizes suggest that 10 cm size nodules were typical in prehistory.

Locality 081199-1 south of Rabbit Mountain in the ash flow tuff. This locality has the highest density of artifact quality glass of the Rabbit Mountain ash flow area. The apparent black soil is actually all geological and archaeological glass; one of the highest densities of geological and archaeological obsidian in the Southwest.

Mix of high density geological obsidian and artifact cores and debitage (test knapping) at Locality 081199-1 south of Rabbit Mountain. Nodules » 200/m², cores and debitage » 100/m², some of the latter could be modern. 

Cerro Toledo Rhyolite obsidian both from the northern domes and Rabbit Mountain varies from an excellent aphyric translucent brown glass to glass with large devitrified spherulites that make knapping impossible. This character of the fabric is probably why there is so much test knapping at the sources – a need to determine the quality of the nodules before transport. While spherulitic fabric occurs in all the Jemez Mountain obsidian, it seems to be most common in the Cerro Toledo glass and may explain why Valle Grande obsidian occurs in sites a considerable distance from the caldera even though it is not secondarily distributed outside the caldera while Cerro Toledo obsidian is common throughout the Rio Grande alluvium. Indeed, in Folsom period contexts in the Albuquerque basin, only Valle Grande obsidian was selected for tool production even though Cerro Toledo obsidian is available almost on-site in areas such as West Mesa (LeTourneau et al. 1996). So, while Cerro Toledo Rhyolite obsidian is and was numerically superior in the Rio Grande Basin, it wasn’t necessarily the preferred raw material.

Valles Rhyolite (Cerro del Medio)

While the primary domes like Cerro del Medio of Valle Grande Rhyolite were not visited due to restrictions on entry to the caldera floor, surveys of the major stream systems radiating out from the caldera were examined for secondary deposits; San Antonio Creek and the East Jemez River, as well as the canyons eroding the outer edge of the caldera rim.

In 1956 two geology graduate students from the University of New Mexico published the first paper on archaeological obsidian in the American Southwest, a refractive index analysis (Boyer and Robinson 1956). In this examination of the Jemez Mountain sources, they noted that obsidian did not occur in the alluvium of San Antonio Creek where it crosses New Mexico State Highway 126, but did occur "in pieces as large as hen’s eggs, but the material is not plentiful and must be searched for with care" in the East Jemez River alluvium where it crosses State Highway 4 (Boyer and Robinson 1956:336). A return to the latter locality (Locality 102799-2) exhibited about the same scenario as that recorded 43 years earlier. The alluvium exhibits nodules up to 40 mm in diameter at a density up to 5/m², but generally much lower. Boyer and Robinson did find nodules up to 15.5 cm in diameter along the upper reaches of San Antonio Creek as shown in their plate reproduced here (Boyer and Robinson 1956:337; Figure below).

Valle Grande Rhyolite obsidian nodules photographed by Boyer and Robinson collected along San Antonio Creek in the caldera (1956:337).

My survey along San Antonio Creek from its junction with State Highway 126 for two miles upstream did not reveal any obsidian, as in the Boyer and Robinson study. It appears then that Valle Grande Rhyolite obsidian does not enter secondary contexts outside the caldera, at least in nodules of any size compared to Cerro Toledo Rhyolite.

Valle Grande Rhyolite obsidian exhibits a fabric that seems to be a combination of El Rechuelos and Cerro Toledo. Some of the glass has that granular texture of El Rechuelos and some has devitrified spherulites similar to Cerro Toledo, and much of it is aphric black glass. Flakes of Valle Grande obsidian can be indistinguishable from El Rechuelos or Cerro Toledo in hand sample. An elemental analysis of samples collected by Dan Wolfman from Cerro del Medio and the nodules in San Antonio Creek in this study are identical indicating that Cerro Toledo glass does not enter the East Jemez River system (see Appendix).

Canovas Canyon Rhyolite and Bear Springs Peak

Perlitic lava with remnant marekanites at Bear Springs Peak.  Large marekanite in the center of image is 47 mm in diameter.

The oldest (Tertiary, ca. 8.8-9.8 mya) artifact quality obsidian source in the Jemez Mountains is the Bear Springs Peak dome complex, part of the Canovas Canyon Rhyolite domes and shallow intrusions (Tcc) as reported by Smith et al. (1970), and more recently by (Kempter et. al. 2004; see also Gardner 1985; Figure 3 here).  These pre-caldera domes appear to be overlain by a rhyolite tuff, but field inspection suggests to Kempter et al. (2004) that they are contemporaneous.  Our work at Bear Springs Peak and domes structures to the north suggest that the obsidian was produced from obsidian zones on the margins of the domes as part of a lava eruptive event.  The lava has mostly devitrified into perlite, leaving the most vitreous portions of the flow as remnant marekanites (Figure 4). 

            Located at the far southern end of the Jemez Mountains, just south and adjacent to Jemez Pueblo (Walatowa) Nation land, this Tertiary Period source exhibits only relatively small marekanites now, most smaller than 2 cm in diameter (Figures 3 and 4).  Although the nodule size was apparently small, Bear Springs Peak obsidian was used in prehistory as a toolstone, and was recovered in samples analyzed from early historic period contexts at Zuni Pueblo, probably a result of relationships between the Zuni and Jemez in the 17^th Century (see Howell and Shackley 2003; Shackley 2005:102-105).   The source data as analyzed by Craig Skinner and my lab, suggest that this may be the “Bland Canyon & Apache Tears” source as reported by Glascock et al. (1999:863), collected by Wolfman and reported by him in 1994, and reported as “Canovas Canyon” by Church in his 2000 study of secondary deposits in the lower Rio Grande (Tables 1 and 2 here).

            As with many of the Tertiary Period sources in the Southwest, Bear Springs Peak obsidian is present as marekanites in perlitic lava at the Bear Springs Peak dome proper and domes trending along north-south faults active during the Mid-Miocene toward and into Jemez Pueblo land.  Biotite in the crystalline rhyolite that underlies the obsidian zone yielded ⁴⁰Ar/³⁹Ar dates of 8.81±0.16 and 9.75±0.08 mya (Kempter et al. 2004).

            Nodules up to 5 cm occur as remnants in the perlite not unlike the environment at Sand Tanks, Superior (Picketpost Mountain), Devil Peak, and many other Tertiary period sources in western North America (Shackley 2005; Shackley and Tucker 2001).  The density of the nodules from pea gravel to 5 cm in the perlite lava is as high as 100/m², although most of the marekanites are under 2 cm (Figure 4).  The glass itself is nearly transparent in thin flakes similar to Cow Canyon and Superior obsidian, and some have noticeable dark black and nearly clear banding.  It is an excellent media for tool production so that bipolar flakes are easily produced and pressure flaking is effective, typical of these high silica rhyolite glasses (Table 2).

            The marekanites have eroded into the stream systems to the south as far as the lower Rio Grande (Church 2000).  Church recovered two specimens called “Canovas Canyon” in the Vado (Camp Rice) collection area in the Rio Grande gravels, south of Las Cruces; these two samples match the Bear Springs Peak data as presented here (Church 2000:660, Table IV).  This suggests that the Bear Springs Peak material has eroded into the ancestral Rio Grande between about 9 million years ago and the present, although based on Church’s (2004) study are available today in very low proportions compared to Cerro Toledo Rhyolite, El Rechuelos, and Mount Taylor obsidian.   Probably the reason this source is not found in archaeological contexts more frequently is that it simply cannot “compete” with the more recent large nodule, high quality sources just to the north including Cerro Toldedo Rhyolite, Valles Rhyolite (Cerro del Medio) and El Rechuelos obsidian (Shackley 2005).  However, the high quality, and perhaps its location on Jemez Pueblo territory, may have caused it to be viable as a toolstone, at least in the historic period, perhaps at a time when the Jemez and Zuni enjoyed an exchange relationship.  The Zuni could have made trips into the Jemez Mountains for any number of reasons as a result of this relationship.

Bear Springs Peak (Shackley and Skinner analyses combined).

Magmatic relationships between the glass sources

The relatively short time period of the eruptive events that produced artifact quality obsidian in the Jemez Mountains from El Rechuelos to Valle Grande Rhyolite is reflected in the elemental chemistry as reported by a number of others discussed above (i.e. Baugh and Nelson 1987; Gardner et al. 1986; Glascock et al. 1999). This relationship is readily evident in three dimensional and biplots of the composition of these sources as shown in Figures 3.17 and 3.18. Rubidium and yttrium are most sensitive in separating these sources, with zirconium nearly so. Indeed, a biplot of the elemental concentrations Rb versus Y can effectively separate these sources, although it is NOT sufficient to eliminate the possibility that the analyzed artifacts could be from outside the Jemez Mountains group (Figures below).

Rb, Y, Zr three dimensional plot of Valle Grande, El Rechuelos and Cerro Toledo Rhyolite obsidian source standards. High variability in Valle Grande and El Rechuelos data are the result of the analysis of small secondary distribution nodules (see Davis et al. 1998).

Rb versus Y biplot of the elemental concentrations for Valle Grande, El Rechuelos and Cerro Toledo Rhyolite obsidian source standards. High variability in Valle Grande and El Rechuelos data are the result of the analysis of small secondary distribution nodules (see Davis et al. 1998).

  Back to SW Obsidian Sources page

To the XRF Lab Homepage

Revised: 17 October 2007