Soil Science Society of America JournalBusiness and Editorial Offi ces at677 South Segoe Road, Madison, WI 53711(www.soils.org)

Soil Science Society of America Journal Editorial Board

S.D. Logsdon, Editor-in-chiefD.D. Myrold, Editor

Technical EditorsG.V. Wilson (Div. S-1) T.L. Thompson (Div. S–4, S–8)P. A. McDaniel (Div. S-5, S-9, S-10) S.M. Dabney (Div. S-6)S.R. Goldberg (Div. S-2) S.C. Hart (Div. S-3, S-7)L. Ahuja (media review)

Associate Editors

Ellen G.M. Bergfeld, Chief Executive Offi cerFrances Katz, Director of PublicationsRebecca Funck, Managing Editor (rfunck@agronomy.org)

2008 Offi cers of SSSAG. A. Peterson, PresidentColorado State UniversityFort Collins, CO

R. Lal, Past-PresidentOhio State UniversityColumbus, OH

P. M. Bertsch, President-ElectUniversity of KentuckyLexington, KY

ISSN 0361-5995

M.A. ArthurJ.R. BachmanR.L. BaumhardtH. Blanco-CanquiJ.L. BoettingerS.M. BrouderK.R. BryeF.X. CaseyM.A. CavigelliF.D. ChoroverT.J. CloughC.B. CraftH.M. DionM.E. EssingtonA. FaresT.P.K. FerréJ.M. GalbraithS. GaoH.H. GerkeT.A. GhezzeheiD. GimenezP.R. GrosslS. Grunwald

N.C. HansenP.A. HoldenW.R. HorwathT.A. HowellM.G. JohnsonT.C. KasparQ. KetteringsR. KhoslaR.T. KoenigS. KuoB.D. LeeG. LevyA. LeytemH. LinD.J. LoweM.A. MayesC.H. NakatsuB.A. NeedelmanA. NemesK. NoborioJ.M. NortonT.A. ObrezaL. Prunty

N. QafokuS. QuideauC. ReddyT. RenW.R. RoyP.C. ScharfW.F. SchillingerL.A. SchipperA. SchlegelT.E. SchumacherG.J. SchwabR.C. SchwartzD.A. ScottB. SiD.G. StrawnC.C. TrumanJ.J. VarcoO. WendrothT.O. WestJ.R. WhiteH ZhangX.-C. Zhang

Published bimonthly by the Soil Science Society of America, Inc. Periodicals paid at Madison, WI and at additional mailing offi ces.

Postmaster: Send address change to SSSA Journal, 677 S. Segoe Rd., Madison, WI 53711.

Subscription rates (nonmember): $635 per year, within the USA; all oth-ers $683. Single copies, $100 USA; elsewhere, $115. Members are eligible for reduced subscription rates. New subscriptions, renewals, and new memberships that include the SSSA Journal begin with the fi rst issue of the current year. Claims for copies lost in the mail must be received with 90 days of publication date for domestic subscribers, and within 26 weeks of publication date for foreign subscribers.

Membership in the Society is not a requirement for publication in SSSA Journal; however, nonmembers will be charged an additional amount to publish the manuscript. To qualify for member rates, the corresponding author must be an active, emeritus, graduate student, or undergraduate student member of SSSA beginning in 2008.

Volunteered papers submitted in 2008 and thereafter will be assessed a charge of $80 per printed page. In addition to the total page charges, nonmem-bers will be assessed a $100 fee per paper. No charge will be assessed against com-ments, letters to the editor, or book reviews. An open access option is available for an additional $800.

Contributions to the SSSA Journal may be (i) papers and notes on original research; (ii) soil science issues papers and review papers; and (iii) “Comments and Letters to the Editor” containing (a) critical comments on papers published in one of the Society outlets or elsewhere, (b) editorial comments on papers pub-lished by Society offi cers, or (c) personal comments on matters having to do with soil science. Letters to the Editor are limited to one printed page. Contributions need not have been presented at annual meetings. Original research fi ndings are interpreted to mean the outcome of scholarly inquiry, investigation, or experi-mentation having as an objective the revision of existing concepts, the develop-ment of new concepts, or the improvement of techniques in some phase of soil science. Refer to SSSA Publication Policy (Soil Sci. Soc. Am. J. 72(1):280–282, 2000) and to the Publications Handbook and Style Manual (ASA-CSSA-SSSA, 2004; available online at https://www.soils.org/publications/style/).

Keep authors anonymous from reviewers by listing title, author(s), author-paper documentation, and acknowledgements on a detachable title page. Repeat manuscript title on the abstract page.

Manuscripts are to be submitted electronically via ManuscriptTracker (http://www.manuscripttracker.com/ver2/default.asp?journal=sssaj). Please sub-mit submission questions to Dr. David D. Myrold, Editor, SSSA Journal, (email: david.myrold@oregonstate.edu). All other correspondence should be directed to the Managing Editor, 677 S. Segoe Rd., Madison, WI 53711.

Opinions and conclusions expressed by authors are their own and are not considered as those of the Soil Science Society of America, its staff, its mem-bers, or its advertisers. Advertisements in this publication do not constitute an endorsement of the product. Trade names are sometimes mentioned in papers published in this journal. No endorsement of those products by the publisher is intended, nor is any criticism implied of similar products not mentioned.

The Soil Science Society of America considers the printed version of Soil Science Society of America Journal as the offi cial version of record.

Copyright © 2008 by the Soil Science Society of America, Inc. All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, record-ing, or any information storage and retrieval system, without permission in writ-ing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

The cover: A 2-mm fi eld of view showing radiating columnar and needlelike crystals of calcic clinoamphibole in a thin section by polarized light microscopy. This mineral, tremolite, is an example of an accessory calcium bearing mineral in serpentinite parent material that contributes to variability in the total and extractable Ca/Mg of soils on serpentinite landscapes. Please see pages 837–846, “Tectonic Inclusions in Serpentinite Landscapes Contribute Plant Nutrient Calcium” by Donald G. McGahan, Randal J. Southard, and Victor P. Claassen.

838 SSSAJ: Volume 72: Number 3 • May–June 2008

Soil Sci. Soc. Am. J. 72:838–847doi:10.2136/sssaj2007.0159Received 1 May 2007.*Corresponding author (soilman@ucdavis.edu).© Soil Science Society of America677 S. Segoe Rd. Madison WI 53711 USAAll rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

Serpentinized ultramafi c lithologies give rise to botani-cally distinct ecosystems (Walker, 1948). Debates have

abounded regarding edaphic infl uences for this effect, but the Ca/Mg ratio has been used to express the high Mg and low Ca condition that exists in these soils. Early researchers attrib-uted the poor growth of vegetation on serpentine soils to Mg toxicity (Loew and May, 1901; Gordon and Lipman, 1926). Vlamis and Jenny (1948) demonstrated that Ca defi ciency, rather than Mg toxicity, was the primary cause of poor plant growth. Walker et al. (1955) demonstrated that non-native species yields were reduced when the exchangeable Ca was 20% or less of the cation exchange capacity, and had little or no growth or yield below 10% exchangeable Ca. Native plant species, however, were better able to extract Ca; yield was only reduced 24% in the 5 to 3% exchangeable Ca range, whereas non-native plant species had a 90% yield reduction.

Soils dominated by Mg silicates, with a potential Ca defi -ciency and its resulting effects on plant growth in agricultural or native settings, are recognized in Soil Taxonomy at the fam-

ily mineralogical class level as “magnesic” (Soil Survey Staff, 1999). Identifi cation of Mg silicates requires XRD and PLM. Still, total or extractable Ca/Mg has been a sometimes useful indicator of serpentinite-derived soils. An NH4OAc-extractable Ca/Mg of 0.7 or greater is generally desired for agricultural crop production (Alexander et al., 1985; Brooks, 1987).

Serpentinite is a product of the low temperature and pres-sure metamorphism and metasomatism of ultramafi c rocks (e.g., peridotite). Metasomatism is a process of indefi nite replacement, loss, or addition of elements as a result of perco-lating solutions (Merrill, 1906). As a result, during metamor-phism or metasomatism, the anhydrous peridotite minerals become more hydrous and Ca content decreases relative to the original rocks, resulting in a relative enrichment of Mg (Page, 1966, 1967; Coleman and Keith, 1971; Alexander et al., 2007). Olivine is the dominant mineral in peridotite, the idealized ser-pentinite precursor, and is readily altered to serpentine during metasomatism. Serpentinous pseudomorphs after the peridot-ite chain (pyroxene and amphibole) and layer (talc) silicates are commonly identifi ed in thin section studies of serpentinized rocks. These pseudomorphs of chain and layer silicates in ser-pentinites are called bastites (Merrill, 1906; Wicks and Whittaker, 1977; O’Hanley, 1996). Wicks and Whittaker (1977) asserted that bastites should be considered a textural rather than a min-eralogical term because once serpentinization is complete and they become serpentine, it is often impossible to distinguish a pyroxene bastite from an amphibole bastite. If distinguishable, however, bastites after clinopyroxene or calcic-clinoamphibole may indicate that, before metasomatism, the peridotite con-tained more Ca than if the bastites identifi ed in the serpentinite were after orthopyroxene or orthoamphibole. The pyroxene and

Donald G. McGahan*Randal J. SouthardVictor P. ClaassenUniv. of CaliforniaDavis, CA 95616

SOIL

MIN

ERA

LOG

Y

Tectonic Inclusions in Serpentinite Landscapes Contribute Plant Nutrient Calcium

Serpentinite-derived soils give rise to botanically distinct systems primarily because of inad-equate parent material Ca content. We hypothesized that Ca content varies widely in what have been mapped as soils derived from serpentinite. An exchangeable Ca/Mg ratio <0.7 is often used to relate the imbalance of these nutrient elements in serpentinite-derived soils. We sampled six parent materials and soils from the Coast Ranges of California in Henneke soil series (clayey-skeletal, magnesic, thermic Lithic Argixerolls) modal location map unit poly-gons. Parent material total CaO content varied from 1.0 to 230 mg kg−1, and CaO/MgO varied from <0.1 to 4. A combination of x-ray diffraction (XRD), polarized light microscopy (PLM), and electron microscopy was used to identify the Ca-bearing accessory minerals diopside, grossularite, andradite, and tremolite. Accessory mineral content was often too low to be detected by XRD or minerals were too fi nely disseminated and diffi cult to detect in thin section by PLM. Electron microscopy, in concert with XRD and PLM, was needed to fully characterize the mineral assemblage. Two sites, Napa and Tehama, contained no serpentine minerals, were not serpentinites, and were tectonic inclusions in the serpentinite landscape. Napa rocks contained almost no Ca-bearing minerals and would be identifi ed as a serpentinite if relying on elemental analysis CaO/MgO ratio alone. Tectonic inclusions and Ca-bearing accessory minerals affect Ca distribution and presumably its availability for plants. Careful mineralogical analysis may be required to identify Ca-bearing accessory minerals.

Abbreviations: BSE, backscatter electron microscopy; CL-VR, interstratifi ed chlorite–vermiculite; EDX, energy dispersive x-ray spectroscopy; PLM, polarized light microscopy; XRD, x-ray diffraction.

SSSAJ: Volume 72: Number 3 • May –June 2008 839

amphibole in a peridotite are less susceptible to serpentiniza-tion than olivine, and it is feasible that not all the Ca in the clinopyroxene or clinoamphibole was completely removed from the lithologic unit during metasomatism to serpen-tinite. Together with metamorphic secondary minerals such as garnets that may contain Ca, it is reasonable to assume that the Ca content of soils on serpentinitic landscape is variable and, therefore, infl uences plant nutrient supplying status as a result of serpentinite weathering.

Less mafi c lithic inclusions in serpentinite landscapes can infl uence the total and extractable Ca/Mg contents of the bulk parent material and the soil. Rabenhorst and Foss (1981) attempted to predict mafi c or ultramafi c par-ent lithology while mapping soils of the eastern Piedmont of Maryland. Based on their study of 39 samples, if the exchangeable Ca/Mg ratio ranged from 0.0 to 0.1, the prob-ability that the soil was formed from serpentinite rather than a mafi c lithology was 98%. An exchangeable Ca/Mg ratio of 0.2 to 0.3 yielded a probability of 53%, whereas a 0.6 to 0.7 exchangeable Ca/Mg ratio reduced the probability to 21%. Clearly at exchangeable Ca/Mg ratios >0.3, the likelihood that the soil is derived solely from ultramafi c parent material is low.

A basic question is “Do some lithologies underlying ser-pentinitic landscapes contain Ca-bearing minerals that can sig-nifi cantly alter extractable Ca/Mg ratios in the soils weathered from the serpentinites?” We studied the variation in parent material (nominally serpentinite) mineralogy beneath magne-sic pedons from the Coast Ranges of California. This focus on bulk mineralogy and Ca-bearing accessory mineralogy of the parent material is necessary as a requisite to follow-up pedo-logic studies at these sites. We expected to fi nd a variation in accessory minerals, specifi cally Ca-bearing minerals, in these serpentinite parent materials and that these accessory minerals

would have a profound infl uence on the total and extractable Ca/Mg ratios in the soils weathered from the serpentinites.

MATERIALS AND METHODSField

Sampling locations were from six California soil survey areas within Henneke soil series (clayey-skeletal, magnesic, thermic Lithic Argixerolls) map units (Fig. 1). These are “soils formed in material weathered from serpentinite and rocks of similar mineralogy” (Soil Survey Staff, 2005). We sampled pedons near the location of the modal pedon for the Henneke series for each soil survey. To minimize the impacts of colluviation, we sampled pedons on summits above the modal pedon location and within the polygon containing the modal location. Pits were excavated by hand tools. Soils were described and sampled by horizon using conventional procedures, and rocks samples were collected within the Cr or R layers (Soil Survey Division Staff, 1993).

Woody vegetation species composition and percentage of cover at the sites is presented in Table 1.

Fig. 1. Pedon sampling locations from Henneke soil series (clayey-skeletal, magnesic, thermic Lithic Argixerolls) map units within six California soil survey areas.

Table 1. Woody vegetation coverage at sites on serpentinite-derived landscapes.

Site Coverage†%

Tehama County Garrya congdonii Eastw. 4 Quercus durata Jepson 37 Eriodictyon californicum (Hook. & Arn.) Torr. 4 Ceanothus cuneatus (Hook.) Nutt. 1 Ceanothus jepsonii Green var. jepsonii 8 Coverage 54Kings County Quercus durata Jepson 1 Quercus douglasii Hook. & Arn. 1 Ceanothus cuneatus (Hook.) Nutt. <1 Cupressus macnabiana A. Murray <1 Pinus sabiniana Douglas ex Douglas 6 Eriodictyon californicum (Hook. & Arn.) Torr. 1 Juniperus californica Carrière <1 Adenostoma fasciculatum Hook. & Arn. <1 Baccharis pilularis DC. <1 Rhamnus californica Eschsch. <1 Coverage 10Shasta County Eriodictyon californicum (Hook. & Arn.) Torr. 7 Coverage 7Glenn County Garrya congdonii Eastw. 4 Quercus durata Jepson 35 Eriodictyon californicum (Hook. & Arn.) Torr. 2 Ceanothus jepsonii Green var. jepsonii 3 Zigadenus fremontii (Torr.) Torr. ex S. Watson <1 Arctostaphylos viscida Parry ssp. pulchella (Howell) P.V. Wells 12 Pinus sabiniana Douglas ex Douglas <1 Coverage 56Napa County Adenostoma fasciculatum Hook. & Arn. 60 Ceanothus cuneatus (Hook.) Nutt. 15 Pinus sabiniana Douglas ex Douglas 5 Coverage 80San Benito County Arctostaphylos viscida Parry ssp. pulchella (Howell) P.V. Wells 7 Quercus durata Jepson 1 Adenostoma fasciculatum Hook. & Arn. 2 Pinus sabiniana Douglas ex Douglas 1 Coverage 11† Percentage of cover determined by line intercept (Bonham, 1989).

840 SSSAJ: Volume 72: Number 3 • May–June 2008

LaboratoryThree rocks from each parent material were ground with an agate

mortar and pestle to pass a 0.106-mm sieve. The three rock powders for each parent material were mixed, ground for 10 min with an agate mortar and pestle (Jackson, 2005), mounted on porous ceramic tiles, washed with MgCl2 or KCl salt solutions, rinsed with deionized water to remove excess salts, and then reoriented by smoothing with a glass slide held at an angle to the sample (Whittig and Allardice, 1986). X-ray analyses were made with a Diano XRD 8000 diffractometer (Diano Corp., Woburn, MA) fi tted with a nickel fi lter and curved graphite monochromator to produce Cu Kα radiation. After the ini-tial diffraction analysis, the MgCl2-treated samples were treated with glycerol and reanalyzed. The KCl samples were reanalyzed after 350 and 550°C heat treatments (Whittig and Allardice, 1986). Major oxide content of the rock powders was determined by inductively coupled plasma emission spectrometry following a LiBO2 fusion and dilute HNO3 digestion (Sawhney and Stilwell, 1994).

Three rock samples from each R or Cr layer were impregnated with Petropoxy-154 resin (Palouse Petro Products, Palouse, WA). Thin sections of these samples were prepared and examined with a polarizing light microscope (Drees and Ransom, 1994; Stoops, 2003). Selected thin sections were polished and analyzed using backscatter elec-tron microscopy (BSE) and energy dispersive x-ray spectroscopy (EDX) on a Cameca SX-100 electron probe microanalyzer (Cameca, Paris).

Selected soil properties are presented in Table 2. Soils were air dried, sieved to pass through a 2-mm sieve, and the fi ne earth fraction was ana-lyzed for particle size distribution by the pipette method as described by Gee and Bauder (1986). Fine-earth-fraction extractable cations were measured by displacement with pH 7.0, 1 mol L−1 NH4OAc; concen-trations of Na, Ca, Mg, and K in the leachate were measured with fl ame atomic absorption or emission spectrometry. Cation exchange capacity (CEC-7) was measured by saturating samples with 1 mol L−1 NH4OAc, at pH 7, washing with 95% ethanol, extracting with 2 mol L−1 KCl to

remove adsorbed NH4 (Soil Survey Staff, 2004), and determining the NH4 concentration in the leachate conductimetrically (Carlson, 1978). The pH of the fi ne earth fraction was determined at 1:1 soil/water on a mass basis (Soil Survey Staff, 2004).

RESULTS AND DISCUSSIONNon-serpentinites

Napa and Tehama County parent materials were not domi-nated by serpentine minerals and, therefore, not serpentinites. The Tehama and Napa County parent materials contained clinopyrox-ene but were quite different in the composition and proportions of the remainder of the parent material mineral suite.

Tehama CountyThe Tehama County parent material contained garnet

(0.299- and 0.266-nm peaks), clinopyroxene (0.32-nm peak), interstratifi ed chlorite–vermiculite (CL-VR) (1.4-nm peak), and pumpellyite (0.29-nm peak), but no serpentine. The CL-VR was characterized by a 1.4-nm Mg-saturated peak that collapsed with K saturation and heating treatments, leaving a plateau of peaks between 1.4 and 1.0 nm (Fig. 2A).

Garnets have the idealized formula of X3Y2(SiO4)3 and are further divided into the pyralspite and ugrandite groups where Y is Al in pyralspite, X is Ca in the ugrandite group, and X is not Ca in the pyralspite group (Table 3). Energy-dispersive x-ray spectroscopy of the Tehama County garnet corroborated XRD results indicating that the garnet was grossularite of the ugrandite group (Fig. 3A).

In thin sections with plane-polarized light at lower mag-nifi cations, the small euhedral interlocking grains of grossular-ite were nearly indistinguishable and formed large cinnamon-brown, somewhat granular masses. The masses were isotropic under crossed polarizers (Fig. 2B).

Table 2. Selected chemical properties, particle size distribution, and colors of soils from serpentinite-derived landscapes.†

Horizon Depth K‡ Mg‡ Ca‡ CEC-7§ Ca/Mg‡ pH Sand Silt Clay Dry color Moist colorcm ————– cmolc kg−1 ————— ————–%————–

Tehama CountyA 0–3 0.4 12.4 16.3 40.2 1.31 6.4 57 19 24 7.5YR 4/3 5YR 3/3Bt1 3–7 0.5 20.2 16.5 32.7 0.82 6.3 55 20 25 7.5YR 4/4 5YR 3/2Bt2 7–16 0.2 27.7 13.7 30.0 0.50 6.1 51 17 32 7.5YR 3/3 7.5YR 3/3

Kings CountyA 0–4 0.2 12.1 1.4 23.7 0.58 6.1 43 55 3 10YR 4/2 10YR 2/2Bt1 4–13 0.6 12.1 7.0 18.0 0.51 6.2 36 49 15 10YR 5/3 7.5YR 3/2Bt2 13–40 0.8 17.7 4.4 20.0 0.25 6.2 27 45 27 10YR 4/2 7.5YR 3/2

Shasta CountyA 0–3 0.3 9.7 3.5 13.2 0.36 5.7 37 49 14 5YR 4/6 2.5YR 3/4Bt 3–40 0.2 10.1 2.0 12.2 0.20 6.0 39 46 14 5YR 4/6 2.5YR 3/4

Glenn CountyA 0–9 0.2 21.9 4.4 18.8 0.20 6.3 54 20 25 7.5YR 4/4 2.5YR 3/3Bt 9–34 0.3 18.9 4.5 27.5 0.20 6.3 49 18 34 5YR 4/4 2.5YR 2.5/3

Napa CountyA 0–6 0.4 16.4 24.1 31.1 1.47 5.8 43 42 15 7.5YR 4/4 5YR 3/3Bt 6–20 0.1 16.2 21.5 27.9 1.32 5.7 54 29 17 5YR 4/6 5YR 3/3

San Benito CountyA 0–8 <0.1 3.0 0.7 3.1 0.24 6.6 80 8 12 10YR 6/3 10YR 4/3Bt 8–20 0.0 0.1 5.6 5.6 0.11 6.7 74 7 19 10YR 5/3 7.5YR 3/1

† Na was measured but was 0.1 cmolc kg−1 or less in all cases except the San Benito County Bt horizon (0.6 cmolc kg−1).

‡ Extracted by 1 mol L−1 NH4OAc at pH 7.

§ Cation exchange capacity at pH 7 may underestimate the actual CEC because both the index cation (NH4+) and the displacing cation (K+) can

cause collapse of vermiculite and hydroxy-interlayered materials, and these were present in most horizons.

SSSAJ: Volume 72: Number 3 • May –June 2008 841

Pyroxene is a common inclusion in ultramafi c serpen-tinite protoliths and is more resistant to alteration than is olivine, the most common mineral in the ultramafi c proto-lith (Goldich, 1938; Huang, 1989). The idealized formula for pyroxene is XYSi2O6. Clinopyroxene is distinguished opti-cally from orthopyroxene by inclined, rather than parallel,

extinction and by its higher interference colors as viewed in cross-polarized light. For orthopyroxene, X is Mg or Fe, but for clinopyroxene X is Ca, Na, or Li and therefore a potential Ca source on weathering (Table 4).

The clinopyroxene diopside was identifi ed in thin sections by inclined extinction. The clinopyroxene was determined by EDX to be a member of the diopside–hedenbergite series with considerably greater Ca than Fe and therefore may be referred to as diopside (Fig. 3B).

Pumpellyite, a sorosilicate [idealized formula Ca2MgAl2(SiO4)(Si2O7)(OH)2·(H2O)] could be easily missed in the x-ray diffractogram and was not identifi ed in thin sec-tion, but EDX confi rmed its presence (Fig. 3C). The BSE micrograph clearly showed that it was intergrown with the CL-VR and chemistry was confi rmed by EDX (Fig. 3F).

Among the six parent materials, the Ti and Al contents were highest in the Tehama County parent material, while Si content was the lowest (Table 5). The Ti was a trace element in grossularite (Fig. 3A) but a major element in ilmenite and titanite (Fig. 3D and 3E). Titanite (sphene) was fi nely dissemi-nated and associated with the diopside (Fig. 2C).

Ugrandite garnets are characteristic of rodingites (Coleman, 1979, 1980). Rodingites are a massive, dense, buff-to-pink rock, typically rich in grossular garnet and calcic pyroxene, and enveloped in serpentinite. They are the result of localized metasomatism of non-peridotites in association with the metasomatic alteration of peridotite to serpentinite. These rodingites can be tectonic inclusions within the serpentinite mass (endogenous) or can occur at metasomatic contacts with the country rock (exogenous). Tectonic inclusions within ser-pentinite bodies could be from mafi c dikes in the peridotite or country rock inclusions during emplacement of the more plas-tic serpentinite. The replacement or invasion by calcisilicate minerals, such as Ca-garnets, into the protomineralogy of the tectonic inclusions is coupled with the Ca lost from the perido-tite during metasomatism. Chlorite and calcic clinopyroxenes are commonly associated with ugrandite garnets in rodingites (Coleman, 1979, 1980). This mineral suite fi ts well with the observed mineralogy of the Tehama County parent material, which was dominated by grossularite and also contained calcic clinopyroxene. The CL-VR in the parent materials is probably the result of interlayer stripping from a precursor chlorite in the hard rock underlying the Cr layer we analyzed.

Rodingites are not serpentinites but probably occur fre-quently enough in serpentinitic landscapes to supply signifi -cant amounts of Ca to the soil solution on weathering.

Fig. 2. Mineralogical properties of Tehama County parent material: (A) x-ray diffractograms of interstratifi ed chlorite–vermiculite, grossularite, diopside, and pumpellyite; (B) photomicrograph under crossed polarizers—the garnets (GN) are isotropic dark purplish black tightly packed small interlocking euhedral crys-tals and the ilmenite (IL) is brown; (C) backscatter electron (BSE) micrograph of diopside (Di), titanite (Tt), ilmenite (IL), and gar-net (grossularite, GN); and (D) BSE micrograph of lighter gray pumpellyite (PU) intergrown into darker gray interstratifi ed chlorite–vermiculite (CL-VR) upper left and center, with smaller masses of euhedral grossularite garnet crystals (GN) lower cen-ter but predominantly tightly packed and interlocked into larger masses (lower right).

Table 3. Chemical composition of garnets (reproduced after Hurlbut and Klein, 1977).

Mineral Chemical formula

Pyralspite group Pyrope Mg3Al2Si3O12 Almandite Fe3Al2Si3O12 Spessartite Mn3Al2Si3O12Ugrandite group

Grossularite Ca3Al2Si3O12 Andradite Ca3Fe2Si3O12 Uvarovite Ca3Cr2Si3O12

Hydrogrossularite Ca3Al2Si3O8(SiO4)1−m(OH)4m; m = 0–1

842 SSSAJ: Volume 72: Number 3 • May–June 2008

Napa CountyBased on XRD, the Napa County parent material was

dominantly vermiculite (1.4-nm peak with Mg treatment col-lapsed to 1.0 nm with K treatment and heat) and subdomi-nant plagioclase feldspar (0.630-, 0.374-, 0.365-, 0.318-, and 0.293-nm peaks) (Fig. 4A). Energy-dispersive x-ray spectros-copy confi rmed the plagioclase, and elemental ratios suggested that the plagioclase was toward the albite end of the plagioclase solid-state solution (Fig. 4C).

The mineralogy, based on XRD peak intensity, suggested a greater amount of plagioclase than would be expected from the elemental analysis (Table 5). This is probably due to attenuated vermiculite peaks due to interstratifi cation by chlorite and its presumed precursor mica in amounts too small to be detected by XRD (Moore and Reynolds, 1997).

A pyroxene XRD peak was not evident, but a pyroxene with inclined extinction was clearly identifi ed by PLM. The clinopyroxene, like the clinopyroxene in the Tehama County parent material, contained more Ca than Fe and had a small amount of Al (Fig. 4D). The vermiculite contained more Al and less O (Fig. 4E) than the Tehama County CL-VR (Fig. 3F). The dominant source of Ca was probably the clinopyrox-ene (diopside) identifi ed by PLM and electron microscopy (Fig. 4C). This mineral was not identifi ed by XRD and therefore was apparently only a minor component of the parent mate-rial (Fig. 4A). Low abundance of diopside, together with high vermiculite content, was probably responsible for a CaO/MgO < 0.1 (Table 5). Napa County parent material contained no detectable serpentine, so was not a serpentinite. It did not have Ca-garnets and was not a rodingite. It was an example of a

Fig. 3. Energy dispersive x-ray spectra (EDX) of Tehama County parent material minerals: (A) grossularite demonstrating some inclusion of Ti; (B) diopside; (C) pumpellyite; (D) titanite; (E) ilmenite; and (F) chlorite–vermiculite.

Table 4. Chemical composition of pyroxenes and amphiboles (reproduced after Hurlbut and Klein, 1977).

Mineral Chemical formula W X Y Example

Orthopyroxene XYSi2O6 Mg, Fe Mg, Fe EnstatiteClinopyroxene XYSi2O6 Ca, Na, Li Mg, Fe, Al Diopside

Orthoamphibole W0–1X2Y5Si8O22(OH)2 Mg, Fe Mg, Fe, Al Anthophyllite

Clinoamphibole W0–1X2Y5Si8O22(OH)2 Na, K Ca, Na Mg, Fe, Al, Ti Tremolite

SSSAJ: Volume 72: Number 3 • May –June 2008 843

mafi c tectonic inclusion in the serpentinitic terrain. Weathering of this parent material resulted in soil Ca enrichment (Table 2), probably as a result of rapid weathering of the plagioclase feldspar and biocycling of the Ca as the diopside weathered (McGahan, 2007).

SerpentinitesShasta County

Shasta County parent material was dominated by ser-pentine (XRD peaks at 0.724, 0.455, and 0.362 nm) and talc (0.93-, 0.466-, and 0.31-nm peaks) with a subdominant com-ponent of chlorite (persistent 1.425-nm peak and 0.475-, 0.71-, 0.355-, and 0.284-nm peaks) (Fig. 5A). Not identifi ed by XRD but clearly distinguishable in thin sec-tions by PLM and BSE were masses of radiating columnar or needlelike crystals of calcic clinoamphibole, probably tremolite, as confi rmed by EDX (Fig. 5B).

Tremolite contains Ca, and yet the CaO content of the parent mate-rial was very low (Table 5); therefore, the tremolite was probably not an abundant accessory component of the rock.

Glenn CountyGlenn County parent mate-

rial was dominated by serpen-tine (0.731-, 0.457-, 0.363-, and 0.250-nm peaks) with a trace of magnetite (0.253-nm peak) (Fig. 6A). Enstatite was not identifi ed by XRD but was clearly identifi ed by PLM by its parallel extinction. Energy-dispersive x-ray spectros-copy confi rmed that it contained no Ca (Fig. 6B).

The CaO content of the Glenn County parent material was simi-

lar to the CaO content of the Shasta County parent material (Table 5), and no Ca-containing minerals were identifi ed in the Glenn County parent materials.

Kings CountyKings County parent material was dominated by serpen-

tine (0.724, 0.455, 0.363, and 0.250 nm) (Fig. 7A). There were also traces of andradite garnet (0.302, 0.271, and 0.246 nm) and magnetite (0.297- and 0.253-nm XRD peaks) detected by XRD (Fig. 7A). The andradite XRD peaks were very weak and could easily be missed. Polarized light microscopy clearly showed garnet interspersed with magnetite among the serpen-

Table 5. Content of selected elements in “serpentinite” parent material rocks beneath the six pedons.

Element Tehama County Kings County Shasta County Glenn County Napa County San Benito County

Depth, cm 16–60 40–50 40–50 34–55 20–52 50–70——————————————————————————- g kg−1——————————————————————————-

SiO2 363 498 423 387 390 394Al2O3 151 11 9.0 3.8 23 5.4Fe2O3 84 95 84 110 101 91MgO 57 248 331 325 319 352CaO 229 3.2 1.0 1.4 0.2 4.4Na2O <0.1 0.1 0.3 0.2 0.1 <0.1K2O <0.2 0.3 <0.2 <0.2 <0.2 <0.2TiO2 15 0.2 0.2 <0.1 0.2 <0.1P2O5 2.6 0.5 0.6 0.6 0.8 0.4MnO 1.3 1.3 0.8 1.5 1.4 1.0Cr2O3 0.3 4.5 3.7 4.5 5.3 4.2CaO/MgO 4.0 < 0.1 <0.1 <0.1 <0.1 <0.1LOI† 95 135 142 160 149 134† Loss on ignition.

Fig. 4. Mineralogical properties of Napa County parent material: (A) x-ray diffractograms of vermiculite and albite plagioclase; (B) backscatter electron micrograph of Napa County parent material (Di = diopside, P = plagioclase feldspar, VR = vermiculite); (C) energy dispersive x-ray spectra (EDX) of Napa County parent material plagioclase; (D) EDX of Napa County parent material diopside; and (E) EDX of Napa County parent material vermiculite.

844 SSSAJ: Volume 72: Number 3 • May–June 2008

tine (Fig. 7C). The Fe content of andradite increases backscat-tered electron fl uorescence and therefore the andradite may easily be mistaken for magnetite in BSE images (Fig. 7D). Andradite is easily distinguished from magnetite by PLM, EDX analysis, or by adjusting the contrast of the BSE image.

The serpentine EDX was representative of the serpentine in all of the parent materials (Fig. 8B). Talc [idealized formula Mg3Si4O10(OH)2] was not identifi ed by XRD but could be

seen in thin section by PLM (Fig. 7C) and in BSE images (Fig. 7B). The talc chemistry was confi rmed by EDX (Fig. 8A). Magnetite chemistry was also confi rmed by EDX (Fig. 8D). The garnet was confi rmed as andradite by EDX and contained more Fe than gros-sularite (Fig. 8C and 3A, Table 3). Unlike the grossularite in the Tehama County parent mate-rial, the andradite garnet in Kings County occurred as individual crys-tals or in small clusters in the dom-inantly serpentine Kings County parent material (Fig. 7C).

Talc does not contribute to soil Ca (Fig. 8A) nor does serpen-tine (Fig. 8B), but the andradite could contribute to the soil solu-tion Ca on weathering (Fig. 8C). The total CaO content was far less (3.2 mg kg−1) in the Kings County parent material than in the Tehama County parent material (229 mg kg−1), and the CaO/MgO ratio was markedly lower (<0.1) than for the Tehama County par-ent material (4.0) (Table 5).

Fig. 5. Mineralogical properties of Shasta County parent material: (A) x-ray diffractograms of serpentine, talc, and chlorite; and (B) energy dispersive x-ray spectra of calcic clinoamphibole (tremolite). Fig. 6. Mineralogical properties of Glenn County parent material:

(A) x-ray diffractograms of serpentine and magnetite; and (B) energy dispersive x-ray spectra of enstatite.

Fig. 7. Mineralogical properties of Kings County parent material: (A) x-ray diffractograms of serpentine, andradite, and magnetite; (B) backscatter electron (BSE) micrograph of talc (TA) among serpen-tine (SY) of varying shades of gray—bright white spots are either magnetite or andradite garnet; (C) photomicrograph shown in plane-polarized light of a mass of talc (TA), dark magnetite (MG), and individual dodecahedral crystals of andradite garnet (GN) in a serpentine (SY) matrix; and (D) BSE micrograph with bright white magnetite and andradite garnet among serpentine.

SSSAJ: Volume 72: Number 3 • May –June 2008 845

San Benito CountySan Benito County parent material was dominated by ser-

pentine (0.73-, 0.45-, 0.36-, and 0.27-nm peaks) (Fig. 9A). No accessory pyroxenes or amphiboles were identifi ed. Polarized light microscopy clearly identifi ed opaque inclusions (Fig. 9B). The opaque minerals seen in PLM were resolved by BSE (Fig. 9C) and EDX to be magnetite (Fig. 10A), chromite spinel (Fig. 10B), and Cr-rich andradite garnet (Fig. 10C). The andradite contained some Ca and probably contributed to the parent material CaO content of 14 mg kg−1 (Table 5, Fig. 10C).

Implications for Calcium Supply and Plant GrowthWe asked a most basic question: How do you know if a

soil is derived from serpentinite? The simple answer is to char-acterize the parent material and ensure that it is dominated by serpentine minerals. If a parent material is dominated by serpentine minerals and is therefore a serpentinite, could it

Fig. 8. Energy dispersive x-ray spectra of Kings County parent mate-rial minerals: (A) talc; (B) serpentine; (C) andradite garnet; and (D) magnetite.

Fig. 9. Mineralogical properties of San Benito County parent material: (A) x-ray diffractogram of serpentine; (B) photomicrograph un-der plane-polarized light of serpentine (SY) with opaque inclu-sions (15-mm-wide fi eld of view); and (C) backscatter electron micrograph increased contrast and lowered brightness elimi-nates outlines of serpentine (SY) crystals that dominate the par-ent material but distinguishes between chromite spinel (SP) at center surrounded by chrome-rich andradite garnet (GN), while lightest gray is magnetite (MT).

846 SSSAJ: Volume 72: Number 3 • May–June 2008

still

contain Ca-bearing minerals? Do accessory minerals infl uence the Ca/Mg ratio?

Elemental analysis clearly demonstrated that Ca content was variable on nominally serpentinite landscapes. The parent material of two out of six pedons contained no serpentine minerals. One parent material, from Tehama County, had abundant Ca-bearing minerals, and the other, from Napa County, contained very few Ca-bearing minerals.

It is clear that reliance solely on CaO/MgO from parent material analysis would have incorrectly identifi ed the Napa County parent material as serpentinite. It was not a serpen-tinite and despite the low parent material total CaO/MgO (<0.01), the soils derived from it had relatively high extractable Ca/Mg (Table 2).

The Tehama County parent material total CaO/MgO (4.0) (Table 5) readily indicated a non-serpentinite parent material. The soil extractable Ca/Mg was more favorable to vegetation and increased with proximity to the surface (Table 2).

Further complicating interpretation of the total CaO/MgO is the fact that serpentinite parent materials can contain Ca-bearing minerals. We could not identify Ca-bearing trace accessory minerals in many of the parent materials using XRD. Kings, Shasta, and San Benito County parent materials were

examples of serpentinites that had no obvious accessory miner-als, other than magnetite, as detected by XRD. They did have trace accessory minerals detected by PLM or BSE and EDX. Soil extractable Ca/Mg ratios at these sites increased with proximity to the surface (Table 2). Kings County soil extract-able Ca/Mg approached 0.6 in the A horizon, a ratio that sug-gests a low probability of serpentinite (Rabenhorst and Foss, 1981), whereas this ratio in the A horizons of the San Benito and Glenn County pedons was 0.2. The variability of the soil extractable Ca/Mg in pedons clearly derived from serpentinite underscores the infl uence of Ca-bearing accessory minerals on the variation of Ca as a potential plant nutrient. Our results suggest that total or soil extractable Ca/Mg measured on sur-face horizon grab samples is an unreliable predictor of the pres-ence of serpentine in nominal serpentinitic landscapes.

Weathering of these parent materials, like the non-serpen-tinite tectonic inclusions such as rodingites (Tehama County) and mafi cs (Napa County), could result in downslope Ca enrichment. Their impact by soil solution or sediment trans-port could be further varied based on downslope dynamics, e.g., water-gathering vs. water-spreading landscape positions.

With other supporting analyses such as PLM and elec-tron microscopy, we identifi ed the minerals contributing Ca to the soils: clinopyroxene (diopside), ugrandite garnets (gros-sularite and andradite), and calcic clinoamphibole (tremolite). Ugrandite garnets were a common Ca source in serpentinites and rodingites. Identifi cation of these Ca-bearing accessory minerals by a screening process could be especially benefi cial to researchers investigating “serpentine soil” to avoid anomalous results arising from tectonic inclusions on serpentinitic land-scapes or calcic accessory minerals in serpentinites. Tectonic inclusions in the landscape and, to a lesser extent, Ca-bearing accessory minerals have the potential of acting as “landscape fertilizers” of Ca. Landscape managers or revegetation efforts may also benefi t from identifi cation of site sources of Ca, as such identifi cation may help to adjust practices of managing amendment application rates.

REFERENCESAlexander, E.B., R.G. Coleman, T. Keeler-Wolf, and S.P. Harrison. 2007. Serpentine

geoecology of western North America. Oxford Univ. Press, New York.Alexander, E.B., W.E. Wildman, and W.C. Lynn. 1985. Ultramafi c

(serpentinitic) mineral class. p. 135–146. In J.A. Kittrick (ed.) Mineral classifi cation of soils. SSSA Spec. Publ. 16. SSSA, Madison, WI.

Bonham, C.D. 1989. Measurements for terrestrial vegetation. John Wiley & Sons, New York.

Brooks, R.R. 1987. Serpentine and its vegetation: A multidisciplinary approach. Dioscorides Press, Portland, OR.

Carlson, R.M. 1978. Automated separation and conductimetric determination of ammonia and dissolved carbon dioxide. Anal. Chem. 50:1528–1531.

Coleman, R.G. 1979. Tectonic inclusions in serpentinites. p. 89–102. In J. Bertrand and J. Deferne (ed.) Proc. Symp. on Tectonic Inclusions and Associated Rocks in Serpentines. 17–29 Sept. 1979. Swiss Natl. Sci. Foundation, Geneva.

Coleman, R.G. 1980. Ophiolite tectonics and metamorphism. p. 167–184. In J. Aubouin and C.J. Allegre (ed.) Orogenic mafi c ultra mafi c association. Colloques internationaux du centre national de la recherche scientifi que 272, Grenoble. 6–11 June 1977. Centre Natl. de la Recherche Scientifi que, Paris.

Coleman, R.G., and T.C. Keith. 1971. A chemical study of serpentization: Burro Mountain, California. J. Petrol. 12:311–328.

Drees, L.R., and M.D. Ransom. 1994. Light microscope techniques in quantitative soil mineralogy. p. 137–176. In J.E. Amonette and L.W. Zelazny (ed.) Quantitative methods in soil mineralogy. SSSA, Madison, WI.

Gee, G.W., and J.W. Bauder. 1986. Particle size analysis. p. 383–411. In A.

Fig. 10. Energy dispersive x-ray spectra of San Benito County parent material: (A) chromite spinel; (B) chrome-rich andradite; and (C) magnetite.

SSSAJ: Volume 72: Number 3 • May –June 2008 847

Klute (ed.) Methods of soil analysis. Part 1. Physical and mineralogical methods. 2nd ed. Agron. Monogr. 9. ASA and SSSA, Madison, WI.

Goldich, S.S. 1938. A study of rock weathering. J. Geol. 46:17–58.Gordon, A., and C.B. Lipman. 1926. Why are serpentine and other magnesian

soils infertile? Soil Sci. 22:291–302.Huang, P.M. 1989. Feldspars, olivines, pyroxenes and amphiboles. p. 975–1050.

In J.B. Dixon and S.B. Weed (ed.) Minerals in soil environments. 2nd ed. SSSA Book Ser. 1. SSSA, Madison, WI.

Hurlbut, C.S., and C. Klein. 1977. Manual of mineralogy (after James D. Dana). 19th ed. John Wiley & Sons, New York.

Jackson, M.L. 2005. Soil chemical analysis: Advanced course. Rev. 2nd ed. Parallel Press, Madison, WI.

Loew, O., and D.W. May. 1901. The relation of lime and magnesia to plant growth: I. Liming of soils from a physiological standpoint; II. Experimental study of the relation of lime and magnesia to plant growth. USDA Bureau of Plant Industry Bull. 1. U.S. Gov. Print. Offi ce, Washington, DC.

McGahan, D.G. 2007. A survey of soils formed on serpentinitic landscapes in California. Ph.D. diss. Univ. of California, Davis (Diss. Abstr. 3261178).

Merrill, G.P. 1906. A treatise on rocks, rock-weathering and soils. Macmillan, New York.

Moore, D.M., and R.C. Reynolds. 1997. X-ray diffraction and the identifi cation and analysis of clay minerals. 2nd ed. Oxford Univ. Press, New York.

O’Hanley, D.S. 1996. Serpentinites: Records of tectonic and petrological history. Oxford Univ. Press, New York.

Page, N.J. 1966. Mineralogy and chemistry of the serpentine group minerals and the serpentinization process. Ph.D. diss. Univ. of California, Berkeley (Diss. Abstr. 6615461).

Page, N.J. 1967. Serpentinization considered as a constant volume metasomatic process: A discussion. Am. Mineral. 52:545–549.

Rabenhorst, M.C., and J.E. Foss. 1981. Soil and geologic mapping over mafi c and ultramafi c parent materials in Maryland. Soil Sci. Soc. Am. J. 45:1156–1160.

Sawhney, B.L., and D.E. Stilwell. 1994. Dissolution and elemental analysis of minerals, soils and environmental samples. p. 49–82. In J.E. Amonette and L.W. Zelazny (ed.) Quantitative methods in soil mineralogy. SSSA, Madison, WI.

Soil Survey Division Staff. 1993. Soil survey manual. Agric. Handbk. 18. U.S. Gov. Print. Offi ce, Washington, DC.

Soil Survey Staff. 1999. Soil Taxonomy: A basic system of soil classifi cation for making and interpreting soil surveys. Agric. Handbk. 463. 2nd ed. U.S. Gov. Print. Offi ce, Washington, DC.

Soil Survey Staff. 2004. Soil survey laboratory methods manual. Soil Surv. Invest. Rep. 42. Version 4.0. Available at soils.usda.gov/technical/lmm/ (accessed 20 May 2007; verifi ed 15 Mar. 2008). Natl. Soil Survey Ctr., Lincolin, NE.

Soil Survey Staff. 2005. Offi cial soil series descriptions. Available at soils.usda.gov/technical/classifi cation/osd/index.html (accessed 12 Apr. 2007; verifi ed 15 Mar. 2008). Natl. Soil Survey Ctr., Lincolin, NE.

Stoops, G. 2003. Guidelines for analysis and description of soil and regolith thin sections. SSSA, Madison, WI.

Vlamis, J., and H. Jenny. 1948. Calcium defi ciency in serpentine soils as revealed by absorbent technique. Science 107:549–551.

Walker, R.B. 1948. A study of serpentine soil infertility with special reference to edaphic endemism. Ph.D. diss. Univ. of California, Berkeley.

Walker, R.B., H.M. Walker, and P.R. Ashworth. 1955. Calcium–magnesium nutrition with special reference to serpentine soils. Plant Physiol. 30:214–221.

Whittig, L.D., and W.R. Allardice. 1986. X-ray diffraction techniques. p. 331–362. In A. Klute (ed.) Methods of soil analysis. Part 1. Physical and mineralogical methods. 2nd ed. Agron. Monogr. 9. ASA and SSSA, Madison, WI.

Wicks, F.J., and E.J. Whittaker. 1977. Serpentine textures and serpentinization. Can. Mineral. 15:459–488.