VOLUME 57, NUMBER 3 JULY-SEPTEMBER 2010 \ fan f Ce} JAN TAXONOMY OF CALYPTRIDIUM PARRYI (MONTIACEAE) Michael G. Simpson, Michael Silveira, and C. Matt Guilliams................ 145 Do NATIVE ANTS PLAY A SIGNIFICANT ROLE IN THE REPRODUCTIVE SUCCESS OF THE RARE SAN FERNANDO VALLEY SPINEFLOWER, CHORIZANTHE PARRYI VAR. FERNANDINA (POLYGONACEAE)? C. Eugene Jones, Youssef C. Atallah, Frances M. Shropshire, Jim Luttrell, Sean E. Walker, Darren R. Sandquist, Robert L. Allen, Jack H.. Burk, and Leo C.. SONG, Ii. @ag hgcI ass oussgaih GPM A asian sveacresvenes 161 GaAs EXCHANGE RATES OF THREE SUB-SHRUBS OF CENTRAL TEXAS SAVANNAS Mitsuru Furuya and OF W..VGR AUke Nie dic 3 PEAR el bacnnnnssvessssceecccnsenenes 170 DOCUMENTATION OF THE CHROMOSOME NUMBER FOR THE CALIFORNIA ENDEMIC, TOXICOSCORDION EXALTATUM (LILIALES: MELANTHIACEAE) Dale W. MeNeaband Wendi Be ZOnmieperoe ii ei cisg ao snhe 0 00etugs 08 Sbnraaazesteeedss 180 THE IDENTITY AND NOMENCLATURE OF THE PACIFIC NORTH AMERICAN SPECIES ZELTNERA MUHLENBERGII (GENTIANACEAE) AND ITS DISTINCTION FROM CENTAURIUM TENUIFLORUM AND OTHER SPECIES WITH WHICH IT HAS BEEN CONFUSED James S. PHN Qe Aes Gc ccwect ee ics S a a cise Oe a sea stoke own vou one ns 184 LOMATIUM TAMANITCHII (APIACEAE) A NEW SPECIES FROM OREGON AND WASHINGTON STATE, USA Mark Darrach, Krista K. Thie, Barbara L. Wilson, RICH OT BE BIOINerd, ONG INICK OTe 2. van ce: sos eae tena sa iedsyieutetecnesusaner ete: 203 ENGI RINT Net ens cic Nn ic Sg fees Ps Se ee aaa ee esis 209 ORE GO Neer ea ee RNR a OE Me etc etn aol coe et Se eS 210 VAAN ree errere ican ontrcan cede a mine nian ame tc itaias Seaceddule deen chuetee castaennec eee tsee 2k MADRONO (ISSN 0024-9637) is published quarterly by the California Botanical Society, Inc., and is issued from the office of the Society, Herbaria, Life Sciences Building, University of California, Berkeley, CA 94720. Subscription information on inside back cover. Established 1916. Periodicals postage paid at Berkeley, CA, and additional mailing offices. Return requested. POSTMASTER: Send address changes to MADRONO, Kim Kersh, Membership Chair, Uni- versity and Jepson Herbarium, University of California, Berkeley, CA 94720-2465. kersh@berkeley.edu. Corresponding Editor—TIMOTHY LOWREY Copy Editor—RICHARD WHITKUS Museum of Southwestern Biology Department of Biology MSCO03 2020 Sonoma State University University of New Mexico 1801 E. Cotati Avenue Albuquerque, NM 87131-0001 Rohnert Park, CA 94928-3609 madrono@unm.edu whitkus @ sonoma.edu Book Editor—JON E. KEELEY Noteworthy Collections Editors—DIETER WILKEN, MARGRIET WETHERWAX Board of Editors Class of: 2010—FReED Hrusa, California Department of Food and Agriculture, Sacramento, CA RICHARD OLMSTEAD, University of Washington, Seattle, WA 2011—JAMIE KNEITEL, California State University, Sacramento, CA KEVIN Rice, University of California, Davis, CA 2012—GRETCHEN LEBUHN, San Francisco State University, CA ROBERT PATTERSON, San Francisco State University, CA 2013—ErIc ROALSON, Washington State University, WA KRISTINA SCHIERENBECK, California State University, Chico, CA CALIFORNIA BOTANICAL SOCIETY, INC. OFFICERS FOR 2010-2011 President: V. Thomas Parker, Department of Biology, San Francisco State University, San Francisco, CA 94132, parker @sfsu.edu First Vice President: Andrew Doran, University and Jepson Herbaria, University of California, Berkeley, CA 94720, andrewdoran @berkeley.edu Second Vice President: Rodney G. Myatt, Department of Biological Sciences, San Jose State University, San Jose, CA 95192, rgmyatt @ gmail.com Recording Secretary: Mike Vasey, Department of Biology, San Francisco State University, San Francisco, CA 94132, mvasey @sfsu.edu Corresponding Secretary: Heather Driscoll, University Herbarium, University of California, Berkeley, CA 94720, hdriscoll @berkeley.edu Treasurer: Thomas Schweich, California Botanical Society, Jepson Herbarium, University of California, Berkeley, CA 94720, tomas @schweich.com The Council of the California Botanical Society comprises the officers listed above plus the immediate Past President, Dean Kelch, Jepson Herbarium, University of California, Berkeley, CA 94720, dkelch @berkeley.edu; the Membership Chair, Kim Kersh, University and Jepson Herbaria, University of California, Berkeley, CA 94720, kersh @ berkeley. edu; the Editor of Madrono; and three elected Council Members: Chelsea Specht, Department of Plant and Microbial Biology, University of California, Berkeley, CA 94720-2465, cdspecht @berkeley.edu; Ellen Simms, Department of Intergrative Biology, 3060 Valley Life Sciences Bldg., #3140, University of California, Berkeley, CA 94720, esimms @berkeley.edu. Staci Markos, University and Jepson Herbaria, University of California, Berkeley, CA 94720, smarkos @berkeley.edu. Graduate Student Representatives: Ben Carter, Department of Integrative Biology and University Herbarium, University of California, Berkeley, CA 94720, bcarter@berkeley.edu. Webmaster: Susan Bainbridge, Jepson Herbarium, University of California, Berkeley, CA 94720-2465, sjbainbridge @ berkeley.edu. This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). Join us for the California Botanical Society 2011 Annual Banquet & Meeting Please help us plan the celebration of the 100" year of the Society (1913-2013)! Saturday, February 12, 2011 University Center Ballroom, Room 102 California State University, Monterey Bay, Seaside, CA 5:00pm Social (No-host Bar) & Meeting California Botanical Society Centennial Planning 7:00pm Dinner with lecture to follow Dr. Joshua Tewksbury University of Washington, Seattle Why Chilies are Hot: The Evolutionary Ecology of a Major Spice Ticket prices are $35 for non-students and $25 for students For more information, please contact Marc Los Huertos Division of Science and Environmental Policy, CSUMB mloshuertos@csumb.edu or 831-582-3209 For information and registration forms visit http://www.calbotsoc.org/banquet Questions? Contact admin@calbotsoc.org Join us in Chico for the California Botanical Society 1** Annual Mixer in conjunction with The Northern California Botanists Symposium http://www.norcalbotanists.org The CBS Council invites you to join us for an evening of discussion focused on the Centennial Celebration of the California Botanical Society. We'll share ideas about upcoming plans and look forward to hearing from members of the community about how CBS can better serve botanists, ecologists, and conservationists in the coming years. Sunday, January 9, 2011 6:30 PM to 8:00 PM Hotel Diamond, wine cellar of Johnnie's Restaurant, Chico http://www.johnniesrestaurant.com/ A fun evening of discussion and ideas exchange Appetizers will be provided and the bar will be open to purchase beverages of your choice. Suggested donation of $10 Mail to: California Botanical Society c/o Jepson Herbarium 1001 Valley Life Sciences Building Berkeley, CA 94720-2465 Please RSVP to Staci Markos at smarkos@berkeley.edu Hope to see you there! MADRONO, Vol. 57, No. 3, pp. 145—160, 2010 TAXONOMY OF CALYPTRIDIUM PARRYI (MONTIACEAE) MICHAEL G. SIMPSON, MICHAEL SILVEIRA, AND C. MATT GUILLIAMS' Department of Biology, San Diego State University, San Diego, CA 92182 msimpson@sunstroke.sdsu.edu ABSTRACT Calyptridium parryi (Montiaceae, formerly Portulacaceae) putatively comprises four varieties: arizonicum, hesseae, nevadense, and parryi. We performed a detailed phenetic analysis of seed, fruit, and sepal morphology in order to assess the distinction and rank of these varieties. We treated the San Pedro Martir Mountains, Baja California populations of C. parryi as a separate taxonomic unit in these studies, given their great disjunct distribution. In addition, we included the very similar and largely sympatric C. monandrum in these analyses. Results from ANOVA and principal components analysis indicate that the six entities examined are, to various degrees, morphologically diagnosable. Calyptridium parryi varieties hesseae, nevadense, and parryi show some overlap in features, and we propose keeping these as varieties. Populations from the San Pedro Martir Mountains are distinct from other varieties of C. parryi, possibly warranting varietal status, but further studies with more samples are needed to confirm this. However, we conclude that C. parryi variety arizonicum 1s different enough in several features and geographic range to warrant species status, by a taxonomic/ morphologic species concept. Calyptridium monandrum should be retained as a separate species. Key Words: Calyptridium, Calyptridium monandrum, Calyptridium parryi, Montiaceae, Portulacaceae, species, taxonomy. Calyptridium Nutt. ex Torr. & A. Gray (FI. N. Am. 1: 198 [1838]) is a genus of the family Montiaceae (formerly Portulacaceae s./.; see Nyffeler 2007; Nyffeler et al. 2008; Ogburn and Edwards 2009; APG III 2009; Nyffeler and Egglhi 2010). Calyptridium is currently recognized to have eight species (Guilliams 2009), one of which has four varieties, for a total of eleven taxa (Table 1; see Hinton 1975 for a nomenclatural history of the group). The genus is distinguished from closely related taxa in having a two-valved, dehiscent capsule. Many other related genera (e.g., Calandrinia, Cistanthe) have dehiscent capsules with three valves, or in Philippiamra [Cistanthe] an indehiscent capsule of two carpels. Members of Calyptridium, like some other Montiaceae, have petals (2-4 in this case) that become apically coherent after anthesis, forming a persistent cap-like structure covering the fruit tip (accounting for the etymology of the name: Greek kaluptra, veil, cap + -idion, little). Based on evidence from morphology and anatomy, Hershkovitz (1990) transferred the species of Calyptridium to section Calyptridium (Nutt.) Hershkovitz of an expanded genus Cistanthe s.l. However, in subsequent molecular analyses of the “‘portulacaceous alliance,” Cis- tanthe s.l. is paraphyletic (Applequist and Wal- lace 2001; Hershkovitz and Zimmer 2000; Hersh- kovitz 2006). Although these studies stop short of making any definitive nomenclatural changes, it "Current address: Department of Integrative Biology, 3060 Valley Life Sciences Building, University of California, Berkeley, CA 94720. is clear that Cistanthe s.l. must be reevaluated, and Hershkovitz (2006), in a study of the western American Portulacaceae, suggested that Calyp- tridium be again classified as a genus distinct from Cistanthe. Based on these studies, we have chosen to recognize Calyptridium at the genus level (in contrast to the treatment in Flora of North America; see Packer 2003). The eight species of Calyptridium have been sorted into two groups, based on plant habit and morphological features of the flower, fruit, and seed (Hinton 1975; Guilliams 2009). One group, comprised of annual plant species with reduced flowers, consists of C. monandrum Nutt., C. parryi A. Gray, C. pygmaeum Parish ex Rydb., C. quadripetalum S. Watson, and C. roseum S. Watson. These five species are similar in lacking a style and in having a fruit that ranges from ovate to narrowly oblong (Table 1). The other group, corresponding to the “Calyptridium umbellatum complex” of Hinton (1975), consists of one annual and two perennial species: C. monosper- mum Greene, C. pulchellum (Eastw.) Hoover, and C. umbellatum (Torr.) Greene. These three species are similar in having filiform styles and an elliptic to orbicular fruit shape (Table 1). Some authors have chosen to recognize the genus Spraguea Torr. (Smithson. Contrib. vi. [Pl. Fremont.] [1854] 4. t. 1.) for this ““Calyptridium umbellatum complex” in recognition of these putative mor- phological differences. However, the validity of Spraguea as a genus distinct from Calyptridium was called into question by Watson (1885), Greene (1886), and Hoover (1940). These authors argue that some of these morphological features 146 TABLE 1. Abbreviation: marg. = marginally; narr. = narrowly. Plant Flower Taxon duration attachment C. monandrum Nutt. annual sessile C. parryi A. Gray var. annual sessile arizonicum J. T. Howell C. parryi A. Gray var. hesseae — annual sessile J. H. Thomas C. parryi A. Gray var. annual sessile nevadense J. T. Howell C. parryi A. Gray var. parryi annual sessile C. pygmaeum Parish ex Rydb. — annual pedicellate C. quadripetalum S. Watson annual pedicellate C. roseum S. Watson annual pedicellate C. pulchellum (Eastw.) Hoover — annual sub-sessile C. monospermum Greene perennial — sub-sessile C. umbellatum (Torr.) Greene perennial — sub-sessile are not fully diagnostic for the group. Most recent treatments (e.g., Wilken and Kelley 1993) of the group follow Greene (1886) in lumping members of Spraguea within an expanded genus Calyptridium (or Cistanthe in Packer 2003). The interrelationships of the species of Calyp- tridium have only recently been the subject of molecular phylogenetic studies. In an analysis. using molecular data from the nrDNA internal transcribed spacer (ITS) and ycf3-trnS chloro- plast DNA intergenic spacer, Hershkovitz (2006) showed that Calyptridium (as defined here) is weakly supported as a monophyletic group. The “C. umbellatum [former genus Spraguea] com- plex” is strongly supported as monophyletic, with C. monospermum and C. umbellatum weakly supported as sister taxa and these sister to C. pulchellum. The other five species are weakly supported as a monophyletic group as well, but with little resolution among them. Calyptridium roseum and C. pygmaeum are weakly supported as sister taxa. Three investigated varieties of C. parryi and C. monandrum form a weakly sup- ported monophyletic group but with no resolu- tion among them. Guilliams (2009), using ITS, ETS, and three non-coding chloroplast regions, obtained results similar to those of Hershkovitz (2006), including the resolution of a monophyletic complex con- sisting of the four varieties of C. parryi plus C. monandrum, but with no resolution of relation- ships within this complex. However, in a com- bined molecular-morphological study, in which previously named taxa were a priori assumed to be monophyletic, he obtained support for a clade MADRONO [Vol. 57 MORPHOLOGICAL FEATURES OF CALYPTRIDIUM TAXA, ARRANGED BY DIAGNOSTIC FEATURES. Petal Style shape/ Seed number presence Fruit shape sculpturing 5 absent ovate - narr. marg. papillate oblong 3-4 absent ovate - narr. smooth oblong 3-4 absent ovate - narr. marg. papillate oblong 3-4 absent ovate - narr. marg. papillate oblong 3-4 absent ovate - narr. papillate oblong 4 absent ovate - narr. smooth oblong 4 absent ovate - narr. papillate oblong 2 absent ovate - narr. marg. papillate oblong 4 filiform — elliptic - smooth orbicular 4 filiform — elliptic - marg. papillate orbicular 4 filiform — elliptic - marg. papillate orbicular in which C. parryi var. hesseae J. H. Thomas is sister to C. monandrum, with these two taxa sister to a clade consisting of a basal C. parryi var. arizonicum J. T. Howell sister to C. parryi var. nevadense J. T. Howell (including C. parryi populations from the San Pedro Martir Moun- tains, Baja California, Mexico) and C. parryi var. parryi (Guilliams 2009). Calyptridium parryi (see Fig. 1A—C for habit and general morphology) has four varieties, which have been separated from one another primarily based on the size and sculpturing pattern of the seeds (Thomas 1956): variety arizonicum having smooth seeds (Fig. 1D); vari- ety hesseae having relatively small, marginally papillate seeds (Fig. 1E); variety nevadense with marginally papillate seeds (Fig. 1F); and variety parryi with seeds papillate over the entire surface (Fig. 1G). Features of the pedicel (‘‘articulate” in C. parryi var. arizonicum, not in the others) and flower duration (‘‘persistent” in C. parryi var. parryi, not in C. parryi vars. hesseae or nevadense) have also been used to separate the varieties (Thomas 1956; Wilken and Kelley 1993), but we found these features difficult to evaluate from herbarium material. Calyptridium monandrum has been separated from C. parryi primarily on fruit morphology (e.g., Wilken and Kelley 1993), the former having fruits longer than 2x the (abaxial) sepal length, the latter <2. The purpose of this study is to investigate the taxonomy of C. parryi in order to evaluate the delimitation, taxonomic ranking, and biogeogra- phy of its infraspecific taxa. Because of the similarity of C. monandrum to C. parryi in a Fic. 1. SIMPSON ET AL.: TAXONOMY OF CAL YPTRIDIUM PARRYI 147 ; ‘fruit (capsule) abaxial A-B. Photographs of Calyptridium parryi var. parryi, showing plant habit in the wild (Simpson 2766, SDSU 18974). C. Close-up of fruit and abaxial sepal from herbarium specimen (RSA 225930). D—G. Seeds of the four varieties of Calyptridium parryi, all to scale. D. C. parryi var. arizonicum; E. C. parryi var. hesseae; F. C. parryi var. nevadense; G. C. parryi var. parryi. number of features (Table 1) and because these two taxa are nested together in recent molecular analyses, C. monandrum was included in all comparative analyses. MATERIALS AND METHODS A total of 109 specimens of Calyptridium parryi and C. monandrum from nine herbaria (ARIZ, ASU, CAS-DS, RSA, SD, SDSU, UC-JEPS, UCR, UTC) were studied, with permission granted to remove seed material for this study. Accession numbers (Appendix |) were recorded and latitude/ longitude recorded or inferred from label informa- tion. Seeds were removed from individual plant specimens having mature fruits and mounted on double-stick tape on microscope slides, labeled with the specimen accession number. Seeds were photographed using a Nikon 990 digital camera on a Wild dissecting microscope. Measurements were made using Image J software (Rasband 1997-2007; see Abramoff et al. 2004) on a Macintosh computer. Seeds are lenticular in shape and have a small notch at the hilar end; a line from this notch through the seed center forms an approximate sagittal section, dividing the seed into two mirror image halves (Fig. 2A). Depending on available material and proper developmental stage, 10-25 seeds per specimen were analyzed. For each seed, the sagittal diameter (SD, including notch length), transverse diameter (TD), notch length (NL), and distance from the perimeter to the central-most extent of papillae (if present) along both the transverse plane (TP) and sagittal plane (SP) were measured from digital images (Fig. 2A). In addition, approximately 5—10 fruits and their corresponding persistent, abaxial sepals (Fig. 2B) were photographed (intact) for each specimen (see Fig. 3 for variation among taxa). In some cases, fewer than five fruits and sepals were measured if suitable material could not be located on the specimen. Fruit outlines could be seen behind the abaxial sepal or determined with minimal manipulation. The abaxial sepal has a distinctive scarious margin. From the images acquired we measured fruit length (FL), maximum fruit width (FW), sepal blade length (SBL), sepal stalk length (SSL), sepal width at widest point (SW), scarious sepal margin width (SM, measured on one side only, the one most intact), and distance from the sepal blade base to the point of greatest sepal width (SB-SW) (Fig. 2B). 148 MADRONO [Vol. 57 FIG. 2. A. Seed image illustrating features measured: sagittal diameter (SD), transverse diameter (TD), notch length (NL), distance from perimeter to the central-most extent of papillae (if present) along both the transverse (TP) and sagittal (SP) planes. B. Fruit and sepal, illustrating features measured: fruit length (FL), fruit width (FW), sepal blade length (SBL), sepal width at widest point (SW), sepal stalk length (SSL), sepal scarious margin width (SM), distance from sepal base to sepal maximum width (SB-SW). Highly disjunct populations of C. parryi from the San Pedro Martir Mountains of Baja California, Mexico were initially identified as C. parryi var. nevadense. However, to investigate possible differences between these Mexican pop- ulations with those of the U.S. populations of C. parryi var. nevadense, these two groups were separated in the analyses, the former labeled C. parryi “‘martirense.”’ To visualize character distributions by taxon, box plots showing the median and the four quartiles of distribution were prepared for 1) average seed papillation; 2) average seed diameter (mm); 3) fruit length (mm); 4) fruit width at widest point (mm); 5) sepal length (mm); 6) ratio of fruit length:total sepal length; 7) sepal width (mm); 8) ratio of total sepal length:sepal width; 9) relative width of scarious margin region of sepal:; and 10) distance from the sepal blade base to the widest point (mm). Average seed papillation was quantified as the relative distance that papillae extend from the seed margin to its center = 2(SP + TP)(SD + TD). A bivariate plot of sepal width versus total length was prepared to visualize trends in these features. Average seed diameter was calculated as (SD + TD)/2. Relative width of the scarious margin region was calculated as SM*2/SW. Each of these morphological charac- ters were evaluated for statistically significant differences by taxon using analysis of variance (ANOVA), with multiple comparisons made between the taxon means using the Tukey post hoc test. Taxa that were statistically different from all other taxa in a particular character are indicated as such (at probabilities <0.01 and <0.05) in the box plot diagrams and in a table, summarizing the mean values of these characters. In addition to comparisons of one character, a principal components analysis (PCA) was con- ducted using 12 characters: 1) seed sagittal diameter; 2) seed transverse diameter; 3) sagittal seed papillation; 4) transverse seed papillation; 5) sepal length; 6) sepal width; 7) distance from sepal blade base to widest point; 8) extent of sepal scarious margin; 9) fruit length; 10) fruit width; 11) fruit length:sepal length; and 12) sepal stalk length. Variables were standardized by subtract- ing the variable mean from each individual measurement, then dividing by the variable standard deviation. This transformation results in all variables having a mean of zero and a standard deviation of 1. The resulting factor scores of this PCA were plotted for the Ist versus 2nd components and 2nd versus 3rd components. Only samples with complete data sets were used in the PCA analyses. All statistical analyses were performed in SYSTAT, Version 11 (Systat Software, Inc., San Jose, CA USA, http://www. systat.com). Maps were prepared showing the distribution of the four varieties of C. parryi and the C. parryi ‘“‘martirense’’ populations. Circles denote relative seed size and shading represents relative papilla- tion of the seed surface (white = smooth region, black = papillate region). The latter was dia- gramed in clock fashion, with, e.g., shading from 12 to 3 o’clock representing 25% average papillation, meaning that the band of papillation extends 25% from the seed periphery to the seed center. 2010] SIMPSON ET AL.: TAXONOMY OF CAL YPTRIDIUM PARRYI 149 FIG. 3. Images, showing examples of fruits with subtending abaxial sepal (scarious margin indicated), all to scale. A. Calyptridium parryi var. arizonicum (ARIZ 200373). B. C. paryii var. hesseae (DS 266493). C. C. paryyi “martirense” (SD 97873). C. C. paryyi var. nevadense (CAS 318495). D. C. parryi var. parryi (RSA 13478). F. C. monandrum (JEPS 48483). Scale bar = 1 mm. RESULTS Three of the four varieties of Calyptridium parryi, plus the C. parryi ‘“‘martirense”’ popula- tions from the San Pedro Martir Mountains and C. monandrum, exhibit statistically significant differences for one or more variables examined using ANOVA (P-values <0.05 and <0.01). Tukey HSD post hoc tests indicate which, if any, of the taxon means are statistically different. Mean values with statistical differences for characters examined are listed in Table 2. The seed sculpturing of C. parryi var. arizoni- cum and C. parryi var. parryi are confirmed to be different from all other taxa examined in having, respectively, totally smooth versus totally papil- late surfaces (both P < 0.01; Fig. 4A). Calyp- tridium monandrum, C. parryi var. hesseae, C. parryi ““martirense,” and C. parryi var. nevadense are very similar in seed sculpturing, having between 25-55% papillation (Fig. 4A). The seed size of C. parryi var. hesseae is overlapping but significantly smaller than all other taxa (P < 0.05), whereas that of C. parryi var. arizonicum 1s 150 =0:01. <0.05; ** = MEAN VALUES FOR CALYPTRIDIUM TAXA OF CHARACTERS EXAMINED USING ANOVA. * TABLE 2. Distance Average Average D 3 Bo OUsE BE o 2 nas eae oe a. cD) AS 2 n E < ~ _ vz ss Pe Y bp | a 2: as £3 & Om Ae = 33 Fae Hx ep 0 S YL — CS aN Bm & oe & nor aoe eee ale sae = mee 2s ee moe — on ca og oa as seed papillation (%) Taxon 0.633* 0.885 16.1 [A293 0.785 25012 1.419 as 4.165* 2757" 3.249 3.249 3.515 4.384 0.894** 1.742 1.449 1.481 0.624 35.4 drum 0.894 1.089 L119 41.5 16.0 Lt22 [263 0.751 3.942 1.838 Bai2o 1.620 1.445 L611 C0607 4.468 0.754** 0.540* 0.685 O0"* 32.) onicum var. hesseae var. aril a4 Be: 4.643 34.5 var. nevadense var. parryi 61.1 42.9 0.983 0.932 2.814 2.882 1.447 925" 1.574 222) 4.666 O39) ese Ages 0.677 Se as 35,9 ‘“martirense”’ la a a eo eas tex ek ~~ ~~ ~~ ~~ Hh & MADRONO [Vol. 57 overlapping but significantly larger (P < 0.01; Fig. 4B). The other four taxa, although showing some variation, generally overlap in seed size, with no significant differences between them (Fig. 4B). Fruits show significant differences in both length and width among investigated taxa. The fruits of C. parryi var. arizonicum are significantly longer than, and (barely) non-overlapping with any other taxon (P < 0.01; Fig. 4C). In contrast, fruits of C. parryi “‘martirense” are significantly shorter (P < 0.01), although overlapping in range with those of C. monandrum, C. parryi var. nevadense, and C. parryi var. parryi (Fig. 4C). Fruit width is significantly smaller (P < 0.01) in C. monandrum and signficantly larger (P < 0.05) in C. parryi ““martirense”’ (Fig. 4D). Abaxial sepals show significant differences in total length. The sepal length of C. monandrum is significantly shorter (P < 0.01) than that of all other taxa, with virtually no overlap in range (Fig. 5A). Sepal length of C. parryi var. hesseae is generally smaller than the remaining taxa (both P < 0.05), and that of C. parryi var. arizonicum is generally greater than all taxa, but both with considerable overlap of ranges (Fig. 5A). In comparing fruit:sepal length the fruits of C. monandrum are 2.1—3.0<* longer than the total abaxial sepal length, significantly greater (P < 0.01) than any other taxon (Fig. 5B). This is due not to its having longer fruits (Fig. 4C), but to its having smaller sepals (Fig. 5A). In contrast, the fruit length of C. parryi “‘martirense” is <1.1X longer than the abaxial sepal length, significantly less (P < 0.01) than any other taxon (Fig. 5B). This is due to its significantly smaller fruit length (Fig. 4C), and a relatively large sepal length (Fig. SA). The three other taxa examined overlap in fruit:abaxial sepal length ratio (Fig. 5B). Abaxial sepals are also variable among taxa with regard to width (at widest point), length:width ratio, relative width of the scarious margin region, and distance from the sepal blade base to the widest point. No taxa are significantly different from all other taxa in sepal width (Fig. 5C). However, both C. monandrum and C. parryi var. hesseae are similar in having relatively narrow sepals, each significantly narrower (P < 0.01) than the other four taxa (Fig. 5C). A plot of abaxial sepal length to width ratio (Fig. 5D) shows that C. parryi var. nevadense has the (relatively) widest abaxial sepals and C. parryi var. hesseae has the (relatively) narrowest sepals. However, there is considerable overlap among taxa, and no taxon is significantly different from any other, although C. monandrum and C. parryi var. hesseae have generally greater ratios, indi- cating that their sepals are relatively narrow (Fig. 5D). A bivariate plot of sepal length versus width (Fig. 6A) shows a general grade with respect to these features, but with definite trends in the complex. Calyptridium monandrum has 2010] SIMPSON ET AL.: TAXONOMY OF CAL YPTRIDIUM PARRYI 15] 100 0.8 5 78 5 = OF : : £ = 2 , s 34 ras) a 3 a0 nm 0.5 12 vo MN -10 0.4 * Fruit Length (mm) Fruit Width (mm) q aa ee C ; Taxa D Fic. 4. Box plots of single characters. A. Percentage seed papillation. B. Seed diameter (mm). C. Fruit length (mm). D. Fruit width (mm). Note: box plots show median (horizontal line), first and third quartiles (boxes above and below median), and second and fourth quartiles (vertical lines); x = outliers; C. parryi ““martirense” = San Pedro Martir Mountains populations of Calyptridium parryi. Statistical difference between a given taxon and all other taxa (via ANOVA Tukey post hoc test) indicated as: * = P < 0.05; ** = P < 0.01. Taxa relatively small sepals (in length and width) as cum have a relatively large sepal length and does C. parryi var. hesseae, the two taxa width. Calyptridium parryi var. parryi and C. overlapping in these two features. Calyptridium parryi ““martirense”’ are intermediate in these two parryi var. nevadense and C. parryi var. arizoni- — sepal metrics. 152 MADRONO [Vol. 57 Sepal Length (mm) 6 5 E E, ai me —_ * a>) = =a 3 2. <>) N Z 1 ri o& Aa ”” os Ros a Om re & . a A ve . a * o & w& ol os of + ar = < : < ay CY) “} o re < q <4 g < C U' U C Taxa Sepal Length ies) NS Fruit Length Width Sepal Length D Taxa FiG. 5. Box plots of single characters. A. Sepal length (mm). B. Fruit length:sepal length ratio. C. Sepal width (mm). D. Sepal length:width ratio. Symbols as in Fig. 4. With regard to relative width of the scarious abaxial sepal margin, C. parryi var. hesseae and C. monandrum have very similar, relatively thin scarious margins (Fig. 6B). These two taxa are not significantly different from each other in this feature, but each is significantly different (P < 0.01) from the other four taxa (C. parryi var. arizonicum, C. parryi var. nevadense, C. parryi 2010] SIMPSON ET AL.: TAXONOMY OF CAL YPTRIDIUM PARRYI bse) 6.0 | © 5.0 + ros © ro A E 4.0 4 eh, A A = | hie § + monandrum = © @ A arizonicum = 3.0 - O ee Gl hesseae a. e Se O martirense N i > nevadense 20 @ parryi 1.0 | T | T | a | 1.0 2.0 3.0 4.0 A Sepal Length (mm) E 1.0 me : 0.9 , z a 0.8 r 5 07 = = : 4 9 06 £ ps v 2 05 5 o eal x 0.4 S -_ ¥ . = 03 2 o S z 02 5 = 0.1 2 a4 (an) 0.0 ny * ea * & s _ g : ~ ee C: LF ay L re Aa v \ S © ¥ ¢ 6 a o8 8 B Taxa C Taxa FIG. 6. A. Bivariate scatter plot of abaxial sepal blade length versus maximum sepal width. B—-C. Box plots of single characters. B. Relative width of sepal scarious margin. C. Distance from sepal base to widest point (mm). Symbols as in Fig. 4. var. parryi, and C. parryi “‘martirense’’). In the comparison of distance from the sepal base to the widest point, taxa generally overlap and show no significant differences, except for C. monandrum, which is significantly less in this feature (but at the P < 0.05 level). The multivariate principal components analy- sis (PCA) using twelve quantified features (Fig. 7) accounts for ca. 44% of the variation explained by the first loading, ca. 23% by the second loading, and ca. 13% by the third loading (Table 3). A plot of the first and second factors 154 MADRONO [Vol. 57 2D + monandrum A arizonicum LJ hesseae O martirense > nevadense @ parryi “2.5 =2.) -2.5 | 5. Y 2.5 + monandrum Aarizonicum hesseae O martirense nevadense @ parryi B Fic. 7. Graph of results of a Principal Components Analysis. A. Plot of first and second factors. Note clear separation of C. parryi var. arizonicum. B. Plot of second and third factors. Note separation of all C. parryi taxa, overlap of C. monandrum and C. parryi var. nevadense. 2010] TABLE 3. PRINCIPAL COMPONENTS ANALYSIS LOADINGS FOR TWELVE CHARACTERS USED. Percent of total variance explained: axis 1 = 44.365, axis 2 = 23.348, and axis 3 = 12.759. Component loadings Character | ) 3 Seed sagittal diameter 0.726 —0.293 0.540 Seed transverse diameter 0.742 —0.150 0.580 Sagittal seed papillation —0.068 0.870 0.362 Transverse seed papillation —0.096 0.867 0.348 Sepal length O914° <0:126 —0126 Sepal width 0.911 —0.069 0.043 Distance from sepal base to widest point 0.656 0.524 —0.137 Extent of sepal scarious margin 0.795 0.218 0.128 Fruit length 0.545 —0.577 0.188 Fruit width 0.619 0.153 =—0.542 Fruit length:sepal length =). 577 —0.611 0:383 Sepal stalk length O70L —0.336° —0:350 shows non-random clusters for all taxa. There 1s some overlap of taxa clusters with the exception for C. parryi var. arizonicum (Fig. 7A). Calyp- tridium monandrum overlaps with C. parryi var. hesseae. A plot of the second and third factors shows a separation of all C. parryi taxa from one another; C. monandrum overlaps with C. parryi var. nevadense (Fig. 7B). A distribution map shows minimal overlap in range among the five C. parryi taxa (Fig. 8). Calyptridium parryi var. parryi 1s found mainly in the transverse range and northern peninsular range, with one known outlier in the South Coast Ranges. Calyptridium parryi var. nevadense 1s found in the Sierras, Inyo, White and desert ranges of California (with one outlier in the transverse range) and in the adjacent mountains of western Nevada, with a couple of outliers in more eastern states. Ca/yptridium parryi var. hesseae is restricted to a few localities in the Santa Lucia, Diablo, and Santa Cruz ranges of the Central Coast, California. Calyptridium parryi var. arizonicum is found both in southern Arizona and in north-central Baja California, Mexico (with two examined outliers in Sonora, Mexico). As mentioned earlier, the “‘martirense”’ populations of C. parryi are restricted to high elevations of the San Pedro Martir Mountains of Baja California, Mexico. Finally, the distribution of C. monandrum overlaps with that of most varieties of C. parryi (not shown; see Consortium of California Herbaria 2009), although C. mon- andrum generally occurs at lower elevations. DISCUSSION The results of these analyses suggest that the four varieties and “‘martirense’’ populations of Calyptridium parryi and the presumed closely SIMPSON ET AL.: TAXONOMY OF CAL YPTRIDIUM PARRYI 15D related C. monandrum each show various differ- ences in sepal, fruit, and seed morphology. Calyptridium parryi var. arizonicum is perhaps the most distinctive of the taxa examined (Table 2). Its seeds are totally smooth, lacking any evidence of papillation (Fig. 4A), and are significantly larger in size (both P < 0.01), although overlapping in size range with C. parryi var. nevadense and C. parryi var. parryi (Fig. 4B). Fruits of C. parryi var. arizonicum are signifi- cantly longer (P < 0.01; Fig. 4C) and sepals are generally longer (P < 0.05) but greatly overlap- ping with those of other taxa (Fig. 5A). Both PCA component plots show a marked separation of C. parryi var. arizonicum in the combination of 12 characters examined (Fig. 7A, B). Calyptridium parryi var. hesseae has signifi- cantly smaller seeds (P < 0.05), although overlapping slightly with two other taxa, partic- ularly with C. monandrum (Fig. 4B). Variety hesseae has sepals shorter than all other taxa except C. monandrum (P < 0.05; Fig. 5A). Variety hesseae is also similar to C. monandrum in having narrow sepals (Fig. 5C) and a relatively narrow sepal scarious margin (Fig. 6B), each significantly different (P < 0.01) from the other four taxa. The sepal length to width ratio of these two taxa is also greater than that of the other four Calyptridium taxa (Fig. 5D; see also Fig. 6A), but without statistical significance. Calyptridium parryi “‘martirense” was surpris- ingly different in some respects from any of the known varieties of that species. The seed sculpturing of this group is very similar to that of C. monandrum and C. parryi varieties hesseae and nevadense (Fig. 4A), and its seed size is somewhat intermediate (Fig. 4B). However, the fruit length of C. parryi “‘martirense”’ 1s signifi- cantly smaller than any of the other five taxa examined (P < 0.01; Fig. 4C), as 1s its fruit:sepal ratio (Fig. 5B). In the PCA analysis C. parryi ‘““martirense’’ shows overlap with other taxa in the plot of the Ist and 2nd factors (Fig. 7A), but is discontinuous 1n the plot of 2nd and 3rd factors (Fig. 7B). Calyptridium parryi var. parryi was confirmed to be discontinuous with respect to seed papilla- tion; all seeds examined had ca. 95—100% papillation (Fig. 4A). The few seeds that were problematic in our study proved to be shrunken, due we think to immaturity or abortion during development. In all other features C. parryi var. parryi shows overlap with one or more taxa. However, in the PCA analysis it shows a discontinuous clustering in the plot of factors 2 and 3 (Fig. 7B). Calyptridium parryi var. nevadense is the only taxon that lacks any unique character states. It 1s very similar to C. monandrum, C. parryi var. hesseae, and C. parryi “‘martirense”’ in seed sculpturing (Fig. 4A) and to C. parryi var. parryi 156 MADRONO [Vol. 57 faites Oe hy C. parryi 7 N var. sida ; C. parryi var. hesseae C. parryi var. parryi Fic. 8. Detailed map of quantified seed data for the Ca/yptridium parryi groups. Circles are proportional to seed diameter. Black coloring is indicative of percentage of distance of papillae from seed periphery to seed center. in seed size (although also slightly with most discontinuous distribution in the plot of factors other taxa; Fig. 4B). Variety nevadense overlaps 2 and 3 among the other C. parryi varieties/ considerably with other taxa in fruit length groups, but overlaps with C. monandrum (Fig. 4C), fruit width (Fig. 4D), fruit length:sepal (Fig. 7B). length (Fig. 5B), and all sepal features (Figs. 5A, Calyptridium monandrum has a_ significantly C-D, 6A-C). In the PCA analysis it has a smaller fruit width (P < 0.01; Fig. 4D) and sepal 2010] length (P < 0.01; Fig. SA) than any C. parryi taxon. It also has a significantly greater fruit length:sepal length (P < 0.01; Fig. 5B). It is this large fruit to sepal ratio that has been used (Wilken and Kelley 1993; Guilliams 2009) to distinguish this species from C. parryi. It is interesting, however, that the fruit length of C. monandrum is quite similar to that of C. parryi vars. hesseae, nevadense, and parryi (Fig. 4C); the greater fruit to sepal ratio is due to C. monandrum having shorter sepals, not to having longer fruits. Finally, as already discussed, C. monandrum and C. parryi var. hesseae are similar in having relatively narrow sepals (Fig. 5C) with relatively narrow scarious margins (Fig. 6B). Biogeographically, the four varieties of C. parryi have mostly discontinuous, well-separated ranges (Fig. 8). Calyptridium parryi var. hesseae is biogeographically unique in being restricted to lower elevations (600—1050 m) in the Santa Cruz, Santa Lucia, and Diablo ranges of the Central Western California Region (biogeographic re- gions after Hickman 1993). It is interesting that C. parryi var. arizonicum occurs in two disjunct regions, one in southern Arizona to Sonora, Mexico and the other in north-central Baja California; we note that another population of C. parryi var. arizonicum was discovered in 2009 in mountains of Anza Borrego Desert State Park, San Diego Co., California (not illustrated; see Consortium of California Herbaria 2009). Thus, C. parryi var. arizonicum 1s restricted to habitats of the Sonoran (including Coloradan) and Viscaino deserts. Calyptridium parryi var. neva- dense and var. parryi have fairly discontinuous ranges, with slight overlap. Calyptridium parryi var. parryi is mostly restricted to mid to high elevations (1400-3500 m) in the transverse range of southern California, whereas C. parryi var. nevadense 1s largely restricted to mid to high elevations (1100-3000 m) in the Sierra Nevada mountains of California and adjacent ranges of the Mohave and Great Basin, east into Nevada, with one examined population in Utah and one in Arizona. One sample each of C. parryi var. parryi and of C. parryi var. nevadense occurs in the western transverse range of southern California, and one sample of C. parryi var. parryi occurs in the South Coast Ranges. As pointed out earlier, the “‘martirense’”” populations of C.. parryi, endemic to the San Pedro Martir Mountains of northern Baja California (2200-2600 m eleva- tion), are quite disjunct from the other popula- tions (Fig. 8). Of the taxa we studied, only C. parryi var. hesseae is rare and appears threatened. This variety was previously described as “‘has not been seen in many years” and “‘likely extirpated locally and possibly near extinction” (Morgan et al. 2005). In earlier field work (Guilliams 2009), we were only able to relocate only one population of SIMPSON ET AL.: TAXONOMY OF CAL YPTRIDIUM PARRYI iS7 this taxon, warranting subsequent solicitation for a California Native Plant Society Inventory listing of 1B.1 (meaning that it is “rare, threatened, or endangered in California and elsewhere” and “‘seriously endangered in Califor- nia, over 80% of occurrences threatened/high degree and immediacy of threat’’); see California Native Plant Society (2009) and Guilliams (2009). NOMENCLATURAL CONSIDERATIONS Based on the presented quantitative morpho- logical data, our study has demonstrated the existence of six groups — Calyptridium monandrum, the four C. parryi varieties, and the C. parryi ‘““martirense”’ populations. The question becomes what to call them and at what rank? In evaluating nomenclatural changes resulting from this study, we here review the concepts for recognizing species and infraspecies (see McDade 1995). Evaluation of the six entities as “biological species” (Mayr 1969) is currently untestable in this complex because we know nothing about reproductive biology of these species and cannot evaluate whether any particular group might be reproductively isolated from any other. The only Calyptridium taxa studied reproductively are C. monospermum and C. umbellatum, found to be outcrossing or have insect-mediated self-pollina- tion (Hinton 1976). The reduced flowers of the C. monandrum - C. parryi complex studied here may imply self-pollination, but this is only speculative, and reproductive isolation among its members has not been tested. Recognition of “phylogenetic species” (e.g., Mishler and Donoghue 1982; de Queiroz and Donoghue 1988), necessitates a phylogenetic analysis and the identification of at least one apomorphy characterizing a lineage. However, the molecular phylogenetic studies of both Hershkovitz (2006) and Guilliams (2009) are inconclusive about the monophyly of any of the taxonomic entities of this study. Thus, a phylo- genetic taxon concept based on current molecular data alone is unworkable for this group, perhaps because of the lack of fine enough molecular marker or because the entities in this complex have evolved relatively recently. Thus, we feel that the only viable procedure is to use a “taxonomic” or “‘morphological”’ species concept (Cronquist 1978, 1988) in which taxa are circumscribed based on a demonstrated disconti- nuity of morphological features. Species are sets of populations that are clearly discontinuous from one another in at least one morphological feature, showing no intergradation. Varieties and subspecies (in practice interchangeable; see Ham- ilton and Reichard 1992) are typically considered to be forms that show differences from one another, but the feature or features that make them different have some intergradation. Varie- 158 ties could represent genetically distinct ecological variants or populations that have either diverged incompletely or become geographically fragment- ed from a larger, ancestral population, possibly representing incipient speciation. Of the six groupings recognized in this study, four — C. parryi var. hesseae, C. parryi var. nevadense, C. parryi var. parryi, and the C. parryi ‘““‘martirense’” populations — are more similar to one another than the other two in the features examined. We feel that the similar seed papilla- tion between varieties hesseae, “‘martirense,’’ and nevadense and the contiguous geographic ranges (with slight interdigitation) between varieties nevadense and parryi support the retention of these taxa as varieties at this time. Within this complex, C. parryi var. hesseae differs from C. parryi var. nevadense and C. parryi var. parryi to a greater degree in its moderately (P < 0.05) significantly smaller (but overlapping) seed size and, along with C. monandrum, its strongly (P < 0.01) significantly narrower (but slightly overlap- ping) sepals and scarious sepal margin. However, because of overlap in these features, we believe that C. parryi variety hesseae should be retained at the varietal rank, given the uncertainty of recent molecular evidence. In addition, we point out that C. parryi var. parryi is completely non- overlapping in one feature: having an entirely papillate seed surface. This is not only a morphological distinction, but could represent a possible apomorphy, as the only other Calyp- tridium taxon to have a uniformly papillate seed surface is C. quadripetalum, which is more distantly related, having evolved uniform seed papillation independently (Guilliams 2009). How- ever, we choose to retain C. parryi var. parryi as a variety, given its very slight interdigitation in range (Fig. 8) and otherwise greatly overlapping morphology with C. parryi var. nevadense. Based on the morphological analyses present- ed, it is clear that the C. parryi “‘martirense” populations are different from C. parryi var. nevadense (to which they were originally identi- fied) within the C. parryi complex. Samples of these populations are identical to C. parryi var. hesseae and C. parryi var. nevadense in seed sculpturing and intermediate to these two in seed size. However, C. parryi ““martirense”’ has signif- MADRONO [Vol. 57 icantly shorter fruits (P < 0.01) and a signifi- cantly smaller, but overlapping, fruit:sepal ratio (P < 0.01). It also exhibits some discontinuity in the PCA analysis. At this stage, we feel that these populations could be treated as another variety of C. parryi (tentatively to be named C. parryi var. martirense), but we wish to conduct further analyses with a greater sample size before publishing a new name. We feel that C. monandrum is justified in its continued recognition as a separate species, due its significantly greater fruit:sepal ratio, smaller fruit width, and smaller sepal size. Calyptridium mon- andrum is more widespread and overlaps in range with most members of the C. parryi complex. However, as pointed out earlier, it shows some similarity in other features to C. parryi var. hesseae. Our strongest conclusion is that C. parryi var. arizonicum is markedly distinct from all other varieties or forms in its smooth (P < 0.01, non- overlapping) and significantly larger (P < 0.01, but overlapping) seeds and its significantly longer (P < 0.01, essentially non-overlapping) fruits. It has the most discontinuous clustering of any taxa in both plots of the PCA analysis using 12 combined characters (Fig. 7). In addition, the range of this taxon, although split by the Sea of Cortez, is completely non-overlapping with the range of C. parryi vars. hesseae, nevadense, and parryi and the “‘martirense”’ populations (Fig. 8), occurring in desert habitats. Thus, based on this study in which seed, sepal, and fruit morphology were quantified, we conclude, based on a taxo- nomic species concept, that C. parryi var. arizoni- cum should be elevated to the rank of species; the necessary new combination is published below, along with a key to this complex of species. Calyptridium arizonicum (J. T. Howell) M. G. Simpson, M. Silveira, & Guilliams, comb. nov., stat. nov. —Type: USA, Arizona, Pima Co., Pima, hills above Rosemont, 13 March 1903 to 23 April 1903, D. Griffiths 4125 (holotype: US 00497458, barcode 00103153). Basionym: Calyptridium parryi A. Gray var. arizonicum J.T. Howell, Leafl. W. Bot. 4: 215. 1945. Synonym: Cistanthe parryi var. arizonica (J. T. Howell) Kartesz & Gandhi, Phytologia TAS 62, 1991; Key to the Calyptridium parryi Species Complex 1. Fruit at maturity gen. 2 longer than abaxial sepal; abaxial sepal <2.2 mm long; Widespread: 2.4.43 5 6e eet dw we oe oes C. monandrum 1’ Fruit at maturity gen. >1 mm wide, <2 longer than abaxial sepal; abaxial sepal 2.2-5 mm long; restricted in range 2. Seeds completely smooth, lacking tubercles; fruit gen. 5-8 mm long ................ C. arizonicum 2' Seeds not completely smooth, with tubercles at least along the margin; fruit 2—5.7 mm long 3:,. Seeds. with tubercles throughout . 5.4... 24 3’ Seeds with tubercles only on the margin C. parryi var. parryi 4. Fruit ca. 2-4 mm long, =| longer than abaxial sepal; San Pedro Martir Mtns., Baja California, Mexico................ ee ea peer epee C.parryi ““martirense”’ 2010] SIMPSON ET AL.: TAXONOMY OF CAL YPTRIDIUM PARRYI 159 4’ Fruit 3-6 mm long, 1—2.4 longer than abaxial sepal; mostly California and Nevada (rarely Utah, Arizona), USA 5. Seeds >0.6 mm in diam.; abaxial fruiting sepal reniform with a wide scarious margin; mostly Sierra Nevada and Panamint ranges, eastern California to western Nevada (rarely Tans PATA ON) ce hac me oe Re Sete Pees Sale eects a ee eee, ty te C. parryi var. nevadense 5S’ Seeds <0.6 mm. in diam.; abaxial fruiting sepals usually ovate with a very narrow scarious margin or none; South Coast Ranges, California......... ACKNOWLEDGMENTS We thank several students at SDSU who helped with plant specimen measurements: Monica Bilodeau, Rose Hipskind, Cathy Luong, Katherine Molfino-Silveira, Ashley Stein, Kana Tran, and Christa Zacharias. We also thank Dr. Jon Rebman, San Diego Natural History Museum, for piquing our interest in this group. LITERATURE CITED ABRAMOFF, M. D., P. J. MAGELHAES, AND S. J. RAM. 2004. Image Processing with ImageJ. Biophotonics International 11:36—42. APG III. 2009. An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG III. Botanical Journal of the Linnean Society 161:105—121. APPLEQUIST, W. L. AND R. S. WALLACE. 2001. Phylogeny of the Portulacaceous cohort based on ndhF sequence data. Systematic Botany 26:406—419. CALIFORNIA NATIVE PLANT SOCIETY (CNPS). 2009. Inventory of rare and endangered plants. Califor- nia Native Plant Society, Sacramento, CA. Website http://cnps.site.aplus.net/cgi-bin/inv/inventory.cgi [accessed 20 July 2010]. CONSORTIUM OF CALIFORNIA HERBARIA. 2009. Data provided by the participants of the Consortium of California Herbaria. Reagent of the University of California, Berkeley, CA. Website http://ucjeps. berkeley.edu/consortium/ [accessed 20 July 2010]. CRONQUIST, A. 1978. Once again, what is a species? Pp. 3-20 in J. A. Ramberger (ed.), Biosystematics in agriculture. Allanheld & Osmun, Montclair, NJ. 1988. The evolution and classification of flowering plants, 2nd ed. New York Botanic Garden, New York, NY. DE QUEIROZ, K. AND M. J. DONOGHUE. 1988. Phylogenetic systematics and the species problem. Cladistics 4:317—338. GREENE, E. L. 1886. Some Californian Polypetalae. Bulletin of the Torrey Botanical Club 13:141—144. GUILLIAMS, C. M. 2009. Phylogenetic reconstruction, character evolution, and conservation in the genus Calyptridium (Montiaceae). M.S. thesis. San Diego State University, San Diego, CA. HAMILTON, C. W. AND S. H. REICHARD. 1992. Current practice in the use of subspecies, variety, and forma in the classification of wild plants. Taxon 41: 485-498. HERSHKOvVITZ, M. A. 1990. Nomenclatural changes in Portulacaceae. Phytologia 68:267—270. . 2006. Ribosomal and chloroplast DNA evi- dence for diversification of western American Portulacaceae in the Andean region. Gayana Botanica 63:13—74. AND E. A. ZIMMER. 2000. Ribosomal DNA evidence and disjunctions of western American Portulacaceae. Molecular Phylogenetics and Evo- lution 15:419-439. C. parryi var. hesseae HICKMAN, J. C. 1993. The Jepson manual: higher plants of California. University of California Press, Berkeley, CA. HINTON, W. F. 1975. Systematics of the Calyptridium umbellatum complex (Portulacaceae). Brittonia 27:197—208. . 1976. The evolution of insect-mediated self- pollination from an outcrossing system in Calyp- tridium (Portulacaceae). American Journal of Botany 63:979—986. Hoover, R. F. 1940. New information regarding Calyptridium and Spraguea. Leaflets of Western Botany 2:222-—225. MAyYR, E. 1969. Principles of systematic zoology. Harvard University Press, Cambridge, MA. MCDADE, L. A. 1995. Species concepts and problems in practice: insight from botanical monographs. Systematic Botany 20:606—622. MISHLER, B. D. AND M. J. DONOGHUE. 1982. Species concepts: a case for pluralism. Systematic Zoology 31:491—503. MORGAN, R., SANTA CRUZ FLORA COMMITTEE. 2005. An annotated checklist of the vascular plants of Santa Cruz County, California. California Native Plant Society, Sacramento, CA. NYFFELER, R. 2007. The closest relatives of cacti: insights from phylogenetic analyses of chloroplast and mitochondrial sequences with special emphasis on relationships in the tribe Anacampseroteae. American Journal of Botany 94:89—101. AND U. EGGLI. 2010. Disintegrating Portula- caceae: a new familial classification of the suborder Portulacineae (Caryophyllales) based on molecular and morphological data. Taxon 59:227—240. , M. OGBURN, AND E. EDWARDs. 2008. Variations on a theme: repeated evolution of succulent life forms in the Portulacineae (Caryo- phyllales). Haseltonia 14:26—36. OGBURN, M. AND E. J. EDWARDS. 2009. Anatomical variation in Cactaceae and relatives: trait lability and evolutionary innovation. American Journal of Botany 96:391—408. PACKER, J. G. 2003. Portulacaceae. Pp. 457-504 in Flora of North America Editorial Committee (eds.), Flora of North America north of Mexico, Vol 4. Oxford University Press, New York, NY. RASBAND, W. S. 1997-2007. ImageJ. U.S. National Institutes of Health, Bethesda, MA. Website http:// rsb.info.nih.gov/ij/ [accessed 20 July 2010]. THOMAS, J. H. 1956. A review of Calyptridium parryi. Leaflets of Western Botany 8:9-11. WATSON, S. 1885. Contributions to American botany. 2. Descriptions of some new species of plants, chiefly from our western territories. Proceedings of the American Academy of Arts and Sciences 20:354-378. WILKEN, D. H. AND W. A. KELLEY. 1993. Calyp- tridium. Pp. 896-898 in J. C. Hickman (ed.), The Jepson manual: higher plants of California. Uni- versity of California Press, Berkeley, CA. 160 APPENDIX | LIST OF VOUCHER SPECIMENS EXAMINED IN THIS STUDY Calyptridium monandrum. Bettys s.n. (JEPS 48483); Gander 1829 (SD 15248); Gander 3302 (SD 16938); Gander 7419 (SD 24927); Gander 7461 (SD 24971); Gander s.n. (SD 11394); King s.n. (SEPS 52574); Moran 14874 (SD 67528); Taylor 14877 (JEPS 105328); Taylor 14886 (JEPS 100387); Taylor 18468 (JEPS 100249); Taylor 6329 (JEPS 101046); Taylor 6457 (JEPS 101043). Calyptridium parryi var. arizonicum. Darrow s.n. (ARIZ 04242): Felger 93-230 (ARIZ 368100, RSA 571207); Moran 12296 (SD 64971); Moran 12438 (SD 64964, UC 1345807); Moran 16872 (ARIZ 185335, RSA 231377, SD 76798, UC 1384544); Moran 16898 (SD 76799); Moran 16914 (ARIZ 185334, SD 76797); Moran 20237 (SD 92613); Moran 20548 (ARIZ 200373, SD 87201); Moran 20604 (ARIZ 200375, CAS 711439, DS 711439, SD 88866); Moran 20634 (SD 88867); Moran 21768 (SD 91249); Moran 8036 (ARIZ 185333, RSA 231378, SD 61137, UC 1384545); Moran 8149 (SD 79003); Peebles 11454 (ARIZ 96164, CAS 324943); Raven 12549 (CAS 531411, UC 115367); Rebman 9976 (SD 157331); Webster 22341 (ARIZ 269658). Calyptridium parryi var. hesseae. Boyd 2645 (RSA 519524); Griffin 3751 (JEPS 73809); Griffin 3799 (JEPS 73810); Guilliams 338 (SDSU 17444); Hesse 1283 (UC 1001290); Hesse 1288 (DS 370154); Hesse 1298 (JEPS 6079); Hesse 1302 (UC 1019332); Hesse 1360 (DS 583117); Hesse 1370 (CAS 388412); Hesse 1942 (DS 583118); Hesse 972 (DS 344294); Howitt 1617 (CAS 531410); Mecmurphy §s.n. (DS 596821); Sharsmith 3381 (DS 266493, UC 724493); Thomas 6001 (CAS 380418, CAS 385058, CAS 408297, DS 380418, DS 385058, DS 587627, UC 1060541): Thomas 6028 (CAS 393436, DS 393436). Calyptridium MADRONO [Vol. 57 parryi “martirense.”” Boyd 2706 (RSA 519494); Moran 14185 (RSA 225157); Moran 24462 (SD 97873); Moran 24489 (SD 97766); Olmsted 4603 (RSA 170797); Reb- man 4174 (SD 142899); Rebman 5407 (SD 144223); Rebman 5579 (SD 145561); Thorne 60834 (RSA 346089); Witham 384 (SD 74689). Calyptridium parryi var. nevadense. Beatley 4079 (DS 611973); Beatley 4122 (DS 591783); Beatley 4373 (DS 591784); Beatley 5732 (UTC 122167); Bostick 5474 (DS 591786); Christy 915B (ASU 200628); DeDecker 1481 (RSA 618440); De- Decker 4336 (RSA 273689, RSA 621364); DeDecker 5448 (RSA 625413); DeDecker 6244 (RSA 627058); Duran 1568 (UC 1297315); Henrickson 17838 (RSA 292001); Hoffman s.n. (RSA 466626); Howell 3912 (CAS 180644); Howell 42888 (CAS 903599); Howell 42894 (CAS 911905); Keck 559 (UC 423468); Kerr 2 (CAS 318495); Morefield 3521 (RSA 387582, UC 1535078); Morefield 4439 (ARIZ 286380, RSA 437274, UC 1545438); Munez 14802 (RSA 231264); Munez 14864 (CAS 366672, RSA 231261); Munez 2630 (RSA 229494); Pollard s.n. (CAS 415968); Romspert 424 (RSA 552078); Stewart s.n. (UTC 237486); Taylor 6429 (JEPS 101044); Taylor 6507 (JEPS 101042); Tiehm 13605 (CAS 1013547, RSA 667225); Wallace K152 (RSA 466627); Zembal 468 (RSA 274018). Calyptridium parryi var. parryi. Duran 1568 (UC 1297315); Elmer 3948 (CAS 141555, DS 45657); Guilliams 381 (SDSU 17439); Hannon s.n. (SD 135302); Hardham 1867 (CAS 414369); Jones s.n. (DS 154025); Munz 5726 (RSA 13478, UC 851936); Munz 6004 (UC 218257); Parish 10964 (DS 92647); Parish 1800 (UC 192468); Parish 3081 (UC 82802); Parish 3725 (UC 7758); Parish 800 (DS 92645); Peirson 3124 (UC 501529); Peirson s.n. (RSA_ 13534); Peirson s.n. (RSA 65395); Roos 3035 (UCR 24222); Tilforth 41066 (RSA 225930); Wheeler s.n. (RSA 620156); Wilder 942 (UC 130501). MADRONO, Vol. 57, No. 3, pp. 161—169, 2010 DO NATIVE ANTS PLAY A SIGNIFICANT ROLE IN THE REPRODUCTIVE SUCCESS OF THE RARE SAN FERNANDO VALLEY SPINEFLOWER, CHORIZANTHE PARRYI VAR. FERNANDINA (POLYGONACEAE)? C. EUGENE JONES, YOUSSEF C. ATALLAH, FRANCES M. SHROPSHIRE, JIM LUTTRELL, SEAN E. WALKER, DARREN R. SANDQUIST, ROBERT L. ALLEN, JACK H. BURK, AND LEO C. SONG, JR. Department of Biological Science, California State University, Fullerton, CA 92831 ceyones@fullerton.edu ABSTRACT Previous field studies of the reproductive biology of the San Fernando Valley spineflower, Chorizanthe parryi var. fernandina (S. Watson) Jeps. suggested that pollination by ants might be an important feature of this endangered polygonous taxon. This conclusion was based on observations that native ants were abundant floral visitors and constant to this species. We conducted the current study to explore more closely the possibility that native ants were facilitating pollination and resulting in viable seed set. Based on our data, ants can indeed be effective pollinators of spineflower. Fruit set was 57% higher in flowers exposed to ant visitation, compared to 27% in control flowers where ants were excluded. Further, a 25.7% germination rate was observed for achenes produced in the absence of ants, in contrast to a 61% rate in those produced in the presence of ants. We suggest that ant pollination may be more prevalent in drier climates, ant production of inhibitory substances may not be a severe limitation to their function as pollinators, invasive Argentine ants may pose a threat to plants pollinated by ants, and self-pollination may not be a negative attribute for ant pollinated plants. Key Words: Allee effect, ant pollination, bet-hedging, Chorizanthe parryi var. fernandina, mixed mating strategies, Polygonaceae, San Fernando Valley spineflower, selfing. Our recent studies have investigated a variety of factors related to the reproductive success of Chorizanthe parryi S. Watson var. fernandina (S. Watson) Jeps., the San Fernando Valley spine- flower (SFVS), an endangered California species formerly thought to be extinct (Jones et al. 2009, C. E. Jones unpublished report'). These studies have demonstrated that the SFVS can self- pollinate and possesses a general mixed mating strategy. Abundant fruit is set and it is visited by a variety of potential pollinators, including native ants. Indeed, ants are among the most frequent visitors to the flowers of the SFVS (Jones et al. 2009). This finding prompted the current investi- gation into whether or not ants facilitate success- ful pollination and fruit set for this species. Ants are rarely considered to be effective pollinators (Hdlldobler and Wilson 1990; Peakall et al. 1991) due to their small size, which can allow them to maneuver in and out of flowers without contacting anthers or stigma (Faegri and van der Pil 1979; Inouye 1980), their smooth bodies, which are not well suited to pollen transport (Schubart and Anderson 1978; Puterbaugh 1998), and chemical secretions from the metapleural gland that reduce pollen viability, germination, and pollen tube growth (Beattie et al. 1984; Gomez and Zamora 1992). These chemical secretions ‘Unpublished reports by CEJ are available upon request from the senior author. contain myrmicacin (3-hydroxydecanoic acid, a broad-spectrum antibiotic), which has been shown to disrupt the flow of components to cell wall formation, the function of Golgi vesicles, and mitosis (Iwanami and Iwadare 1978; Nakamura et al. 1982), thereby affecting pollen germination and pollen tube growth. In addition, ants groom themselves frequently, decreasing the likelihood of transferring pollen from one plant to another (Beattie et al. 1985). Rico-Gray and Oliveira (2007) list sixteen plant species in which ant pollination has been demon- strated including two well-documented cases of ant pollination by Peakall et al. (1991) and Puterbaugh (1998). Hickman (1974) previously noted ten adaptive characteristics commonly shared by such ant-pollinated plants: 1) plants are found in hot and dry climates where ant activity is high; 2) nectaries are accessible to short- tongued insects; 3) plants are short or prostrate; 4) there are dense populations of plants with interdigitating branches; 5) few blooms occur at once per plant; 6) the flowers on erect plants are sessile or are found on the surfaces of low-growing matted plants; 7) pollen volume per flower is small; 8) few seeds are produced per fruit; 9) flowers are small with minimal visual attraction; and 10) small amounts of nectar are produced. Hickman (1974) developed this list during his studies on Polygonum cascadense W. H. Baker (Polygonaceae), an ant-pollinated annual found in the hot dry climate of the Western Cascades of 162 Oregon. In populations where the ants were abundant, he showed that plants had 85—100% seed set, whereas greenhouse plants, exposed to flying pollinators only, exhibited 0—7% seed set. Hagerup (1932), Faegri and van der Pil (1979), and Rico-Gray (1989) have also noted that ant pollination is more likely to occur in dry and hot climates, where flying pollinators often are not abundant. Small flowers growing near to the ground with minimal visual attractants may also be associated with ant pollination (Faegri and van der Pil 1979; Gomez et al. 1990a, b; Garcia et al. 1995; Proctor et al. 1996). Also, among the described examples of an ant- pollinated species is Ptilotrichum spinosum (L.) Boiss. (Brassicaceae), a low-growing, woody plant found in the Sierra Nevada of southern Spain (Gomez et al. 1990a, b). This species bears small hermaphroditic flowers that are frequently visited by ants and exclusion experiments were employed to study the possible effects of such visits. During our previous investigations, we noted that the SFVS appears to share many of the characteristics observed in these published stud- ies, with the exception of number 5 (few blooms occur at once per plant) on Hickman’s (1974) list. Furthermore, the diameter of the SFVS floral tube is only slightly larger than the head widths of the native ants frequently found visiting the plant (C. E. Jones unpublished report), a characteristic noted as important by L. LaPierre (unpublished report). Additionally, these native ants were observed moving in and out of the flowers and did contact both the anthers and stigma in the process (C. E. Jones unpublished report), sup- porting the suggestion that ants may indeed be significant pollination vectors of the SFVS. Based upon the high visitation rates document- ed by Jones et al. (2009, C. E. Jones unpublished report) and Wyatt and Stoneburner (1981), we predict that a significantly higher fruit-set will occur in the plants exposed to ant vectors compared to plants where all vectors are exclud- ed. A previous study showed that a fruit set of about 25% occurs even when all pollination vectors are excluded (Jones et al. 2009, C. E. Jones unpublished report). A significantly higher fruit-set in the ant vector exposed group than that found in the absence of all pollinators would strongly suggest that ant species, specifically Dorymyrmex insanus, play a significant role in the pollination biology of the SFVS. Since at least three different species of native ants were very common visitors to the flowers of the SFVS at the Ahmanson and Newhall Ranch sites (Jones et al. 2009), we decided that follow-up studies were warranted to address the following questions: 1) Is the SFVS adapted for ant pollination? 2) Do ants serve as effective pollina- tors in the absence of other vectors? 3) Is any seed produced by ant pollination viable? 4) Can this MADRONO [IMol-S7 species self without a vector? 5)What are the possible evolutionary implications of ant pollina- tion for the SFVS? MATERIALS AND METHODS Plant Species Formerly distributed in Southern California from Lake Elizabeth in Los Angeles Co. to near Del Mar in San Diego Co. (Munz and Keck 1959; C. E. Jones unpublished report; Glenn Lukos Associates, Inc. unpublished report), Chorizanthe parryi var. fernandina, the San Fernando Valley spineflower (SFVS) is an herbaceous annual found within coastal sage communities at eleva- tions below 350 m (Munz and Keck 1959; C. E. Jones unpublished report). After being considered extinct for a time (Hickman 1993), the SFVS has been found in two locations (the Ahmanson and Newhall Ranches — sites of our previous studies), where it occurs primarily in dry, sandy places within coastal sage in dense patches of several hundred plants (CNDDB 2001). It is currently designated as a List 1B.1 plant (Rare, Threatened, or Endangered in California or Elsewhere; seriously endangered in California) by the Cali- fornia Native Plant Society and is State-listed Endangered (CNPS 2001) and a Federal candi- date for similar listing (CNPS 2005). Stems of the SFVS mostly spread horizontally from the base to form a low, flat-topped, grayish plant 0.2—0.8 (1) dm high and 0.5—4 (6) dm across (Jepson 1925; Reveal 1989). The predominantly sessile, single-flowered involucres are more or less openly distributed in small clusters (Munz and Keck 1959) at branchlet ends (Jepson 1925) and are urn-shaped, bearing six bracts and three awns (Reveal 1989; see Fig. 1). These involucral awns are straight rather than hooked in the SFVS, a trait that distinguishes C. parryi var. fernandina from the more widely distributed C. parryi var. parryi (Reveal 1989). The sessile flowers are 2.5—3 mm long with a greenish-white tube and six white, sparsely hairy lobes, occurring in two series of three (Reveal 1989; Hickman 1993). Nectar is present around the base of the ovary and between the filaments. The flowers are protandrous (Taylor-Taft 2003) and are produced in late spring, April—June (Munz and Keck 1959). Voucher SFVS speci- mens were deposited in the Fay A. MacFadden Herbarium (MACF) at California State Univer- sity, Fullerton, CA. Ant Species The small, medium brown pyramid ant, Dorymyrmex insanus (Hymenoptera: Formicidae; Dolichoderinae) was selected as the model ant for our investigations since it was the dominant ant Fic. 1. € ce 4 éi 012 3mm Photo of Chorizanthe parryi ssp. fernandina with pollinator. (A) Unopened flower bud. (B) Open flower with dehiscing anthers — flowers protandrous. (C) Open flower with receptive stigma. (D) Post-pollinated flower — perianth retained. (E) Pollen on head of Pyramid Ant, Dorymyrmex insanus. Photo by Robert L. Allen. visitor to the SFVS flowers at the Ahmanson Ranch (C. E. Jones unpublished report) and, on our preliminary visits to the Newhall Ranch, colonies of this ant were found near, although not visiting, the populations of the SFVS that we subsequently investigated there (C. E. Jones unpublished report). Snelling (1995) describes the Dorymyrmex insanus worker neotype measurements as: head length 0.90 mm, head width 0.79 mm, scape length 0.87 mm, eye length 0.26 mm, and total length 3.1 mm. Most importantly, Snelling (1995) further notes that the entire head (except the clypeus, frontal area and gena), mesosoma and gastral terga are pubescent, which would provide a surface where pollen grains could adhere. These ants are predaceous, but are also attracted to sugary substances like honeydew and, presum- ably, nectar (Wheeler and Wheeler 1973). Experimental Design This controlled study was carried out between May and July of 2004 and between May and July of 2005 and consisted of two enclosures each composed of two sections. The top portion, measuring 91.44 cm X 30.48 cm x 10.16 cm, consisted of four pine boards and a 1.59 mm hardware mesh top allowing the entry of light. The bottom of each enclosure consisted of four pine boards and a plywood base of equal size to the top and contained approximately 4 in. of Sta- Green Premium Container Mix with Fertilizer (United Industries, Atlanta, GA) with more or less 7 mm of commercially available sterile sand spread evenly over the entire surface area. As an added precaution, these screened enclosures were chemically treated using ““Cooks Ant Barrier”, a commercially available insect repellent. The sides of the enclosures were initially sponged with the solution until the wood was heavily saturated and this treatment was repeated at regular intervals (Luttrell 2006). SFVS achenes were extracted from plants collected from the Ahmanson Ranch site 3 (34°25.12'’N, 118°35.14’W) during the first week of April 2004 and held in refrigerated storage until removal from the inflorescences. The inflorescences were first wetted with de-ionized water and placed in a strainer, where rotational hand pressure was applied to the wet clumps and the loosened seeds settled into a bowl of de- ionized water. The collected achenes were then rinsed five times again with de-ionized water in order to leach out any possible chemical germi- nation inhibitors (Luttrell 2006). 164 The extracted achenes were divided into two equal groups and each group (approximately 100 fruits per group) was evenly planted in a black plastic tray (43.5 cm square X 60 mm deep) filled with same premium grade potting soil as before and top layered with the same commercially produced sterile sand. Planting was completed during the last week of April 2004 and 2005. Subsequent germination took place in a con- trolled environment (see details below) and began to occur after 9 d and was considered complete after 14 d. In 2004, seedlings were allowed to grow and resulted in 32 seedlings surviving in one flat and 27 in the second flat. In 2005, seedlings were allowed to grow until overcrowding prompted the removal of all but 40 plants per container. The SFVS plants were next placed in the two separate screened enclosures, 32 plants in Enclo- sure 1, and 27 in Enclosure 2. Finally, the plants, trays and enclosures were placed in an indoor controlled setting equipped with timed fluores- cent lighting (grow lamps, T-12, 40 watt, Sylvania fluorescent tubes, approximately 61 cm long). Pollination and Fruit Production Observations For each trial, approximately 500 sterile female Dorymyrmex insanus worker ants were collected from the Ahmanson Ranch site 3 (34°25.12'N, 118°35.14’W) at the end of the second week of May 2004 (first trial) or the second week of May 2005 (second trial) and were introduced into Enclosure 1 — Experimental, whereas no vectors were introduced into Enclosure 2 — Control. The screening and chemical treatment, combined with the indoor controlled setting, ensured that no outside pollination vectors would be capable of entry. This extra precaution was taken to rule out the possibility that some smaller flower-visiting potential pollinators might be able to gain access to the flowers in Enclosure 2. Additionally, the chemical treatment prevented the escape of the introduced potential ant pollination vector, Dorymyrmex insanus. In each trial, the number of flowers on each plant and subsequent seed set per flower were counted in both enclosures. After all plants had died in each of the two enclosures, each individual plant was harvested and seeds from these plants were removed and counted. To assess seed viability, three hundred achenes from Enclosure 1 — Experimental (with ants) and three hundred from Enclosure 2 — Control (with no potential vectors) were each divided into replicates of 15 achenes each and placed on moistened 38 Ib., 8.9 cm circles of regular seed germination paper (Anchor Paper Company, St. Paul, MN) in 100 x 15 mm Fisherbrand disposable sterile Petri dishes (Fisher Scientific, Los Angeles Office, Tustin, CA). A total of 40 Petri dishes (replicates) were utilized, with 15 achenes from each enclosure. MADRONO [Vol. 57 Each Petri dish was watered with 5 ml of deionized water and placed in an individual Ziploc® one quart storage bag (S. C. Johnson & Sons, Inc., Racine, WI) and randomly placed in one of four Percival Model E-30B growth chambers (Percival Scientific, Inc., Perry, IA). Each growth chamber was then programmed for 11 hr of daylight with 15C daytime temperature and 10C nighttime temperature. Germination was monitored and recorded for each Petri dish replicate daily from 12 December 2005 through 6 January 2006. Statistical Analysis Data on post-experimental seed set were analyzed using a paired-t test (Excel). Data from all 40 replicates from the seed viability experi- ment were pooled into those produced with ants (20 replicates) versus those produced without ants (20 replicates) and then were compared using a chi-square goodness of fit test. RESULTS Pollination and Fruit Production We found a significant difference between fruit set in control and pollinator exclosure treatments in 2004 (P < 0.001, t = 20.387, df = 38) and 2005 (P < 0.001, t = 24.612, df = 38, Table 1). Approximately 22% of the SVFS flowers within the control group, which lacked ants, set fruit without a vector as compared to 78% fruit-set in the Experimental flowers exposed to the ant species Dorymyrmex insanus. Fruit set in the presence of the ant Dorymyrmex insanus 1s significantly higher by approximately 56% in the 2004 trial, and is also significantly higher by about 57% in the 2005 trial (Table 1). These differences in seed set occurred despite the fact that the number of flowers produced per plant was not significantly different for the 2004 trial (P < 0.05, t = 0.82, df = 1) or for the 2005 trial (P < 0.05, t = 0.28, df = 1), Table 1). We found significant differences in germination under controlled growth chamber conditions of achenes harvested from each enclosure experiment (STATS). Of the 300 achenes produced in control Enclosure 2 (1.e., in the absence of any vector, 77 seeds germinated (25.7% germination). In con- trast, of the 300 achenes produced in Experimental Enclosure | (i.e., in the presence of ants), 183 germinated (61% germination). Achenes produced without a vector were less than half as likely to germinate than those produced in association with ant vectors (X° = 43.22, P < 0.01, df = 1). DISCUSSION In terms of pollination interactions, of the ten characteristics noted by Hickman (1974) in -his 2010] TABLE |. JONES ET AL.: ANTS AS POLLINATORS — SAN FERNANDO SPINEFLOWER 165 COMPARISON OF NUMBER OF FLOWERS PRODUCED PER PLANT AND FRUIT SET PER PLANT FOR THE Two ENCLOSURES FOR THE 2004 AND 2005 TRIALS. Difference in percent fruit set is significant at P < 0.001 for both 2004 and 2005 trials (t = 20.387, df = 38) for the 2004 trial and (t = 24.612, df = 38) for the 2005 trial. Note: difference in average number of flowers per plant is not significant for either trial year, (P < 0.05, t = 0.82, df = 1 for 2004 and P < 0.05, t = 0.28, df = 1 for 2005. Total produced on all plants Character Enclosure | with ants present. (n = 32 for 2004 trial vs. n = 40 for 2005 trial). Number of flowers produced — (2004) 2977 Number of fruit produced — (2004) 1922 Number of flowers produced — (2005) 4422 Number of fruit produced — (2005) 3728 Enclosure 2 without ants present. (n = 27 for 2004 trial vs. 40 for 2005 trial). Number of flowers produced — (2004) 2480 Number of fruit produced — (2004) q23 Number of flowers produced — (2005) 448 1 Number of fruit produced — (2005) 1196 description of what has been termed the “ant pollination syndrome” (Hickman 1974; Faegri and van der Pil 1979), all except number 5 (few co-occurring blooms per plant) were fulfilled in the SFVS. What we found in the SFVS may significantly increase the floral display, which would seem to be adaptive for attracting a diverse group of flying insects that we found (Jones et al. 2009). Post-pollination retention of the perianth (see Fig. 1) also serves to increase the floral display (Jones and Cruzan 1999). In other studies, such as those by Garcia et al. (1995) on Borderea pyrenaica Miégev. (Dioscor- eaceae) and Mayer and Gottsberger (2002) on Arenaria serpyllifolia L. (Caryophyllaceae), there was substantial or complete congruence with all ten tenets postulated by Hickman (1974) as characteristic of myrmecophilous species. Data from our previous observational surveys clearly support the hypothesis that ants play a significant role in the pollination biology of this taxon (Jones et al. 2009, C. E. Jones unpublished report). In terms of the entire flowering season, ants accounted for 51% of visitors and 37% of visits in the Ahmanson Ranch study (C. E. Jones unpublished report) and 21% of the visitors and 34% of the visits in the Newhall Ranch study (C. E. Jones unpublished report). Ants were especially important during early and late por- tions of the season at the Ahmanson Ranch (C. E. Jones unpublished report), providing 84% of the visitors and 54% of visits during the former period and 77% of visitors and 71% of visits during the latter. More visits were made by the ant species Dorymyrmex insanus (3711 of 9830 or 38%) than by any other ant taxon. Seasonal mean number of SFVS flowers visited per Dorymyrmex individual per observation period was 3.1 in that study. A second ant species, Solenopsis xyloni, was also a prominent visitor albeit in much smaller numbers (257 visits) (C. E. Jones Range Average Standard per plant per plant deviation (in %) (in %) per plant 49-132 109.1 23:6 51-97 78.3 152 51-144 110.6 225 55-100 84.3 13.8 42-138 103-9 24.7 14.5—33.8 2271 6.3 44-142 112.0 22.3 16.8—38.5 26.7 a2 unpublished report). Yet a third species of ant (Forelius mccooki) was an important visitor at the Newhall Ranch (C. E. Jones unpublished report). Photographic evidence (see Fig. 1 — note pollen being carried by the individual of D. insanus) and SFVS pollen removed from collected ant speci- mens support visual observations that pollen is indeed being carried by ant visitors (C. E. Jones unpublished report). Although we examined a relatively small sample of ants captured on SFVS flowers at the Ahmanson Ranch, those that were examined had pollen loads that were 98% specific to the flowers of the SFVS, indicating that individual ants were purposefully visiting these plants for nectar and in the process, picking up pollen and very likely facilitating the successful reproduction of the SFVS (C. E. Jones unpublished report). Jones and colleagues (C. E. Jones unpublished report) found that the ant species Forelius mccooki visited SFVS flowers at the Newhall Ranch (C. E. Jones unpublished report) and showed that 13 of 17 individuals collected on SFVS flowers carried one or more pollen grains of the SFVS. Further, of the 13 that carried pollen, nine bore only SFVS pollen and were, therefore, deemed 100% con- stant to the SFVS. The remaining four specimens carried mixed loads but all included some SFVS pollen (C. E. Jones unpublished report). We demonstrated that individuals of the ant species Dorymyrmex insanus were effective pollen transporters and facilitated fruit-set in over 50% more flowers than in the case where all pollina- tors were excluded. Furthermore, ants move extensively among flowers on any given plant and, in doing so, likely promote geitonogamy. Therefore, it is highly likely that the ants promoted an increase in overall selfing, as well. However, ants also move between flowers on different plants when those plants are in close proximity to one another. In doing so, ants 166 MADRONO facilitate xenogamy. Therefore, ants very likely do facilitate some level of out-crossing, though that level is likely to be much less than a bee pollination vector lke the honey bee. The percentage of viable fruit, as judged by germina- tion of the SFVS achenes produced by ant visitation, was 61% - over double the germination rate found for achenes produced in flowers by selfing without a vector. The potential capacity to set seed in the absence of any pollination vector further increas- es the likelihood of successful SFVS reproduc- tion. A fruit-set of approximately 33% is normally expected for out-crossing plant species (Sutherland 1986). Values above that number are suggestive of a plant that is at least a facultative selfer. Given a fruit set close to 60% in the SFVS and given the small size of the flowers, the SFVS would seem to be a facultative selfer. However, it is unlikely that it would be an obligatory selfer since the flowers are protandrous. The occurrence of selfing without a vector within a single flower might not be possible unless the pollen remains viable until the stigma 1s receptive and, in addition, the anthers are positioned close enough to the stigmatic surface to facilitate pollen transfer. Our data would indicate that at least some portion of the pollen does remain viable long enough to result in selfing without a vector. Further, the anthers are positioned close enough to the stigmatic surface to result in transfer of pollen without the necessity of a vector. Given the manner in which ants and other small insects move among the SFVS flowers, vector-assisted selfing should be considered likely. Our data indicate that 27% of the seed set occurred within plants in which all potential pollinators were prohibited from visiting the flowers. The significantly lower number of fruits produced by the selfing treatment versus in the enclosure with ants indicates that the SFVS is not a productive selfer without a vector. However, 27% seed set is probably sufficient to ensure reproductive success of the SFVS in unfavorable years. Furthermore, in a germination test carried out on a single sub-sample of the seeds produced by selfing without a vector, approximately one third (47 of 150 or 31%) did germinate. Our data seem to indicate that achenes produced by selfing without a vector have a lower viability, as judged by germination rates, than achenes produced with the aid of ants. Unresolved, then, is the question of the viability and/or fertility of adult plants produced from such selfing. Our results support the postulate by Stebbins (1957) that geographically restricted plants are likely to be self-compatible. Self-pollination or autogamy appears to be quite common in ant-pollinated species (examples include Wyatt and Stoneburner 1981; Gomez et [Vol. 57 al. 1990a; Peakall and Beattie 1991; Gomez and Zamora 1992, 1999; Gomez et al. 1996; Bosch et al. 1997; Gomez 2000, 2002; Buide and Guitian 2002; and Kawakita and Kato 2002). Why is selfing so common in these taxa? In part, this may be attributed to the relatively short distances that crawling ants travel between flowers and to the fact that the frequency distributions of these visits are strongly leptokurtic (Wyatt and Stoneburner 1981). As was pointed out by Gomez (2002) in his study of selfing in Euphrasia willkommii Freyn (Scrophulariaceae), an endemic alpine species of the Spanish Sierra Nevada, selfing may represent an “ecological mechanism to ensure successful reproduction in a harsh environment where pollinator availability is low.” Affre et al. (1995) also suggested that a scarcity of pollinators, as well as fragmentation and isolation of popula- tions, could increase the frequency of self- pollination in a Mediterranean endemic Cycla- men. We would only add that in unpredictable environments, such as Mediterranean ecosys- tems, where annual rainfall seems to be a major limiting factor, differences in annual survivorship can result in dramatic fluctuations in plant and pollinator densities. Small populations resulting from such unpredictable conditions are more likely to experience Allee effects due to pollen limitations caused by reduced mate availability (Groom 1998; Moeller 2004). This leads to uncertain reproductive success if the plant requires a pollination vector and, therefore, has important consequences regarding the population dynamics of the species (Clauss and Venable 2000). In the SFVS, significant annual variation in population numbers has been recorded and reflects variation in seasonal rainfall (Dudek, Dudek and Associaties, Inc., and Sapphos Environmental, Inc. unpublished reports). This variation in plant densities also seems to be associated with substantial variation in pollinator availability both in terms of species composition and total numbers (Jones et al. 2009, C. E. Jones unpublished report). Therefore, non-facilitated autogamy would appear to be functioning as a bet-hedging pollination strategy in arid regions, similar to the variation in seed germination strategies found in desert annuals (Clauss and Venable 2000). Such a strategy in the annual SFVS assures some successful fruit/seed set in the face of potentially reduced numbers of both plants and pollinators (reproductive assurance as discussed by Jarne and Charlesworth 1993). The preservation of genotypes that are well adapted to survival and reproduction under drought conditions may be another advantage of selfing in the SFVS (Jarne and Charlesworth 1993). In this regard, it is interesting to note in Peakall and Beattie’s (1991) study of ant-facilitated 2010] selfing in the orchid Microtis parviflora R. Br. that, although their electrophoretic analysis indicated that the populations were highly inbred, some outcrossing was occurring. They indicated that the ant foraging witnessed on this species would have yielded a mixed mating system similar to those reported for a variety of other insect pollinators (e.g., Vogler and Stephenson 2001). Thus, in the case of a facultative selfer like the SFVS, ants appear to provide a reliable pollination vector that ensures successful repro- duction via both selfing (fitness in times of stress) and outcrossing (production of genetic variation for possible adaptation to future environmental fluctuations). Our data indicate that during harsh, dry, growing seasons, the SFVS may survive by producing a significant number of progeny via auto-fertility or by utilizing various native ant species as major pollination vectors. A decrease in the number of floral visitors or the production of a significant number of progeny via selfing with or without a vector would have important genetic implications in terms of inter-population gene flow (Ellstrand and Elam 1993; Jarne and Charlesworth 1993). A detailed analysis of the population genetics of this species throughout its extant range would help determine its genetic status and to establish management strategies to maintain or enhance population genetic diversity. Although ants may not visit as many flowers per foraging bout as other pollinators, they are present in greater abundance and are clearly superior in facilitating pollination. In addition, ants tend to be consistently present throughout the flowering season in contrast to other visitors that often display more limited availability, frequently appearing only during the peak of the flowering season. Certainly this is the case in the SFVS (Jones et al. 2009). Further, ant- pollinated systems are low-energy systems allow- ing for a reduced energetic commitment (e.g., in the production of very small quantities of nectar per flower as was documented in the SFVS — Jones et al. 2009) on the part of the plant species (Hickman 1974). Such energetic savings could be extremely important to the survival of plants living in unpredictable environments and may aid in ensuring that the plant will have sufficient resources for at least some reproduction even in very dry years (Heinrich and Raven 1972; Svensson 1985). A potential problem of concern for ant pollinated species like the SFVS is the invasive alien Argentine ant. Argentine ants (Linepithema humile) are considered to be among the top 100 worst invasive alien species globally (Lowe et al. 2000). A major practical concern related to ant- pollinated systems is the question of how interactions between invasive Argentine ants and native ants will affect the reproductive JONES ET AL.: ANTS AS POLLINATORS — SAN FERNANDO SPINEFLOWER 167 biology of these plants (Lack 2003). Argentine ants have been shown to significantly reduce the foraging success of native ant species by being more efficient at exploiting food sources and thereby displacing native ant species from areas where they successfully invade (Human and Gordon 1996, 1997; Suarez et al. 1998). Al- though, Suarez et al. (1998) found that Argentine ants normally penetrated only approximately 100 m into Mediterranean type ecosystems in San Diego Co., CA, it is unclear whether the process of invasion 1s not yet complete or whether they were not penetrating into these habitats because they lacked water. Holway (2005) found that Argentine ants were able to move into coastal sage scrub habitats in southwestern San Diego Co., CA by using riparian corridors; however, their numbers decreased with increasing distance from anthropomorphic influences as they moved into drier scrub habitats. Although Argentine ants apparently have not yet successfully invaded the drier areas in the Mediterranean ecosystems found in southern California, they have invaded and colonized other dry habitats in Hawaii (Cole et al. 1992; Reimer 1993), the fynbos in South Africa (Giliomee 1986; Lack 2007), and the matorral in Chile (Fuentes 1991), so they may very well eventually invade the similar dry habitats in Southern California. No data are currently available regarding whether or not Argentine ants could serve as an effective replacement for native pollination vectors for species like the SFVS. They may be less effective than the native ants because of their much smaller size, which might result in their not contacting both anthers (for pollen pickup) and/or stigmas (for pollen deposition) in the SFVS. Clearly, the issue of potential Argentine ant impacts on the reproduc- tive biology of plants like the SFVS, which seems to depend on ants for a significant part of their reproductive effort, requires further investigation. We show that native ants are effective pollina- tors of SFVS, transporting pollen between flowers and leading to significant increases in seed set and seed viability, compared to when ants are excluded from flowers. Taken together, our results provide evidence that although the SFVS can set fruit without a vector, fruit set in the presence of the ant Dorymyrmex insanus 1s significantly higher. Results from the current investigation contribute significant information to the SFVS database and should prove useful to conservation biologists charged with protecting this rare taxon. Based on our study and a review of the literature, we stress the importance of investigat- ing the reproductive biology of other plants found in dry, harsh, variable environments and predict that many more species will be found to receive a significant portion of the pollen 168 deposited on their stigmatic surfaces via ant vectors. Thus, a better understanding of the role of ants in the successful reproduction of native plants of dry environments is crucial to our comprehension of how these communities func- tion and the ecological tradeoffs made under the variable and unpredictable environmental condi- tions that prevail in such habitats. ACKNOWLEDGMENTS The authors thank Kelly Argue for laboratory assistance, the staffs at the Ahmanson and Newhall Ranches, Dudek and Associates, Inc., Sapphos Envi- ronmental, Inc., and Lenora O. Kirby of the Las Virgenes Institute for their assistance, as well as Roy Snelling, for the ant identifications. Contract grant support from Newhall Land and Farming Corp., Newhall, CA, via Dudek and Associates, Inc., En- cinitas, CA and from the Ahmanson Corporation via Sapphos Environmental Inc., Pasadena, CA is grate- fully acknowledged. LITERATURE CITED AFFRE, L., J. D. THOMPSON, AND M. DEBUSSCHE. 1995. The reproductive biology of the Mediterra- nean endemic Cyclamen balearicum (Primulaceae). Botanical Journal of the Linnean Society 118:309—330. BEATTIE, A. J., C. TURNBULL, R. B. KNOX, AND E. G. WILLIAMS. 1984. Ant inhibition of pollen function: a possible reason why ant pollination is rare. American Journal of Botany 71:421—426. T. HOUGH, S. JOBSON, AND R. B. KNox. 1985. The vulnerability of pollen and fungal spoes to ant secretions: evidence and some evolu- tionary implications. American Journal of Botany 72:606-614. BoscHu, J., J. RETANA, AND X. CERDA. 1997. Flowering phenology, floral traits and pollinator composition in a herbaceous Mediterranean plant community. Oecologia 109:583—591. BUIDE, M. L. AND J. GUITIAN. 2002. Breeding system in the dichogamous hermaphrodite Si/ene acutifolia (Caryophyllaceae). Annals of Botany 90:691—699. CALIFORNIA NATURAL DIVERSITY DATA BASE (CNDDB). 2001. Chorizanthe parryi var. fernan- dina. California Department of Fish and Game, Sacramento, CA. CALIFORNIA NATIVE PLANT SOCIETY (CNPS). 2001. Inventory of rare and endangered plants of California, 6th ed. Rare plant scientific advisory committee, David P. Tibor, Convening Editor. California Native Plant Society, Sacramento, CA. . 2005. Inventory of rare and endangered plants (online edition, v6-05a). California Native Plant Society, Sacramento, CA, Website http://www. cnps.org/inventory/ [accessed 21 January 2005]. CLAuss, M. J. AND D. L. VENABLE. 2000. Seed germination in desert annuals: an empirical test of adaptive bet hedging. American Naturalist 155:168—186. COLE, F. R., A. C. MEDEIROS, L. L. LOOPE, AND W. W. ZUEHLKE. 1992. Effects of the Argentine ant on arthropod fauna of Hawaiian high-elevation shrub- land. Ecology 73:1313—1322. MADRONO [Vol. 57 ELLSTRAND, N. C. AND D. R. ELAM. 1993. Population genetic consequences of small population size: implications for plant conservation. Annual Re- view of Ecology and Systematics 24:217—242. FAEGRI, K. AND L. VAN DER PIJL. 1979. The principles of pollination ecology, 3rd revised ed. Pergamon Press, Oxford, UK. FUENTES, E. R. 1991. Central Chile: how do introduced plants and animals fit into the landscape? Pp. 43— 49 in R. H. Groves and F. di Castri, (eds.), Biogeography of Mediterranean invasions. Cam- bridge University Press, Cambridge, UK. GARCIA, M. B., R. J. ANTOR, AND X. ESPADALER. 1995. Ant pollination of the palaeoendemic dioe- cious Borderea pyrenaica (Dioscoreaceae). Plant Systematics and Evolution 198:17—27. GILIOMEE, J. H. 1986. Seed dispersal by ants in the Cape flora threatened by Jridomyrmex humilis (Hymenoptera: Formicidae). Entomologia Gener- alis 11:217-219. GOMEZ, J. M. 2000. Effectiveness of ants as pollinators of Lobularia maritime: effects on main sequential fitness components of the host plant. Oecologia 122:90-97. . 2002. Self-pollination in Euphrasia willkommii Freyn (Scrophulariaceae), an endemic species from the alpine of the Sierra Nevada (Spain). Plant Systematics and Evolution 232:63—71. AND R. ZAMORA. 1992. Pollination by ants: consequences of the quantitative effects on a mutualistic system. Oecologia 91:410-418. AND . 1999. Generalization vs. special- ization in the pollination of Hormathophylla spinosa (Cruciferae). Ecology 80:796—80S. , AND J. A. HODAR. 1996. Experimental study of pollination by ants in Mediterranean high mountain and arid habitats. Oecologia 105:236— 242. , AND A. TINAUT. 1990a. Pollination of Ptilotrichum spinosum by ants: experimental study of the quality component. Pp. 445-446 in G. K. Veeresh, B. Mallik, and C. A. Viraktamath, (eds.), Social insects and the environment: proceedings of the 11th international congress of IUSSI, Banga- lore, India. Oxford & IBH Publishing Co., New Delhi, India. , AND . 1990b. Pollination of Ptilotrichim spinosum by ants: quantity compo- nent. Pp. 447— in G. K. Veeresh, B. Mallik, and C. A. Viraktamath, (eds.), Social insects and the environment: proceedings of the 11th International Congress of TUSSI, Bangalore, India. Oxford & IBH Publishing Co., New Delhi, India. GRooM, M. J. 1998. Allee effects limit population viability in an annual plant. American Naturalist 151:487-496. HAGERUP, O. 1932. On pollination in the extremely hot air at Timbuctu. Dansk Botanisk Arkiv 8 1:1—20. HEINRICH, B. AND P. H. RAVEN. 1972. Energetics and pollination ecology. Science 176:597—602. HICKMAN, J. C. 1974. Pollination by ants: a low-energy system. Science 184:1290—1292. 1993. Polygonaceae. Pp. 854-895 in J. C. Hickman (ed.), The Jepson manual: higher plants of California. University of California Press, Berkeley, CA. HOLLDOBLER, B. AND E. O. WILSON. 1990. The ants. Belknap Press, Cambridge, MA. 2010] HoLway, D. A. 2005. Edge effects of an invasive species across a natural ecological boundary. Biological Conservation 121:561—567. HUMAN, K. G. AND D. M. GORDON. 1996. Exploita- tion and interference competition between the invasive Argentine ant, Linepithema humile, and native ant species. Oecologia 105:405—412. AND . 1997. Effects of Argentine ants on invertebrate biodiversity in northern California. Conservation Biology 11:1242—1248. INOUYE, D. W. 1980. The terminology of floral larceny. Ecology 61:1251—1253. IWANAMI, Y. AND T. IWADARE. 1978. Inhibiting effects of myrmicacin on pollen tube growth and pollen tube mitosis. Botanical Gazette 139:42-45. JARNE, P. AND D. CHARLESWORTH. 1993. The evolu- tion of selfing rate in functionally hermaphrodite plants and animals. Annual Review of Ecology and Systematics 24:441—466. JEPSON, W. L. 1925. A manual of the flowering plants of California. University of California Press, Berkeley, CA. JONES, C. E. AND M. CRUZAN. 1999. Floral morpho- logical changes and reproductive success in Deer Weed. American Journal of Botany 86:273—277. , F. M. SHROPSHIRE, L. L. TAYLOR-TAFT, S. E. WALKER, L. C. SONG, Jr., Y. C. ATALLAH, R. L. ALLEN, D. SANDQUIST, J. LUTTRELL, AND J. H. BURK. 2009. Reproductive biology of the San Fernando Valley Spineflower, Chorizanthe parryi var. fernandina (Polygonaceae). Madrono 56:23- 42. KAWAKITA, A. AND M. KATO. 2002. Floral biology and unique pollination system of root holopar- asites, Balanophora kuroiwai and B. tobiracola. American Journal of Botany 89:1164—1170. LACK, L. 2003. Invasive ants: unwanted partners in ant- plant interactions? Annuals of the Missouri Botan- ical Garden 90:91—108. . 2007. A mutualism with a native membracid facilitates pollinator displacement by Argentine ants. Ecology 88:1994—2004. LoweE, S., M. BROWNE, AND S. BOUDLEJAS. 2000. 100 of the world’s worst invasive alien species. Aliens P2120: LUTTRELL, J. 2006. Ant facilitated fruit-set and the fitness of selfed progeny in the San Fernando Valley Spineflower. M.S. thesis. California State University, Fullerton, CA. MAYER, E. AND G. GOTTSBERGER. 2002. Die Auswir- kung von Ameisen auf den Reproduktionserfolg des Quendelblattrigen Sandkrautes (Arenaria ser- pyllifolia, Caryophyllaceae). Botanische Jahrbtcher 124:31-47. MOELLER, D. A. 2004. Facilitative interactions among plants via shared pollinators. Ecology 85:3289— B50 lk Munz, P. A. AND D. D. Keck. 1959. A California flora. University of California Press, Berkeley, CA. NAKAMURA, S., H. MIKE-HIROSIGE, AND Y. IWA- NAMI. 1982. Ultrastructural study of Camellia japonica pollen treated with myrmicacin, an ant origin inhibitor. America Journal of Botany 69: 538-545. JONES ET AL.: ANTS AS POLLINATORS — SAN FERNANDO SPINEFLOWER 169 PEAKALL, R. AND A. J. BEATTIE. 1991. The genetic consequences of worker ant pollination in a self- compatible, clonal orchid. Evolution 45:1837—-1848. , S. N. HANDEL, AND A. J. BEATTIE. 1991. The evidence for, and importance of, ant pollination. Pp. 421-429 in C. R. Huxley and D. F. Cutler (eds.), Ant-plant interactions. Oxford University Press, Oxford, UK. Proctor, M., P. YEO, AND A. LACK. 1996. The natural history of pollination. Timber Press, Inc., Portland, OR. PUTERBAUGH, M. N. 1998. The roles of ants as flower visitors: experimental analysis in three alpine plant species. Oikos 83:36—46. REIMER, N. J. 1993. Distribution and impact of alien ants in vulnerable Hawaiian ecosystems. Pp. 1 1—22 in D. F. Williams (ed.), Exotic ants: biology, impact, and control of introduced species. West- view Press, Boulder, CO. REVEAL, J. L. 1989. The eriogonoid flora of California (Polygonaceae: Eriogonoideae). Phytologia 66: 295-414. RICcO-GRAY, V. 1989. The importance of floral and circum-floral nectar to ants inhabiting dry tropical lowlands. Biological Journal of the Linnean Society 38:173-181. AND P. S. OLIVEIRA. 2007. The ecology and evolution of ant-plant interactions. The University of Chicago Press, Chicago, IL. SCHUBART, H. AND A. ANDERSON. 1978. Why don’t ants visit flowers? A reply to D.H. Janzen. Biotropica 10:310-311. SNELLING, R. 1995. Systematics of Neartic ants of the genus Dorymyrmex (Hymenoptera: Formicidae). Contributions in Science, Natural History Museum of Los Angeles County 454:1—14. STEBBINS, G. L. 1957. Self fertilization and population variability in the higher plants. American Natural- ist 91:337-354. SUAREZ, A. V., D. T. BOLGER, AND T. J. CASE. 1998. Effects of fragmentation and invasion on native ant communities in coastal southern California. Ecol- ogy 79:2041—2056. SUTHERLAND, S. 1986. Patterns of fruit-set: what controls fruit-flower ratios in plants? Evolution 40: 117-128. SVENSSON, L. 1985. An estimate of pollen carryover by ants in a natural population of Scleranthus perennis L. (Caryophyllaceae). Oecologia 66:373—377. TAYLOR-TAFT, L. 2003. The reproductive biology of the San Fernando Valley Spineflower (Chorizanthe parryi var. fernandina (S. Watson) Jepson (Poly- gonaceae). M.S. thesis. California State University, Fullerton, CA. VOLGER, D. W. AND A. G. STEPHENSON. 2001. The potential for mixed mating in a self-incompatible plant. International Journal of Plant Sciences 162:801—805. WHEELER, G. C. AND J. WHEELER. 1973. Ants of Deep Canyon: Colorado Desert California. Philip L. Boyd Deep Canyon Desert Research Center, University of California, Riverside, CA. WYATT, R. AND A. STONEBURNER. 1981. Patterns of ant-mediated pollen dispersal in Diamorpha smallii (Crassulaceae). Systematic Botany 6:1—7. MADRONO, Vol. 57, No. 3, pp. 170-179, 2010 GAS EXCHANGE RATES OF THREE SUB-SHRUBS OF CENTRAL TEXAS SAVANNAS MITSURU FURUYA! AND O. W. VAN AUKEN? Department of Biology, University of Texas at San Antonio, San Antonio, TX 78249 oscar.vanauken@utsa.edu ABSTRACT Savannas are open communities with a woodland and grassland phase. Some species are restricted to one phase or the other with some in both. The central Texas Edwards Plateau is mainly a savanna community with an east to west rainfall gradient and a tree density gradient with both factors being reduced in the west. We examined gas exchange rates of sun and shade leaves of three perennial species of Asteraceae, Simsia calva (bush sunflower), Wedelia texana (hairy wedelia,) and Brickellia cylindracea (brickell-bush). Maximum photosynthetic rates (A; ax), light saturation points (Lga;), ambient light levels, leaf mass per area, leaf conductance, and transpiration rates (E) for sun leaves were significantly greater than shade leaves for all three species. Mean A,,ax rates were 32.8, 23.2 and 21.2 umolCO;/m/’/s for sun leaves of S. calva, W. texana and B. cylindracea respectively. For shade leaves, rates were 21.3, 15.2 and 12.9 umolCO;/m’/s respectively for the same species. Mean Rg (dark respiration) rates were 3.00 umolCO>/m/7/s for sun leaves and 1.1 umolCO>/m/’/s for shade leaves of S. calva. Rg rates for sun and shade leaves of W. texana and B. cylindracea were mostly lower but only significantly different for B. cylindracea. There were no significant differences between sun and shade leaves for the initial slope or quantum yield for any of the species. Leaf mass per area was greater for sun leaves of all species. Ajax rates for sun leaves were high as expected for sun plants, but A,,ax rates for shade leaves were fairly high as well, probably because of the relatively open nature of the sub- shrub canopy. Most factors measured were high and suggest that these plants are sun plants or facultative species. These measurements would help explain why these species are present in the full sun, open grassland phase of these central Texas savanna communities and not below the woodland canopy where their growth and survival would be reduced because of lower light levels. Key Words: CO; uptake, photosynthesis, savannas, shade leaves, sun leaves. Composition and structure of grassland com- munities Worldwide have changed and are con- tinuing to change (Bond 2008; Van Auken 2009). These transformations include grasslands from California through Arizona, New Mexico and central Texas (Van Auken and McKinley 2008; Maze 2009). Many of the changes involve the encroachment or invasion of a few woody species; however, as many as 10% of the herbaceous species in southwestern arid or semiarid grasslands are invaders from all over the World (Van Auken 2009). The composition and structural changes have occurred because of modifications in the level and kind of herbivory, other disturbances and the reduced frequency and intensity of grassland fires. However, some still suggest that global phenom- ena such as elevated levels of atmospheric CO; and concomitant temperature increases are the primary cause of the changes in these grasslands (Van Auken 2009). Regardless of the specific cause or causes, the mechanisms directing or driving the conversion in these grasslands have been difficult to identify. ‘Current address: 7-11-20 Nakagawa, Tsuzuki-ku Yokohama-shi Kanagawa-ken, 224-0001 Japan. > Author for correspondence. Simple as well as complex gradients of rainfall, temperature, light, soil resources, biotic factors or combinations occur throughout the world, deter- mining to a large extent the kind of communities and species of plants that are present in given areas (Begon et al. 2006). Woody plant density and cover change along rainfall gradients with desert shrubs increasing on the dry end of the gradient, woodlands or forests on the wetter end with grasslands or savannas in between (Board- man 1977; Larcher 2003; Breshears 2008). This type of gradient occurs, proceeding from west to east, on the Edwards Plateau in central Texas (Van Auken and McKinley 2008). Although the structure of some of the communities along this rainfall gradient have been described (Van Auken and McKinley 2008; Van Auken and Smeins 2008; Wayne and Van Auken 2008), the reasons that various species occur in defined places and at certain densities along this gradient are not clear and predicting the dynamics of this biphasic system is challenging (House et al. 2003). Three shrubs or sub-shrubs found in the grassland phase of many of these central Texas savanna communities are Simsia calva (Engelm. & A. Gray) A. Gray, (bush sunflower, Aster- aceae), Wedelia texana (A. Gray) B. L. Turner, (=Zexmenia hispida, hairy wedelia, Asteraceae) and Brickellia cylindracea A. Gray and Engelm., 2010] (brickell-bush, Asteraceae) (Correll and Johnston 1979; Enquist 1987; USDA 2009). These taxa are not reported from the woodlands or forest communities (Van Auken et al. 1979; Van Auken et al. 1980; Van Auken et al. 1981). They are usually not dominant but secondary species in these communities. They can be found at high density in some open habitats on dry, shallow soil in areas that have been disturbed (personal observation). They may be found at the edge of woodland or motts, but not below the canopy in these communities. They may require disturbanc- es or low nutrient soils, but no studies were found concerning their requirements for establishment or their successional status. In addition, we have not found any studies of these species that might suggest limiting physiological factors or condi- tions. Physiological differences between plants native to open, full sun habitats compared to those found in shady, understory communities are fairly well known (Begon et al. 2006; Valladares and Niinemets 2008), but gas ex- change rates of the species studied here have not been examined. Sun plants usually have higher photosynthetic rates at high light levels, they light saturate at higher light levels, have higher light compensa- tion points (photosynthetic rate equals respira- tion rate), higher dark respiration rates, higher transpiration rates and higher stomatal conduc- tance rates compared to shade plants (Board- man 1977; Young and Smith 1980; Larcher 2003; Valladares and Niinemets 2008). Some species display what is called adaptive crossover and are capable of acclimating to high or low light environments, thus they could have a broader ecological niche (Givnish 1988; Givnish et al. 2004). Some plants, especially many trees start their growth in the low light environment of the canopy understory (Spurr and Barnes 1973), grow slowly for many years until they grow into a light gap. A light gap is created via a tree fall or plants reach the open, high light above the canopy (Ryniker et al. 2006; Van Auken and Bush 2009). In addition, shade leaves from plants grown in full sun have been used as surrogates for plants grown in shade conditions to understand a species’ ecological requirements (Hamerlynck and Knapp 1994; Furuya and Van Auken 2009). Gas exchange rates and other characteristics of sun and shade leaves of Simsia calva, Wedelia texana and Brickellia cylindracea were measured. Ambient light levels, xylem water potential and soil water content were also measured. Based on the habitats where these species were found, we expected that they were sun plants and would have higher maximum photosynthetic rates, light saturation points, light compensation points, respiration rates, conductance, and transpiration compared to shade-adapted species. FURUYA AND VAN AUKEN: GAS EXCHANGE OF SUB-SHRUBS Wa | METHODS Study Species Simsia calva, (bush sunflower, Asteraceaae), 1s a native, warm-season, perennial, sub-shrub with orange-yellow ray and disk flowers on long peduncles. It grows on disturbed, sand or heavy clay limestone soils and appears drought tolerant, widespread through central, southern, and south- western Texas and southeastern New Mexico (Correll and Johnston 1979; Enquist 1987; Turner et al. 2003). Wedelia texana (=Zexmenia hispida, hairy wedelia, Asteraceae), is a native, warm-season, perennial, sub-shrub. It has bright yellow-orange flowers, blooming from March to December in South Texas, and May to Septem- ber in central Texas (Correll and Johnston 1979; Enquist 1987). It is hardy, grows in disturbed areas and tolerates a variety of limestone soils (USDA 2009). It is found in central and west Texas and northeastern Mexico (Turner et al. 2003). Brickellia cylindracea (brickell-bush, As- teraceae), 1s a perennial, sub-shrub with consid- erable variability. Flowers are yellow, blooming from August through November (Correll and Johnston 1979; Enquist 1987). It is drought tolerant, has a strong preference for limestone soil, open habitats and is only found in central Texas (Turner et al. 2003). Study Area All plants in this field study were found on the southern edge of the Edwards Plateau region of central Texas just below the Balcones Escarpment in northern Bexar Co. (approximately 29°68’N and 98°50'W) (Correll and Johnston 1979; Van Auken et al. 1981; Van Auken and McKinley 2008; Furuya and Van Auken 2009). The Balcones Escarpment is rough, well-drained, with elevations increasing abruptly from approximately 200 m above mean sea level (AMSL) to 500 to 700 m AMSL. Most of the subsurface of the area is Cretaceous limestone, and soils are usually shallow, limestone derived, rocky or gravelly on slopes, and deep in broad valleys and flats (Taylor et al. 1962; NRCS 2006). Soils are dark colored, calcareous with usually neutral or slightly basic pH. Area mean annual temperature is 20.0°C with monthly means ranging from 9.6°C in January to 29.4°C in July (NOAA 2004). Mean annual precipitation is 78.7 cm per year and bimodal, with peaks occurring in May and September (10.7 cm and 8.7 cm, respectively). There is little summer rainfall, high evaporation and high inter-annual variation (Thornthwaite 1931; NOAA 2004). Juniperus-Quercus savanna or woodland is the major vegetation type in the study area and is representative of savannas and woodlands found throughout this region, but higher in woody plant 172 MADRONO density than savanna communities farther to the west (Van Auken et al. 1979; Van Auken et al. 1980; Van Auken et al. 1981; Smeins and Merrill 1988). The high density woody species are Juniperus ashei J. Buchholz (Ashe juniper) and Quercus virginiana Mill. (=Q. fusiformis Small., live oak) followed by Diospyros texana Scheele. (Texas persimmon) and Sophora_ secundiflora (Ortega) DC. (Texas mountain laurel) with other oaks and elms in sheltered habitats or on deeper soil. Associated with these woodlands are rela- tively small grasslands and sparsely vegetated intercanopy patches or gaps (openings between the canopy trees) (Van Auken 2000). Carex planostachys Kunze (cedar sedge, Cyperaceae) (Wayne and Van Auken 2008) is the major herbaceous species below the canopy, but Verbe- sina virginica L. (frost weed, Asteraceae) is found in places on deeper soil (Gagliardia and Van Auken in press). In the grasslands and gaps Aristida purpurea Nutt. var. longiseta (Steud.) Vasey (red three-awn), Bouteloua curtipendula (Michx.) Torr. (side-oats grama), Bothriochloa laguroides (Steud.) Allred & Gould (silver blue- stem), B. ischaemum (L.) Keng (KR bluestem), various other Cy, grasses, and a variety of herbaceous annuals are common (Van Auken 2000; USDA 2009). Measurements Gas exchange rates as a function of light level or photosynthetic flux density (PFD) were measured and plotted for sun and shade leaves of all three species (light response curves) (Hamerlynck and Knapp 1994; Furuya and Van Auken 2009). Five plants or replications were used. Individual sun and shade leaves were measured on each plant. All plants sampled were 0.5 tol.0 m tall. All plants were growing in full sun. Sun leaves were on the outermost, southern facing canopy branches and shade leaves were on the lower, innermost branches of these plants. Shade leaves from full sun plants were used as surrogates for shade plants as reported by others (Hamerlynck and Knapp 1994; Furuya and Van Auken 2009). Light response measurements were made in June within +3 hr of solar noon with a LI-COR® infrared gas analyzer (LI-6400). Irradiances were generated by the LI-COR LED red-blue light source using a light curve program with the LI- COR, a gas flow rate of 400 umol/s, and a CO; concentration of 400 umol/mol. One mature, undamaged, fully expanded leaf per replication and leaf type was used with the 2 xX 3 cm chamber. The LI-COR 6400 was run at approx- imate ambient summer, midday, daytime tem- perature (35°C) and relative humidity (50%), and was calibrated daily. All response data were recorded after at least two minutes when a stable [Vol. 57 total coefficient of variation was reached (<0.3%), usually less than five minutes. All light response curves were started at a PFD of 2000 umol/m’/s for sun leaves and shade leaves and then decreased to the following: 1800, 1400, 1200, 1000, 800, 600, 400, 200, 100, 75, 50, 25, 10, 5, and 0 umol/m?’/s (16 total measurements). Measurements for each species and leaf type included net photosynthesis, stomatal conduc- tance, and transpiration. Repeated measure ANOVA was utilized to determine if significant differences occurred between leaf types. A one way ANOVA was used to determine if net photosynthesis, stomatal conductance, and tran- spiration were significantly different between the PFD’s tested and also between leaf types (Sall et al. 2001). If significant differences were found within a leaf type, Tukey-Kramer Honestly Significant Difference test was used to determine where significant differences occurred. Shapiro- Wilks tests were used to test for normal distributions and the Bartlett’s test was used to test for homogeneity of variances. Data were log transformed for analyses due to unequal varianc- es as necessary. Maximum photosynthesis (Amax), PFD at Amax» transpiration at Ajax, conductance at Amax light saturation point, dark respiration, light compensation point, and the quantum yield efficiency (initial slope) were determined for each replicate, and means were calculated. The Ajax was the highest net photosynthesis rate. Light saturating photosynthesis was the PFD when the slope of the initial rate line reached the Amax. Dark respiration was the gas exchange rate at a PFD of 0 umol/m’/s (y-intercept of the line for the initial slope or rate). The light compensation point was calculated as the PFD when the photosynthetic rate = O wmol CO,/m’/s (x- intercept of the line for the initial slope or rate). The quantum yield efficiency or initial slope was calculated using the dark value and increasing PFDs until the regression coefficient of the slope decreased (150 umol/m’/s PFD) (Furuya and Van Auken 2009; Wayne and Van Auken 2009). A one-way ANOVA (Sall et al. 2001) was used to detect significant difference between species and between leaf types for maximum photosyn- thetic rates (Ajax), light saturation, dark respi- ration, transpiration at Ajax, conductance at Amax» and quantum yield efficiency. Significance level for all tests was 0.05. Ambient PFD was also measured for each sun and shade leaf with the LI- COR® integrating quantum sensor at the time the light response curves were initiated (LI-COR, Inc, Lincoln, NE). Pre-dawn Xylem Water Potential (‘¥,) Measurements of pre-dawn xylem water po- tential were made for leaves of each plant 2010] (Scholander et al. 1965; Furuya and Van Auken 2009) with the model 1000 PMS® pressure chamber (PMS, Instrument, Co. Corvallis, OR). Samples of each leaf type were collected with a sharp knife and put in a zip lock plastic bag with a wet paper towel between 4:30 and 5:00 a.m. The plastic bag was put 1n a cooler with ice to insure that ¥Y%. would not change. ¥%, measurements were made within 45 min of harvest. A one-way ANOVA (Sall et al. 2001) was used to detect significant difference between species and_ be- tween leaf types. Soil Moisture Measurements Volumetric soil moisture measurements were made using time domain reflectometry (TDR) with a TRIME portable TDR soil moisture meter (TRIME-FM) (MESA System Co. Medfield, ME). The TDR is a transmission line technique used to determine soil water content by inserting two parallel metal rods in a soil matrix to make measurements (Topp and Reynolds 1998; No- borio 2001). Soil water content was measured in five positions below the canopy of each plant. The five positions were the four cardinal compass points and the site next to the bole of the plant. Soil water content of 5 plants was sampled between 10:00 and 11:00 a.m. The site next to the bole of the plant was on the south side, and the other locations were approximately 10 cm from the bole. Data from 5 sites (north, south, east, west, and the site next to the plant) around each plant were pooled and a mean and standard error was determined. A one-way ANOVA (Sall et al. 2001) was used to detect significant difference between species. Leaf Area and Total Leaf Dry Mass Area (LA) and total dry mass (Mooney and Gulmon 1982) of five sun and five shade leaves were measured to determined mass per unit area (LMA) (g/cm). The sun and shade leaves used to make gas exchange measurements were collected after gas exchange measurements were completed and used to determine leaf area and total dry mass. Leaves were dried at 60C to a constant mass prior to weighing. A one-way ANOVA (Sall et al. 2001) was used to detect significant difference between species and between leaf types. RESULTS The photosynthetic light response curves for the sun and shade leaves were significantly different over the light levels measured for all three species (Repeated Measures ANOVA, P < 0.05; Fig. 1A, B, C). At PFD’s above 400-600 tumol/m?/s (depending on the species), sun leaves had significantly higher photosynthetic rates than FURUYA AND VAN AUKEN: GAS EXCHANGE OF SUB-SHRUBS 73 A -—®-SUN —O— SHADE Phtosynthetic Rate (umolCO2/m7/s) Repeated Measures ANOVA F = 39.9, P =0.0002 0 500 1000 1500 2000 B —®-SUN —O— SHADE e* e* ef e* e* Phtosynthetic Rate (umolCO2/m7/s) Repeated Measures ANOVA F = 9.8, P =0.0139 0 500 1000 1500 2000 GH HI HU O O O O O ie aii (mas Galt fal Phtosynthetic Rate (umolCO2/m7/s) Repeated Measures ANOVA F = 33.9, P =0.0004 0 500 1000 1500 2000 PFD (ymol/m7/s) Fic. 1. Photosynthetic light response curves for sun (@) and shade (O) leaves of (A) Simsia calva, (B) Wedelia texana and (C) Brickellia cylindracea. There were significant differences in photosynthetic rates for sun and shade leaves of the three species (repeated measures ANOVA, P < 0.05). Error bars are examples and represent + or — one standard error of the mean. Different letters between light levels indicate significant differences between sun leaves (upper case) or shade leaves (lower case) and between leaf types (*). Rates were not significantly different for sun leaves between PDF’s of zero and 100 umol/m/’/s or shade leaves either for S. clava or W. texana. There were significant but small differences in rates for B. cylindracea over the same range. shade leaves, while at PFD’s lower than 400— 600 umol/m?/s, shade leaves generally had higher rates than sun leaves. Photosynthetic rates for the sun leaves continued to increase from 400—600 to 174 A —®-SUN —O— SHADE Repeated Measures ANOVA F =12.2,P =0.0081 Stomatal Conductance O 0 500 1000 1500 2000 Repeated Measures ANOVA F=5.1, P=0.0546 B -—#-SUN —O- SHADE AB AB ~aB AB AB AB Stomatal Conductance 2 1000 1500 2000 Repeated Measures ANOVA —O— SHADE F = 24.3, P=0.0012 BcD CD ABCD ABCD ABCD ABCD A acp Stomatal Conductance 0 500 1000 1500 2000 PFD (ymol/m?/s) FIG. 2. Stomatal conductance curves for sun (@) and shade (O) leaves of (A) Simsia calva, (B) Wedelia texana and (C) Brickellia cylindracea. There were significant differences in stomatal conductance rates for sun and shade leaves of each species (repeated measures ANOVA, P < 0.05). Error bars are examples and represent + or — one standard error of the mean. Different letters between light levels indicate significant differences between sun leaves (upper case) or shade leaves (lower case) and between leaf types (*). Rates were not significantly different for sun leaves between PDF’s of zero and 100 umol/m’/s or shade leaves either for any of the three species. 2000 umol/m?/sec, with few significant differences (Fig. 1A, B, C). The same was true for the shade leaves. Stomatal conductance of the sun and shade leaves were significantly different over the light levels examined (Repeated Measures ANOVA, P < 0.05; Fig. 2A, B, C). However, there were MADRONO [Vol. 57 few significant differences between light levels for sun or shade leaves over the light levels tested (One way ANOVA). When significant differences were detected, they were usually at the lowest light levels examined. The sun leaves had higher conductance rates at most light levels tested, but they were not always significantly different. Transpiration rates of the two leaf types (sun and shade) were also significantly different over the light levels measured for all three species (Repeated Measures ANOVA; F = 6.6, P = 0.0245; F = 10.7, P = 0.0112, F = 16.9, P = 0.0034). At all light levels tested, the transpiration rate was significantly higher for the sun leaves compared to shade leaves for W. texana and B. cylindracea. For S. calva, transpiration rates were higher for sun leaves at all light levels tested, but differences were significant only at the two highest and the two lowest light levels (data not presented). Transpiration rates were 1—2% of the conduction rates (Table 1). The maximum photosynthetic rate (Ajax) of S. calva sun leaves was 32.8 umol CO>/m/7/s (Fig. 3A) and occurred at the maximum PFD (light level) measured (2000 umol/m/?’/s). This rate was 1.41 times higher than the A,,,x rate for sun leaves of W. texana and 1.55 times the rate for sun leaves of B. cylindracea. Simsia calva also had the highest Amax for shade leaves at 21.3 umol CO2/m7/s. This rate was significantly higher than the rates for the other two species. The A,,,x for the sun leaves of all three species was significantly higher than their respective shade leaves (Fig. 3A). The A,,ax rates for the shade leaves was 61—66% of the rates for the sun leaves at PFDs of 1400 or 1600 umol/m’/s. Ambient light levels were significantly different for the two leaf types, with the sun leaves being exposed to 6.1—8.8 times more light (Fig. 3A). There were no species differences in exposure to ambient full sun or shade. Light saturation (L,,;) for sun leaves of S. calva was 672 umol/m/?/s, which was significantly higher than the L,,, for B. cylindracea but not W. texana (Fig. 3B). The L,,, for the sun leaves of all three species was higher than the shade leaves of the respective species. Brickellia cylindracea shade leaves had the lowest L,,; (260 umol/m?/s). The light compensation point (Lep) of sun leaves of all three species was not significantly different and was between 18 and 39 umol/m?/sec (Fig. 3B). The L,, of shade leaves was similar with only B. cylindracea having a significantly lower value at 10 umol/m?/s (Fig. 3B). The dark respiration (Rg) of sun leaves was not significantly different, ranging from 1.3—3.0 umol CO,/m7/s (Table 1). The Rg of shade leaves was not significantly different ranging from 0.7— 1.1 umol CO;/m?/s. The only significant differ- ence between sun and shade leaves was for B. cylindracea and the Rg was 35% of the value for the sun leaves. The mean (+1 SE) quantum yield 2010] FURUYA AND VAN AUKEN: GAS EXCHANGE OF SUB-SHRUBS 175 Fee pes efficiency or the initial slope (QY or IS) (slope of S <3 a = ss S S the line from 0—150 umol/m’/s) was 0.041 + 0.009 xe 2 S ssssss to 0.061 + 0.013 and not significantly different ee o 2 225008 between the species for sun or shade leaves (data Sears is Syed ae not shown). _ | x 3 c VY cone ae Conductance at Amax was significantly higher a8 = Nee for sun leaves compared to shade leaves for all Sea a three species. Simsia calva had the highest SS = p g = a = i conductance rate for sun or shade leaves of all os? a Fs S Gacsss three species. Transpiration rates were similar ae o> ; 7 S 2roeeeess with higher rates for sun leaves compared to ae ae Diz| - meoanze shade leaves and S. calva had the highest rates 22S sc) a Sen oon and B. cylindracea had the lowest rates (Table 1). ~ £0 a a Soa 3 is Brickellia cylindracea had the largest leaf area a&as So |! (LA) for sun and shade leaves and there were no S Oe 5 = significant differences in sun and shade LA s a = oe _=s = except for S. calva, which had larger shade D ae 2 = Sees = = = leaves. Shade leaves of B. cylindracea were larger Z S EP wa NE) 0) Se ee than the shade leaves of the other two species and a eg = n Al 4 eeneae shade leaves of W. texana were smallest. There mw 2 Bor lS Se ee were no significant differences in the sun leaf a5 = Sa ZSs¢ mass for any of the species (Table 1). The leaf Aad 3 2 S | mass/area (LMA) for B. cylindracea sun and 5 ne S 9 = shade leaves was significantly lower than the < = o | FF = 2 S aS other species and all three species shade leaves See es g N oncseoonr had lower LMA than their respective sun leaves. 6 ; E> ‘a O = ener ewem — The xylem water potential (¥%,) of the sun and 9 S25 S pA z arr me aX shade leaves was not significantly different, but aie eS ae Ser ene W. texana had the most negative Y. at —1.84 nO : = > ASS < MPA for sun leaves. The surface soil was dry, a = having only 11.7-16.1% _ soil moisture and no em A 2 _o significant differences between species. Hato G ~~-~38R5 QAZ25 p SC HTTS SHO Hon o Se ecccece DISCUSSION S455) |8\2| & g288ee~ : : Za S oo ee ee All three of the species studied are commonly a an 2 a Poise = found growing in high light environments in Z i = = 2 oe 7 Central Texas savannas, but not in_ shaded Gin. habitats (Van Auken et al. 1981; Enquist 1987), p Pica ee ae and sun leaves of all three species had high z 2 B = + =ece S aS maximum photosynthetic rates (Ajax), typical of a6 S F S nsessses species of open habitats (Begon et al. 2006) uP? 52 v fc ssossse (Fig. 3). Other photosynthetic parameters, in- 2 - 5 es = = a a es * aN ss cluding light saturation, light compensation, dark Faas SG anwksgta respiration, conductance, and transpiration, were Az ay SS | high for sun adapted leaves (Fig. 3, Table 1). a2 8 2 = However, shade leaves had relatively high gas KOE < = exchange parameters as well, probably because of 2 age = _ the relatively open nature of the sub-shrub a 38 5 s canopy and the presence of light flecks (Hull re 4G 5 2002). These responses are not consistent with S PS Z 5 a E 52 findings for shade plants, but for plants that are 2 < 5 = ° ee Ss sun plants or intermediate or facultative species 5 ae =| 2 z Wenge 5 y 3B g (Boardman 1977; Hull 2002; Larcher 2003; ae 2 E Ze > ES = < eS Givnish et al. 2004; Begon et al. 2006; Valladares = 18y CERO RSa R420 and Niinemets 2008). Although Simsia_ calva, Sas 56 2 _ aoe) 8 Wedelia texana and Brickellia cylindracea are all ae S & 2 ais ES * 7 l@2e% native species with a fairly broad distribution in BS : & | e | ales sears E central Texas, west Texas, New Mexico and ao & S Re SS lass Ce northern Mexico, very little is known about their rR < 3 6. mb MW SIH photosynthetic capabilities. No studies were 176 40 = 35 2000 > : 2 ~< 30 @ Amax Neo "e = 25 Ab 0 Ambient Light 2 S 1200 = = 20 2 te =* 800 = Ss (3) = 10 = = 400 Lsat (mol CO2/m?/sec) Lep (PFD-, mol/m?/sec) sc WT BC sc WT BC Sun leaves Shade leaves Fic. 3. Mean A,,ax values + one standard error (A) and ambient light levels (PFD) for sun and shade leaves of Simsia calva (SC), Wedelia texana (WT) and Brickellia_ cylindracea (BC). Mean light saturation (L,,) + one standard error (B) and light compensation point (L,,) for the same species. Bars for the same parameter and the same leaf type with the same upper case letter are not significantly different at the 0.05 level (one-way-ANOVA). Bars for the same parameter and the same species followed with the same lower case letter are not significantly different at the 0.05 level (one-way-ANOVA, Tukey-Kramer honestly Significant Difference test). identified which evaluated their physiological responses or growth responses to light levels or other factors. The parameters measured for both leaf types suggest that these species are sun or faculative species, because some individuals may be found in partial shade or at the edge of woodland canopies or in light gaps (personal observation). In general, true understory species have much lower photosynthetic rates than the rates report- ed for the three species in the current study. Photosynthetic rates of three understory montane spruce forests species found in central Europe had CO, uptake rates of 3.4—5.5 umol CO>/m?/s (Hattenschwiler and Korner 1996). In addition, the European forest species reached light satura- tion at lower light levels (~200 umol/m/7’/s) than sun leaves of the species reported here (435— 672 umol/m’/s). Even shade leaves of S. calva, W. texana and B. cylindracea light saturated at higher light levels (260-375 pmol/m’/s) than the forest understory species. Arnica cordifolia Hook. MADRONO [Vol. 57 (Asteraceae), an herbaceous perennial which grows in the understory of lodgepole pine forests in southeastern Wyoming, also had photosyn- thetic rates (3.5-4.2 umolCO,/m7/s) that were 18— 30% of the rates of shade adapted leaves of the species in the current study and reached light saturation at light levels at about the same levels as shade leaves in the current study (~330 umol CO,/m?’/s) (Young and Smith 1980). Polygonum virginianum L. (Polygonaceae), found in the forest understory and at the forest edge in the eastern United States, had an A,,,, of ~3 umol CO,/m’/s at a light saturation of ~500 umol/m’/s (Zangerl and Bazzaz 1983). Carex planostachys from the central Texas Edwards Plateau Juniperus woodland understory had an Ajax value of 4.9 + 0.3 umol CO,/m?/s which was 23.0—38.0% of the value for shade leaves of the three species in the current study. In addition, light saturation for C. planostachys was about half the value of the current species studied at 151 + 43 umol/m?/s (Wayne and Van Auken 2009). While S. calva, W. texana and B. cylindracea are typically found growing in open habitats and sometimes at the edge of woodlands, their high A,,,x for shade adapted leaves com- pared to other herbaceous shade plants would suggest they are sun species, and would not grow in low light environments. True sun plants are adapted to high light conditions and consequently have high rates of gas exchange. For example, an early successional herbaceous perennial, Abutilon theophrasti Medik., had Ajax rates between 15-25 umol CO,/m?7/s (Wieland and Bazzaz 1975; Bazzaz 1979; Munger et al. 1987a, b; Hirose et al. 1997; Lindquist and Mortensen 1999; Van Auken and Bush (in press). Two oaks found mostly in gallery forest in tall grass prairies of northeastern Kansas, Quercus muehlenbergii Engelm. and Q. macrocarpa Michx. had Ajax rates of 18—26 umol CO,/m/7/s for sun leaves and 11—13 umol CO>/m/?/ s for shade leaves (Hamerlynck and Knapp 1994). Some plants, particularly early successional species or plants from disturbed (open) communi- ties can acclimate to variability of the lght environment in which they live (Bazzaz and Carlson 1982). For example, Polygonum pensylva- nicum L., a colonizing annual of open fields, had an Ayax Of ~12 umol CO,/m?/s at ~1500 umol/ m’/s when plants from a shaded-habitat (200 umol/m?/s) were measured (Bazzaz and Carlson 1982; Zangerl and Bazzaz 1983). How- ever, the rate was ~24 umol/m?’/s at ~1800 umol/ m?/s when plants from a full sun habitat were examined (Bazzaz and Carlson 1982). The light — levels of the sun leaves we reported in the present — study received 1389 + 83 to 1594 + 103 umol/m7/s (~65-80% full sunlight). Individuals from higher | light environments could have higher maximum | photosynthetic rates, while those from lower light | 2010] environments would probably be lower. Further studies would be needed to determine if these species acclimate to variability in the light environment as reported for other species (Hull 2002; Valladares and Niinemets 2008). Dark respiration of sun leaves of SS. clava (3.0 + 0.8 umol CO,/m?/s) are similar to other sun- adapted plants (Hamerlynck and Knapp 1994). This rate is 2.72 times higher than the Rg of its shade adapted leaves. The Rg for shade adapted leaves of all three species examined is about five times higher than rates for true shade adapted species (Hirose and Bazzaz 1998; Hull 2002). Dark respiration for shade-adapted species is typically lower than sun-adapted species, due to the lower metabolism of shade-adapted plants (Bjorkman 1968; Bazzaz and Carlson 1982). P. pensylvanicum grown at 200 umol/m?/s had a respiration rate of ~0.5 umol CO;/m/’/s, and the rate for sun leaves was twice this level (Bazzaz and Carlson 1982). Values of other photosynthetic parameters reported in this study for the three savanna sub- shrubs are similar to those reported for sun plants or sun leaves from the literature. Quan- tum yield efficiency reported here (0.041— 0.061 umol CO>/umol quanta, sun leaves and shade leaves) are within the range or similar to values reported for other species (0.035— 0.052 umol CO>/umol quanta) (Hirose et al. 1997). Stomatal conductance and transpiration reported from the species in the current study were similar to other studies and indicate open stomates; however, many factors affect the levels of these parameters (Wieland and Bazzaz 1975; Zangerl and Bazzaz 1984; Yun and Taylor 1986; Munger et al. 1987a, b; Stafford 1989). The sun and shade leaves of S. calva, W. texana and B. cylindracea have relatively high gas exchange rates. These physiological responses to various light levels more than likely are contrib- utors to the niche observed for this species. In the field they are usually found in disturbed grass- lands, gaps and other high light environments. In general, resource utilization is spatially parti- tioned among species along complex environ- mental gradients, such as changes in light from Open areas to woodland or forest edges (Van Auken and Bush 2009; Wayne and Van Auken 2009; Gagliardia and Van Auken in press). The ability of these three species to reach high photosynthetic rates in partial shade and having relatively low light compensation points would allows them to exist at the canopy edge or in partially shaded savanna communities. At light levels below approximately 400 umol CO>/m/’/s, more shade tolerant species would probably be able to out-compete these three sub-shrubs. At light levels above 400-500 umol CO;/m7/s, S. calva, W. texana and B. cylindracea could be community dominates and out-compete other co- FURUYA AND VAN AUKEN: GAS EXCHANGE OF SUB-SHRUBS 177 occurring species, especially in disturbed areas or in areas where grass biomass and density is reduced because of heavy grazing and a low fire frequency. This would occur in part because they have photosynthetic rates as high as or higher than most co-occurring species (Grunstra 2008; Valladares and Niinemets 2008). These central Texas savannas have apparently changed considerably over the past 200+ years for a number of reasons (Van Auken and McKinley 2008; Van Auken and Smeins 2008). Recently, heavy grazing, tree cutting and reduced fire frequency have been major factors contributing to community alterations. These are proximate factors, but they are not the only factors. Some still blame global climatic changes, including elevated CO, levels and increased temperature levels, but these factors appear to be background conditions and not the conditions directly causing the decrease in many native grasses (Van Auken 2009). We expect that these three sub-shrubs have increased in density, cover and possibly range because of the reduction in grass biomass because of herbivory and the lack of grassland fires and their apparent resistance to both native and domestic herbivory. ACKNOWLEDGMENTS We would like to thank the Center for Water Research at The University of Texas at San Antonio for support provided for the senior author. Help from M. Grunstra and J. K. Bush during various stages of the study is truly appreciated. LITERATURE CITED BAZZAZ, F. A. 1979. The physiological ecology of plant succession. Annual Review of Ecology and Sys- tematics 10:351—371. AND R. W. CARLSON. 1982. Photosynthetic acclimation to variability in the light environment of early and late successional plants. Oecologia 54:313-316. BEGON, M., C. R. TOWNSEND, AND J. L. HARPER. 2006. Ecology: from individuals to ecosystems. Blackwell Publishing, Malden, MA. BJORKMAN, O. 1968. Carboxydismutase activity in shade-adapted and sun-adapted species of higher plants. Physiologia Plantarum 21:1—10. BOARDMAN, N. K. 1977. Comparative photosynthesis of sun and shade plants. Annual Review of Plant Physiology 28:355—377. BOND, W. J. 2008. What limits trees in Cy grasslands and savannas? Annual Review of Ecology, Evolu- tion, and Systematics 39:641—659. BRESHEARS, D. D. 2008. Structure and function of woodland mosaics: consequences of patch-scale heterogeneity and connectivity along the grassland- forest continuum, Pp. 58-92 in O. W. Van Auken (ed.), Western North American Juniperus communities: a dynamic vegetation type. Springer, New York, NY. CORRELL, D. S. AND M. C. JOHNSTON. 1979. Manual of the vascular plants of Texas. The University of Texas at Dallas, Richardson, TX. 178 ENQUIST, M. 1987. Wildflowers of the Texas Hill Country. Lone Star Botanical, Austin, TX. FURUYA, M. AND O. W. VAN AUKEN. 2009. Gas exchange rates of sun and shade leaves of Sophora secundiflora. Texas Journal of Science 61:243—258. GAGLIARDIA, J. W. AND O. W. VAN AUKEN. In Press. Distribution of Verbesina virginica (Asteraceae, Frost Weed) in Central Texas and possible causes. Texas Journal of Science 62. GIVNISH, T. J. 1988. Adaptation to sun and shade - a whole-plant perspective. Australian Journal of Plant Physiology 15:63—92. , R. A. MONTGOMERY, AND G. GOLDSTEIN. 2004. Adaptive radiation of photosynthetic physi- ology in the Hawaiian lobeliads: light regimes, static light responses, and whole-plant compensa- tion points. American Journal of Botany 91: 228-246. GRUNSTRA, M. B. 2008. Investigation of Juniperus woodland replacement dynamics Ph. D. Disserta- tion. University of Texas at San Antonio, TX. HAMERLYNCK, E. P. AND A. K. KNAPP. 1994. Leaf- level responses to light and temperature in two co- occurring Quercus (Fagaceae) species: implications for tree distribution patterns. Forest Ecology and Management 68:149—159. HATTENSCHWILER, S. AND C. KORNER. 1996. Effects of elevated CO, and increased nitrogen deposition on photosynthesis and growth of understory plants in spruce model ecosystems. Oecologia 106: 172-180. HIROSE, T., D. D. ACKERLY, M. B. TRAw, D. RAMSEIER, AND F. A. BAZZAZ. 1997. CQ, elevation, canopy photosynthesis, and optimal leaf area index. Ecology 78:2339—2350. AND F. A. BAZZAZ. 1998. Trade-off between light- and nitrogen-use efficiency in canopy photo- synthesis. Annals of Botany 82:195—202. House, J. L., S. R. ARCHER, D. D. BRESHEARS, AND R. J. SCHOLES. 2003. Conundrums in mixed woody-herbaceous plant systems. Journal of Bio- geography 30:1763—1777. HULL, J. C. 2002. Photosynthetic induction dynamics to sunflecks of four deciduous forest understory herbs with different phenologies. International Journal of Plant Sciences 163:913—924. LARCHER, W. 2003. Physiological plant ecology: ecophysiology and stress physiology of functional groups. Springer, New York, NY. LINDQUIST, J. L. AND D. A. MORTENSEN. 1999. Ecophysiological characteristics of four maize hybrids and Abutilon theophrasti. Weed Research 39:271-285. MAZE, D. M. 2009. Effect of terrestrial mollusc herbivory on Holocarpha macradenia (Asteraceae) seedlings in California coastal prairie under differ- ent clipping regimes. Madrono 56:1-—7. Mooney, H. A. AND S. L. GULMON. 1982. Constraints on leaf structure and function in reference to herbivory. BioScience 32:198—206. MUNGER, P. H., J. M. CHANDLER, AND J. T. COTHREN. 1987a. Effect of water stress on photosynthetic parameters of soybean (Glycine max) and velvetleaf (Abutilon theophrasti). Weed Science 35:15—21. , AND F. M. Hons. 1987b. max) - velvetleaf (Abutilon Soybean (Glycine MADRONO [Vol. 57 theophrasti) interspecific competition. Weed Sci- ence 35:647—653. NOAA. 2004. Meteorological data. National Climatic Data Center, National Oceanic and Atmospheric Administration. Website http://www.ncdc.noaa. gov/oa/ncdc.html [accessed 19 August 2010]. NOBORIO, K. 2001. Measurement of soil water content and electrical conductivity by time domain reflec- tometry: a review. Computers and electronics in agriculture 31:213—237. NRCS. 2006. Web Soil Surveys. Soil Survey Staff, Natural Resources Conservation Service, United States Department of Agriculture. Website http:// websoilsurvey.nrcs.usda.gov/app/ [accessed 19 Au- gust 2010]. RYNIKER, K. A., J. K. BUSH, AND O. W. VAN AUKEN. 2006. Structure of Quercus gambelii communities in the Lincoln National Forest, New Mexico, USA. Forest Ecology and Management 233:69-77. SALL, J., A. LEHMAN, AND L. CREIGHTON. 2001. JMP start statistics: a guide to statistics and data analysis using JMP and JMP IN software. Dux- bury Thomson Learning, Pacific Grove, CA. SCHOLANDER, P. F., H. T. HAMMEL, E. D. BRAD- STREET, AND F. A. HEMMINGSEN. 1965. Sap pressure in vascular plants. Science 148:339—346. SMEINS, F. E. AND L. B. MERRILL. 1988. Long-term change in semi-arid grasslands, Pp. 101-114 in B. B. Amos and F. R. Gehlbach, (eds.), Edwards Plateau vegetation: plant ecological studies in central Texas. Baylor University Press, Waco, TX. SPURR, S. H. AND B. V. BARNES. 1973. Forest ecology. The Roland Press, New York, NY. STAFFORD, R. A. 1989. Allocation responses of Abutilon theophrasti to carbon and nutrient stress. American Midland Naturalist 121:225—231. TAYLOR, F. B., R. B. HAILEY, AND D. L. RICHMOND. 1962. Soil survey of Bexar County, Texas. United States Department of Agriculture. Soil Conserva- tion Service, Washington D.C. THORNTHWAITE, C. W. 1931. The climates of North America: according to a new classification. Geo- graphical Review 21:633-—655. Topp, G. C. AND W. D. REYNOLDS. 1998. Time domain reflectometry:a seminal technique for measuring mass and energy in soil. Soil tillage research 47:125—132. TURNER, B. L., H. NICHOLS, G. DENNY, AND O. Doron. 2003. Atlas of vascular plants of Texas. Brit Press, Fort Worth, TX. USDA, NRCS. 2009. The PLANTS Database, Na- tional Plant Data Center, Baton Rouge, LA. Website http://plants.usda.gov/index.html. [ac- cessed 13 August 2010]. VALLADARES, F. AND U. NIINEMETS. 2008. Shade tolerance, a key plant feature on complex nature and consequences. Annual Review of Ecology and Systematics 39:237—257. VAN AUKEN, O. W. 2000. Characteristics of inter- canopy bare patches in Juniperus woodlands of the southern Edwards Plateau, Texas. Southwestern Naturalist 45:95—110. . 2009. Causes and consequences of woody plant encroachment into western North American grass- lands. Journal of Environmental Management 90:2931—2942. 2010] AND J. K. BUSH. 2009. Gas exchange rates of Quercus gambelii in Q. gambelii woodlands. Journal of the Torrey Botanical Society 136:465-478. AND . In Press. Photosynthetic rates of two species of Malvaceae, Malvaviscus arboreus. var. drummondii (Wax mallow) and Abutilon theophrasti (Velvetleaf). Southwestern Naturalist 56. AND D. C. MCKINLEY. 2008. Structure and composition of Juniperus communities and factors that control them, Pp. 19-47 in O. W. Van Auken (ed.), Western North American Juniperus communities: a dynamic vegetation type. Springer, New York. AND F. SMEINS. 2008. Western North American Juniperus communities: patterns and causes of distribution and abundance, Pp. 3-18 in O. W. Van Auken (ed.), Western North American Juni- perus communities: a dynamic vegetation type. Springer, New York, NY. , A. L. FORD, AND J. L. ALLEN. 1981. An ecological comparison of upland deciduous forests of central Texas. American Journal of Botany 68:1249-1256. , AND A. G. STEIN. 1979. A comparison of s some woody upland and riparian plant commu- nities of the southern Edwards Plateau. Southwest- ern Naturalist 24:165—180. : , A. STEIN, AND A. G. STEIN. 1980. Woody vegetation of upland plant communities in FURUYA AND VAN AUKEN: GAS EXCHANGE OF SUB-SHRUBS 179 the southern Edwards Plateau. Texas Journal of Science 32:23-35. WAYNE, E. R. AND O. W. VAN AUKEN. 2008. Comparisons of the understory vegetation of Juniperus woodlands, Pp. 93-110 in O. W. Van Auken (ed.), Western North American Juniperus communities: a dynamic vegetation type. Springer, New York, NY. AND . 2009. Light responses of Carex planostachys from various microsites in a Juniperus community. Journal of Arid Environments 73:435-443. WIELAND, N. K. AND F. A. BAZZAZ. 1975. Physiolog- ical ecology of three codominant successional annuals. Ecology 56:68 1—688. YOUNG, D. R. AND W. K. SMITH. 1980. Influence of sunlight on photosynthesis, water relations, and leaf structure in the understory species Arnica cordifolia. Ecology 61:1380—1390. YUN, J. I. AND S. E. TAYLOR. 1986. Adaptive implications of leaf thickness for sun- and shade- grown Abutilon theophrasti. Ecology 67:1314—-1318. ZANGERL, A. R. AND F. A. BAZZAZ. 1983. Plasticity and genotypic variation in photosynthetic behav- iour of an early and a late successional species of Polygonum. Oecologia 57:270-273. AND 1984. Effects of short-term selection along environmental gradients on varia- tion in populations of Amaranthus retroflexus and Abutilon theophrasti. Ecology 65:207—217. MADRONO, Vol. 57, No. 3, pp. 180-183, 2010 DOCUMENTATION OF THE CHROMOSOME NUMBER FOR THE CALIFORNIA ENDEMIC, TOXITCOSCORDION EXALTATUM (LILIALES: MELANTHIACEAE) DALE W. MCNEAL Department of Biological Sciences, University of the Pacific, 3601 Pacific Avenue, Stockton, CA 95211-0197 WENDY B. ZOMLEFER' Department of Plant Biology, University of Georgia, 2502 Miller Plant Sciences, Athens, GA 30602-7271 wendyz@plantbio.uga.edu ABSTRACT The mitotic chromosome count for Toxicoscordion exaltatum (2n = 22), documented for the first time here, matches the number formerly reported for other species in the genus. A base number of x = 11 is synapomorphic for Toxicoscordion. An updated overview of this recently resurrected segregate genus is given. Key Words: California endemic, chromosome number, Melanthiaceae, Toxicoscordion, Zigadenus. Tribe Melanthieae (Liliales: Melanthiaceae) comprises seven genera (ca. 65-95 spp.) of predominately woodland and/or alpine perennial herbs occurring mainly in the temperate to Arctic zones of the Northern Hemisphere: Amianthium A. Gray (1 sp.), Anticlea Kunth (9-11 spp.), Schoenocaulon A. Gray (26 spp.), Stenanthium (A. Gray) Kunth (3 spp.), Toxicoscordion Rydb. (8 spp.), Veratrum L. s.l. (17-45 spp.), and Zigadenus Michx. s.s. (1 sp.). These generic circumscriptions (most novel) are supported by parsimony analyses of trnL-F (plastid) DNA and ITS (nuclear ribosomal) sequence data (Zomlefer et al. 2001, 2003, 2006a, b). A significant consequence of the molecular studies was the reassessment of the traditional Zigadenus s.1., a poorly defined assemblage with a complex taxonomic history involving several segregate genera (summaries in Zomlefer 1997; Zomlefer et al. 2006a). Contemporary treatments (e.g., Schwartz 2002) have typically accepted the monotypic segregate Amianthium with the re- maining ca. 20 species maintained in Zigadenus s.. Based on these molecular data, however, Zigadenus s.l. is polyphyletic and forms five strongly supported clades correlating with certain geographical distribution, morphological charac- ters, and chromosome number. One well-defined clade corresponds to To.x- icoscordion, originally described by Rydberg (1903), and sometimes recognized as Zigadenus sect. Chitonia (Salisb.) Baker (Preece 1956; Schwartz 1994). This Zigadenus segregate is defined by the morphological synapomorphies of conspicuously clawed tepals (especially the ' Corresponding author. inner three) and one obovate gland per tepal (Fig. 1B; Zomlefer et al. 2001; Zomlefer and Judd 2002). Toxicoscordion comprises ca. eight species (Table 1) restricted to the midwestern United States to western North America (Fig. 1A) and includes the widely distributed poisonous “‘death camas” plants of the rangelands such as 7, nuttallii (A. Gray) Rydb., T. paniculatum (Nutt.) Rydb., and 7. venenosum (S. Watson) Rydb. (see Marsh et al. 1915, 1926). However, the focal taxon of this paper, 7. exaltatum (Eastw.) A. Heller (giant death camas), has a more limited distribution: mid-elevations (1000 to 1800 m) along margins of mixed coniferous forest on the western slopes of the Sierra Nevada Mountains in California (Fig. 1A; Walsh 1940; Preece 1956; Schwartz 1994, 2002). (Reports of this taxon from the foothills of the Sierra Nevada near Chinese Camp, Tuolumne Co. [see Preece 1956] are based on _ several collections of 7. fremontii, a species otherwise known only from the Coast Ranges in California [McNeal and Zomlefer 2009].) Toxicoscordion exaltatum is the largest species in the genus with relatively large bulbs (6-10 cm long; 3—5 cm in diam.) and long flowering stems to 100 cm tall (McNeal 1993). These robust plants often occur in large populations of hundreds to thousands of individuals—creating a remarkable display (McNeal personal observation). The chromosome number 27 = 22 (orn = 11) has been verified for all species of Toxicoscordion (Table 1; Fig. 2; see Zomlefer 2003), except for T- exaltatum with an undocumented citation of 7 = 11 by McNeal (1993; Schwartz 2002). Since chromosome number is a significant and likely an invariable apomorphy for genera of tribe Melanthieae (e.g., Zomlefer and Smith 2002; 2010] MCNEAL AND ZOMLEFER: TOXICOSCORDION EXALTATUM CHROMOSOME NUMBER - 181 CENTRAL NORTH AMERICA Toxicoscordion FIG. 1. Distinctive characteristics of Toxicoscordion. A. General distribution map of Toxicoscordion (gray shading) with dots indicating localities of the California endemic 7: exaltatum. (The south-midwestern disjunct range of the genus comprises 7: nutta/lii.) Data from Walsh (1940), Preece (1956), and Schwartz (1994, 2002). B. Adaxial surface of an outer (left) and an inner (right) tepal of 7: exa/tatum showing the generic autapomorphies: claw plus a single obovate gland. Dashed ellipse = filament insertion. The claws of the inner tepals are typically more pronounced than those of the outer whorl. Drawn from live material (McNeal & Smith 4749; CPH, GA) by W. B. Zomlefer. Zomlefer 2003), 22 is the predicted mitotic count months, and then placed on moist peat moss in a for all Zigadenus species now transferred to growth chamber at 4°C with 16 hr of light and Toxicoscordion. We here document the chromo- 8 hr of darkness per day. Growing root tips some number of 7. exaltatum. (1-2 cm long) were collected at ca 7:00 a.m. and soaked in 0.2% colchicine for 26—28 hr, rinsed in MATERIALS AND METHODS deionized water, and fixed in Carnoy’s solution (3 ethanol: | acetic acid). Root tips thus prepared Several bulbs of Toxicoscordion exaltatum were were stored in the freezer for at least 24 hr, collected during field work by the first author on transferred to 70% ethanol to slow tissue 12 May 2006, stored in paper bags for three hardening, and returned to the freezer. Squash TABLE 1. THE SPECIES OF TOXICOSCORDION AND THE ORIGINAL DOCUMENTATION OF THE MITOTIC (2) AND/OR MEIOTIC (VN) CHROMOSOME NUMBERS. See also Zomlefer (2003). Chromosome number Taxon n 2n Original source(s) T. brevibracteatus (M. E. Jones) R. R. 1] — Lewis (1959); Cave (1970) Gates T. exaltatum (Eastw.) A. Heller 11 — this paper —T. fontanum (Eastw.) Zomlefer & Judd 1] — Preece (1956, as Zigadenus venenosus var. fontanus); Cave (1970) T. fremontii (Torr.) Rydb. 11 — Miller (1930); Preece (1956, as Zigadenus fremontii var. fremontii); Cave (1970, as Z. fremontii var. fremontii and var. inezianus) TT. micranthum (Eastw.) A. Heller 1] — Preece (1956, as Zigadenus venenosus var. | micranthus) —T. nuttallii (A. Gray) Rydb. —- Ze Zomlefer (2003) | T. paniculatum (Nutt.) Rydb. 1] De Preece (1956) | T. venenosum (S. Watson) Rydb. 1] 22 Preece (1956, as Zigadenus venenosus var. gramineus and var. venenosus); Cave (1970); Taylor & Taylor (1977, as Z. venenosus Var. gramineus); | Hartman & Crawford (1971, as Z. venenosus var. gramineus ) MADRONO [Vol. 57 10 um FIG. 2. Summary illustration of previously documented meiotic (7 = 11) and/or mitotic (2n = 22) chromosomes of Toxicoscordion (A—B, D-—I) and the new report for T. exaltatum (C: 2n = 22). A. T. brevibracteatus, diakinesis (n = 11). B. T. fontanum, anaphase II (n = 11). C. T. exaltatum, metaphase (2n = 22), (McNeal & Smith s.n. [12 May 2006], CPH). D. 7. fremontii, metaphase I (n = 11). E. 7. micranthum, metaphase I (n = 11). F. 7. nuttallii, metaphase (2n = 22). G. T. paniculatum, metaphase (2n = 22). H. T. paniculatum, metaphase I (n = 11). L. 7: venenosum, metaphase I (n = 11). A, B, and I modified from Cave (1970); D, E, G, and H modified from Preece (1956); F from Zomlefer (2003). preparations were made within 30-60 days ac- cording to protocols outlined by Brooks et al. (1963). Following fixation, roots were rinsed several times in deionized water and hydrolyzed in 1.0 N HCL at 60°C for 5 min and rinsed several times again. Root tips were trimmed to ca. 3 mm and placed on slides in saturated iron aceto-carmine and macerated with fine-tipped forceps. After application of a cover slip, the preparation was squashed using thumb pressure. Slides were mounted with euparal for future reference. Well-spread metaphase chromosomes were traced by the second author under a Leica DMLB Research Microscope with a camera lucida attachment. The herbarium voucher spec- imen, McNeal & Smith s.n. (12 May 2006), is deposited at CPH. RESULTS AND DISCUSSION The mitotic chromosome number of 27 = 22 for Toxicoscordion exaltatum is confirmed with plants from a population in El Dorado Co., California (Fig. 2C; Table 1), and the previous report by McNeal (1993) is now documented with a voucher specimen. The validation of the diploid number for 7. exa/tatum strengthens support for the monophyly of Toxicoscordion, as well as the phylogenetic significance of chromosome num- bers as generic synapomorphies for tribe Mel- anthieae. A probable base chromosome number of x = 8 | is often cited for the tribe (Summary in Zomlefer et al. 2006a). The diploid numbers of 22 for | Toxicoscordion (Zomlefer 2003) and 20 for | 2010] MCNEAL AND ZOMLEFER: TOXICOSCORDION EXALTATUM CHROMOSOME NUMBER 183 Stenanthium (Zomlefer and Judd 2002) are exceptions to the multiples of eight prevalent in the other genera (Sen 1975; Tamura 1995; Lowry et al. 1987; Zomlefer 1997): Amianthium (2n = 32), Anticlea (2n = 32), Schoenocaulon (2n = 16), and Veratrum (including Melanthium; 2n = 16, 32, 64, 80, 96). Due to the small size of the chromosomes of this tribe (ca. 2.0-4.0 um in length), the few detailed karyological studies (e.g., Lee 1985) lack the detail to infer mecha- nisms of chromosomal evolution, although these chromosome numbers indicate the prevalence of polyploidy and/or aneuploid variation of the prospective basic number. ACKNOWLEDGMENTS The authors acknowledge the following undergradu- ate students who prepared slides and/or chromosome counts while enrolled in Undergraduate Research under the guidance of DWM: Christine Ahn, Jagmeet Chauhan, and Hein Ha. Barney L. Lipscomb provided constructive review of the manuscript. LITERATURE CITED BROOKS, R. M., M. V. BRADLEY, AND T. I. ANDER- SON. 1963. Plant microtechnique manual. Depart- ment of Pomology, University of California-Davis, Davis, CA. CAVE, M. S. 1970. Chromosomes of the California Liliaceae. University of California Publications in Botany 57:1—48. HARTMAN, R. L. AND D. J. CRAWFORD. 1971. In: IOPB Chromosome number reports XXXI. Taxon 20:157—160. LEE, N. S. 1985. A cytotaxonomic study of the Korean Veratrum species. Korean Journal of Plant Taxon- omy 15:155—161. LEwIs, H. 1959. Jn: Documented chromosome numbers of plants. Madrono 15:49—52. Lowry, P. P., P. GOLDBLATT, AND H. TOBE. 1987. Notes on the floral biology, cytology and embry- ology of Campynemanthe (Liliales: Campynemata- ceae). Annals of the Missouri Botanical Garden 74:573-576. MARSH, C. D., A. B. CLAWSON, AND H. MARSH. 1915. Zygadenus [sic], or death camas. Bulletin of the United States Department of Agriculture 125:1-46. : , AND G. C. ROE. 1926. Nuttall’s death camas (Zygadenus [sic] nuttallii) as a poisonous plant. Bulletin of the United States Department of Agriculture 1376:1—13. MCNEAL, D. W. 1993. Zigadenus. Pp. 1210-1211 in J. C. Hickman (ed.), The Jepson manual: higher plants of California. University of California Press, Berkeley, CA. AND W. B. ZOMLEFER. 2009. Toxicoscordion, death camas. Jn T. J. Rosatti (ed.), Second edition of the Jepson manual: vascular plants of California [on-line]. Treatments for public viewing. Regents of the University of California, Oakland, CA, Website http://ucjeps.berkeley.edu/jepsonmanual/review/ [accessed 15 July 2010]. MILLER, E. W. 1930. A_ preliminary note on the cytology of the Melanthioideae section of the Liliaceae. Proceedings of the University of Durham Philosophical Society 8:267—271. PREECE, S. J. 1956. A cytotaxonomic study of the genus Zigadenus. Ph.D. dissertation. State College of Washington, Pullman, WA. RYDBERG, P. A. 1903. Some generic segregations. Bulletin of the Torrey Botanical Club 30:271—281. SCHWARTZ, F. C. 1994. Molecular systematics of Zigadenus section Chitonia (Liliaceae). Ph.D. dis- sertation. University of Washington, Seattle, WA. . 2002. Zigadenus. Pp. 81-89 in Flora of North America Editorial Committee (eds.), Flora of North America north of Mexico, Vol. 26, Magno- liophyta: Liliidae: Lilales and Orchidales. Oxford University Press, Oxford, U.K. SEN, S. 1975. Cytotaxonomy of Liliales. Repertorium 86:255—305. TAMURA, M. N. 1995. A karyological review of the orders Asparagales and Liliales (Monocotyledo- nae). Feddes Repertorium 106:83—111. TAYLOR, R. L. AND S. TAYLOR. 1977. Chromosome numbers of vascular plants of British Columbia. Syesis 10:125—138. WALSH, O. S. 1940. A systematic study of the genus Zigadenus Michx. Ph.D. dissertation. University of California, Berkeley, CA. ZOMLEFER, W. B. 1997. The genera of Melanthiaceae in the southeastern United States. Harvard Papers in Botany 2:133—177. . 2003. Documented chromosome numbers 2003: 1. Chromosome number of Toxicoscordion nuttallii (Liliales: Melanthiaceae) and clarification of the genus. Sida 20:1085—1092. AND W. S. JUDD. 2002. Resurrection of segregates of the polyphyletic genus Zigadenus s.1. (Liliales: Melanthiaceae) and resulting new combi- nations. Novon 12:299—308. AND G. L. SMITH. 2002. Documented chromo- some numbers 2002: |. Chromosome number of Stenanthium (Liliales: Melanthiaceae) and its sig- nificance in the taxonomy of tribe Melanthieae. Sida 20:221—226. , W. S. JUDD, W. M. WHITTEN, AND N. H. WILLIAMS. 2006a. A synopsis of Melanthiaceae (Liliales), with focus on character evolution in tribe Melanthieae. Aliso 22:566—578. , W. M. WHITTEN, N. H. WILLIAMS, AND W. S. JUDD. 2003. An overview of Veratrum s.l (Liliales: Melanthiaceae) and an infrageneric phylogeny based on ITS sequence data. Systematic Botany 28:250-269. Feddes _ ——-, -- , AND . 2006b. Infra- generic phylogeny of Schoenocaulon (Liliales: Melanthiaceae) with clarification of cryptic species based on ITS sequence data and geographical distribution. American Journal of Botany 93:1178—-1192. , N. H. WILLIAMS, W. M. WHITTEN, AND W. S. JUDD. 2001. Generic circumscription and relation- ships in the tribe Melanthieae (Liliales, Melanthia- ceae), with emphasis on Zigadenus: evidence from ITS and trnL-F sequence data. American Journal of Botany 88:1657—1669. MADRONO, Vol. 57, No. 3, pp. 184-202, 2010 THE IDENTITY AND NOMENCLATURE OF THE PACIFIC NORTH AMERICAN SPECIES ZELTNERA MUHLENBERGII (GENTIANACEAE) AND ITS DISTINCTION FROM CENTAURIUM TENUIFLORUM AND OTHER SPECIES WITH WHICH IT HAS BEEN CONFUSED JAMES S. PRINGLE Royal Botanical Gardens, P.O. Box 399, Hamilton, Ontario, Canada L8N 3H8 jpringle@rbg.ca ABSTRACT The name Ze/tnera muhlenbergii (Griseb.) G. Mansion is here applied to a species native from Monterey County, California, to southwestern British Columbia. From Z. davyi (Jeps.) G. Mansion, a California endemic to which the name Z. muhlenbergii has sometimes been misapplied, Z. muhlenbergii differs in that its calyx lobes are not keeled or have keels proximally only, to 0.25 mm wide, and its corolla lobes are 1—2 mm wide; in Z. davyi the keels of the calyx lobes are 0.3—0.6 mm wide and the corolla lobes are 2—3 mm wide. Ze/ltnera muhlenbergii has often been confused with Centaurium tenuiflorum (Hoffmanns. & Link) Fritsch ex Janch. s./, a species naturalized from Eurasia, now established in Pacific and Gulf coastal North America. From C. tenuiflorum, Z. muhlenbergii differs most notably in its flabelliform rather than ovate stigmatic lobes and in its more open, non-corymboid cymes. Key Words: California, Centaurium, Gentianaceae, nomenclature, taxonomy, Ze/tnera. Centaurium Hill s.l. was recently divided by Mansion (2004; Mansion and Struwe 2004) into four genera, two of which are pertinent to the present discussion. The species native to North America north of Mexico were placed in Ze/tnera G. Mansion and the naturalized species discussed here were retained in Centaurium. Although Mansion’s division of Centaurium was _ based largely on nrDNA and cpDNA sequences, Broome (1973) had commented earlier that ‘‘the New and the Old World taxa (of Centaurium s.1.) have totally different constellations of charac- ters.” The name Ze/tnera is used here for the accepted identifications of native species. Except for the gender endings, the specific epithets remain the same whether the respective species are placed in Erythraea Borkh., Centaurium, or Zeltnera. (The genus name Erythraea, with its authorship variously attributed, was widely used for Centaurium s./. during the nineteenth century, prior to a consensus as to the effective and valid publication of the genus name Centaurium Hill [Heller 1908], but is no longer in use.) Taxonomic treatments of Centaurium s.l. in North America differ greatly. Among them, discrepant applications of the name C. muhlen- bergii (Griseb.) W. Wight ex Piper have led to uncertainty as to whether plants so designated constitute a native species appropriately of conservation concern or an introduced species of which the North American range is expanding. This paper discusses the correct application of the name Zeltnera muhlenbergii (Griseb.) G. Mansion, distinguishes Z. muhlenbergii from the species with which it has been confused, correlates historic applications of names with the nomenclature accepted here, and provides a published equivalent to the personal communi- cation from me cited by Mansion (2004). Specimens examined include those of the Ze/tnera and naturalized Centaurium species from North America north of Mexico, or those of the species discussed in this paper, at BM, CAN, CHSC, DAO, DUKE, GH, HAM, ILL, JEPS, K, MICH, MO, NY, OSH, TRT, and UC. TYPIFICATION OF THE NAME ZELTNERA MUHLENBERGII In the original description of Erythraea muh- lenbergii, Grisebach (1838) cited specimens col- lected by Gotthilf Henry Ernest Muhlenberg in Pennsylvania and by David Douglas in Califor- nia. He also cited ““E. Centaurium Beck” (1.e., (L.) Pers. sensu Beck [1833], who had thus identified plants from New York), but he expressed uncertainty as to whether Beck’s New York plants were of the same species. In 1839, Hooker and Arnott (in Hooker and Arnott 1830-1841) identified Beck’s plants as FE. ramosissima Pers. (=Centaurium pulchellum (Sw.) Hayek ex Hand.-Mazz. et al., a species native to Eurasia, naturalized in North and South America and Australia; the epithet ramosissima, illegitimate under current rules of nomenclature, was subse- quently displaced in general use for this species by | the earlier, legitimate epithet pulchella). Gray | (1848) placed the Pennsylvania component of | Grisebach’s E. muhlenbergii in the synonymy of E. ramosissima. These identifications have 2010] consistently been accepted (as C. pulchellum) by later authors. Torrey (1857), who was aware of the exclusion of the eastern plants by previous authors, accepted the name E. muhlenbergii “quoad pl. Calif.” for plants from Benicia, Solano Co., California. In the interim Wood (1845 and later editions) had given “‘N.Y., Penn.”’ as states in which E. muhlenbergii occurred, without mentioning California, but because the early editions of Wood’s Class-Book of Botany did not cover western North America and the later editions included only selected western species, he did not address the question of whether the eastern and western specimens cited by Grisebach as E. muhlenbergii were conspecific. Later authors have not interpreted Wood’s treatment as having restricted the circumscription of E. muhlenbergii by excluding the California specimens. Gray (1876, 1878) explicitly excluded both Beck’s and Muhlenberg’s specimens from E. muhlenbergii, despite the epithet, and retained the name for the only remaining element cited by Grisebach, Douglas’s plants from California. Piper (1906) formalized this typification by citing Douglas’s collection as the type, with no mention of Beck’s or Muhlenberg’s, and Gillett (1963), in accord with Grisebach’s (1838) statement that he had seen the Douglas collection in the herbarium of W.J. Hooker, specified the component at K, into which repository Hooker’s herbarium has been incorporated. At K, plants perhaps from a single collection by Douglas (although C. Rose Broome expressed uncertainty in 1979 annotations) are present on two sheets. The specimen bearing the bar code number K000195655, is from W.J. Hooker’s herbarium (Fig. 1), and is appropriately consid- ered the lectotype. This is the only such collection from Hooker’s herbarium, and it uniquely is labeled with a provisional name in handwriting identified by Otto Stapf (annotation) as Grise- bach’s and matching that in an attached note. This note, presumably sent by Grisebach to Hooker, states that he would henceforth use the name £. muhlenbergii (** Mtihlenbergii’’) instead of the provisional name, and contains wording similar to that used by Grisebach (1838) in the original description of the species. The combina- tion of the note and Grisebach’s label bearing the provisional name indicates that Grisebach saw this specimen and called it E. muhlenbergii shortly before he published that name in 1838. On this herbarium sheet the name Douglas appears only in Grisebach’s note and in a 1979 annotation by Broome, but Grisebach’s citation of Douglas in both the note and his published description of the species indicates that he understood these plants to have been collected by Douglas, presumably from information provided by Hooker. A stigma recognizable as that of a Ze/tnera species rather PRINGLE: ZELTNERA MUHLENBERGII 185 than a Centaurium (discussed below) is visible on one of these plants. The possibility that they represent Beck’s or Muhlenberg’s collection, or any naturalized species from New York or Pennsylvania, can thereby be eliminated from consideration. The specimen on the other sheet at K, with the bar code number K000195658, is from George Bentham’s herbarium, with the printed label “Douglas 1833.’ The date probably indicates when a shipment reached a recipient, as Douglas did not collect specimens in California in 1833 (McKelvey 1956). It also bears a label in Bentham’s handwriting identifying it as Eryth- raea, with the original specific epithet crossed out and replaced with “Muhlenbergii,” and the citation ““Griseb. Gent. 146.” In what may be a later annotation, Bentham added “‘[ditto] in DC. Prod. 9.60.” Typified by the Douglas collection at K, the name Ze/tnera muhlenbergii is correctly applied to a species native to western North America from California north to southwestern British Colum- bia, with most of its populations in California. A representative well-developed plant of this species is illustrated in Figure 2. (Depending on condi- tions of the habitat and the time of seed germination in relation to photoperiod, plants of this and other Ze/tnera species are sometimes smaller than well-developed plants, with shorter internodes and less dichasial branching, as in the type collection.) The distribution of Z. muhlen- bergii in California is mapped in Figure 3. A few records exist from scattered localities farther north. Most Oregon records are from the western part of the state, but it is also known from Harney Co., Oregon, and from Washington Co. in adjacent Idaho. There are historic records from eastern Washington, but no recent collec- tions from that state have been encountered in this study. The northernmost records are from Vancouver Island, British Columbia. All records of this species from Nevada appear to have been based on misidentified plants of Z. exaltata, Z. namophila (Reveal, C.R. Broome, & Beatley) G. Mansion, and perhaps other native species. IDENTITY OF THE TYPE SPECIMEN: ZELTNERA MUHLENBERGII VS. Z. DAV YI Successive annotations by W. L. Jepson (in JEPS) reflect a change in his opinion as to the correct application of the name Centaurium muhlenbergii. His descriptions of C. muhlenbergii (Jepson 1901 [as Erythraea], 1911) and his original identification of Jepson 7624 (JEPS), from Mendocino Co., California, as C. muhlen- bergii indicate that initially he applied that name to the species called Zel/tnera muhlenbergii in this paper. Later (annotations in 1935), he concluded that the name C. muhlenbergii was applicable, 186 MADRONO [Vol. 57 6) 1 2 Ea eae lo eel 2cm FIG. 1. Holotype collection of the name Erythraea muhlenbergii Griseb. (Douglas s.n., K). instead, to the taxon he (Jepson 1925) had davyi is exactly the original C. muhlenbergii! as previously described as C. exaltatum var. davyi compared at Kew, 1935.” He reidentified Fergu- Jeps., a California endemic here treated as son & Ferguson 294 (JEPS; identified as Z. davyi Zeltnera davyi (Jeps.) G. Mansion. He annotated — in this study), which he had originally called C. the type of the latter name (cited below) as “‘true exaltatum var. davyi, as C. muhlenbergii, with the C. muhlenbergii!”’ with the statement that ““C. comment “very close to Douglas type at Kew.” 2010] PRINGLE: ZELTNERA MUHLENBERGII 187 Fic. 2. Representative specimen of Ze/tnera muhlenbergii (Oswald & Ahart 9267A, CHSC). MADRONO tla i +, t. (a : ee cd an) wr pl Sacramento ' i i ana mo | 1 1 i i cb ar 1 i ' i i é ad , é ener rane abnor see cea os its ms ps Sm we ear een Ne 100Miles = " SanDiegoft == = eee yo 100KM.,. ee Fic. 3. Documented distribution of Zeltnera muhlenbergii in California. Jepson 7624, he then stated (annotation), was not interpreted here as a heterotypic synonym of Z. C. muhlenbergii, but was instead C. floribundum —muhlenbergii, is discussed below.) More recently, (Benth.) B. L. Rob., as he had determined from C. Rose Broome and James L. Reveal (annota- examining the type of the latter name at Kew. tions in JEPS, MO, and UC in 1988, 1992, and | From 1935 on he applied the name C. muhlen- undated, probably ca. 1980) likewise applied the bergii to the species called Z. davyi in this paper name C. floribundum to the species treated here as — and the name C. floribundum to the taxon treated Z. muhlenbergii and the name C. muhlenbergii to | here as Z. muhlenbergii, and discontinued all use the species treated here as Z. davyi, whereas of the epithet davyi. (The name C. floribundum, Hickman (1993), Beidleman and Kozloff (2003), 2010] FIG. 4. c. Z. davyi (West 140, JEPS). and Mansion (2004) retained the epithet davyi for the latter species. Jepson (1925), as noted above, originally described the taxon davyi as a variety of Centaurium exaltatum (Griseb.) W. F. Wight ex Piper but later treated it as a distinct species, to which he applied the name C. muhlenbergii. Dunbar (1929) treated it as a variety of C. curvistamineum (Wittr.) Druce (species name discussed below; varietal combination unpub- lished), although his concept of var. davyi included some plants from eastern California referable to Z. exaltata (Griseb.) G. Mansion as well as plants identified as Z. davyi in this study. Abrams (1951) accorded the taxon the rank of species as C. davyi (Jeps.) Abrams. This status, including the distinctness of this species from the one treated here as Z. muhlenbergii, has generally been accepted by subsequent authors, including Munz and Keck (1959), Broome (1973), Hickman (1993), and Mansion (2004). Nevertheless, the species respectively designated Z. davyi (Figs. 4 and 5) and Z. muhlenbergii (Figs. 2 and 4) in this paper are often similar in aspect. The question as to which of these species is represented by the type of the name Z. muhlenbergii is addressed below. The Douglas specimens at K ex herb. Bentham (as distinguished from those ex herb. Hooker) appear definitely to represent the species called Z. muhlenbergii in this paper rather than Z. davyi. This was acknowledged by Broome in an annotation in 1979. She likewise annotated the replicate at E in 1987. Determination that the Douglas specimens ex herb. Hooker also repre- sent this species rather than Z. davyi has required careful examination. Plants that small are not ideal for identification, and the distal leaves are proportionately wide and broad-based, ap- proaching in shape those more common in Z. PRINGLE: ZELTNERA MUHLENBERGII 189 5mm Flowers of Ze/tnera spp. a. Holotype, Z. muhlenbergii. b. Z. muhlenbergii (Oswald & Ahart 9267A, CHSC). davyi. In Z. davyi elliptic to ovate leaves generally retain their width (usually over 5 mm except on the smallest plants) well into the inflorescence, sometimes to the summit. In Z. muhlenbergii of this paper elliptic to narrowly ovate leaves, when present, are usually limited to the proximal third or less of the plant, with at least the distal leaves being narrower, but both species vary in leaf proportions and occasionally depart from these generalizations. In the species here designated Z. muhlenbergii, the larger plants are usually branched from near but not at the base to the summit. Branches from the base are less common in Z. muhlenbergii than in Z. davyi, although not rare, and when present are often more slender than the main stem. The proximal branching of the inflorescence of Z. muhlenbergii is usually dichasial, with the central flower in the divisions sessile or on a pedicel to 5 mm or occasionally to 12 mm long on the larger plants. Distal branching is often monochasial, with a branch developing only on one side of each flower. The flowers 1n the distal portions of the inflorescences are sessile or on pedicels to 4 mm. The inflorescences of small plants are often monochasial throughout. In Z. davyi, medium-sized and larger plants are usually several-stemmed from the base, although smaller plants are often single-stemmed. The branching of the inflorescence is similar to that of Z. muhlenbergii, but pedicels (1.5) 4-25 mm are generally present even in the distal portions of the inflorescence. In the holotype collection of the name Z. muhlenbergii, one plant is_ basally several-stemmed and the other is single-stemmed, but with plants of that size the branches are too few and too short to exhibit the characteristic branching patterns of larger plants of the respective species, so this condition is of little 190 MADRONO [Vol. 57 | West 35cm 140 FiG. 5. Representative specimen of Ze/tnera davyi (West 140, JEPS). diagnostic value. On the type plants, pedicels sufficiently well exposed to be measured range from 4-5 mm in the proximal divisions of the inflorescence to ca. 1.5 mm distally, and, although not definitive, are compatible with the interpretation of Z. muhlenbergii in this paper. Zeltnera davyi differs from all similar species in that the calyx lobes are distinctly keeled along the midveins for much of their length, with the keels of all or at least the outer lobes proximally being 0.3—-0.6 mm wide, and the calyx consequently is ovoid to ellipsoid. In Z. muhlenbergii, in contrast, the keels are absent or weakly developed, if present being confined to the proximal plant of the calyx and less than 0.25 mm wide. The outer calyx lobes of the type plants of Z. muhlenbergiti are slightly keeled near the base, but even there | the keels are ca. 0.2 mm wide, and the calyx | appears narrowly cylindric, as in Z. muhlenbergii — of this paper. Calyces of one of the type plants » and of representative plants of Z. muhlenbergii | and Z. davyi are compared in Figure 4. 2010] The corolla tube of Z. davyi flares at ca. 40° almost immediately above the summit of the ovary. That of Z. muhlenbergii at a comparable stage of floral development flares more gradually between the summit of the ovary and the base of the lobes, forming a longer, more distinct, slender neck (Fig. 4b, c). The corolla lobes of Z. davyi are ovate-oblong to ovate-elliptic, 3-7 =< 2-3 mm, slightly less to more than half as long as the tube, rounded at the apex. Those of Z. muhlenbergii are lance-elliptic, 2-7 =< 1—2 mm, less than half as long as the tube, tapering to a subacute apex. The corolla lobes of the type specimen that are pressed flat enough for measurement are 1— 1.2 mm wide. In these floral characters, the flowers of the type plants are consistent with the interpretation of Z. muhlenbergii in this paper. It has sometimes been assumed that Douglas collected the type of the name FE. muhlenbergii in the vicinity of Monterey Bay, California, which is in the heart of the range of Z. davyi but at the southern limit of the range of the species called Z. muhlenbergii in this paper (Figs. 3 and 6). In that area, both species are known from Gilroy, and three replicates of Elmer 4378 (JEPS, MICH, UC) suggest that intergradation or hybridization has occurred at Pacific Grove. Douglas did much of his California botanizing in the vicinity of Monterey Bay, but in 1831, at the time of year when Z. muhlenbergii would have been in flower, he traveled north from Monterey via the vicinity of San Francisco to the site of present-day Sonoma (McKelvey 1956), which is well within the range of the species interpreted here as Z. muhlenber git. I have, therefore, concluded that despite the relatively wide leaves of the type plants, the name Z. muhlenbergii is correctly applicable to the more northern of these two species, to which the name Z. muhlenbergii is therefore applied in the remainder of this paper. This conclusion pre- serves the widely accepted usage of the epithet davyi, including that by Abrams (1951), Munz and Keck (1959), Hickman (1993), Beidleman and Kozloff (2003), and Mansion (2004). It also preserves the well-established use of the epithet muhlenbergii for plants of British Columbia and Washington, e.g., by Piper (1906), Hitchcock (1959), Gillett (1963), Straley et al. (1985), Douglas (1999), and Kozloff (2005). HETEROTYPIC SYNONYMY OF ZELTNERA MUHLENBERGII Nomenclatural complexity with regard to heterotypic synonyms of the name Zel/tnera muhelnbergii has several causes. Initially, in some cases, variation within the species, or the failure to recognize plants of true Z. muhlenbergii as representing a species already described and named, has led to the publication of new, PRINGLE: ZELTNERA MUHLENBERGII 19] heterotypic names for the species as it Is circumscribed here. Later, 1n such cases, the use of a heterotypic synonym for the true Z. muhlenbergii has been perpetuated, even though the circumscription of the species may have been equivalent to that accepted here, because the name Centaurium muhlenbergii had incorrectly become associated with some other species. The name Erythraea floribunda Benth., treated by Holmgren (1984) as a heterotypic synonym of C. muhlenbergii and interpreted as such in this study, is typified by specimens collected ‘‘in valle Sacramento,” California, by Karl Theodor Hart- weg. As Holmgren recognized, these specimens represent the species treated here as Z. muhlen- bergii. Although differences in plant size may have contributed to Bentham’s (1849, in Bentham 1839-1857) perception that the specimens he identified, respectively, as EL. muhlenbergii and E. floribunda represented different species, plants of the latter being larger, his concept of E. muhlenbergii appears to have been based in part on plants that would now be segregated as Z. davyi. Some post-1920 authors who have distin- guished between plants respectively designated C. floribundum (Benth.) B. L. Rob. and C. muhlen- bergii have based their concepts of one or the other in whole or in part on C. tenuiflorum (Hoffmanns. & Link) Fritsch ex Janch. s./. (discussed below). Centaurium curvistamineum (Wittr.) Druce was accepted as a species by Dunbar (1929), Abrams (1951), and Munz and Keck (1959), but was included in C. muhlenbergii by Piper (1906), Hitchcock (1959), Holmgren (1984), and Douglas (1999) and in Z. muhlenbergii by Mansion (2004). The type collection comprises small plants from Lincoln Co., Washington. Wittrock (1886) con- trasted Erythraea curvistaminea Wittr. only with E. douglasii A. Gray (=Zeltnera exaltata; the epithet douglasii is illegitimate, because the species as described by Gray [1876] included the type of the older name Cicendia exaltata Griseb., as noted, e. g., by Broome [1973, as Centaurium] and Mansion [2004]), which was the only representative of the genus recorded from Wash- ington at the time. Wittrock characterized E. curvistaminea by its incurved filaments that brought the anthers into contact with the stigma, which was borne on a short style that was erect from the first, thereby effecting self-pollination. He described the flowers of E. douglasii as differing in that the style was at first deflected in one direction and the stamens in the opposite, with both the style and the stamens later becoming erect. Broome (1973) found that floral morphology conducive to autogamy, similar to that attributed to E. curvistaminea by Wittrock, prevails among the smaller-flowered species now placed in Ze/tnera and also occurs in small flowers on plants of the predominantly larger-flowered a gE San Francisco ¢ ‘ees + 100 Miles 100KM 123 FIG. 6. Documented distribution of Ze/tnera davyi. species, and that the latter morphology, condu- cive to xenogamy, prevails among the larger- flowered species. Dunbar’s (1929) and Abrams’ (1951) statements that the anthers of C. muhlen- bergii, unlike those of C. curvistamineum, do not spiral following anthesis may have been based on observations of newly opened flowers or on a misinterpretation of Wittrock. The anthers coil helically in all species of Centaurium and Zeltnera MADRONO Nec cee ones me nas we geen s me sens ene pe est ee Se ee cee ve er etm _— —-_-=— ed — eo” _— in North America, although not until they dehisce. From my examination of the type collection, I concur with the inclusion of C. curvistamineum in Z. muhlenbergii. The inflorescences of Centaurium and Zeltnera species are cymes. Grisebach (1838, 1845) de- scribed E. muhlenbergii as having pedicellate central flowers in the dichasial divisions of the cymes. Gray (1878) and Jepson (1901, quoted ———— 2010] here) described E. muhlenbergii as having the ‘flowers in the forks with short pedicels or hardly any; lateral flowers with pedicels often as long as the flowers and with 2 bractlets at summit.” Jepson (1925) later modified this to “flowers in the forks sessile or subsessile, the others sessile or shortly pedicelled.”” Gray and Jepson (in 1901) interpreted the ultimate branches of the mono- chasially dividing portions of the cymes, termi- nating in sessile flowers, as pedicels. It is the central flowers at the proximal divisions of a Zeltnera cyme—those “‘in the forks’”—that are most likely to be pedicellate or to have the longest true pedicels, i.e., between the most distal bractlets and the base of the calyx. Dunbar (1929) described C. muhlenbergii as having pedicels absent or to 0.5 mm long and C. curvistamineum as having pedicels to 3 mm, whereas Howell (1939), mindful of Grisebach’s description, con- sidered pedicellate central flowers to be a distinctive trait of C. muhlenbergii. Abrams (1951) described C. muhlenbergii as having sessile flowers and C. curvistamineum as having. all flowers on pedicels (1) 3—12 mm. In this study I have found that plants of Z. muhlenbergii with all of the flowers sessile or nearly so are otherwise indistinguishable from plants with at least the central flowers in the proximal cyme divisions on true pedicels usually less than 10 mm, rarely to 12 mm. No discontinuity is apparent in this variation, and neither condition appears to prevail in any part of the range of the species. Plants identified as C. muhlenbergii or C. floribundum in which all or most of the true pedicels are over 10 mm, however, are actually small plants of Z. exa/tata or other species. DISTINCTION OF ZELTNERA MUHLENBERGII FROM CENTAURIUM TENUIFLORUM Much of the confusion associated with the names Centaurium muhlenbergii and C. floribun- dum is due to the absence of published reports of C. tenuiflorum (Fig. 7) as a naturalized species in North America prior to 1990, and the subsequent limitation of such reports to county checklists until plants from California and Texas were so designated by Mansion and Zeltner (2004). It was not included in The Jepson Manual: Higher Plants of California (Hickman 1993), nor by Hrusa et al. (2002) or Dean et al. (2008) in their lists of species more recently reported naturalized in California. It was recently included in the revised edition of the Marin Flora (Howell et al. 2007). Pending the availability of a more _ Satisfactory classification, the name C. tenui- _ florum is used here in a broad sense. According to Mansion et al. (2005), C. tenuiflorum s. 1. includes a diploid entity, C. tenuiflorum subsp. acutiflorum (Schott) Zeltner; a probable auto- tetraploid, C. tenuiflorum subsp. tenuiflorum; and PRINGLE: ZELTNERA MUHLENBERGII 193 an unnamed entity believed to be an allotetra- ploid derivative of diploid C. tenuiflorum X C. erythraea Rafn. Of these, the first two are native to Europe and are not known from North America. The allotetraploid is a colonizing taxon, native to Europe, western Asia, and northern Africa and naturalized in Australia (Adams 1996), New Zealand (Sykes 1981), and North America. The earliest North American specimen of C. tenuiflorum that I have seen is Davy & Blasdell 5696 (UC), collected in the North Coast Ranges of California, probably in Humboldt Co., in 1896—-notably, in the context of typification, decades after the names Erythraea muhlenbergii and E. floribunda had been published. The next earliest 1s Jepson 2022 (JEPS), from Humboldt Co., which dates from 1902. Pre-1935 records exist only for Humboldt, Butte, and Yuba counties, California, and Douglas Co., Oregon. The earliest records from the San Francisco Bay region are from the 1940’s. Except for the one record from Douglas Co., Oregon, which dates from 1916 (Peck 3649, GH), all known records of C. tenuiflorum in the Pacific states are from California (Fig. 8), but a 2005 collection from southeastern Lassen Co., California (Ahart & Dittes 11977, CHSC), indicates that this species is continuing to colonize new localities distant from previously known occurrences. The ranges of C. tenuiflorum and Zeltnera muhlenbergii now over- lap extensively in California (Figs. 3 and 8). These species sometimes occur at the same site, without intergrading, as was the case with the Ahart specimens from Sutter Co., California, cited below. Centaurium tenuiflorum and Z. muhlenbergii differ in branching pattern and in details of floral morphology. In C. tenuiflorum (Fig. 7), branch- ing is usually restricted to the distal one-eighth to one-third of the plant. The inflorescences are densely many-flowered. The flowers are sessile, or the central flowers at the proximal divisions may be on pedicels to | mm or rarely to 2 mm long. Most of the flowers in each inflorescence or major division thereof are borne at nearly the same level. The inflorescences are consequently corymboid, and the aspect of the plant is often reminiscent of Silene armeria L. Exceptional plants of C. tenuiflorum are more diffusely branched. In extreme cases the plants may be branched from near the base, and large numbers of flowers may be borne in non-corymboid inflorescences, e.g., Jepson 16757 and 19454 (JEPS). Whether such plants represent the results of viral infections, injuries, or unusual environ- mental conditions is not known. The inflorescences of Z. muhlenbergii are more open and fewer-flowered. The flowers are borne mostly singly or in groups of two or three at several to many levels in the inflorescences, which 194 MADRONO [Vol. 57 + Jepson $ 5cm 12361 FIG. 7. Representative specimens of Centaurium tenuiflorum (Jepson 12361, JEPS). 2010] PRINGLE: ZELTNERA MUHLENBERGII 195 aA ‘ ee pe SH 50 cm tall 2. Ultimate leaflet segments =18 mm long; vestiture glabrous; plants of western Klickitat Co., Washington and adjacent Hood River and Wasco Cos., Oregon iimitaes & Dai L. suksdorfii (Watson) J. M. Coult & Rose 2’ Ultimate leaflet segments =18 mm long; vestiture short hairy/cinereous; plants of Chelan Co., WA i OD Oe ee re eee L. thompsonii (Mathias) Cronquist 1’ Mature fruit length <15 mm; plants <50 cm tall; plants sometimes growing on clay-rich substrates MADRONO [Vol. 57 3. Mature fruit glabrous and narrowly oblong; <3.5mm wide and sessile or subsessile on the umbellets; plants typically growing on clay-rich substrates Sheen BMG ca EN cae. dyke a cues Oh ee ee a ena: L. bicolor var. leptocarpum (J. M. Coult. & Rose) Schlessman 3’ Mature fruit 3.5—7.5 mm wide; glabrous to sparsely to densely hirtellous; fruit clearly pedicellate on the umbellets 4. Fruit glabrous; vestiture (sub)glabrous with ultimate leaflet segments long - usually all well in excess of 10 mm L. triternatum (Pursh) J. M. Coult. & Rose var. triternatum 4’ Fruit and leaves sparsely to (usually) densely finely hairy 5. Developing and mature fruit without narrow raised dorsal ribs; caudex single; fruit trichomes linear; plants not growing on clay-rich soils; plants usually >50 cm tall; plants of central and western Klickitat Co., Washington .... ... L. triternatum (Pursh) J. M. Coult. & Rose var. brevifolium (J. M. Coult. & Rose) Mathias 5’ Mature and developing fruit with very narrow raised ribs; plants often with multicipital caudex; fruit trichomes mostly narrowly triangular; plants strictly of clay-rich soils; eastern Klickitat Co., Washington and one known locale in Union Co., Oregon Pe ee Le ee a L. tamanitchii Darrach & Thie ACKNOWLEDGMENTS We thank Johnson Meninick and Gladys Sohappy Wiltse from the Yakama Indian Nation Cultural Resources Program for assistance in providing an appropriate specific epithet for the species. We also acknowledge Mary Schlick, Cheryl Mack, Deanie Johnson, and many others for additional assistance. We are indebted to the various botanists who have provided us with input regarding this new and interesting addition to the Pacific Northwest flora; in particular, discussions with Joy Mastroguiseppe at WS were of value. Linda Brooking provided superb illustrations. Jeanette Burkhardt contributed signifi- cantly to improving the Latin description. Thanks are also owed to James Kagan, Lisa Vogler, John Luginbuhl, Rachel Chambers, Brett Anderson, and Jerry Igo. LITERATURE CITED CONSTANCE, L. 1993. Apiaceae. Pp. 136-166 in J. C. Hickman (ed.), The Jepson manual: higher plants of California. University of California Press, Berkeley, CA. COULTER, J. M. AND J. N. ROSE. 1900. Monograph of the Umbelliferae. Contributions from the U.S. National Herbarium 7:9—256. EVERT, E. F. 1983. A new species of Lomatium (Umbelliferae) from Wyoming. Madrono 30:143—146. GILL, S. J. AND J. D. MASTROGUISEPPE. 1983. Lomatium tuberosum Hoover (Apiaceae). Madrono 30:259. HARTMAN, R. L. AND L. CONSTANCE. 1988. A new Lomatium (Apiaceae) from the Sierran Crest of California. Madrono 35:121—125. HUNN, E. AND D. FRENCH. 1981. Lomatium: a key resource for Columbia Plateau native subsistence. Northwest Science 55:87—94. INTEGRATED TAXONOMIC INFORMATION SYSTEM (ITIS). 2009. On-line database. Smithsonian Insti- tution, Washington, D.D. Website http://www. itis. gov [accessed 23 March 2009]. KAGAN, J. S. 1986. A new species of Lomatium (Apiaceae) from southwestern Oregon. Madrono 33:71—75. KRUCKEBERG, A. R. 1984. California serpentines: flora, vegetation, geology, soils and management problems. University of California Press, Berkeley, CA. . 2002. Geology and plant life: the effects of landforms and rock types on plants. University of Washington Press, Seattle, WA. MATHIAS, M. E. 1938. A revision of the genus Lomatium. Annals of the Missouri Botanical Garden 25:225—297. NATURESERVE. 2009. NatureServe consevation status assessments: rank calculator version 2.0. NatureServe, Arlington, VA. Website http://www.natureserve.org/ publications/ConsStatusAssess_RankCalculator-v2. jsp [accessed 13 November 2009]. RABINOWITZ, D. 1981. Seven forms of rarity. Pp. 205— 217 in H. Synge (ed.), The biological aspects of rare plant conservation. John Wiley & Sons, Chichester, NY. SAFFORD, H. D., J. H. VIERS, AND S. P. HARRISON. 2005. Serpentine endemism in the California flora: a database of serpentine affinity. Madrono 927222-25 7. SCHLESSMAN, M. A. 1984. Systematics of tuberous lomatiums (Umbelliferae). Systematic Botany Monographs 4:1—55. AND L. CONSTANCE. 1979. Two new species of tuberous lomatiums (Umbelliferae). Systematic Botany 7:134—-149. SIMMONS, K. S. 1985. Systematic studies in Lomatium (Apiaceae). Ph.D. dissertation. Washington State University, Pullman, WA. SOLTIS, P. S. AND S. J. NOVAK. 1997. Polyphyly of the tuberous lomatiums (Apiaceae): cp DNA evidence for morphological convergence. Systematic Botany 22:99-112. THIE, K. K. 2000. Lomatium. Pp. 159-166 in P. Hirsch, P. Gladstar, and R. Gladstar, (eds.), Planting the future: saving our medicinal herbs. Healing Arts Press, Rochester, VT. USDA, NRCS. 2009. The PLANTS Database, Na- tional Plant Data Center, Baton Rouge, LA. Website http://plants.usda.gov [accessed 23 March 2009]. WASHINGTON DEPARTMENT OF NATURAL RESOURCES (WANDR). WASHINGTON INTERACTIVE GEOLOG- Ic MAP. 2009. On-line geologic map. Washington State Department of Natural Resources, Olympia, WA. Website http://wigm.dnr.wa.gov [accessed 23 March 2009]. WEINER, J. 1988. The influence of competition on plant reproduction. Pp. 228-245 in J. L. Doust and L. L. Doust, (eds.), Plant reproductive ecology: patterns and strategies. Oxford University Press, Cary, NC. MADRONO, Vol. 57, No. 3, p. 209, 2010 NOTEWORTHY COLLECTION CALIFORNIA VIOLA HOWELLII A. Gray (VIOLACEAE).—Siskiyou Co., on Elliott Creek Road (Forest Service 1050) near a signpost labeled 928 (New London Trail) and 930 (Carlton Pasture Trail), near ‘Joe Bar,’ plants scattered on shady, forest floor beneath Pseudotsuga menziesii and ferns, at 41 59.870’N, 123 06.657'W, 1295 m elev., 25 May 2008 (+30 plants, many in flower) and 22 June 2008 (numerous plants in fruit, none in flower), R. J. Little 12222 and R. J. Little 12236 (CAS, DAV, and UC). Previous knowledge. This taxon is known from British Columbia, Washington, and Oregon. It occurs in moist coniferous forests, prairies, and occasionally along creeks at 50-1500 m. Viola howellii resembles V. adunca Sm., with which it is sometimes confused. Significance. First documented record for California. In his treatment of Viola in A Manual of the Flowering Plants of California Jepson (1925) stated for V. howellii, ‘“North coastal swamps, rare, Pt. Reyes, Marin Co., and Noyo, Mendocino Co. (acc. E. Brainerd); n. to Wash.” Jepson did not provide a citation for E. Brainerd and thus this source 1s unknown. Other than this brief discussion in Jepson, there was no evidence that this species occurred in California. In A Flora of California Jepson (1936) did not mention V. howellii. In Violets of North America, Brainerd (1921) mentioned that V. howel/ii occurs in British Columbia, Oregon, and Washington, but he did not mention California. In Wild violets of North America, Viola B. Baird (Brainerd’s daughter) also noted that V. howellii occurs in British Columbia, Oregon, and Washington, but she did not mention California (Baird 1942). For V. adunca, Howell (1949) stated, “Viola howellii Gray, a closely related northern species with a short blunt spur, has been reported from Point Reyes Peninsula by Brainerd and by Mason but no plants differing from V. adunca have been seen there.’ Howell (1949) wrote that he had visited Point Reyes many times. Certainly he would be an authority on the flora of this area. The lack of V. howellii specimens in any major herbaria in California (Consortium of California Herbaria 2010) confirms Howell’s observation that V. howellii does not occur at Point Reyes. Because no evidence could be found of its presence in California, V. howellii was not treated in The Jepson manual (Hickman 1993). The potential presence of V. howellii in Siskiyou Co. was brought to my attention in 2008 by Dr. Ray Collette and is gratefully acknowledged. Specimens of V. howellii were collected at the Siskiyou Co. location by John Little in support of research on the Viola of California for the second edition of the Jepson Manual and the Flora of North America projects. The specimens were identified based on floristic treatments in Hitchcock et al. (1961), Ballard (1992), and Douglas et al. (2000). Herbarium specimens have been deposited at CAS, DAV, and UC (RJL 12236a, b, and c, respectively). —R. JOHN LITTLE, MagnaFlora, 16 Pebble River Circle, Sacramento, CA 95831. John.Little@SycamoreEnv.com. MADRONO, Vol. 57, No. 3, p. 210, 2010 NOTEWORTHY COLLECTION OREGON ALLIUM PENINSULARE Lemmon ex Greene (LILIA- CEAE).—Jackson Co., northwest flank of Soda Moun- tain in the Cascade-Sisktyou National Monument, Medford District BLM, T40S, RO3E, Section 29, in open forb meadow surrounded by Abies concolor- dominated forest, associated species: Symphoricarpos mollis, Rosa pisocarpa, Amelanchier alnifolia, Veratrum californicum, Hydrophyllum fendleri, Perideridia sp. Madia sp., full sun to filtered light, moist loam soil (vernally wet), andesite parent material, 8% slope, aspect 315°, 1670 m. elev., 25 July 2007, Scot Loring. Collected by Mark Mousseaux and Wayne Rolle 25 July 2008 (OSC 219500). Dr. Dale McNeal (Prof. Emeritus, Dept. of Biological Sciences, Univ. of the Pacific) verified the identification, and vouchers depos- ited at Oregon State University. In 2007, the collector estimated 75 plants; in 2008 we counted about 200 in the population. Cattle had grazed the area. Previous knowledge. The type locality is Las Cruces Canon, E of Ensenada, Lower California. In Jackson Co., M. E. Peck collected a specimen along Hwy. 66 near its summit across the Cascade Mountains of southern Oregon (near Klamath Co.) on 10 July 1931 (WILLU 15941). Common in central, southern and Baja California, the populations closest to Oregon are in the foothills of the Sierra Nevada near Chico, (ca. 275 km S), at Shingletown, California, off Hwy. 44 (ca. 200 km S), and in Siskiyou Co. at Kidder Creek in the Marble Mountains, 3 km W of Scott Valley and in the Scott Mountains, bordering Shasta Valley, ca. 7 km S of Gazelle (ca. 60 km S of Oregon report). Vouchers of the latter two collections by D. W. Taylor (research associate, retired, Jepson Herbarium, Univ. of Califor- nai, Berkeley) in 1996 are at JEPS. Significance. After failing to find it for over 70 yr in Oregon, botanists were beginning to speculate that the previous lone Oregon collection was either an aberrant form of another taxa, or that the species had been extirpated in Oregon. This collection confirms the presence of A. peninsulare in Jackson Co., and changes its ONHIC (Oregon Natural Heritage Information Center) status from S-H (state historical) to S-1 (imperiled because of extreme rarity). In addition to its status as rare and endangered for Oregon, it is on the BLM/Forest Service Sensitive List. —MARK MOUSSEAUX, BLM District Botanist, 3040 Biddle Road, Medford, OR 97504. Mark_Mousseaux@ or.blm.gov. MADRONO, Vol. 57, No. 3, p. 211, 2010 NOTEWORTHY COLLECTION MEXICO GUTIERREZIA RAMULOSA (Green) M. A. Lane (AS- TERACEAE).—Baja California, Isla Angel de la Guarda, elev. 113 m, 29°20'45.2”N, 113°30'22.4"W, 1 February 2005, Peter Garcia 6 (SD). ATRIPLEX HYMENELYTRA (Torr.) S. Watson (CHE- NOPODIACEAE).— Baja California, Isla Angel de la Guarda, elev. 10 m, 29°27.679'N, 113°24.460'W, 26 March 2006, Peter Garcia 10 (SD). PROSOPIS GLANDULOSA (Torr.) var. TORREYANA (L. D. Benson) M. C. Johnst. (FABACEAE).—Baja California, Isla Angel de la Guarda, elev. 16 m, 29°30'55.9"N, 113°34'06.3"W, 5 March 2006, Peter Garcia 5 (SD, BCMEX). DODONAEA VISCOSA (L.) Jacq. var. ANGUSTIFOLIA (L. f.) Benth. (SAPINDACEAE).—Baja California, Isla Angel de la Guarda, elev. 11 m, Palm Canyon, 2 mi S of Punta Rocosa, | January 2003, Peter Garcia 9 (SD, BCMEX). TAMARIX RAMOSISSIMA Ledeb. (TAMARICA- CEAE).—Baja California, Isla Angel de la Guarda, elev. 26 m, 29°27.579'N, 113°24.830'W, 26 March 2006, Peter Garcia § (SD, BCMEX). Previous knowledge. All taxa mentioned above are located on the opposite mainland of Mexico and some islands in the Sea of Cortez. Significance. All taxa previously mentioned are first collections for Isla Angel de la Guarda. The following five plants have been seen by me (and photographed), so future visitors to the island should be aware of their potential presence: Sonchus oleraceus L., Erodium texanum A. Gray, Malacothrix californica DC., Helio- tropium curassavicum L., Psittacanthus sonorae (S. Watson) Kuijt. —PETER J. GARCIA, independent botanist, Berkeley, CA. donpedro@Imi.net. ACKNOWLEDGMENTS I thank Dr. Jon Rebman (SD) for information on Baja plants on islands in the Sea of Cortez, Dr. José Delgadillo (BCMEX) for help in obtaining collection permits, and Dr. Staci Markos (UC/JEPS) and the Annetta Carter Fund of the California Botanical Society for funding. LITERATURE CITED BAIRD, V. B. 1942. Wild violets of North America. Univer- sity of California Press, Berkley and Los Angeles, CA. BALLARD, H. E. 1992. Systematics of Viola section Viola in North America North of Mexico. M.S. thesis. Central Michigan University, Mount Pleasant, MI. BRAINERD, E. 1921. Violets of North America. Ver- mont Agricultural Experiment Station. Bulletin 224. Free Press Printing Company, Burlington, VT. CONSORTIUM OF CALIFORNIA HERBARIA. 2010. Regents of the University of California. Website http://ucjeps. berkeley.edu/consortium/ [accessed 14 July 2010]. DOUGLAS, G. W., D. V. MEIDINGER, AND J. POJAR (EDS.). 2000. Illustrated flora of British Columbia, Vol. 5: Dicotyledons. B.C. Ministry of Environment, Lands and Parks, and B.C. Ministry of Forests, Victoria, B.C. HICKMAN, J. C. (ED.). 1993. The Jepson manual: higher plants of California. University of California Press, Berkeley, CA. HITCHCOCK, C. L., A. CRONQUIST, M. OWNBEY, AND J. W. THOMPSON. 1961. Vascular plants of the Pacific Northwest, Part 3. University of Washing- ton Press, Seattle, WA. JEPSON, W. L 1925. A manual of the flowering plants of California. University of California Press, Berkeley, CA. . 1936. A flora of California. Vol. 2. Associated Students Store, University of California, Berkeley, CA. MADRONO, Vol. 57, No. 3, p. 212, 2010 N ANNOUNCEMENT EDITORS’ NOTE Dear Madrono Reader In consultation with the Board of the Califor- nia Botanical Society, this volume of Madrono begins using a different style for the Noteworthy Collections. In the past, condensed in text references were provided. This typically led to confusion with authors (and the editors) on the correct style for these references. We are now employing the standard in text call out for references and a single compiled list of references for all the Noteworthy Collections in a single issue. Our goal is to make references more consistent throughout the issue, but realize our readers may have a different view. Please provide feedback to let us know if this 1s perceived as a welcomed change in style. Timothy Lowrey Richard Whitkus Volume 57, Number 3, pages 145—212, published 21 December 2010 SUBSCRIPTIONS — MEMBERSHIP The California Botanical Society has several membership types (individuals ($35 per year; family $40 per year; emeritus $27 per year; students $27 per year for a maximum of 7 years). Late fees may be assessed. Beginning in 2011, rates will increase by $5 for all membership types except life memberships, for which rates will increase by $100, and student memberships, which will not show a rate increase. Members of the Society receive MADRONO free. Institutional subscriptions to MADRONO are available ($70). Membership is based on a calendar year only. Life memberships are $750. Applications for membership (including dues), orders for subscriptions, and renewal payments should be sent to the Membership Chair. Requests and rates for back issues, changes of address, and undelivered copies of MADRONO should be sent to the Corresponding Secretary. INFORMATION FOR CONTRIBUTORS Manuscripts submitted for publication in MADRONO should be sent to the editor preferably as Microsoft Word (.doc), Rich Text Format (.rtf), or Portable Document Format (.pdf) files. It is preferred that all authors be members of the California Botanical Society. Manuscripts by authors having outstanding page charges will not be sent for review. Manuscripts may be submitted in English or Spanish. English-language manuscripts dealing with taxa or topics of Latin America and Spanish-language manuscripts must have a Spanish RESUMEN and an English ABSTRACT. For all articles and short items (NOTES, NOTEWORTHY COLLECTIONS, POINTS OF VIEW, etc.), fol- low the format used in recent issues for the type of item submitted. Allow ample margins all around. Manuscripts MUST BE DOUBLE-SPACED THROUGHOUT. For articles this includes title (all caps, centered), author names (all caps, centered), addresses (caps and lower case, centered), abstract and resumen, five key words or phrases, text, acknowledgments, literature cited, tables (caption on same page), and figure captions (grouped as consecutive paragraphs on one page). Order parts in the sequence listed, ending with review copies of illustrations. The title page should have a running header that includes the name(s) of the author(s), and a shortened title. Avoid foot- notes except to indicate address changes. Abbreviations should be used sparingly and only standard abbreviations will be accepted. Table and figure captions should contain all information relevant to information presented. All measurements and elevations should be in metric units, except specimen citations, which may include English or metric measurements. Authors are encouraged to include the names, addresses, and e-mail addresses of two to four potential reviewers with their submitted manuscript. Authors of accepted papers are required to submit an electronic version of the manuscript. Microsoft Word 2000 or later or WordPerfect 9.0 (or later) for Windows is the preferred software. Line copy illustrations should be clean and legible, proportioned to the MADRONO page. Scales should be in- cluded in figures, as should explanation of symbols, including graph coordinates. Symbols smaller than | mm after reduction are not acceptable. Maps must include a scale and latitude and longitude or UTM references. Presentation of nomenclatural matter (accepted names, synonyms, typification) should follow the format used by Sivinski, Robert C., in MADRONO 41(4), 1994. Institutional abbreviations in specimen citations should follow Index Herbariorum. Names of authors of scientific names should be abbreviated according to Brummitt and Powell, Authors of Plant Names (1992) and, if not included in this index, spelled out in full. Titles of all periodicals, serials, and books should be given in full. Books should include the place and date of publication, publisher, and edition, if other than the first. All California Botanical Society members current in the volume year that their contributions are published are allowed five free pages per volume year. Additional pages will be charged at the rate of $40 per page. Joint authors may apply all or a portion of their respective five-page allotments to a jointly-published article. Partial pages will be charged as full. The purpose of this fee is not to pay directly for the costs of publishing any particular paper, but rather to allow the Society to continue publishing MADRONO on a reasonable schedule, with equity among all members for access to its pages. Printer’s fees for color plates and other complex matter (including illustrations, charts, maps, photographs) will be charged at cost. Author’s changes after typesetting @ $4.50 per line will be charged to authors. Page charges are important in maintaining Madrono as a viable publication, and timely payment of charges is appreciated. At the time of submission, authors must provide information describing the extent to which data in the manuscript have been used in other papers that are published, in press, submitted, or soon to be submitted elsewhere. SMITHSONIAN INSTITUTION LIBRARIES VALI 3 9088 01553 710