Crossosoma Journal of the Southern California Botanists, Inc. Volume 34, Number 1 Spring-Summer 2008 Southern California Botanists, Inc. — Founded 1927 — CROSSOSOMA (ISSN 0891-9100) is published twice a year by Southern California Botanists, Inc., a California nonprofit organization of individuals devoted to the study, preservation and conservation of the native plants and plant communities of southern California. SCB Board of Directors for 2008 President Vice President Secretary Treasurer Webmaster Editor of CROSSOSOMA Editor of Leaflets Directors-at-large Ex officio Board Members Gary Wallace Naomi Fraga Linda Prince Alan P. Romspert Naomi Fraga Denise Knapp Kerry Myers David Bramlet Sara Baguskas Terry Daubert Elizabeth Delk Michael Honer Charlie Hohn Carrie Kiel Diane Menuz Orlando Mistretta Bart O’Brien Fred Roberts Darren Sandquist Susan Schenk Allan A. Schoenherr Paul Schwartz Sula Vanderplank (Past President) Articles, book reviews, or other items for submission to CROSSOSOMA can be sent to the incoming editor Michael Honer, at mihoner@earthlink.net or 874 Fortuna Lane, Isla Vista, California 93117, USA. Electronic submission is preferred. Please see our website, www.socalbot.org, for format guidelines. Notices of a time-dated nature (field trips, workshops, symposia, etc.) to be included in the newsletter Leaflets should be submitted to Kerry Myers, Editor of Leaflets , 215 North Fifth St., Redlands, California 92374, USA. Views published in CROSSOSOMA are those of the contributing author(s) and are not necessarily those of the editors, the membership of Southern California Botanists Inc., or the SCB Board of Directors, unless explicitly stated. Copyright © 2008 by Southern California Botanists, Inc. All rights reserved. Permission to reproduce items in CROSSOSOMA, in whole or in part, should be requested from the current Editor. Volume 34, Number 1 Spring- Summer 2008 CONTENTS Preliminary morphometric analysis of Eriastrum densifolium (Polemoniaceae) populations from Lytle Creek and La Cadena Avenue, Santa Ana River watershed Sarah J. De Groot 1 New California records of lichens and lichenicolous fungi Jana Kocourkova, Kerry Knudsen, James C. Lendemer, and Alan M. Fry day 19 Plant succession in the eastern Mojave Desert; An example from Lake Mead National Recreation Area, southern Nevada (revised) Scott R. Abella, Alice C. Newton , and Dianne N. Bangle 25 Noteworthy Collections: New records of lichens and lichenicolous fungi from California Kerry Knudsen and Jana Kocourkova 37 Book Review: Introduction to the geology of southern California and its native plants by Clarence A. Hall, Jr. (2007) 40 Cover: Eriastrum densifolium , photographed by Sarah De Groot http://www.socalbot.org Crossosoma 34(1), Spring-Summer 2008 1 PRELIMINARY MORPHOMETRIC ANALYSIS OF ERIASTRUM DENSIFOLIUM (POLEMONIACEAE) POPULATIONS FROM LYTLE CREEK AND LA CADENA AVENUE, SANTA ANA RIVER WATERSHED Sarah J. De Groot Rancho Santa Ana Botanic Garden and Claremont Graduate University 1500 N. College Ave. Claremont, CA 91711 sarah.degroot@cgu.edu ABSTRACT: Plants of E. densifolium growing at Lytle Creek and the La Cadena Avenue crossing of the Santa Ana River have been suggested to be hybrids involving E. densifolium subsp. sanctorum , a federally endangered subspecies. Using multivariate analysis of morphological characters, relative warp analysis, and elliptic Fourier functions, these two populations were found to be morphologically intermediate between subsp. sanctorum and a group containing samples of subspp. elongatum and austromontanum. These results could be explained by a hybrid origin of each population, or the populations may be simply intermediate forms. E. densifolium is a highly variable species that appears to be composed of a number of ecological races, which are not well characterized and warrant further study. KEYWORDS: Eriastrum densifolium , E. densifolium subsp. sanctorum , hybrid, intermediate, La Cadena Avenue, Lytle Creek, morphology, Santa Ana River INTRODUCTION Gilia densifolia subsp. sanctora was first described by Milliken in 1904 (p. 39), and later transferred to Eriastrum densifolium by Mason (1945: 75). It was distinguished from typical E. densifolium by the size of the corolla, “...fourteen to fifteen lines long [about 30 mm] and proportionately ample” (Milliken 1904: 39; type specimen [UC 52454] has corollas >25 mm). These characters seem to hold fairly well, because subsp. sanctorum generally has been recognized as distinct from other subspecies by the vast majority of authors who have treated it (e.g., Craig 1934; Jepson 1943; Mason 1945; Harrison 1972). Plants usually occur on higher floodplains above washes, in sandy alluvial soils (Zembal and Kramer 1984). The only range or locality information listed in the protologue was “Santa Ana River near Riverside...” However, from field notes and herbarium specimens, it is apparent that the type collection was made at the Spanishtown Crossing of the Santa Ana River, which today is where Riverside Ave./Main St. crosses the river. Plants of subsp. sanctorum occurred from the foot of the San Bernardino Mountains all the way down into Orange County. For example, J.T. Howell 2985, collected in Santa Ana Canyon in 1927 (RSA), has corollas 27-29.5 mm long, and was cited by Craig (1934: 390) as typical sanctorum. More recent surveys and collections have shown that the present distribution of this subspecies has become restricted to the upper portions of the Santa Ana River drainage in the San Bernardino Valley, namely, the area near Highland, Mentone, and Redlands in San Bernardino County, California (Zembal and Kramer 1984; California Natural Diversity Database [CNDDB]). A few plants have been reported near Colton and Riverside (CNDDB), and although plants had been collected in the Santa Ana River 2 Crossosoma 34(1), Spring-Summer 2008 Canyon in Orange County, all Orange County populations by now are probably extirpated (Zembal and Kramer 1984; Marsh 1988). Outside of the Santa Ana River bed, plants have been reported from Plunge Creek, Cajon Wash, Lytle Creek, and several other localities (CNDDB; Zembal and Kramer 1984). A shrinking range and loss of suitable habitat were reasons cited for its listing as endangered by the federal government (Kramer 1987). Almost since the time they were first collected, the plants at Lytle Creek have been suspected hybrids. Craig (1934) cited them as intermediate between subsp. sanctorum and subsp. elongatum. Subsequent authors have reiterated this (Wheeler 1988; Burk et al. 1989). However, little quantitative study has been done involving these plants. The U.S. Fish and Wildlife Service continues to recognize them as the endangered subsp. sanctorum. At another site of purported hybrids along Cajon Wash, a number of corollas were measured and many were found to be intermediate in length between subsp. sanctorum and subsp. elongatum (La Pre and Pendleton 1988). More recently, plants found near the La Cadena Avenue crossing of the Santa Ana River also appeared to be intermediates between subsp. sanctorum and possibly subsp. austromontanum. Although five subspecies of E. densifolium have been recognized by most taxonomists (e.g., Craig 1934; Mason 1945; Harrison 1972), the species is remarkably variable and some authors have found it difficult to define precisely the circumscriptions of the subspecies (e.g. Craig 1934; Brunell and Whitkus 1997, 1999a, 1999b). Identification also has been problematic. For example, the RSA herbarium has two sheets of Swinney 2294, one identified as subsp. elongatum , and the other identified as austromontanum (det. by O. Mistretta, Oct. 2005). It does not appear to be a mixed collection. A sampling of the morphological diversity in corolla shape and size is illustrated in Figures 1-5, and some leaves are shown later in the paper, in Figure 8. Geometric morphometric data has been used successfully to distinguish plant species or hybrid populations (Premolil996; Olsson et al. 2000; Dickinson et al. 1987; Shipunov and Bateman 2005). McLellan and Endler (1998) classified leaves based on landmarks and elliptic Fourier functions. In a study of two species of Acer and their hybrid, Jensen et al. (2002) used traditional measurements, elliptic Fourier coefficients, and relative warp scores to demonstrate that the hybrids were morphologically intermediate between the parent species. Thus, quantitative morphological data can be informative in studies of potential hybrid populations. This paper reports preliminary findings of a morphometric study involving the Lytle Creek population, the La Cadena Avenue population, and a population each of subsp. sanctorum , subsp. elongatum , and subsp. austromontanum , for reference. The main goal was to see how each intermediate population was related to each of the three subspecies and, in particular, to the endangered subsp. sanctorum. METHODS Fifteen flowering individuals per population were selected arbitrarily and sampled from one population each of E. densifolium subsp. sanctorum , subsp. elongatum , and subsp. austromontanum , along with two intermediate populations: one at Lytle Creek just south of Interstate 210, and one at the La Cadena Avenue crossing of the Santa Ana River (Table 1; Figure 6). Populations of the subspecies were chosen based on their resemblance to type material and proximity to type localities. Crossosoma 34(1), Spring-Summer 2008 3 Figures 1-5 (top to bottom). Sample dissected corollas of Eriastrum densifolium subspecies, with landmarks used in this analysis. 1. Subsp. austromontanum (Mormon Rocks). 2. Subsp. elongatum (Vineyard CynB). 3. Subsp. sanctorum (Alabama St). 4. Intermediate from Lytle Creek. 5. Intermediate from La Cadena Ave. Scale (in mm) is the same for all photos. Table 1. Summary of localities sampled. Subspecies Population Name County (California) Latitude N Longitude W Elevation (m) sanctorum Alabama Street San Bernardino 34.09694 117.20972 347 elongatum Vineyard Cyn B Monterey 35.83702 120.62572 336 austromontanum Mormon Rocks San Bernardino 34.31639 117.49333 988 intermediate La Cadena San Bernardino 34.04590 117.32332 889 intermediate Lytle Creek San Bernardino 34.13264 117.35462 1284 An open corolla from each of fifteen individuals per population was chosen arbitrarily. It was dissected open by removing the petal to the lower left (when the flower was viewed from the front), and photographed along with a metric rule at lOx through an Olympus SZH stereo microscope, using a SPOT RT Color 2.2.1 digital camera and SPOT software (version 4.1.1, © 1997-2004 Diagnostic Instruments, Inc.). Forty-four landmarks and semi-landmarks were placed on the same two petals and three stamens in each image (those in the center after the removal of one petal), with TpsDIG (1.40, Rohlf 2004). Landmarks were placed at presumably homologous points, such as bi- or trifurcations of 33 e 54mi 34 WN 34 WN 34*1 2KTN 34‘1S0"N 4 Crossosoma 34(1), Spring-Summer 2008 120 t, 54'0"W i20“3ctrvv Miles Mormon Rocks Camp Roberts 120°4ZO"w _W3Q'[TVU Lytle Creek Alabama St 120WW I20WW ii5‘0'(rw 117"3S'0"W 117*24'0"W 117*1Z0"W ■ l i ii7*tzcrw imrecm 117"30'(rw 117 e 30’crw ii7*ie'a"w 117*10U'W vineyard Canyon Mieuel Woolly Star Mitigation + ^ ,ba Cadena Miles Figure 6. Map of Eriastrum densifolium sites sampled in this study. Main map shows the upper Santa Ana River drainage area, mitigation land and sites sampled in this study. Top inset shows the location of the Vineyard Canyon B ( elongatum ) population. Bottom inset shows the general location of the five populations in southern California (created with ArcMap® 9.1). Crossosoma 34(1), Spring-Summer 2008 5 veins, anastomoses, sinuses, filament divergence points, and where veins met the margins of the corolla. Semi-landmarks were placed along the veins or margins of the corolla, to capture more general aspects of the form (Figure 7). Landmarks were rescaled from pixels to millimeters in CoordGen6 (part of the Integrated Morphometries Package [IMP], Sheets 2002), and saved in X1Y1...CS format. A matrix of 95 measurements in a modified truss pattern was calculated from these landmarks with tMorphGen6c (Sheets 2002; Strauss and Bookstein 1982; Figure 7). The measurements were ln-tansformed and used in a principal components analysis (PC A) with varimax rotation (StatView version 5.0.1 Power PC version ©1992-98 SAS Institute, Inc.) and a discriminant analysis (DA) with all variables entered simultaneously, and classification cross-validated by leaving out each successive observation to test functions created by the other samples (SPSS version 15.0 for Windows® ©2006 SPSS Inc.). Figure 7. Diagram of interlandmark measurements (white lines) for the corollas. Numbered dots and notes refer to landmarks. To obtain a more visual idea of how the corolla shapes differ among the populations, the original landmark files were concatenated and used in a relative warp (RW) analysis using the program tpsRelw (1.46, Rohlf 2008; Rohlf 1993), with landmarks for the filaments removed because of artificial variation. Settings were default: Alpha=0, uniform component= complement, scale aligned (centroid size)= 1, projection= orthogonal. This analysis superimposed point configurations for each specimen, then calculated singular values based on the shape differences between configurations (Rohlf 1993). Since the number and position of lobes on the leaves was too variable to permit the placement of homologous landmarks (Figure 8), a boundary analysis was used instead. One mature leaf, i.e. below the inflorescence but part of the current year’s growth, was chosen arbitrarily from each individual and scanned on a Hewlett-Packard ScanJet 3970 or G3010 flatbed scanner at 400 dpi in grayscale, using HP Scanning (version 2.2.1 © 1996-2003 Hewlett-Packard Co.). Individual images were converted to black and white in Adobe® Photoshop® (CS2 Version 9.0 ©1990-2005 Adobe Systems, Inc.). Outlines were captured with TpsDIG (1.40, Rohlf 2004), and 40 elliptic Fourier (EF) harmonics 6 Crossosoma 34(1), Spring-Summer 2008 Figure 8. Sample outlines of leaves for each population: a=subsp. austromontanum Q=elongatum s=sanctorum int ^intermediate were calculated with the program EFA (Rohlf and Ferson 1993; Kuhl and Giardina 1982), although only the first eight harmonics showed variation (see Figure 13) and were used in the analysis. This program was set to be invariant to size (by dividing all points by the square root of the area of the first harmonic) and translation (by subtracting the coordinates of the centroid from each pair of points on the outline), but allowed to vary with rotation and starting position of points. Corrections to make the analysis invariant to rotation and starting position of points both rotate the object so that its major axis is parallel to the x-axis, meaning that final (rotated) position is dependent on the shape of the object. Since this can introduce an artifact into the analysis if there is wide variation among shapes, the corrections for rotation and starting position were not applied (Rohlf and Ferson 1993). In this analysis, all leaves were oriented with the apex up and points were started in approximately the same place. The matrix of EF coefficients was analyzed with PCA (StatView; Al-8, Bl-8, CO-8, Dl-8) and discriminant analysis, with variables entered simultaneously (SPSS; Al-8, Bl-8, CO-8, Dl-8). RESULTS Principal components analysis of the corolla measurements showed some differentiation among populations (Figure 9). Although 47 components were recovered in the analysis, components 13-47 each explained less than 1% of the variance (not shown). The first axis accounted for 52.6% of the total variation (Table 2), and because nearly all characters had high (>0.5), positive loadings on it, probably referred to corolla size. The Alabama Street population was distinguished from the other populations by PC 1, although there was also Crossosoma 34(1), Spring-Summer 2008 Figure 9. Ordinations of scores of individuals on principal components 1-8, based on corolla measurement data. Note that axes are not necessarily isometric. (La Cadena and Lytle Creek= intermediates. Vineyard CB= subsp. elongatum. Mormon R= austromontanum. Alabama St= sanctorum.) some separation of the two intermediate populations from a group containing the Vineyard Canyon B and Mormon Rocks populations. PC 2, which was concerned with the distance from the sinus to the filament divergence point, also separated La Cadena and Lytle Creek from Vineyard Canyon B and Mormon Rocks, but with overlap. PC 4 also had high (>0.3) loadings for measurements of the distances from the sinus to the filament divergence point, along with corolla lobe length, and partially distinguished the La Cadena and Lytle Creek populations. Vineyard Canyon B and Mormon Rocks were not readily distinguished by PCA, although PC 6 may show partial differentiation. 8 Crossosoma 34(1), Spring-Summer 2008 Table 2. Eigenvalues, percent variance explained, and some characters described by each axis (based on high loadings) from a PC A of 95 corolla measurements. PC Eigenvalue Variance explained Characters described 1 49.986 52.6% Size (loadings >0.5 for all but 15 variables); variables associated with tube length have highest loadings (gen. >0.8) 2 8.122 8.5 Distance from sinus to filament divergence point (loadings >0.5) 3 6.735 7.1 Trifurcation position, throat position (loadings generally >0.4) 4 3.867 4.1 Distance from sinus to filament divergence point, anastomoses position (loadings >0.3) 5 3.606 3.8 Shape of the tips of the lobes (loadings >0.4) 6 2.730 2.9 Curvature/asymmetry of corolla lobes and filament lengths 1, 2 (loadings >0.3) 7 2.213 2.3 Possibly corolla asymmetry (loadings >0.3 or <-0.3) 8 1.935 2.0 Width of middle of corolla lobes (loadings >0.4) 9 1.671 1.8 Filament lengths or exsertion, lobe width at base and tip (loadings >0.3 or <-0.3) 10 1.596 1.7 Position of middle sinus and overlap between lobes (loadings >0.3) 11 1.226 1.3 Base of corolla (-.303), tip of right lobe (-0.314) 12 1.006 1.1 Position of trifurcation on left petal, throat (loadings >0.2 or <-0.2) In contrast to the PCA, the discriminant analysis was able to discriminate all five populations, with an original correct classification rate of 96.0% and a cross- validated rate of 73.3% (random cross-validated data was 20.0% correct; Figure 10, Table 3). All populations had several misclassifications, except for Alabama Street (sanctorum), which had 100% correct classification in the cross-validation. Lytle Creek had the worst classification (8 of 15 correct). Function 1 described overall corolla size, and separated the populations fairly neatly into three groups: (1) Alabama Street, (2) La Cadena and Lytle Creek, and (3) Vineyard Canyon B and Mormon Rocks. Function 2 further separated the intermediate populations from Vineyard Canyon B and Mormon Rocks, based on the positions of the sinuses and filament divergence points. Vineyard Canyon B and Mormon Rocks were distinguished by function 3, which referred to the width of the corolla lobes and the position of the midvein trifurcation. La Cadena and Lytle Creek were discriminated by function 3 in combination with function 4, which describes the distance from the throat to the sinuses. Bartlett’s Test indicated that all the functions showed significant discrimination between groups (Table 4; Zelditch et al. 2004). The results of the relative warp analysis largely concurred with the PCA and DA (Figures 11, 12, Table 5). Relative warp 1 distinguished Alabama Street from Vineyard Canyon B and Mormon Rocks, with the La Cadena and Lytle Creek populations in between. Relative warp 2 differentiated the intermediate populations from the others. Vineyard Canyon B and Mormon Rocks were slightly separated on RW 3 and RW 4, while RW 5 partially discriminated La Cadena from Lytle Creek. Relative size of the tube versus the lobes seemed to play a role for RW 1, while characters represented by RW 2 included sinus position and corolla lobe shape (Figure 12). Crossosoma 34(1), Spring-Summer 2008 9 Figure 10. Ordination of scores for individuals on discriminant functions, based on corolla measurement data. Centroids are also plotted. Note that axes are not isometric. Mormon Rocks (a) MR centroid Vineyard Cyn B (e) VCB centroid Alabama St (s) ALB centroid La Cadena (int) LCD centroid Lytle Creek (int) LTC centroid Table 3. Summary of discriminant functions and characters described, from a discriminant analysis using 95 corolla measurements. Function Eigenvalue %of Variance Canonical Correlation Characters described (correlations with discriminant functions >0.3 unless noted) 1 14.135 53.2 0.966 Corolla size (length and width characters) 2 9.126 34.4 0.949 Sinus and filament divergence positions 3 2.067 7.8 0.821 Corolla lobe width, position of trifurcation (correlations <-0.2) 4 1.236 4.7 0.743 Distance from throat to sinus or furcation Table 4. Results of Bartlett’s Test for discriminant functions. Function(s) Wilks’ Lambda DF p 1 through 4 0.001 389.618 124 2 through 4 0.014 237.467 90 3 through 4 0.146 107.821 58 4 0.447 45.060 28 0.022 RW6 RW2 10 Crossosoma 34(1), Spring-Summer 2008 0.05 -0.05 e T 0.04 □ • dV ° 0.02 X □ + . O VJLtt _ + « lA* + GO ♦ A x " r X tz -0.02 x o fee ^ A D O X O -0.04 -0.05 0 0.05 RW 5 0.1 -0.05 0 RW 7 0.05 Figure 11. Ordinations of scores of individuals on relative warps 1-8, based on corolla landmark data. Axes are not necessarily isometric. D. IS 01 D.05 Consensus t La Cadena and Lytle Creek (int.) : i s * ■* # * ' 1 * I * ; i t i * * 1: ; \ # Mormon Rocks fllsbamg SI (s> + La CadsnsflnO Ly0eCn&ek(jnH Mormon Rocks (a) and Vineyard Cyn B (e) I □ -0.05 -01 -G05 0 0.05 0.1 0.15 RW 1 \ ' _ Alabama St (s) w S r* y* * ^ 2 ! t ■*, j \ * S f \ Figure 12. Vector deformations associated with relative warps 1 and 2. The consensus was the average configuration and would correspond to the configuration at the origin of the plot (0,0). The other vector plots show the deformations associated with moving from the consensus shape to the approximate centroid of each of the three main groupings of points in the ordination. Crossosoma 34(1), Spring-Summer 2008 11 Table 5. Singular values and percent variance explained for the first 20 relative warps, based on 38 corolla landmarks. There were 72 relative warps in total, but warps 21-72 each accounted for less than 0.2% of the variance (not shown). warp Singular value % Variance Cum. % warp Singular value % Variance Cum. % 1 0.57479 46.13 46.13 11 0.07329 0.75 94.52 2 0.36031 18.12 64.25 12 0.06989 0.68 95.20 3 0.26584 9.87 74.12 13 0.06463 0.58 95.78 4 0.22737 7.22 81.33 14 0.06391 0.57 96.35 5 0.1718 4.12 85.45 15 0.05424 0.41 96.76 6 0.14471 2.92 88.38 16 0.05148 0.37 97.13 7 0.11038 1.70 90.08 17 0.05063 0.36 97.49 8 0.10357 1.50 91.58 18 0.04641 0.30 97.79 9 0.09415 1.24 92.81 19 0.04359 0.27 98.06 10 0.08264 0.95 93.77 20 0.03813 0.20 98.26 The first eight elliptic Fourier functions of the leaves appeared to capture most of the variation, and thus only these were used in statistical analyses (Fig. 13). Principal components analysis of the EF coefficients of the leaves showed little differentiation among populations (Figure 14). Component 1 accounted for 21.2% of the variance, PC 2 for 14.9 %, PC 3 for 1 1.0%, PC 4 for 9.0%, PC 5 for 8.7%, PC 6 for 5.5%, PC 7 for 4.8% and PC 8 for 3.6%. However, the DA distinguished the populations fairly well (Figure 15; Table 6). The original classification rate was 82.7% correct, although the cross- validated rate was 32.0% correct (random cross- validated data was 22.7% correctly classified). While each group had misclassifications in the other four groups, Alabama Street (, sanctorum ) had the highest number of individuals correctly assigned (7 of 15), and Mormon Rocks ( austromontanum ) had the lowest (3 of 15). Function 1 separated Vineyard Canyon B and Mormon Rocks from the other three populations. Function 2 discriminated the intermediates from Alabama Street. Vineyard Canyon B and Mormon Rocks were mostly separated by function 3, and La Cadena and Lytle Creek were distinguished (but with a little overlap) by function 4. Although some discrimination was apparent for all four functions, Wilks’ Lambda was significant for only the first function (Table 7; Zelditch et al. 2004). DISCUSSION La Cadena and Lytle Creek as hybrid populations Both corolla and leaf data point to the La Cadena and Lytle Creek populations as morphologically intermediate between subsp. sanctorum and the group of subspp. elongatum and austromontanum , based on the first axis of PCA and DA of both data sets. While morphological intermediacy is not conclusive proof that these populations are of hybrid origin, many hybrids are morphologically intermediate between parents, at least initially (McDade 2000, and citations therein; Jensen et al. 2002). However, it is also very possible that these populations are stable entities, and just as old as the other subspecies, and are simply morphologically intermediate between described subspecies. That the intermediates are of hybrid origin, or that they are not, cannot be shown with these data. 12 Crossosoma 34(1), Spring-Summer 2008 1.00 o.so 0.60 0.40 0.20 0.00 - 0.20 -0.40 -0.60 -0.30 - 1.00 ■ LaCadena (int) ■ VineyardCB (e) MormonR (a) ■ AlabamaSt (s) ■ LytleCr (int) liorMo^^oiolfO'j-oioiNco in h n n 01 ■ niTrrl TTTTTH Iln 1-h rv co cn m -fh -fh eg eg m <<<<< CO CO ED ED CO UUUUUj 1 Q Q Q Q Q 1 1 Figure 13. Graph showing the amplitudes of each elliptic Fourier coefficient, averaged across each population. + La Cadena □ Vineyard CB O Mormon R A Alabama St X Lytle Cr PC 5 PC 7 Figure 14. Ordinations of scores of individuals on principal components 1-8, based on elliptic Fourier coefficients for the leaves. Note that axes are not necessarily isometric. (La Cadena and Lytle Creek= intermediates. Vineyard CB= subsp. elongatum. Mormon R= austromontanum. Alabama St= sanctorum .) Crossosoma 34(1), Spring-Summer 2008 13 $ Mormon Rocks (a) * MR centroid El Vineyard Cyn B (e) * VCB centroid A Alabama St (s) # ALB centroid + La Cadena (int) LCD centroid X Lytle Creek (int) * LTC centroid Figure 15. Ordinations of scores of individuals on discriminant functions 1-4, based on elliptic Fourier coefficients for the leaves. Note that axes are not necessarily isometric. Group centroids are also plotted. Table 6. Summary of discriminant functions for elliptic Fourier coefficients. Function Eigenvalue %of Variance Canonical Correlation 1 2.521 47.6 0.846 2 1.335 25.2 0.756 3 0.754 14.3 0.656 4 0.681 12.9 0.637 Table 7. Results of Bartlett’s Test for discriminant functions. Function(s) Wilks’ Lambda Chi-square DF P 1 through 4 0.041 175.377 132 0.007 2 through 4 0.145 106.141 96 0.225 3 through 4 0.339 59.496 62 0.567 4 0.595 28.576 30 0.540 It appears that there is little or no reproductive barrier to forming hybrids between subspecies of E. densifolium. Brunell and Whitkus (1999a) found that “...populations have not diverged sufficiently to cause a reduction in crossing efficiency with other populations” (250; Brunell and Whitkus 1997). While their results were somewhat inconsistent between years and not all possible crosses were attempted, they did find that many crosses between subspecies produced seed, including crosses between subsp. sanctorum and austromontanum and crosses between sanctorum and elongatum (Brunell and Whitkus 1999a). Neither intermediate population groups well with either subspp. sanctorum , austromontanum , or elongatum , suggesting that plants at the Lytle Creek and La Cadena sites do not fit well into any of these subspecies. However, given that the intermediate populations fall in between subsp. sanctorum and the other two subspecies in multivariate 14 Crossosoma 34(1), Spring-Summer 2008 analyses, it seems likely that any hybridization events involved subsp. sanctorum for both populations, along with subspp. elongatum , austromontanum , both, or other subspecies that were not sampled in this study. How different are the intermediates from the other subspecies? While plants growing at the Lytle Creek and La Cadena sites can be distinguished, they are morphologically more similar to each other than to subspp. sanctorum , austromontanum , or elongatum. In all analyses, subsp. sanctorum and the intermediates as a group were more readily differentiated from each other and from other subspecies than subsp. elongatum is from subsp. austromontanum. Teasing apart the two hybrid populations, however, was as difficult as discriminating subsp. elongatum from subsp. austromontanum. Further sampling of additional populations of E. densifolium would be necessary to determine more precisely the relationships of the La Cadena and Lytle Creek populations to these and other subspecies (in progress). Eriastrum densifolium seems to fit well into a model of ecological races, such as those described for North American Achillea millefolium (Ramsey et al. 2008). In that study, five non-coding regions of chloroplast DNA and amplified fragment length polymorphisms (AFLPs) showed little structure among populations, leading the authors to conclude that while A. millifolium rapidly diversified, it had also expanded its range. Similarly, very little structure was found among populations of E. densifolium in Brunell and Whitkus’ (1997) study using random amplified polymorphic DNA markers (RAPDs). However, the species is remarkably morphologically diverse, and occurs in a great diversity of habitats, from coastal sand dunes to deserts to montane forests. Morphological adaptations to particular environments may occur at a more rapid rate than molecular evolution or speciation (e.g., Bhagat et al. 2006). Thus, as difficult as ecological races may be to tease apart with molecular techniques, they may be morphologically distinct (Ramsey et al. 2008). How can these populations be distinguished? For the populations studied here, corolla size and relative proportions seemed to be the main defining characters (Figures 16, 17). Corollas of the intermediate populations were intermediate in size between subsp. sanctorum and the group of subspp. austromontanum and elongatum (see also La Pre and Pendleton 1988). In the Jepson Manual key (Hickman 1993), the intermediates are likely to lie right in between the corolla lengths described in couplet 4 (<20mm vs. 25-30 mm). For the populations in this study, the total corolla lengths ranged from 11.0 - 18.0 mm in subspp. austromontanum and elongatum , and from 22.5 - 30.0 mm in subsp. sanctorum. The corollas of the intermediates were 13.0 - 21.5 mm long. The La Cadena population had more within-population variation than the Lytle Creek site, based on the sum of ranked (662 vs. 466) or raw (117.1 vs. 37.5) variances calculated for each corolla measurement. The broader range of variation in corolla size is illustrated in Figure 16. The intermediates also appeared to have a shorter distance between the divergence points of the filaments and the sinuses (Figure 17). Proportionally, these distances were very similar to subsp. sanctorum , but the actual distances were smaller because the corollas were smaller (Vineyard Cyn B 0.6 - 1.7 mm; Mormon Rocks 0.3 - 1.7 mm; Alabama St 0.2 - 1.8 mm; Lytle Creek 0.1 - 0.6 mm; La Cadena 0.2 - 0.8 mm). The intermediate Crossosoma 34(1), Spring-Summer 2008 15 qt; n 30.0 © ~ 25,0 £ t 20 0 1 j -2 150 | 1 T B 1 m away from any shrub) at the northwestern and southeastern corners of each plot and composited these samples on a plot basis. We also selected a dominant Larrea tridentata on each plot on the control and on the treated 1998 corridor (the untreated corridor contained no L. tridentata ) and collected four, 0-10 cm soil samples (composited on a plot basis) halfway between the main stem and the canopy edge. Laboratory and Data Analysis Air dried < 2 mm fractions of soil samples were analyzed for pH (saturated paste), total P and K (Olsen NaHCOs method), total C and N (Leco C/N analyzer), and texture (hydrometer method). We compared mean (n = 3 for each area) species richness and total mean shrub density among the control, treated and untreated 1998 corridor, and the 1968 Crossosoma 34(1), Spring-Summer 2008 29 corridor using one-way analysis of variance and Tukey’s test in JMP (SAS Institute 2004). Analysis of variance also was used to compare interspace soils among the control and 1998 corridor areas. For the control and the treated 1998 corridor, we used paired t tests to compare soil properties between interspaces and below Larrea tridentata. Statistical results should not be extrapolated to other sites since treatments were not replicated, but mean comparisons are presented as interpretational aids. RESULTS 1998 Corridor Exotic species richness/m 2 was similar between treated and untreated areas in the 1998 corridor but was greater than in the control (Figure 2). Total species richness/ 100m 2 was similar among treatments, ranging from 8-9.3 species. The exotic annual grasses Schismus spp. exhibited the highest relative cover on the 1998 corridor compared to the control, but total absolute cover for all species on the corridor was only 5-6% (Figure 3). Relative cover of the native annual Plantago ovata Forsskal was highest in the control, intermediate in the treated corridor, and lowest in the untreated corridor. Perennial forbs and grasses were sparse or absent from all treatments. Shrub density was eight times higher in the control than in the untreated corridor, which contained no Larrea tridentata (Figure 4). Ambrosia dumosa and Encelia farinosa Torrey & A. Gray were the only shrubs inhabiting the untreated corridor, and these species did not occupy plots on the treated corridor or on the control. Larrea tridentata exhibited a density of 300/ha on the treated corridor, which was 36% of the density on the control. In interspaces, soil properties were similar among the treated, untreated, and control areas except for K, which was significantly greater on the untreated corridor than on the control (Table 1). Sand concentration was 10% higher and silt 9% lower on the untreated corridor compared to the control, but all soils were still sandy loams. P and K both tended to be greater below Larrea tridentata than in interspaces for the treated corridor and the control, but the only difference that was statistically significant (p < 0.05) was for P for the control. 1968 Corridor Exotic species richness was lowest in the 1968 corridor relative to the control or the 1998 corridor, and total richness/100 m 2 was comparable to both the 1998 corridor and the control (Figure 2). Similar to the 1998 corridor, Plantago ovata was a major contributor to relative cover, although Stephanomeria pauciflora exhibited the highest relative cover. Total shrub density averaged 3134/ha, 31 times more than the untreated 1998 corridor, nine times more than the treated 1998 corridor, and four times more than the control. Stephanomeria pauciflora and Hymenoclea salsola contributed 76% of the total shrub density. Relative cover (%) No. species No. species 30 Crossosoma 34(1), Spring-Summer 2008 Figure 2. Mean plant species richness at (a) 1 m 2 and (b) 100 m 2 scales among water pipeline corridors and a control area, Lake Mead National Recreation Area, southern Nevada. Error bars are 1 SD for total mean richness. In comparisons within native or exotic categories, only exotic species per m 2 and per 100 m 2 differed significantly (p < 0.05) among the four areas. S Other SSTEPAU ■SCHSPP □ PLAOVA 0 LARTRI □ ERIDEF □ CRYSPP DDCHASPP 19 wmy/f/M 1968 corridor Control Figure 3. Relative cover of dominant plant species and genera among water pipeline corridors and a control area, Lake Mead National Recreation Area, southern Nevada. CHASPP = Chamaesyce spp., CRYSPP = Cryptantha spp., ERIDEF = Eriogonum deflexum, LARTRI = Larrea tridentata, PLAOVA = Plantago ovata, SCHSPP = Schismus spp., and STEPAU = Stephanomeria pauciflora. Numbers at the top of each bar represent total mean absolute % cover. Shrubs/ha Crossosoma 34(1), Spring-Summer 2008 31 Table 1. Comparison of 0-10 cm soil properties among treatments and between interspaces and below Larrea tridentata within treatments on an eight-year-old (1998) water pipeline corridor, Lake Mead National Recreation Area, southern Nevada. Property Untreated corridor Treated corridor 1 Control Interspace Larrea Interspace Larrea Interspace Larrea pH 8.1±0.1 2 — 3 8.1±0.2 8.1±0.1 8.1±0.1 7.9±0.1 P (mg/kg) 4.0±1.2 — 3.5±0.7 3.7±0.8 4.1±1.6 11.5±2.9 K (mg/kg) 555±13a — 491±62ab 575±204 400±61b 552±42 C (mg/kg) 942±35 — 7 1 6± 118 954±107 686±180 736±90 N (mg/kg) 27±5 — 37±14 43±8 51 ±21 50±15 Sand (% wt.) 70±2a — 65±3b 67±4 60±lb 66±6 Silt (% wt.) 24±lb — 29±3a 27±3 33±2a 28±5 Clay (% wt.) 6±2 6±0 6±1 7±1 6±2 1 Restoration treatments included raking the soil surface, applying artificial desert varnish, and planting four species of native shrubs. 2 Vales are mean ± SD (n = 3 within each treatment and canopy combination). Letters within a row compare means among treatments for interspaces only. Values in bold denote significant differences at p< 0.05 between interspaces and below Larrea tridentata within treatments. 3 Not measured because L. tridentata did not occur in this treatment. 1 998 untreated corridor 1 998 treated corridor 1 968 corridor Control Figure 4. Shrub densities among water pipeline corridors and a control area, Lake Mead National Recreation Area, southern Nevada. Error bars are 1 SD for total mean density. Means without shared letters differ at p < 0.05 for total density. 32 Crossosoma 34(1), Spring-Summer 2008 DISCUSSION Although this assessment of an existing disturbance and an unreplicated operational management activity supported only limited statistical inference, our findings represent a case study of succession after land clearing in the eastern Mojave Desert, and how a particular set of restoration treatments may influence succession. Effects of individual restoration treatments cannot be discerned in this study, but the set of treatments including surface raking, applying artificial desert varnish, and planting of shrubs, appeared to make shrub composition on the treated 1998 corridor converge with that of the control (Figure 4). Although our study was not designed to track survival of individual plants in the planting, we found that Larrea tridentata established on the treated 1998 corridor at a density 36% of that of the control. No L. tridentata established on the untreated 1998 corridor. Previous studies of L. tridentata outplanting have produced widely differing results, ranging from complete mortality (Graves et al. 1978) or < 2% survival (Brum et al. 1983), to > 90% survival (Wallace et al. 1980; Clary and Slayback 1984; Newton 2001). In our view, the restoration treatments also made the 1998 corridor appear more similar to surrounding L. tridentata communities, an important consideration on National Park Service lands (Figure 1). This visual blending resulted from the L. tridentata establishment and also probably from the artificial desert varnish. Potential ecological effects of this darkening varnish remain unclear, but it could result in warmer soil temperatures or other effects. One of the largest aesthetic differences between the treated and untreated 1998 corridor and the control was that the control contained desert pavement. Pavement can require millennia to form (Elvidge and Iverson 1983), and it is unclear whether the raking portion of the restoration suite had an effect or will have an effect on surface layers such as desert pavement. The approximately 20 cm of upper soil was salvaged, stockpiled, and reapplied after blading on both the treated and untreated 1998 corridor. Although salvage operations add logistical challenges and expense to projects, ecological effects of soil salvage are not well known in the Mojave Desert and require further study. Soil salvage effects cannot be evaluated in this study of an operational project because this would have required areas on the 1998 corridor that were bladed but did not have soil replaced. The effects of soil salvage and replacement after disturbance may depend on several factors, such as soil type, depth of salvage, and length of time soil is stored (Bainbridge et al. 1998). Effects also hinge on whether or not nutrients and seed banks are diluted upon reapplication by mixing upper and lower soil layers (Nelson and Chew 1977). Based on soil seed bank sampling in the northern Mojave Desert, Guo et al. (1998) reported that 91% of the total seeds were in the upper 2 cm of soil and only 9% occurred from 2-10 cm. Using these data and assuming that the 20 cm of salvaged soil in our study was evenly mixed during salvage operations, the upper 2 cm (likely the germination zone) of the salvaged soil would contain only 10% of the original seeds. Further reductions in viable seeds may occur during topsoil handling or storage. However, it is possible that soil salvage could result in nutrient retention. For example, Rundel and Gibson (1996) reported that total N below shrubs in the northern Mojave Desert was approximately two or more times more concentrated in the upper 5-9 cm of soil than in deeper layers. Shrub species composition on the 38-year-old 1968 corridor is largely consistent with Vasek’s (1983) classification of the successional status of species on an abandoned borrow pit in the Sacramento Mountains in the southeastern Mojave Desert. Three (Stephanomeria pane (flora, Hymenoclea salsola , and Encelia farinosa) of the five shrub species on plots within the 1968 corridor in our study were classified as “pioneer Crossosoma 34(1), Spring-Summer 2008 33 perennials” by Vasek (1983). A fourth species, Ambrosia dumosa , was classified by Vasek (1983) as a long-lived opportunistic species that can be both an early and late- successional species. The last species, Bebbia juncea (Benth.) Greene, was not abundant in Vasek’ s (1983) study and was the least abundant of the five shrub colonizers of the 1968 corridor in our study. Bebbia juncea also was uncommon in a study of abandoned roads in Lake Mead National Recreation Area (Bolling and Walker 2000) and in a study of abandoned military camps in the eastern Mojave Desert (Prose et al. 1987). This species does appear capable of colonizing disturbed areas at low densities, however. Also similar to Vasek’ s (1983) classification, Larrea tridentata , categorized as a late- successional, long-lived perennial, was uncommon even after 38 years on the 1968 corridor. Our findings on both the 1968 and the 1998 corridors concur with the long time scales reported in the literature for Mojave Desert plant succession (Lovich and Bainbridge 1999). For example, Webb and Wilshire (1980) found that perennial species composition on dirt roads abandoned 51 years previously still sharply differed from adjacent control areas at the Wahmonie ghost town site in the northern Mojave Desert. However, these early successional shrub communities are not necessarily “bad,” depending on ecological management objectives. In fact, in our study, plant species richness in early successional shrub communities on the 1968 corridor was similar to the control, and exotic richness was actually lower (Figure 2). Plant assemblages similar to those on this corridor also characterize natural washes in this region (Wells 1961). Based on minimal colonization on the untreated 1998 corridor, however, these shrub communities take more than eight years to develop under the climatic and site conditions characterizing our study. It is possible that the direct planting of Larrea tridentata seedlings on the treated corridor bypassed the development of an early successional shrub stage. The planting allowed the late-succesional L. tridentata to circumvent high-mortality germination and early seedling phases that make natural regeneration an infrequent event (Barbour 1968). ACKNOWLEDGEMENTS We thank Stacey Provencal, Mike Boyles, and Mark Sappington with the National Park Service for facilitating our research permit for this study; the Southern Nevada Water Authority for enabling sampling on their right-of-way; David Connally (Southern Nevada Water Authority) for providing the disturbance history of the 1968 corridor; Utah State Analytical Laboratories for analyzing soil samples; Sharon Altman (University of Nevada Las Vegas) for creating Figure 2; and Jill Craig, Jef Jaeger, Denise Knapp, and three anonymous reviewers for reviewing the manuscript. Funding was provided by the National Park Service through a cooperative agreement with the University of Nevada Las Vegas. LITERATURE CITED Bainbridge, D., R. MacAller, M. Fidelibus, A.C. Newton, A.C. Williams, L. Lippitt, and R. Franson. 1998. A beginner’s guide to desert restoration. 2nd edition. U.S. Department of the Interior, National Park Service. 35 pp. Baldwin, B.G., S. Boyd, B.J. Ertter, R.W. Patterson, T.J. Rosatti, and D.H. Wilken (eds.). 2002. The Jepson desert manual: vascular plants of southeastern California. University of California Press, Berkeley, CA. 624 pp. Barbour, M.G. 1968. Germination requirements of the desert shrub Larrea divaricata. Ecology 49:915-923. 34 Crossosoma 34(1), Spring-Summer 2008 Bolling, J.D., and L.R. Walker. 2000. Plant and soil recovery along a series of abandoned desert roads. Journal of Arid Environments 46:1-24. Brum, G.D., R.S. Boyd, and S.M. Carter. 1983. Recovery rates and rehabilitation of powerline corridors. In Environmental effects of off-road vehicles , eds. R.H. Webb and H.G. Wilshire, 303-314. Springer- Verlag, New York. Carpenter, D.E., M.G. Barbour, and C.J. Bahre. 1986. Old field succession in Mojave desert scrub. Madrono 33:1 1 1-122. Clary, R.F., and R.D. Slayback. 1984. Revegetation in the Mojave Desert using native woody plants. In Proceedings of the native plant revegetation symposium , eds. J.P. Rieger and B.A. Steele, 42-47. California Native Plant Society, San Diego, CA. Elvidge, C.D., and R.M. Iverson. 1983. Regeneration of desert pavement and varnish. Pp. 225-243 in Environmental effects of off-road vehicles , eds. R.H. Webb and H.G. Wilshire, 225-243. Springer- Verlag, New York. Graves, W.L., B.L. Kay, and W.A. Williams. 1978. Revegetation of disturbed sites in the Mojave Desert with native shrubs. California Agriculture 32:4-5. Guo, Q., P.W. Rundel, and D.W. Goodall. 1998. Horizontal and vertical distribution of desert seed banks: patterns, causes, and implications. Journal of Arid Environments 38:465-478. Hunter, R., F.B. Turner, R.G. Lindberg, and K.B. Hunter. 1987. Effects of land clearing on bordering winter annual populations in the Mohave Desert. Great Basin Naturalist 47:234-238. Kemp, P.R., and M.L. Brooks. 1998. Exotic species of California deserts. Fremontia 26:30-34. Lathrop, E.W., and E.F. Archbold. 1980. Plant responses to utility right of way construction in the Mojave Desert. Environmental Management 4:215-226. Lato, L.J. 2006. Soil survey of Clark County area, Nevada. U.S. Department of Agriculture, Natural Resources Conservation Service. 1801 pp. Lovich, J.E., and D. Bainbridge. 1999. Anthropogenic degradation of the southern California desert ecosystem and prospects for natural recovery and restoration. Environmental Management 24:309-326. Moore, C.B., and C. Elvidge. 1982. Desert varnish. In Reference handbook on the deserts of North America , ed. G.L. Bender, 527-536. Greenwood Press, Westport, CT. Nelson, J.F., and R.M. Chew. 1977. Factors affecting seed reserves in the soil of a Mojave Desert ecosystem, Rock Valley, Nye County, Nevada. American Midland Naturalist 97:300-320. Newton, A.C. 2001. DRiWATER: an alternative to hand-watering transplants in a desert environment (Nevada). Ecological Restoration 19:259-260. Prose, D.V., S.K. Metzger, and H.G. Wilshire. 1987. Effects of substrate disturbance on secondary plant succession: Mojave Desert, California. Journal of Applied Ecology 24:305-313. Rundel, P.W., and A.C. Gibson. 1996. Ecological communities and processes in a Mojave Desert ecosystem: Rock Valley, Nevada. Cambridge University Press, New York. 369 pp. SAS Institute. 2004. JMP user’s guide. SAS Institute, Inc., Cary, NC. 402 pp. Vasek, F.C. 1979/80. Early successional stages in Mojave Desert scrub vegetation. Israel Journal of Botany 28:133-148. Vasek, F.C. 1983. Plant succession in the Mojave Desert. Crossosoma 9:1-23. Wallace, A., E.M. Romney, and R.B. Hunter. 1980. The challenge of a desert: revegetation of disturbed desert lands. Great Basin Naturalist Memoirs 4:2 lb- 225. Crossosoma 34(1), Spring-Summer 2008 35 Webb, R.H., and H.G. Wilshire. 1980. Recovery of soils and vegetation in a Mojave Desert ghost town, Nevada, U.S.A. Journal of Arid Environments 3:291-303. Webb, R.H., J.W. Steiger, and R.M. Turner. 1987. Dynamics of Mojave Desert assemblages in the Panamint Mountains, California. Ecology 68:478-490. Wells, P.V. 1961. Succession in desert vegetation on streets of a Nevada ghost town. Science 134:670-671. Crossosoma 34(1), Spring-Summer 2008 37 NOTEWORTHY COLLECTIONS New Records of Lichen and Lichenicolous Fungi from California ARTHONIA VARIANS (Davies) NyL, San Diego County: Cuyamaca Mountains, Cuyamaca State Park, Fire Lookout Road, top of Cuyamaca Peak, 32° 56’ 55” N 116° 36’ 18” W, 1949 m, on apothecia of Lecanora rupicola on large granite boulders, 10 Oct. 2007, Knudsen 9138 (UCR). Previous knowledge. Arthonia varians, a lichenicolous fungus, has been previously reported from North America from Arizona (Gila, Santa Cruz, and Coconino Counties) and Baja California (Guadalupe Island) (as Arthonia glaucomaria (NyL) NyL, a synonym, Triebel et al. 1991; Grube 2007). Opegrapha glaucomaria (NyL) Kallsten ex Hafellner also occurs on Lecanora rupicola and has not yet been reported from California (Ertz & Egea 2007). The ascospores of the two species are similar but easily distinguished: A. varians is characterized by ascospores 3-septate, usually constricted at septa, narrowly obovate, 13-18 x 4-7 pm, persistently hyaline (Grube 2007), while ascospores of O. glaucomaria are ovoid-oblong, sometimes ellipsoid or clavate, straight, hyaline, 3-4-septate, not or slightly constricted at the septa, larger, 18-26(-29) x 6.5-9 pm, and becoming dark when mature (Ertz & Egea 2007). Significance. Arthonia varians is reported new to California, and is expected to be frequent. The host Lecanora rupicola and the related host L. bicinta are common lichens in montane habitats, esp. above 1500 m. Opegrapha glaucomaria is also expected. BACIDIA BAGLIETTOANA (A. Massal. & De Not. ex A. Massal.) Jatta. Santa Barbara County: Santa Rosa Island, Channel Islands National Park, near Smith Highway between Lobo and Cow Canyons, 34° O’ 18” N 120° 5’ 30” W, 150 m, on uplifted slabs of the Beecher Bay formation with Niebla homalea, 16 Oct 2006, Knudsen 7545.1 w/ Sara Baguskus (UCR). Previous knowledge. Bacidia bagliettoana is a crustose lichen with a granular thallus and usually black apothecia, with a blue-green epihymenium, hyaline ascospores with 3 to 9 septa, 33-56 x 2-3 pm, and an orange-brown hypothecium. It occurs on bryophytes, decaying vegetation, plant debris, decaying lichens and calcareous soil, as well as occasionally on the bark of tree bases covered with bryophytes. It is known from Africa, Europe, North America, and New Zealand (Ekman 2004). It is locally common in eastern North America (pers. comm., J.C. Lendemer, NY). Significance. Bacidia bagliettoana is reported new for California and Channel Islands National Park on outcrops of the Beecher Formation. In western North America it has been reported as infrequent in the mountains of Arizona (Ekman 2004). BUELLIA SCHAERERI De Not. San Diego County: Point Loma Ecological Reserve, on Navy land near water treatment plant on hillside, 32° 41 ’ 0” N, 117° 14’ 51” W, 51 m, on Euphorbia misera, 25 Jan. 2006, Knudsen 4984 w/ Andrea Compton (UCR); Point Loma, Cabrillo National Monument, coastal slope above second parking lot on Gatchell Road, 32° 40’ 13” N, 117° 14’ 23” W, 128 m, on Euphorbia misera, 15 May 2007, Knudsen 8427.1 (UCR). Previous knowledge. Buellia schaereri is a widespread lichen species on bark and occasionally on wood, but is either rare or infrequently collected. It has small ascomata 0.2-0.4 mm with dark one-septate ascospores, mostly 7-9 x 3-4.5 pm (Bungartz et al. 2007). The ascospores are not constricted at the septum and have a narrow septum not thickening during spore ontogeny. Significance. Buellia schaereri is reported new for California. It is expected to be frequent and under-collected or mis-identified as Buellia punctata.. 38 Crossosoma 34(1), Spring-Summer 2008 Note : The first author (K.K.) collected extensively off Euphorbia misera, cutting branches for specimens with pruning shears. He got small amounts of sap on his hands. This was transferred to his eyes after sap dried causing drying out tear ducts then corrosive action on eye surface and severe pain. He ended up in emergency and had to have eyes flushed. He was fine by next day and his eyes had no permanent damage. ENDOCOCCUS INCRASSATUS Etayo & Bruess. San Diego County: Henderson Canyon, Anza Borrego State Park, north-facing slope, 33° 18' 35" N, 116° 25' 21" W, 427 m, on Placidiopsis cinerascens on soil over granite, March 15, 2008, Knudsen 9538 w/ Tom Chester & Wayne P. Armstrong (PRM , UCR). Previous knowledge. Endococcus incrassatus , a lichenicolous fungus, was only known from its type locality on the lower slopes of Cerro los Enjambress ca. 2 km WNW of Laguna Chapala in Baja California (Etayo & Breuss 2001). The species is distinguished by the superficial ascomata with thickening around the ostiole and dark one-septate ascospores 10.5-14 x 6-8 pm as well as its host, Placidiopsis cinerascens. Significance. Endococcus incrassatus is reported new to California and the continental United States. This represents only the second collection of this species. The host Placidiopsis cinerascens is rare in California and E. incrassatus is also expected to be rare. LICHENOSTIGMA RADIO AN S Calatayud & Barreno. Santa Barbara County: Santa Rosa Island, Channel Islands National Park, Lopez Road above Jolla Veija Canyon, 33° 54’ 55” N, 120° 4’ 40” W, 66 m, on Aspicilia pacifica , 15 Aug. 2007, Knudsen 8778.2 w/ Sarah Chaney & Silke Werth (UCR, PRM). Previous knowledge. Lichenostigma radicans, a lichenicolous fungus, was previously known only from holotype collection from Spain on a vagrant Aspicilia (Calatayud & Barreno 2003). Significance. Lichenostigma radicans is only the second known collection of this species and it is reported new to California and North America. The specimens match the original description as well as photographs of type. The determination of the host is tentative because ascospores were rare, conidia not found, stictic acid levels were low or non- existent, and only a small amount of norstictic acid was found in medulla. Lichenostigma radicans may be infrequent, nonetheless it can easily be overlooked and if ascomata are not ripe, impossible to determine. LICHENOSTIGMA RUGOSUM Thor. Orange County: Santa Ana Mountains, Fremont Canyon, south ridge, above main truck trail, 33° 47’ 24” N, 117° 41’ 33” W, 490 m, locally abundant on Diploschistes species on decomposing sandstone slabs in shade above main truck trail, Dec. 3, 2007, Knudsen 9279 (PRM, UCR). Previous knowledge. Lichenostigma rugosum belongs to the subgenus Lichenostigma with the ascomata not connected by superficial vegetative hyphae. The species is cosmopolitan in distribution and known from nine species of Diploschistes. It has been reported from the Middle East (Iran and Saudi Arabia), Australia, Europe, and North Africa. In North America it has been reported from Arizona, British Columbia, Colorado, Utah as well as Greenland (Alstrup and Cole 1998; Alstrup and Hawksworth 1990; Calatayud et al. 2004; Thor 1985 & 1995; Triebel et al. 1991). For detailed distribution see Kocourkova (2000). Significance. The species is reported new for California. It appears to be rare or infrequent in southern California, although it is a common lichenicolous fungus in other parts of its range such as the Czech Republic (Kocourkova 2000). Crossosoma 34(1), Spring-Summer 2008 39 Kerry Knudsen, Lichen Curator, UCR Herbarium, Dept, of Botany and Plant Sciences, University of California, Riverside, California 92521 Knuds en(cbucr. edu Jana Kocourkova, Lichen Curator, National Museum, Department of Mycology, Vaclavske nam. 68, 115 79 Praha 1, Czech Republic iana kocourkova(a) nm.cz Acknowledgements Special thanks to Kim Marsden (California State Parks), Mary Ann Hawke and the Plant Atlas Program at the San Diego Natural History Museum, Sarah Chaney (Channel Islands National Park), Andrea Compton (Cabrillo National Monument) and Trish Smith (The Nature Conservancy). We thank Shirley Tucker for reviewing this paper. The work of J. Kocourkova was financially supported by a grant from Ministry of Culture of the Czech Republic (MK0000237201). Cited Literature Alstrup, V. and M.S. Cole. 1998. Lichenicolous fungi of British Columbia. The Bryologist 101(2):221-229. Alstrup, V. and D.L. Hawksworth. 1990. The lichenicolous fungi of Greenland. Meddelelser om Gr0nland, Bioscience 31: 1-90. Bungartz, F., A. Nordin, and A. Grube. 2007 (2008). Buelllia. In Lichen Flora of the Greater Sonoran Desert Region, Vol. 3 , eds. T.H. Nash III, C. Gries, and F. Bungartz, 1 13-179. Lichens Unlimited, Arizona State University, Tempe, Arizona. Calatayud, V. and E. Barreno. 2003. A new Lichenostigma on vagrant Aspicilia species. Lichenologist 35(4):279-285. Calatayud, V., J. Hafellner, and P. Navarro-Rosines. 2004. Lichenostigma. In Lichen Flora of the Greater Sonoran Desert Region, Vol. 2, eds. T.H. Nash III., B.D. Ryan, P. Diederich, C. Gries, and F. Bungartz, 664-669. Lichens Unlimited, Arizona State University, Tempe, Arizona. Ekman, S. 2004. Bacidia. In Lichen Flora of the Greater Sonoran Desert Region, Vol. 2, eds. T.H. Nash III., B.D. Ryan, P. Diederich, C. Gries, and F. Bungartz, 18-28. Lichens Unlimited, Arizona State University, Tempe, Arizona. Ertz, D. and J.M. Egea. 2007 (2008). Opegrapha In Lichen Flora of the Greater Sonoran Desert Region, Vol. 3, eds. T.H. Nash III, C. Gries, and F. Bungartz, 255-266. Lichens Unlimited, Arizona State University, Tempe, Arizona. Etayo, J. and O. Breuss. 2001. Endococcus incrassatus , a new lichenicolous fungus (Dothideales). Osterreichische Zeitschrift fur Pilzkunde 10: 315-317. Grube, M. 2007 (2008). Arthonia In Lichen Flora of the Greater Sonoran Desert Region, Vol. 3 , eds. T.H. Nash III, C. Gries, and F. Bungartz, 39-61. Lichens Unlimited, Arizona State University, Tempe, Arizona. Kocourkova, J. 2000. Lichenicolous fungi of the Czech Republic (the first commented checklist). Acta Musei Nationalis Pragae, Serie B, Historia Naturalis (1999): 59-169. Thor, G. 1985. A new species of Lichenostigma , a lichenicolous ascomycete. Lichenologist 17: 269-272. Thor, G. 1995. Additional lichen records from Australia. 21. Lichenostigma rugosa Thor in Australia. Australasian Lichenological Newsletter 36: 20. Triebel, D., G. Rambold, and T.H. Nash, III. 1991. On lichenicolous fungi from continental North America. My cotaxon 42: 263-296. 40 Crossosoma 34(1), Spring-Summer 2008 BOOK REVIEWS Introduction to the Geology of Southern California and its Native Plants by Clarence A. Hall, Jr. 2007. University of California Press, Berkeley, CA. 493 pp. $75.00. UC Press has essentially cornered the market on massive treatises on the natural history of California. Introduction to the Geology of Southern California and its Native Plants is their latest hit, but it is more a ground-rule double than a homerun. When I first opened the book, I unconsciously expected it to be a Southern California version of Hall’s classic Natural History of the White-Inyo Range (Hall was the editor and wrote the section on geology), but Hall’s most recent contribution is a different kind of book and a very different read. Introduction to the Geology of Southern California and its Native Plants includes an absolutely massive amount of information. There are 69 tables, 32 plates of color photos, and more than 70 figures. There are two glossaries, there are both a species and subject index, and there are more than 350 cited references. The book contains five sections and 21 chapters. It begins with a short overview, and then proceeds to a three-chapter section devoted to geologic concepts, a five-chapter section on geologic history in Southern California, a nine-chapter section covering the major geomorphic provinces, and then two chapters treating basic botany and the major Southern California plant families. Although I originally anticipated that Introduction to the Geology of Southern California and its Native Plants might be a sort of geobotanical tour through Southern California, it never really succeeds in combining the two stated subjects, and personally I found the book somewhat schizophrenic. In this reviewer’s opinion, Introduction to the Geology of Southern California and its Native Plants is chiefly valuable as a compendium of information on the geology of the southern half of the State. There is really only a smattering of interesting but disjointed botanical info, and the botany and plant family chapters are almost gratuitous. Many of the tables in the book are simply very long lists (some of these run 16-19 pages!) of selected plant taxa from different geomorphic provinces that Hall apparently assembled from different floras or from his field work. The lists do not include grasses or sedges (with one exception), and there is no information as to why certain species may have been selected for inclusion and others not. Indeed, it is not at all clear what purpose the lists serve. If you want to know something about Southern California geology, you are likely to read about it in this book, if you can find it! Hall’s book is confoundingly difficult to read (the syntax is awkward and complex and there are long digressions), confusingly organized, and - even with two glossaries! - undefined terms surface everywhere. For example, in the section devoted to a summary of the Ordovician Period in Southern California, Hall writes: “The sequence is composed of a relatively conformable succession of genetically related strata bounded at its top and base by unconformities or their correlative conformities, stratal surfaces, or bedding planes. The sequence boundary is an unconformity and correlative conformity marking a significant basinward shift in facies patterns. A depositional sequence is composed of sequence tracts. A sequence tract is a linkage of contemporaneous depositional systems. Each highland, lowland, or transgressive tract is defined by characteristic parasequence stacking patterns and each is interpreted to be associated with a specific part of the eustatic sea-level cycle. Parasequences are a relatively Crossosoma 34(1), Spring-Summer 2008 41 conformable succession of genetically related, shoaling-upwards beds bounded by marine-flooding surfaces and their correlative surfaces. These depositional sequences have a time-stratigraphic or chronostratigraphic significance in that all of the strata within a parasequence were deposited in a given broad interval of time.” (italics mine) None of the italicized terms in the above selection have entries in the glossaries (although, to be fair, some of them have a quick explanation of the term in parentheses hidden somewhere in the main text). The Introduction suggests the book is written for undergraduate-level students with no geology background, but there is some really dense stuff in here, and in this reviewer’s humble opinion the layperson will find him/herself quickly lost. Introduction to the Geology of Southern California and its Native Plants is apparently meant to serve both as library resource and field guide, although it is too big to easily carry in the field, and it simply isn’t organized like a field guide ought to be. Typically, the geology of an area will be thoroughly (if confusingly) covered, then a very short paragraph will draw attention to a long list of plant species provided in an adjoining table. In a few cases some tidbits on plant ecology of an area are provided. I only found one road log-type entry, in the first chapter of the geomorphic provinces section (“Peninsular Ranges and Colorado Desert”), and it was confusingly done, with a very difficult map to understand. It almost appears as if the author intended to carry this through the other chapters, but ran out of time, patience, or interest. In summary, Introduction to the Geology of Southern California and its Native Plants could have used a heavy dose of old-fashioned editing. Hall’s intentions are admirable, and his book is a veritable bible of geological information, but as with the Bible itself, this book will require a translation to the vernacular before it becomes really useful to the common man. Hugh D. Safford , USDA-F orest Service, Pacific Southwest Region, 1323 Club Drive, Vallejo, CA 94592, and Department of Environmental Science and Policy, University of California, 1 Shields Avenue, Davis, CA 95616. hughsajford@fsfed.us. Become a Member of Southern California Botanists Today! □ Individual (family) $15.00 a New Member □ Organization $25.00 □ Renewal Name Address 1 Address 2 City State Zip code Phone (optional) E-mail address In addition, I would like to give $ to help support SCB. Make checks payable to Southern California Botanists and mail to: Alan P. Romspert Department of Biology California State University Fullerton, CA 92834 Southern California Botanists, Inc. — Founded 1927 — Memberships, Subscriptions, and Back Issues Individual and Family Memberships in SCB are $15 per calendar year domestic, and $20 per year to foreign addresses. Membership includes two issues of Crossosoma, and 5 or 6 issues of Leaflets , the newsletter of SCB. Leaflets provides time-dated information on activities and events that may be of interest to our membership. A subscription to Crossosoma is available to libraries and institutions at the domestic rate of $25 per calendar year, and $30 to foreign institutions. Back issues (Volumes 18-present) are available for $5 each, or $10 for the volume, postpaid. Prior to Volume 18, Crossosoma included time-dated notices to the membership and was published six times a year. These back issues of Volumes 1-17 are $1 each, or $6 per volume, postpaid. Some back issues that are out of stock may be provided as photocopies. Available SCB Special Publications No. 1 A Flora of the Santa Rosa Plateau, by Earl W. Lathrop and Robert F. Thorne, 39 pp $7.00 No. 3 Endangered Plant Communities of Southern California, Proceedings of the 15 th Annual SCB Symposium, edited by Allan A. Schoenherr, 1 14 pp $12.00 No. 4 Flora and Ecology of the Santa Monica Mountains, 194 pp $40.00 By request, the following article has been reprinted separately, with covers, and is available for plant collecting workshops: Herbarium specimens as documents: purposes and general collecting techniques, by T.S. Ross [from crossosoma 22(l):3-39, 1996] $3.95 each; 10 for $22.50 Prices include California state sales tax, handling, and domestic postage. Applications for membership, requests for purchases of Special Publications and back issues, name or address corrections, and requests for replacement of lost or damaged Crossosoma issues should be sent to Alan Romspert, Treasurer Southern California Botanists, Inc., Department of Biology, California State University, Fullerton, CA 92834, USA. Consult the website for current email contact information. Make your check or money order payable to Southern California Botanists , or SCB. http://socalbot.org Southern California Botanists < cn