Pecos – and from the Seven Rivers Formation outcrops of southeastern New Mexico

James L. Albright* and Virgil W. Lueth, New Mexico Bureau of Geology and Resources, New Mexico Institute of and Technology, 801 Leroy Place, Socorro, NM 87801

Abstract Chaves County (Locality 6) and followed Distribution that report with a description of pseudocu- Pecos diamonds, also known as Pecos valley bic quartz crystals from Artesia in Eddy Large, authigenic quartz and dolomite diamonds are colorful, doubly terminated County (Tarr and Lonsdale, 1929). The crystals appear to be confined exclusively quartz crystals that occur in scattered out- term “Pecos Diamonds” appears to have to the Permian Seven Rivers Formation. crops of the Permian Seven Rivers Formation Essentially all occurrences are confined to (Permian) along the Pecos River valley in been first mentioned in print by Tarr and the back segment of the Guadalupe southeastern New Mexico. Although authi- Lonsdale (1929), where they note that the genic quartz is relatively common in evapor- crystals were described by that name by reef complex starting in the south at the ite sequences worldwide and throughout local collectors. Other names ascribed to “beginning of the facies” of Kelley geologic history, Pecos valley diamonds are the crystals include “Indian diamonds” (1971) near Dark Canyon and terminating unique for their large size, variable color, and “Pecos valley diamonds,” the latter near the De Baca and Guadalupe County and morphologies. Single crystals of term favored by collectors today. Practical line in the vicinity of Salado Creek to the dolomite of variable morphologies also uses for these objects are mainly decorative north. In Pecos country, the back reef seg- occur with the Pecos valley diamonds but with present day uses confined to ment of the Seven Rivers Formation con- are of much more limited distribution. sists predominantly of gypsum with sub- pseudohexagonal crystals are also and jewelry. There is no evidence for their ordinate amounts of dolomitic , present near one locality, but they have not use, decorative or otherwise, by pre-Euro- been found to coexist in the same unit. pean people. red and gray gypsiferous , and fine- Although the outcrop area encompasses Initial descriptions of euhedral authi- grained . East of the study area, an area 100 mi long by as much as 25 mi genic quartz crystals noted the rarity of and salt become prevalent. Ward wide along the margins of the river valley, these in sequences et al. (1986) documented two original the distribution of Pecos valley diamonds is (Tarr, 1929). However, subsequent work depositional modes for the gypsum; these limited to specific depositional environ- has shown that euhedral quartz crystals include subaqueous salina and subareal ments that resemble salinas and/or salt pans are relatively common in ancient shallow sabkha environments. within a larger sabkha setting. Replacement Meinzer et al. (1927) coined the term features within the quartz indicate an authi- marine and evaporite sequences Seven Rivers gypsiferous member for the genic origin with formation soon after (Folk, 1952; Grimm, 1962; Wilson, 1966; of the host rock. However, Zenger, 1976; Ulmer-Scholle et al., 1993). later discredited Chupadera Formation. the large size and suite of inclusions may They are found throughout the Phanero- Tarr (1929) referred to the authigenic indicate a deep burial, late diagenetic origin zoic and exist in rocks as old as quartz occurrence at Acme (Locality 6) as for these crystals. The presence of organic (Grimm, 1962). Other famous occurrences being in the Manzano series of red beds. In matter, formation of dolomite, and the oxi- of authigenic quartz include Herkimer dia- the same year, Tarr and Lonsdale (1929) dation of appear to be important monds hosted by Cambrian dolomites in continued to use the term Chupadera components to the ultimate formation of New York (Zenger, 1976). Large authigenic Group in describing the pseudocubic crys- Pecos valley diamonds. quartz crystals have been described in tals near Artesia (Localities 1 and 2). These young carbonate-evaporite names were changed to Whitehorse, Introduction in the Arabian Gulf (Chafetz Bluff, and Bernal by various authors and and Zhang, 1998). One of the most com- petroleum geologists over the years. Sub- Colorful and doubly terminated authi- prehensive studies of worldwide occur- sequently, Tait et al. (1962) proposed the genic quartz crystals with variable crystal- rences of quartz crystals in was presently used stratigraphic names based lographic forms occur in and weathered published by Grimm (1962). He noted over on the nomenclature applied to the subsur- from scattered outcrops of the Seven 150 localities that displayed similar geo- face rocks of the Artesia Group. Rivers Formation in southeastern New logic and depositional characteristics. Kelley (1971) mapped the surface out- Mexico (Fig. 1). In places, when the sun’s Personal collecting trips by the authors crops of the study area. During his map- rays are at low angles, the desert appears more precisely define the geological set- ping, Kelley (1971, p. 18) reported, “in time paved with diamonds. However, most tings of the localities of Tarr (1929) and Tarr it was noted that their [the Pecos dia- sparkles are only broken or small and and Lonsdale (1929) in addition to docu- monds] distribution is stratigraphically imperfect crystals. Only a small percentage menting 10 other in situ occurrences along related. This is so commonly true that they of crystals are large enough to be of inter- the Pecos River valley. These new localities might be used as a stratigraphic indicator. est to the collector. Crystals in a of highlight areas where authigenic quartz, Most all occurrences are in part of a zone gypsum are only rarely found, but at dolomite, and/or aragonite crystals are rel- perhaps 100 to 200 feet thick from the Locality 1, a 10-cm layer consists of a mass atively abundant, large, or morphological- upper part of the Seven Rivers into the of quartz crystals. ly unique. A discussion of color, crystal lower part of the Yates.” Of the 12 in situ A Spanish miner, Don Antonio de Espe- shape, and inclusion variations is provided occurrences of authigenic quartz crystals jo, first described these crystals in 1583 for each occurrence. Ultimately, we will described in this paper, 10 are within a (Albright and Bauer, 1955). In 1929 Tarr speculate on the origin of these minerals mapped area of Kelley (1971). Localities 12 described the occurrence near Acme in and their significance in sedimentological and 13 also occur in the Seven Rivers For- interpretations. mation based on the mapping of Kelley *Author deceased

August 2003, Volume 25, Number 3 NEW MEXICO GEOLOGY 63 FIGURE 1—Location map illustrating the distribution of the Seven Rivers dix. Outcrop segments that contain Pecos valley diamonds are identified and Yates Formations in southeastern New Mexico. Numbers refer to with arrows. Tectonic elements adapted from Kelley (1971). Pecos localities discussed in the text and presented in the appen-

64 NEW MEXICO GEOLOGY August 2003, Volume 25, Number 3 (1972) on the Fort Sumner sheet. We have of matrix pieces can be found on the mar- Features of the authigenic not been able to document any occurrences gins of drainages that dissect the bluffs. quartz crystals of Pecos valley diamonds in the Yates For- Roswell segment mation during our study. The authigenic quartz crystals from each North of the K-M fault at the mouth of the Pecos valley locality are distinctive enough Outcrop segments Rio Felix, the Seven Rivers Formation as to habit, size, color, and inclusions that reappears in the Roswell outcrop segment they can be grouped into suites. Some Both tectonic and geomorphic features east of the Pecos River (Fig. 1). The Roswell suites are traceable for several kilometers, along the Pecos valley control outcrop dis- segment is continuous over a north-south whereas others are restricted to a few tribution of the Seven Rivers Formation distance of 47 mi with an outcrop width square meters. Some localities contain spe- (Fig. 1). Pecos diamonds occur sporadical- 1 1 varying from ⁄6 to 3 ⁄2 mi. The segment dis- cific forms and colors, whereas others may ly within particular segments of an out- appears in the north in the vicinity of the contain a diverse range. crop belt that is almost continuous. These terminus of the Six Mile Buckle of Kelley outcrop segments, and gaps that separate (1971). Pecos diamond occurrences tend to Size them, are caused by specific tectonic and be confined to the top of the bluffs in this Crystals in the New Mexico Bureau of geomorphic controls that were imposed segment except along the Pecos River cut Geology and Mineral Resources Mineral after Pecos valley diamond development. bank at Locality 4. Museum range from microscopic to a max- Descriptions of Pecos diamond localities imum of 6.5 cm (~2.5 inches) along the c Dunnahoo Hills segment within specific outcrop segments are pre- axis. Median length for perfect crystal sented in the appendix. West of the Pecos, between Roswell and forms is approximately 2.5 cm (~1 inch); Seven Rivers segment Acme, a triangular-shaped outcrop is pres- those larger tend to be distorted. The ent in the Dunnahoo Hills (Fig. 1). No largest single crystals found thus far come The Seven Rivers segment starts at the authigenic quartz or dolomite has been from Locality 13, and the largest clusters dolomite-gypsum transition near Dark found in this segment. The Dunnahoo are from Locality 3. Canyon north to the McMillan escarpment. Hills segment is surrounded by Quater- The southernmost documented occurrence nary terrace and alluvium deposits and of Pecos diamonds was mentioned by not defined by tectonic elements, in con- Dake et al. (1938). They report, “at Seven trast with the southern segments. The lack Rivers, small doubly terminated crystals of Pecos valley diamonds in this segment are found, averaging half an inch in length, may be due to the fact that this segment which have a uniform red-brown color, only exposes the lower portions of the and are quite perfect. The larger crystals Seven Rivers Formation. 1 from the same locality averaging 1 ⁄2 inches in length are of a dirty brown color with Dunlap segment faces as bright and perfect as the smaller North of Dunnahoo Hills and west of the ones...these crystals sometimes assume a Pecos River, the Dunlap segment of the FIGURE 2—An example of color variation of cubical aspect.” The original Seven Rivers Seven Rivers Formation forms an irregular Pecos Valley diamonds from Locality 9. 1 townsite is now 1 ⁄2 mi south-southwest of fan-shaped outcrop covering over 165 mi2 the present village along the transition (Fig. 1). No tectonic elements, as defined zone from dolomite to gypsum facies. We by Kelley (1971), are apparent in this broad Color have not been able to confirm the presence segment. Kelley (1971, 1972) noted an Pecos diamonds take on a wide range of of Pecos diamonds at Dake’s locality nor abundance of Pecos diamonds in the upper colors (Fig. 2). Perfectly transparent crys- have subsequent studies in the area (e.g., 1 portions of the Seven Rivers Formation tals are usually less than 4 mm (~ ⁄8 inch) Sarg, 1981). The segment is terminated by within this area especially south of Arroyo long. Larger crystals take on the color of 1 a 7 ⁄2-mi gap in the outcrop near Lake Yeso (Localities 8–12). The majority of the gypsum matrix, usually with a slight McMillan caused by the presence of the Pecos diamond occurrences are found increase in intensity. Occasionally a bleach- Fourmile Draw syncline (Kelley, 1971). The within this outcrop segment, particularly ing of the host gypsum adjacent to the downwarping of the syncline and subse- over broad areas in the southern part. The quartz crystals is observed. Color bands in quent Quaternary alluvial fill from a mul- northernmost occurrence of Pecos dia- the host rocks, present as laminations or titude of streams that drain the Seven monds is west of El Morro Mesa in De Baca along joints, commonly cross the Rivers embayment cover the outcrop. County (Locality 13). The Seven Rivers included quartz crystals without interrup- Artesia segment Formation thins to the north (Kelley, 1972), tion. Opaque crystals are milky , and Pecos valley diamond occurrences light to dark gray, red, pink, yellow, North of the Fourmile Draw syncline, the decrease in that direction. orange, light to dark red to Seven Rivers Formation reappears on the brown, and light to nearly black magenta. east side of the Pecos River as the Artesia Salado Draw segment Translucent crystals tend to be white, outcrop segment (Fig. 1). This outcrop seg- The northernmost occurrence of the Seven white with pink or orange streaks, or light ment, exposed by the Artesia–Lovington Rivers Formation is confined to the to medium honey brown. Nearly transpar- arch, is 17 mi long by approximately 3 mi drainage valley of Salado Creek at the De ent crystals from Locality 7 contain cloudy wide. The northern margin of this segment Baca and Guadalupe County line. Kelley gypsum inclusions along with greenish- is truncated in the vicinity of the K-M fault (1972) notes that the Seven Rivers, along black material possibly sapropel or hydro- and covered by an embayment of alluvium with the Yates Formation, pinches out . Some crystals show color zonation. deposited by drainages on both sides of beneath the Santa Rosa Sandstone under Very large prismatic crystals from Locality the Pecos River. Tarr and Lonsdale (1929) the pediment cap of Guadalupe Mesa. No 12 are variegated creamy gray green with published their early work on the south- authigenic quartz or dolomite occurrences pink points. ernmost occurrence of Pecos diamonds are known from this segment. Strikingly similar authigenic quartz that we can document. Most Pecos valley crystals are common in the Triassic Keuper diamond occurrences are limited to the gypsiferous facies of Spain, where they are upper benches of small bluffs on the east known as “Jacintos de Compostela”– liter- side of the Pecos River. Excellent exposures ally, Hyacinths from Compostela, an allu-

August 2003, Volume 25, Number 3 NEW MEXICO GEOLOGY 65 minerals. Most common are included Crystal forms grains of gypsum in the outer portions of One of the distinguishing features of Pecos some crystals. Nissenbaum (1967) noted valley diamonds when compared with abundant anhydrite inclusions in Pecos other occurrences of authigenic quartz is valley diamonds and authigenic quartz the great variation in crystal forms (Fig. 6). crystals contained in evaporite clasts from The vast majority of quartz crystals are typ- Israel. The anhydrite inclusions are orient- ically hexagonal prisms (Figs. 5, 6B) termi- ed along growth planes and are preferen- nated on both ends by positive (r) and neg- tially found in the cores of the crystals but ative (z) rhombs (Grimm, 1962; Nis- absent on the rims. Anhydrite is notably senbaum, 1967; Chafetz and Zhang, 1998). absent in the surrounding evaporite in Tarr (1929) reported a sequential develop- both the Pecos valley and Israel occur- ment of crystals from Acme (Locality 1). rences. Kelley (1971) noted gypsum Based on size and morphology, he stated after anhydrite in places that initially the quartz precipitated as a FIGURE 3—Color changes in Pecos valley dia- within the Seven Rivers Formation, how- pseudocube (r), afterwards the quartzoid (r monds after heating at 450°C. All samples illus- ever. Large (up to 3 mm, typically 1 mm) + z) and, finally, the doubly terminated trated as pairs with the heated sample on the rounded, green-black inclusions are rarely prism (r + z + m) (Figs. 6 B–D). Unequivo- left. Sample pairs A and B from Locality 2, C noted in some crystals and may be of D cal development of any other crystal forms from Locality 12, and from Locality 11. The organic origin, although they do not fluo- upper left sample is 1 cm long. of quartz, although observed elsewhere in resce under UV light. These dark inclu- the Seven Rivers Formation in our study, sions are always restricted to the outer- was not observed from specimens in the sion to red-orange precious stones from most portions, and some intersect the sur- Tarr (1929) study area around Acme. A that locality popular during the Middle face of the crystals (Fig. 4). basis for size and sequential development Ages (Febrel, 1963; Rios, 1963). A compari- is rendered false at other localities, howev- son of Pecos diamonds from diverse local- er. Some of the largest crystals found are ities cannot be distinguished from the pseudocube specimens (as long as 6 cm Spanish Jacintos insofar as size, color, [~2.5 inches] along the c axis) at Locality 13 form, and inclusions. in association with large authigenic Chaves (1896) observed that the Spanish dolomite crystals. Jacintos turned gray on heating and sug- Prismatic forms. The most common gested that the were in part megascopic in Pecos valley organic. This heating experiment was diamonds is the regular prism (m) pointed repeated on representative Pecos valley at both ends by hexagonal pyramids diamonds by Abraham Rosenzweig (at the formed by equal, or near equal, develop- University of New Mexico) by raising the ment of both the (r) and (z) rhombs (Fig. 5). temperature in an electric oven to 450° C Grimm (1962) measured 1,000 authigenic and maintaining the temperature for 12 hrs. quartz crystals that were associated with All samples exhibited loss of color, with the FIGURE 5—Prismatic forms of Pecos diamonds evaporite deposits from around the world least change in the hematite red varieties showing rough prism faces (m) and smooth (including the Pecos valley) and found that rhombohedron faces (r and z), even on second- (Fig. 3). Loss of luster occurred primarily the axial ratios range from 1.5 to 3.0 in the from fracturing of the crystal faces follow- ary growths. Note the left and center crystals also show rough (z) faces. In contrast, both (r) majority of cases, the shortest being 1.1 ing decrepitation of fluid, hydrocarbon, or and the most being 5.5. Common- hydrous mineral inclusions. and (z) rhombohedron faces on the crystal on the right are smooth. Quartz crystals illustrated ly the (r) rhomb is relatively large, even are from Locality 11. The left crystal is 1.7 cm though the prism (m) faces are equal or long. nearly so. Most of the (r) faces have a bright luster, whereas the (z) faces tend to Crystal surface features be dull. A significant portion of the pris- matic crystals have prism faces that are Megascopic surface features include varia- alternately wide and narrow; thus, in cross tions in reflectivity, growth lines, negative section, the prism zone appears as if trigo- crystal pits, crystal mold impressions, and nal prisms are present (Figs. 7 and 6E–plan linear depressions. Tarr (1929) noted that the prismatic faces, when present, are typ- ically rough, whereas the rhombohedron faces tend to be smooth (Fig. 5). Even among the rhombohedron faces, Tarr and Lonsdale (1929) noted in pseudocubic vari- eties that the negative rhomb (z) tends to be dull compared to the positive rhomb (r). FIGURE 4—Dark inclusions of organic material When present, growth lines on the prism observed in some specimens. Note the quartz faces (m) tend to be subdued and broad crystal clusters illustrating the tendency for compared with most common quartz. subsidiary growth on the prism faces of the Negative crystal pits are common on prism quartz crystal. Illustrated examples from Local- faces and locally attain large sizes (7 mm ity 3. The lower left crystal is 2 cm long. 1 [~ ⁄4 inch] on a 5-cm [~2-inch] crystal). These pits may be casts of dissolved dolomite or other evaporite minerals and Inclusions FIGURE 7—Quartz crystals displaying partial represent mold impressions. Linear de- development of a pseudotrigonal bipyramid In addition to color, many Pecos valley dia- pressions are typically zones of crystal- habit from Locality 1. The crystal on the left is 2 monds contain discrete inclusions of other growth overlap. cm long.

66 NEW MEXICO GEOLOGY August 2003, Volume 25, Number 3 Pseudotrigonal Pyramids

FIGURE 6—Crystal diagrams of the various forms of Pecos valley diamonds noted in the study area. Diagram modified from quartz crystal incremental growth concepts of Grimm (1962).

August 2003, Volume 25, Number 3 NEW MEXICO GEOLOGY 67 quartz known as beta quartz (Frondel, rhombohedra (r) and (z) are bright and 1962). This form is best thought of as smooth. This is especially true when the matching (r) and (z) faces at the ends of a axial ratio is greater than three. At Locality very narrow prism zone. Localities 2 and 6, 1, tapered prismatic crystals of this type, as 3 and especially at the latter, contain abun- long as 3.5 cm (1 ⁄8 inches) with axial ratios dant examples of this form (Fig. 9). The averaging 3.5, also give the illusion of obvious low-temperature environment being bent or twisted along the c axis (Fig. required by the mineral assemblage in the 11). Pecos valley suggests a high-temperature Clusters of quartz crystals are some- origin for beta quartz forms is not times observed, especially at Localities 2 required. and 3. Interestingly, these clusters are com- posed of one large core crystal with small- er crystals growing from the (m) faces almost exclusively (Fig. 4). These smaller crystals tend to grow perpendicular to the prism (m) face. These subsidiary crystals radiate from the (m) faces parallel to the bedding planes within the host gypsum, a FIGURE 8—Equant pseudocubic quartz crystals feature also noted in authigenic aragonite from Locality 13, placed at various orientations. The large crystal on the top is 2.5 cm across. crystals found in the Grayburg–Queen Formation stratigraphically lower in the section at Locality 13. It is important to view). The slanted (r) and (z) faces are also note, however, authigenic quartz is only equally wide and narrow, forming match- found with dolomite and never observed ing trigonal pyramids at each end. Tabular with aragonite. crystals flattened parallel with a pair of coplanar prism faces (m) are rare. Only one such tabular Japan-Law twin has been found at Locality 12; however, normal pris- FIGURE 10—Trigonal prism form of Pecos dia- matic Japan-Law twins are fairly common monds from Locality 1. Note the color zonation at Locality 11. in the upper portion of the crystal. Maximum Equant forms. Equant pseudocubic dimension is 2.2 cm. habit results when the faces of (r) are developed to the exclusion, or near exclu- Pseudotrigonal bipyramids (Figs. 10, 6E) sion, of both (z) and (m). Because the inter- at Localities 1 and 13 are the result of the face angle is 85°46’, these crystals appear rare over development of (r) and diminu- cubic to the unaided . Vestigial devel- tion of (z), to the exclusion or near exclu- opments of (z) and/or (m) are nearly sion of the prism faces (m). These are also always present. Locality 1, first described the localities with abundant pseudocubic by Tarr and Lonsdale (1929), yields types and may represent modification of 3 pseudocubic crystals as great as 1 cm (~ ⁄8 previously formed pseudocubes (Albright FIGURE 12—Examples of loose authigenic inch) in size. Pseudocubes as long as 2.5 cm and Krachow, 1958) in a manner similar to dolomite from the study area. Maximum length of the largest crystal is 2 cm. (~1 inch) on an edge also occur at Locality that described by Tarr (1929). 13 (Fig. 8). Equal or near equal development of (r) and (z) to the exclusion or near exclusion Features of the authigenic of (m), simulates a quartzoid habit, which dolomite crystals is more characteristic of high-temperature Euhedral, authigenic dolomite crystals also occur in the Seven Rivers Formation (Fig. 12). Sparse, light to medium honey-yellow crystals have been found at Locality 3. At Locality 4, medium to dark honey-brown 3 crystals as great as 2 cm (~ ⁄4 inch) in size occur in and weathered from massive white gypsum slump blocks. These crystals do not withstand and are diffi- cult to free from matrix. Nevertheless, both fine matrix specimens and loose crystals from this locality are found in museums around the world. Small (5 mm), scarce, transparent to translucent, euhedral FIGURE 11—“Bent or twisted” appearing dolomite crystals occur in white gypsum FIGURE 9—Pecos diamonds exhibiting the quartz crystals from Locality 1. These crystals with abundant medium to dark honey- quartzoid and pseudo-octahedral habits from tend to be more mottled than typical Pecos val- brown prismatic quartz crystals that aver- various localities. Top crystal group represents a ley diamonds. Largest crystal is 4 cm. age 2.5 cm (~1 inch) in length at Locality 12. common twinning habit in beta quartz. Crystal bottom center is a pseudo-octahedral type. Note initiation of prism face growth on left and right Distorted crystals. The prism (m) faces FIGURE 13—Crystal drawings of the variations quartzoid crystals. The crystal on the lower left are not always completely developed and in authigenic dolomite crystal morphology from is 2 cm long. may be rough and pitted, even when the the study area.

68 NEW MEXICO GEOLOGY August 2003, Volume 25, Number 3 August 2003, Volume 25, Number 3 NEW MEXICO GEOLOGY 69 probably a function of the scattered nature of salinas or salt pans that served as catch- ments for meteoric water and perhaps fine- grained wind-blown and silt. The silty units could serve as a potential source for silica although other workers have cited a biogenic source of silica, mainly from spicules (Ulmer-Scholle et al., 1993; Chafetz and Zhang, 1998). Alternatively, the distribution of the Pecos valley diamonds may represent zones of diagenetic alteration or fluid migration in the deep subsurface. Fluid or hydrocarbon migration may have been FIGURE 14—Authigenic dolomite (A.) and Pecos valley diamond (B.) in matrix from Locali- confined to the upper portions of the ty 1 an 11 respectively. The dolomite crystal in A. is a Type 4 (Fig. 13), pseudo-octahedral Seven Rivers Formation, hence the absence shape. Note the rhombohedral shaped vug in sample B with quartz occupying only part of of Pecos valley diamonds in the lower por- the space. A small rim of gypsum marks the extent of the original crystal that dissolved and tions. The presence of anhydrite inclusions left the vug later partially filled by quartz crystal with a pseudo-trigonal prism form. Both may be indicative of deeper burial, and the specimens are 7 cm in the longest dimension. silica could be derived from diagenetic alteration of clays. The gypsum-rich outer zones observed in some Pecos valley dia- The geometry of the dolomite crystals is processes that may have led to the forma- monds may be linked to subsequent uplift formed by combinations of the positive tion of the Pecos valley diamonds. and the conversion of anhydrite rhombohedron {4041} truncated by posi- A fundamental question that remains to gypsum during final quartz growth. tive and negative basal pinacoids spaced at unanswered is the timing of Pecos valley The presence of dolomite and absence of specific axial ratios (Fig. 13). Crystal Types diamond precipitation. The presence of aragonite are important indicators for the 1 through 7 are found at Locality 3, includ- authigenic quartz in deeply buried evapor- presence of Pecos valley diamonds. In ing one specimen with hemimorphic trun- ites and is well known (e.g., evaporite units that underlie the Seven cations by Types 7 and 8 pinacoids. These Ulmer-Scholle et al., 1993). However, the Rivers Formation (e.g., the Grayburg– 3 crystals average less than 1 cm ( ⁄8 inch) in recent discovery of modern megaquartz in Queen Formation) large and euhedral length. Types 4 and 5 dolomite forms occur dolomite in the Arabian Gulf (Chafetz and are fairly common and quartz is at Locality 4. Locality 12 contains Types 1 Zhang, 1998) reveals the possibility of absent. Chafetz and Zhang (1998) also note through 5, and they average 2.5 cm (~1 early diagenetic formation near the sur- an absence of quartz in aragonite and low- inch). No Type 9 dolomite forms have been face. The features exhibited by the Pecos units that overlie the recognized in the study area. Where gyp- valley diamonds and their host rocks have quartz-bearing dolomites in the Gulf of sum matrix pieces are available, careful much in common with those described by Arabia. They suggest the aragonite low- examination of vugs occupied by Pecos other workers as characteristic of both Mg units were deposited in a marine envi- valley diamonds in a couple of locations (2 near-surface and deep environments. ronment, more similar to the Grayburg– and 9) suggests that the space may have These competing modes of formation will Queen Formation. This relationship sug- been originally a dolomite rhomb (Fig. 14). be discussed with regard to the possible gests that the Seven Rivers Formation may Known as “las Teruelitas” in Spain, black origin of these fascinating quartz and have undergone dolomitization and that pseudo-octahedral Type 4 dolomite crys- dolomite crystals. the Pecos valley diamonds formed during tals were first described by D. Amalio or shortly after this process. The formation Depositional environment Maestra in 1845 from a similar geologic set- of the quartz in this environment would ting. At the type locality (Barranco de Sola- Previous sedimentological studies of the not require deep burial and is probably bral, northeast of Teruel in the province of Permian Seven Rivers Formation inferred characteristic in portions of the sabkha the same name) the Teruelitas occur with the depositional environment to be similar environment. Nissenbaum (1967) also black Jacintos de Compostela in a gray to to modern sabkha environments present in rejected a deep burial replacement origin black Keuper gypsum (Muñoz and Piñero, the Arabian Gulf today (Kendall, 1969; Till, for Pecos valley diamonds based on the 1951). Chemical analyses of the Teruelitas 1978; Ward et al., 1986; Warren, 1989). The presence of growth lines and zonal anhy- show them to be calcian-dolomite. The distribution of authigenic dolomite and drite found in the quartz crystals. He black color results from a small and quartz crystals appears to be related to a believed the quartz was formed during content. Rogers (1949) first specific depositional environment within primary anhydrite precipitation on reported on the similarities of Pecos valley the Seven Rivers Formation. The crystals supratidal salt flats, consistent with the dolomite to Spanish Teruelitas. are not distributed uniformly throughout sabkha depositional model. the evaporitic sequence stratigraphically However, the very large size of the Discussion or across the study area. The crystals tend authigenic dolomite and quartz crystals to be concentrated near the top of the for- are not common in low-temperature, sedi- During his study of over 150 authigenic mation (Kelley, 1971) and are only sporad- mentary environments. Large crystal sizes occurrences of quartz in evaporites, ically distributed within the entire outcrop in these minerals are more characteristic of Grimm (1962) noted features that were area. In general, their distribution pattern the hydrothermal environment and may common to all localities: 1) presence of is typically circular to elongate. indicate deep burial and formation during saline facies; 2) evidence of euxinic envi- The distribution patterns of Pecos valley diagenesis at higher temperatures. ronments; and 3) evidence of petroleum diamonds, and their tendency to be most migration and entrapment. We will pro- abundant stratigraphically adjacent to thin The role of organic matter vide more information on these features in shale units, suggests they were confined to Significant changes in oxidation and pH addition to specifying some of the details subenvironments of the sabkha, most like- are required for the formation of dolomite of the depositional environment, timing of ly shallow salinas or salt pans. The spotty and later formation of quartz. Stable iso- the quartz formation, and speculate on the distribution of Pecos valley diamonds is tope work has identified organic carbon as

70 NEW MEXICO GEOLOGY August 2003, Volume 25, Number 3 involved in the anaerobic bacterial reduc- edly exceeded the space of the precursor rocks would be useful in testing the specu- tion of sulfur to sulfide before chert forma- minerals and further engulfed the evapor- lations presented in the discussion above. tion in some chalk units (Clayton, 1986). ite minerals. Some quartz crystals appear Variations in both carbon and iso- Similar studies on authigenic quartz in to have grown entirely within the gypsum, topes should fall within values observed settings (Palmer, 1995) noted an especially those with abundant inclusions. by other workers (e.g., Clayton, 1986; important contribution by organic carbon If the quartz replaced earlier dolomite Palmer, 1995) and help identify the type compounds in the formation of quartz. forms, some of the unusual quartz forms and role of organic matter in this process. Bennett et al. (1988) also identified the abil- (e.g., pseudocubes, etc.) may actually be Detailed mapping of quartz, dolomite, ity of organic compounds to dissolve and quartz pseudomorphs after dolomite. The and aragonite occurrences may prove use- later reprecipitate quartz. presence of quartz in vugs with rhombohe- ful for delineating different depositional Abundant organic matter is observed in dral outlines, similar to dolomite, suggests environments within the Seven Rivers For- partially aragonitized mats presently the precipitation of quartz at least accom- mation or pathways for diagenetic fluids, forming in the Macleod evaporite basin of panied the dissolution of dolomite. depending on the results of fluid inclusion Australia (Logan, 1987), another potential Tarr (1929) suggested hematite as the analysis. The limited distribution of the modern analog to the Seven Rivers Forma- main coloring agent in the crystals and Pecos valley diamonds suggests that they tion. This material forms a sapropel, a jelly- speculated that the coloration was induced are unique to particular depositional/dia- like ooze composed of algal remains mac- from the outside of the evaporite unit after genetic environments within fairly monot- erating and putrefying in an anaerobic quartz formation. Later the same year, Tarr onous evaporite units. Detailed petrogra- environment in shallow water. Other and Lonsdale (1929) noted that coloration phy of the evaporites and associated silty workers have noted periods of hydrocar- of the rock preceded the formation of the units may also show evidence of quartz bon migration and associated evaporite quartz crystals as the color banding in the within the host rocks (e.g., Ben- silicification in the Seven Rivers Formation host rock was continuous through inter- nett and Siegel, 1987). south of the study area (Ulmer-Scholle et vening quartz. They speculated that the al., 1993). Regardless of the type of organic source of the color was from adjacent red Acknowledgments matter, it appears it had an important role beds. Nissenbaum (1967) analyzed both in the reduction of sulfate to sulfide in authigenic quartz crystals from Israel and The second author (VWL) would like to parts of the study area and may have Pecos valley diamonds to find some crys- thank the Albright family, especially Jim’s mobilized silica for later precipitation as tals contained as much as 0.3% Fe2O3. We widow Carrie and son Jon, for permission Pecos valley diamonds. speculate an influx of meteoric water oxi- to publish this work with modifications. In dized , creating locally acidic addition, the manuscript reviews of Drs. Dolomite and quartz formation waters that dissolved the dolomite and led Brian Brister, NMBGMR, and Dana Ulmer- Authigenic dolomites within the Seven to the precipitation of quartz. Stable iso- Scholle, NMIMT, are gratefully acknowl- Rivers Formation are unusually large and tope work by Chafetz and Zhang (1998) edged along with the help and suggestions have a wide variety of morphologies not documented periodic episodes of meteoric from the editorial (Jane Love and Nancy typical of sedimentary environments. water flushing during in the modern Gilson) and cartography (Leo Gabaldon) These crystals are typically dark when megaquartz occurrence in the Arabian staff members of the New Mexico Bureau compared to the matrix. This is in contrast Gulf. A similar episode or episodes of oxi- of Geology and Mineral Resources. to the similar colors exhibited between dation would explain the near ubiquitous host rock matrix and quartz crystals. This red coloration seen in many Pecos valley References dark coloration may be due to small diamonds. Alternatively, migration of dia- amounts of manganese or iron like Spanish genetic fluids could also to changing Albright, J. L. and Bauer, R. M., Jr., 1955, Pecos Val- “Teruelitas,” although the matrix in the ley diamonds: Rocks and Minerals, v. 30, nos. 7–8, fluid chemistry that would favor quartz pp. 346–350. Pecos valley is not dark like that in Spain stability over dolomite. Albright, J. L. and Krachow, T. 1958, Schwebend (Muñoz and Piñero, 1951). Alternatively, Suggestions for further work gebildete quartzcristalle in New Mexico: Der they may be dark colored because of Aufschluss, v. 5, pp. 98–101. hydrocarbon or sulfide mineral inclusions The varied morphology of the quartz crys- Bennett, P. C., Melcer, M. E., Siegel, D. I., and Has- indicative of reducing conditions during tals, especially variations in crystal face sett, J. P., 1988, The dissolution of quartz in dilute luster and orientation of subsidiary quartz aqueous solutions of organic at 25°C: their formation. In contrast, authigenic Geochimica et Cosmochimica Acta, v. 52, pp. quartz is typically associated with oxi- growth on (m) faces suggests that unique 1521–1530. dized colors and mineral phases. We have controls on were present Bennett, P. C. and Siegel, D. I., 1987, Increased solu- not observed dolomite and quartz in the during quartz precipitation. Surface stud- bility of quartz in water due to complexing by same sample or stratigraphic layer, ies on these crystals may reveal features organic compounds: Nature, v. 326, no. 6114, pp. although they can be found together at a that lead to a better understanding of both 684–686. Chafetz, H .S. and Zhang, J., 1998, Authigenic euhe- particular locality. The mutually exclusive internal and external controls on crystal dral megaquartz crystals in a Quaternary occurrence within a particular stratigraph- morphology. dolomite: Journal of Sedimentary Research, v. 68, ic horizon may indicate an aspect of each A fluid inclusion study on the dolomites no. 5, pp. 994–1000. mineral’s formation that precludes the and quartz would greatly help to constrain Chaves, 1896, Effect of heating on Jacintos de Com- presence of the other. Accordingly, we the origin of the quartz crystals. Homoge- postela: Am. Soc. Esp. Hist. Nat., v. 25, p. 243. speculate that the dolomite and quartz nization temperatures and salinities could Clayton, C. J., 1986, The chemical environment of flint formation in Upper ; in formed at different times based on the differentiate between crystals formed near Sieveking, G. deG., and Hart, M. B. (eds.), The presence of rhomb-shaped vugs occupied surface and at low temperatures (e.g., scientific study of flint and chert: Cambridge by quartz. Chafetz and Zhang, 1998) and those University Press, pp. 45–54. Many of the features within the quartz formed during deeper burial and at higher Dake, H. C., Fleener, F. L., and Wilson, B. H., 1938, crystals suggest that they formed by both temperatures (e.g., Ulmer-Scholle et al., Quartz family minerals—a handbook for the 1993). Perhaps the Pecos valley diamonds mineral collector: McGraw-Hill, 304 pp. open space precipitation and in situ Febrel, T., 1963, Jacintos contenidos en anhyritas: replacement of gypsum and dolomite. have components that suggest they formed Materales salinos del Suelo Espanol, Apedice 2: Some of the Pecos valley diamonds are initially near the surface and grew larger Instituto Geologico y Mineros de España, Memo- found in vugs with rhombohedral outlines with diagenesis. Stable isotope studies of ria 64. similar in form to the authigenic dolomites organic materials contained in inclusions Folk, R. L., 1952, and petrology of the (Fig. 14). Larger quartz crystals undoubt- within the Pecos valley diamonds and host Lower Beekmantown carbonate

August 2003, Volume 25, Number 3 NEW MEXICO GEOLOGY 71 rocks in the vicinity of State College, Pennsylva- las teruelitas: Inst. Geo y Minero de España, in gypsum from Acme, New Mexico: American nia: Unpublished Ph.D. dissertation, Pennsylva- Notas y Comuus. 25, pp. 3–8. Mineralogist, v. 14, no. 1, pp. 19–25. nia State University, 366 pp. Nissenbaum, A., 1967, Anhydrite inclusions in Tarr, W. A. and Lonsdale, J. T., 1929, Pseudocubic Frondel, C., 1962, The system of ; v. 3, idiomorphic quartz in gypsum from quartz crystals from Artesia, New Mexico: Amer- Silica minerals: John Wiley and Sons, 334 pp. Makhtesh Ramon, Israel: Israel Journal of ican Mineralogist, v. 14, no. 2, pp. 50–53. Grimm, W. D., 1962, Idiomorphe quarze als leeit- Sciences, v. 16, pp. 30–33. Till, R., 1978, Arid shorelines and evaporites; in mineralien fur salinare fazies: Erdöl Kohle, v. 15, Palmer, A. N., 1995, Geochemical models for the Reading, H. G. (ed.), Sedimentary Environments pp. 880–887. origin of macroscopic solution in car- and Facies: Elsevier, pp. 178–206. Kelley, V. C., 1971, Geology of the Pecos country, bonate rocks; in Budd, D. A., Saller, A. H., and Ulmer-Scholle, D. S., Scholle, P. A., and Brady, P. V., southeastern New Mexico: New Mexico State Harris, P. M. (eds.), Unconformities and porosity 1993, Silicification of evaporites in Permian Bureau of Mines and Mineral Resources, Memoir in carbonate strata: American Association of (Guadalupian) back-reef carbonates of the 24, 78 pp. Petroleum Geologists, Memoir 63, pp. 77–101. , west Texas and New Mexico: Kelley, V. C., 1972, Geology of the Fort Sumner Rios, J. M., 1963, Materiales salinos del Suelo Journal of Sedimentary Petrology, v. 63, no. 5, pp. Sheet, New Mexico: New Mexico State Bureau of Español: Inst. Geol. y Minero de España, Memo- 955–965. Mines and Mineral Resources, Bulletin 98, 55 pp. ria 64 p. Ward, R. F., Kendall, C. G. St. C., and Harris, P. M., Kendall, C. G. St. C., 1969, An environmental re- Rogers, A .F., 1949, Dolomite crystals of the Teruel 1986, Upper Permian (Guadalupian) facies and interpretation of the Permian evaporite/carbon- habit from Lake Arthur, Chaves County, New their association with hydrocarbons—Permian ate shelf sediments of the Guadalupe Mountains: Mexico (abs.): Geological Society of America, Basin, west Texas and New Mexico: American Geological Society of America, Bulletin, v. 80, no. Abstracts with Programs, v. 60, no. 12, p. 1943. Association of Petroleum Geologists, Bulletin, v. 12, pp. 2503–2526. Sarg, J. F., 1981, Petrology of the carbonate-evapor- 70, no. 3 pp. 239–262. Logan, B. W., 1987, The Macleod evaporite basin, ite facies transition of the Seven Rivers Formation Warren, J. K., 1989, Evaporite sedimentology— western Australia—Holocene environments, sed- (Guadalupian, Permian), southeast New Mexico: importance in hydrocarbon accumulation: Pren- iments and geological evolution: American Asso- Journal of Sedimentary Petrology, v. 51, no. 1, pp. tice Hall, 185 pp. ciation of Petroleum Geologists, Memoir 44, 140 73–96. Wilson, R. C. L., 1966, Silica diagenesis in Upper pp. Tait, D. B., Ahlen, J. L., Gordon, A., Scott, G. L., of southern England: Journal Meinzer, O. E., Renick, B. C., and Bryan, K., 1927, Motts, W. S., and Spitler, M. E., 1962, Artesia of Sedimentary Petrology, v. 36, pp. 1036–1049. Geology of no. 3 reservoir site of the Carlsbad Group (Upper Permian) of New Mexico and west Zenger, D. H., 1976, Definition of type Little Falls irrigation project, New Mexico, with respect to Texas: American Association of Petroleum Geolo- Dolostone (Late Cambrian), east-central New water tightness: U.S. Geological Survey, Water- gists, Bulletin, v. 46, no. 4, pp. 504–517. York: American Association of Petroleum Geolo- supply Paper 580-A., pp. 1–39. Tarr, W. A., 1929, Doubly terminated quartz crystals gists, Bulletin, v. 60, pp. 1570–1575. Muñoz, C., and Piñero, A., 1951, Breve nota sobre

James Lofton Albright was born in Kingwood, West Virginia, on December 26, 1922, to Charles and Hazel Albright. He developed an interest in minerals at an early age after seeing the crystals in a new railroad cut near his home. After graduating from high school at age 16, Albright spent a short amount of time at the University of West Virginia before the hard economic times of the depression caused him to leave school and seek employ- ment in the naval shipyards in Orange, Texas. There he met his future wife, Carrie Willey, and they were married in June 1942. After spending 3 years in the South Pacific during World War II, Albright returned to Texas and enrolled at the Texas School of Mines, now the University of Texas at El Paso. There he earned a B.S. degree in in 1949. In El Paso, Albright was able to practice his hobby of mineral collecting in earnest, soon amassing an impressive collection with emphasis on specimens from New Mexico, Texas, Arizona, and northern Mexico. After graduation, Albright began a long and varied career in the oil and gas industry as a geologist and geophysicist. He worked for Pan American Petroleum in Roswell, New Mexico, from 1950 until 1959, and then for Pubco Petroleum in Albuquerque until 1966. After a short stint with a NASA project at the University of New Mexico, Albright and his family moved to Texas where he worked for Petty Geophysical in Houston and San Antonio. In 1971, Albright began work as an independent consultant in Hous- ton. He retired to Canyon Lake, Texas, in 1993 where he passed away in September 2000. Albright had a life-long passion for minerals and mineral col- James Lofton Albright lecting. Most of his collection he gathered himself, including a large collection of Pecos diamonds. He proudly displayed his most interesting and favorite samples in special glass cabinets at his home. Albright’s minerals are now part of the collection at the mineral museum of the New Mexico Bureau of Geology and Min- eral Resources at New Mexico Institute of Mining and Technology in Socorro. The donation represents one of the most significant in recent museum history consisting of many singularly unique and historic pieces. A review of Stuart Northrop’s book, The Minerals of New Mexico, reveals many entries with described occurrences attributed to James L. Albright—many of them one of a kind. In his tradition of sharing mineralogical knowledge, this paper is his last work, most of it written shortly after retirement.

72 NEW MEXICO GEOLOGY August 2003, Volume 25, Number 3 Appendix Localities of authigenic quartz and dolomite crystals.

Artesia outcrop segment slump blocks of Seven Rivers vicinity of the gaging station, Sample localities 1–4 gypsite. These slumps result they contain abundant world- from continuing erosion, aug- famous-euhedral Types 4 and Locality no. 1 Locality no. 3 mented by solution, along the 5 pseudo-octahedral dolomite Artesia, NM Lake Arthur, NM trace of the K-M fault. Near crystals. secs. 19, 30, 31 secs. 33, 34 T15S R26E river level, especially in the 1 T17S R27E Artesia NE 7 ⁄2-min quad. 1 Spring Lake 7 ⁄2-min quad. 32°56’57”N 32°48’59”N 104°20’06”W Roswell outcrop segment 104°18’35”W Sample localities 5 & 6 Locality no. 5 cast gypsum ashtrays, book Roswell, NM ends, etc. for the tourist trade. secs. 3, 10, 11, 14 They are, or have been, so plen- T11S R26E tiful that this type has become Bitter Lake, Bottomless Lakes, synonymous with “Pecos dia- Comanche Spring, and South monds” in the minds of most 1 Spring 7 ⁄2-min quads. mineralogists. 33°23’45”N 104°22’38”W Locality no. 6 Acme, NM sec. 30 T8S R26E 1 Deeply weathered Seven Ri- Acme 7 ⁄2-min quad. Tarr and Lonsdale’s (1929) vers gypsite high on a topo- 33°35’18”N 1 104°20’48”W paper, as well as the fine crys- graphic rise in the SE ⁄4 sec. 35 tals that can be found here, T16S R26E along the Eddy– especially the pseudocubes, Chaves County line contains have made this locality the sparse light to medium honey- destination of collectors over yellow euhedral dolomite the years. With a little work crystals as large as 2 cm. These this locality can still yield are combinations of the rhom- good examples of authigenic bohedron M{4041} truncated quartz forms, both loose and by a pair of basal pinacoids On Comanche Hill, 10 mi east in matrix. Types 1 through 7. of Roswell in the vicinity of Sparse, poorly developed, Bottomless Lakes State Park, brownish prismatic authigenic drussy, light to dark hematite- Locality no. 2 quartz crystals, as large as 2 red quartz crystals were for- Artesia, NM cm along the c axis, also occur merly abundant. The drussy sec. 18 T17S R27E in and weathered from the faces are the terminations of In the first professional paper 1 gypsite. Spring Lake 7 ⁄2-min quad. prismatic crystals radiating published on in situ authi- 32°49’39”N about the center of a single genic quartz crystals in the 104°18’35”W short to long prismatic crystal. Seven Rivers Formation, Tarr Locality no. 4 If short, the radiating cluster (1929) describes white to pink Lake Arthur, NM may approximate a sphere. If drussy crystals “...one mile secs. 26, 35 T15S R26E long, the drussy area is usual- 1 southwest of Acme along the Artesia NE 7 ⁄2-min quad. ly insignificant. highway to Roswell....The 32°59’20”N In the 1930s and 1940s quartz crystals range from .075 104°19’11”W drussy crystals from Coman- millimeter to two centimeters che Hill were used to decorate in length.” Dunlap outcrop segment Sample localities 7–13 Locality no. 7 Selmen Draw, Chaves County sec. 4 T7S R24E Locality 2 is separated from Coyote Draw, Shannon Draw 1 Locality 1 based on a distinc- and Marley Draw 7 ⁄2-min tive assemblage of attractive, quads. small, medium to dark blood- 33°44’21”N red crystals that can be 104°31’34”W screened from loose silty sand in large numbers. These are In secs. 25, 26, and 27 T15S Abundant, nearly transparent, mainly equant forms averag- R26E the Pecos River makes a prismatic quartz crystals, both ing 5 mm with rare pseudo- sharp west-southwest turn for loose and in matrix, occur in the 1 1 1 . No bedrock is exposed. 2 ⁄2 mi around large obsequent SW ⁄4 sec. 34 T6S R24E and SW ⁄4

August 2003, Volume 25, Number 3 NEW MEXICO GEOLOGY 73 sec. 3 T7S R24E. These attractive Locality no. 10 these form normal prismatic Locality no. 13 crystals, averaging 2 cm along Huggins Draw, Chaves County Japan-Law twins. Overton Ranch the c axis, contain cloudy-white sec. 10 T5S R25E De Baca County 1 inclusions of gypsum and Deering Place 7 ⁄2-min quad. secs. 7, 8 T1N R21E 1 greenish-black material resem- 33°52’45”N Locality no. 12 Yeso Mesa SE 7 ⁄2-min quad. bling sapropel. Large, dry sink- 104°18’13”W Gibbin Ranch 34°18’50”N holes in gypsite, comparable in De Baca County 104°46’11”W 1 size to those at Bottomless SW ⁄4 sec. 10, T3S R24E 1 Lakes State Park, suggest origi- Lovelady Draw 7 ⁄2-min quad. nal deposition in a salina envi- 34°03’05”N ronment. 104°24’32”W

Locality no. 8 Shannon Draw, Chaves County sec. 2 T6S R24E Cottonwood Draw, Coyote Draw and Eightmile Draw 1 7 ⁄2-min quads. 33°49’10”N 104°26’13”W Plentiful drussy, medium- to dark-orange, prismatic crystals Abundant prismatic quartz occur in and weathered from crystals as long as 70 mm and light-orange gypsite bluffs on pseudocubes as great as 25 mm both sides of Huggins Creek in on an edge occur in and weath- sec. 11 T5S R25E. Average size On the Gibbin Ranch abundant ered from medium- to dark- of these colorful crystals is 18 very good, light to dark honey- gray gypsite in sec. 8 T1N R21E mm. brown prismatic quartz crys- on the Overton Ranch. These tals occur in and weathered crystals are accompanied by from light-gray to tan gypsite Locality no. 11 scarce pseudotrigonal bipyra- in sec. 10, T3S R24E. These Old Cavel School, Chaves mids as long as 35 mm along crystals contain copious gyp- County the c axis that are formed by sum inclusions. Average size is secs. 23, 24 T4S R23E unequal development of the (r) 25 mm. Largest crystal found Dunlap Sill and Swallow and (z) rhombohedra and the 1 measures 50 mm (Albright and absence or near absence of the Nest Canyon 7 ⁄2-min quads. Sparse pink, orange, red, and Bauer, 1955). 33°56’43”N prism (m) (Albright and Kra- white prismatic crystals aver- Scarce, small (5 ± mm), trans- 104°29’21”W chow, 1958). aging 15 mm along the c axis lucent, white euhedral dolo- The larger prismatic crystals occur loose and in white to mite crystals Types 1 through 5 are generally mottled reddish pink gypsum matrix in sec. 36 are commonly embedded in creamy gray or white with T6S R25E. the gypsum matrix along with pink points, whereas the small- the quartz crystals. These do er crystals of all types are about not weather well and are easily evenly divided between white Locality no. 9 overlooked. Very fine matrix and light to dark pink. Huggins Hill, Chaves County specimens containing both sec. 36 T5S R24E authigenic quartz and dolo- All illustrations in the article and appen- Cottonwood Draw and mite crystals are easily dix with the exception of Figure 1 were 1 Shannon Draw 7 ⁄2-min quads. obtained. created by Virgil Lueth. 33°49’23”N 104°26’18”W

Abundant dark-magenta to nearly black, short to long, stout, doubly terminated pris- matic crystals occur in and weathered from medium- to light-magenta gypsite in sec. 13 T4S R23E south of the aban- doned Old Cavel School. These crystals average 2 cm but may be as long as 4 cm along the c axis. A very few of these form normal prismatic Japan-Law twins. Sparse loose, white and dark- Southeast of the abandoned brown, almost black, prismatic schoolhouse, plentiful hand- crystals averaging 15 mm some, opaque, medium hema- along the c axis have been col- tite-red prismatic crystals as lected in a shallow ravine in long as 3 cm occur loose in the sec. 36 T5S R25E. . Approximately 5% of

August 2003, Volume 25, Number 3 NEW MEXICO GEOLOGY 74