-r^X MM THE TEXAS JOURNAL OF SCIENCE GENERAL INFORMATION MEMBERSHIP.— Any person or member of any group engaged in scientific work or interested in the promotion of science is eligible for membership in The Texas Academy of Science. For more information regarding membership, student awards, section chairs and vice-chairs, the annual March meeting and author instructions, please access the Academy’s homepage at: http: //hsb.baylor . edu/html/tas/ Dues for regular members are $30.00 annually; supporting members, $60.00; sustaining members, $100.00; patron members, $150.00; associate (student) members, $15.00; family members, $35.00; affiliate members, $5.00; emeritus members, $10.00; corporate members, $250.00 annually. Library subscription rate is $50.00 annually. The Texas Journal of Science is a quarterly publication of The Texas Academy of Science and is sent to most members and all subscribers. Payment of dues, changes of address and inquiries regarding missing or back issues should be sent to: Dr. Fred Stevens, Executive Secretary The Texas Academy of Science CMB 5980 Schreiner University Kerrville, Texas 78028-5697 E-mail: FStevens@schreiner.edu AFFILIATED ORGANIZATIONS American Association for the Advancement of Science, Texas Council of Elementary Science Texas Section, American Association of Physics Teachers Texas Section, Mathematical Association of America Texas Section, National Association of Geology Teachers Texas Society of Mammalogists The Texas Journal of Science (ISSN 0040-4403) is published quarterly at Lubbock, Texas, U.S.A. Periodicals postage paid at San Angelo, Texas and additional mailing offices. POSTMASTER: Send address changes and returned copies to The Texas Journal of Science, Dr. Fred Stevens, CMB 5980, Schreiner University, Kerrville, Texas 78028-5697, U.S.A. The known office of publication for The Texas Journal of Science is the Department of Biology, Angelo State University, San Angelo, Texas 76909; Dr. Ned E. Strenth, Managing Editor. THE TEXAS JOURNAL OF SCIENCE Volume 54, No. 1 February, 2002 CONTENTS A Review of Species Names for Ammonia and Elphidium, Common Foraminifera along the Texas Gulf Coast. By Pamela Buzas-Stephens , Emile A. Pessagno, Jr. and C. Jerry Bowen . 3 Analysis of Horse ( Equus ) Metapodials from the Late Pleistocene of the Lower Nueces Valley, South Texas. By Jon A. Baskin and Antonia E. Mosqueda . 17 Silica-scaled Chrysophytes and Synurophytes from East Texas. By Daniel E. Wujek, James L. Wee and James E. Van Kley . . 27 Enzymatic Variation in the Land Snail Euglandina texasiana (Gastropoda: Pulmonata) from South Texas and Northeastern Mexico. By Kathryn E. Perez and Ned E. Strenth . 37 Spatial Associative Learning in the Crevice Spiny Lizard, Sceloporus poinsettii (Sauria: Iguanidae). By Fred Punzo . . . 45 Long-term Structural Habitat Use of Male Specimens of Two Native and One Introduced Anolis (Iquanidae) Species on the North Coast of Jamaica. By Allan J. Landwer and Gary W. Ferguson . 51 Habitat Utilization by Eastern Yellowbelly Racers ( Coluber constrictor flaviventris) in Southwest Dallas County, Texas. By Richard D. Reams and William H. Gehrmann . 59 Effects of Temperature and Light on Chinese Tallow ( Sapium sebiferum) and Texas Sug'arberry ( Celtis laevigata) Seed Germination. By Somereet Nijjer, Richard A. Lankau, William E. Rogers and Evan Siemann ... 63 Upstream Changes and Downstream Effects of the San Marcos River of Central Texas. By Richard A. Earl and Charles R. Wood . . 69 General Notes The Ghost-faced Bat, Mormoops megalophylla, (Chiroptera: Mormoopidae) from the Davis Mountains, Texas. By Robert S. DeBaca and Clyde Jones . 89 Membership Application . 92 Author Instructions 93 THE TEXAS JOURNAL OF SCIENCE EDITORIAL STAFF Managing Editor: Ned E. Strenth, Angelo State University Manuscript Editor: Patrick L. Odell, Baylor University Associate Editor for Botany: Janis K. Bush, The University of Texas at San Antonio Associate Editor for Chemistry: John R. Villarreal, The University of Texas-Pan American Associate Editor for Computer Science: Nelson Passos, Midwestern State University Associate Editor for Environmental Science: Thomas LaPoint, University of North Texas Associate Editor for Geology: Ernest L. Lundelius, University of Texas at Austin Associate Editor for Mathematics and Statistics: E. Donice McCune, Stephen F. Austin State University Associate Editor for Physics: Charles W. Myles, Texas Tech University Manuscripts intended for publication in the Journal should be submitted in TRIPLICATE to: Dr. Patrick L. Odell TJS Manuscript Editor Institute of Statistics Baylor University - P.O. Box 97225 Waco, Texas 76798 Pat_Odell@bay lor . edu Scholarly papers reporting original research results in any field of science, technology or science education will be considered for publication in The Texas Journal of Science. Instructions to authors are published one or more times each year in the Journal on a space-available basis, and also are available from the Manuscript Editor at the above address. They are also available on the Academy’s homepage at: http : / /hsb . bay lor . edu/html/tas/ The Texas Journal of Science is published quarterly in February, May, August and November for $30 per year (regular membership) by The Texas Academy of Science. Periodical postage rates (ISSN 0040-4403) paid at Lubbock, Texas. Postmaster: Send address changes and returned copies to Dr. Fred Stevens, Executive Secretary, CMB 5980, Schreiner University, Kerrville, Texas 78028- 5697, U.S.A. TEXAS J. SCI. 54( 1): 3-16 FEBRUARY, 2002 A REVIEW OF SPECIES NAMES FOR AMMONIA AND ELPHIDIUM , COMMON FORAMINIFERA ALONG THE TEXAS GULF COAST Pamela Buzas-Stephens, Emile A. Pessagno, Jr.* and C. Jerry Bowen Department of Biology, Midwestern State University 3410 Taft Blvd. , Wichita Falls, Texas 76308 and * Department of Geosciences , University of Texas at Dallas P. O. Box 830688, Richardson, Texas 75083 Abstract.— Species names for Ammonia and Elphidium have continually changed since these taxa were first described in Texas coastal environments. As a result, classification is problematic and the literature is inconsistent. The purpose of this paper is to evaluate the taxonomic status of species currently assigned to Ammonia and Elphidium. This task has been accomplished through extensive literature review and through comparison of specimens from this study with those in the Cushman Collection at the National Museum of Natural History. Most Elphidium found along the Texas coast are assignable to either Elphidium gunteri or E. excavatum, and the Ammonia present are assignable to Ammonia parkinsoniana and A. tepida. Present geographic, molecular and reproductive evidence shows that the species names A. parkinsoniana and A. tepida, not A. beccarii, should be used to describe these morphotypes of Ammonia wherever they occur, including the Gulf of Mexico, the east coast of North America, the Caribbean and the Pacific. Since recent foraminifera on the Texas Gulf Coast were first described (Kornfeld 1931; Phleger & Parker 1951; Post 1951; Parker et al. 1953), species names for the most common estuarine taxa, Ammonia and Elphidium , have continually changed in the literature, making it difficult for subsequent workers to identify specimens. Specific names of these foraminifera have changed as it was recognized that many of the Gulf Coast morphotypes (shell types) are also found around the world (Miller et al. 1982; Buzas et al. 1985; Walton & Sloan 1990; Hayward et al. 1997; 1999), and that rather than being many separate species, a single species can have a variety of morphotypes (Schnitker 1974; Poag 1978; Miller et al. 1982; Walton & Sloan 1990; Stouff et al. 1999a). The purpose of this paper is to assess the species names for the most commonly encountered morphotypes of Ammonia and Elphidium found in Texas coastal environments. The taxonomy proposed for Ammonia herein also affects the species name used for this morphotype on a worldwide basis. To determine current taxonomy, a survey of the literature was undertaken and specimens from Laguna Madre, Nueces Bay, the Arroyo Colorado, and Laguna Atascosa were compared with specimens in the Smithsonian National Museum of Natural History. 4 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 1, 2002 Figure 1 . Map of the southern Texas coast showing the four sampling localities (Nueces Bay, southern Laguna Madre, the Arroyo Colorado and Laguna Atascosa). Methods Sediment cores, varying from 33 cm to 95 cm in length depending on bottom conditions, were taken at four sites along the south Texas coast: BUZAS-STEPHENS ET AL. 5 Nueces Bay, southern Laguna Madre, the Arroyo Colorado, and Laguna Atascosa (Fig. 1). Since the spatial distribution of foraminifera is not uniform (Buzas 1968), four cores were taken at each locality in order to get an accurate representation of the taxonomic assemblages. To evaluate living foraminifera present during sampling, the top 2 cm of the cores was stained with rose Bengal (Walton 1952; Murray & Bowser 2000), a protein- specific stain. Cores were sliced in 1 to 5 cm intervals (depending on other analyses performed), and 20 ml of sedi¬ ment from selected intervals was washed through a 230 mesh (.062 mm) sieve. If foraminifera were not abundant, as in the Arroyo Colorado samples, they were concentrated for ease of counting using a sodium poly tungstate flotation technique (Callahan 1987). From each locality, approximately 300 foraminifera were counted from the surface sediments of each core and from different depths in at least one core per site. Specimens of Elphidium gunteri, E. excavatum, Ammonia parkin- soniana and A. tepida figured herein have been deposited in the Cushman Collection, National Museum of Natural History (USNM). Results Elphidium gunteri , E. excavatum , Ammonia parkinsoniana and A. tepida (Figure 2) are the most common taxa at the sampling localities, together comprising from 60% to over 90% of the total assemblage in surface sediments and at depth. A range of shell types is exhibited within each of these species. Other taxa present at the sites are Quinqueloculina seminula , Hccynesina germanica and Ammotium salsum (for a complete account of taxa, see Buzas-Stephens 2001). Living Ammonia and Elphidium were found at all localities except for the Arroyo Colorado, where there were no living Ammonia and overall numbers (living plus dead) of Ammonia are low. The river has a history of poor water quality conditions (Bryan 1971) that may be responsible for the paucity of Ammonia (Buzas-Stephens 2001). Interestingly, perfectly preserved Cretaceous through Eocene age foraminifera are present in the samples from southern Laguna Madre, the Arroyo Colorado, and Laguna Atascosa. Since much of the sedi¬ ment in southern Laguna Madre is derived from offshore, many of these specimens are probably reworked at least a couple times from old Rio Grande delta distributary deposits (Rusnak 1960). 6 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 1, 2002 Figure 2. Most common species of Ammonia (except for Ammonia beccarii) and Elphidium present in Nueces Bay, Laguna Madre, the Arroyo Colorado and Laguna Atascosa (southern Texas). (2a) Elphidium gunteri Cole, scale = 200 pm, USNM 517707. (2b, c) Elphidium excavatum (Terquem), scale = 100 pm, (2b) USNM 517708; (2c) USNM 517709. (2d,e) Ammonia beccarii (Linne'), (2d) spiral view; scale = 907 ^m; (2e) umbilical view, scale = 880 pm. Micrographs scanned from Walton & Sloan 1990 (Plate 3, Figs. 3a & 3b). This species is not present at the above localities, but is included here for comparison with A. parkinsoniana and A. tepida. (2f,g) Ammonia parkins oniana (d’Orbigny), (2f) spiral view, scale = 200 pm, USNM 517710; (2g) umbilical view, scale = 100 pm, USNM 517711. (2h,i) Ammonia tepida (Cushman), (2h) spiral view, scale = 200 pm, USNM 517712; (2i) umbilical view, scale = 100 pm, USNM 517713. BUZAS-STEPHENS ET AL. 7 Systematics Identification of the different species of Ammonia and Elphidium was accomplished through comparison of specimens from this study with those housed in the Cushman Collection, National Museum of Natural History (NMNH) in Washington, D.C. To ascertain identifications, hypotypes and specimens from Phleger & Parker (1951), Parker et al. (1953) and Poag (1981) were examined. Although many other authors have recorded species of Ammonia and Elphidium in Texas coastal environments (for a compilation, see Culver & Buzas 1981), most of these taxa are not illustrated and thus not available for comparison in future studies such as this. Preliminary attempts to classify specimens from this work through illustrations and micrographs from Phleger & Parker (1951), Parker et al. (1953) and Poag (1981) were largely unsuccessful. Correct identifi¬ cations were made only through direct comparison of individuals under the light microscope. Unless a worker is thoroughly familiar with an assemblage, identifications should be based on actual specimens, and not on figures from the literature. The following taxonomy through the genus level is from Loeblich & Tappan (1987). Synonymies are presented for foraminiferal studies specific to the Texas Gulf Coast. Micrographs were taken with a Philips XL30ESEM. Order FORAMINIFERIDA Eichwald Suborder ROTALIINA Delage & Herouard Superfamily ROTALIACEA Ehrenberg Family ELPHIDIIDAE Galloway Subfamily ELPHIDIINAE Galloway Genus Elphidium de Montfort Elphidium gunteri Cole Figure 2a Elphidium gunteri Cole, 1931:34, pi. 4, figs. 9, 10.— Parker et al., 1953:8, pi. 3, figs. 18- 19. -Parker, 1954:508, pi. 6, fig. 16. Elphidium gunteri Cole var. galvestonensis Kornfeld (part), 1931:87-88, pi. 15, figs. 2a, b, 3a, b (not figs, la, b).— Post, 1951:172, pi. 1, fig. 14. Elphidium gunteri Cole var. galvestonense— Phleger & Parker, 195 1 : 10, 8 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 1, 2002 pi. 5, figs. 13-14. Cellanthus gunteri Cole, Wantland, 1969:109, pi. 3, fig. 5. Elphidium gunteri forma typicum—Poag, 1978:402, pi. 2, figs. 13-16, 22-23. -Poag, 1981:61-62, pis. 37-38, figs, la, b. Material examined. — USNM 517707. Distribution. — Elphidium gunteri comprises 8% of the total taxa in the surface sediments (0-2 cm) of Nueces Bay, 17% in Laguna Madre, 80% in the Arroyo Colorado and 10% in Laguna Atascosa. Remarks.— As in coastal environments worldwide (Miller et al. 1982; Hayward 1997), the wide range of Elphidium shell types present along the Texas Gulf Coast provides a continual challenge to researchers attempting to assign them to species. One of the most common estuarine forms, which has straight sutures, prominent sutural bridges and umbilical bosses, and large, widely spaced pores has been consistently assigned to Elphidium gunteri Cole (Kornfeld 1931; Phleger & Parker 1951; Post 1951; Parker et al. 1953; Wantland 1969; Poag 1978; 1981) (Figure 2a). However, the above characteristics defining E. gunteri grade into other common morphotypes, including E. excavatum (Ter- quem) (Figures 2b & c). Since the characteristics of E. gunteri and E. excavatum can overlap (Buzas et al. 1985), it is sometimes difficult to assign a specimen to a species, even after counting thousands. Elphidium excavatum (Ter quern) Figures 2b & c Polystomella excavata Ter quern, 1875:25, pi. 2, fig. 2. Elphidium translucens (Natl and).— Post, 1951:173, pi. 1, fig. 17.— Parker et al, 1953:9, pi. 3, fig. 27. Protelphidium delicatulum (Bermudez) . — W anti and , 1969: 1 10, pi. 3, fig. 7. Elphidium delicatulum (Bermudez) . —Parker et al., 1953:7, pi. 3, figs. 12, 17. Elphidium gunteri Cole forma salsum. — Poag, 1978:402, pi. 2, figs. 1-12, 17-21. -Poag, 1981:61, pis. 37-38, figs. 2, 2a, b. Material examined. —USNM 517708, 517709. BUZAS-STEPHENS ET AL. 9 Distribution.— In the surface sediments Elphidium excavatum makes up 15% of the assemblage in Nueces Bay, 17% in Laguna Madre, 8% in the Arroyo Colorado and 1 1 % in Laguna Atascosa. Remarks.— Also ubiquitous to coastal environments worldwide (Miller et al. 1982; Buzas et al. 1985; Hayward 1997), Elphidium excavatum can generally be distinguished from E. gunteri by the presence of more numerous and smaller pores, lesser development of sutural bridges and fewer umbilical bosses. Because these features can vary widely within the species, E. excavatum has “probably been misidentified more than any other foraminiferal species” (Buzas et al. 1985). Until this study, the name E. excavatum has never been assigned to Texas coastal Elphidium , though Culver & Buzas (1981) provided synonymies in their compilation of foraminiferal studies from the Gulf of Mexico. Instead, E. excavatum has mostly been referred to as E. translucens (Post 1951; Parker et al. 1953) or E. delicatulum (Parker et al. 1953; Wantland 1969), though it surely has had other synonyms as well. In the current project, E. excavatum and E. gunteri together comprise about 98% of all Elphidium found at each site, making these two species the most widespread and abundant species of Elphidium in southern Texas estuaries. Although some authors have correlated the different shell types of E. gunteri and E. excavatum with environmental variables (Poag 1978; Miller et al. 1982), a similar correlation cannot be seen in this study. Perhaps future DNA and/or reproductive studies will help clarify the extent of genetic versus environmental control over the production of different Elphidium shell types. Family ROT ALII DAE Ehrenberg Subfamily AMMONIINAE Saidova Genus Ammonia Brunnich Ammonia parkinsoniana (d’Orbigny) Figures 2f & g Rosalina parkinsoniana d’Orbigny, 1839:99, pi. 4, figs. 25-27 . Rotalia beccarii (Linne' var. parkinsoniana (d’Orbigny).— Kornfeld, 1931:90-91, pi. 13, figs, la, b, c.— Phleger & Parker, 1951:23, pi. 12, figs. 6a, b. Rotalia beccarii (Linne').— Post, 1951:176, pi. 1, fig. 20. 10 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 1, 2002 Rotalia beccarii (Linne') variant A.— Parker et al., 1953:13, pi. 4, figs. 20-22. Rotalia beccarii (Linne') variants.— Parker, 1954:531, pi. 10, figs. 1, 2, 5, 6. Ammonia beccarii (Linne') variants.— Wantl and, 1969:109, pi. 3, figs. la, b, c. Ammonia parkinsoniana (d’Orbigny) forma typica. — Poag, 1978:397, pi. 1, figs. 5-9, 13-16, 19-21.— Poag, 1981:38, pis. 45-46, figs. 1, la, lb. Material examined.— USNM 517710, 517711. Distribution.— In surface sediments, Ammonia parkinsoniana comprises 63% of the total fauna in Nueces Bay, 21% in Laguna Madre, 4% in the Arroyo Colorado and 32% in Laguna Atascosa. Remarks. — Among the different species of Ammonia found world¬ wide, there are three very common forms (Walton & Sloan 1990): (1) An ornamented shell morphotype that has distinct beading, fluting and/or furrowing along the sutures on one or both sides, and an umbilical plug (Figures 2d, e). This form was first described from the Mediterranean as A. beccarii (Linne' 1758); (2) An unornamented morphotype with an umbilical plug that lacks the above beading/fluting/furrowing along sutures (Figures 2f, g). This form was first described from the Carib¬ bean as A . parkinsoniana (d’Orbigny 1839); and (3) A smaller, more lobate morphotype that has neither ornamentation nor an umbilical plug (Figures 2h, i), also first described from the Caribbean. This third morphotype is probably the A. catesbyana of d’Orbigny (1839). When updating d’Orbigny’s Cuban collection, Le Calvez (1977) designated a neotype for A. parkinsoniana , but apparently no identifiable specimens of A. catesbyana were available for the designation of a neotype for this species. In addition, d’Orbigny’s type figures are vague (Poag 1978), leaving the formal species name for the small lobate morphotype in question. As well as having differences in overall appearance, the three morphotypes vary in their geographic ranges. The ornamented form occurs mainly in northern waters such as the Mediterranean, the north¬ eastern Atlantic (no references are available south of 27N in the eastern Atlantic), and the western Atlantic near Cape Cod (Walton & Sloan BUZAS-STEPHENS ET AL. 1 1990). The other two morphotypes are found around the world (Walton & Sloan 1990), and are exclusive to southern oceans including the Pacific, Indian, and south Atlantic (Walton & Sloan 1990). They also occur in Texas estuaries and near shore in the Gulf of Mexico (Poag 1981; Walton & Sloan 1990). In the early 1900s, Cushman (1926) began using the species name A. beccarii for all three forms, using variety names to distinguish them, and this terminology persisted. Thus the ornamented form was called A. beccarii var. beccarii, the one without ornamentation was A. beccarii var. parkinsoniana, and the lobate one without a plug was A. beccarii var. tepida. Many workers today refer to the smaller, lobate form as A. tepida, the varietal name first used by Cushman (1926). Neither "variety" nor "forma" names is governed by the International Code of Zoological Nomenclature (Ride et al. 1985). In their classic studies documenting Texas Gulf Coast foraminifera, Kornfeld (1931), Phleger & Parker (1951) and Parker et al. (1953) also used the names A. beccarii var. parkinsoniana and A. beccarii var. tepida (or "Rotalia" beccarii , as the generic name became for awhile). Poag (1978) reinstated d’Orbigny’s original name, Ammonia parkinsoni¬ ana , for the unornamented shell type, designating it as A. parkinsoniana forma typica. He called the lobate form Ammonia parkinsoniana forma tepida . Poag (1978) asserted that A. parkinsoniana was a different species than A. beccarii because the ornamented form does not occur in southern oceans (Poag 1978; Walton & Sloan 1990). Since Poag’s re-introduction of the name A. parkinsoniana , both A. parkinsoniana (see Yuill 1991; Colburn & Baskin 1998) and A. beccarii (see Williams 1995) have been used to describe Texas Gulf Coast Ammonia. More recent investigations into the DNA (Pawlowski et al. 1995; Holzmann & Pawlowski 1997; Holzmann 2000) and the life cycle (Goldstein & Moodley 1993; Stouff et al. 1999a) of Ammonia are help¬ ing to clarify the role genetics plays in producing these shell types. Through ribosomal DNA sequencing, Pawlowski et al. (1995) deter¬ mined that the three morphotypes have distinct sequences indicating that they are separate species. Another molecular study (Holzmann & Pawlowski 1997) again shows that A. tepida can be distinguished genetically from another Ammonia morphotype. Holzmann (2000) further shows how different species of Ammonia found worldwide can be characterized according to morphology and DNA. 12 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 1, 2002 In the most current reproductive study, Stouff et al. (1999a) found that the different stages in the life cycle of Ammonia tepida have a characteristic shell forms. The asexual, diploid schizont stage typically has one or more umbilical plugs, while the sexual, haploid gamont stage lacks an umbilical plug. They did not observe the ornamented form as part of the life cycle of A . tepida or as a result of ontogeny. Likewise, ornamented offspring have not been produced from any cultures of A. tepida to date (Bradshaw 1957; Schnitker 1974; Goldstein & Moodley 1993; Stouff et al. 1999). However, Schnitker (1974) cultured two individuals of the ornamented form (A. beccarii), and reported "most of their offspring were similar to the tepida offspring". To summarize, present geographic, reproductive, and molecular evidence show that the ornamented, unornamented and lobate morpho- types of Ammonia are separate species. Since the name A. beccarii was first assigned to the ornamented morphotype (Linne' 1758), it should be reserved exclusively for this form. The name A. parkinsoniana (d’Orbigny 1839) is the valid name for the unornamented form, and apparently other authors agree with this conclusion (Jorissen 1988; Sen Gupta et al. 1996; Colburn & Baskin 1998; Hayward et al. 1999). Ammonia tepida (Cushman) Figures 2h & i Rotalia beccarii (Linne') var. tepida Cushman, 1926:79, pi. 1.— Kornfeld, 1931:91, pi. 13, figs. 3a, b, c.— Phleger & Parker, 1951:23, pi. 12, figs. 7a, b.-Post, 1951:176, pi. 1, figs. 21, 22. Rotalia beccarii (Linne') variants B, C.— Parker et al., 1953:13, pi. 4, figs. 25-30. Rotalia pauciloculata (Phleger & Parker).— Phleger & Parker, 195 1 : 23, pi. 12, figs. 8a, b, 9a, b.— Parker et al., 1953:13-14, pi. 4, figs. 31, 37. Ammonia pauciloculata (Phleger & Parker).— Poag, 1981:38-39, pis. 45-46, figs. 3, 3a, b. Ammonia beccarii (Linne') variants.— Wantl and, 1969:109, pi. 3, figs. 2a, b, c, 3a, b, c. Ammonia parkinsoniana (d’Orbigny) forma tepida.— Poag, 1978:397, pi. 1, figs. 1-4, 10-12, 17, 18.— Poag, 1981:37-38, pis. 45-46, figs. 2, 2a, b. BUZAS-STEPHENS ET AL. 13 Table 1 . Percent of Ammonia parkinsoniana versus Ammonia tepida as compared to salinity. Nueces Bay Laguna Madre Arroyo Colorado Laguna Atascosa Salinity 27 ppt 26 ppt 14 ppt 6 ppt A. parkinsoniana 96% 82% 45% 45% A. tepida 4% 18% 55% 55% Material examined. — USNM 517712, 517713. Distribution.— Ammonia tepida comprises 2% of the assemblage in the surface sediments of Nueces Bay, 4% in Laguna Madre, 4% in the Arroyo Colorado and 39% in Laguna Atascosa. Ammonia tepida is more abundant than A. parkinsoniana at the sites with lower salinities (the Arroyo Colorado and Laguna Atascosa) (Table 1). Hayward et al. (1999) note a similar correlation with salinity, but Poag (1978) observed the opposite. Remarks. — Although the taxonomy for the small, lobate morphotype will not be stabilized until a neotype is established, it is recommended that this form be assigned to Ammonia tepida since this name is already widely used (Pawlowski et al. 1995; Yanko et al. 1994; 1998; Geslin et al. 1998; Stouffetal. 1999a; 1999b). Conclusions The most common estuarine foraminifera found along the Texas Gulf Coast are Elphidium gunteri , E. excavation, Ammonia parkinsoniana and A. tepida. These species are also found worldwide. Since the defining characteristics of E. gunteri and E. excavatum overlap, care must be exercised when assigning species names to these forms. Elphidium excavatum , in particular, exhibits a wide range of shell morphotypes and thus can be quite challenging to identify. Current geographic, molecular and reproductive evidence shows that A. parkinsoniana and A. tepida have erroneously been called A. beccarii for most of the 20th century. The unornamented morphotype of Ammonia should be called A. parkin¬ soniana and the small, lobate form is A. tepida. Acknowledgments This research was partially funded by Geological Society of America grant 6148-97. The authors also wish to thank Martin A. Buzas for kindly reviewing the manuscript. 14 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 1, 2002 Literature Cited Bradshaw, J. S. 1957. Laboratory studies on the rate of growth of the foraminifer, "Streblus beccarii (Linne') var. tepida (Cushman)". J. Paleo., 3 1 (6) : 1 1 38- 1 1 47. Buzas, M. A. 1968. On the spatial distribution of foraminifera. Contributions from the Cushman Foundation for Foraminiferal Research, 19(1): 1-11. Buzas, M. A., S. J. Culver & L. B. Isham. 1985. A comparison of fourteen elphidiid (Foraminiferida) taxa. J. Paleo., 59(5): 1075-1090. Buzas-Stephens, P. 2001. Foraminiferal analysis of sediment cores from Laguna Madre, Nueces Bay, the Arroyo Colorado, and Laguna Atascosa: South Texas Coast. Unpublished Ph.D. Dissertation, Univ. of Texas at Dallas, 173 pp. Bryan, C. E. 1971. An ecological survey of the Arroyo Colorado, Texas. Texas Parks and Wildlife Dept. Technical Series, no. 10, 28 pp. Callahan, J. 1987. A nontoxic heavy liquid and inexpensive filters for separation of mineral grains. J. Sed. Petrol., 57:765-766. Colburn, D. F. & J. A. Baskin. 1998. A morphological study of Ammonia parkins oniana from Laguna Madre and Baffin Bay, Texas. Gulf Coast Assoc, of Geol. Societies Transaction, XLVIII: 1 1-19. Cole, W. S. 1931. The Pliocene and Pleistocene foraminifera of Florida. Florida Geol. Surv. Bull., 6:1-79. Culver, S. J. & M. A. Buzas. 1981. Distribution of Recent Benthic Foraminifera in the Gulf of Mexico, Volume II. Smithsonian Contributions to the Marine Sciences, no. 8, Smithsonian Institution Press, Washington DC :41 3-898. Cushman, J. A. 1926. Recent Foraminifera from Porto Rico. Carnegie Institution of Washington Publications, no. 344:73-84. Geslin, E., J.-P. Debenay & M. Lesourd. 1998. Abnormal wall textures and test deformation in Ammonia (hyaline foraminifer). J. Foraminiferal Research, 28(2): 148-156. Goldstein, S. T. & L. Moodley. 1993. Gametogenesis and the life cycle of the foraminifer Ammonia beccarii (Linne') forma tepida (Cushman). J. Foraminiferal Research, 23(4):213-220. Hayward, B. W., C. J. Hollis & H. R. Grenfell. 1997. Recent Elphidiidae (Foraminiferida) of the South-west Pacific and fossil Elphidiidae of New Zealand. Institute of Geological and Nuclear Sciences, monograph no. 16:166 pp. Hayward, B. W., H. R. Grenfell, C. M. Reid & K. A. Hayward. 1999. Recent New Zealand shallow-water benthic foraminifera: taxonomy, ecologic distribution, biogeography, and use in paleoenvironmental assessment. Institute of Geological and Nuclear Sciences, monograph no. 21:258 pp. Holzmann, M. 2000. Species concept in foraminifera: Ammonia as a case study. Micropaleontology, 46(l):21-37. Holzmann, M. & J. Pawlowski. 1997. Molecular, morphological and ecological evidence for species recognition in Ammonia (Foraminifera). J. Foraminiferal Research, 27 (4) : 3 1 1-318. Jorissen, F. J. 1988. Benthic foraminifera from the Adriatic Sea; principles of phenotypic variation. Ultrecht Micropaleo. Bull., 37: 176 pp. Kornfeld, M. M. 1931. Recent littoral foraminifera from Texas and Louisiana. Contrib. Dept. Geol. Stanford Univ., 1:77-101. Le Calvez, Y. 1977. Revision des Foraminiferes de la collection d’Orbigny. II. Foraminiferes de file de Cuba. Cahiers de Micropaleontologie, C. N. R. S., 2, part 2, 130 pp. Linne', C. 1758. Systema Naturae. Edition 10, vol. 1, Holmiae ( = Stockholm): L. BUZAS-STEPHENS ET AL. 15 Salvii.G. Engleman (Lipsiae), 824 pp. Loeblich, A. R., Jr. & H. Tappan. 1987. Foraminiferal genera and their classification. Van Nostrand Reinhold Co., New York, 970 pp. Miller, A. A. L, D. B. Scott & F. S. Medioli. 1982. Elphidium excavatum (Terquem): ecophenotypic versus subspecific variation. J. Foraminiferal Research, 12:116-144. Murray, J. W. & S. S. Bowser. 2000. Mortality, protoplasm decay rate, and reliability of staining techniques to recognize "living" Foraminifera: a review. J. Foraminiferal Research, 30(l):66-70. Orbigny, A. D., D\ 1839. Foraminiferes in De la sagra, historie physique, politique et naturelle de file de Cuba. Paris, 224 pp. Parker, F. L. 1954. Distribution of the foraminifera in the northeastern Gulf of Mexico. Bull, of the Museum of Comparative Zoology, 111:453-588. Parker, F. L., F. B Phleger & J. F. Peirson. 1953. Ecology of foraminifera from San Antonio Bay and environs, southwest Texas. Cushman Foundation for Foraminiferal Research Special Publication No. 2:1-75. Pawlowski, J., I. Bolivar, J. Farhni & L. Zaninetti. 1995. DNA analysis of "Ammonia beccarii" morphotypes: one or more species? Marine Micropaleontology, 26:171-178. Phleger, F. B & F. L. Parker. 1951. Ecology of foraminifera, northwest Gulf of Mexico. Geol. Soc. of America, Memoir 46, part 1, 1-88; part 2, 1-64. Poag, C. W. 1978. Paired foraminiferal ecophenotypes in gulf coast estuaries: ecological and paleoecological implications. Transactions Gulf Coast Assoc, of Geol. Societies, 28:395-421. Poag, C. W. 1981. Ecologic atlas of benthic foraminifera of the Gulf of Mexico. Hutchinson Ross Publishing Company, 174 pp. Post, R. J. 1951. Foraminifera of the south Texas coast. Publications of the Institute of Marine Science, 2:165-176. Ride, W. D. L., C. W. Sabrosky, G. Bernardi & R. V. Melville. 1985. International code of zoological nomenclature. Univ. of Calif. Press, 338 pp. Rusnak, G. A. 1960. Sediments of Laguna Madre, Texas, in Recent Sediments, Northwest Gulf of Mexico. (F. P. Shepard et al., eds.) Amer. Assoc, of Petroleum Geol. Bull., Tulsa, 153-195. Schnitker, D. 1974. Ecotypic variation in Ammonia beccarii (Linne). J. Foraminiferal Research, 4:216-223. Sen Gupta, B. K., R. E. Turner & N. N. Rabalais. 1996. Seasonal oxygen depletion in continental-shelf water of Louisiana: historical record of benthic foraminifera. Geology, 24:227-230. Stouff, V., M. Lesourd & J.-P. Debenay. 1999a. Laboratory observations on asexual reproduction (schizogony) and ontogeny of Ammonia tepida with comments on the life cycle. J. Foraminiferal Research, 29(l):75-84. Stouff, V., E. Geslin, J.-P. Debenay & M. Lesourd. 1999b. Origin of morphological abnormalities in Ammonia (Foraminifera): studies in laboratory and natural environments. J. Foraminiferal Research, 29(2): 152-170. Terquem, M. O. 1875. Essai sur le classement des animaux qui vinent sur la plage et dans les environs de Dunkerque, 153 pp. Walton, W. R. 1952. Techniques for the recognition of living foraminifera. Contributions from the Cushman Foundation of Foraminiferal Research, 3:56-60. Walton, W. R. & B. J. Sloan. 1990. The genus Ammonia Brunnich, 1772: its geographic distribution and morphologic variability. J. Foraminiferal Research, 20(2): 128-156. Wantland, K. F. 1969. Distribution of modern brackish-water foraminifera in Trinity Bay, in Holocene Geology of the Galveston Bay area (R. R. Lankford & J. J. W. Rogers, eds.), Houston Geol. Society, 93-117. 16 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 1, 2002 Williams, H. F. L. 1995. Foraminiferal record of recent environmental change: Mad Island Lake, Texas. J. Foraminiferal Research, 25(2): 167-179. Yanko, V., J. Kronfeld & A. Flexer. 1994. Response of benthic foraminifera to various pollution sources; implications for pollution monitoring. J. Foraminiferal Research, 24(1): 1-17. Yanko, V., M. Ahmad & M. Kaminski. 1998. Morphological deformities of benthic foraminiferal tests in response to pollution by heavy metals: implications for pollution monitoring. J. Foraminiferal Research, 28(3): 177-200. Yuill, R. M. 1991. Paleoecological evidence of salinity changes in Galveston Bay. Galveston Bayline, Summer 1991, Galveston Bay National Estuary Program Newsletter, 9-11. PB-S at: pamela.stephens@mwsu.edu TEXAS J. SCI. 54(1): 17-26 FEBRUARY, 2002 ANALYSIS OF HORSE ( EQUUS) METAPODIALS FROM THE LATE PLEISTOCENE OF THE LOWER NUECES VALLEY, SOUTH TEXAS Jon A. Baskin and Antonia E. Mosqueda Department of Biology, Texas A&M University -Kingsville Kingsville, Texas 78363 Abstract.— Ninety-eight relatively complete metapodials (29 metacarpals and 69 metatarsals) of Equus were recovered from late Pleistocene terrace and valley fill deposits along the Nueces River in western Nueces and San Patricio counties, Texas. Sixteen measurements were taken on each metapodial. Three species of Equus were determined to be present using discriminant functions and bivariate and multivariate plots of the data. Equus cf. conversidens , the most abundant species, is a small- to average-sized horse with normal length metapodials. It is similar to members of the E. alaskae group. The second species, represented by 24 metapodials, is assigned to E. cf. scotti. These are larger horses with robust limbs that resemble members of the E. scotti and E. laurentius groups. The third, represented by six specimens, is a stilt-legged horse of the E. francisci group. The Wright Material Inc., sand and gravel pits along the Nueces River in western Nueces and San Patricio counties, Texas have produced a diverse assemblage of late Pleistocene fossils. Twenty-six species of mammals have been identified from here (Baskin 2000). Equids are among the most common fossils recovered. Living Equus includes horses, asses and zebras. Identification of fossil Equus to species from isolated teeth and bones is difficult at best (Winans 1989; Dal quest & Schultz 1992). Additionally, although most of the approximately 60 North American species that have been named are junior synonyms or invalid, the taxonomy of Equus itself is far from agreed on. Dalquest & Schultz (1992) identified seven or eight species of Equus from the Pleistocene (Irvingtonian and Rancholabrean) of northwestern Texas alone. Azzaroli (1998) recognized up to ten Pleistocene North American species. Winans (1989) recognized five species groups of North American Equus , of which no more than four groups were extant at a given time. Winans (1989) considered the possibility that each group represented a single species and that therefore only four species of Equus were present in North America during the Pleistocene. The purpose of this paper is to determine how many species of Equus were present in the late Pleistocene Nueces River Valley deposits of south Texas and identify them. Voucher specimens are deposited with the holdings of the Texas Memorial Museum (TMM) of the University of Texas, Austin. 18 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 1, 2002 Geologic Setting Four alluvial terrace units and three younger valley fill units are recognized from late Pleistocene and Holocene sediments in the lower Nueces River Valley, Nueces and San Patricio counties, west of Corpus Christi, Texas, between Odem and Mathis, where the Nueces River is entrenched in the late Pleistocene Beaumont Formation (Cornish & Baskin 1995). The valley fill units are included in the Cayamon Creek Alloformation. Most of the metapodials described in this paper come from channel fill and point bar sands and gravels of the Cayamon Creek allomember 1 at the Wright Materials, Inc. quarries (TMM localities 43059 and 43064), approximately 4 km north of Bluntzer, Nueces County, Texas. A log buried in this unit has been carbon dated at 13,230 ±110 YBP (Baskin 1991). Seven metatarsals were recovered across the river in the Angel ita Terrace, San Patricio County (TMM locality 18594). These late Quaternary terraces and valley fill deposits have produced a mixed assemblage of early Pliocene and Pleistocene fossil vertebrates. The Pliocene horses are reworked from older updip deposits, presumably of the upper Goliad Formation (Baskin 1991). Whether the Pleistocene vertebrates are all contemporaneous with the latest Pleistocene alluvium or are to some degree reworked cannot be easily determined. There is a wide variation in the nature of the preservation. Some of the bones and teeth are darkly stained and appear to be partly mineralized. Other specimens are quite fresh in appearance. The fossils consist mainly of isolated teeth and durable postcranial elements such as astragali, phalanges and metapodials that indicate that transportation and sorting of specimens has occurred (Hanson 1980). Some of the specimens are water worn, but most are not. The fact that there are more than twice as many complete metatarsals as metacarpals is further evidence of hydrodynamic sorting. The presence of jaws of Equus, Bison , Tap inis and Came lops and a mammoth skull and associ¬ ated partial skeleton suggests that some, if not most, of the Pleistocene specimens were not transported very far. Bison in the fauna is also indicative of a Rancholabrean (late Pleistocene) age. The presence of both Bison latifrons and B. antiquus may indicate some degree of mixing for the Rancholabrean fauna, because B. latifrons is sometimes con¬ sidered an early Rancholabrean species (Guthrie 1970). However, B. latifrons may have survived into the late Rancholabrean (Pinsof 1991; Wyckoff & Dal quest 1997). BASKIN & MOSQUEDA 19 Methods and Materials Metapodials are usually the most useful skeletal element available for identifying fossil Equus (Winans 1989). Skulls are rarely preserved and isolated teeth are highly variable. Ninety-eight complete metapodials (29 metacarpal s and 69 metatarsals) were analyzed. Sixteen measurements (Eisenmann 1979; Winans 1989) were taken on each (Tables 1,2). All measurements are in mm. Winans (1989: table 14.1) developed dis¬ criminant functions based on eight of these measurements to assign specimens to one of her five species groups: the E. simplicidens (early Blancan), E. scotti (late Blancan to early Rancholabrean) , E. laurentius (Rancholabrean), E. francisci (Irvingtonian to Rancholabrean), and E. alaskae (Irvingtonian to Rancholabrean) groups. The Nueces River metapodials were initially assigned to one of the five species groups using these discriminant functions. Because reliability of these discrimi¬ nant functions to correctly assign specimens to species varied from 61-91 % (Winans 1989), the measurements were analyzed graphically using bivariate and multivariate plots (Figs. 1, 2) of the data to look for groupings of specimens and to emend species assignments. For the principal components analyses (Fig. 2), measurements (Tables 1, 2) used are 1, 4, 5, 10, 12, 14 and 15 for the metacarpals, and 1, 4, 5, 6B, 7, 10, 11, and 15 for the metatarsals (Fig. 2). This was done to compare the Nueces River results with the average values Winans (1989) used to determine her discriminant functions. Material examined.— Equus cf. conversidens metacarpals TMM 43059-19 to -21, metatarsals TMM 43059-10 to -35 and TMM 18594-19 to -24; Equus cf. scotti metacarpals TMM 43059-22 to -32, metatarsals TMM 43059-36 to -59; Equus cf francisci metacarpals TMM 43059-33 to -35, metatarsals TMM 43059-61 to -62 and TMM 18594-25. Results The most common equid species from the Nueces River gravel pits is represented by 13 metacarpals and 42 metatarsals. These form relatively distinct clusters (Figs. 1, 2) centered near average values for the E. alaskae group of Winans (1989: table 14.2). They are assigned to Equus cf. conversidens, a small- to medium- sized horse with normally proportioned (i.e., stout-legged) metapodials. Winans (1989) assigned the small, stout-legged horses of Rancholabrean age to the Equus alaskae (originally E. niobrarensis alaskae) group. Azzaroli (1998) synonymized E. niobrarensis alaskae and E. laurentius with E. ferns Table 1. Univariate statistics on metacarpal Ill’s of Equus from the Nueces River Valley. Measurements are 1, greatest length [1, 1]; 2, lateral length [2,- ]; 3, mid-shaft width [3, 6]; 4, mid-shaft anteroposterior breadth [4, 7]; 5, proximal articular width [5, 2]; 6, proximal articular breadth [6, 3]; 7, width of magnum facet [7, 5]; 8, width of anterior unciform facet [8, -] ; 8’, width of posterior unciform facet [8’, -]; 9, width of trapezoid facet [9, -]; 10, distal supra-articular width [10, 8]; 11, distal articular width [11, 9]; 12, anteroposterior 20 o co 13 vo 18 a § C T3 3 loC £ >n oo i On c o .-w 5 o g "O ^ 03 1 2 1 JO THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 1, 2002 ON 3 NO ON q CO q q NO CM q q 00 q 4 NO* oo co q 4 4 4 4 oo o O Pi (M CM CO CM Tf CO q ON NO q CO CM q q •a o 4 4 ON ON ON o o q NO 4 q q 4 4 q r (N (N r-H 4 co o 00 co 4 NO oo q O NO 4 NO 53 CO CM ■^r co CO co co CM CM CM O CO CO 8 8 o CM co co co CO q co co s O 8 O q r- o r- 0°. co q § X NO CM in NO CM ON co 00 NO CM CM 4 oo’ 00 00 CO CM Tf CO CO M- Tf co CM CM CM CM CM R co co co co co CO CO CO co - co co co CO CO co r- 00 co CO q CM NO CM r- q 00 co q q o CO CO 4 4 CM >ri 4 q NO in co co 4 CO co in (N in M- CO q 00 o CM q o NO _ q q CM in CM q 00 q q ON 4 r- in 4 4 4 NO NO c* 1 CM q CO in q in On in m q CO CO CO $5 O CM r- CO q CM 4 ON 4 4 O in in CO NO o oo § CM CM 4 oo’ co in CO CO ON NO ON 00 4 CO o CO CM in co ^r '~H "Cf ^r CO co co CO NO in o co >n On m 00 NO Tf ■'t in O CM s r- NO f- o in NO co Tf- 1—1 s NO r- CM 3 Q 00 00 r- CM in ON 00 00 CM ON r- o q 00 or1 W < 1 CM 1 >— 1 1 *— 1 1 >— 1 — i +1 +! +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 x NO ^t CM CM ON ON Tj- >n m ON 00 CM I - . ON CM ON q NO q o CO o q q q OO q oo CO ON ON in 4 4 NO q 4 4 4 4 4 CO CM CM co CM m co >n m co CO CO R _ o o 00 _ *> q O N CM NO q q CO CO q OO q q 00 q o CM CM in 4 NO »ri NO 4 CM 4 o 4 oo NO oo 4 CM co NO q CM o q o q q CM oo q 00 q 00 co co CM 4 NO o 4 CM >n >ri 4 4 ON o ON § C4 CM CM CO CM m CO q — 1 On 4 q q CO CM CO CM £ O 4 CM 00 o CM co 4 4 CM NO o 00 6 CM in o NO q 4 CO ON NO «n CM 4 q 4 q 4 CM CO CM CM CM co CM co CM CM CM CM > r 8 00 NO ON 00 8 t-~ ON CO 00 r- NO On t"- r- CM r- r- Tf in o 00 1—1 CM r- 00 00 m <4-3 Q 00 o oo o ON q q q NO o CO o q oo CM O o in in 4 4 CM 4 CM 4 4 4 4 CO 4 4 4 4 s a +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 kl X _ q CO ON NO _ 00 m CM CO O 00 00 8 ON q i—l q CM q NO »n q oo CM CM i— I 4 in 00 CM 4 in 8 8 o NO 4 NO CM CM CO CM CO CO 1—1 co CM CM CM R CO CO co CO co CO CO _ 'Cf NO CO CO co co CO co CM CO ^f- m NO r- 00 00 ON o CM CO >n BASKIN & MOSQUEDA 'g "3 ON O — Is 2 g Tts g ov ON •n 00 co o CM CO 00 °° CM CM CO o in o CO CO 'St o co q CM ON 4 CO ON q ~ ~ O o ^ n in co o O co C- O O co o o NO ON rc i °° o 00 'St co q q o s X co o — a o n q co CM NO NO S o in 0.-3 r-n " c3 5 00 00 co co tj- 'cr 'St — H co co CM CM CM >. O co ON co CM in ON 00 * q ON in in r- aj p sg TJ co .« x g* a-“|-8 •g OR CO 1 in NO co i ON in rf co CO NO l n in i 00 n fl n q q 00 q NO q NO 'St CO CM CO 1 o CO ON -I"4I§ S3 a 8 6 i 6 s c2 S o - jj oo CM ON NO r- s CM ON in NO >n ON t" r- c (S- Q oo +1 +1 oo' +1 co +1 q CM +1 ^t +1 o ’St +1 CM +1 q CO +1 q CM +1 q +1 o "St +1 'St co +1 q CM +1 ON +1 CM +1 co +1 c/3 n CM 00 r-H -H 8 00 in O 8 *«.31 'S | NO o r- q CM co ON 'St q q q q l-c S S| ^ 1 Z £ jC "C o u p j8 Ja^E •5 g- - 2-0 2^ 00 00 »n ti- NO ON o CO oo CM —H ON — H 'St in !C (N in CM ■ct CO in CO m >n in ^a- in 'St 00 in in in in CO m CO in CO 'St CM 't (N CM CM CM CM CM CM CM CM CM CM CM CM CM CM s ^ g-g-^s - 3«f £1 S3 e, j3 sS T3 — q T3 m P -Sfi trl i Wh - rr\ 2 r- 00 r- q o q q in CO q ,-H q co O N co > o CO CO oo r- 00 00 00 CM CO 00 oo o 00 CM °'s *71 pp ,■ CO „ g, - ,_ • HH Tt -J- ‘ ’— 1 ^ <£ CM — r^; 2 21 ri - . CO o ON n o NO o in CO 'St q q q q q avo a H — Jl* a r- r^ n o Tf- NO 00 in o 00 00 00 CM 'St 'St «2 P _j CO g 04 (N CM co co in ,^ ^t 'St 'St 'St co co CO q s a a a | ss -S O NO CM rl o On 00 o ON q in co co o q CO q 13 g t« fl o O g gT3 - E o C «-h C (_i C co B co (N CO CM ^j- u-i 00 >n ON 'St 'St CO in CM CM CM > CM CM co CM CO co co CO CM CM CM c £ -o <£ £ «- 2 o $S ^ % o a . K 8 NO ON NO ON r- NO On ON NO 00 CM r- 00 CM CA Cd rH . G NO r- CO NO 'St CM Tt 'St o 00 'St .2-5—3 'tj Q in in CM n CO q q in q 'St CO O co q MS '-Bfe-2 -5 ■— vo ts G -s • a g g3 w 1/3 „4“ o 2 > ^ 1 ^ 3 g 's -<* s a to +1 ON +1 ON +1 CM +1 CM +1 co +1 co +1 co +1 co +1 +1 +1 co +1 co* +1 co +1 CM +1 CM +1 CM +1 t*3 X 00 r- CO 00 r- CM ON r- co CM ON t- 00 Tf On q °N 00 o q q q q 'St CO ON |-§ f £28 in in t-H o Tf 00 — H CM CM NO CM CM NO 00 ON ^ia a X) £ i — i ^ S < OQ E-1 CM CO M- in NO NO r- 00 ON o CM co 'St »n 21 22 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 1, 2002 60 Q 50 § < te 40 Q 30 210 220 230 240 250 260 METACARPAL 3 LENGTH 60 Q 30 - 1 - 1 — — i - 1 - ' - 1 - ' - 230 250 270 290 310 METATARSAL 3 LENGTH Figure 1 . Scatterplots of greatest length versus distal articular width for medial metacarpals and metatarsals of Equus. Circles are specimens from the Nueces River Valley assigned to E. cf. conversidens , to E. cf. scotti; inverted triangles, to E. cf. francisci; and stars, to E. sp. The letters P, S, L, F and A refer to the average values for these measurement for the E. simplicidens , E. scotti , E. laurentius, E. francisci and E. alaskae groups (Winans 1989), respectively. (=E. caballus - the extant domestic horse). He retained E. conversidens for the horse from San Josecito Cave and the smaller horse from Slaton (Dal quest & Hughes 1965). The dimensions of these metapodials (Tables 1, 2) are similar to those of Equus species A from the Irving- tonian Leisey Shell Pit of Florida. Hulbert (1995) noted that this taxon was intermediate in tooth row length between E. conversidens and E. scotti. However, metatarsals from Slaton assigned to E. conversidens are similar in length and distal width to the Leisey and Nueces River taxon. Length and distal width of metapodials of E. conversidens from San Josecito Cave are also within this range (Lundelius 1984). These metapodials have somewhat higher coefficients of variation (Tables 1, 2), even for length, when compared to samples that are BASKIN & MOSQUEDA 23 METACARPAL PRINCIPAL COMPONENT I METATARSAL PRINCIPAL COMPONENT I Figure 2. Principal components analysis using a covariance matrix on six measurements taken on medial metacarpals and metatarsals. See Fig. 1 for an explanation of the symbols and Tables 1 and 2 for measurements used. Principal component one accounted for 72% and 82% of the variance, and principal component 2, 24% and 14% for the metacarpals and metatarsals respectively. probably not as temporally mixed, such as Rancho La Brea (Willoughby 1948) or Leisey (Hulbert 1997). This may indicate that more than one species is represented or time averaging within this species has occurred. Removing the six smallest metatarsals results in CV’s that are compara¬ ble to the Rancho La Brea or Leisey samples. Although, these six are within the size range given by Winans for the E. alaskae (or even E. scotti ) group, they could represent early Pliocene Dinohippus or a smaller species of Equus (e.g., E. tau , although Dalquest [1979] assigned relatively elongate metatarsals to that species). Two of these specimens are very worn distally, three display average wear, and one has no evidence of transport. Mooser & Dalquest (1975) attributed metatarsals and dentitions of the smallest horses from Cedazo to E. conversidens . These are similar in size to these six smallest Nueces River specimens. Somewhat larger, but similarly proportioned meta¬ tarsals, were assigned to E. excelsus (Mooser & Dalquest 1975). Rodruiguez Avalos (1999) referred all horses from Cedazo to a single species. Dalquest (1979) referred metatarsals 240-265 mm in length to E. conversidens , which would also include E. excelsus from Cedazo. Howe (1970) reported high CV values for E. simplicidens from Broad¬ water, comparable to the Nueces River sample, and concluded that because measurements were normally distributed, only a single species was present. Likewise, the smallest Nueces metatarsals are retained in E. cf. conversidens . 24 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 1, 2002 Equus cf. scotti is represented by 1 1 metacarpals and 24 metatarsals that are for the most part intermediate in size between the mean measurements determined by Winans (1989) for her E. scotti and E. laurentius groups. These are large horses with normally proportioned to robust limbs. Measurements (Tables 1, 2) are similar to those for E. occidentalis from Rancho La Brea (Willoughby 1948). The discriminant functions of Winans (1989) assigned the smaller of these individuals to the E. scotti group, the larger to the E. laurentius group. Four of the five metacarpals assigned to the E. scotti group are relatively short and stout. Winans (1989) used the E. scotti group for late Blancan to early Rancholabrean large, stout-legged horses and included E. hatched and E. niobrarensis in this group. Winans used the E. laurentius group for later Rancholabrean large horses which are similar to E. scotti. Equus laurentius is probably based on a recent specimen (Winans 1989; Azzaroli 1998) and is therefore an invalid name. Winans (1989) referred E. occidentalis to the E. laurentius group. Scott (1998) stated that E. scotti was replaced by E. occidentalis in the late Pleistocene of California. The most important previous collections of Rancholabrean Equus from South Texas are from Ingleside (Lundelius 1972), Berclair terraces (Quinn 1957), and Cueva Quebrada (Lundelius 1984). The Ingleside sample, which does not include any metapodials, was originally referred to three species: most of the specimens to E. complicatus, a few larger specimens to E. pacificus and a few smaller to E. fratemus. Winans (1989) referred the entire sample to the E. scotti group. Two species were described from Cueva Quebrada: a stouter E. cf. scotti and a slenderer E. francisci. Winans (1989) referred the stout-limbed species to the E. laurentius group; the stilt-limbed is not discussed, but belongs to the E. francisci group. The stout-legged metapodials are similar to those from the Nueces River Valley. Azzaroli (1998) referred the large stout-limbed horses from the late Pleistocene of South Texas to E. excelsus, which he considered a senior synonym of E. scotti. He differentiated E. excelsus from E. occidentalis on the basis of dentition, but suggested the two were closely related. Dal quest & Schultz (1992) stated that E. excelsus was a medium-sized horse and that the larger caballine horses of the late Pleistocene may have been E. scotti. The third species, represented by six specimens, is a stilt-legged ass of the E. francisci group. The metapodials are similar in size to specimens assigned to E. francisci (e.g., Lundelius & Stevens 1970; BASKIN & MOSQUEDA 25 Lundelius 1984). The type of E. francisci is from the Lissie Forma¬ tion, Wharton County, Texas (Lundelius & Stevens 1970). Dalquest and Schultz (1992) recognized three or four species of stilt-legged horses, including E. pseudaltidens . Equus pseudaltidens (Hulbert 1995) was described from the late Pleistocene Berclair terrace, Bee County, Texas as Onager altidens Quinn (1957). The reported metatarsal length is 283 mm, similar in size to specimens referred to E. francisci. Based on dental characteristics, Hulbert considered E. pseudaltidens to be distinct from E. francisci. Acknowledgments Dedicated to the memory of W. W. Dalquest. I am very grateful to the management of Wright Materials, Inc., particularly M. Truesdale and L. and R. Wright, for permission to collect on their property. A. Mosqueda was funded by the Ronald E. McNair Scholars Program. Without the dedication of R. Thomas of TAMUK, who collected most of the metapodials, this paper would not have been possible. The North American Mammalian Paleofaunal Database (Alroy 2000) was consulted for background information. Constructive comments were offered by Richard Hulbert and Ernest Lundelius. This study was funded in part by Texas A&M University-Kingsville. Literature Cited Alroy, J. 2000. North American fossil mammal systematics database. http://www.nceas.ucsb.edu/ ~ alroy/nafmsd.html Azzaroli, A. 1998. The genus Equus in North America - The Pleistocene species. Pal. Italica, 85:1-60. Baskin, J. A. 1991. Early Pliocene horses from late Pleistocene fluvial deposits, Gulf Coastal Plain, South Texas. J. Pal., 65(6): 995 -1006. Baskin, J. A. 2000. The Pleistocene fauna of South Texas. http : //users . tamuk . edu/kfjab02/SOTXF AUN . htm Cornish, F. G. & J. A. Baskin. 1995. Late Quaternary sedimentation, lower Nueces River, South Texas. Texas J. Sci., 47(3): 191-202. Dalquest, W. W. 1979. The little horses (genus Equus) of the Pleistocene of North America. Amer. Mid. Nat., 101(l):241-244. Dalquest, W. W., & J. T. Hughes. 1965. The Pleistocene horse Equus conversidens. Am. Mid. Nat., 74(2):408-417. Dalquest, W. W., & G. E. Schultz. 1992. Ice age mammals of northwestern Texas. Midwestern St. Univ. Press, Wichita Falls, 309 pp. Eisenmann, V. 1979. Les metapodes d 'Equus sensu lato (Mammalia, Perissodactyla). Geobios, 12(6): 863-886. Eisenmann, V. 1986. Comparative osteology of modern and fossil horses, half-asses, and asses. Pp. 67-116, in Equids in the ancient world (R. H. Meadow & H.-P. Uerpmann, eds.), Dr. Ludwig Reichert Verlag, Wiesbaden, 421 pp. 26 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 1, 2002 Guthrie, R. D. 1970. Bison evolution and zoogeography in North America during the Pleistocene. Quart. Rev. Biol., 45(1): 1-15. Hanson, C. B. 1980. Fluvial taphonomic processes: Models and experiments. Pp. 156-181, in Fossils in the making (A. K. Behrensmeyer & A. P. Hill, eds.), Univ. Chicago Press, Chicago, Illinois, 338 pp. Howe, J. A. 1970. The range of variation in Equus (Plesippus) simplicidens Cope from the Broadwater quarries of Nebraska. J. Pal., 44(5):958-968. Hulbert, R. C. 1995. Equus from the Leisey Shell Pit 1A and other Irvingtonian localities from Florida. Bull. Florida Mus. Nat. Hist., 37(17):553-602. Lundelius, E. L. 1972. Fossil vertebrates from the late Pleistocene Ingleside Fauna, San Patricio County, Texas. Univ. Texas Bur. Econ. Geol. Rept. Invest., 77:1-74. Lundelius, E. L. 1984. A late Pleistocene mammalian fauna from Cueva Quebrada, Val Verde County, Texas. Pp. 456-481, in Contributions in Quaternary vertebrate paleontology: a volume in memorial to John E. Guilday (H. H. Genoways & M. R. Dawson, eds.), Carnegie Mus. Nat. Hist., Special Pub., 8, 538 pp. Lundelius, E. L. & M. S. Stevens. 1970. Equus francisci Hay, a small stilt-legged horse, middle Pleistocene of Texas. J. Pal., 44(1): 148-153. Mooser, O. & W. W. Dalquest. 1975. Pleistocene mammals from Aguascalientes , Central Mexico. J. Mammal., 56(4): 78 1-820. Pinsof, J. D. 1991. A cranium of Bison alaskensis (Mammalia: Artiodactyla: Bovidae) and comments on fossil Bison in the American Falls area, southeastern Idaho. J. Vertebrate Pal., 1 1(4):509-514. Quinn, J. H. 1957. Pleistocene Equidae of Texas. Univ. Texas Bur. Econ. Geol. Rept. Invest., 33:1-51. Rodriguez Avalos, J. 1999. Population structure and biological implications of Equus conversidens, Cedazo Local Fauna (Pleistocene), Aguascalientes, Mexico. J. Vert. Paleontol., 19(3):71A. Scott, E. 1998. Equus scotti from southern California. J. Vert. Paleontol., 18(3):76A Willoughby, D. P. 1948. A statistical study of the metapodials of Equus occidentalis Leidy. Bull. Southern California Acad. Sci., 47(3):84-94. Winans, M. C. 1989. A quantitative study of the North American fossil species of the genus Equus. Pp. 262-297, in The evolution of perissodactyls (D. R. Prothero & R. M. Schoch, eds.), Oxford Monographs Geol. Geophysics, no. 15, 537 pp. Wyckoff, D. G. & W. W. Dalquest. 1997. From whence they came: the paleontology of southern plains bison. Plains Anthropologist, 42(l):5-32. JAB at: J-Baskin@tamuk.edu TEXAS J. SCI. 54(l):27-36 FEBRUARY, 2002 SILICA-SCALED CHRYSOPHYTES AND SYNUROPHYTES FROM EAST TEXAS Daniel E. Wujek, James L. Wee and James E. Van Kley Department of Biology, Central Michigan University Mt. Pleasant, Michigan 48859 Department of Biological Sciences, Loyola University 6363 St. Charles St., New Orleans, Louisiana 70118 and Department of Biology, Stephen F. Austin State University Nacogdoches, Texas 75962 Abstract.— A total of 27 scale-bearing species of the algal classes Chrysophyceae and Synurophyceae, referred to herein as scaled chry sophytes , were recorded in 35 water bodies from 11 eastern Texas counties using transmission electron microscopy. These were distri¬ buted between the Chrysophyceae (one Chrysosphaerella sp., one Paraphysomonas sp. and two Spiniferomonas sp.) and Synurophyceae (14 Mallomonas sp. and nine Synura sp.). The number of taxa per collection varied from zero to six. Twenty-three taxa are new records for Texas. Mallomonas multisetigera is reported for the first time from North America. Scales of the colorless free-living flagellate Gyromitus disomatus, an organism of uncertain taxonomic affinity, were also observed. Surveys of the freshwater algal flora of Texas were initiated through a series of investigations by H. C. Bold and his students (Deason & Bold 1960; Chantanachat & Bold 1962; Bischoff & Bold 1963; Brown & Bold 1964; Cox & Bold 1966; Smith & Bold 1966; Groover & Bold 1969; Archibold & Bold 1970; Baker & Bold 1970. Other studies have included Texas collections, but were not directed specifically on the Texas flora (Flint 1955; Nicholls 1964; Hoffman 1967; Ott 1976; Carty 1989; Carty & Cox 1985; 1986; Sheath et al. 1993a; 1993b; Swamikan- nu & Hoagland 1990; Vis & Sheath 1996). All of these reports are based on light microscopy. Silica-scaled chrysophytes are taxonomically placed in the division Chry sophy ta, classes Chrysophyceae and Synurophyceae, based on ultrastructural and biochemical characteristics (Andersen 1987). The majority of chry sophy te genera lack a covering of siliceous scales and exhibit a great plasticity in regard to both morphology and nutrition (Hoek et al. 1995). Most of the common chrysophytes are flagellated, occurring as single cells or are colonial. The first electron micrographs of silica-scaled chrysophytes from Texas are in the unpublished report of Marquis (1977) based on col- 28 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 1, 2002 lections made in the Edwards Plateau located at the southern most end of the High Plains province (Cole 1966). The chrysophyte flora of neighboring Louisiana (Wee et al. 1993), the nearby state of Arizona (Gretz et al. 1979; 1983; 1985), and the country of Mexico (Kristiansen & Tong 1995) also have been studied previously using electron micros¬ copy. Adjacent states, such as New Mexico and Oklahoma remain to be investigated. In this study, the silica-scaled algal flora from 11 counties represent¬ ing 35 eastern Texas locations in the western Coastal Plain physio¬ graphic province (Cole 1966) were examined using transmission electron microscopy (TEM). Materials and Methods Phytoplankton samples were collected with a plankton net (10/xm mesh size) in mid-March of 1996 from 35 ponds and lakes in 11 Texas counties (Table 1). Samples were fixed in acid Lugol’s (Wee 1983). For TEM, subsamples were placed on Formvar-coated, carbon- stabilized grids, air dried and examined with a Philips 300 transmission electron microscope. All identifications were based on TEM. Percentages were based on the number of samples in which a taxon was observed divided by the total number of samples. Physical/chemical parameters taken in the field were surface water temperature, pH (Markson model 85), and specific conductance (Oakton WD-60). In an attempt to identify coordi¬ nated variation between physical /chemical parameters and species composition of the samples, the data were subject to multivariate analyses, including Detrended Correspondence Analysis (DC A, Hill 1979) and Canonical Correspondence Analysis, (CCA, ter Braak 1992). Species occurring in only one sample were eliminated prior to analysis, resulting in a data set of 17 samples and 31 taxa. All micrographs documenting this study are on deposit in the herbarium of Central Michigan University. Results and Discussion The taxa identified, and the collection sites for each taxon are listed in Table 2. Twenty-seven silica-scaled synurophycean and chryso- phycean taxa from five genera, Mallomonas (14 taxa), Synura (10 taxa), Spiniferomonas (two taxa) , Paraphysomonas and Chrysosphaerella (one taxon each) were observed from the 35 samples. All scale morphologies were similar to those in the published literature. WUJEK, WEE & VAN KLEY 29 Table 1 . Eastern Texas plankton collection sites containing silica-scaled chrysophytes and synurophytes, plus physical/chemical data and number of taxa observed for each site, 17 March to 20 March, 1996. Sample No. Location pH Temperature °C Conductivity fiS/M #taxa obs. Shelbv County - 17 March, 1996 1 Toledo Bend Reservoir 6.8 24.3 371.0 3 Just southwest of County 139 bridge across arm of the reservoir 2 William Roberts Pond 6.7 21.2 117.6 2 At Shelbyville - spring fed 3 Center City Reservoir 7.1 23.8 117.2 1 xh mile east of US 96, 3 miles south of Center 4 Pinkston Reservoir 7.7 22.8 105.8 1 Near Aiken, southwest corner of county Nacogdoches County - 17 March, 1996 5 Sam Rayburn Reservoir 6.8 21.9 150.0 1 Jasper County - 17 March, 1996 6 BA Steinhagen Lake 6.5 22.3 147.0 2 Martin Dies Jr. State Park, south of US 190 Nacogdoches County - 17 March, ] [996 7 Lake Nacogdoches 6.5 13.8 103.9 4 Rusk County - 17 March, 1996 8 Lake Striker 6.7 16.5 279.0 1 9 Craig’s Pond 6.7 18.3 92.2 4 10 Willow Lake 6.6 14.4 157.5 1 In Henderson 11 Martin Lake 6.7 17.2 172.9 3 Martin Lake State Park Panola County - 17 March, 1996 12 Lake Murvaul 6.7 16.9 190.3 5 Rusk County - 19 March, 1996 13 Doc Young Pond 6.6 14.8 89.8 6 Impoundment, about 2 + miles west of Tabem on north side of Texas Hwy 43 (P4 miles east of Road 1716) 14 Pond 6.7 18.8 73.0 3 Northeast side of Road 1716, IV2 miles northwest of Texas Hwy 43 15 Lake Cherokee 6.7 16.7 92.6 6 Small arm crossing Farm Road 1716 16 Long Glade Lake 6.7 19.6 60.2 1 Boat launch, west side of Road 2127, 2 miles south of Lake Cherokee 17 J.W. Walters Pond 6.7 16.4 86.9 2 East side of Road 2127, 2 miles north of Road 1727 18 Lake Forest Park 6.6 16.3 103.1 2 In west Hendersonville, south side of Texas Hwy 64 Smith County - 19 March, 1996 19 Pleasure Acres Lake 6.7 18.3 101.3 1 Lake at subdivision, north of Texas 64, northwest of New Chapel Hill 30 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 1, 2002 Table 1 cont. Sample pH Temperature Conductivity #taxa No. Location °C /liS/M obs. 20 Pond 7.8 West of Pleasure Acres Lake entry road 16.6 128.7 1 21 Lake Tyler 6.7 Near southeast side of Farm Road 848 18.2 95.3 4 Henderson County - 19 March, 1996 22 Lake Palestine 6.5 15.0 176.9 0 West side of main lake, southeast side of Texas 155 Cherokee County - 19 March, 1996 23 Lake Jacksonville 6.5 15.9 84.7 2 Northwest part just east of Farm Road 747 Nacogdoches County - 19 March, 1996 24 Whisper Oaks Pond 8.3 16.5 88.1 Harrison County - 20 March, 1996 25 Lake 7.0 13.5 396.0 On south side of access road that parallels I 20, just east of US 259, exit at Lakeport 26 Privately owned lake 6.8 13.8 72.1 Between I 20 and US 80 east of Longview, west of Ring Road [281] just south of County 3417 27 Highway Lake 6.9 17.4 113.1 In small subdivision, east of Longview, south side of Highway 80, accessed via County 3427, composite sample from both sides of dam between 2 biggest ponds 28 Big Rock Lake 6.9 13.5 47.4 East of Farm Road 450, east of Longview, between US 80 and Farm Road 449 Marion County - 20 March, 1996 29 Lake of the Pines 6.5 14.2 134.1 At Island View boat ramp, south side of lake Upshur County - 20 March, 1996 30 Barton Lake 6.5 16.4 97.7 East of Gilmer on south side of Texas 154 bridge 31 Beaver Pond 6.8 16.1 136.0 South of Gilmore, north side Bluebird Road between US 271 and Texas 300 32 South Twin Lake 6.4 15.7 77.1 33 North Twin Lake 6.6 13.6 88.1 34 Spencer’s Pond 7.5 West side of US 217, south of Gilmore, 12.0 619.0 1 mile north of Eagle and Evergreen Roads 35 Big Sandy Lake 6.6 13.9 93.3 1 2 3 3 4 1 2 1 6 2 1 WUJEK, WEE & VAN KLEY 31 Table 2. Species of silica-scaled chrysophytes and synurophytes from eastern Texas. See Table 1 for description of locations. Taxa indicated with an asterisk (*) are new reports for Texas; double asterisks (**) are new for North America. Taxon Collection Locations Chrysophyceae Chrysosphaerella *C. coronacircumspina Wujek & Kristiansen 11 Paraphysomonas *P. vestita (Stokes) de Saedeleer 6, 8, 11, 34 Spiniferomonas *5'. crucigera Takahashi 15 *S. trioralis Takahashi 1, 5, 11, 12, 15, 18, 21, 23, 34 Synurophyceae Mallomonas *M. akrokomos Ruttner in Pascher 7, 15 *M. annulata (Bradley) Harris 31 M. caudata Ivanov em. Krieger 7, 26, 27, 28, 30, 32 *M. crassisquama (Asmund) Fott 7, 12, 23, 32 M. doignonii Bourrelly em. Nicholls 16 *M. elongata Reverdin 10 *M. hamata Asmund 2, 17, 31, 32 *M. heterospina Lund 12 M. mangofera Harris & Bradley 25, 32 *M. papillosa Harris & Bradley 18 **A/. multisetigera Diirrschmidt 9 *M. portae-ferreae Peterfi & Asmund 4, 12 M. transsylvanica Peterfi & Momeu 3, 35 *M. tonsurata Teiling 21, 25, 26, 28, 29, 32, 33 Synura *S. australiensis Playfair em. Croome & Tyler 13, 14, 27, 28, 32 *S. curtispina (Petersen & Hansen) Asmund 24, 32 *S. echinulata Korshikov 1, 2, 13, 19, 20, 21, 35 *S. mollispina Korshikov 13 *S. petersenii f. petersenii Korshikov 1, 3, 7, 12, 13, 14, 29 *S. petersenii f. glabra (Korshikov) Siver 1, 5, 15 *S. sphagnicola Korshikov 13 *S. spinosa f. spinosa Korshikov 15, 21, 29 *S. uvella Stein em. Korshikov 1, 13, 14, 15, 17, 26, 27, 28, 29, 30, 32, 33 Classis insertae *Gyromitus disomatus Skuja 21 The number of scale-bearing chrysophyte taxa observed per sample varied from zero to six (Table 1). Species richness was greatest at three sites, Doc Young Pond, Lake Cherokee and South Twin Lake, where some of the lowest water temperatures, ranging from 14.8-15.7°C, were observed (Table 1). The most frequent Mallomonas species were M. tonsurata (20%), M. caudata (17%) and M. crassisquama and M. hamata (11% each) . Common species from other genera were Synura uvella (34%), S. echinulata and S. petersenii f. petersenii (both 20%), 32 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 1, 2002 S. australiensis (14%) and Spiniferomonas trioralis (23%). Species observed from only one collection site included Mallomonas annulata , M. doignonii , M. elongata , M. heterospina , M. mangofera , M. papil- losa, Synura curtispina, S. mollispina, Spiniferomonas crucigera and Chrysosphaerella coronacircumspina. Additionally, Mallomonas multisetigera is newly reported for North America (Fig. la). Scaled chrysophytes were observed in every collection except Lake Palestine (sample 22). The reason(s) for this is unknown. In the scanning electron microscope examination of samples from the Edward’s plateau through central and northeastern Texas, Marquis (1977) observed nine Mallomonas taxa. This current study observed four of Marquis’ (1977) taxa: Mallomonas caudata , M. doignonii , M. mangofera (as M. texensis) and M. transsylvanica. Mallomonas asmundiae , M. corymbosa , M. lychenensis, M. pseudocoronata and M. teilingii var. papillosa nomen nudum were not observed during this current study. Scales of Gyromitus disomatus Skuja (Fig. lb), a colorless free-living flagellate of unknown taxonomic affinity, were observed in the Lake Tyler sample. This organism has no obvious affinities with any taxonomic group (Swale & Belcher 1974). Nicholls (1979), using X-ray emission spectra, has shown that the scales are composed of silica, but not calcified, and hence do not represent coccoliths. Specific conductance across all collections ranged from 47.4 to 619.0 /xS/M. For the ten most frequently observed species, Synura echinulata and Mallomonas crassisquama had the narrowest ranges, 89.8 to 128.7 and 49.4 to 109.3 /xS/M, respectively. The largest ranges were ob¬ served in Synura uvella (47.4 - 371.0 pS/M) and Spiniferomonas trioralis (84.7 - 371.0 /xS/M). These are very close to the ranges reported by Wujek & Menapace (1998). As Siver (1993) reported, until more conductivity studies are published, it is unknown "at this time whether individual taxa are responding to specific anions and cations or some combination thereof. " The range in water temperatures (12.0 to 24.3 °C) and the time of year may indicate that the collections contained elements of both late spring and summer floras. Species observed, such as Mallomonas akrokomos and M. transsylvanica , have been observed under the ice in more northern regions (Cronberg & Kristiansen 1980; Siver 1991). WUJEK, WEE & VAN KLEY 33 Figure 1. (a) Scale from Mallomonas multisetigera, a newly reported silica-scaled chrysophyte for North America, (b) Scales from Gyromitus disomatus, a flagellate of unknown taxonomic affinity. Scale bar = 1 /xm. Taxa such as M. crassisquama , M. tonsurata and Synura curtispina , are more commonly observed during the summer (Siver 1991), supporting the hypothesis that collections made during this current study contain some taxa common in summer and others more common in the spring or winter. With additional sampling, especially in colder waters, given that collections were taken during mid March when the surface water had begun to warm (Table 1), it is suspected that additional warm water species and cooler water taxa known to occur in the southeastern U.S. will be found to occur in Texas. The pH of the collections ranged from 6.4 to 8.3. Many of the species observed in localities with low pH values, including Mallomonas hamata, M. transsylvanica , Synura echinulata and S. sphagnicola , have been reported previously as common in acidic habitats (Siver 1988; 1989; 1991; Wujek & Menapace 1998). Observations made during this study clearly support these earlier findings. Synura petersenii f. peters enii occurred in the largest pH range, 6.5 to 7. 1 , for the most prevalent taxa followed by S. australiensis , S. uvella and Mallomonas caudata , 6.4 to 6.9 and then M. tonsurata with 6.6 to 7.0. Both Mallomonas crassisquama and Synura echinulata had the narrowest ranges (Table 1). The data for all nine species are within the values reported in the literature. Multivariate analysis did not demonstrate clear relationships between the measured physical/chemical parameters and the scaled chrysophyte 34 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 1, 2002 and synurophyte species composition of the samples. The first, second and third axes of a DC A ordination of the samples, which was based on the presence of scaled chry sophy te and synurophyte taxa, were not significantly correlated with any of the measured parameters. Likewise, a Canonical Correspondence Analysis showed low species-environment correlations. A Monte Carlo test of the CCA model using 100 random iterations failed to reject the hypothesis that no relationship existed between the species-samples matrix and the matrix of physical/chemical parameters. Several explanations exist for the apparent lack of correspondence between the species composition of the samples and the environmental parameters. First, the relatively small number of samples taken at each lake may have resulted in taxa being missed. The sparse species- samples matrix and the large number of samples with only two or three taxa may have skewed ordination results. Additionally, collecting data that included an abundance measurement for taxa might result in more sensitive ordinations than the current data which were based on species presence or absence. Moreover, sampling during the period of turnover from the winter flora to the summer flora may have obscured species- environment patterns that would otherwise exist. It also is possible that the limiting environmental factor (s) structuring the scaled chry sophy te and synurophyte community were not measured or that scaled chryso- phytes and synurophytes are not strong indicators of environment in eastern Texas lakes within the measured ranges of the environmental parameters. In conclusion, as has been demonstrated in other regions of the U.S., Texas contains a diverse flora of scaled chry sophy tes. This investigation is by no means exhaustive, and it is believed that further collections and observations from eastern Texas and other Texas physiographic pro¬ vinces will yield additional species and possibly clarify species- environment relationships. Including this paper, the silica-scaled chry sophy tes known from Texas, based on electron microscopy, now comprise 33 taxa. Acknowledgments The authors wish to thank M. Wujek for all aspects of the field collections, Alexandra Van Kley for her assistance in some of the field collections, P. Eisner for grid coating, K. Jeisel for some of the TEM observations, and CMU MultiMedia Production in helping with the WUJEK, WEE & VAN KLEY 35 preparation of the illustrations. The senior author thanks the FRCE Committee of Central Michigan University for partial funding of this study. We thank Sara Marquis Burgin for making available her unpublished observations. Literature Cited Andersen, R. A. 1987. Synurophyceae Classis Nov., a new class of algae. Amer. J. Bot., 74(3): 337-353. Archibald, P. A. & H. C. Bold. 1970. Phycological Studies XI. The genus Chlorococcum Meneghini. Univ. Texas Publ. 7015, 115 pp. Baker, A. F. & H. C. Bold. 1970. Phycological Studies X. Taxonomic studies in the Oscillatoriaceae. Univ. Texas Publ. 7004, 105 pp. Bischoff, H. W. & H. C. Bold. 1963. Phycological Studies IV. Some soil algae from Enchanted Rock and related algal species. Univ. Texas Publ. 6318, 95 pp. Brown, Jr., R. M. & H. C. Bold. 1964. Phycological Studies V. Comparative studies of the algal genera Tetracystis and Chlorococcum. Univ. Texas Publ. 6417, 213 pp. Carty, S. 1989. Thompsodinium and two species of Peridinopsis (Dinophyceae): taxonomic notes based on scanning electron micrographs. Trans. Am. Micros. Soc., 108(l):64-73. Carty, S. & E. R. Cox. 1985. Observations on Lophodinium polylophum (Dinophyceae). J. Phycol., 2 1(3): 396-401. Carty, S. & E. R. Cox. 1986. Kansodinium gen. nov. and Durinskia gen. nov.: two genera of freshwater dinoflagellates (Pyrrhophyta). Phycologia, 25(2): 197-204. Chantanachat, S. & H. C. Bold. 1962. Phycological Studies II. Some algae from arid soils. Univ. Texas Publ. 6218, 75 pp. Cole, G. A. 1966. The American Southwest and Middle America. Pp. 393-434, in Limnology in North America (D.G. Frey, ed.), Univ. Wisconsin Press, Madison, Wisconsin, 734 pp. Cox, E. R. & H. C. Bold. 1966. Phycological Studies VII. Investigation of Stigeoclonium. Univ. Texas Publ. 6618, 167 pp. Cronberg, G. & J. Kristiansen. 1980. Synuraceae and other Chrysophyceae from central Smaland, Sweden. Bot. Notiser, 1 33(4) :595-6 1 8 . Deason, T. R. & H. C. Bold. 1960. Phycological Studies I. Exploratory studies of Texas soil algae. Univ. Texas Publ. 6022, 72 pp. Flint, L. H. 1955. Hildenbrandia in America. Phytomorphology, 5(3): 185-189. Hoek, van den C., D. G. Mann & H. M. Jahns. 1995. Algae, an introduction to phycology. Cambridge University Press, N.Y., 623 pp. Gretz, M. R., M. R. Sommerfeld & D. E. Wujek. 1979. Scaled Chrysophyceae of Arizona: A preliminary Survey. J. Arizona-Nevada Acad. Sci., 14(3):75-80. Gretz, M. R., D. E. Wujek & M. R. Sommerfeld. 1983. Scaled Chrysophyceae of Arizona: Further additions to the aquatic flora. J. Arizona-Nevada Acad. Sci., 18(1): 17-21 . Gretz, M. R., M. R. Sommerfeld & D. E. Wujek. 1985. Light and electron microscopical observations of Mallomonas portae-ferreae var. reticulata var. nov. (Chrysophyceae). Phycologia, 24(4) : 478-48 1 . Groover, R. D. & H. C. Bold. 1969. Phycological Studies VIII. The taxonomy and comparative physiology of the Chlorosarcinales and certain other edaphic algae. Univ. Texas Publ. 6907, 165 pp. Hill, M. O. 1979. DECORANA - a FORTRAN program for detrended correspondence analysis and reciprocal averaging. Cornell ecology programs, Department of Ecology and 36 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 1, 2002 systematics, Cornell university, Ithaca, New York. Hoffman, L. R. 1967. Four species of Oedogonium. Can. J. Bot., 45(4):405-412. Kristiansen, J. & D. Tong. 1995. A contribution to the knowledge of the silica-scaled chrysophytes in Mexico. Algol. Stud., 77(1-6): 106. Marquis, S. A. 1977. Taxonomic investigations of Mallomonas. Unpublished M.S. thesis, Texas A&M, 57 pp. Nicholls, H. W. 1964. Culture and developmental morphology of Compsopogon coeruleus. Amer. J. Bot., 51(2): 180-188. Nicholls, K. H. 1979. Is Hymenomonas prenanti Lecal (Prymnesiophyceae) really the colourless flagellate Gyromitus disomatus Skuja? Phycologia, 18 (4): 420-423. Ott, F. D. 1976. Further observations on the freshwater alga Flintiella sanguinaria Ott in Bourrelly 1970 (Rhodophycophyta, Porphyridales). Arch. Protistenk., 1 18(1 /2) : 34-52. Sheath, R. G., D. Kaczmarczyk & K.M. Cole. 1993a. Distribution and systematics of freshwater Hildenbrandia (Rhodophyta, Hildenbrandiales) in North America. Eur. J. Phycol., 28(2): 115-121. Sheath R. G., M. L. Vis & K. M. Cole. 1993b. Distribution of Batrachospermum (Batrachospermales, Rhodophyta) in North America. 3. Section Setacea. J. Phycol., 29(5) :7 19-725. Smith, R. L. & H. C. Bold. 1966. Phycological Studies VI. Investigations of the algal genera Eremosphaera and Oocystis. Univ. Texas Publ. 6612, 121 pp. Swale, E. M. F. & J. H. Belcher. 1974. Gyromitus disomatus Skuja-a free-living colourless flagellate. Arch. Protistenk., 1 1 6(3) :21 1-220. Siver, P. A. 1988. Distribution of scaled chrysophytes in 17 Adirondack (New York) lakes with special reference to pH. Can. J. Bot., 66(7): 1391-1403. Siver, P. A. 1989. The distribution of scaled chrysophytes along a pH gradient. Can. J. Bot., 67(7):2120-2130. Siver, P. A. 1991. The biology of Mallomonas : morphology, taxonomy and ecology. Kluwer, The Netherlands, 230 pp. Siver P. A. 1993. Inferring the specific conductivity of lake water with scaled chrysophytes. Limnol. Oceangr., 38(7): 1480-1492. Swamikannu, X. & K. D. Hoagland. 1990. Effects of snail grazing on the diversity and structure of a periphyton community in a eutrophic pond. Can. J. Fish. Aquat. Sci., 46(10): 1698-1704. Ter Braak, C. J. 1992. CANOCO - a FORTRAN program for canonical community ordination. Microcomputer Power, Ithaca New York. Vis, M. L. & R. G. Sheath. 1996. Distribution and systematics of Batrachospermum (Batrachospermales, Rhodophyta) in North America. 9. Section Batrachospermum : description of five new species. Phycologia, 35(2): 124-134. Wee, J. L. 1983. Specimen collection and preparation for critical light microscope examination of Synuraceae (Chrysophyceae). Trans. Am. Micro. Soc., 102(l):68-76. Wee, J. L., D. J. Booth & M. A. Bossier. 1993. Synurophyceae from the southern Atlantic Coastal Plain of North America: a preliminary survey in Louisiana. Nord. J. Bot., 13(1):95-106. Wujek, D. E. & F. J. Menapace. 1998. Silica-scaled chrysophytes from Alabama. J. Alabama Acad. Sci., 69(l):33-43. JEVK at: jvankley@sfasu.edu TEXAS J. SCI. 54(l):37-44 FEBRUARY, 2002 ENZYMATIC VARIATION IN THE LAND SNAIL EUGLAND1NA TEXASIANA (GASTROPODA: PULMONATA) FROM SOUTH TEXAS AND NORTHEASTERN MEXICO Kathryn E. Perez* and Ned E. Strenth Department of Biology, Angelo State University San Angelo, Texas 76909 * Current Address: Department of Biological Sciences University of Alabama, P. O. Box 870345 Tuscaloosa, Alabama 35487 Abstract.— Enzymatic variation in four specimens of the land snail Euglandina texasiana (Pfeiffer) from south Texas and northeastern Mexico (150 km distant) was examined using cellulose acetate gel electrophoresis. Ten of the 15 loci examined were found to be monomorphic for all specimens. Considerable variation was observed to occur in the remaining five loci. A computer analysis of the resulting enzymatic variation revealed that specimens from these two locations were 94.5% genetically similar. A single specimen of Euglandina singleyana (Binney) from New Braunfels in central Texas was found to be 47.6% similar to specimens of Euglandina texasiana. Resumen.— La variacidn enzimdtica en cuatro especfmenes del caracol terrestre Euglandina texasiana (Pfeiffer) del sur de Texas y del nordeste de Mexico (a 150 km distante) fue examinada usando electrofdresis de gel de acetato celuloso. Se encontrd que diez de los 15 lugares examinados son monombrficos para todos los especfmenes. Un andlisis de computadora de la variacidn enzimStica resultante reveld que los especfmenes de estas dos localidades fueron el 94.5% gendticamente similares. Se encontrd que un solo espdcimen de Euglandina singleyana (Binney) de New Braunfels en Texas central es el 47.6% parecido a especfmenes de Euglandina texasiana. Two naturally occurring widespread species of the predaceous land snail Euglandina are currently recognized from Texas. Euglandina singleyana is reported from a large area of central Texas. It ranges from Terrell County in the west to Fayette County in the east, and south to Refugio County (Strecker 1935; Pilsbry & Ferriss 1906; Pilsbry 1946; Fullington & Pratt 1974; Neck 1980; Hubricht 1985: Map 342). Euglandina texasiana inhabits areas of Hidalgo, Cameron and Willacy counties in the Rio Grande Valley of south Texas (Pilsbry 1946; Fullington & Pratt 1974; Harry 1983; Neck 1984; Hubricht 1985: Map 341). These two species of Euglandina are allopatric and separated by a zone of over 200 km. 38 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 1, 2002 Euglandina texasiana also inhabits a region of coastal lowlands in Mexico which extends from the Rio Grande Valley south through Tamaulipas to eastern San Luis Potosi and northern Veracruz (Pilsbry 1907-08; 1946; Pilsbry & Vanatta 1936; Correa 1999; 2000; Correa et al. 1998). It also ranges westward to Nuevo Leon (Correa 1999). While northeastern Mexico in general is characterized by the presence of numerous conspecifics (Pilsbry 1907-08; Pilsbry & Vanatta 1936; Correa 1993; 1996-97; 1999; 2000; Correa et al. 1998), these coastal lowlands of northern Tamaulipas appear to lack additional species and subspecies of Euglandina. Collections made during this study at both San Fernando and Soto la Marina yielded only specimens of E. texasiana. This study was undertaken to examine and determine the level of enzymatic variation among specimens of Euglandina texasiana from two distant collection localities in south Texas and northern Tamaulipas. The collection site of San Fernando represents a location near the center of the distributional range of Euglandina texasiana in Mexico. The selection of this collection location also appears to minimize any possible influence of the numerous additional species and subspecies which are present in areas to both the south and west of this region of northeastern Mexico. In addition, these electrophoretic results are compared with those from a single specimen of E. singleyana from near its type-locality in central Texas. It should be noted that the habitat of E. texasiana in south Texas is rapidly being eliminated due to agricultural clearing (Fullington & Pratt 1974; Neck 1984; 1988) and that Euglandina specimens in Texas are generally considered to be uncommon (Singley 1893; Neck 1984; 1988). As a result of their very specialized feeding habits, rarity, habitat preferences, as well as the results of human activities, the five specimens examined during this study represent a significant collection of living specimens of Euglandina. Additionally, Gorman & Renzi (1979) support the validity of the use of small sample sizes in electrophoretic studies such as this one. PEREZ & STRENTH 39 Table 1 . Enzymes with buffer system used. The buffer system used for all enzyme systems was Tris-Glycine pH 8.5. Recipe from Hebert & Beaton (1993). Enzyme (E.C. No.) Abbreviation Adenylate kinase (2. 7. 4. 3) ADK Aspartate aminotransferase (2. 6. 1.1) AAT-1 AAT-2 Catalase (1.11.1.6) CAT Glucose-6-phosphate Dehydrogenase (1.1.1 .49) G6PDH Glucose-6-phosphate Isomerase (5. 3. 1.9) GPI Glutamate Dehydrogenase (1.4. 1.2) GTDH Hexokinase (2. 7. 1.1) HK Isocitrate Dehydrogenase (1.1.1.42) IDH Malate Dehydrogenase (1.1.1.37) MDH-1 MDH-2 Malate Dehydrogenase (NADP+) (1.1.1.40) MDHP-1 MDHP-2 Mannose Phosphate Isomerase (5.3. 1.8) MPI Phosphoglucomutase (5. 4. 2. 2) PGM Materials and Methods Two specimens each of Euglandina texasiana were collected from Mission in Hidalgo County of south Texas and San Fernando (150 km distant to the south) in Tamaulipas, Mexico. One specimen of Euglandina singleyana was collected from New Braunfels in Comal County of central Texas. A single specimen of Rabdotus altematus from Nacimiento de Rio Frio in Tamaulipas was selected as an out¬ group. Following collection, individual specimens were held without feeding for 7-10 days. They were then removed from their shells and the tissue was frozen in cryotubes in liquid nitrogen and stored at -80°C in an ultracold freezer until analysis. Samples of muscular foot tissue were homogenized in two volumes of distilled water using a glass rod and centrifuged to obtain an aqueous extract. Procedures for cellulose acetate electrophoresis followed those of Hebert & Beaton (1993). Gels were purchased from Helena Laboratories Inc. (Beaumont, Texas). The stain and buffer recipes used follow those of Shaw & Prasad (1970) and Hebert & Beaton (1993). The buffer used was Tris-Glycine pH 8.5. Scorable data for fifteen loci (Table 1) were obtained and analyzed using 40 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 1, 2002 Similarity Rabdotus San Fernando, Mx Mission, Tx E. singleyam J0CT+ +\T+ +33+~ +50 +67 +!§3+ t.00 Figure 1 . Phenogram of genetic similarity based upon cellulose acetate gel electrophoresis of specimens of Euglandina texasiana from Texas (Mission) and Tamaulipas (San Fernando), Euglandina singleyana from central Texas (New Braunfels) and Rabdotus alternatus (outgroup) from Tamaulipas, Mexico. the BIOSYS-1 computer program (S wofford & Selander 1981). To determine genetic similarity, Rogers’ genetic similarity (1972) was calculated. An unweighted pair group method using arithmetic averages (UPGMA) cluster analysis was then performed using Rogers’ genetic similarity matrix. The shells of all specimens examined during this study are deposited with the holdings of the Strecker Museum (SM) of the Mayborn Museum Complex of Baylor University. Material Examined Euglandina texasiana.— Two specimens (SM 32449, 32450), Mission, Hidalgo County, Texas, 1 July 1992; two specimens (SM 32447, 32448), 5 km S of San Fernando, Tamaulipas, Mexico, 20 May 1992. Euglandina singleyana.— One specimen (SM 32451), New Braunfels, Comal County, Texas, 12 March 1991. Rabdotus alternatus. — One specimen, Nacimiento de Rio Frio (22 km NNW of Ciudad Mante), Tamaulipas, Mexico, 24 May 1991. Results and Conclusions The results of this study (Figure 1, Table 2) reveal the presence of a moderately high degree of genetic similarity in all four specimens of Euglandina texasiana examined from both Texas and Tamaulipas. Ten PEREZ & STRENTH 41 Table 2. Allele frequencies in specimens of Euglandina texasiana (San Fernando and Mission), E. singleyana and Rabdotus alternatus from Texas and Mexico. R. alter. S. Fern., Mission, E. (outgrp) Tamp. Texas sing. n 1 2 2 1 Locus PGM A .000 .000 .000 .000 B .000 .000 .000 .000 C .500 1.000 .500 .000 D .500 .000 .000 .000 E .000 .000 .000 .500 F .000 .000 .500 .500 GTDH A 1.000 .000 .000 .000 B .000 1.000 .750 1.000 C .000 .000 .250 .000 MDHP-1 A .000 1.000 1.000 1.000 B 1.000 .000 .000 .000 MDHP-2 A 1.000 .000 .000 .000 B .000 1.000 1.000 1.000 HK A .000 .750 .000 .000 B 1.000 .000 .500 .000 C .000 .250 .000 1.000 D .000 .000 .500 .000 IDH A .000 .000 .000 .000 B .000 .000 .000 1.000 C .000 1.000 1.000 .000 D 1.000 .000 .000 .000 ADK A .000 .500 .750 .000 B .000 .500 .250 1.000 C 1.000 .000 .000 .000 AAT-1 A .000 1.000 .500 .000 B 1.000 .000 .000 .000 C .000 .000 .500 1.000 R. alter. S. Fern., Mission, E. (outgrp) Tamp. Texas sing. n 12 2 1 Locus AAT-2 A .000 1.000 1.000 .000 B .000 .000 .000 1.000 c 1.000 .000 .000 .000 CAT A 1.000 .000 .000 .000 B .000 1.000 1.000 1.000 MDH- A .000 1.000 1.000 1.000 B 1.000 .000 .000 .000 MDH-2 A .000 .000 .000 .000 B .000 .000 .000 .500 c .000 1.000 1.000 .000 D 1.000 .000 .000 .500 GPI A .000 1.000 1.000 .000 B .000 .000 .000 1.000 C 1.000 .000 .000 .000 G6PD A .000 1.000 1.000 .000 B .000 .000 .000 .000 c .000 .000 .000 1.000 D .000 .000 .000 .000 E .500 .000 .000 .000 F .500 .000 .000 .000 MPI A 1.000 .000 .000 .000 B .000 1.000 1.000 .000 c .000 .000 .000 1.000 42 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 1, 2002 of the 15 loci examined were monomorphic for all specimens; variation was observed in five loci (PGM, GTDH, HK, ADK, AAT-1). The two specimens from San Fernando were identical at 14 of the 15 loci and differed only when stained for hexokinase (Table 2). The two speci¬ mens from Mission exhibited a greater degree of variation than the San Fernando specimens with variation observed at four loci (GTDH, HK, ADK, AAT-1). An analysis of the resulting enzymatic variation (Figure 1) revealed that specimens from the two collection localities were 94.5 % genetically similar. This overall genetic similarity of 94.5% for all specimens from both locations is well within the range expected for genetic variation within a single species (Quicke 1993) and compares with similar results in populations of Helix aspersa by Selander & Kaufman (1975) and Helicina orbiculata by Strenth & Littleton (2000). The single specimen of Euglandina singleyana from central Texas was found to be only 47.6% genetically similar (Figure 1) to specimens of E. texasiana from south Texas and northern Tamaulipas. This low level of genetic similarity supports the validity of the results of earlier workers in maintaining the distinction and separation of these two species of land snails based upon differences in shell morphology and geographical distribution. Acknowledgments The authors wish to thank the Beta Beta Beta organization for the research scholarship that provided financial support for this research. For help in procuring specimens we wish to thank Thomas G. Littleton, Dr. Brad C. Henry of UT Pan-American and Lynn McCutchen of Kilgore College. We also thank Dan Webb and John Beatty for labora¬ tory assistance and Dr. Neil Devereaux for translation of the Spanish resumen. Literature Cited Correa-Sandoval, A. 1993. Caracoles terrestres (Mollusca: Gastropoda) de Santiago, Nuevo Leon, Mexico. Revista de Biologia Tropical, 4 1(3): 683-687. Correa-Sandoval, A. 1996-97. Caracoles terrestres (Mollusca: Gastropoda) de Iturbide, Nuevo Leon, Mexico. Revista de Biologia Tropical, 45(1): 137-142. PEREZ & STRENTH 43 Correa-Sandoval, A. 1999. Zoogeografia de los gastropodos terrestres de la region oriental de San Luis Potosf, Mexico. Revista de Biologia Tropical, 47(3): 493 -502. Correa-Sandoval, A. 2000. Gastropodos terrestres del norte de Veracruz, Mexico. Acta Zoologica Mexicana (nueva serie), 79:1-9. Correa-Sandoval, A., A. Garcfa-Cubas & M. Reguero-Reza. 1998. Gastropodos terrestres de la region oriental de San Luis Potosf, Mexico. Acta Zoologica Mexicana (nueva serie), 73:1-17. Fullington, R. W. & W. L. Pratt. 1974. The aquatic and land mollusca of Texas. Dallas Mus. Nat. Hist., Bull. 1 (part 3): 1-48. Gorman, G. C. & J. R. Renzi, Jr. 1979. Genetic distance and heterozygosity estimates in electrophoretic studies: effects of sample size. Copeia, 2:242-249. Harry, H. W. 1983. Notes on the flesh-eating land snail, Euglandina rosea in Texas, and its feeding habits. Texas Conchologist, 20(l):23-27. Hebert, P. & M. J. Beaton. 1993. Methodologies for Allozyme Analysis Using Cellulose Acetate Electrophoresis. Helena Laboratories. Beaumont, Texas, pp. 1-32. Hubricht, L. 1985. The distribution of the native land molluscs of the eastern United States. Fieldiana Zool., No. 24:1-191. Neck, R. W. 1980. Two disjunct populations of Euglandina singleyana (W. G. Binney). The Veliger, 23:112. Neck, R. W. 1984. Restricted and declining nonmarine molluscs of Texas. Texas Parks and Wildlife Department Technical Series No. 34: 17 pp. Neck, R. W. 1988. Urban Refugia which support dense concentrations of Euglandina singleyana. Texas Conchologist, 24(3):78-82. Nei, M. 1972. Genetic distance between populations. Amer. Nat., 106:283-292. Pilsbry, H. A. 1907-08. Oleacinidae, Ferussacidae. Vol. XIX in G. W. Tryon’s Manual of Conchology, Academy of Natural Sciences of Philadelphia, 366 pp. +52 plates. Pilsbry, H. A. 1946. Land mollusca of North America (north of Mexico). Acad. Nat. Sci. Philadelphia. Monograph 3 (Vol. 2-part 1): 188-199. Pilsbry, H. A. & J. H. Ferriss. 1906. Mollusca of the southwestern states, II. Proc. Acad. Nat. Sci. Phila. , 58:123-175. Pilsbry, H. A. & E. G. Vanatta. 1936. Three Mexican Euglandinas. The Nautilus, 49(3): 97-98 + plate 7. Quicke, D. L. J. 1993. Principles and techniques of contemporary taxonomy. Blackie Academic & Professional, Glasgow, pp. 167-189. Rogers, J. S. 1972. Measures of genetic similarity and genetic distance. Stud. Genet. VII, Univ. Texas, Pub., 7213:145-153. Selander, R. K. & D. W. Kaufman. 1975. Genetic structure of populations of the brown snail {Helix aspersa). I. Microgeographic variation. Evolution, 29(3) : 385-401 . Shaw, C. R. & R. Prasad. 1970. Starch gel electrophoresis of enzymes - A compilation of recipes. Biochemical Genetics, 4:297-320. Singley, J. A. 1893. Texas Molluscs. Geological Survey of Texas. Fourth Annual Report., Pt 2, 1892:301-302. Strecker, J. K. 1935. Land and fresh-water snails of Texas. Trans. Texas Acad. Sci., 17:4-50. 44 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 1, 2002 Strenth, N. E. & T. G. Littleton. 2000. A revision of the land snail Helicina orbiculata (Gastropoda: Prosobranchia) from the southern United States. Texas Jour. Sci. , 52(1) :25- 32. Swofford, D. L. & R. B. Selander. 1981. BIOSYS-1: A Fortran program for the comprehensive analysis of electrophoretic data in population genetics and systematics. J. Hered., 72:281-283. KEP at: Kathryn_Perez@excite.com TEXAS J. SCI. 54(l):45-50 FEBRUARY, 2002 SPATIAL ASSOCIATIVE LEARNING IN THE CREVICE SPINY LIZARD, SCELOPORUS POINSETTII (SAURIA: IGUANIDAE) Fred Punzo Box 5F, Department of Biology University of Tampa Tampa , Florida 33606 Abstract.— Studies were conducted to assess the spatial learning ability of adults of Sceloporus poinsettii. Experimental design was such that it tested the ability (discrimination ratio, DR) of specimens to re-visit sites that had provided food on the previous day. DRs plotted as a proportion of correct responses were significantly greater than chance for all lizards, showing that these animals returned to a location where they had found food 24 hr earlier. This 24 hr period was longer than those previously reported for reptiles on other types of spatial learning tasks. The adaptive significance of spatial learning in lizards is discussed. The ability of an animal to associate specific locations with the availability of food or some other required resource (spatial learning) would certainly contribute to its overall fitness. Spatial learning has been reported in fish (Reebs 1994), birds (Wilkie & Willson 1992) and mammals (Leonard & McNaughton 1990; Poucet 1993; Janson 1998), as well as insects (Punzo 1985a; 1996; Beugnon et al. 1996) and spiders (Punzo & Kukoyi 1997), but less information is available on amphibians (Brattstrom 1990; Punzo 1991) and reptiles (Burghardt 1977; Brattstrom 1978; Kirkish et al. 1979; Punzo 1985b; Holtzman et al. 1999). Spatial learning can significantly reduce the amount of time spent in random searching patterns and as a result maximize foraging activities (Stephens & Krebs 1986). Previous research has suggested that reptiles learn and remember a variety of spatial tasks encountered under natural conditions. These include the location of water and food by turtles (Yeomans 1995) and snakes (Weatherhead & Robertson 1990), the location of escape routes in snakes (Holtzman et al. 1999), orientation and navigation in alligators (Rodda 1985) and lizards (Adler & Phillips 1985), location of conspeci- fics in lizards (Korning et al. 2000), and homing behavior in sea turtles (Lohmann & Loll m arm 1996). In the present study laboratory experiments were conducted to assess the ability of the crevice spiny lizard, Sceloporus poinsettii, to return to 46 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 1, 2002 a specific food source based on previous experience (spatial associative learning). This lizard typically inhabits rocky canyons, hillsides and outcrops of limestone or granite, as well as lava formations in mesquite grasslands and arid woodlands (Bartlett & Bartlett 1999). It is a wary animal that frequently perches on rocks and boulders, quickly retreating into crevices when disturbed. Materials and Methods The subjects were seven adult male lizards (6 - 7 cm, SVL) that were raised as hatchlings in the laboratory. Their maternal parents ( n = 4) were collected from several locations in Big Bend Ranch State Park (Presidio County, Texas) in 1997. The lizards were housed separately in plastic cages and maintained on a diet of mealworm larvae and adults, as well as crickets, katydids, harvestmen and spiders. The lizards were tested in a Plexiglass chamber (60 by 60 by 30 cm). The floor of the chamber was covered with a piece of synthetic turf grass carpeting. A small glass dish was placed on the floor at the center of each wall, and positioned so that it made contact with the wall. These dishes were used to provide food reinforcement during training sessions. All experiments were conducted in a room with no windows. A cool fluorescent light was positioned 120 cm directly above the center of the chamber. Under these conditions, the temperature on the floor of the chamber was 28° ± 2°C, with a relative humidity of 65 - 72%. In order to acclimate the lizards to the test chamber (pretrial sessions), two mealworm larvae ( Tenebrio molitor , 12 - 15 mm in length) were placed in each dish, and each lizard was introduced separately into the center of the chamber (one at a time, with its head facing the east wall of the room) and allowed to feed from any dish of its choosing. These pretrial sessions (once per day) consisted of a period of 2 hr over a three-day period. The lizards typically ran from one dish to another, eating from two to five mealworms per session, and all of the lizards had visited each of the dishes at least twice over the course of three days. This feeding regime allowed the lizards to maintain their body weight at approximately 95% of their prior free-feeding weights. All lizards were deprived of food for 72 h prior to trials. For all spatial associative learning trials each lizard received 25 daily sessions, and each session was divided into two stages. No food was available in any of the dishes during the initial stage. The lizard was PUNZO 47 manually placed in the center of the chamber, and the initial stage began with the first visit to any food dish. This initial stage (which lasted an average of 90 sec) was considered terminated when a lizard made contact with a food dish with its snout. Data from these trials allowed an assessment of whether or not the lizards would continue to choose to visit a dish that had provided food during a previous session. In all experiments, observations on the lizards were made behind a one-way mirror so as to minimize distraction of the animals. During the second stage of each session (lasting 30 min), only one dish (randomly chosen, using a table of random numbers) provided food each day. During each session a discrimination ratio was calculated for responses during the initial no-food period as described by Wilson & Wilkie (1993). The total number of visits made to the dish that had provided food on the previous day was divided by the total number of visits made to all four dishes during this period. This ratio is a measure of the animals’ tendency to persevere at the dish that provided food (positive reinforcement) on the previous day and can be used as an index of the animals’ capacity to remember place-food associations from day to day (Mistlberger 1994). In the absence of such memory, this ratio would be expected to have a value of 0.25. These discrimination ratios were expressed as the proportion of correct responses averaged over all training sessions for each subject. Significance was assessed using a Chi Square test (Sokal & Rohlf 1995). Results and Discussion Figure 1 illustrates the discrimination ratios plotted as the proportion of correct responses averaged over all sessions for each lizard. These ratios are significantly greater than chance ( X 2 = 13.24, P < 0.01). The results indicate that these lizards will return to a location (dish) where they had found food 24 hr earlier. This 24 hr period is longer than those previously reported for lizards (Burghardt 1977; Kirkish et al. 1979; Punzo 1985b) and other reptiles (Yeomans 1995; Ishida & Papini 1997) in other types of spatial learning tasks. This memory capacity compares favorably with results reported for birds and mam¬ mals (Sherry et al. 1992; Benhamou & Poucet,1996). For many animals, food availability may vary both spatially and temporally over the course of a day or several days. For example, prey may be more abundant at one particular location in the morning or afternoon. If this spatial and/or temporal pattern remains consistent on 48 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 1, 2002 Subjects Figure 1 . Proportion of visits to a food dish during the initial non-rewarded stage of each session that had provided food during the previous session. See text for details. a daily basis, an efficient forager should learn to visit those sites where prey abundance is highest. An even greater foraging efficiency could be attained by learning to visit a specific location during a particular time interval when food is more readily available, a behavior known as time-place learning (TPL). To date, TPL has been shown only in honey bees (Gould 1986) and ants (Beugnon et al. 1996), as well as some species of birds (Biebach et al. 1989; Wilkie & Wilson 1992) and mammals (Benhamou & Poucet 1996). This study shows that S. poinsetti learned to associate specific locations characterized by a higher probability of finding prey. Temporal aspects of learning have not yet been clearly demonstrated in reptiles and TPL should be further investi¬ gated in this group. In experiments on reversal training, another type of spatial learning task, investigators had concluded that reptiles did not perform very well (Bitterman 1965; 1975). However, Kirkish et al. (1979) showed that the ability of the gecko, Coleonyx variegatus to learn a spatial reversal task was similar to that shown by birds. Thus, it appears that the capacity of reptiles to modify their behavior based on past experience (behavioral plasticity) is more highly developed than previously thought, and future studies should focus on other types of learning tasks that have ecological relevance. PUNZO 49 The results of the present study also may suggest that under natural conditions, S. poinsetti may be expected to revisit locations where food was successfully procured on the previous day. Further work is needed to determine if these lizards will move to different patches when prey density falls below a certain level and capture rates decline over a certain period of time. Spatial associative learning allows foraging animals to revisit those locations which are most likely to obtain a source of food while minimizing the costs associated with search time. Acknowledgments I thank two anonymous reviewers as well as R. A. Seigel and C. Bradford for commenting on earlier drafts of the MS. I also thank T. Punzo for his assistance in running some of the learning sessions and recording data, and the University of Tampa for a Faculty Development Grant which made much of this work possible. This research was conducted with the permission of the Texas Parks & Wildlife Department, Permit # 44-97. Literature Cited Adler, K. & J. B. Phillips. 1985. Orientation in a desert lizard ( Uma notata ): time-compensated compass movement and polarotaxis. J. Comp. Physiol. (A), 156(4):547-552. Bartlett, R. D. & P. P. Bartlett. 1999. A field guide to Texas reptiles and amphibians. Gulf Publ. Co., Houston, Texas, 289 pp. Benhamou, S. & B. Poucet. 1996. A comparative analysis of spatial memory processes. Behav. Process., 35(1): 1 13-126. Beugnon, G. P., B. Isabelle, B. Schatz & J. P. Lachaud. 1996. Cognitive approach of spatial and temporal information in insects. Behav. Process., 35(1) :55-62. Biebach, H., M. Gordjin & J. R. Krebs. 1989. Time-and-place learning by garden warblers, Sylvia borin. Anim. Behav., 37:353-360. Bitterman, M. E. 1965. The evolution of intelligence. Sci. Amer., 212:92-100. Bitterman, M. E. 1975. Phyletic differences in learning. Amer. Psychol., 2(2):396-409. Brattstrom, B. H. 1978. Learning studies in lizards. Pp. 173 - 181, in Behavior and neurology oflizards (N. Greenberg & P. D. MacLean, eds.), National Institute of Mental Health, Rockville, Maryland, 321 pp. Brattstrom, B. H. 1990. Maze learning in the fire-bellied toad, Bombina orientalis. J. Herpetol., 24(l):44-47. Burghardt, G. M. 1977. Learning processes in reptiles. Pp. 555-681, in Biology of the Reptilia. Vol. 7 (C. Gans & D. W. Tinkle, eds.), Academic Press, New York, 614 pp. Gould, J. L. 1986. The locale map of honeybees: do insects have cognitive maps? Science, 232:861-863. Holtzman, D. A., T. W. Harris, G. Aranguren & E. Bostocks. 1999. Spatial learning of an escape task by young com snakes, Elaphe guttata guttata. Anim. Behav., 57(l):51-60. Ishida, M. & M. R. Papini. 1997. Massed-trial overtraining effects on extinction and 50 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 1, 2002 reversal performance in turtles ( Geoclemys reevesii). Quart. J. Exper. Psychol., 8(1): 1-16. Janson, C. H. 1998. Experimental evidence for spatial memory in foraging wild capuchin monkeys, Cebus apella. Anim. Behav., 55(6): 1229-1243. Kirkish, P. M., J. L. Forbes & A. M. Richardson. 1979. Spatial reversal learning in the lizard Coleonyx variegatus. Bull. Psychon. Soc., 13(2):265-267. Korning, P., M. J. Whiting & J. W. Ferguson. 2000. Interspecific aggression in flat lizards suggests poor species recognition. Afr. J. Herpetol., 49:139-146. Leonard, B. & B. L. McNaughton. 1990. Spatial representation in the rat: Pp. 363-422, in Conceptual, behavioral and neurophysiological perspectives (R. P. Kesner & D. Olton, eds.), Eribaum, Hillsdale, New Jersey, 483 pp. Lohmann, K. J. & C. M. Lohmann. 1996. Orientation and open-sea navigation in sea turtles. J. Exp. Biol., 199(1):73-81 . Mistleberger, R. E. 1994. Circadian food-anticipatory activity: formal models and physiological mechanisms. Neurosci. Biobehav. Rev., 18(1): 171-195. Poucet, B. 1993. Spatial cognitive maps in animals - new hypotheses on their structure and neural mechanisms. Psychol. Rev., 100(1): 163-182. Punzo, F. 1985a. Recent advances in behavioral plasticity in insects and decapod crustaceans. Florida Sci., 68(1):89-102. Punzo, F. 1985b. Neurochemical correlates of learning and role of the basal forebrain in the brown anole, Anolis sagrei (Lacertilia, Iguanidae). Copeia, 1985(3):409-414. Punzo, F. 1991. Group learning in tadpoles of Rana heckscheri (Anura:Ranidae). J. Herpetol., 25(2):214-217. Punzo, F. 1996. Localization of brain function and neurochemical events associated with learning in insects. Trends Comp. Biochem. Physiol., 2(1):9-15. Punzo, F. & O. Kikoyi. 1997. The effects of prey chemical cues on patch residence time in the wolf spider Trochosa parthenus (Chamberlin) (Lycosidae) and the lynx spider Oxyopes salticus Hentz (Oxyopidae). Bull. Br. Arachnol. Soc., 10(3):323-326. Reebs, S. G. 1994. A test of time-place learning in a cichlid fish. Behav. Process., 30(2):273-281 . Rodda, G. H. 1985. Navigation in juvenile alligators. Zeitschr. Tierpsychol . , 68(l):65-77. Sherry, D. F., L. Jacobs & S. Gaullin. 1992. Spatial memory and adaptive specialization of the hippocampus. Trends Neurosci., 15(2):298-303. Sokal, R. R. & F. J. Rohlf. 1995. Biometry: The principles and practice of statistics in biological research. W. H. Freeman and Co., New York, 818 pp. Stephens, D. W. & J. R. Krebs. 1986. Foraging theory. Princeton University Press, Princeton, New Jersey, 408 pp. Weatherhead, P. J. & I. C. Robertson. 1990. Homing to food by black rat snakes ( Elaphe obsoleta). Copeia, 1990(4): 1 164-1 165. Wilkie, D. M. & R. J. Wilson. 1992. Time-place learning by pigeons, Columba/ivia. J. Exp. Anim. Behav., 57(1): 145-158. Wilson, R. J. & D. M. Wilkie. 1993. Pigeons remember briefly trained spatial location-food associations over extended periods of time. J. Exp. Psychol. (Anim. Behav. Proc.), 19(2):373-379. Yeomans, S. R. 1995. Water-finding in adult turtles: random search or oriented behavior? Anim. Behav., 49(5):977-987. FP at: tpunzo@ut.edu TEXAS J. SCI. 54( 1):5 1-58 FEBRUARY, 2002 LONG-TERM STRUCTURAL HABITAT USE OF MALE INDIVIDUALS OF TWO NATIVE AND ONE INTRODUCED ANOLIS (IGUANIDAE) SPECIES ON THE NORTH COAST OF JAMAICA Allan J. Landwer and Gary W. Ferguson Department of Biology, Box 16165, Hardin-Simmons University Abilene, Texas 79698 and Department of Biology, Texas Christian University Fort Worth, Texas 76129 Abstract.— This study compares the perch heights and densities of male Anolis grahami, A. sagrei and A. lineatopus at four localities near Ocho Rios, Jamaica during the spring of 1983, 1987, 1994, 1996, 1998 and 2000. This is the first report of an Anolis perch height study sampling the same study area for a period of almost two decades. The finding of stability over the long-term in this system lends validity to numerous short-term studies. Data analyses included assessment of perch height and densities. Mean perch heights were signifi-cantly different among species during the study. However, there was still significant overlap in this niche dimension in many years of the study. Anolis sagrei is an invader that has integrated into the Anolis community at this locality. Implications of these findings are discussed regarding reasons for coexistence and potential competition between these species. Community structure may arise by both rapid processes occurring in ecological time, as well as by more long term processes acting over evolutionary time (Roughgarden et al. 1983; Grant 1986). Communities may be structured by invaders, where to be successful as a colonist, an invading species must be pre-adapted to fit in with other members of the community (Rummel & Roughgarden 1983; 1985), or structured through mutual co-adaptation of community members (Beuttell & Losos 1999). Caribbean Anolis lizards provide good model systems for analyzing community structure because of their simplicity, ease of observation, and because anoles are likely to be relatively insensitive to unobtrusive observation (Sugerman 1990). However, the extent to which such communities are the result of coevolution among species is ambiguous (Williams 1972; Roughgarden et al. 1989; Losos 1992a; 1992b; Butler et al. 2000). Investigation of the effects of invasion by Anolis sagrei may provide insight into the evolution of Anolis communities in the Caribbean. This is the first long-term report in the literature studying the Anolis community structure at the same study area for nearly two decades. This study examines perch heights and densities of three common Anolis species {Anolis grahami, A. sagrei and A. lineatopus) at four 52 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 1, 2002 localities of a study site on the north coast of Jamaica (see Underwood & Williams 1959 for a complete description of these species). Rand (1964) originally used perch height and perch diameter to distinguish the spatial niche dimensions used by the anole species that he studied, and other authors have used these measures to quantify habitat use (Schoener 1967; Schoener & Schoener 1971; Losos et al. 1993). Spatial habitat use has also been described by perch type and by the degree of insola¬ tion (Schoener & Schoener 1971; Losos et al. 1993). Perch height may be the most conveniently quantified measure of structural habitat because these characteristics are correlated with other habitat features and, if partitioned between or within species, will allow a large number of lizard species to coexist in a relatively small area. For example, in studying five Anolis species at three lowland localities on Jamaica, Schoener & Schoener (1971) found that anoles have partitioned their spatial niches to reduce spatial overlap and avoid potential intraspecific and interspecific competition. Results of these studies have led to the classification of Anolis lizards into different ecomorph classes (Butler et al. 2000). Because A. sagrei is a successful invader that is part of the landscape at these localities and is expanding its range across the island (Under¬ wood & Williams 1959; Williams 1969; Mayer 1989; Schwartz & Hen¬ derson 1991), there is reason to expect changes in the Anolis communi¬ ty. This study quantifies perch height and examines density of male Anolis species and compares these data among the species. Materials and Methods Observations were made at the Hofstra University Marine Laboratory campus at Priory, Saint Ann’s Parish, Jamaica, West Indies. The study began in March 1983, but these data were not included in the analysis because it was a project development year. Data were collected on 10- 13 March 1987, 4-9 January 1994, 16-22 March 1996, 16-20 March 1998 and 13-25 March 2000 between 0800-1800 hours from four localities on the campus. Each locality represented a study area that included a beach and three highly disturbed areas. The disturbed areas included laboratory buildings, concrete walls, telephone poles, a barb wire fence, lumber and brick piles and a low wooden pier. Flora present in all of these areas included red mangrove, coconut and banana trees and several varieties of cultivated garden plants. The study areas were termed "Beach", "Executive Suite", "Boathouse" and "Driveway". LANDWER & FERGUSON 53 Potential habitat availability and maximum available perch heights were similar in all of these study areas. One to four observers carefully searched each study area several times during the day. The junior author participated in data collection in all study years and the senior author participated in 1987 and 2000. When a lizard was observed, its perch height and perch diameter was recorded by tape measure to the nearest cm. Lizards observed on the ground were assigned a perch height of zero. Only large, adult males were used in the study because they are larger and more conspicuous due to their social dewlap displays. Several other studies on Anolis have focused exclusively on adult males (Rand & Williams 1969; Williams 1972; 1983). Perch data were used to calculate and compare mean perch heights for each species within each year of observation. Lizard densities were also calculated for all study areas in all years that observations were made. Perch heights between species for each year were compared using one¬ way ANOVA. When the same data were used in pairwise multiple comparisons (LSD Tests), sequential Bonferroni correction (alpha = 0.05) was used to judge statistical significance (Rice 1989). Probabili¬ ties reported remain significant with the Bonferroni correction. Results The mean perch heights for each Anolis species during all years of the study are shown in Fig. 1. One-way ANOVA tests indicate highly statistically significant differences in perch heights within years among all three species during each year of the study (1987 /=34.167, PCO.OOOl; 1994/= 14.938, P<0.001; 1996/= 12.216, P=0.001; 1998 /= 16.960, PcO.001; 2000/= 16.900, P<0.001). Multiple comparisons using post-hoc LSD with sequential Bonferroni correction indicated many significant differences in perch height between species for each year of observation (Table 1). Anolis grahami tended to occupy the highest perches compared to the other species studied, with A sagrei occupying the lowest and A. lineatopus occupying a mid¬ range perch height. In every year of the study, A. grahami occupied significantly higher perches than A. sagrei , but A. grahami ’s mean perch height was significantly higher than A. lineatopus only in 1987, 1998 54 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 1, 2002 300.00 250.00 200.00 *5 150.00 100.00 50.00 Mean Perch Heights — A. grahaml — A. lineatopus ■ - A. sagrei ^29.62 \ / / / / / / t T ______ - 1 162.72 1 1 ^442.15^ h44AJ_4_____ pr^64 r ] ^6Q,08 r i9»723 - [53.93 ! ~ ~ £34.54 1987 1994 1996 1998 Year of Observation 2000 Figure 1. Mean perch heights and standard errors for male Anolis grahami, Anolis lineatopus and Anolis sagrei from observations taken between 1987 - 2000. and 2000 (Table 1). Anolis lineatopus occupied significantly higher perches than A. sagrei in 1994, 1996 and 2000 (Table 1). Observed densities of the three Anolis species across all sites were similar and not significantly different from one other across years of the study. Density data are shown for all years of the study in Table 2. Discussion The data analysis results show that there are statistically significant differences among these species in perch heights. However the degree of perch height overlap varies by year (Fig. 1) and only A. grahami and A. sagrei show significantly different mean perch heights at these localities in every year studied. The reasons for this variation is not entirely clear and because of this result, it may be useful to examine other niche axes, such as degree of insolation or temperature, to explain the coexistence of these species. Statistical differences in perch heights may not solely be biologically meaningful measures of habitat use because they do not adequately or accurately represent the structural LANDWER & FERGUSON 55 Table 1. Differences in perch height between species for each year of observation 1987 - 2000. Mean values with asterisks (*) are significantly different than mean values without asterisks based on post-hoc LSD tests with sequential Bonferroni correction (P<0.05). Year Anolis grahami Mean Perch Heights (cm) Anolis lineatopus Anolis sagrei 1987 229.62 90.75* 69.31* 1994 142.15* 114.14* 58.08 1996 125.64* 101.93* 34.54 1998 152.84 61.87* 75.71* 2000 162.72 99.23 53.93 niche of a lizard. Other niche axes such as degree of thermal prefer¬ ence, and/or finer scale habitat measures may be needed to fully quantify habitat use. For example, thermal preference data taken from Anolis at these sites in 1996 and 1998 show that A. grahami is a thermo¬ regulator that prefers to bask in the sun while A. lineatopus is a thermo- conformer (Huey & Webster 1976) and prefers more shaded habitats. Anolis densities were similar in all years of this study and although there was a trend towards A. sagrei having the highest density, its density was not significantly different than that of the other species. Those factors which allow these species to coexist remain unresolved. There may be ecological differences among these species. For example, A. sagrei exhibited tremendous phenotypic plasticity in hindlimb length when raised under laboratory conditions in different structural habitats leading to phenotypes well adapted to particular environments (Losos et al. 2000). Such plasticity is likely to be advantageous to these colonists both in getting established and in subsequent persistence. Pre-invasion body- size differences may affect the size of prey items consumed by these lizards (Schoener 1967; Roughgarden 1974; but see Floyd & Jenssen, 1983). Also, other unmeasured niche axes on which these species may be very different in their overlap include differences in insolation and preferred body temperatures. Anolis lineatopus was observed far more frequently in the shade than the other two species, and hence may not compete with the other species for shady perches. Cuban and Floridian A. sagrei are thermophilic and utilize sunny low habitats (Ruibal 1961; Salzburg 1984). In this study, A. sagrei and A. grahami were both seen frequently at these types of sites and especially on the beach in direct sunlight. Hence, preferred body temperatures 56 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 1, 2002 Table 2. Mean densities of three Anolis species during all years of the study. Data are mean numbers of male lizards observed per census across all sites. Year Mean Density Across Years at all Sites Anolis grahami Anolis lineatopus Anolis sagrei 1987 3.4 1.1 5.3 1994 3.9 1.9 7.7 1996 2.6 1.3 4.8 1998 2.5 1.3 5.9 2000 4.2 2.2 4.2 may differ and prevent overlap among these species. All of these factors may allow the coexistence that is currently observed in this Anolis community. Long-term studies like this one are essential in understanding community structure and the observed stability at this study area for almost two decades validates numerous short-term studies. Future studies should involve detailed microhabitat use quantification and account for other important niche axes that may help to understand the ecological and ultimately the evolutionary factors that are responsible for Anolis community structure. Such information may be useful in pre¬ dicting the outcome of invasions on lizard community structure. Finally, it must be noted that conclusions reported herein about competitive effects must remain tentative in the absence of data on the entire marked populations, data from controlled field manipulations, and data on the effects of climate, resource availability, and the ongoing habitat distur¬ bances on these species. Such thorough investigations are necessary to reveal post- invasion responses to interactions among these species. Acknowledgments We are grateful to Melissa Garretson, Mark Brewer, Amy Foster, Natalie Stevens, Heather Terrill, Todd Harris, Bill Gehrmann and Kris Karsten for help with lizard observations and perch height measure¬ ments. We thank Roy Vogtsberger for critiquing the manuscript and thank the Hofstra Marine Station officials and personnel, including Eugene Kaplan, Deb Bidwell and Gorka Sancho for making this study possible. We also thank Joe Darnall for statistical advice. Partial funding for this study was made possible by W. Craig Turner, LANDWER & FERGUSON 57 Chief Academic Officer of Hardin-Simmons University as a grant to the senior author, and a TCU Research Foundation grant to Gary Ferguson. Literature Cited Beuttell, K. & J. B. Losos. 1999. Ecological Morphology of Caribbean Ancles. Herpetological Monographs, 13:1-28. Butler, M. A., T. W. Schoener & J. B. Losos. 2000. The relationship between sexual size dimorphism and habitat use in greater antillean Anolis lizards. Evolution, 54(l):259-272. Floyd, H. G. & T. A. Jenssen. 1983. Food habits of the Jamaican lizard, Anolis opalinus : resource partitioning and seasonal effects examined. Copeia, 1983:319-331. Grant, P. R. 1986. Ecology and Evolution of Darwin’s Finches. Princeton Univ. Press, Princeton, New York, 458 pp. Huey, R. B. & T. P. Webster. 1976. Thermal Biology of Anolis lizards in a Complex Fauna: The Cristatellus Group on Puerto Rico. Ecology, 73:985-994. Losos, J. B. 1992a. A critical comparison of the taxon-cycle and character displacement models for the size evolution of Anolis lizards in the Lesser Antilles. Copeia, 1992:279-288. Losos, J. B. 1992b. The evolution of convergent structure in Caribbean Anolis communities. Syst. Biol., 41:403-420. Losos, J. B., J. C. Marks & T. W. Schoener. 1993. Habitat use and ecological interactions of an introduced and a native species of Anolis lizard on Grand Cayman, with a review of the outcomes of anole introductions. Oecologia, 95:525-532. Losos, J. B., D. A. Creer, D. Glossip, R. Goellner, A. Hampton, G. Roberts, N. Haskell, P. Taylor & J. Ettling. 2000. Evolutionary Implications of Phenotypic Plasticity in the hindlimb of the lizard Anolis sagrei. Evolution, 54(1) :301 -305. Mayer, G. C. 1989. Deterministic Patterns of Community Structure in West Indian Reptiles and Amphibians. Ph.D. Dissertation, Harvard University, Cambridge, Massachusetts, 294 pp. Rand, A. S. 1964. Ecological distribution in anoline lizards of Puerto Rico. Ecology, 45:745-752. Rand, A. S. & E. E. Williams. 1969. The anoles of La Palma: aspects of their ecological relationships. Breviora 327:1-19. Rice, W. R. 1989. Analyzing tables of statistical tests. Evolution, 43:223-225. Roughgarden, J. 1974. Niche width: biogeographic patterns among Anolis lizard populations. Am. Nat., 108:429-442. Roughgarden, J., J. D. Rummel & S. W. Pacala. 1983. Experimental evidence of strong present-day competition between Anolis populations of the Anguilla Bank - a preliminary report. Pp. 499-506, in Advances in Herpetology and Evolutionary Biology: Essays in honor of Ernest E. Williams (A. Rhodin and K. Miyata, eds.), Museum of Comp Zool, Cambridge, Massachusetts, 725 pp. Roughgarden, J. 1989. The structure and assembly of communities. Pp. 203-226, in Perspectives in Ecological Theory (J. Roughgarden, R.M. May and S.A. Levin, eds.), Princeton Univ. Press, Princeton, New Jersey, 394 pp. Ruibal, R. 1961. Thermal relations of five species of tropical lizards. Evolution, 15:98-111. Rummel, J. D. & J. Roughgarden. 1983. Some differences between invasion-structured and coevolution-structured competitive communities: a preliminary theoretical analysis. Oikos, 41:477-486. Rummel, J. D. & J. Roughgarden. 1985. A theory of faunal build-up for competition 58 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 1, 2002 communities. Evolution, 39:1009-1033. Salzburg, M. A. 1984. Anolis sagrei and Anolis cristatellus in southern Florida: a case study in interspecific competition. Ecology, 65:14-19. Schoener, T. W. 1967. The ecological significance of sexual dimorphism in size of the lizard Anolis consperus. Science, 155:474-478. Schoener, T. W. & A. Schoener. 1971. Structural habitats of West Indian Anolis lizards. Jamaican lowlands. Breviora, 368:1-53. Schwartz, A. & R. W. Henderson. 1991. Amphibians and Reptiles of the West Indies: Descriptions, Distributions, and Natural History. University of Florida Press, Gainesville, Florida, 736 pp. Sugerman, R. A. 1990. Observer effects in Anolis sagrei. J. Herp., 24(3):316-317. Underwood, G. & E. E. Williams. 1959. The anoline lizards of Jamaica. Bull. Inst. Jamaica, Science Series, 9:1-48. Williams, E. E. 1969. The ecology of colonization as seen in the zoogeography of anoline lizards on small islands. Q. Rev. Biol., 44:345-389. Williams, E. E. 1972. The origin of faunas. Evolution of lizard congeners in a complex island fauna: a trial analysis. Evolutionary Biology, 6:47-89. Williams, E. E. 1983. Ecomorphs, faunas, island size and diverse end points in island radiations of Anolis. Pp. 326-370, in Lizard Ecology: Studies of a model organism (R. B. Huey, E. R. Pianka, and T. Schoener, eds.), Harvard University Press: Cambridge, Massachusetts, 501 pp. AJL at: alandwer@hsutx.edu TEXAS J. SCI. 54(l):59-62 FEBRUARY, 2002 HABITAT UTILIZATION BY EASTERN YELLOWBELLY RACERS {COLUBER CONSTRICTOR FLAVIVENTRIS) IN SOUTHWEST DALLAS COUNTY, TEXAS Richard D. Reams and William H. Gehrmann Department of Herpetology , Dallas Zoo 650 South R. L. Thornton Freeway Dallas, Texas 75203 Abstract. — A population of Coluber constrictor flaviventris was surveyed in both relic and disturbed Blackland Prairie habitats during the spring, summer and fall of 1999 and 2000 at Cedar Hill State Park in Dallas County, Texas. Drift fences with funnel traps and coverboards were used to live-trap snakes. During the two years, 69 specimens were recorded. The study demonstrated that the disturbed Blackland Prairie had significantly more specimens then the relic Blackland Prairie. There are several publications discussing reptiles and amphibians utilizing and adapting to disturbed habitats (Dyrkacz 1977; Enge 1998; Fitch 1999; Kaufmann 1992.). Bury (1983) reported that the amphibian populations in an old growth forest (both species richness and abun¬ dance) were drastically altered after logging had occurred. However, comparisons of reptiles or amphibians utilizing both relic and disturbed habitats of the same habitat type are uncommon. Coluber constrictor is found in a variety of habitats throughout the United States including grasslands, prairies and forested areas (Conant & Collins 1998). In Cedar Hill State Park, where C. constictor flaviventris occurs, they are abundant in these habitats but were noted to exhibit a strong tendency towards grasslands. Coluber constrictor are strictly diurnal with a relatively high body temperature preference (Fitch 1999). Their diet consists of a large variety of food items including insects, reptiles, amphibians and rodents in adult specimens (Fitch 1999). This study compares habitat usage by Coluber constrictor flaviventris in both relic and disturbed prairie habitats. Study Site Cedar Hill State Park is located in southwest Dallas County and encompasses 1826 acres of disturbed prairie, relic prairie, oak forest and disturbed hillside forest. Adjacent to the park is Joe Pool Lake, which was created in 1981 for water consumption by the Dallas/Fort Worth metroplex residents. The topography is gently rolling hills except for 60 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 1, 2002 steep west-facing bluffs associated with a chalk outcrop of the White Rock Escarpment (Carr 1993). Elevation varies from 115 m in the prairies to 250 m on the escarpment. The two habitats studied were relic Blackland Prairie (RBP) and disturbed Blackland Prairie (DBP). The RBP (32° 37’N, 96° 60’W) is characterized by nearly level to gently rolling hills with extremely fertile soil. The soil of the Blackland Prairie within the park is considered deep, moderately well drained, moderately alkaline and contains black calcareous clay (Coffee et al. 1980). The blackland habitat is true tallgrass prairie, with the little blue- stem ( Schizachyrium scoparium) as the climax dominant vegetation (Simpson 1998). This relatively undisturbed habitat has never been plowed or grazed by livestock and is maintained with controlled burns every three to five years. DBP (32° 36’N, 96° 60’W) is similar to RBP in that the soil content and elevation are comparable. However, DBP has been grazed by livestock and the soil was once extensively farmed. The year of the last disturbance on this habitat was ca. 1980. At present, the habitat consists of mostly introduced grasses or non- woody native and introduced weeds. No dominant weed species could be discerned. There are no controlled burns in the disturbed prairie. Because of this the vegetation is much thicker and higher than that of the relic prairie. Methods and Materials Within both habitat types, four 50 m2 plots were established. These plots were approximately 20 m from the edge of any other plots. All plots were no less than 50 m from an adjacent habitat edge to avoid sampling species utilizing edge habitat. In the DBP, an unpaved park road measuring four meters wide divides a portion of the habitat sampled. One drift fence made of aluminum flashing measuring 15.5 m long and 0.5 m high was erected in the center of each of the four plots within each habitat type. A funnel trap measuring 60 cm by 25 cm was used at both ends of each drift fence. Due to the lack of shade and the heightened chances of a snake exceeding its thermal limit while confined in a funnel trap, boards (approximately the same size as the traps) were placed on top to provide shade. In addition to the plots, four-one m2 cover boards were placed on the prairie ground. These cover boards were placed 10 m from the drift fences. While checking traps, visual searches were conducted on the way to and from the traps. If specimens were observed within these plots, they were recorded from that habitat type. REAMS & GEHRMANN 61 Table 1 . Numbers of individual specimens of Coluber constrictor flaviventris collected in relic Blackland Prairie and disturbed Blackland Prairie habitats during the spring, summer and fall of 1999 and 2000 in Dallas County, Texas. Spring Summer Fall Totals Relic Blackland Prairie 1999 (3) 1999 (1) 1999 (1) 5 2000 (4) 2000 (2) 2000 (1) 7 Disturbed Blackland Prairie 1999 (18) 1999 (2) 1999 (7) 27 2000 (17) 2000 (4) 2000 (9) 30 All traps were checked three times per week in the spring (April - mid June), once a week during the hotter summer months (Mid June - 1 September) and twice a week in autumn (September - 1 December), for both years. The traps were checked a total of 134 days during the two years. Specimens were tagged with passive integrated transponders (pit-tags) to record recaptures. Recaptures were not included in the data presented in this study. Results and Discussion In 1999, five specimens were captured in the RBP, compared to 27 specimens in the DBP for a total of 32 specimens. The number of captures between the two habitats were significantly different (x2— 15.2, P<0.01). In year 2000, seven specimens were captured in the RBP, compared to 30 specimens in the DBP for a total of 37 specimens. Captures between the two habitats was again significantly different (x2 = 14.3, P<0.01). For the two years sampled, a considerably higher number of C. constictor flaviventris were captured in the disturbed habitat compared to the relic habitat (Table 1). Fitch (1999) reported that formerly cultivated fields, woodlands and open pastures that were heavily grazed by livestock had very few C. constrictor. However, when the grazing ended and the grasses returned the C. constrictor population size increased for several years (Fitch 1999). The disturbed habitat in this study was observed as being considerably more dense with meter high vegetation, which created more ground cover than that of the relic habitat. In addition, there was man¬ made debris in the form of lumber and old car tires in the disturbed habitat only. This debris was approximately 100 m from any of the four plots sampled. Two different species of rodents were observed in the DBP and three different species were found in the RBP. However, much larger numbers of rodents were observed in the DBP compared to 62 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 1, 2002 the RBP. Insects appeared in be more common in the DBP as well. Further studies on the prevalence of food items and/or vegetative cover within these two habitats needs to be undertaken in order to determine the importance of these factors in contributing to the differences in the number snakes observed. Acknowledgements We would like to thank Ruston W. Hartdegen, Winston Card, Cynthia Bennett, the Staff of the Dallas Zoo Department of Herpetology, the staff of the Cedar Hill State Park and Texas Parks and Wildlife Department (Permit # 51-98). Literature Cited Bury, R. B. 1983. Differences in amphibian population in logged and old growth redwood forest. Amer. Midi. Natur., 103(2):412-416. Coffee, D. R., R. H. Hill & D. S. Ressel. 1980. Soil survey of Dallas County, Texas. United States Department of Agriculture, Soil Conservation Service, 153 pp. Conant, R. & J. T. Collins. 1998. A field guild to reptiles and amphibians. Houghton Mifflin Company, Boston, Massachusetts, 616 pp. Dyrkacz, S. 1977. The natural history of the eastern milk snake in a disturbed environment. J. Herpetol., 1 1(2): 155-159. Enge, K. M. 1998. Herpeto faunal survey of an upland hardwood forest in Gadsden County, Florida. Florida Scient., 61(3/4): 141-159. Fitch, H. S. 1999. A Kansas snake community: Composition and changes over 50 years. Krieger Publishing Company, Malabar, Florida, 165 pp. Kaufmann, J. H. 1992. Habitat use by wood turtles in central Pennsylvania. J. Herpetol., 26(3)315-321. Lewis, S. D., R. R. Fleet & F. L. Rainwater. 2000. Herpetofaunal assemblages of four forest types in the Big Sandy Creek Unit of the Big Thicket National Preserve. Texas J. Sci., 52(4) Supplement: 1 39- 1 50 . Simpson, B. J. 1988. A field guild to Texas trees. Gulf Publishing Company, Houston, Texas, 372 pp. RDR at: hylareams@aol.com TEXAS J. SCI. 54(l):63-68 FEBRUARY, 2002 EFFECTS OF TEMPERATURE AND LIGHT ON CHINESE TALLOW {SAPIUM SEB1FERUM) AND TEXAS SUGARBERRY {CELTIS LAEVIGATA) SEED GERMINATION Somereet Nijjer, Richard A. Lankau, William E. Rogers and Evan Siemann Department of Ecology and Evolutionary Biology Rice University, Houston, Texas 77005 Abstract. — Experiments were performed to assess germination requirements of seeds of Chinese tallow (Sapium sebiferum (L.) Roxb.) and Texas sugarberry ( Celtis laevigata Willd.). Sapium and Celtis seeds were exposed to different combinations of light and tem¬ perature. It was predicted that Sapium would germinate under a variety of environmental conditions, but Sapium seeds germinated predominantly in fluctuating temperature conditions. Celtis seeds also germinated readily in such conditions but had less restrictive germination requirements. Since Celtis appears to be better adapted to a variety of germination condi¬ tions, a broader range of environmental germination tolerances does not explain Sapium' s greater establishment success as an alien invader. Nevertheless, seeds requiring oscillating temperatures to germinate are most commonly found in canopy gaps or open areas suggesting that Sapium invasions may be especially problematic in disturbed habitats. Chinese tallow tree {Sapium sebiferum) is an invasive deciduous tree species in ecosystems throughout the southeastern United States. Origi¬ nally from Asia, it has extended its distribution considerably throughout the southeastern United States since its introduction in 1772 (Bruce et al. 1997; Barrilleaux & Grace 2000). Sapium displaces many plant species and drastically alters community structure by transforming native grass¬ lands and other habitats into monospecific woodlands (Harcombe et al. 1993; Bruce et al. 1995; Grace 1998). The objective of this study was to assess the effects of light and temperature on seed germination of Sapium and an ecologically similar native, Texas sugarberry {Celtis laevigata). Celtis and Sapium are both small deciduous, fast growing trees that are insect pollinated and have bird dispersed seeds (Van Auken & Lohstroh 1990; Jubinsky & Anderson 1996; Renne et al. 2000). In the absence of proper management regimes, Celtis may also invade grassland ecosystems (Van Auken & Bush 1990; Harcombe et al. 1993). Understanding what factors control Sapium germination is necessary for understanding the success of this invasive species and may provide insights for managing its growth at early developmental stages. Study¬ ing germination of Celtis in conjunction with Sapium provides a model for comparison with an ecologically similar native species. Germination 64 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 1, 2002 requirements for Sapium seeds were predicted to be governed less by environmental conditions than Celtis seeds and these differences were expected to partially explain the success of Sapium as an alien invader. Materials and Methods Sapium sebiferum and Celtis laevigata seeds were collected from several uncultivated trees at the University of Houston Coastal Center in Galveston County, Texas in December 1999. Seeds were stored in the dark at room temperature for two months. In February 2000, seeds were planted in subdivided germination trays lined with a thin layer of peat moss and then filled with a mixture of 2/3 commercial top soil and 1/3 humus. Seeds of each species were randomly assigned to a tempera¬ ture treatment (hot, cold and cycling) and a light treatment (light, dark and cycling) in a full-factorial design. Within each treatment, pairs of Sapium and Celtis seeds were randomly assigned to one of 240 cells (2 cm by 2 cm by 5 cm deep) in germination trays. The trays were kept in a temperature controlled room without windows for the duration of the experiment. Seeds in the hot treatment were warmed with a germination mat that maintained the soil at a constant 32 °C. Seeds in the cold treatment were kept constant at 16°C. Seeds in the cycling temperature treatment were subjected to 16 h of 32° C and 8 h of 16°C daily. Seeds assigned to the light treatment received 24 h of continuous light supplied by commer¬ cially available wide-spectrum plant grow lights suspended 20 cm above the tray surface (average PAR = 50 /xmol/m2/sec). The cool fluorescent bulbs in the cold room emitted a negligible amount of heat that was unlikely to influence soil temperatures. Seeds in the dark treatment were kept dark 24 h per day. Although the dark treatments were isolated by opaque barriers, they were periodically exposed to low levels of diffuse light. Seeds assigned to the light cycle received 16 h of light and 8 h of dark per day. The temperature and light cycles were synchronous. All treatments were lightly watered and checked for germination daily. Newly germinated seeds were removed from their cells. The experiment was conducted for 175 days, but no seeds germinated after 120 days. Separate Kruskal-Wallis nonparametric tests (i.e. nonparametric ANOVAs) were used to examine the effects of temperature treatment and light treatment on Celtis and Sapium germination (Statview 5.0, SAS Institute). Each cell was treated as an experimental unit and it was assigned a value of "yes" if a seedling germinated and a value of "no" if no seedling germinated in the cell. NIJJER ET AL. 65 Table 1 . Separate Kruskal-Wallis nonparametric tests for dependence of the probability of a seed germinating in a cell for (a) Sapium sebiferum and (b) Celtis laevigata in response to temperature treatments (continuous cold, temperature-cycle, continuous heat) and light treatments (continuous dark, light/dark-cycle, continuous light). Species Factor df adjusted H P-value (a) Sapium Temperature 2 83.7 <0.0001 Light 2 8.2 <0.05 (b) Celtis Temperature 2 375.2 <0.0001 Light 2 36.0 <0.0001 Results A significantly higher proportion of Sapium seeds germinated in the temperature-cycle treatment (Table 1, Figure la). Germination in the hot treatment and the cold treatment were extremely low. Within the temperature-cycle treatment, more Sapium seeds germinated in constant light and constant dark treatments than the light-cycle treatment (Figure la). Likewise, a significantly higher proportion of Celtis seeds germi¬ nated in the temperature-cycle treatment (Table 1 , Figure lb). Germina¬ tion of Celtis in constant temperature treatments was lower than the temperature-cycle treatment, but some constant temperature and light combinations had moderate levels of germination (Figure lb). Within temperature treatments, Celtis germination tended to be higher as light levels increased. Sapium germination was lower than Celtis germina¬ tion in all treatment combinations (Figure 1). Discussion Germination requirements for Celtis and Sapium appear to be primari¬ ly affected by fluctuating temperatures. For both species, germination in the temperature-cycle treatment far exceeded the combined germina¬ tion total in continual cold and continual heat treatments. Light treat¬ ments also had a significant, albeit less prominent, effect on germina¬ tion. Celtis seeds appeared to have less restrictive germination require¬ ments than Sapium seeds. This contrasted with predictions that Sapium ’s success as an invader may be partially due to a greater breadth of acceptable germination conditions than similar native woody species. Many plant species use temperature fluctuations to monitor seasonal changes and assess growing conditions (Fenner 1985; Baskin & Baskin 1989). Additionally, soil and canopy vegetation can insulate seeds against daily temperature fluctuations under natural conditions. Ade¬ quate temperature fluctuations likely indicate appropriate burial depth and microhabitat conditions for successful establishment and new seed¬ ling growth (Fenner 1985; Baskin & Baskin 1989). For these reasons, 66 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 1, 2002 70 6o 0s 03 C 30 i 20 2800 (>100,000) 1998 Guadalupe at Gonzales 15.4 (50.4) 9600 (340,000) 1929 Guadalupe at Gonzales 15.0 (49.3) — 1998 Guadalupe at New Braunfels 11.7 (38.5) 6300 (222,000) 1913 Guadalupe at New Braunfels 12.0 (39.5) — 1869 Guadalupe at New Braunfels 12.0 (39.5) — 1998 Guadalupe at Victoria 10.3 (33.8) 13,200 (466,000) 1936 Guadalupe at Victoria 9.5 (31.1) 5100 (179,000) * USGS Flood peaks data base gives lower value than USGS 1999 report. verticillata (hydrilla), Potamogeton illinoiesnsis (pondweed) and Colacasia escuelenta (elephant ear) (Lemke 1989; Staton 1992; Angerstein & Lemke 1994). In an analysis of this problem, Spain (1994) concluded that any program to control these exotics, including mechanical removal, biologic controls or herbicides, would be either expensive, controversial or illegal. EARL & WOOD 81 Another change observed by the public and verified by quantitative measurement has been sedimentation of the channel. Between 1990 and 1995, the City conducted annual sedimentation measurements at sites between University Drive and Hopkins Street. Employing a 0.63 cm diameter rod-penetrometer, sediment depth to the closest 0.1 m was measured every 3 meters across the channel at cross-sections every 25 meters in SWT Sewell Park and every 50 meters farther downstream. Between the scouring floods December 1991- January 1992 and 1995, there was an average channel deposition of 0.50 meters or 13,000 m3 in the 800 meter reach of the stream immediately downstream from Spring Lake (Wood & Gilmer 1996). Analysis of Headwater Changes A hypothesis of this study is that the decrease in drainage area has reduced the number and maximum energy of stream erosional events. Furthermore, continued construction in the headwaters area downstream of the flood control dams has increased sediment input introduced into the river. Employing the slope-area method (Dalyrimple & Bentson 1967) for calculating bankfull discharge from channel geometry, the bankfull discharge in Sewell Park, immediately downstream of Spring Lake, is 33 m3/sec (1150 ft3/sec). Because of the changes in the location and instrumentation of gage number 0817000, it is difficult to determine the number of times that this discharge has been exceeded since the completion of the upstream dams. According to the formulas developed by Slade et al. (1995) for calculating peak discharge for a given recurrence interval (Q2 = 252A0 721 (SF)'° 326 where A = area in mi2 and SF = shape factor) with the tributary area of this segment reduced from 77.0 to 15.6 km2, the bankfull discharge for the two-year event comes out to 35 m3/sec (1240 ft3/sec) or approximately bankfull discharge. As shown in Table 1, only the October 1998 event exceeded bankfull discharge since the December 1991 flood. In contrast, prior to closure of the dams, bankfull discharge was exceeded in over 82% (23/28) of the years with a mean annual flood of 510 m3/sec (18,000 ft3/sec). This 510 m3/sec (18,000 ft3/sec) value has a recurrence interval of 3.6 years. If Slade et al. ’s (1995) equations are used, the present-day 3.6 year flood or mean annual flood with the reduced basin comes out to 43 m3/sec (1,500 ft3/sec). The maximum drain release from the "Site #3 Sink Creek" dam, approximately two kilometers upstream of Spring Lake is 27 m3/sec (936 ft3/sec) but this discharge is only produced when the structure has received 254 mm 82 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 1, 2002 basin precipitation, the "100-year, 24-hr" event (US SCS 1978). Consequently, significant runoff below the flood control dams would be required to produce bankfull discharge in the main channel. Employing the TR 55 model, with a curve number of 75, between 38 and 54 mm/hr of precipitation is required to produce sufficient runoff to yield bankfull discharge (33 m3/sec or 1150 ft3/sec) downstream of Spring Lake (US SCS 1986). This precipitation rate has a recurrence interval of approximately two years, and thus, the decrease in bankfull events is at least partially due to a climatic perturbation that produced few high- intensity rainfalls between 1991 and 1998 (Hershfield 1961). Thus, both hydrologically (Slade et al. 1995) and rainfall frequency based models suggest that the 1991-1997 period of stream sedimentation was noted for its lack of large flood events. Sedimentation results when sediment supply exceeds the ability of flood events to remove the sediment supply. A major source of the sediment introduced into the river is Sessoms Creek, with a basin area of 147 ha (363 acres) and which receives runoff from much of the SWT campus. Particularly noteworthy has been the recent construction on campus. An analysis of the sediment production from the site of the new student union and other facilities was undertaken with the Universal Soil Loss Equation (USLE) (Dunne & Leopold 1978). These areas had a bare, disturbed surface area of 7.3 ha and produced an estimated sediment yield of 121 tonnes/ha/year in contrast to an estimated pre¬ construction value of 0.22 tonnes/ha/year. Based upon a density of 2.0 g/ cm3, the annual sediment production from the construction site would have a volume of 780 m3 compared to 15 m3 for the remaining 140 ha of the basin. This sediment production rate over the three years of construction that began in 1995 would produce an annual sedimentation in the channel of the San Marcos River of 16 cm/year in the upper 250 meters of the river or at least 48 cm since 1995. The failure of the 1998 flood event to scour this sediment suggests that the sediment volume from Sessoms Creek is greater that the ability of even a major flood of the San Marcos River to erode and transport in the concrete lined main channel and farther downstream. A potentially beneficial impact of the flood control project is the enhanced recharge to the Edwards Aquifer that occurs behind the USMWRFCD dams. Water that would have flowed down the San Marcos River as quickflow is now infiltrated and contributes to the water that is withdrawn by wells or emerges from San Marcos Springs. EARL & WOOD 83 The actual amount of enhanced recharge is a function of the mean annual runoff estimate. Interpolation between the 5 1 and 127 mm runoff lines on the runoff map produced by the USGS in 1970 produces an estimated 76 mm runoff for the region (USGS 1970). If runoff is calculated from the nearby gaged discharge/drainage basin area relation¬ ships, the regional runoff is approximately 127 mm (USGS 1921; 1956- 1999). Lastly, if the regional annual runoff equation (Qj = 10° 211 A0 846) developed by Lanning-Rush (2000) is employed, the 217 km2 above the dams have an estimated 287 mm runoff. If one uses the value of 127 mm runoff calculated from the adjacent basins, the mean annual runoff from the controlled 217 km2 regulated portion of the upper San Marcos basin is 27 million m3 (22,000 acre feet) / year. Other than aquifer seepage, the only natural outlet for this water is San Marcos Springs. If all of this enhanced recharge were to become springflow, it would increase the average baseflow of the river by 0.8 m3/sec (30 ft3/sec) or about 18%. Unfortunately, this recharge is dependent upon precipitation and this enhanced flow could not be expected during drought times when little surface runoff is produced. Looking at this enhanced recharge in terms of water supply, this additional water, based upon a value of $0.37 to $1. 10/ m3 ($100 to $300/acre foot), is worth between $2.2 and 6.6 million per year. Presently, Texas law does not grant ownership to the party responsible for enhanced recharge, but the volatile status of Texas water law forced upon a reluctant state by increasing demand for water and periodic droughts will probably clarify the ownership of enhanced recharge (Kaiser 1986; Votteler 1998). Once this issue is resolved, the increased groundwater would become a resource for future water demand for San Marcos or could generate revenues. Management Options The altered hydraulic regimen of the San Marcos River brought about by the upstream flood control project can be accepted and adjusted to, or can be modified. Already the City and SWT have implemented a hydrilla "mowing" program to reduce vegetation clogging (Anonymous 1994). Other than accepting the sedimentation, a channel dredging program has been proposed and could be implemented. Dredging would raise legal concerns over the “taking” of endangered species, and consequently, other options need to be considered to shift the sediment supply /stream energy relationship. A different stream regime with some combination of increased total stream energy or a reduction of sediment supply is necessary to reverse the persistent sedimentation. Dam design 84 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 1, 2002 and operation could be modified to increase the amount of flow passed through the flood control dams on Sink Creek. On the other side of the equation, the amount of sediment entering the river could be reduced by enhanced efforts to reduce erosion and or implementation of a series of sediment check dams immediately upstream of where high sediment load tributaries enter the San Marcos River. The failure of the 1998 storm to produce a net scour of sediment deposited since 1991 suggests that with the present stabilized channel, the only means to reduce/reverse the sedimentation problems is to reduce sediment yield from the contributing watersheds. An unresolved issue is the near autonomous status of SWT in terms of its actions affecting the river. University land constitutes 192 of the 940 ha of the drainage of Sessoms Creek, which includes the potential to cause considerable changes or damage to the river such as the leaky gasoline storage tank incident of 1994 proved (Dreckman 1994). City, county and even some state agencies cannot force SWT to modify its actions to protect the river because it is a state entity. This presents a fundamental problem for management in that such a significant landuser is immune from rules that are designed to protect such a valuable resource. With implementation of the next set of federal Water Quality Act (1987) rules in 2003, SWT will formally be required to treat its runoff from all new developments and be required to implement sedi¬ ment control measures from all construction sites larger than 0.4 ha (1 acre) (U.S. Environmental Protection Agency 2000). Perhaps, enforce¬ ment of these rules on the SWT campus and elsewhere in the watershed of the upper San Marcos River watershed will significantly reduce the hydrologically overwhelming sediment load presently introduced into the river. Discussion The recent changes in the San Marcos River illustrate a number a principles in environmental management. Hydrologically, the dams have reduced the magnitude and frequency of scouring flood events. Landuse changes downstream of the flood control dams have increased sediment yield so that the channel has adjusted through deposition. The reduction in peak flood energy has also led to in increase in exotic vegetation. These changes mimic those observed in the Grand Canyon of the Colorado following the closure of Glen Canyon Dam (Graf 1985; Carothers & Brown 1991). The trunk stream is unable to remove the sediment that is deposited at the mouths of the tributaries and there is EARL & WOOD 85 insufficient energy to flush out undesirable vegetation. Perhaps, a program as dramatic as the recent peak flood simulation in the Grand Canyon would be required (Webb et al. 1999). Public authorities would be reluctant to create or increase a flood that would further endanger lives and increase property damage for some uncertain benefit to the river. Sediment control through expanded implementation of the Water Quality Act (1987) will help with the sediment problem but other measures, perhaps more costly or controversial, will be required to deal with the increase in exotic vegetation. The recent changes in the San Marcos River brought about by the upstream flood control project illustrate the hydrologic and ecologic principle that reducing flood discharges establishes an entirely new stream regime that is often associated with a new set of management problems. Acknowledgments We would like to thank Jeff Wilson, Karim Aziz and Andy Skadberg for helping with the maps and data analysis for this report. Melanie Howard, the San Marcos River steward, provided data for this report. John Dudik, Ryan Kainer and Barry Kolarik, past and present SWT students, assisted with the fieldwork and data analysis. Literature Cited Adamietz, G. 1999. An analysis of the October 1998 flood and the 100-year FEMA floodplain along the San Marcos River using DOQQ imagery. Unpublished Masters of Applied Geography Directed Research, Department of Geography, Southwest Texas State University, San Marcos, Texas, 40 pp. Angerstein, M. B. & D. E. Lemke. 1994. First records of the aquatic weed Hyrophila polysperma (Acanthaceae) from Texas. SIDA, 16(2):365-371 . Anonymous. 1994. Mower keeps river clear while protecting rare wild rice. American Town & City, September 1994, p. 80. Baker, V. R. 1975. Flood Hazards along the Balcones Escarpment in central Texas: Alternative approaches to their recognition, mapping and management. University of Texas, Bureau of Economic Geology, Circular, 75-5, 22 pp. Carothers, S. W. & B. T. Brown. 1991. The Colorado River through the Grand Canyon: Natural history and human change. University of Arizona Press, Tucson, xix+235 pp. Chow, V. T., D. Maidment & L. Mays. 1988. Applied Hydrology. McGraw-Hill Inc., New York, xiii+572 pp. City of San Marcos. 1995. San Marcos City Code. Municipal Code Corporation, Tallahassee, Florida, xxi-f 665 pp. City of San Marcos. 1996. San Marcos Horizons: City Master Plan. City of San Marcos, Planning and Development Services Department, San Marcos, Texas, xviii + 252 pp. Dalrymple, T. & M. A. Benson. 1967. Measurement of Peak Discharge by the Slope-Area Method. Chapter A2, Book 3 U.S. Geological Survey Techniques on Water-Resources Investigations, 12 pp. Dingman, S. L. 1996. Physical Hydrology. Prentice Hall, Englewood Cliffs, New Jersey, 86 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 1, 2002 xiv + 575 pp. Dreckman, M. 1994. SWT Continues Gasoline Leak Probe and Clean-up. San Marcos Daily Record, June 16, 1994, p.l. Dunne, T. & L. B. Leopold. 1978. Water in Environmental Planning. W.H. Freemand and Co. San Francisco, xxvii + 818 pp. Edwards Aquifer Research and Data Center. 1999. Aftermath— Addressing the south central Texas Floods of October 1998. Edwards Aquifer Research and Data Center, Southwest Texas State University: San Marcos, Texas, 38 pp. Goudie, A. 1990. Geomorphological Techniques, second edition. George Allen and Unwin, London, xx + 570 pp. Graf, W. L. 1985. The Colorado River: Change and basin management. Association of American Geographers, Washington, DC, vi + 86 pp. Greater San Marcos Economic Development Council. 2000. San Marcos, Texas Community Profile. Greater San Marcos Economic Development Council, San Marcos, Texas, 4 pp. Groeger, A. W., P. F. Brown, T. E. Tictzen & T. C. Kelsey. 1997. Water Quality of the San Marcos River. Texas J. of Sci., 49(4): 279-294. Hanson, J. A. & T. A. Small. 1995. Geologic framework and hydrogeologic characteristics of the Edwards Aquifer Outcrop, Hays County, Texas. U.S. Geological Survey Water- Resources Investigations Report 95-4265, 10 pp. Herschfield, D. M. 1961. Rainfall Frequency Atlas of the United States. Technical Paper No. 40, U.S. Weather Bureau, Department of Commerce, 111 pp. Kaiser, R. A. 1986. Handbook of Texas water law: Problems and needs. Texas Water Resources Institute. College Station, Texas, 46 pp. Lanning-Rush, J. 2000. Regional equations for estimating mean annual and mean season runoff for natural basins in Texas, base period 1961-1990. U.S. Geological Survey Water- Resources Investigations Report 00-4064, 27 pp. Lemke, D. E. 1989. Aquatic macrophytes of the upper San Marcos River, Hays Co., Texas. Southwestern Naturalist, 34:289-291. Lower Colorado River Authority WaterCo-River Management. 1999. Storm Report: Lower Colorado River, Central and Southeast, Texas, October 1998 flood event, Austin, Texas. Lower Colorado River Authority, 23+pp. Mays, K. B., K. S. Saunders, R. E. Moss & K. Aziz. 1996. Effects of changes in Spring Lake Headgate elevation on water levels in the San Marcos River, Hays County, Texas. Texas Parks and Wildlife Department, Resources Protection Division, River Studies Report No. 13, 31 pp. McGehee, A. S. 1982. A River Reflects on Peppers Past. Peppers at the Falls, San Marcos, Texas (Accessed at Southwest Texas State University Library Archives, San Marcos, Texas), 13 pp. Miller, A. 1996. A dirty secret: sedimentation. San Marcos Daily Record, September 6, 1996, pp. 1-2. Pulich, W. , S. Perry & D. German. 1994. Habitat and Land Use Inventory and Change Detection Analysis of the San Marcos River Corridor. Pp. 11-33, in The San Marco River: A Case Study (Robert W. Spain, Project Coordinator). Texas Parks and Wildlife Department, Austin, Texas. U.S. Environmental Protection Agency, Region 6, Cooperative Agreement Number X-006603-01-0. Sands, S. 1998. Weather Report for the City of San Marcos. City of San Marcos Public Electic Utility, San Marcos, Texas. Saunders, K. S. 1992. An Annotated Bibliography of the San Marcos River, its Source and Tributaries. Texas Parks and Wildlife Department, Austin, Texas. U.S. Environmental Protection Agency, Region 6, Cooperative Agreement No. X-006603-01-0, iii-f-77 pp. EARL & WOOD 87 Slade, R. M., Jr. 1986. Large rainstorms along the Balcones Escarpment in Central Texas. Pp. 86-91, in The Balcones Escarpment, Geological Society of America Annual Meeting Guidebook (Abbott, P.L. & C.M. Woodruff, Jr., eds), Geological Society of America, Boulder, Colorado. Slade, R. M., Jr., W. H. Asquith & G. D. Tasker. 1995. Multiple-regression equations to estimate peak-flow frequency for streams in Hays County, Texas. U.S. Geological Survey Water Resources Investigations Report 95-4019, 4 pp. Southwest Texas State University. 2001. Southwest Texas At-A-Glance: Tables and Statistics, Fall 2000. Southwest Texas State University, San Marcos, Texas, 26 pp. Spain, R. W. 1994. The San Marcos River: A Case Study. Texas Parks and Wildlife Department, Austin, Texas. U.S. Environmental Protection Agency, Region 6, Cooperative Agreement Number X-006603-01-0, 169 pp. Staton, L. L. 1992. Assessment of the changes in the macrophyte community in the upper San Marcos River. Unpublished M.S. thesis, Southwest Texas State University, San Marcos, Texas, 74 pp. Stovall, F. 1986. San Marcos, The Town. Pp. xix + 1-195, in Clear Springs and Limestone Ledges, A History of Hays County, Texas (Stovall, F., G. Johnson, D. Schwartz, & D.W. Kerbow, eds.). Hays County Historical Commission, San Marcos, Texas. Texas Parks and Wildlife Department. 1993. Computerized Element Occurrences of Special Concern in Travis, Hays and Williamson Counties. Texas Parks and Wildlife Department, Austin, Texas. Upper San Marcos Watershed Reclamation and Flood Control District. 1991. Protecting People, Property and the Environment: The Upper San Marcos Watershed Reclamation and Flood Control District, 1971-1991. The Upper San Marcos Watershed, Reclamation and Flood Control District. San Marcos, Texas, 17 pp. U.S. Bureau of the Census. 2001 . United States Census 2000. Basic Facts > Quick Tables, QT-PL Race, Hispanic or Latino, and Age 2000, San Marcos city, Texas. http://factfinder.census.gov/bf. U.S. Department of Agriculture Natural Resources Conservation Service. 1999. Storm Report: October 17-19, 1998 Central Texas Flood. U.S. Department of Agriculture Natural Resources Conservation Service, Temple, Texas, 17 pp. U.S. Department of Agriculture Soil Conservation Service. 1976. Erosion & Sediment Control Guidelines for Developing Areas in Texas. U.S. Department of Agriculture Soil Conservation Service, Temple Texas, x + 165 pp. U.S. Department of Agriculture Soil Conservation Service. 1978. Watershed Plan and Environmental Impact Statement USDA-SCS-EIS-WS-(ADM)-78-2-(F)-(TX): Upper San Marcos River Watershed, Comal and Hays Counties, Texas. U.S. Department of Agriculture Soil Conservation Service, Temple, Texas, iv + 139 pp. U.S. Department of Agriculture Soil Conservation Service. 1986. Urban Hydrology for Small Watersheds. Soil Conservation Service, Engineering Division Technical Release 55, Second Edition, iv + 160 pp. U.S. Environmental Protection Agency, Office of Wastewater Management. 2000. Phase II of the NPDES Storm Water Program, 4 pp. http : //www . epa . go v/O WM/sw/phase2/index . htm . U.S. Geological Survey. (1956-1974). Surface Water Supply of the United States, Volume 8. U.S. Geological Survey Water Supply Papers. U.S. Geological Survey. 1970. The National Atlas of the United States of America. Washington, DC: GPO, xiii + 417 pp. U.S. Geological Survey. (1971-1999). Water Resources Data— Texas, Volume 3. Surface Water. U.S. Geological Survey Water-Data Report TX-(year)-3. U.S. Geological Survey. 1999. Floods in the Guadalupe and San Antonio River Texas, 88 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 1, 2002 October 1998. U.S. Geological Survey Fact Sheet FS-147-99, 7 pp. U.S. National Climate Data Center. 2002. Daily Precipitation, San Marcos, Texas, December 1, 1896-November 30, 2001. http://www.ncdc.noaa.gov. Votteler, T. H. 1998. The Little Fish that Roared: The Endangered Species Act, State Groundwater Law, and Private Property Rights Collide over the Edwards Aquifer. Environmental Law 28:845-880. Webb, R. H., J. C. Schmidt, F. R. Marzolf & R. A. Valdez, eds. 1999. The Controlled Flood in Grand Canyon. American Geophysical Union Geophysical Monograph 1 10, 367 pp. Wood, C. R. & S. Gilmer. 1996. San Marcos River Stewardship Research Findings and Progress Report for 1995. City of San Marcos Parks and Recreation Department, San Marcos, Texas, 16+ pp. RAE at re02@swt.edu TEXAS J. SCI. 54(1), FEBRUARY, 2002 89 GENERAL NOTES THE GHOST-FACED BAT, MORMOOPS MEGALOPHYLLA, (CHIROPTERA: MORMOOPIDAE) FROM THE DAVIS MOUNTAINS, TEXAS Robert S. DeBaca and Clyde Jones Department of Biological Sciences and the Museum Texas Tech University Lubbock, Texas 79409-3131 The ghost-faced bat (. Mormoops megalophylla) is broadly distributed in the tropics of southwestern North America and a portion of northern South America. The North American range of this species includes the southern three- fourths of Baja California, mainland Mexico and northern Central America (Hall 1981; Cameron 1993). The northern distribu¬ tional limit of M. megalophylla reaches southeastern Arizona and southern Texas near the Rio Grande (Cameron 1993). In Trans-Pecos Texas, M. megalophylla has been recorded in only Brewster, Presidio and Culberson counties (Davis & Schmidly 1994). This new record from Jeff Davis County fills a gap in distribution between northern, extralimital records for the ghost- faced bat at Elephant Mountain in Brewster County (Bradley et al. 1999) and the Apache Mountains in Culberson County (Stangl et al. 1994), which are, respectively, 72 km SE and NW of the new record. On 4 October 2000, two female ghost- faced bats (one non-lactating adult and one sub-adult) were captured in a 12 m, four- tiered mist net set across Limpia Creek at Davis Mountains State Park (DMSP) (eleva¬ tion 1522 m, UTM coordinates: 13 603632E 3385791N). The first ghost- faced bat came in at 2122 hr, which was about 90 minutes after sunset, and the second came in at 2220 hr. The sky was clear, the temperature was 24 °C, and the wind blew about 15 km/h during the times of capture. Eighty-four other bats (one Myotis velifer , 22 Nyctinomops macrotis and 61 Tadarida brasiliensis) flew into this single net between 2010 hr and 2300 hr. The net was located over an open, slow moving, ephemeral pool that was 13 m wide by about 55 m long. The site of capture was in a broad, sloping canyon with a mosaic of oak-juniper and montane grassland that had been invaded by shrubs. 90 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 1, 2002 Yancey (1997) recorded M. megalophylla as the second most common bat at Big Bend Ranch State Park, which indicates a healthy population of this species 83 km south of the Davis Mountains. Furthermore, this record for the Davis Mountains is the third northern, extralimital record for the species in less than 20 years. These data support the idea that the ghost-faced bat is undergoing an expansion of range northward (Bradley et al. 1999). However, the lack of breeding individuals among these records contradicts this evidence. Alternatively, these records might be of individuals searching for food or water outside the normal range for the species during times of drought. Additionally, this net site was about 250 m upstream from a locality where Jones et al. (1999) recorded the first capture of the western yellow bat, Lasiurus xanthinus, from the Davis Mountains. In contrast to this Mormoops locality, that site was fed by a permanent spring and supported mature, riparian woodland that was nearly closed canopied. These records are too few to draw many comparisons between the two species at the two sites. However, the captured individuals occurred in habitats with potential roost sites respective to each species. The skins, skulls, post-cranial skeletons and frozen tissues (TK 92913, 92914) of these two specimens are deposited in the Collection of Recent Mammals in the Natural Science Research Laboratory, the Museum of Texas Tech University. The materials are cataloged under the numbers TTU 82461 and TTU 82462. Acknowledgments These specimens were obtained from DMSP in accordance with scientific collecting permit SPR-0790-189, which was issued by the Texas Parks and Wildlife Department. Kelly Bryan, Linda Hedges and John Holland facilitated the overall project at DMSP. Ernest W. Valdez helped prepare the specimens, and Linda Hedges and Kelly Bryan helped characterize the habitat. This project was in accordance with a partnership between the Texas Parks and Wildlife Department, the Nature Conservancy of Texas, and Texas Tech University. Literature Cited Bradley, R. D., D. S. Carroll, M. L. Clary, C.W. Edwards, I. Tiemann-Boege, M. J. Hamilton, R. A. Van Den Bussche & C. Jones. 1999. Comments on some small mammals from the Big Bend and Trans-Pecos regions of Texas. Occ. Pap. Mus., Texas Tech Univ., 193:1-6. Cameron, G. N. 1993. Mormoops megalophylla. Mamm. Species, 448:1-5. TEXAS J. SCI. 54(1), FEBRUARY, 2002 91 Davis, W. B., & D. J. Schmidly. 1994. The mammals of Texas. Texas Parks and Wildlife Dept., Austin, 338 pp. Hall, E. R. 1981. The mammals of North America. John Wiley & Sons, New York, 2nd ed., 1:1-600 + 90. Jones, C., L. Hedges & K. Bryan. 1999. The western yellow bat, Lasiurus xanthinus (Chiroptera: Vespertilionidae) , from the Davis Mountains, Texas. Texas J. Sci., 51(3):267-269. Stangl, F. B., W. W. Dalquest & R. R. Hollander. 1994. A 10,000-year history of mammals from Culberson and Jeff Davis Counties, Trans-Pecos, Texas. Midwestern State Univ. Press, Wichita Falls, xix + 264 pp. Yancey, F. D., II. 1997. The mammals of Big Bend Ranch State Park, Texas. Spec. Publ. Mus., Texas Tech Univ., 39:210 pp. CJ at: cjones@packrat.musm.ttu.edu 92 THE TEXAS JOURNAL OF SCIENCE- VOL. 54. NO. 1, 2002 THE TEXAS ACADEMY OF SCIENCE http : //hsb . bay lor . edu/html/tas/ Membership Information and Application MEMBERSHIP.— Any person or member of any group engaged in scientific work or interested in the promotion of science is eligible for membership in The Texas Academy of Science. (Please print or type) Name Last First Middle MailingAddress City State Zip Ph FAX E-mail: Regular Member $30.00 Supporting Member $60.00 Sustaining Member $100.00 Patron Member $150.00 Student-U ndergraduate $15.00 Student-Graduate $15.00 Joint $35.00 Emeritus $10.00 Corporate Member $150.00 Affiliate $5.00 (list name of organization) Contribution AMOUNT REMITTED SECTIONAL INTEREST AREAS: A. Biological Science B. Botany C. Chemistry D. Computer Science E. Conservation & Management F. Environmental Science G. Freshwater & Marine Sciences H. Geography I. Geology K. Mathematics L. Physics M. Science Education O. Systematics & Evolutionary Biology P. Terrestrial Ecology Q. Anthropology R. Threatened or Endangered Species Please indicate your Sectional interest(s) below: 1. _ 2. _ 3. _ Send Application Form and Check or Money Order to: Dr. Fred Stevens, Executive Secretary The Texas Academy of Science CMB 5980 Schreiner University Kerrville, Texas 78028-5697 Please photocopy this Application Form AUTHOR GUIDELINES 93 INSTRUCTIONS TO AUTHORS Scholarly manuscripts reporting original research results in any field of science or technology will be considered for publication in The Texas Journal of Science. Prior to acceptance, each manuscript will be reviewed both by knowledgeable peers and by the editorial staff. Authors are encouraged to suggest the names and addresses of two potential reviewers to the Manuscript Editor at the time of submission of their manuscript. No manuscript submitted to the Journal is to have been published or submitted elsewhere. Excess authorship is dis¬ couraged. Manuscripts listing more than four authors will be returned to the corresponding author. Upon completion of the peer review process, the corresponding author is required to submit two letter quality copies of the final revised manuscript as well as a diskette (3.5 inch) copy. Format Except for the corresponding author’s address, manuscripts must be double-spaced throughout (including legends and literature cited) and submitted in TRIPLICATE (typed or photocopied) on 8.5 by 11 inch bond paper, with margins of approximately one inch and pages numbered. Scientific names of species should be placed in italics. Words should not be hyphenated. The text can be subdivided into sections as deemed appropriate by the author(s). Possible examples are: Abstract; Methods and Materials; Results; Discussion; Summary or Conclusions; Acknowledgments; Literature Cited. Major internal headings are centered and capitalized. Computer generated manuscripts must be reproduced as letter quality or laser prints. References References must be cited in the text by author and date in chronological (not alphabetical) order; Jones (1971); Jones (1971 ; 1975); (Jones 1971); (Jones 1971; 1975); (Jones 1971; Smith 1973; Davis 1975); Jones (1971); Smith (1973); Davis (1975); Smith & Davis (1985); (Smith & Davis 1985). Reference format for more than two authors is Jones et al. (1976) or (Jones et al. 1976). Citations to publications by the same author(s) in the same year should be designated alphabetically (1979a; 1979b). 94 THE TEXAS JOURNAL OF SCIENCE- VOL. 54. NO. 1, 2002 Literature Cited Journal abbreviations in the Literature Cited section should follow those listed in BIOSIS Previews ® Database (ISSN: 1044-4297). All libraries receiving Biological Abstracts have this text; it is available from the interlibrary loan officer or head librarian. Otherwise standard recognized abbreviations in the field of study should be used. All citations in the text must be included in the Literature Cited section and vice versa. Consecutively-paged journal volumes and other serials should be cited by volume, number and pagination. Serials with more than one number and that are not consecutively paged should be cited by number as well. The following are examples of a variety of citations: Journals & Serials. — Jones, T. L. 1971. Vegetational patterns in the Guadalupe Mountains, Texas. Am. J. Bot., 76(3): 266-278. Smith, J. D. 1973. Geographic variation in the Seminole bat, Lasiurus seminolus J. Mammal., 54(l):25-38. Smith, J. D., & G. L. Davis. 1985. Bats of the Yucatan Peninsula. Occas. Pap. Mus., Texas Tech Univ., 97:1-36. Books. — Jones, T. L. 1975. An introduction to the study of plants. John Wiley & Sons, New York, 386 pp. Jones, T. L., A. L. Bain & E. C. Burns. 1976. Grasses of Texas. Pp. 205-265, in Native grasses of North America (R. R. Dunn, ed.), Univ. Texas Studies, 205:630 pp. Unpublished. — Davis, G. L. 1975. The mammals of the Mexican state of Yucatan. Unpublished Ph.D. dissertation, Texas Tech Univ. , Lubbock, 396 pp. In the text of the manuscript, the above unpublished reference should be cited as Davis (1975) or (Davis 1975). Unpublished material that cannot be obtained nor reviewed by other investigators (such as unpublished or unpublished field notes) should not be cited. AUTHOR GUIDELINES 95 Graphics, Figures & Tables Every table must be included as a computer generated addendum or appendix of the manuscript. Computer generated figures and graphics must be laser quality and camera ready, reduced to 5.5 in. (14 cm) in width and no more than 8.5 in. (20.5 cm) in height. Shading is unacceptable. Instead, different and contrasting styles of crosshatching, grids, line tints, dot size, or other suitable matrix can denote differences in graphics or figures. Figures, maps and graphs should be reduced to the above graphic measurements by a photographic method. A high contrast black and white process known as a PMT or Camera Copy Print is recommended. Authors unable to provide reduced PMT’s should submit their originals. Do not affix originals or PMT’s to a cardboard backing . Figures and graphs which are too wide to be reduced to the above measurements may be positioned sideways. They should then be reduced to 9 in. (23 cm) wide and 5 in. (12.5 cm) in height. Black and white photographs of specimens, study sites, etc. should comply with the above dimensions for figures. Color photographs cannot be processed at this time. Each figure should be marked on the back with the name of the author(s) and figure number and top of figure. All legends for figures and tables must be typed (double-spaced) on a sheet(s) of paper separate from the text. All figures must be referred to in text as "Figure 3" or "(Fig. 3)"; all tables as "Table 3" or "(Table 3)". Galley Proofs & Reprints The corresponding author will receive galley proofs prior to the final publishing of the manuscript. Proofs must be corrected and returned to the Managing Editor within five days; failure to return corrected galley proofs promptly will result in delay of publication. The Academy will provide 100 reprints without charge for each feature article or note published in the Journal. These will be mailed to the corresponding author or the designated contact person following the publishing of each issue of the Journal. The distribution of reprints among coauthors is the responsibility of the cor¬ responding author. Authors will have the opportunity to purchase additional reprints (in lots of 100) at the time that the corrected galley proofs are returned to the Managing Editor. Voucher Specimens When appropriate, such as new records, noteworthy range extensions, or faunal or floral listings for an area, the author(s) should provide proper information (to include accession numbers) relative to the deposition of voucher specimens. Specimens should be placed with the holdings of a recognized regional or national museum or herbarium. The name(s) and designated initials used by the museum should be given as part of the introduction or methods section. Do not site the deposition of voucher specimens in personal collections. 96 THE TEXAS JOURNAL OF SCIENCE- VOL. 54. NO. 1, 2002 The Editorial Staff is very aware that many members of the Academy work with organisms that are protected by state or federal regulations. As such, it may not be possible to collect nor deposit these specimens as vouchers. In the interest of maintaining credibility, authors are expected to provide some alternate means of verification such as black and white photographs, list of weights or measurements, etc. The Editorial Staff retains the option to determine the validity of a record or report in the absence of documentation with a voucher specimen. Page Charges Authors are required to pay $50 per printed page. While members of the Academy are allowed four published pages per year free of charge on publications with one or two authors, all authors with means or institutional support are requested to pay full page charges. Full payment is required for all pages in excess of four. All publications authored by three or four persons will incur full page charges. Nonmembers of the Academy are required to pay full page charges for all pages. The Academy, upon written request, will subsidize a limited number of pages per volume. These exceptions are, however, generally limited to students without financial support. Should a problem arise relative to page charges, please contact: Dr. Ned E. Strenth TJS Managing Editor Department of Biology Angelo State University San Angelo, Texas 76909 E-mail: ned.strenth@angelo.edu Additional Guidelines An expanded version of the above author guidelines which includes instructions on style, title and abstract preparation, deposition of voucher specimens, and a listing of standardized abbreviations is available on the Academy’s homepage at: http : //hsb . bay lor . edu/html/tas/ A hard copy is also available upon request from: Dr. Patrick L. Odell TJS Manuscript Editor Institute of Statistics Baylor University - P.O. Box 97225 Waco, Texas 76798 E-mail: Pat_Odell@baylor.edu THE TEXAS ACADEMY OF SCIENCE, 2001-2002 OFFICERS President: President Elect'. Vice-President: Immediate Past President : Executive Secretary: Corresponding Secretary: Managing Editor: Manuscript Editor: Treasurer: AAAS Council Representative: DIRECTORS 1999 Norman V. Horner, Midwestern State University John A. Ward, Brooke Army Medical Center 2000 Bobby L. Wilson, Texas Southern University John P. Riola, Texaco Exploration 2001 David S. Marsh, Angelo State University Felipe Chavez-Ramirez, International Crane Foundation SECTIONAL CHAIRPERSONS Anthropology: Roy B. Brown, Institute Nacional de Antropologia y Historia Biological Science: David S. Marsh, Angelo State University Botany: Cyndy Galloway, Texas A&M University-Kingsville Chemistry: Mary A. Kopecki-Fjetland, St. Edward’s University Computer Science: John T. Sieben, Texas Lutheran University Conservation and Management: Felipe Chavez-Ramirez, International Crane Foundation Environmental Science: Haydn A. Fox, Texas A&M University-Commerce Freshwater and Marine Science: Joseph L. Kowalski, University of Texas-Pan American Geology and Geography: James W. Westgate, Lamar University Mathematics: William D. Clark, Stephen F. Austin State University Physics: David L. Bixler, Angelo State University Science Education: Julie F. Westerlund, Southwest Texas State University Systematics and Evolutionary Biology: Allan Hook, St. Edward’s University Terrestrial Ecology: Monte Thies, Sam Houston State University Threatened or Endangered Species: Donald L. Koehler, Austin Parks and Recreation Dept. COUNSELORS Collegiate Academy: Jim Mills, St. Edward’s University Junior Academy: Vince Schielack, Texas A&M University Nancy Magnussen, Texas A&M University David R. Cecil, Texas A&M University-Kingsville Larry D. McKinney, Texas Parks and Wildlife Department John T. Sieben, Texas Lutheran University Thomas L. Arsuffi, Southwest Texas State University Fred Stevens, Schreiner University Deborah D. Hettinger, Texas Lutheran University Ned E. Strenth, Angelo State University Patrick L. Odell, Baylor University James W. Westgate, Lamar University Sandra S. West, Southwest Texas State University SMITHSONIAN INSTITUTION LIBRARIES 3 9088 01569 1959 THE TEXAS JOURNAL OF SCIENCE PERIODICAL POSTAGE PrinTech, P. O. Box 43151 PAID AT LUBBOCK Lubbock, Texas 79409-3151 TEXAS 79402 RETURN SERVICE REQUESTED T^X [Off THE TEXAS JOURNAL OF SCIENCE GENERAL INFORMATION MEMBERSHIP.— Any person or member of any group engaged in scientific work or interested in the promotion of science is eligible for membership in The Texas Academy of Science. For more information regarding membership, student awards, section chairs and vice-chairs, the annual March meeting and author instructions, please access the Academy’s homepage at: http : //hsb . baylor . edu/html /tas/ Dues for regular members are $30.00 annually; supporting members, $60.00; sustaining members, $100.00; patron members, $150.00; associate (student) members, $15.00; family members, $35.00; affiliate members, $5.00; emeritus members, $10.00; corporate members, $250.00 annually. Library subscription rate is $50.00 annually. The Texas Journal of Science is a quarterly publication of The Texas Academy of Science and is sent to most members and all subscribers. Payment of dues, changes of address and inquiries regarding missing or back issues should be sent to: Dr. Fred Stevens, Executive Secretary The Texas Academy of Science CMB 5980 Schreiner University Kerrville, Texas 78028-5697 E-mail: FStevens@schreiner.edu AFFILIATED ORGANIZATIONS American Association for the Advancement of Science, Texas Council of Elementary Science Texas Section, American Association of Physics Teachers Texas Section, Mathematical Association of America Texas Section, National Association of Geology Teachers Texas Society of Mammalogists The Texas Journal of Science (ISSN 0040-4403) is published quarterly at Lubbock, Texas, U.S. A. Periodicals postage paid at San Angelo, Texas and additional mailing offices. POSTMASTER: Send address changes and returned copies to The Texas Journal of Science, Dr. Fred Stevens, CMB 5980, Schreiner University, Kerrville, Texas 78028-5697, U.S. A. The known office of publication for The Texas Journal of Science is the Department of Biology, Angelo State University, San Angelo, Texas 76909; Dr. Ned E. Strenth, Managing Editor. THE TEXAS JOURNAL OF SCIENCE Volume 54, No. 2 May, 2002 CONTENTS The Effect of the Number of Sensors on the Magnitude of the Maximum Observed Ozone. By D. Dorsett and N. Miserendino . . . . . 99 A Preliminary Checklist of the Fishes of Caddo Lake in Northeast Texas. By Clark Hubbs . . . Ill Mortality of Black Bass Captured in Three Fishing Tournaments on Lake Amistad, Texas. By Gene R. Wilde, David H. Larson, William H. Redell and Gene R. Wilde, III . 125 Theropod Dinosaur Trackways in the Lower Cretaceous (Albian) Glen Rose Formation, Kinney County, Texas. By Jack V. Rogers, II . . . . . . . 133 Reproduction in the Coachwhip, Masticophis flagellum (Serpentes: Colubridae), from Arizona. By Stephen R. Goldberg . . . 143 Mitochondrial DNA Analysis of Gene Flow Among Six Populations of Collared Lizards ( Crotaphytus collaris ) in West Central Texas. By James H. Campbell and J. Kelly McCoy . 151 Evaluation of Facilitated Succession at Las Palomas Wildlife Management Area in South Texas. By Frank W. Judd, Robert /. Lonard and Gary L. Waggerman . . 163 Minimum Flow Considerations for Automated Storm Sampling on Small Watersheds. By R. Daren Harmel, Kevin W. King, June E. Wolfe and H. Allen Torbert . . . . . . 177 General Notes Reexamination of the Range for the Northern Pygmy Mouse, Baiomys taylori (Rodentia: Muridae), in Northeastern Texas. By Joel G. Brant and Robert C. Doxvler . 189 THE TEXAS JOURNAL OF SCIENCE EDITORIAL STAFF Managing Editor: Ned E. Strenth, Angelo State University Manuscript Editor: Patrick L. Odell, Baylor University Associate Editor for Botany: Janis K. Bush, The University of Texas at San Antonio Associate Editor for Chemistry: John R. Villarreal, The University of Texas-Pan American Associate Editor for Computer Science: Nelson Passos, Midwestern State University Associate Editor for Environmental Science: Thomas LaPoint, University of North Texas Associate Editor for Geology: Ernest L. Lundelius, University of Texas at Austin Associate Editor for Mathematics and Statistics: E. Donice McCune, Stephen F. Austin State University Associate Editor for Physics: Charles W. Myles, Texas Tech University Manuscripts intended for publication in the Journal should be submitted in TRIPLICATE to: Dr. Patrick L. Odell TJS Manuscript Editor Institute of Statistics Baylor University - P.O. Box 97225 Waco, Texas 76798 Pat_Odell@bay lor . edu Scholarly papers reporting original research results in any field of science, technology or science education will be considered for publication in The Texas Journal of Science. Instructions to authors are published one or more times each year in the Journal on a space-available basis, and also are available from the Manuscript Editor at the above address. They are also available on the Academy’s homepage at: http://hsb.baylor.edu/html/tas/ The Texas Journal of Science is published quarterly in February, May, August and November for $30 per year (regular membership) by The Texas Academy of Science. Periodical postage rates (ISSN 0040-4403) paid at Lubbock, Texas. Postmaster: Send address changes and returned copies to Dr. Fred Stevens, Executive Secretary, CMB 5980, Schreiner University, Kerrville, Texas 78028- 5697, U.S.A. TEXAS J. SCI. 54(2):99-l 10 MAY, 2002 THE EFFECT OF THE NUMBER OF SENSORS ON THE MAGNITUDE OF THE MAXIMUM OBSERVED OZONE D. Dorset! and N. Miserendino Department of Information Systems Baylor University, Waco, Texas 76798 Abstract.— For environmental studies of pollution, the number of sensors involved can vary considerably. This paper investigates the effect when two regions with the same level of ozone but differing number of sensors are measured. The method indicated in the analysis allows for the adjustment of measurement levels for different number of sensors. It seems reasonable that the larger a geographical region the greater the chance that somewhere within that region one will find a high concentration of ozone. Similarly, a greater number of observation stations (sensors) within the region should also increase the chance that one out of the set of sensors will observe a high ozone concentration. The analysis in this paper is motivated by comments by a statistician and meteorologist at the Texas Natural Resources Conservation Commis¬ sion and by the work of David P. Chock, Research Laboratory, Ford Motor Company, Dearborn, Michigan and his research associates. The principal purpose of this paper is to define a relationship between the expected maximum observed daily ozone (03) and the number of sensors and the number of observations per sensor in the region. The analysis is relatively simple but hopefully sufficiently useful and applicable when extended to more realistic and operational situations. Also, if one is convinced that regulations for improving air quality are better applied and understood when standards are written with respect to mean concentration and the variability of concentrations and not with respect to maximum observed concentrations, then one can use a method indicated by the analysis formulated in this paper to relate expected high level ozone when various number of sensors are being used to monitor air quality to the desired mean level of ozone allowed. The method allows one to adjust the estimates from two regions having the same mean level of ozone, yet different numbers of sensors, which are judged regions with unequal concentration with respect to observed maximum ozone. Conversely, areas with different observed levels of ozone can be judged the same with respect to maximum ozone levels when the number of sensors in each region are different. 100 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 2, 2002 Currently, two regions with the same mean level of ozone, and same standard deviation of ozone may not be judged equal due solely to the fact that there are different numbers of sensors in each of the two regions. The greater number of sensors in a region increases the chance for a greater number of exceedances in that region even though the distribution of concentrations of the two regions are the same. The analysis formulated in this paper is based on the fundamental concepts of order statistics (Sarban & Greenbing 1962; Gibbons 1971; Hogg & Craig 1978). Mathematical Preliminaries Let Xl9 X2,..., Xn denote a random sample from a pdf f^x)\ the cumulative pdf is denoted by F^x) = Pr[X < x], and X(1) < X(2) < ...< X{n) denote the sample ordered with respect to magnitude. For con¬ venience of notation, let Yt = X(i), Then Yn = max[2Q. The pdf of Yn is (1) and if f^x) is known, then E[Yn\ and V[Yn\ (the mean and variance) could be computed. If f^x) is not known, approximate techniques are available. For purposes here a normality assumption is made for f^x) to obtain illustrative numerical values (Gibbons 1971). The asymptotic formulas are as follows: (2) (3) Suppose thatZ, = ( Xfiix)/ax ~ M0,1), /=1,2,..., n, then (4) DORSETT & MISERENDINO 101 -2 (5) where and Fz = [Z fz (Z)dz> J- 00 ^Knaxl = /** + °X^Knax] 5 (6) V[XmJ=o2xV[ZmJ. (7) In Table 1, the values of E[Zmax\ and V[Zmax] are tabulated when Z ~ A(0,1). The entries in Table 1 are used later to calculate £[2^] and V[Xmax] when Xt ~ These quantities appear in Table 2. Thus, suppose there are eleven sensors and one observation per sensor (n= 11), one would expect E[Zmax] = 1.38, or 1.38 sigma’s above the mean E[Z\ = 0. The variance of Zmax is 0.250, or the standard devia- tion of Zmax is Wn,J = {HZmjr=0.05. A 2-sigma interval estimate for Zmax is 1.38 - 2(.05) < Zmax < 1.38+2(.05), or 1.28 < Zmax < 1.48. And finally, if X, ~ N(. 060, (,05)2), then £[Xmax] = . 060 +1.38(.05) V[Xmax]=(.05)2(.25) = .000625, ££>[*_] = 025. A two-sigma interval estimate for X„lax is or, in ppb, 79 ppb < Xmax < 1 19ppb. 102 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 Table 1. The Mean, E[Z n ] , and Variance, V[Zn], of the Max Z, , t= 1 ,2, _ «, when Z~N{0,\) versus the number of observations ( n ). n E[Zd] nzH] n E[Z n] VTZJ 1 0 .523 23 1.73 .201 2 .431 .420 24 1.75 .199 3 .675 .371 25 1.76 .196 4 .841 .339 26 1.78 .194 5 .967 .317 27 1.80 .192 6 1.06 .300 28 1.81 .190 7 1.15 .286 29 1.83 .188 8 1.22 .275 30 1.84 .187 9 1.28 .265 31 1.86 .185 10 1.33 .257 32 1.87 .183 11 1.38 .249 33 1.89 .182 12 1.42 .243 34 1.90 .180 13 1.46 .237 35 1.91 .178 14 1.50 .232 36 1.92 .177 15 1.53 .227 37 1.93 .176 16 1.56 .223 38 1.94 .175 17 1.59 .219 39 1.96 .174 18 1.62 .216 40 1.97 .172 19 1.64 .212 41 1.98 .171 20 1.66 .209 42 1.99 .170 21 1.69 .206 43 2.00 .169 22 1.71 .204 44 2.01 .168 The Main Results Table 3 lists several locations of interest, the number of sensors at each location, and the approximate sizes of the region defining the location. Figure 1 gives an indication of how ozone varies diurnally by graphing the amounts of ozone by hour of the day. This shows that maximal ozone occurs between 1300 and 1800 (Cleveland et al. 1976). Hence one can expect four observations per station to be especially relevant for the analysis here. Using Table 1 and equations (6) and (7), Table 2 gives the £[Xmax] for n= 1 ,...44 and for fjix= 0.06, 0.08 and 0.10, or = 60, 80, and 100 ppb, and ax= 0.02, 0.04, and 0.06, or =20, 40, and 100 ppb. DORSETT & MISERENDINO 103 Table 2. for various values of /xx a°d ax versus number of observations. 8 ii b ff=40 o=60 n <7=60 Q 11 00 o <7=100 a = 60 o 00 II b (7=100 a =60 o 00 II b (7= 100 1 60 80 100 60 80 100 60 80 100 2 68 88 108 77 97 117 85 105 125 3 73 93 113 87 107 127 100 120 140 4 76 96 116 93 113 133 110 130 150 5 79 99 119 98 118 138 118 138 158 6 81 101 121 102 122 142 124 144 164 7 83 103 123 106 126 146 129 149 169 8 84 104 124 108 128 148 133 153 173 9 85 105 125 111 131 151 136 156 176 10 86 106 126 113 133 153 140 160 180 11 87 107 127 115 135 155 142 162 182 12 88 108 128 117 137 157 145 165 185 13 89 109 129 118 138 158 147 167 187 14 90 110 130 120 140 160 150 170 190 15 90 110 130 121 141 161 152 172 192 16 91 111 131 122 142 162 153 173 192 17 91 111 131 123 143 163 155 175 195 18 92 112 132 124 144 164 157 177 197 19 92 112 132 125 145 165 158 178 198 20 93 113 133 126 146 166 160 180 200 21 93 113 133 127 147 167 161 181 201 22 94 114 134 128 148 168 162 182 202 23 94 114 134 129 149 169 163 183 203 24 95 115 135 130 150 170 165 185 205 25 95 115 135 130 150 170 166 186 206 26 95 115 135 131 151 171 167 187 207 27 96 116 136 132 152 172 168 188 208 28 96 116 136 132 152 172 169 189 209 29 96 116 136 133 153 173 170 190 210 30 96 116 136 133 153 173 170 190 210 31 97 117 137 134 154 174 171 191 211 32 97 117 137 135 155 175 172 192 212 33 97 117 137 135 155 175 173 193 213 34 98 118 138 136 156 176 174 194 214 35 98 118 138 136 156 176 174 194 214 36 98 118 138 137 157 177 175 195 215 37 98 118 138 137 157 177 176 196 216 38 98 118 138 137 157 177 176 196 216 39 99 119 139 138 158 178 177 197 217 40 99 119 139 138 158 178 178 198 218 41 99 119 139 139 159 179 178 198 218 42 99 119 139 139 159 179 179 199 219 43 100 120 140 140 160 180 180 200 220 44 100 120 140 140 160 180 180 200 220 Table 3. Locations, number of sensors and sizes. Size of Region Location 1990-91 1992-93 (sq. miles) Houston 11 11 1777 Dallas/ FW 5 7 1808 Austin 2 2 1000 San Antonio 2 2 1247 El Paso 3 4 526 104 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 ♦ 0 to 155 langleys ■ 155 to 335 langleys A Above 335 langleys Hour of the Day Figure 1. Dirunal curves of Bayonne 03 concentrations. Days stratified according to the total solar radiation in Central Park, New York City from 7 a.m. to 7 a.m. (in langleys). For example, Houston has eleven sensors with four observations of interest per day per sensor, giving a total of 44 observations each day to order with respect to magnitude. The expected maximum ozone read¬ ing can be read from Table 2 for the various situations. Also, the variation in the distribution of the maximum value due to different numbers of observations are illustrated in Figures 2 and 3. Currently, a day in which Xmax> 120 ppb is declared an exceedance day or ozone episode. The probability of observing and exceedane day is also a function of the number of observations. Using this as the cutoff, and the distributions in Table 2, Table 4 gives the />[^nax> 120] ppb. The effect of increasing the number of observations from which the maximum is determined is clearly seen under varying distributions. Some Suggested Strategies There are several reasonable ways to write air quality (ozone) regula¬ tions, some based on the average of several hourly ozone measurements, where s is the number of sites (sensors) in the region and nt is the total number of observations per site. Usually the are constant for a region. DORSETT & MISERENDINO 105 Figure 2. Comparison of probability density functions for A^ax, when n = 4, 1 1 and 44, and [xx = 80, ox = 20. Figure 3. Probability of an exceedance judgement as a function of N. Thus, a strategy to define an ozone episode is: If X > xc for some critical value xc; that is, if a daily average (or a partial daily average) X exceeds a pre-assigned amount xc, then that day is said to be an ozone exceedance day or an episode has occurred. 106 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 2, 2002 Table 4. P[^as > 120] ppb. n Q II O a = 40 a = 60 n o 00 II ^=ioo fix= 60 o 00 II fxx= 100 /V=60 o 0Q II =£ „, = 100 1 0.001 0.067 0.159 0.023 0.159 0.252 0.159 0.309 0.369 2 0.003 0.129 0.292 0.045 0.292 0.441 0.292 0.522 0.602 3 0.004 0.187 0.404 0.067 0.404 0.582 0.404 0.669 0.749 4 0.005 0.242 0.499 0.088 0.499 0.688 0.499 0.771 0.842 5 0.007 0.292 0.578 0.109 0.578 0.767 0.578 0.842 0.900 6 0.008 0.340 0.645 0.129 0.645 0.826 0.645 0.891 0.937 7 0.009 0.384 0.702 0.149 0.702 0.870 0.702 0.924 0.960 8 0.011 0.425 0.749 0.168 0.749 0.903 0.749 0.948 0.975 9 0.012 0.463 0.789 0.187 0.789 0.927 0.789 0.964 0.984 10 0.013 0.499 0.822 0.206 0.822 0.946 0.822 0.975 0.990 11 0.015 0.533 0.850 0.224 0.850 0.959 0.850 0.983 0.994 12 0.016 0.564 0.874 0.241 0.874 0.970 0.874 0.988 0.996 13 0.017 0.593 0.894 0.259 0.894 0.977 0.894 0.992 0.998 14 0.019 0.620 0.911 0.275 0.911 0.983 0.911 0.994 0.998 15 0.020 0.646 0.925 0.292 0.925 0.987 0.925 0.996 0.999 16 0.021 0.669 0.937 0.308 0.937 0.990 0.937 0.997 0.999 17 0.023 0.691 0.947 0.324 0.947 0.993 0.947 0.998 1.000 18 0.024 0.712 0.955 0.339 0.955 0.995 0.955 0.999 1.000 19 0.025 0.731 0.962 0.354 0.962 0.996 0.962 0.999 1.000 20 0.027 0.749 0.968 0.369 0.968 0.997 0.968 0.999 1.000 21 0.028 0.766 0.973 0.383 0.973 0.998 0.973 1.000 1.000 22 0.029 0.782 0.978 0.397 0.978 0.998 0.978 1.000 1.000 23 0.031 0.796 0.981 0.411 0.981 0.999 0.981 1.000 1.000 24 0.032 0.810 0.984 0.424 0.984 0.999 0.984 1.000 1.000 25 0.033 0.822 0.987 0.437 0.987 0.999 0.987 1.000 1.000 26 0.035 0.834 0.989 0.450 0.989 0.999 0.989 1.000 1.000 27 0.036 0.845 0.991 0.463 0.991 1.000 0.991 1.000 1.000 28 0.037 0.856 0.992 0.475 0.992 1.000 0.992 1.000 1.000 29 0.038 0.865 0.993 0.487 0.993 1.000 0.993 1.000 1.000 30 0.040 0.874 0.994 0.499 0.994 1.000 0.994 1.000 1.000 31 0.041 0.883 0.995 0.510 0.995 1.000 0.995 1.000 1.000 32 0.042 0.891 0.996 0.521 0.996 1.000 0.996 1.000 1.000 33 0.044 0.898 0.997 0.532 0.997 1.000 0.997 1.000 1.000 34 0.045 0.905 0.997 0.543 0.997 1.000 0.997 1.000 1.000 35 0.046 0.911 0.998 0.553 0.998 1.000 0.998 1.000 1.000 36 0.047 0.917 0.998 0.563 0.998 1.000 0.998 1.000 1.000 37 0.049 0.923 0.998 0.573 0.998 1.000 0.998 1.000 1.000 38 0.050 0.928 0.999 0.583 0.999 1.000 0.999 1.000 1.000 39 0.051 0.933 0.999 0.592 0.999 1.000 0.999 1.000 1.000 40 0.053 0.937 0.999 0.602 0.999 1.000 0.999 1.000 1.000 41 0.054 0.941 0.999 0.611 0.999 1.000 0.999 1.000 1.000 42 0.055 0.945 0.999 0.620 0.999 1.000 0.999 1.000 1.000 43 0.056 0.949 0.999 0.628 0.999 1.000 0.999 1.000 1.000 44 0.058 0.952 1.000 0.637 1.000 1.000 1.000 1.000 1.000 Concluding Remarks Consider for example the locations Houston, Dallas/FW and El Paso. Houston has eleven sensors, Dallas/FW has seven and El Paso has four. Then let the average ozone be /jlx = 60, 80, and 100 ppb, and the standard deviation ox = 20 ppb. The following values in Table 5 were obtained using Table 2. DORSETT & MISERENDINO 107 Table 5. Expected Maximum (ox ppb) — 20 ppb). Location jxx = 60 o 0© II 5 o o II 5 Houston («= 11) 115 135 155 11.6 D/FW (n= 7) 106 126 146 10.7 El Paso (n=4) 93 113 136 10.0 For example, if fix — 60 and ox = 20 ppb and one assumes that the pdf of observations is normal, then Pr[20 < X < 100] = 0.95 for all three locations, but Pr[Xmax > 120 1 Houston] * 0.337, Pr[Xmax > 120|D/FW] i 0.096, and Pr[Xmax > 120 1 El Paso] ® 0.004. The probabilities are approximations and the last three are not equal since the number of sensors at each location are not equal ; even though the ozone distributions are the same for each location. The average of ozone measurements for each location will be "close" to 60 ppb at each of the three locations, but the expected maximum and the probabilities of an ozone episode varies due solely to the unequal number of sensors at the different locations. Conversely, consider two regions, one with eleven sensors (like Houston) and one with two sensors (like Austin) and suppose that E[X\ = 80 ppb and ox = 40 at both Houston and Austin. Then if Xmax = 135 at Houston and Xmax = 107 at Austin, Houston is declared in exceedance and Austin is not, even though Austin's air quality is the same as that of Houston's. These results are from Table 2. It is important to note that in this case, Houston would be in exceedance while Austin would not, yet both would be experiencing the same distribution of ozone in the sense that E[X\ Houston] = E[X\ Austin] = 80 ppb and ox = 40 ppb. It is true that if the concentrations of ozone are bounded by the true largest value (which we denote by the parameter, ft), then the greater number of sensors will give a better estimate for the actual largest value, ft . For example let the observations be distributed uniform on the interval (0, ft); that is, the pdf of X is fx(x) = 1/ft, 0 95%) largemouth bass M. salmoides , but some spotted bass M. punctulatus also were captured. All tournament-caught fish held for observation of delayed mortality, as well as control fish, were largemouth bass; therefore, results of this study can be interpreted as providing estimates of tournament-associated mortality for largemouth bass. Water temperature appears to be the single most important factor influencing mortality of tournament-caught fish (Wilde 1998). However, several other factors are known to affect mortality of black bass captured and released in fishing tournaments: number and size of fish captured (Meals & Miranda 1994; Weathers & Newman 1997; Wilde et al. 2002), number of participants (Schramm et al. 1985; 1987; Hartley & Moring 1995; Ostrand et al. 1999), live- well conditions (Plumb et al. 1988; Kwak & Henry 1995), and organizational effects and tournament rules (Kwak & Henry 1995; Weathers & Newman 1997; Wilde et al. 1998b; Ostrand et al. 1999). 130 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 2, 2002 Table 1. Initial, delayed and total mortality of black bass captured in three fishing tournaments at Lake Amistad, Texas, during August 1998 through March 1999. Date No. of teams Water temp. (°C) No. of fish caught No. of tournament fish held for observation Total pre-release mortality (%) Delayed mortality (%) Total mortality (%) SE total mortality (%) 8-9 Aug. 1998 527 27.9 2409 64 8.2% 61.6 64.8 5.60 19-20 Sept. 1998 187 25.6 504 62 5.8% 44.0 47.2 9.16 27-28 Mar. 1999 241 16.8 506 64 0.3% 0.0 0.3 0.00 Among the tournaments studied at Lake Amistad, a combination of high water temperatures, overland transportation in live wells, and the holding period of 2-4 hours prior to release most likely are responsible for the high mortality observed in the August 1998 tournament. Unex¬ pectedly high mortality in the September 1998 tournament may be related to the depth at which fish were captured. Total mortality among fish captured on the first day of this tournament (31.6%) was slightly greater than that predicted (24.4%) based on water temperature; how¬ ever, total mortality among fish captured on the second day of the tournament (73.5%) was substantially greater than predicted (24.4%). Tournament participants reported that most fish captured on the second day of the tournament were taken from waters deeper than 9 m. Black bass captured from this depth would be expected to show signs of depressurization sickness (Feathers & Knable 1983), which include a distended abdomen and swollen air bladder that can cause death. Obser¬ vation of black bass floating at the lake surface, struggling to submerge, at the release site is evidence of depressurization sickness among these fish. Overland transport of fish in the August tournament and depres¬ surization sickness in the September tournament probably account for the large discrepancies between observed mortality and that predicted by Wilde’s (1998) regression models. Tournament anglers’ estimates of total mortality for the August 1998 tournament ranged from 2 to 95% (mean = 29.7%, SE = 2.76, N = 77). Logistic regression of the acceptability of different rates of mortality to anglers was highly significant (Fig. 1. x2 = 1 1.096, df = 1, P < 0.0001). Based on the logistic regression, 28% total mortality (predicted based on water temperature at the time of the tournament) would be acceptable to only 33% of tournament anglers, 30% total mortality (the average of anglers estimates) would be accepted to 28% of tournament anglers, and 65% total mortality (observed) would be acceptable to only 10% of tournament anglers. WILDE ETAL. 131 Total Mortality (%) Figure 1. Logistic regression showing the proportion of anglers that found as acceptable various rates of total mortality (TM) of tournament-caught black bass. Equation for the regression is: Acceptable = e0463 ' 0 041 *TM / (1 + eU463 “ 0 041 *TM), N = 77, X2 = 11.096, df= 1, P < 0.0001). Total mortality observed at two of the three black bass fishing tournaments studied were near (September, 47.2%) or exceeded (August, 64.8%) 50%, a level that was acceptable to fewer than 20% of tournament anglers. This level of mortality also is consistent with the belief held by many nontournament anglers that most tournament- caught fish do not survive (Wilde et al. 1998a). Although there remains some question whether tournament-associated mortality is harmful to black bass populations (e.g., Schramm et al. 1991), results of this study suggest the need for continued improvement of tournament rules and procedures to reduce mortality to a level acceptable to both tournament and nontournament anglers. Acknowledgments We thank Bill Sontag, Todd Brindle and Joe Kraai for logistical support; Jimmy Dean, John Dennis and Bob Zerr, for help in collecting control fish; Tim Bonner, Ken Ostrand, Kevin Pope and Gregory Wilde for help in various portions of the study; and Ron Hilliard, Dane Widner and Harvey Holmes for assistance in collecting information at their tournaments. We also thank David O’Keefe and other Del Rio, Texas, sportsmen for partial financial support of this work. This is contribution 132 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 T-9-000 of the College of Agricultural Sciences and Natural Resources, Texas Tech University. Literature Cited Duttweiler, M. W. 1985. Status of competitive fishing in the United States: trends and state fisheries policies. Fisheries, 10(5): 5-7. Feathers, M. G. & A. E. Knable. 1983. Effects of depressurization upon largemouth bass. N. Am. J. Fish. Manage., 3(l):86-90. Hartley, R. A. & J. R. Moring. 1995. Differences in mortality between largemouth and smallmouth bass caught in tournaments. N. Am. J. Fish. Manage., 15(3):666-670. Holbrook, J. A., II. 1975. Bass fishing tournaments. Pp. 408-414, in Black bass biology and management (H. Clepper, ed.). Sport Fishing Institute, Washington, D.C., 534 pp. Kwak, T. J. & M. G. Henry. 1995. Largemouth bass mortality and related causal factors during live-release fishing tournaments on a large Minnesota lake. N. Am. J. Fish. Manage., 15(3): 62 1-630. Meals, K. O. & L. E. Miranda. 1994. Size-related mortality of tournament-caught largemouth bass. N. Am. J. Fish. Manage., 14(2):460-463. Ostrand, K. G., G. R. Wilde, D. W. Strickland & M. I. Muoneke. 1999. Initial mortality in Texas black bass fishing tournaments. N. Am. J. Fish. Manage., 19(4): 1 124-1 128. Plumb, J. A., J. M. Grizzle & W. A. Rogers. 1988. Survival of caught and released largemouth bass after confinement in live wells. N. Am. J. Fish. Manage., 8(3): 324-328. Schramm, H. L., Jr., P. J. Haydt & N. A. Bruno. 1985. Survival of tournament-caught largemouth bass in two Florida lakes. N. Am. J. Fish. Manage., 5(4) : 606-61 1 . Schramm, H. L., Jr., P. J. Haydt & K. M. Porter. 1987. Evaluation of prerelease, postrelease, and total mortality of largemouth bass caught during tournaments in two Florida lakes. N. Am. J. Fish. Manage., 7(4): 394-402. Schramm, H. L., Jr., M. L. Armstrong, N. A. Funicelli, D. M. Green, D. P. Lee, R. E. Manns, Jr., B. D. Taubert & S. J. Waters. 1991. The status of competitive fishing in North America. Fisheries, 1 6(3) :4- 1 2. Shupp, B. D. 1979. 1978 status of bass fishing tournaments in the United States. Fisheries, 4(6): 1 1-19. Weathers, K. C. & M. J. Newman. 1997. Effects of organizational procedures on mortality of largemouth bass during summer tournaments. N. Am. J. Fish. Manage., 17(1): 131- 135. Wilde, G. R. 1998. Tournament-associated mortality in black bass. Fisheries, 23(10): 12-22. Wilde, G. R., R. K. Riechers & R. B. Ditton. 1998a. Differences in attitudes, fishing motives, and demographic characteristics between tournament and nontournament black bass anglers in Texas. N. Am. J. Fish. Manage., 18(2): 422-431 . Wilde, G. R., D. W. Strickland, K. G. Ostrand & M. I. Muoneke. 1998b. Characteristics of Texas black bass fishing tournaments. N. Am. J. Fish. Manage., 18(4):972-977. Wilde, G. R., C. E. Shavlik & K. L. Pope. 2002. Initial mortality of black bass in B.A.S.S. fishing tournaments. N. Am. J. Fish. Manage., Vol. 22 (in press). GRW at: gene.wilde@ttu.edu TEXAS J. SCI. 54(2): 133-142 MAY, 2002 THEROPOD DINOSAUR TRACKWAYS IN THE LOWER CRETACEOUS (ALBIAN) GLEN ROSE FORMATION, KINNEY COUNTY, TEXAS Jack V. Rogers, II Department of Geological Sciences Southern Methodist University Dallas, Texas 75275 Abstract.— Two parallel theropod dinosaur trackways are preserved in the Albian Glen Rose Formation of Kinney County in southwest Texas. Although track size and depth indi¬ cate that one of the trackways represents a larger, heavier dinosaur, track morphology of the two trackways is similar. The tracks are referred to Grallator sp. Gregarious behavior is suggested by the direction of travel and uniform spacing of the trackways; however, speed estimates calculated utilizing stride and footprint length indicate that the trackmakers were moving at different speeds. Dinosaur trackways are common in Texas, known from at least 50 locations (Pittman 1992). One of the more famous of the Texas trackways is now on display at the American Museum of Natural History. This trackway, recovered from the bed of the Paluxy River in what is now Dinosaur Valley State Park in Somervell County, Texas, has been interpreted to document the attack of a predatory theropod, probably Acrocanthosaurus , on its sauropod prey (Thomas & Farlow 1997). This unique interpretation illustrates how behavior can be inferred from trackway analyses. Trackways can provide evidence for gregarious behavior in dinosaurs, which is usually inferred from the occurrence of a number of tracks oriented in the same direction. This is commonly seen in sauropod track sites, such as the Davenport Ranch in Bandera County, Texas, that records the movements of a herd of at least twenty-three sauropods (Lockley & Hunt 1995). Similar records of herding behavior are reported for herbivorous bipedal ornithopod dinosaurs (Ostrom 1972). Upper Cretaceous Western Interior Seaway coastal plain deposits of the Dakota Group in New Mexico and Colorado contain trackways of ornithopod herds in such abundance that the region is known as the Dinosaur Freeway (Lockley & Hunt 1995). Trackway evidence for gregarious behavior among theropod dinosaurs is less common, usually inferred from the occurrence of parallel tracks of two or more theropods found oriented in the same general direction as non-theropod (prey) dinosaur tracks. Lockley (1991) reported the occurrence of trackways 134 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 2, 2002 from the Late Cretaceous of Bolivia that apparently represent a group of theropods following a sauropod herd. Dinosaur Valley State Park in Texas has a series of trackways that have been interpreted as evidence of three theropods following a herd of 12 sauropods (Lockley 1991). Farlow (1987) interpreted these trackways as showing only a single theropod following the sauropods. Paired theropod dinosaur trackways are exposed near the top of the Glen Rose Formation in an ephemeral branch of Live Oak Creek along the eastern edge of Kinney County in southwest Texas (Fig. 1). The site, SMU loc. 330, was made available for this study by Mr. Tom Master son. This study documents the trackways, which were produced by two theropod dinosaurs moving in the same direction, maintaining a consistent spatial separation, and taking strides of equal length. Materials and Methods The site was worked during two visits in 1999-2000. Excavation uncovered additional tracks of the easternmost trackway (2c and 2d). The tracks were individually measured, photographed and mapped in accordance with Lockley & Hunt (1995). Site location data is on file at the Shuler Museum of Paleontology, Southern Methodist University, Dallas, Texas (SMU loc. 330). Geological Setting The Glen Rose Formation crops out along with other Lower Creta¬ ceous sediments in a sinusoidal northeast to southwest trending band across the center of Texas and into Oklahoma. The Glen Rose Forma¬ tion consists of a wedge of limestones, dolomites and sandstones, repre¬ senting a variety of depositional environments ranging from transitional shoreline tidal flats and marshes to open marine. This diversity is the result of the transgressive/regressive nature of encroaching Albian seas that deposited the Trinity Group prior to the initial completion of the Western Interior Seaway (Pittman 1992). The Trinity Group is com¬ posed of the Twin Mountains, Glen Rose and Paluxy Formations. The Glen Rose is underlain by the terrigenous elastics of the Twin Moun¬ tains, with which it has a gradational contact, and is unconformably overlain by the Paluxy, a package of loosely consolidated sediments that ranges from continental elastics to deltaic and beach deposits. In north central Texas the Glen Rose Formation wedge pinches out and disap¬ pears, and the merged Twin Mountains and Paluxy Formations are termed the Antlers Formation (Hayward & Brown 1967). ROGERS 135 Figure 1. (a) Map of the SMU loc. 330 theropod trackway site. The trackways are preserved on a bedding plane that was exposed by intermittent runoff erosion. An undetermined portion of trackway 2 is covered by overburden, (b) Texas map showing location of SMU loc. 330 (scale bar equals 2 meters). The Glen Rose is subdivided into three members, the lower, middle (or Thorp Spring) and the upper member (Davis 1974). Many track¬ ways in the Glen Rose Formation are distributed within the top of the upper Glen Rose Formation, extending over a large area representing an ancient coastal plain (Lockley & Hunt 1995), including the trackways at SMU loc. 330 (Pittman 1992). The Albian age of the upper Glen Rose is derived from ammonite biostratigraphic zones (Jacobs & Winkler 1998). Results and Discussion Two parallel tridactyl trackways are preserved, indicating a direction of movement of 182°. The western trackway, designated trackway 1, 136 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 2, 2002 Figure 2. A well preserved left footprint (le) from trackway 1. Note that the middle digit (III) is turned medially. A claw impression is preserved on digit II. Each alternating dark and light scale bar increment equals one centimeter. consists of eight tracks. The centerline of trackway 2 lies 270 centi¬ meters to the east. Trackway 2 consists of four tracks, two of which were uncovered during this study. The tracks represent a theropod morphology commonly recorded from the Glen Rose Formation. They are distinguished by long, slender toe marks. Phalangeal pads are not well defined. Distal digit imprints are deeper than the mid-foot print. In well-preserved tracks of both track¬ ways, the middle digit (III) is turned medially toward the opposing foot (Fig. 2). Digits II and IV diverge from the longitudinal axis of the foot at about 25 degrees (Table 1). Distinct claw imprints are evident on tracks la, lc, le and 2d. There is no impression of the hallux. A distinct heel imprint is apparent. Preservation and depth among the tracks varies. Tracks la- Id are shallower than the others of that trackway (Table 2). There is no evidence of preserved skin impressions in any track. Prints of trackway 2 are deeper, longer and wider than of trackway 1. Average depth of ROGERS 137 Table 1. Digit divarication angles. Measurements are in compass degrees. DIGITS II-III III-IV II-IV Track Number Divarication angle la 24 24 48 lb 26 28 54 lc 24 25 49 Id 25 25 50 le 24 26 50 If 28 25 53 lg 24 26 50 lh 25 28 53 Mean 25.0 25.9 50.9 2a 28 20 48 2b 24 25 49 2c 24 26 50 2d 25 25 50 Mean 25.3 24.0 49.3 Table 2. Individual tracks dimensions: ments are in centimeters. W = width, L = length, D = depth. All measure- W L D Track Number la 32 50 2 lb 31 50 3 lc 29 49 4 Id 32 38 1 le 33 50 6 If 34 50 5 lg 34 46 6 lh 33 46 5 Mean 32.3 47.4 4.0 2a 37 55 7 2b 34 60 7 2c 36 55 8 2d 35 55 6 Mean 35.5 56.3 7.0 trackway 2 prints is 7 cm, of trackway 1 prints, 4.4 cm. The deeper prints of trackway 2 preserve greater detail of foot morphology (Fig 3), including footpads that presumably correspond with phalangeal joints. Digit divarication angle is similar to trackway 1, about 25 degrees. There is evidence of upward displacement of sediments between joints 138 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 2, 2002 Figure 3. Right footprint (2a) from trackway 2. Note that footpad impressions are more clearly defined in the deeper footprints of trackway two, and there is evidence of sediment displacement. Each alternating dark and light scale bar increment equals one centimeter. and digits not seen in the shallower prints of trackway 1. Trackway 2 prints average 9% wider and 8.7% longer than trackway 1 prints. The two trackmakers were of dissimilar size and weight. Langston (1974) referred theropod tracks from the Glen Rose to the ichnotaxa Irenesauripus (Sternberg 1926), and mentions the theropod Acrocanthosaurus as a trackmaker candidate. Acrocanthosaurus body fossils are reported from the Twin Mountains Formation of Texas (Harris 1998) and the Antlers Formation of Oklahoma (Stovall & Langston 1950; Currie & Carpenter 2000). Pittman (1992) referred tridactyl tracks from the Glen Rose to the theropod ichnotaxa Grallator (Hitchcock 1858), and noted that Irenesauripus displays character states diagnostic of Grallator. Grallator is a bird-like print, characterized by a medially turned digit III, which usually exhibits preserved footpad impressions (Pittman 1992). Eubrontes (Hitchcock 1845) a theropod ichnotaxa typically larger in size than Grallotor but exhibiting similar overall morphology, was originally considered by Olsen (1980) to be synonymous with Grallator. Recent morphometric analyses indicate ROGERS 139 Table 3. Trackway dimensions. Pace length is the distance between the same point on successive footprints of a trackway (left, right); stride length is the distance between the same point on successive same foot footprints (left, left, or right, right); pace angulation measures the angle formed by drawing a line from the most anterior tip of the middle digit of three successive footprints (left, right, left, or right, left, right). All measurements are in centimeters. Dashes indicate non-measured dimensions. Tracks Pace Tracks Pace Tracks Stride Length Angulation Length la, lb 160 la,lb,lc 169 la,lc 321 lb,lc 161 — — — — . lc,ld 162 lc,ld,le 170 lc,le 323 ld,le 161 — — — — le,lf 161 IflgJh 169 — — lf,lg 161 — If, lh 322 lg,lh 161 — — — — Mean 161.0 169.3 322.0 2a, 2b 161 2a, 2b, 2c 155 2a, 2c 322 2b, 2c 161 2b, 2c, 2d 155 2b, 2d 322 2c, 2d 161 — — — — Mean 161.0 155.0 322.0 that the two ichnotaxa exhibit proportional differences that allow them to be differentiated (Olsen et al. 1998). The SMU loc. 330 trackways appear to represent both Grallator (trackway 1) and Eubrontes (trackway 2) . They exhibit variable preservation of a single body morphology caused by disparities in the size and mass of the trackmakers and differences in the substrate. Following Olsen (1998) and Pittman (1992), the trackways are here referred to Grallator , with Irenesauripus considered a Grallator synonym. While the individual SMU loc. 330 tracks exhibit commonly observed theropod morphology, the trackways are parallel for the length of their concurrent exposure, and have identical pace and stride lengths (Table 3) . Mean pace angulation of trackway 2 is more acute than trackway 1 (155° as compared to 169.3°), which indicates a greater displacement between the left and right limbs and provides further evidence of a size differential between the two trackmakers. Hip height of bipedal trackmakers can be estimated using morpho¬ metric ratios derived from measurements of bipedal dinosaur skeletons (Alexander 1976; Thulborn 1989). This hip height estimate can then be combined with stride length to produce a stride length/hip height ratio (X/h). Alexander (1976) demonstrated that in living terrestrial verte- 140 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 brates the X/h ratio for a walking gait is <2.0, for trotting or running > 2.0, and suggested the same was true for dinosaurs. The X/h ratio for SMU loc. 330 trackways 1 and 2 is 1.34 and 1.19 respectively, indicating both trackmakers were moving at a walk. Farlow (1981) estimated speeds for theropod dinosaurs from track¬ ways in the Glen Rose Formation of Kimble County, Texas, using methods from Alexander (1976). Estimates ranged from 1.8 to 11.9 m s'1, with 12 of 15 estimates falling within the 1.8 to 3.4 m s'1 range. Applying this method to the SMU 330 trackway 1 produced an estimated speed of 2.6 m s'1; trackway 2 speed is estimated at 1.8 m s'1. These speeds fall within the walking speed estimates for bipedal dinosaurs provided by Thulborn (1982). Conclusions The SMU loc. 330 trackways were produced by a pair of theropod dinosaurs walking in the same direction. The tracks are referred to the ichnogenus Grallator. The absence of appropriately sized theropod body fossils other than Acrocanthos auras within the Glen Rose Formation and other Trinity Group sediments suggests that this taxon likely made the trackways at SMU loc. 330, a conclusion in concurrence with Farlow (2001). Estimated speed of the trackmakers agrees with previous estimates from Glen Rose Formation trackways. Although the consistent direction and equal spacing of the trackways appear to suggest gregarious be¬ havior, speed estimates for the two trackmakers differ, which indicates that if the trackmakers were moving in concert, they were not doing so at the same speed during the interval represented by the tracks. Varying preservation between the two trackways suggests they may have been made at different times. Acknowledgments Grateful acknowledgment is made of the gracious hospitality of Ann and Tom Masterson of Houston, Texas, who provided access to their property as well as food and lodging to the author during this study. Thanks also to Drs. Dale Winkler and Louis Jacobs of Southern Methodist University who reviewed this manuscript and made helpful suggestions. ROGERS 141 Literature Cited Alexander, R. M. 1986. Estimates of speeds of dinosaurs. Nature, 261:129-130. Currie, P. J. & K. Carpenter. 2000. A new specimen of Acrocanthosaurus atokensis (Theropoda, Dinosauria) from the Lower Cretaceous Antlers Formation (Lower Cretaceous, Aptian) of Oklahoma, USA. Geodiversitas, 22:207-246. Davis, K. W. 1974. Stratigraphy and depositional environments of the Glen Rose Formation, north-central Texas. Baylor Geological Studies, Bulletin 26, 43 pp. Farlow, J. O. 1981. Estimates of dinosaur speeds from a new trackways site in Texas. Nature, 294:747-748. Farlow, J. O. 1987. A guide to Lower Cretaceous dinosaur footprints and tracksites of the Paluxy River Valley, Somerville County, Texas. Field Trip Guidebook, South Central Section, Geological Society of America, Baylor University, Waco, Texas, 50 pp. Farlow, J. O. 2001. Acrocanthosaurus and the maker of Comanchean large theropod Footprints. Pp. 408-427, in Mesozoic Vertebrate Life (D.H. Tanke & K. Carpenter, eds.), Indiana University Press, 577 pp. Harris, J. D. 1989. A reanalysis of Acrocranthosaurus atokensis, its phylogenetic status, and paleobiogeographic implications, based on a new specimen from Texas. New Mexico Museum of Natural History and Science Bulletin 13, 75 pp. Hayward, O. T. & L. F. Brown, Jr. 1967. Comanchean (Cretaceous) rocks of central Texas. Pp. 31-48, in Comanchean (Lower Cretaceous) Stratigraphy and Paleontology of Texas (Hendricks, L., ed.), Society of Economic Paleontologists and Mineralologists Publication No. 67-8, 410 pp. Hitchcock, E. 1845. An attempt to name, classify, and describe the animals that made the fossil footprints of New England. Sixth Annual Meeting of the Association of American Geologists and Naturalists: 23-25. Hitchcock, E. 1858. Ichnology of New England: A report on the Sandstone of the Connecticut Valley, especially its fossil footprints. Natural Sciences of America Reprint. W. White, Boston, 220 pp. Jacobs, L. L. & D. A. Winkler. 1998. Mammals, archosaurs, and the Early to Late Cretaceous transition in north-central Texas. Pp. 253-280, in Advances in Paleontology and Geochronology (Y. Tomida, L. J. Flynn & L. L. Jacobs, eds.), National Science Museum Monographs No. 14, Tokyo, 292 pp. Langston, W., Jr. 1974. Nonmammalian Comanchean tetrapods. Geoscience and Man, 8:77-102. Lockley, M. 1991. Tracking dinosaurs. Cambridge University Press, New York, 238 pp. Lockley, M. & A. P. Hunt. 1995. Dinosaur tracks and other footprints of the western United States. Columbia University Press, New York, 338 pp. Olsen, P. E. 1980. Fossil great lakes of the Newark Supergroup in New Jersey. Pp. 352-398, in Field studies of New Jersey geology and guide to field trips (W. M. Manspeizer, ed.), New York State Geological Association, 52nd Annual Meeting, Rutgers University. Olsen, P. E., J. B. Smith & N. G. McDonald. 1998. Type material of the type species of the classic theropod footprint genera Eubrontes, Anchisauripus, and Grallator (Early Jurassic, Hartford and Deerfield Basins, Connecticut and Massachusetts, U.S.A.). Journal of Vertebrate Paleontology, 18:586-601. Ostrom, J. H. 1972. Were some dinosaurs gregarious? Palaeogeography, Palaeo- climatology, Palaeoecology, 11:287-301. Pittman, J. G. 1992. Stratigraphy and vertebrate ichnology of the Glen Rose Formation, Western Gulf Basin, USA. Unpublished Ph.D. dissertation, The University of Texas at Austin, 726 pp. 142 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 Sternberg, C. M. 1926. Dinosaur tracks from the Edmonton Formation of Alberta. Canadian Geological Survey Bulletin XLIV: 73-84. Stovall, J. W. & W. Langston, Jr. 1950. Acrocanthosaurus atokensis, a new genus and species of Lower Cretaceous Theropoda from Oklahoma. American Midland Naturalist, 43:696-728. Thomas, D. A. & J. O. Farlow. 1997. Tracking a dinosaur attack. Scientific American, December: 74-79. Thulborn, R. A. 1982. Speeds and gaits of dinosaurs. Palaeogeography, Palaeo- climatology, Palaeoecology, 38:227-256. Thulborn, R. A. 1989. The gaits of dinosaurs. Pp. 39-50 in Dinosaur tracks and traces (D. D. Gillette & M. G. Lockley, eds.), Cambridge University Press, New York, 454 pp. JVR at: jack.rogers@attbi.com TEXAS J. SCI. 54(2): 143-150 MAY, 2002 REPRODUCTION IN THE COACHWHIP, MASTICOPHIS FLAGELLUM (SERPENTES: COLUBRIDAE), FROM ARIZONA Stephen R. Goldberg Department of Biology, Whittier College Whittier, California 90608 Abstract.— Reproductive tissue was examined from 145 sexually mature Masticophis flagellum museum specimens from Arizona. Males follow a seasonal testicular cycle in which individuals undergoing spermiogenesis were found April to November. Males with regressed testes were found March-July and December. Males with testes in recrudescence occurred March-August. The female reproductive activity season encompassed April-July. Fifty-three percent of the females sampled from this period were reproductively active. Mean clutch size for nine females was 7.2 + 3.0 SD, range = 2-12. The finding of one female with a clutch size of two is a new minimum clutch size for M. flagellum. The coach whip, Masticophis flagellum ranges through the southern half of the United States from coast to coast, south to the tip of Baja California and Queretaro, Mexico from below sea level to around 2350 m (Stebbins 1985). It frequents a variety of habitats including desert, prairie, scrubland, juniper-grassland, woodland, thornforest and farm¬ land (Stebbins 1985) and has a diurnal activity period (Stebbins 1954). There are many anecdotal reports of reproduction in this species containing information on clutch sizes and mating time in different parts of its range (Force 1930; Brennan 1934; Marr 1944; Clark 1949; Werler 1951; Guidry 1953; Zweifel & Norris 1955; Wright & Wright 1957; Carpenter 1958; Minton 1958; Cunningham 1959; Tennant 1984; Stebbins 1985; Degenhardt et al. 1996). The purpose of this paper is to provide information on the ovarian and testicular cycles of M. flagellum from Arizona from a histological examination of preserved museum specimens. Comparisons are made with the reproductive cycles of other North American species of Masticophis. Materials and Methods A sample of 145 sexually mature specimens of M. flagellum (54 females, mean snout- vent length, SVL = 942 mm ± 102 SD, range = 750-1152 mm; 91 males, SVL = 1003 mm ± 152 SD, range = 642-1527 mm) from Arizona was examined from the herpetology 144 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 collections of Arizona State University, Tempe (ASU), The Natural History Museum of Los Angeles County, Los Angeles (LACM) and The University of Arizona, Tucson (UAZ). Snakes were collected during the years 1950-2000. Over half of the snakes 76/145 (52%) came from Pima County, Arizona. A t test was used to compare means of female and male body sizes (SVL’s). Counts were made of enlarged ovarian follicles (> 12 mm length) or oviductal eggs. The left testis, vas deferens and a portion of the kidney were removed from males; the left ovary was removed from females for histological examination. Slides with tissue sections were stained with Harris’ hematoxylin followed by eosin counterstain. Testis slides were examined to determine the stage of the testicular cycle; ovary slides were examined for the presence of yolk deposition (secondary yolk deposition sensu Aldridge 1979). It was common to observe autolytic changes in the kidneys of road killed specimens, whereas structures in the testis and vas deferens appeared normal. Data on the kidney sexual segment are not presented due to difficulty in distinguishing whether enlargement of kidney tubules from some males was due to reproductive activity or autolytic changes from inadequate fixation or decay prior to fixation (road kills). Because some of the specimens were road kills, not all tissues were available for histological examination due to damage or autolysis. Number of specimens examined by reproductive tissue were: ovary = 54, testis = 91, vas deferens = 79 . Since the M. flagellum samples were from different areas, there is the possibility that reproductive data of large samples from geographic or taxonomic subpopulations may differ distinctly from the pattern described herein. Material examined.— The following sexually mature specimens of Masticophis flagellum from Arizona were examined: COCHISE COUNTY, (UAZ 25167, 25168, 25183, 25529, 25547, 25566, 25568, 25595, 25607, 32434, 33006, 34529, 38097, 41646, 46286, 46529, 46624, 46627, 46649, 50039, 52031); GILA COUNTY, (UAZ 25166); GRAHAM COUNTY, (UAZ 25559, 50337); MARICOPA COUNTY , (ASU 1403, 1675, 14033, 14035, 14093, 14373, 22469, 23617, 23618, 24304, 24382, LACM 125263, UAZ 37430, 40352, 44078); MOHAVE COUNTY, (ASU 24368, LACM 145267, UAZ 40087, 40356, 44859, 44860); LA PAZ COUNTY, (UAZ 35870); PIMA COUNTY, (LACM 64299, 103116, UAZ 23927, 25132, 25169, 25170, 25172, 25175, 25176, 25178, 25185, 25187, 25188, 25191, 25198-25200, 25203, 25213, 25216, 25218, 25219, 25271, 25520, 25521, 25523, 25537, 25539, 25541, 25543, 25544, 25548, 25551, 25556, 25557, 25561, 25562, 25576-25578, 25596, 25598-25600, 25609, GOLDBERG 145 25610, 26834, 28593, 28597, 30225, 30629, 31757, 31763, 41070, 41624-41627, 41632-41634, 41669, 41682, 44256, 44257, 44933, 46413, 48622, 49367, 50059, 50220, 50653, 51709, 51770, 51802, 51907); PINAL COUNTY, (ASU 24391, 28240, 28245, UAZ 25174, 25177, 25209, 25220, 25536, 25603, 30626, 42996, 43003); SANTA CRUZ COUNTY, (UAZ 25173, 25597, 25617, 32155, 49167, 50188); YAVAPAI COUNTY, (ASU 13836, 22139, UAZ 40355); YUMA COUNTY, (UAZ 25190, 25534). Results and Discussion Testicular histology was similar to that reported by Goldberg & Parker (1975) for two colubrid snakes, Masticophis taeniatus and Pituophis catenifer (= P. melanoleucus). In the regressed testes, seminiferous tubules contained spermatogonia and Sertoli cells. In recrudescence, there was renewal of spermatogenic cells characterized by spermatogonial divisions; primary and secondary spermatocytes were typically present. Spermatids were occasionally seen. In spermio- genesis, metamorphosing spermatids and mature sperm were present. Males undergoing spermiogenesis were found April-November. Males with regressed testes were found March-July and December. Males with testes in recrudescence were found March- August (Table 1). The smallest reproductively active male (spermiogenesis, sperm in vas deferens) measured 642 mm SVL. As this was the smallest male examined, there may be yet smaller reproductively mature male M. flagellum in Arizona. Vasa deferentia of 79/79 (100%) males contained sperm: March (5); April (16); May (22); June (7); July (11); August (4); September (5); October (6); November (2); December (1). The presence of sperm in the vasa deferentia suggests M. flagellum has the potential of mating throughout the activity season, although previously reported matings occurred in spring (Wright & Wright 1957; Minton 1958; Degenhardt et al. 1996) and late summer (August) (Zweifel & Norris 1955). Males were significantly larger than females, t = 2.60; 143 df P = 0.01. The testicular cycle of M. flagellum was similar to that of Masticophis bilineatus which was studied by Goldberg (1998). In M. bilineatus , males undergoing spermiogenesis were found in both spring and fall. These cycles differ from those reported for the congeners Masticophis lateralis (Goldberg 1975) and M. taeniatus (Goldberg & Parker 1975) in which spermiogenesis was restricted to fall and testes were regressed 146 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 2, 2002 Table 1. Monthly distribution of stages in the seasonal testicular cycle of Masticophis flagellum from Arizona from examination of 91 adult museum specimens. Values shown are the numbers of males exhibiting each of the three conditions. Month n Regressed Recrudescence Spermiogenesis March 5 1 4 0 April 18 3 13 2 May 27 3 15 9 June 8 1 2 5 July 13 2 4 7 August 4 0 1 3 September 6 0 0 6 October 7 0 0 7 November 2 0 0 2 December 1 1 0 0 in spring, containing spermatogonia and Sertoli cells. Most of the M. bilineatus in Goldberg (1998) were from southern Arizona where both M. flagellum and M. bilineatus may be sympatric. Whether the pro¬ longed periods of spermiogenesis of M. flagellum and M. bilineatus are a response to the climate of southern Arizona (see Lowe 1964) and the resultant food availability from both summer and winter rainfall must await further study. The smallest reproductively active M. flagellum female (enlarged follicles > 12 mm length) measured 781 mm SVL. Females smaller than this size (several were examined) were excluded from the study to avoid including immature females in analysis of the ovarian cycle. There was no suggestion (oviductal eggs and ovarian follicles with yolk deposition in the same female) to suggest M. flagellum produces more than one clutch in a reproductive season. Females with enlarged follicles (> 12 mm length) or oviductal eggs were found April-July (Table 2). Females undergoing early yolk deposition = secondary yolk deposition (sensu Aldridge 1979) were found in April-May and November. It is not known if the yolked follicles of the one November female would have been used in a clutch the following year or would have been reabsorbed (atresia). The timing of the ovarian cycle of M. flagellum from Arizona, with eggs likely being deposited in May through July, appears in synchrony to that which GOLDBERG 147 Table 2. Monthly distributions of stages in the seasonal ovarian cycle of Masticophis flagellum from Arizona from examination of 54 adult museum specimens. Values shown are the numbers of females exhibiting each of the four conditions. Month n Inactive Early yolk deposition Enlarged follicles > 12 mm length Oviductal eggs March 2 2 0 0 0 April 6 3 2 1 0 May 13 5 3 5 0 June 7 3 0 3 1 July 4 3 0 1 0 August 9 9 0 0 0 September 9 9 0 0 0 October 2 2 0 0 0 November 2 1 1 0 0 has been reported in the literature for this species (Force 1930; Brennan 1934; Marr 1944; Clark 1949; Werler 1951; Guidry 1953; Carpenter 1958; Cunningham 1959). Tennant (1984) reported M. flagellum usually deposited eggs in June and July in Texas. Females of three other species of Masticophis had similar periods of female reproductive activity: Masticophis lateralis April -July (Goldberg 1975); M. taeniatus May-July (Parker & Brown 1980); M. bilineatus April-July (Goldberg 1998). Mean clutch size for nine M. flagellum females was 7.2 ± 3.0 SD, range = 2-12. Clutch size data is summarized in Table 3. Sample size is inadequate to determine whether a SVL-clutch size relationship exists for M. flagellum. Fitch (1970) calculated a mean clutch size of 10. 1 for eleven clutches (range 4-16) taken from the literature. Tennant (1984) and Stebbins (1985) gave what are apparently maximum clutch sizes of 20 eggs for M. flagellum. There are reports of clutches of four eggs (Marr 1944; Wright & Wright 1957; Tennant 1984; Stebbins 1985) hence the clutch of two eggs (Table 3) is a new minimum clutch size for M. flagellum. Only a portion of M. flagellum females (16/30) 53 % showed evidence of reproduction (Table 2) during the period of reproductive activity. That not all members of the female population produce eggs in a given year has been reported for other North American colubrid snakes, for 148 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 2, 2002 Table 3. Clutch sizes for 9 Masticophis flagellum from Arizona (estimated from counts of yolked follicles > 12 mm length or oviductal eggs) from museum specimens. Date SVL (mm) Clutch size County Source 29 April 990 10 Pima UAZ 25218 15 May 781 2 Mohave UAZ 40356 16 May 1015 9 Cochise UAZ 33006 22 May 873 _ a Pima UAZ 25578 28 May 1009 4 Pinal UAZ 25220 29 May 1125 12 Pinal UAZ 25536 3 June 985 7 Pima UAZ 25598 4 June 981 _ _b Cochise UAZ 25568 12 June 842 6 Mohave UAZ 44859 13 June 853 8C La Paz UAZ 35870 15 July 855 7 Pima LACM 103116 a Part of clutch was missing. b Snakes with coagulated yolk, follicles could not be counted. c Oviductal eggs, all other females contained enlarged follicles. example see Goldberg (2000a; 2000b) and Goldberg & Rosen (1999). However, in contrast, Parker & Brown (1980) reported 13/14 (93%) M. taeniatus from late May-early July from northern Utah were gravid and that annual reproduction was normal for this species. While some information on snake reproduction can be obtained by histological examination of museum specimens, field studies will be required to elucidate other aspects of M. flagellum reproductive biology such as ages when sexual maturity occurs and frequency of clutch production by females. Since the range of M. flagellum extends through the southern portion of the United States from coast to coast (Stebbins 1985), examination of the reproductive cycle from eastern populations would provide information on variation in reproduction within the same species from different environments. Acknowledgments I thank Michael E. Douglas (formerly of Arizona State University), David A. Kizirian (Natural History Museum of Los Angeles County) and Charles H. Lowe (University of Arizona) for permission to examine M. flagellum . GOLDBERG 149 Literature Cited Aldridge, R. D. 1979. Female reproductive cycles of the snakes Arizona elegans and Crotalus viridis. Herpetologica, 35(3):256-261 . Brennan, L. A. 1934. A check list of the amphibians and reptiles of Ellis County, Kansas. Trans. Kansas Acad. Sci., 37:189-191. Carpenter, C. C. 1958. Reproduction, young, eggs and food of Oklahoma snakes. Herpetologica, 14(2): 1 13-1 15. Clark, R, F. 1949. Snakes of the hill parishes of Louisiana. J. Tenn. Acad. Sci., 24(4): 244-261 . Cunningham, J. D. 1959. Reproduction and food of some California snakes. Herpetologica, 15(1): 17-19. Degenhardt, W. G., C. W. Painter & A. H. Price. 1996. Amphibians and reptiles of New Mexico. Univ. New Mexico Press, Albuquerque, xix + 431 pp. Fitch, H. S. 1970. Reproductive cycles of lizards and snakes. Misc. Publ. Mus. Nat. Hist., Univ. Kansas, 52:1-247. Force, E. R. 1930. The amphibians and reptiles of Tulsa County, Oklahoma, and vicinity. Copeia, 1930(2):25-39. Goldberg, S. R. 1975. Reproduction in the striped racer, Masticophis lateralis (Colubridae). J. Herpetol., 9(4): 36 1-363. Goldberg, S. R. 1998. Reproduction in the Sonoran whipsnake, Masticophis bilineatus (Serpentes: Colubridae). Southwest. Nat., 43(3):412-415. Goldberg, S. R. 2000a. Reproduction in the longnose snake, Rhinocheilus lecontei (Serpentes: Colubridae). Texas J. Sci., 52(4) :3 19-326. Goldberg, S. R. 2000b. Reproduction in the glossy snake, Arizona elegans (Serpentes: Colubridae) from California. Bull. Southern Calif. Acad. Sci., 99(2): 105-109. Goldberg, S. R. & W. S. Parker. 1975. Seasonal testicular histology of the colubrid snakes, Masticophis taeniatus and Pituophis melanoleucus. Herpetologica, 31(3):317-322. Goldberg, S. R. & P. C. Rosen. 1999. Reproduction in the Sonoran shovelnose snake ( Chionactis palarostris) and the western shovelnose snake (Chionactis occipitalis) (Serpentes: Colubridae). Texas J. Sci., 51(2): 153-158. Guidry, E. V. 1953. Herpetological notes from southeastern Texas. Herpetologica, 9(1): 49-56. Lowe, C. H. 1964. Arizona Landscapes and Habitats, Pp. 3-132, in The vertebrates of Arizona. Landscapes and habitats, fishes, amphibians and reptiles, birds, mammals (C. H. Lowe, ed), Univ. Arizona Press, Tucson, 270 pp. Marr, J. C. 1944. Notes on amphibians and reptiles from the central United States. Am. Midi. Nat., 32(2): 478-490. Minton, S. A., Jr. 1958. Observations on amphibians and reptiles of the Big Bend region of Texas. Southwest. Nat., 3(l):28-54. Parker, W. S. & W. S. Brown. 1980. Comparative ecology of two colubrid snakes, Masticophis t. taeniatus and Pituophis melanoleucus deserticola, in northern Utah. Milwaukee Public Museum, Publ. Biol, and Geol. 7 : vii -El- 104. Stebbins, R. C. 1954. Amphibians and reptiles of western North America. McGraw-Hill Book Company, Inc., New York, xxii + 536 pp. Stebbins, R. C. 1985. A field guide to western reptiles and amphibians. Houghton Mifflin Company, Boston, xiv + 336 pp. Tennant, A. 1984. The snakes of Texas. Texas Monthly Press, Austin. 561 pp. Werler, J. E. 1951. Miscellaneous notes on the eggs and young of Texan and Mexican 150 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 reptiles. Zoologica, 36(l):37-48. Wright, A. H. & A. A. Wright. 1957. Handbook of snakes of the United States and Canada, Volume 1, Comstock Publishing Associates, Cornell University Press, Ithaca, New York, xviii + 564 pp. Zweifel, R. G. & K. S. Norris. 1955. Contribution to the herpetology of Sonora, Mexico: descriptions of new subspecies of snakes ( Micruroides euryxanthus and Lampropeltis getulus ) and miscellaneous collecting notes. Am. Midi. Nat., 54(l):230-249. SRG at: sgoldberg@whittier.edu TEXAS J. SCI. 54(2): 151-162 MAY, 2002 MITOCHONDRIAL DNA ANALYSIS OF GENE FLOW AMONG SIX POPULATIONS OF COLLARED LIZARDS {CROTAPHYTUS COLLARIS) IN WEST CENTRAL TEXAS James H. Campbell* and J. Kelly McCoy Department of Biology, Angelo State University Box 10890, San Angelo, Texas 76909 * Current address: Department of Biological Sciences Texas Tech University Lubbock, Texas 79409 Abstract.— Gene flow among six populations of the collared lizard ( Crotaphytus collaris ) was estimated using restriction endonuclease analysis of a segment of the mitochondrial genome. Individuals (/i = 37) were collected from O. C. Fisher Lake (n = 6), Twin Buttes Reservoir (n = 4), E. V. Spence Reservoir (« = 2), Ballinger Municipal Lake {n = 4), Carter Ranch (n — 6) and Sims Ranch (n=15) in Tom Green, Runnels, Coke and Concho counties of west central Texas. Wagner parsimony analysis revealed gene flow between O. C. Fisher Lake and Twin Buttes Reservoir, from Carter Ranch to O. C. Fisher Lake and from Carter Ranch to both Ballinger Municipal Lake and E. V. Spence Reservoir. Sims Ranch individuals displayed no well-supported gene flow affiliations with any other population sampled. It is likely that the Carter and Sims Ranch populations are ancestral to the other populations sampled. Gene flow is an important consideration when examining the popula¬ tion structure of any organism over time. Movement of genetic material between populations of organisms causes them to become more similar in overall genetic composition. Perhaps more importantly, gene flow results in increased genetic heterozygosity that may allow the organisms to better cope with selection pressures. Isolated populations that receive little or no genetic exchange with other populations may experience loss of heterozygosity (inbreeding depression) and encounter a reduced ability to endure selection pressures. Alleles within isolated populations may be driven to fixation through genetic drift or founder effect. However, outbreeding depression can also occur when gene flow disrupts highly specialized adaptive complexes that have been obtained through strong selection pressures associated with some habitats (Templeton et al. 1990). While much work has been done on geographic variation in sexual dimorphism in the collared lizard, Crotaphytus collaris , by McCoy et al. 1994, McCoy 1995, McCoy et al. 1997 and Baird et al. 1997, only limited work has been conducted on gene flow in this species. The 152 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 opportunity for genetic divergence and speciation was postulated in a study by Sexton et al. (1992). This postulate supported the findings of Templeton et al. (1990) when highly fragmented populations of Crotaphytus collaris were studied in the Ozark Mountains. Using both protein and nucleic acid assays, Templeton et al. (1990) noted a lack of gene flow among these isolated populations. Inbreeding depression and genetic drift are postulated to greatly increase the probability of local extinctions in these populations. Templeton et al. (1990) goes on to cite circumstantial evidence for local extinctions that have already occurred. Yoshioka (1996) found evidence of strong gene flow among collared lizards in central Oklahoma populations. In addition, gene flow was stronger among more closely-arranged populations than those distantly spaced. Hranitz & Baird (2000) supported these conclusions in popula¬ tions in central Oklahoma. Moreover, it was found that effective population size was small, and hence, genetic drift could have profound effects on collared lizard populations. However, gene flow was reported to be sufficiently strong to avoid divergence through genetic drift, and therefore, local extinctions (Hranitz & Baird 2000). The current study is the first attempt to measure gene flow among populations of Crotaphytus collaris in unfragmented habitats using mtDNA. In addition, no gene flow studies have been performed in Texas populations. Two ranch sites and four lake sites were used for lizard collection. Because these lakes vary in age, the effect of age on gene flow can be determined. The O. C. Fisher Lake is located northwest of San Angelo, Texas, and was completed in 1951 (Wilde pers. comm.). Twin Buttes Reservoir is southwest of San Angelo and was completed in 1963 (Thornton pers. comm.). Farther northwest of San Angelo than O. C. Fisher Lake is the E. V. Spence Reservoir, which was completed in 1969 (Thornton pers. comm.). Lastly, the Ballinger Municipal Lake is far northeast of San Angelo and was completed in 1984 (New pers. comm.). Methods and Materials Lizards (ft =37) were captured from six local populations in Tom Green, Runnels, Coke and Concho counties including the Sims Ranch near Paint Rock (ft =15), the Carter Ranch near Mertzon (ft = 6), E. V. Spence Reservoir near Robert Lee (ft =2), O. C. Fisher Lake in San Angelo (ft = 6), Twin Buttes Reservoir in San Angelo (ft =4) and Ballinger Municipal Lake near Ballinger (ft =4). After capture, lizards were euthanized with sodium pentobarbital, autopsied and all parasites CAMPBELL & McCOY 153 were removed. Approximately 60 /xL of whole blood was drawn from the orbital sinus (Mac Lean et al. 1973) of each lizard prior to eutha¬ nasia. Blood samples were stored at -70°C in blood buffer (10.0 mM Tris, 1.0 mM EDTA, 120.0 mM NaCl). Digestions consisted of 20 /xL of blood, 90 /x L of lysis buffer (10.0 mM Tris, 1.0 mM EDTA, 120.0 mM NaCl, 0.2% SDS), 50 /xL proteinase K (6.0 mg/mL), and 3.0 (1 CaCl2 (100 mM). Digestions were incubated at 35 °C for 36-48 hours. DNA was harvested from the resulting blood lysate using a standard phenol-chloroform extraction (Mullebach et al. 1989; Sambrook et al. 1989). Harvested DNA was visualized on 1.0% (w/v) agarose gels stained with ethidium bromide. Samples of satisfactory quality were quantified with a mass spectrophotometer (Sambrook et al. 1989). The mitochondrial genome of the collared lizard was chosen for study because of its strictly maternal mode of transmission and highly con¬ served nature (Avise 1986; 1994; Cronin 1991; Cronin et al. 1993). An approximately 2400 base pair (bp) region of the NADH dehydrogenase gene (Lee et al. 1994) was amplified using the polymerase chain reaction (PCR) (Sambrook et al. 1989) and primers from Lee et al. (1994). Optimal reagent concentrations per reaction (50 /xL) were: 50 ng of DNA, primers (5’-TAA GCT ATC GGG CCC ATA CC-3\ 5’- ACT TCA GGG TGC CCA AAG AAT CA-3’) at 0. 1 /xM each, MgCl2 at 0.5 mM, dNTP’s at 2.2 /xM, reaction buffer at IX, and 2.5 units (U) Taq polymerase. Each reaction consisted of 30 cycles (88 °C for 45 sec, 58 °C for 45 sec, 72°C for 2 min). Amplicons were digested with 12 restriction endonucleases (Avise et al. 1979; Lansman et al. 1981), including Cfo I, Hpa II, Sau 96 I, Rsa I, Hin d III, Dde I, Ase I, Hin f I, Alu I, Hin c II, Dpn I and Eco R V. Each amplicon was treated with 1.0 U of restriction enzyme and incubated 12-16 hours at 35 °C. For easy comparisons among individual lizards, digestions of all lizards within an enzyme were loaded onto a 1.0% (w/v) midi agarose gel. Different restriction patterns were coded as different haplotypes and scored for each individual. Binary character matrices included only restriction sites (Hillis 1996) from those enzymes that yielded more than one haplotype from all individuals treated. The presence or absence of each restriction site (from enzymes that yielded more than one haplotype for all individuals sampled) were used as discrete characters in the matrices (Georgiadis 1996). Phylogenetic trees were generated by PHYLIP® (Felsenstein 1993). The documentation information with this software package was 154 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 2, 2002 used to determine which analysis should be used. The Wagner and Dollo parsimony approaches appeared to be the most relevant to binary data sets (Felsenstein 1993). The Wagner parsimony program was chosen over Dollo parsimony because of its allowance for the equal probability that a mutation event will either create or destroy a restriction site. The Dollo program assumes that restriction sites are more likely to be destroyed than created by mutation events. It is the opinion of the authors of this study that this assumption cannot be supported without sequencing the PCR fragment. The more liberal view of mutation event ramifications upheld by the Wagner parsimony program, therefore, appeared more appropriate for this analysis. Bootstrapping (Felsenstein 1985) was performed 100 times on these data. In the program used (PARS), the most parsimonious representa¬ tion for the data is found. Bifurcations and multifurcations in trees are considered, and numerical trees are produced (Felsenstein, 1993). The tree-drawing program produces a graphical illustration of the most parsimonious numerical tree. Results Five ( Cfo I, Sau 96 I, Hpa II, Hin f I and Alu I) of the 12 restriction endonucleases cleaved the ampl icons with more than one haplotype. From these five enzymes, 19 restriction sites were gleaned (Table 1). When considering only the restriction enzymes that cleaved the PCR ampl icon with more than one haplotype, only Haplotype 2 from Hin f I appears to be unique to a single collection site. This haplotype was found in five of the six (83%) individuals captured at the O. C. Fisher Lake. The 12 remaining haplotypes from these five enzymes were shared between individuals of at least two sites. The bootstrapped gene tree (Fig. 1) generated from the restriction site data demonstrated genetic trends among the populations. Much of this tree was supported by bootstrap values of more than 60% . However, some sections of the tree were inferred with values less than 40%. Since much of the tree has a high probability of accuracy, population structure can be inferred. Both individuals from E. V. Spence Reservoir were placed together. This bootstrap value was high (74%) and is likely accurate. Five of the six lizards captured at O. C. Fisher Lake were placed on a single branch, as well. The likelihood of this arrangement was calculated at 90% . The other individual captured from this site was placed on a low- probability branch ( — 40%) with lizards from Ballinger Municipal Lake, Twin Buttes Reservoir and the Sims Ranch. CAMPBELL & McCOY 155 Table 1. Binary character matrix consisting of 19 restriction sites. The presence of a restriction site is denoted by a "1", while "0" denotes the absence of the site. Lizard I.D. Restriction sites 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 1-11-17 1 0 1 1 1 0 0 1 1 1 0 1 1 1 0 0 1 1 1 1-11-19 1 0 1 1 1 0 0 1 1 1 0 1 1 1 0 0 1 1 1 5-10-12 1 0 1 1 0 1 1 1 1 1 0 1 1 1 0 0 1 1 1 5-10-14 1 0 1 1 0 1 1 1 1 1 0 1 1 1 0 0 1 1 1 1-13-18 1 0 0 0 0 0 0 1 1 1 0 1 1 0 0 1 1 1 1 1-11-20 1 0 1 1 1 0 0 1 1 1 0 1 1 1 0 0 1 1 1 5-10-18 1 0 1 1 1 0 0 1 1 1 0 1 1 1 0 0 1 1 1 5-10-11 1 0 1 1 1 0 0 1 1 1 0 1 1 1 0 0 1 1 1 5-9-11 1 0 1 1 1 0 0 1 1 1 0 1 1 1 0 0 1 1 1 5-9-18 1 0 1 1 1 0 0 1 1 1 0 1 1 1 0 0 1 1 1 5-8-16 1 0 1 1 1 0 0 1 1 1 0 1 1 1 0 0 1 1 1 5-8-17 1 0 1 1 1 0 0 1 1 1 0 1 1 1 0 0 1 1 1 1-11-18 1 0 0 0 0 0 0 1 1 1 0 1 1 1 0 0 1 1 1 8-13-17 1 0 1 1 0 1 1 1 1 0 0 1 1 1 0 0 1 1 1 5-10-13 1 0 1 1 0 1 1 1 1 0 0 1 1 1 0 0 1 1 1 5-8-19 1 0 1 1 0 1 1 1 1 0 0 1 1 1 0 0 1 1 1 5-9-19 1 0 1 1 0 1 1 1 1 0 0 1 1 1 0 0 1 1 1 5-9-20 1 0 1 1 0 1 1 1 1 0 0 1 1 1 0 0 1 1 1 1-12-20 1 0 0 0 0 0 0 1 1 0 0 1 1 1 0 0 1 1 1 1-12-19 1 0 1 1 0 1 1 1 1 0 0 1 1 1 0 0 1 1 1 5-10-15 1 0 1 1 0 1 1 1 1 0 0 1 1 1 0 0 1 1 1 5-9-12 1 0 1 1 0 1 1 1 1 0 0 1 1 1 0 0 1 1 1 5-9-13 1 0 1 1 0 1 1 1 1 0 0 1 1 1 0 0 1 1 1 5-9-14 1 0 1 1 0 1 1 1 1 0 0 1 1 1 0 0 1 1 1 5-9-15 1 0 1 1 0 1 1 1 1 0 0 1 1 1 0 0 1 1 1 5-9-17 1 0 1 1 0 1 1 1 1 0 0 1 1 1 0 0 1 1 1 1-11-16 1 1 1 1 0 1 1 1 1 1 0 1 0 0 1 0 1 1 1 5-10-17 1 1 1 1 0 1 1 1 1 1 0 1 0 0 1 0 1 1 1 5-10-19 1 1 1 1 0 1 1 1 1 1 0 1 0 0 1 0 1 1 1 5-8-18 1 1 1 1 0 1 1 1 1 1 0 1 1 1 0 0 1 1 1 5-10-16 1 1 1 1 0 1 1 1 1 1 0 1 0 0 1 0 1 1 1 5-10-20 1 1 1 1 0 1 1 1 1 1 0 1 0 0 1 0 1 1 1 5-8-15 1 1 1 1 1 0 0 1 1 1 0 1 1 0 0 1 1 1 1 1-12-18 1 1 1 1 0 1 1 1 0 0 1 1 1 1 0 0 1 1 0 1-13-16 1 1 1 1 0 1 1 1 0 0 1 1 1 1 0 0 1 1 0 1-12-16 1 1 1 1 0 1 1 1 1 0 0 1 1 1 0 0 1 1 0 7-13-16 1 1 1 1 0 1 1 1 1 0 0 1 1 1 0 0 1 1 0 In contrast, Crotaphytus collaris individuals (n= 4) captured from Twin Buttes Reservoir did not place in well -supported branches on the gene tree. One lizard was placed on the branch shared with the majority of the O. C. Fisher Lake lizards. The bootstrap value generated for this arrangement was calculated at 70% and is likely accurate. However, the other three lizards were placed on a branch of the tree that bootstrapped at approximately 40%. This positioned these three individuals with members of the Ballinger, Sims and O. C. Fisher populations. 156 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 Carter (8) Figure 1 . Gene tree generated by PHYLIPW using a Wagner parsimony method. Probabili¬ ties are provided only for branches of 40% bootstrap value or higher. Values given are percentages. This was also the case for members of the Ballinger Municipal Lake sample (n= 4). One of these lizards was placed on a branch of 63% probability with two Carter Ranch individuals. The remaining three lizards were positioned on a low probability branch with members of the Twin Buttes, Sims and O. C. Fisher samples. CAMPBELL & McCOY 157 Figure 2. Gene tree applied to a geographical map of the collection sites. The lizards captured from the Carter Ranch (n = 6) were among the most dispersed on the gene tree. Two individuals were placed with the two lizards captured at E. V. Spence Reservoir with a probability of 74%. Two more Carter Ranch lizards were placed with a single lizard from the Ballinger sample site with a bootstrap of 63%. The last two Carter Ranch lizards were arranged with 10 lizards from the Sims Ranch. The bootstrap value of this arrangement was lower than 40%. When the gene tree is applied to maps of the collection sites, inter- populational data can be inferred. If all of the tree is applied to the map of the collection sites (Fig. 2), it is obvious that gene flow is occurring between E. V. Spence and the Carter Ranch. In fact, this appears to be the only gene flow to or from E. V. Spence Reservoir. Ballinger Municipal Lake is linked both to the Carter Ranch and to the Sims Ranch. In addition, gene flow to and from O. C. Fisher appears to be linked mainly to Twin Buttes. The Carter Ranch appears to be linked to Ballinger Municipal Lake and E. V. Spence. The Sims Ranch appears to be linked to all collection sites except E. V. Spence. 158 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 The gene tree applied to the geographic locations was also informative when only the most strongly supported clades (>60% bootstrap values) were analyzed (Fig. 3). Again, only the Carter Ranch is connected to the E. V. Spence lizards. The Carter Ranch is clearly linked to Ballinger Municipal Lake. Also, Twin Buttes and O. C. Fisher individuals appear linked. The branch with five of the six O. C. Fisher lizards and a single Twin Buttes lizard also appears. The Sims Ranch has no well -supported gene flow affiliations with any other population when analyzed with this method. Discussion It is likely that many of the 12 restriction endonucleases utilized did not yield differing haplotypes because these areas of the amplicon are highly conserved. Mutations at these sites possibly result in lethal changes in protein structure. Mutations can and will occur at these sites, but individuals are not viable and do not reproduce. For this reason, haplotypes arising from mutations at these sites are not observed in the populations sampled. However, it is possible that mutations at these restriction sites occur at low rates in viable individuals but were not observed due to an insufficient sample size. Because Crotaphytyus collaris individuals captured on the Carter Ranch grouped with individuals from all other populations, one can infer that this site contains all ancestral haplotypes at these restriction sites, relative to the sites sampled. Even though individuals captured at the Sims Ranch did not link to any members of the E. V. Spence popula¬ tion, and all branches containing Sims Ranch lizards were poorly supported, one could infer that the Sims Ranch individuals also con¬ tained all ancestral haplotypes for these restriction sites. When considering the gene tree (Figure 1), only the lizards from E. V. Spence Reservoir form an exclusive group. This group is tied only to the Carter Ranch on a high-probability branch. This indicates strong gene flow between these two populations. An individual captured at the Ballinger Municipal Lake is tied to two Carter Ranch individuals. Other Ballinger Lake individuals are grouped with lizards from Twin Buttes, O. C. Fisher and Sims Ranch. This suggests genetic flow to Ballinger Lake from all of these sources. It is also to be noted that the Carter Ranch is tied to all other collection sites sampled, and the Sims Ranch is tied to all other sample sites except E. V. Spence. One possible explanation for this is that the Carter and Sims Ranches are ancestral to CAMPBELL & McCOY 159 Figure 3. Well-supported gene tree branches applied to a geographical map of the collection sites. The branches used are of a 60% or higher bootstrap value. all other populations sampled. Because of this, the other populations contain subsets of ancestral diversity. The ages of the sites also shed light on the apparent gene flow observed. The Carter and Sims Ranches are the oldest of the sites in this study. Because of this, the strong gene flow apparent between them in Figure 2 is expected. It is possible that both of these populations were founded from the same parent population, and gene exchange has been occurring for many years. The fact that gene flow is occurring between the older ranch populations and the newer lake populations is logical. As the new habitats were founded, the routes of gene flow extended. The most convincing argument for the gene flow observed in this study is illustrated in Figure 3. Clear and logical paths of genetic exchange are visible when the most poorly supported branches of the gene tree are removed from the map of the collection sites. In this representation, the Sims Ranch is left unlinked to any other population sampled. This can be explained again by its age. Its ties to the Carter 160 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 Ranch are likely a result of a common founder population. All links to lizards from Twin Buttes, O. C. Fisher and Ballinger Municipal Lake are likely carried over indirectly from the Carter Ranch exchanges. No gene flow is extended from the Sims Ranch to any of the populations directly, but similarities are observed due to a possible common ancestral population. Sufficient diversity existed within these 19 restriction sites in the Sims Ranch lizards to allow the tree-drawing algorithm of PHYLIP® to group them with the lizards of virtually all other populations. The gene flow paths from the Carter Ranch are likely more accurate in Fig. 3 than Fig. 2. When the poorly supported branches are re¬ moved, it is obvious that the Carter Ranch is linked to E. V. Spence and Ballinger Municipal Lake. E. V. Spence is linked only to the Carter Ranch. O. C. Fisher and Twin Buttes also engage in genetic exchange. Also, the Carter Ranch is the only well-supported genetic link to the Ballinger Municipal Lake (Figure 1). Age of habitat again becomes significant when applied to Figure 3. O. C. Fisher Lake was completed in 1951, and probably was colonized by populations receiving gene flow from the Carter Ranch. Subsequent¬ ly, Twin Buttes Reservoir was completed in 1963. It appears that gene flow extended from the population established 14 years earlier at O. C. Fisher Lake since Twin Buttes lizards do not group with high probability to any lizards from the Carter Ranch. This is expected when the close proximity of O. C. Fisher to Twin Buttes is considered. The Carter Ranch also appears to have extended its gene flow to E. V. Spence Reservoir upon its completion in 1969. It is unlikely that the same geographic path was taken from the Carter Ranch to both O. C. Fisher and to E. V. Spence since lizards from each location do not form a branch of high probability on a gene tree. The same appears to be true of the geographic path of gene flow between the Carter Ranch and Ballinger Municipal Lake. This lake was completed in 1984, but no individuals from either O. C. Fisher or E. V. Spence form high probability branches with those of Ballinger Municipal Lake. It does not appear possible that this hypothetical path extends in a roughly northward direction from the Carter Ranch and trifurcates, turning toward these three lake sites. Lizards from each of these three populations would be placed on separate branches on a gene tree, and each of these branches would contain lizards from the Carter Ranch. CAMPBELL & McCOY 161 However, only Ballinger Municipal Lake and E. V. Spence form branches with Carter Ranch individuals. This indicates that the population of Crotaphytus collaris at O. C. Fisher Lake has diverged to a greater extent from the Carter Ranch than the populations at E. V. Spence and Ballinger Municipal Lake. Again, this could be an artifact of its age, relative to the other sites. It is possible that a geographic boundary has severed gene flow between this population and the one at the Carter Ranch. The resulting isolation could cause this divergence. However, extensive geographic knowledge and the exact position of all satellite populations of C. collaris would be necessary to reach this conclusion. An isolation theory is applicable to the Sims Ranch population, as well. It is closer to Ballinger Municipal Lake than is the Carter Ranch, but branches linking Ballinger Municipal Lake to the Sims Ranch were of a low probability. Again, extensive geographical knowledge would be needed to support this theory. Acknowledgments We wish to thank the Colorado Municipal Water District, Army Corp of Engineers, Ben Sims and Tom Carter for access to the collection sites. In addition, we would like to thank Ned E. Strenth, David S. Marsh, Kathryn Perez and Brandon Speed for their assistance during the study and in the preparation of this manuscript. We also wish to thank T. New (Ballinger City Planner Office), O. Thornton (Colorado Munici¬ pal Water District Office) and W. Wilde (San Angelo Water District Office) for historical data relative to the collection sites. We wish also to thank the anonymous reviewers of this study for their time and effort. Literature Cited Avise, J. C., R. A. Lansman & R. O. Shade. 1979. The use of restriction endonucleases to measure mitochondrial DNA sequence relatedness in natural populations. I. Population structure and evolution in the genus Peromyscus. Genetics, 92:279-295. Avise, J. C. 1986. Mitochondrial DNA and the evolutionary genetics of higher animals. Royal Society of London. Philosophical Transactions. Biological Sciences, B312:325- 342. Avise, J. C. 1994. Introduction - History of Molecular Phylogenetics. Pp. 3-43, in Molecular markers, natural history and evolution. Chapman and Hall. New York, New York, 511 pp. Baird, T. A., S. F. Fox & J. K. McCoy. 1997. Population differences in the roles of size and coloration in intra- and intersexual selection in the collared lizard, Crotaphytus collaris : influence of habitat and social organization. Behavioral Ecology, 8:506-517. Cronin, M. A. 1991. Mitochondrial and nuclear genetic relationships of deer ( Odocoileus spp.) in western North America. Canadian Journal of Zoology, 69:1270-1279. Cronin, M. A., W. J. Spearman, R. L. Wilmot, J. C. Patton & J. W. Bickham. 1993. 162 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 Mitochondrial DNA variation in chinook ( Oncorhynchus tshawytscha ) and chum salmon ( O . keta) detected by restriction enzyme analysis of polymerase chain reaction (PCR) products. Canadian Journal of Fisheries and Aquatic Science, 50:708-715. Felsenstein, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution, 39:783-791. Felsenstein, J. 1993. PHYLIP: Phylogeny inference package, version 3.5. University of Washington, Seattle. Georgiadis, N. 1996. Reconstruction population history using PCR-restriction site data. Pp. 154-166, in Conservation genetics: case studies from nature. Chapman and Hall. New York, New York, 512 pp. Hillis, D. M., C. Moritz & B. K. Mable. 1996. Nucleic acids II: the polymerase chain reaction - Phylogenetic inference. Pp. 205-514, in Molecular systematics. 2nd Ed. Sinauer Associates, Inc. Sunderland, Massachusettes, 655 pp. Hranitz, J. M. & T. A. Baird. 2000. Effective population size and genetic structure of a population of collared lizards, Crotaphytus collaris, in central Oklahoma. Copeia, 2000: 786-791. Lansman, R. A., R. O. Shade, J. F. Shapira & J. C. Avise. 1981. The use of restriction endonucleases to measure mitochondrial DNA sequence relatedness in natural populations. Journal of Molecular Evolution, 17:214-226. Lee, T. E., J. W. Bickham & M. D. Scott. 1994. Mitochondrial DNA and allozyme analysis of North American pronghorn populations. Journal of Wildlife Management, 58:307-318. MacLean, G. S., A. K. Lee & K. J. Wilson. 1973. A simple method of obtaining blood from lizards. Copeia, 1973:338-339. McCoy, J. K. 1995. Mechanisms of selection for the evolution of sexual dimorphism in the collared lizard ( Crotaphytus collaris ). Ph.D. dissertation. Oklahoma State University. Stillwater, Oklahoma, 120 pp. McCoy, J. K., S. F. Fox & T. A. Baird. 1994. Geographic variation in sexual dimorphism in the collared lizard, Crotaphytus collaris (Sauria: Crotaphytidae). The Southwestern Naturalist, 39:328-335. McCoy, J. K., H. J. Harmon, T. A. Baird & S. F. Fox. 1997. Geographic variation in sexual dichromatism in the collared lizard, Crotaphytus collaris (Sauria: Crotaphytidae). Copeia, 1997:565-571. Mullebach, R., P. G. L. Lagoda & C. Welter. 1989. An efficient salt-chloroform extraction of DNA from blood and tissue. Trends in Genetics, 5:391. Sambrook, J., E. F. Fritsch & T. Maniatis. 1989. In vitro amplification of DNA by the polymerase chain reaction. Pp. 14.2-14.5, in Molecular cloning: a laboratory manual. 2nd Ed. Cold Spring Harbor Laboratory Press. Cold Spring Harbor, New York, 1659 pp. Sexton, O. J., R. M. Andrews & J. E. Bramble. 1992. Size and growth rate characteristics of a peripheral population of Crotaphytus collaris (Sauria: Crotaphytidae). Copeia, 1992:968-980. Templeton, A. R., K. Shaw, E. Routman & S. C. Davis. 1990. The genetic consequences of habitat fragmentation. Annals of the Missouri Botanical Garden, 77:13-27. Yoshioka, J. H. 1996. The genetic structure of Oklahoma populations of the collared lizard Crotaphytus collaris. Unpublished master’s thesis. Oklahoma State University. Stillwater, Oklahoma. 54 pp. JHC at: jim_h_campbell@yahoo.com TEXAS J. SCI. 54(2): 163-176 MAY, 2002 EVALUATION OF FACILITATED SUCCESSION AT LAS PALOMAS WILDLIFE MANAGEMENT AREA IN SOUTH TEXAS Frank W. Judd, Robert I. Lonard and Gary L. Waggerman* Department of Biology The University of Texas - Pan American Edinburg, Texas 78539 and *Texas Parks and Wildlife Department 3331 Ranch Road 12, # 102 San Marcos, Texas 78666 Abstract. — This study examined the effectiveness of re- vegetation efforts which have been ongoing in the Lower Rio Grande Valley of Texas since 1958. Species composition, richness and diversity were evaluated in an undisturbed native woodland, a site planted with late successional species in 1961 (facilitated succession) and a farm field abandoned in 1974 (unaided succession) in northwestern Cameron County. Species richness and diversity for both trees and shrubs were greatest in the native woodland site. While there was greater similarity in species composition between the native woodland and the facilitated succession sites, species diversity in the tree and shrub layers of the facilitated succession site is still significantly lower than the native woodland site. The Lower Rio Grande Valley (LRGV) of Texas is both a political and a biogeographic unit. As a political unit it comprises the southern¬ most four counties in Texas, i.e., Cameron, Hidalgo, Starr and Willacy. Biogeographically, it corresponds closely with the Matamoran District of the Tamaulipan Biotic Province (Blair 1950). It includes all of the Pleistocene-Recent delta of the Rio Grande in Texas. This 1,208,530 ha area exhibits great biodiversity. More than 500 vertebrate and 170 woody species occur in the LRGV. Sixty-seven species are considered threatened or endangered (Jahrsdoerfer & Leslie 1988). Because of its high biodiversity, large number of threatened and endangered species, large number of neotropical species that reach the northern limit of their distribution in the area, and small amount of native habitat remaining, the Texas and U.S.A. governments are combining to purchase lands for preservation and re- vegetation. The U. S. Fish and Wildlife Service (USFWS) Land Protection Plan calls for protection of 53,420 ha in the LRGV using island biogeography concepts (Harris 1984) with the Rio Grande serving as the major corridor linking tracts of native and restored vegetation. When completed, the Lower Rio Grande Wildlife Corridor will extend 240 km 164 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 from the mouth of the river in Cameron County to Falcon Dam in Starr County. Lands acquired by USFWS for the corridor become a part of the Lower Rio Grande Valley National Wildlife Refuge. It currently consists of over 100 tracts comprising about 26,000 ha. To date about 2,430 ha have been re-vegetated. Since 1958, Texas Parks and Wildlife Department (TPWD) has been involved in re-vegetation in the LRGV to create white-winged dove (Zenaida asiatica) habitat and to promote biodiversity. TPWD has planted about 260 ha on 12 tracts. The Texas Nature Conservancy (TNC) also has promoted re- vegetation efforts, primarily by encouraging and helping private land owners to plant native woody species. This has culminated in the re-vegetation of 27 tracts comprising 364 ha. The combined efforts of the USFWS, TPWD, and TNC have resulted in the re- vegetation of about 3,054 ha. The lands acquired to create white-winged dove habitat and promote biodiversity are usually recently cultivated fields. Left alone these abandoned fields undergo secondary succession. The first plants to become established typically are herbaceous annuals (Vora & Messerly 1990). In time, these colonizing species are gradually replaced by woody species. The rate at which succession occurs depends, in part, on the ability of mid and late successional species to disperse to a site and successfully compete with colonizing species that are already established. Re-vegetation projects attempt to accelerate succession by introducing climax species into an area. Thus, these re- vegetation efforts are based on the Facilitation Model of succession (Connell & Slay ter 1977). Vora & Messerly (1990) suggested that unaided succes¬ sion in LRGV communities fit the Facilitation Model and Archer et al. (1988) reported that succession at a site 175 km north of the Rio Grande fit the Facilitation Model. Re- vegetation efforts in the LRGV have been ongoing since 1958 and have occurred annually since 1983. However, there has been no assess¬ ment of the effectiveness of these re-vegetation efforts in accelerating succession or in achieving similar composition and structure as existing climax communities. Only one study (Vora & Messerly 1990) provides information on succession in the LRGV, and it covers only a five-year time period. Consequently this study sought information on the effec¬ tiveness of facilitated succession in achieving similar composition, structure, and diversity, in a period of 40 years, to an undisturbed forest at the TPWD’s Longoria Unit of the Las Palomas Wildlife Management Area (LPWMA) in northwestern Cameron County. JUDD, LONARD & WAGGERMAN 165 Figure 1 . Map showing the location of Las Palomas Wildlife Management Area. Methods The Longoria Unit is a 80.9 ha public hunting site located 7.4 km north of Santa Rosa, Cameron County, Texas (Fig. 1). Soils at the site belong to the Raymondville Association, which are characterized by nearly level, moderately drained, clay loams (Williams et al. 1977). The climate is semi-arid and annual precipitation is about 68 cm with a rainfall peak in September and October (Lonard et al. 1991). The mean frost- free period is 330 days. Frequently an entire winter will pass without a freezing temperature. Controls were not established when the re- vegetation was done so this study compared vegetation at three sites: (1) an undisturbed woodland, (2) a re- vegetated site and (3) an unaided secondary succession site. Historically, the Longoria Unit was a small portion of a Spanish land grant. The former owner of the tract indicated that the site selected as undisturbed woodland had never been cleared for agriculture or grazing. The facilitated succession site was formerly a cultivated field that was 166 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 re- vegetated in 1961. Seedlings of anacua ( Ehretia anacua), brasil (Condalia hookeri), Texas ebony ( Chloroleucon ebano ) and granjeno ( Celtis pallida) were hand-dug and planted in rows 3.0 m apart. Plants within rows were spaced 2.4 m apart and approximately equal numbers of each of the species were planted. Seedlings were watered, fertilized, and hand-pruned to enhance branching for white- winged dove nests. Successful nesting first occurred seven years after planting. The unaided succession site was a cultivated field that was abandoned in 1958. It was mowed infrequently from 1958 to 1974. In 1974 all mechanical plant control operations ceased and the area was allowed to undergo secondary succession. Thus, except for the ground layer it had been in succession for 13 fewer years than the facilitated succession site. The sites are all close to each other and separated by dirt roads about 5 m wide. To census the vegetation, ten 10 m by 10 m quadrats were established at regular intervals of 150 m along a north/south axis across each of the three sites. Censusing of tree, shrub and ground layers was done sepa¬ rately. The tree layer consisted of woody plants greater than 3.0 m tall. The shrub layer was comprised of woody plants 1.0 to 3.0 m tall. The ground layer consisted of woody and herbaceous plants less than 1.0 m tall. Heights were determined with a calibrated telescoping pole. Density of trees and shrubs was counts of individuals in the quadrats. Frequency was determined by the presence of a species in the 10 qua¬ drats at a site. Cover was based on diameter at breast/height (dbh = 1.35 m) of trees and basal diameter of shrubs. Multiple stems were summed. Dominance in the tree and shrub layers was determined by calculating an importance value, which was the sum of relative density, relative frequency, and relative cover. Heights of trees and shrubs were determined using a calibrated telescoping pole that had a maximum height of 7.5 m. Height of trees taller than 7.5 m was estimated. The ground layer was censused using the line intercept technique (Canfield 1941). Five 10 m long intervals were established spaced 2 m apart across each quadrat. Thus, at each site there were 50 intervals. Cover was determined by the perpendicular projection of the foliage onto the transect line. Frequency was based on the presence of a species in the 50 intervals of the transects. To determine the density of tree and shrub seedlings, a 10 cm strip on each side of the transect was established. Density and height of tree and shrub seedlings less than 1 m tall were determined in the 20 cm wide belts. For all other ground layer species, density was not determined because of the difficulty in identifying what JUDD, LONARD & WAGGERMAN 167 constituted an individual. Dominance was assessed in the ground layer by calculating an importance value that was the sum of relative fre¬ quency and relative cover. Similarity of floristic composition within layers among sites was determined using Sorenson’s coefficient of community (Krebs 1999). Species diversity and evenness was assessed using the Shannon- Wiener function applied to species importance values (Brower et al. 1998; Krebs 1999) Tests for significant differences follow information in Sokal & Rohlf (1981). Nomenclature follows Jones et al. (1997). Results Species presence, richness, mean height and importance in the tree layer are compared among sites in Table 1 . The native woodland had the greatest number of species (13). All four of the species ( Ehretia anacua , Celtis pallida , Condalia hookeri and Chloroleucon ebano) initially planted at the facilitated succession site were present in the native woodland. Ehretia anacua (anacua) was the dominant species in the tree layer at this site. Condalia hookeri (brasil) was a close second in importance. Prosopis glandulosa (mesquite), Celtis laevigata (hackberry) and Chloroleucon ebano (Texas ebony) were the tallest trees, but density of mesquite and Texas ebony was low. Similarly, all four of the species initially planted at the facilitated succession site were present here (Table 1). Only three tree species were added to the facilitated succession site in the 40 years since planting, and species richness was just seven (54% of the native woodland). Anacua also was the dominant species in the tree layer at this site. Mean height of anacua was similar to the height of the species in the native woodland. Indeed, except for Texas ebony, the heights of other species in the tree layer of the facilitated succession were similar to their counterparts in the native woodland. The unaided succession site (Table 1) had one more species in the tree layer (8) than the facilitated succession site. All four species initially planted at the facilitated succession site were present in the tree layer. Zanthoxylumfagara (colima) was the dominant species. Anacua was second in importance. Except for Celtis pallida (granjeno), mean height of the trees was shorter than their counterparts in the native woodland. Species presence, richness and importance in the shrub layer of the sites are shown in Table 2. Mean height is not presented because this layer was rather narrowly defined as being between 1 .0 and 3.0 m. The 168 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 Table 1. Mean height and species importance in the tree layer. Imp. value = importance value and is the sum of relative frequency, relative density and relative cover. Density is number per 1,000 sq. m. Species Height (m) Freq. (%) Rel. Freq. Den. Rel. Den. Cover (cm) Rel. Cov. Imp. Value Undisturbed Native Woodland Ehretia anacua 4.75 100 14.9 85 29.6 767.7 25.0 69.5 Condalia hookeri 5.60 100 14.9 46 16.0 925.6 30.2 61.1 Sideroxylon celastrinum 4.60 90 13.4 50 17.4 361.0 11.8 42.6 Zanthoxylum fagara 5.46 90 13.4 33 11.5 303.5 9.9 34.8 Celtis pallida 4.61 70 10.4 12 4.2 72.2 2.4 17.0 Celtis laevigata 8.22 50 7.5 11 3.8 167.9 5.5 16.8 Ulmus crassifolia 5.15 30 4.5 18 6.3 177.9 5.8 16.6 Diospyros texana 3.94 60 9.0 10 3.5 78.0 2.5 15.0 Forestiera angustifolia 3.58 30 4.5 11 3.8 88.3 2.9 11.2 Prosopis glandulosa 9.15 20 3.0 2 0.7 74.5 2.4 6.1 Amyris texana 3.44 10 1.5 7 1.5 28.9 0.9 3.9 Chloroleucon ebano 7.30 10 1.5 1 0.3 14.1 0.5 2.3 Acacia greggii 4.80 10 1.5 1 0.3 7.6 0.2 2.0 Total number of plants = 287 Facilitated Succession Site Ehretia anacua 4.85 100 18.5 69 37.7 941.9 37.1 93.3 Zanthoxylum fagara 4.73 90 16.7 41 22.4 461.2 18.2 57.3 Celtis laevigata 8.29 90 16.7 24 13.1 383.1 15.1 44.9 Celtis pallida 4.56 100 18.5 26 14.2 203.2 8.0 40.7 Condalia hookeri 5.73 70 13.0 8 4.4 261.0 10.3 27.7 Chloroleucon ebano 5.96 50 9.3 10 5.5 214.4 8.5 23.3 Ulmus crassifolia 7.77 40 7.4 5 2.7 71.8 2.8 12.9 Total number of plants = 183 Unaided Succession Site Zanthoxylum fagara 4.05 100 23.8 44 29.3 521.4 37.0 90.1 Ehretia anacua 4.10 100 23.8 47 31.3 313.7 22.3 77.4 Baccharis neglecta 3.40 60 14.3 25 16.7 172.9 12.3 43.3 Celtis pallida 4.73 80 19.0 15 10.0 156.8 11.1 40.1 Celtis laevigata 6.76 50 11.9 16 10.7 203.4 14.4 37.0 Condalia hookeri 4.10 10 2.4 1 0.7 26.1 1.9 5.0 Chloroleucon ebano 4.90 10 2.4 1 0.7 7.2 0.5 3.6 Havardia pallens 4.00 10 2.4 1 0.7 6.6 0.5 3.6 Total number of plants = 150 native woodland had the greatest number of species (16) and the unaided succession and facilitated succession sites had equal numbers of species (12). Three of the four species initially planted in the facilitated succession site were present in all three sites (anacua, Texas ebony and granjeno), but brasil was present as a shrub only in the native woodland site (Table 2). Forestiera angustifolia (panalero) was the dominant shrub at the native woodland. Anacua was the dominant species in the shrub layer at the facilitated succession site and colima was the dominant JUDD, LONARD & WAGGERMAN 169 Table 2. Species importance in the shrub layer. Imp. value = importance value and is the sum of relative frequency, relative density and relative cover. Density is number per 1 ,000 sq. m. Species Freq. Rel. Den. Rel. Cover Rel. Imp. (%> Freq. Den. (cm) Cov. Value Undisturbed Native Woodland Forestiera angustifolia 70 9.7 75 23.1 214.9 27.5 60.3 Randia rhagocarpa 100 13.9 76 23.5 122.8 15.7 53.1 Ehretia anacua 90 12.5 51 15.7 153.2 19.6 47.8 Sideroxylon celastrinum 80 11.1 27 8.3 68.0 8.7 28.1 Amyris texana 60 8.3 20 6.2 40.3 5.2 19.7 Zanthoxylum fagara 60 8.3 17 5.2 36.3 4.6 18.1 Diospyros texana 70 9.7 9 2.8 16.8 2.1 14.6 Celtis pallida 40 5.6 10 3.1 37.1 4.7 13.4 Malpighia glabra 30 4.2 19 5.9 14.4 1.8 11.9 Chloroleucon ebano 40 5.6 6 1.9 27.6 3.5 11.0 Ulmus crassifolia 20 2.8 6 1.9 16.9 2.2 6.9 Phaulothamnus spinescens 20 2.8 4 1.2 16.7 2.1 6.1 Bernardia myricifolia 10 1.4 1 0.3 14.2 1.8 3.5 Condalia hookeri 10 1.4 1 0.3 1.9 0.2 1.9 Ziziphus obtusifolia 10 1.4 1 0.3 0.6 0.1 1.8 Celtis laevigata Total number of plants = 324 10 1.4 1 0.3 0.3 <0.1 1.7 Facilitated Succession Site Ehretia anacua 100 14.7 116 42.6 190.0 47.3 104.6 Chloroleucon ebano 90 13.2 37 13.6 30.5 7.6 34.4 Celtis pallida 70 10.3 27 9.9 55.9 13.9 34.1 Randia rhagocarpa 90 13.2 22 8.1 46.4 11.5 32.8 Zanthoxylum fagara 80 11.8 27 9.9 33.9 8.4 30.1 Forestiera angustifolia 70 10.3 12 4.4 19.5 4.9 19.6 Sideroxylon celastrinum 70 10.3 14 5.1 4.4 1.1 16.5 Celtis laevigata 50 7.4 8 2.9 7.0 1.7 12.0 Amyris texana 30 4.4 6 2.2 12.3 3.1 9.7 Malpighia glabra 10 1.5 1 0.4 1.0 0.2 2.1 Diospyros texana 10 1.5 1 0.4 0.5 0.1 2.0 Ulmus crassifolia Total number of plants = 272 10 1.5 1 0.4 0.4 0.1 2.0 Unaided Succession Site Zanthoxylum fagara 100 16.7 64 24.9 208.5 27.9 69.5 Ehretia anacua 100 16.7 58 22.6 112.1 15.0 54.3 Baccharis neglecta 70 11.7 31 12.1 177.3 23.7 47.5 Forestiera angustifolia 70 11.7 31 12.1 112.5 15.0 38.8 Celtis pallida 60 10.0 26 10.1 45.7 6.1 26.2 Celtis laevigata 40 6.7 18 7.0 19.8 2.6 16.3 Diospyros texana 50 8.3 6 2.3 16.4 2.2 12.8 Lantana camara 30 5.0 12 4.7 14.7 2.0 11.7 Phaulothamnus spinescens 20 3.3 3 1.2 21.4 2.9 7.4 Randia rhagocarpa 20 3.3 3 1.2 10.4 1.4 5.9 Chloroleucon ebano 20 3.3 3 1.2 7.9 1.1 5.6 Sideroxylon celastrinum Total number of plants = 257 20 3.3 2 0.8 1.4 0.2 4.3 170 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 Table 3. Species importance in the ground layer. Imp. value = importance value and is the sum of relative frequency and relative cover. Species Freq. Rel. Cover Rel. Imp. <%) Freq. (%) Cover Value Undisturbed Woodland Rivina humilus 92 13.6 8.67 32.8 46.4 Chromolaena odorata 64 9.5 5.90 22.3 31.8 Cocculus diversifolius 90 13.3 1.11 4.2 17.5 Trixis inula 42 6.2 2.88 10.9 17.1 Tamaulipa azurea 44 6.5 2.49 9.4 15.9 Forestiera angustifolia 40 5.9 0.69 2.6 8.5 Malpighia glabra 16 2.4 1.37 5.2 7.6 Salvia coccinea 36 5.3 0.37 1.4 6.7 Randia rhagocarpa 24 3.6 0.65 2.5 6.1 Verbesina microptera 24 3.6 0.59 2.2 5.8 Amyris texana 32 4.7 0.21 0.8 5.5 Clematis drummondii 18 2.7 0.39 1.5 4.2 19 additional species Total cover = 26.47% Unaided Succession Site Chromolaena odorata 82 9.7 9.15 27.0 36.7 Trixis inula 94 11.2 8.94 25.1 36.3 Clematis drummondii 88 10.4 4.96 14.6 25.0 Salvia coccinea 86 10.2 2.41 7.1 17.3 Cardiospermum halicacabum 72 8.6 2.57 7.6 16.2 Tamaulipa azurea 32 3.8 1.40 4.1 7.9 Zanthoxylum f agar a 50 5.9 0.47 1.4 7.3 Rivina humilus 38 4.5 0.67 2.0 6.5 Matelea reticulata 42 5.0 0.24 0.7 5.7 Rhynchosia minima 36 4.3 0.35 1.0 5.3 Leersia monandra 14 1.7 1.12 3.3 5.0 Ehretia anacua 30 3.6 0.42 1.2 4.8 24 additional species Total cover = 33.87% Facilitated Succession Site Rivina humilus 100 15.4 4.39 26.5 41.9 Tamaulipa azurea 76 11.7 4.45 26.8 38.5 Clematis drummondii 54 8.3 2.12 12.8 21.1 Trixis inula 48 7.4 1.83 11.0 18.4 Ehretia anacua 54 8.3 0.85 5.1 13.4 Chloroleucon ebano 35 5.4 0.53 3.2 8.6 Salvia coccinea 42 6.5 0.29 1.7 8.2 Cocculus diversifolius 32 4.9 0.52 3.1 8.0 Zanthoxylum fagara 27 4.2 0.17 1.0 5.2 Celtis pallida 24 3.7 0.10 0.6 4.3 Matelea reticulata 22 3.4 0.15 0.9 4.3 Serjania brachycarpa 18 additional species Total cover = 16.58% 18 2.8 0.16 1.0 3.8 in the shrub layer at the unaided succession site. Baccharis neglecta (jara dulce), a well known successional stage species, was third in JUDD, LONARD & WAGGERMAN 171 importance at this site. The unaided succession was the only one of the three sites to support jar a dulce. Species presence, richness and importance in the ground layer are compared among sites in Table 3. The unaided succession site had the greatest number of species (36) and the native woodland and facilitated succession sites had equal numbers of species (31). Only the unaided succession site had all four of the species initially planted at the facilitated succession site present in the ground layer. Texas ebony was absent in the ground layer of the native woodland and brasil was absent in the ground layer at the facilitated succession. Rivina humilus (coral ito) was the dominant species in the ground layer at the native woodland and facilitated succession, but Chromolaena odorata (crucita) was dominant at the unaided succession. Eleven of the 13 species present in the tree layer of the native woodland were present in the ground layer at this site, which suggests that the tree species are being perpetuated. Ten of the 13 species present in the tree layer of the native woodland were present in the ground layer of the unaided succession site. Ten is two species more than are currently present in the tree layer at the unaided succession site, and it suggests that the species richness and diversity in the tree layer here may improve in the future. Similar¬ ly, 10 of the 13 species present in the tree layer of the native woodland site were present in the ground layer of the facilitated succession. This is three more species than are currently present in the tree layer at the site. Again, this suggests that species richness and diversity of the tree layer at the facilitated succession site may improve in the future. Nine (facilitated succession) to 10 (native woodland and unaided succession) of the tree species also were represented in the shrub stage. Only three species ( Acacia greggii , Havardia pallens and Prosopis glandulosa) were not present in the shrub stage. Acacia greggii and H. pallens were each represented by a single individual in the quadrats. Prosopis glandulosa was represented by two individuals. Table 4 shows a comparison of coefficients of similarity among layers and sites. There was greater similarity in species in the tree and shrub layers between the native woodland and the facilitated succession than there was between the native woodland and the unaided succession. Conversely, similarity in the ground layer was greater between the native woodland and the unaided succession. The two successional sites were more similar to each other in the tree layer and ground layer than either was to the native forest. 172 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 Table 4. Comparison of coefficients of similarity among layers and treatments. Comparison Tree Layer Shrub Layer Ground Layer Undisturbed Woodland and Unaided Succession .571 .714 .746 Undisturbed Woodland and Facilitated Succession .700 .857 .716 Unaided Succession and Facilitated Succession .800 .750 .774 Table 5 shows a comparison of species richness, evenness, species diversity and variance of the species diversity among sites and vegetation layers. Table 6 shows the t- values, degrees of freedom and probabilities for comparisons between sites within layers. Evenness was greater than .75 in all cases and was relatively similar among sites within a layer. Generally, evenness was inversely related to species richness among the layers with evenness being highest for the tree layer and lowest for the ground layer. Species diversity was greater in the ground layer because of the greater species richness there. Within the tree layer, species diversity was significantly greater in the undisturbed woodland than in either of the succession sites (Tables 5 & 6). In the shrub layer, species diversity in the undisturbed woodland was significantly greater than in the facilitated succession site. There were no significant differences in species diversity among sites in the ground layer. Discussion Mesquite is often depicted as a dominant species in the South Texas Plains ( e.g. Kuchler 1964). Clearly, this is not consistently the case in the Lower Rio Grande Valley. This study provides the first quantifica¬ tion of importance of all species comprising a thorn woodland communi¬ ty in the Lower Rio Grande Valley of Texas. It shows that the domi¬ nant tree species in the native woodland is anacua and that mesquite is scarce at this locality. Mesquite is the tallest of the trees, however, and this might tend to make it conspicuous even if less abundant than other species. Vora (1990) gives a canopy cover value for the dominant species (mesquite) in a woodland community at a site 5 km west of Mission, JUDD, LONARD & WAGGERMAN 173 Table 5. Comparison of species richness (Sp), eveness (J1), species diversity (H1) and variance (s2) of the species diversity index among treatments and vegetation layers. Undisturbed Woodland Unaided Secondary Succession Facilitated Succession Layer Sp J' H1 s2 Sp J1 H1 s2 Sp J1 H1 s2 Tree 13 .892 .944 .000365 8 .813 .734 .000260 7 .918 .776 .000178 Shrub 16 .843 1.016 .000308 12 .859 .927 .000324 12 .812 .876 .000456 Ground 31 .773 1.153 .001142 36 .740 1.152 .001130 31 .755 1.126 .001149 Hidalgo County, Texas, but he does not provide a measure of abundance for other species in the community. He stated that the overall density of overstory trees was 250 to 380 per ha. A much higher density of trees of 2,870 per ha was found in this study. The difference may be due, in part, to differences in defining what constituted a tree. A 3.0 m or taller height was used as the criterion in this study. Vora (1990) does not state what criterion he used, but he reports that tree height ranged from 5.3 to 9.6 m. Using a lower limit of tree height of 5.3 m would have markedly reduced the density of trees in this current study. The facilitation was not done as part of a succession experiment, rather it was done to create white- winged dove nesting habitat quickly. Consequently, there was no contemporaneous unaided control. How¬ ever, the immediate proximity of the undisturbed native woodland and the subsequent abandonment of adjacent farmland provided a mature woodland control and a non-contemporaneous unaided control site. In 40 years, the facilitated succession has not achieved species richness or species diversity of trees and shrubs equal to that of native woodland. Indeed, the facilitated succession site is no better in species diversity and richness than an unaided succession site that has had only 27 years of succession. However, composition of the facilitated succession is more similar to the native woodland than is the unaided succession. The native woodland and facilitated succession have the same dominant species in the tree and ground layers, while none of the dominants is the same in the native woodland and unaided succession. Thus, based on similarity of dominant species and similarity of all species, there is greater similarity between the native woodland and the facilitated succession in the shrub and tree layers than between the native woodland and unaided succession. Whether this relationship will hold after the unaided succession has had 13 more years for development remains to be seen. 174 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 2, 2002 Table 6. Student’s r-tests for comparisons of species diversity between treatments within vegetation layers. H1 and s2 values are given in Table 5. Comparison t df P Tree layer Undisturbed Woodland vs Facilitated Succession 7.120 656 <.001 Undisturbed Woodland vs Unaided Succession 8.400 592 <.001 Facilitated Succession vs Unaided Succession 2.000 599 <.05 Shrub layer Undisturbed Woodland vs Facilitated Succession 5.000 583 <.001 Undisturbed Woodland vs Unaided Succession 1.486 399 >.05 Facilitated Succession vs Unaided Succession 1.821 399 >.05 Ground layer Undisturbed Woodland vs Facilitated Succession 1.563 400 >.05 Undisturbed Woodland vs Unaided Succession .021 400 >.05 Facilitated Succession vs Unaided Succession .054 400 >.05 Vora (1990) reported jara dulce, buffelgrass and bermudagrass were dominant species after five years in an old-field succession near Mission, Texas. Jara dulce was third in importance in the tree layer of the unaided succession, but was not present in the facilitated succession or the native woodland. Thus, it appears that jara dulce is eliminated from the woodland succession between 27 and 40 years after initiation of the succession. Clearly, the unaided succession site has not reached the composition or structure of the native woodland in 27 years. And, while the facili- JUDD, LONARD & WAGGERMAN 175 tated succession site has reached a composition and structure similar to that of native woodland it lacks the diversity provided by scarce to rare species that is present in the native woodland. Three lines of evidence suggest that the unaided succession site is still undergoing succession. First, jara dulce is present and it is known to be a successional stage that appears within five years and then does not persist into mature com¬ munities (Vora 1990). Second, all but one of the species present in the tree layer were shorter than their counterparts in the native woodland. Third, anacua was second in importance rather than the dominant as it was in the native woodland. It is thought that establishing tree and shrub cover on newly acquired agricultural fields will provide perches for birds, which will bring in seeds of many additional species (Vora 1992). At a locality 175 km north of the Rio Grande near Alice, Jim Wells County, Texas, Archer et al. (1988) demonstrated that mesquite trees serve as such foci for bird disseminated seeds of other woody species. The resulting tree and shrub clusters eventually coalesce to form closed canopy woodlands (Archer 1989; 1990). In the facilitated succession, three species have been added to the tree layer: hackberry, Ulmus crassifolia (cedar elm) and colima. Colima and hackberry have become more abundant than any of the species originally planted except for anacua. Another six species of shrubs have been added to the facilitated succession. Based on data provided by Van Auken & Bush (1985) for a community on a terrace of the San Antonio River about 380 km north of LPWMA, it might take up to 150 years for a woodland community to reach maturity. However, the dominant species of mature communities there appear to reach dominance in about 30 years. Clearly, anacua has achieved dominace in the facilitated succession at LPWMA in 40 years. However, in the unaided succession it is second in importance. Therefore, these data suggest that dominance by anacua may be expected to occur between 27 and 40 years. Additional data points are needed to ascertain the developmental rate of the woodland succession with respect to species richness and diversity. Acknowledgments We gratefully acknowledge financial assistance provided by Texas Higher Education Coordinating Board Advanced Research Grant 003599-0006-1999. We thank Francisco Chavero, Martin Garcia, Mathew Garcia and Mitchell Sternberg for field assistance. 176 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 Literature Cited Archer, S., C. Scifres, C. R. Bassham & R. Maggio. 1988. Autogenic succession in a sub¬ tropical savanna: conversion of grassland to thorn woodland. Ecol. Monogr. , 58(2): 1 11- 127. Archer, S. 1989. Have southern Texas savannas been converted to woodlands in recent history? American Naturalist, 134(4):545-561 . Archer, S. 1990. Development and stability of grass/woody mosaics in a subtropical savanna parkland, Texas, U.S.A. J. Biogeography, 17(4/5):453-462. Blair, W. F. 1950. The biotic provinces of Texas. Texas J. Sci., 2(1):93-1 17. Brower, J. E., J. H. Zar & C. N. von Ende. 1998. Field and Laboratory Methods for General Ecology. WCB/ McGraw-Hill. Boston, Massachusetts, 273 pp. Canfield, R. H. 1941. Application of the line interception method in sampling range vegetation. J. Forestry, 39(4):388-394. Connell, J. H. & R. O. Slayter. 1977. Mechanisms of succession in natural communities and their role in community stability and organization. American Naturalist, 1 1 1 (982): 1119-1144. Harris, L. D. 1984. The Fragmented Forest. Univ. Chicago Press, Chicago, Illinois, 211 pp. Jahrsdoerfer, S. E. & D. M. Leslie, Jr. 1988. Tamaulipan brushland of the Lower Rio Grande Valley of south Texas: description, human impacts, and management options. U. S. Fish and Wildlife Serv., Biol. Rep., 88(36), 63 pp. Jones, S. D., J. K. Wipff & P. M. Montgomery. 1997. Vascular plants of Texas: a comprehensive checklist including synonymy, bibliography, and index. Univ. Texas Press. Austin. 404 pp. Krebs, C. J. 1999. Ecological Methodology, 2nd Ed., Benjamin/Cummings, 2725 Sand Hill Road, Menlo Park, CA 94025, 620 pp. Kuchler, A. W. 1964. The Potential Natural Vegetation of the Coterminous United States. American Geographical Society Special Publication, Number 36. 116 pp. Lonard, R. I., J. H. Everitt & F. W. Judd. 1991. Woody plants of the Lower Rio Grande Valley, Texas. Misc. Publ. No. 7. Texas Memorial Museum. Univ. Texas at Austin, 179 pp. Sokal, R. R. & F. J. Rohlf. 1981. Biometry. The Principles and Practice of Statistics in Biological Research. 2nd. Ed. W. H. Freeman and Company. New York, New York, 859 pp. Van Auken, O. W. & J. K. Bush. 1985. Secondary succession on terraces of the San Antonio River. Bull. Torrey Botanical Club, 1 12(2): 158-166. Vora, R. S. 1990. Plant communities of the Santa Ana National Wildlife Refuge, Texas. Texas J. Sci., 42(2): 1 15-128. Vora, R. S. 1992. Restoration of native vegetation in the Lower Rio Grande Valley, 1984-87. Restoration & Management Notes, 10(2): 150-157. Vora, R. S. & J. F. Messerly. 1990. Changes in native vegetation following different disturbances in the Lower Rio Grande Valley, Texas. Texas J. Sci., 42(2): 151-158. Williams, D., C. M. Thompson & J. L. Jacobs. 1977. Soil survey of Cameron County, Texas. U.S. Dept. Agric. Soil Conserv. Serv., 92 pp. FWJ at: fjudd@panam.edu TEXAS J. SCI. 54(2): 177-188 MAY, 2002 MINIMUM FLOW CONSIDERATIONS FOR AUTOMATED STORM SAMPLING ON SMALL WATERSHEDS R. Daren Harmel, Kevin W. King, June E. Wolfe* and H. Allen Torbert** USDA-ARS, 808 E. Blackland Road, Temple, Texas 76502 *TAES, 720 E. Blackland Road, Temple, Texas 76502 and **USDA-ARS, P. O, Box 3439, Auburn, Alabama 36830 Abstract.— The issue of a minimum flow threshold (also referred to as enable level) above which to trigger sampling plays an important role in water quality sampling projects; however, guidance on developing appropriate storm sampling strategies for small streams is limited. As a result, arbitrary strategies are used that may not accurately characterize pollutant flux. Therefore, the objectives of this study were to: (1) compare measured nutrient flux data to hypothetical results collected under several alternative minimum flow threshold or enable level scenarios and (2) publish initial guidance on setting minimum flow thresholds for automated storm sampling in small watersheds. Comparison of measured nutrient fluxes for various enable level scenarios illustrated that substantial error is introduced even with relatively small enable level increases. Based on these results, minimum flow thresholds for automated sampling equipment should be set such that even small storms with small increases in flow depth are sampled. In order to manage the number of samples collected, enable levels should be raised only after careful consideration of the resulting consequences. Alternatives for decreasing the number of samples in nutrient flux measurements, such as increasing the time or flow volume between samples or compositing several samples into one collection bottle, introduce substantially less error than does increasing minimum flow thresholds. Monitoring water quality during storm events is becoming increas¬ ingly important in characterization of pollutant loading to water bodies, especially as National Water Quality Inventories (USEPA 1995; USEPA 2000) continue to report that nonpoint source (NPS) pollution adversely impacts rivers, lakes and coastal waters. NPS pollution includes runoff from diffuse sources such as urban areas, farms, and silvicultural operations. Excessive anthropogenic NPS inputs of the macro-nutrients, nitrogen (N) and phosphorus (P) or "cultural eutrophication" can create accelerated algal growth which degrades aquatic ecosystem health, increases water treatment costs and diminishes recreational and aesthetic values (Kolbe & Luedke 1993). The traditional monitoring focus on periodic grab sampling of low flows to characterize point source pollution (discharged from specific locations such as factories and waste water treatment plants) is now often coupled with automated storm flow monitoring to characterize NPS 178 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 2, 2002 pollution. Most commercially available automated samplers contain similar components, including: programmable operation and memory, water level recorder, sample collection pump and sample bottles. Typical storm sampling operation involves setting a minimum flow threshold or enable level to start and finish sampling (either a flow depth or a rainfall depth per specified time) and setting a time or flow interval on which to collect samples after the sampler is triggered. This type of automated storm monitoring is often the cornerstone of small watersheds projects whose objectives are to compare water quality impacts of various land management activities, evaluate water quality improvement following implementation of best management practices and determine annual pollutant fluxes for Total Maximum Daily Load (TMDL) projects (Tate et al. 1999; Robertson & Roerish 1999). On small watershed monitoring projects, however, sampling and fund¬ ing considerations, along with NPS variability, often make it difficult to achieve project objectives (Tate et al. 1999). Budget determination is generally the first step in monitoring projects (Shih et al. 1994). Most sampling proposals specify a maximum number of storms that will be sampled or a maximum number of samples that will be collected, so that a reasonable sampling expectation can be met. Service and mainte¬ nance of automated sampling equipment is labor intensive and expensive, and cost considerations often limit the number of samples that can be collected and analyzed (Robertson & Roerish 1999; Dissmeyer 1994). Another consideration in developing a sampling scheme is the number of samples that can be collected and analyzed by a laboratory in a reasonable time frame (Novotny & Olem 1994). Since a large portion of the cost of a monitoring program is directly related to the number of samples, determination of a proper minimum flow threshold and sample frequency is important in achieving objectives within budget limitations. A high minimum flow threshold and/or low frequency sampling by¬ passes important information and may lengthen the project duration (Novotny & Olem 1994; Shih et al. 1994). However, a low minimum flow threshold and/or high frequency sampling may be inhibited by available financial and laboratory resources. Guidance on developing storm sampling strategies for small streams is limited, but examples for larger perennial streams and rivers are presented by Robertson & Roerish (1999). The United States Geological Survey NPS program in Wisconsin collects 100 to 200 fixed interval HARMEL ET AL. 179 grab samples and storm flow samples per year for small streams (water¬ sheds less than 100 km2). The typical National Water Quality Assess¬ ment strategy collects monthly samples supplemented by four to eight storm samples per year for about 2.5 yr. For larger streams and rivers, precision and accuracy increase with sampling frequency in almost all cases. In smaller watersheds, which are typically more variable in their response than larger ones, more intensive sampling is generally needed to achieve precise and accurate load estimates (Richards & Holloway 1987). Comparisons of specific automated sampling alternatives are also limited. However, issues of discrete (one sample per bottle) versus composite sampling (several samples per bottle) and flow-weighted (based on flow volume) versus time- weighted sampling (based on time intervals) have been addressed by King & Harmel (2001); Shih et al. (1994); Miller et al. (2000) and others. One important question that has not received attention is what storm size should be sampled, which translates into how many storms are sampled. As stated earlier, this issue of a minimum flow threshold above which to trigger sampling plays an important role in developing sampling strategies. However, without published studies on the impact of setting enable levels, arbitrary decisions are made. General guidance on this issue indicated that for determination of annual storm loads, storms with rain exceeding 25 mm/hr or runoff exceeding 13 mm should be sampled and that generally three to five storms per year create about 75% of the annual runoff (Slade pers. comm.). Tate et al. (1999) state that a majority of annual flow and NPS loading occurs during four to six storms per year on California rangelands. For large rivers, commonly as much as 80% of annual NPS load is contributed by 20% of flows (Richards & Holloway 1987). Richards & Holloway (1987) indicated that assessment of the ade¬ quacy of sampling programs for large rivers is needed. That need also exists for small streams, especially since numerous small watershed monitoring programs are underway with limited assessment of sampling program adequacy. No published guidance is available on setting minimum flow thresholds. If they are set too low, samples will be taken on every runoff event even though no significant NPS load is trans¬ ported. In this case, analysis cost and personnel time will be wasted. If enable levels are set too high, substantial portions of runoff events and 180 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 2, 2002 Table 1. Characteristics of watershed study sites. Traditional Precision Airport Mixed Urban Area 5.7 ha 9.1 ha 37.5 ha 66.5 ha Slope 1 -5% 1 - 5% 1 -4% 1 - 8% Soil texture Clay Clay 70% Impervious, silty clay to sandy clay loam 12% Impervious, silty clay to sandy clay loam Landuse Corn Corn Airport Airport, golf course, residential Land management Conventionally- applied fertilizer, terraces, residue management Precision applied fertilizer, terraces, residue management Mowing, limited fertilizer and pesticide use Mowing, aeration, moderate fertilizer and pesticide use, irrigation Flow channel Ephemeral - grass waterway Ephemeral - grass waterway Small perennial stream Irrigation return flow supplements small perennial stream possibly entire events will not be sampled, thus valuable information will be missed. Therefore, the objectives of this study were to: (1) compare measured N03 + N02-N load data to hypothetical load data collected under various enable level scenarios and (2) produce initial guidance on setting minimum flow thresholds for automated storm sampling in small watersheds. Materials and Methods Study site description.— Runoff and water quality data from two nutrient load studies on four watersheds ranging from 5.7 to 66.5 ha in central Texas were used in this analysis (Table 1). Two were agricul¬ tural watersheds located 3 km east of Temple, Texas, and two were urban watersheds in Austin, Texas. The Austin/Temple area receives 813 to 889 mm normal annual precipitation, has an average of 273 growing season days per year, and average maximum daily temperatures from 15°C in January to 35 °C in August (NO A A 1999). Flow measurement and water quality sampling— To monitor surface runoff on the agricultural watersheds near Temple, Texas, a 0.61 m H-flume equipped with an ISCO 4230/3700 flow meter and sampler HARMEL ET AL. 181 system was installed at the "outlet" of each field. An ISCO 674 rain gauge and two HOBO rain event recorders were also installed on site to record rainfall data. From February 1999 through January 2001, flow rates were recorded every five minutes during runoff events. Time- weighted, composite samples with four 200 mL samples per bottle were collected automatically during runoff events. Samplers were pro¬ grammed to sample all runoff events with adequate flow depth to sub¬ merge the sampler intake (approximately 38 mm water depth) and allow sample collection. To provide adequate resolution in short duration events and adequate sampling capacity for longer events, samples were taken in five min intervals for 65 min, 15 min intervals for the next 660 min, and 30 min intervals for the final 1200 min. Similar monitoring strategies were used to measure surface runoff on the urban sites in Austin, Texas. An ISCO 6700 automatic sampler, an ISCO 4150 area velocity flow logger, and an ISCO 674 rain gauge were installed at each site. Two round culverts drain the airport site, and a box culvert drains the mixed urban site. From April 1998 through March 2000, flow rates were recorded every 15 minutes during runoff events. Time- weighted composite samples with six 150 mL samples per bottle were collected automatically during runoff events. As with the agricultural sites, samplers were programmed to sample runoff events with adequate flow depth to submerge the sample intake (38 mm water depth) and allow sample collection. Samples were taken at five min intervals for 120 min, 15 min intervals for the next 720 min, 30 min intervals for the next 1440 min, and 60 min intervals for the next 1440 min. Samples were collected within 48 hr of runoff events, acidified, iced and transported to the laboratory where they were stored at 4°C prior to analysis. Samples were analyzed for dissolved nitrate plus nitrite nitrogen (N03 + N02-N) concentrations using a Technicon Autoanalyzer IIC (Technicon Instruments Corp., Tarrytown, New York) and colori¬ metric methods published by Technicon Industrial Systems (1973). For each of the four watersheds, measured dissolved N03 + N02-N loads were determined by multiplying measured nutrient concentrations by corresponding flow volumes and summing these incremental loads for the duration of the runoff event. This measured load was then compared to loads that would have been measured for increased enable levels. For 182 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 Figure 1. Sample storm illustrating duration of sample collection for various enable levels. the 0.61 m H-flumes on the agricultural watersheds, increased enable levels ranged from 38 to 305 mm (0.001 to 0.06 m3/s). Increased enable levels ranged from 137 to 762 mm (0.02 to 0.49 m3/s) for the airport site and from 519 to 1067 mm (0.04 to 1.06 m3/s) for the mixed urban site. An example storm is presented in Figure 1 to illustrate the duration of sample collection for various hypothetical enable levels. Results and Discussion Runoff events . — Dissolved N03 + N02-N loads for each site were analyzed for a total of 122 measured runoff events over two years. A summary of rainfall and runoff data for events in which samples and flow rate data were collected from both sites in the urban and agricul¬ tural watersheds is presented in Table 2. A wide range of rainfall depths and intensities, runoff volumes, and peak flow rates occurred during the study period. Results from this study match well with information provided by Slade (pers. comm.) and other studies such as Tate et al. (1999) that generally report that three to six events per year create about 75% of the annual storm runoff and NPS load. Our results for these study sites HARMEL ET AL. 183 Table 2. Properties of rainfall and runoff events. Traditional Precision Airport Mixed Urban Number of runoff events 24 18 40 40 Peak flows (m3/s) 0.00 - 0.32 0.00 - 0.39 0.02 - 4.91 0.03 - 9.82 Runoff volumes (m3) 0.14 - 946 0.15 - 2260 39 - 77000 109 - 89000 Runoff depths (mm) 0.00 - 36 0.00 - 25 0.10 - 205 0.20 - 134 Rainfall (mm) 8 - 63 9 - 63 5 - 227 4 - 187 Max 15 min rainfall (mm) 19 19 26 26 showed that three to six events per year produced on average from 74 to 87% of the N03 + N02-N load and that between 64 and 100% of the annual load could have been captured by sampling only the largest six storms each year (Table 3). As enable levels increase, an increasing amount of pollutant flux is not captured; therefore, increasing enable levels results in increased error compared to the true or total load. To quantify these increases, relative errors (percent deviation from the total measured load) and absolute errors (magnitude of deviation from the total measured load) were calculated. Figures 2 and 3 illustrate that errors increase rapidly as enable levels increase, especially for the smaller watersheds. Errors for the smaller agricultural watersheds were substantial even for small increases in enable levels because small increases in enable level resulted in relatively large increases in flow rates and because large changes in nutrient concentration occurred during the storms events; therefore, sub¬ stantial flow volume and nutrient flux were not sampled with increased minimum flow thresholds. In most water quality sampling projects, appropriate sampling to adequately measure loads must be conducted within the constraint of limited project resources. To reduce analysis costs and overcome laboratory time and personnel limitations, the number of samples can be managed by raising enable levels, increasing duration or flow volume between samples and/or compositing several samples together. How¬ ever, when each of these adjustments are made errors in pollutant flux measurements increase. Based on the results of this study and compari¬ sons to King & Harmel (2001), enable levels should be raised only after 184 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 2, 2002 Table 3. Annual N03±N02-N loads determined by measuring the largest storm events. Measure only the Largest (Number of events) Percent of Measured Annual Load (Average) (Standard deviation) (Range) 1 49 ±25 16 - 89 2 64 ±26 27 - 97 3 74 ±23 38 - 98 4 80 ±20 49 - 100 5 84 ±17 58 - 100 6 98 ±15 64 - 100 careful consideration of the resulting consequences, since small increases in enable levels resulted in large errors. King & Harmel (2001) showed that increasing the duration between samples from 5 min to 15 min, which reduced the number of samples by 66 % , resulted in less than 1 % average increases in relative error. Even when samples were composi¬ ted up to six samples per bottle, which further reduced samples numbers by 83%, less than 20% average increases in relative error occurred. In contrast to relatively small increases in relative error for increased duration and flow volume presented by King & Harmel (2001), relative errors increased rapidly when minimum flow thresholds were raised for the watersheds in this study. Figure 4 illustrates that less error is introduced with corresponding reduction in sample numbers by increas¬ ing duration or flow volume between samples, with further reduction possible with composite sampling. This figure presents the most valua¬ ble result of these analyses: alternative strategies are recommended over raising minimum flow thresholds. Minimum flow thresholds should be set at low levels, such that even small storms with small increases in flow depth are sampled. On watersheds of the size studied (6 to 67 ha), minimum flow thresholds of 0.001 to 0.04 m3/s are recommended. Conclusions As human population grows and water resources increase in value from a water supply and an aquatic ecosystem standpoint, accurate characterization of water quality will become more important. In order to correctly quantify total water quality constituent fluxes, the traditional methodology of periodic low flow grab sampling to characterize point sources must be coupled with storm flow monitoring to characterize HARMEL ET AL. 185 Figure 2. Relative and absolute errors the small agricultural watersheds for various minimum flow thresholds. - error (kg/ha) - mixed urban ------- error (kg/ha) - airport — a — error (%) - mixed urban — x — error (%) - airport 0 0.3 0.6 0.9 1.2 minimum flow threshold (m3/s) Figure 3. Relative and absolute errors the larger urban watersheds for various minimum flow thresholds. 186 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 2, 2002 Figure 4. Comparison of relative errors for various sampling strategies to manage the number of samples. nonpoint sources. Since guidance on developing storm sampling strate¬ gies for small streams is limited especially in light of resource con¬ straints in most monitoring projects, appropriate guidance is needed to develop sampling strategies that accurately characterize pollutant flux within budget resources. Guidance such as presented in this study should assist monitoring program developers in setting minimum flow thresholds for automated storm sampling in small watersheds. Comparison of measured nutrient fluxes to hypothetical fluxes col¬ lected under various enable level scenarios in this study showed that substantial error is introduced as minimum flow thresholds are in¬ creased. Based on this comparison, minimum flow thresholds for automated sampling equipment should be programmed such that even small storms with small increases in flow depth are sampled. On smaller watersheds, minimum flow thresholds of 0.001 to 0.04 m3/s are recommended. In order to manage the number of samples collected, enable levels should not be raised above these levels without careful consideration of consequences. Alternatives for managing sample HARMEL ET AL. 187 numbers, such as increasing the time or flow volume between samples, or compositing several samples, introduce substantially less error in nutrient flux measurements for the watersheds studied. Acknowledgments We would like to thank Kirk Dean, PhD, Principal Scientist, Parsons Engineering Science; Christine Kolbe, Aquatic Scientist, Texas Natural Resource Conservation Commission (TNRCC); and Roger Miranda, Geochemist, TNRCC for review of an earlier version of this manuscript. Raymond M. Slade, Jr., hydrologist and Texas District Surface-Water Specialist for the U.S Geological Survey, also deserves credit for providing valuable insight and information on the subject of storm water sampling. Trade names in this manuscript are included for the benefit of the reader and do not imply endorsement by USD A. Literature Cited Dissmeyer, G. E. 1994. Evaluating the effectiveness of forestry best management practices in meeting water quality goals or standards. USDA Forest Service, Southern Region, Atlanta, Georgia Misc. Pub. 1520. King, K. W. & R. D. Harmel. 2001. Considerations in selecting a water quality sampling strategy. ASAE Paper No. 01-2134. St. Joseph, Michigan: ASAE. Kolbe, C. M. & M. W. Luedke. 1993. A guide to freshwater ecology. Texas Natural Resource Conservation Commission. GI-34. Miller, P. S., B. A. Engel & R. H. Mohtar. 2000. Sampling theory and mass load estimation from watershed water quality data. ASAE Paper No. 00-3050. St. Joseph, Michigan: ASAE. National Oceanic and Atmospheric Administration. 1999. Climatological Data Annual Summary Texas 1999. Vol 104. No. 13. Novotny, V. & H. Olem. 1994. Water quality: prevention, identification, and management of diffuse pollution. New York, Van Nostrand Reinhold, 1072 pp. Richards, R. P. & J. Holloway. 1987. Monte Carlo studies of sampling strategies for estimating tributary loads. Water Resources Research, 23(10): 1939-1948. Robertson, D. M. & E. D. Roerish. 1999. Influence of various water quality sampling strategies on load estimates for small streams. Water Resources Research, 35(12):3747-3759. Shih, G., W. Abtew & J. Obeysekera. 1994. Accuracy of nutrient runoff load calculations using time-composite sampling. Transactions of ASAE, 37(2):419-429. Tate, K. W., R. A. Dahlgren, M. J. Singer, B. Allen-Diaz & E. R. Atwill. 1999. Timing, frequency of sampling affect accuracy of water-quality monitoring. California Agriculture, 53(6):44-49. Technicon Industrial Systems. 1973. Nitrate and nitrite in water and waste water. Industrial method no. 100-70w. Technicon Instruments Corp., Tarrytown, New York, 3 PP- United States Environmental Protection Agency. 1995. National water quality inventory 1994 report to Congress. USEPA 841-R-95-005. USEPA, Office of Water, Washington, 188 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 D.C., 572 pp. United States Environmental Protection Agency. 2000. National water quality inventory 1998 report to Congress. USEPA 841-R-00-001 . USEPA, Office of Water, Washington, D.C., 413 pp. RDH at: dharmel@brc.tamus.edu TEXAS J. SCI. 54(2), MAY, 2002 189 GENERAL NOTES REEXAMINATION OF THE RANGE FOR THE NORTHERN PYGMY MOUSE, BAIOMYS TAYLORI (RODENTIA: MURID AE), IN NORTHEASTERN TEXAS Joel G. Brant* and Robert C. Dowler Department of Biology, Angelo State University San Angelo, Texas 76909 * Current address Department of Biological Sciences and Museum Texas Tech University, Lubbock, Texas 79409 The northern pygmy mouse, Baiornys taylori, is a southern species that reaches its northern distributional limits in Texas, New Mexico and Oklahoma (Choate et al. 1990; Stuart & Scott 1992; Tumlison et al. 1993). They usually are found in association with cotton rats (Sigmodon) and harvest mice ( Reithrodontomys ) , and prefer grassy areas such as old fields, pastures and along railroads or highways (Schmidly 1983). Their distribution in Texas currently is thought to range from along the coast and throughout the central portions of the state to western Texas, excluding the Trans-Pecos and northeastern Texas (Davis & Schmidly 1994). Fieldwork in Nacogdoches County has documented the northern pygmy mouse at one site beyond the distribution reported by Davis & Schmidly (1994). On 9 and 10 October 1999, eight Baiomys taylori were collected from Nacogdoches County, Alazon Bayou Wildlife Management Area, near the Angelina River (31° 29.7’ N, 94° 45.2’ W). Specimens were collected in Sherman live traps set in an old field bordered by a pine forest. The specimens were deposited in the Angelo State Natural History Collections (ASNHC 11054-11061) as museum study skins and skeletons. Average measurements (range shown in parentheses), in mm, for the specimens (ft = 8) were: total length, 102 (91-1 10); length of tail, 41 (39-44); length of hind foot, 13 (13-14); length of ear, 10 (10-12). The specimens had an average mass of 7.7 grams (6g-l lg). One female had two embryos, one in each uterine branch, measuring 10 mm (crown-rump length). Other rodents collected at this locality were Sigmodon hispidus , Reithrodontomys Julvescens , Geomys breviceps and Ochrotomys nuttalli. 190 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 2, 2002 These records are approximately 50 miles (80 km) northeast of the reported range for B. taylori by Davis & Schmidly (1994) and 85 miles (136 km) south of the Harrison County record reported by Baccus et al. (1971). The range of this species has been continually expanding north¬ ward (Choate et al. 1990) since it was described from Duval County in southern Texas (Thomas 1887). Bailey (1905) indicated that B. taylori was restricted to southern and coastal Texas. Since then it has been moving northwest into central Texas and the Llano Estacado (Davis 1974; Stangl et al. 1983; Jones & Manning 1989; Pitts & Smolen 1989; Choate et al. 1990). Recently it has extended its range to include New Mexico and Oklahoma (Stangl & Dalquest 1986; Stuart & Scott 1992; Tumlison et al. 1993). The extent of the northeastern range of this species has been unclear in the last 30 years. Baccus et al. (1971) reported a specimen from Harrison County, Texas. This specimen extended the range of B. taylori to the northeast by over 100 miles (>160 km). Davis (1974) reported the range of B. taylori in Texas as extending from Cooke County to Jefferson County excluding northeastern Texas and the Harrison County record (Fig. 1). In Davis & Schmidly (1994), the reported range was similar to that given in Davis (1974), with the Harrison County record listed as outside the range of the species. Hall (1981), Schmidly (1983) and Cameron (1999) depicted the range of B . taylori as including most of northeastern Texas, extending from Cooke County east to Harrison County and continuing south parallel to the Louisiana border to Orange County, Texas (Fig. 1). Recent records from Anderson County (Roberts et al. 1997) and the records reported in this study appear to support the range reported by Hall (1981), Schmidly (1983) and Cameron (1999). Further research into the northeastern extent of the range of Baiomys taylori is needed. Acknowledgments Thanks are due to the Texas Parks and Wildlife Department for allowing research and collection at Alazon Bayou Wildlife Management Area. We thank Michael Poteet and other wildlife biologists at Alazon Bayou Wildlife Management Area, for assistance while at the site. Mark Boyle, Stephanie Franklin, Eddie Lyons, Marisol Salazar, Bill Scoggins and Clay White assisted in field collecting. Richard Humbertson and Brandy Martin were instrumental in preparing and cataloging specimens into the Angelo State Natural History Collections. TEXAS J. SCI. 54(2), MAY, 2002 191 Figure 1. Map showing the northeastern distributional limit of Baiomys taylori in Texas. Solid line indicates the range limit proposed by Davis (1974) and Davis & Schmidly (1994). Dashed line indicates the range limit proposed by Hall (1981), Schmidly (1983) and Cameron (1999). Solid circles indicate county records for Baiomys taylori and solid triangle represents county records reported in this study. Darin Carroll and Clyde Jones provided comments on an earlier draft of this manuscript. Literature Cited Baccus, J. T., R. E. Greer & G. G. Raun. 1971. Additional records of Baiomys taylori (Rodentia: Cricetidae) for northern Texas. Texas J. Sci., 23(2): 148-149. Bailey, V. 1905. Biological survey of Texas. North Amer. Fauna 25: 1-222. Cameron, G. N. 1999. Northern pygmy mouse | Baiomys taylori. Fp. 586-587, in The Smithsonian book of North American mammals (D. E. Wilson and S. Ruff, eds.). Smithsonian Institution Press, Washington D. C., xxv -I- 750 pp. Choate, L. L., J. K. Jones, Jr., R. W. Manning & C. Jones. 1990. Westward ho: continued dispersal of the pygmy mouse, Baiomys taylori , on the Llano Estacado and in adjacent areas of Texas. Occas. Papers Mus., Texas Tech Univ., 134:1-8. Davis, W. B. 1974. The mammals of Texas. Bull. Texas Parks and Wildlife Dept., 41:1-294. Davis, W. B. & D. J. Schmidly. 1994. The mammals of Texas. Texas Parks and Wildlife 192 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 2, 2002 Press, Austin, x + 338 pp. Hall, E. R. 1981. The mammals of North America. John Wiley & Sons, Inc., New York, 1:1-600 + 90. Jones, J. K., Jr. & R. W. Manning. 1989. The northern pygmy mouse, Baiomys taylori, on the Texas Llano Estacado. Texas J. Sci., 41(1): 110. Pitts, R. M. & M. J. Smolen. 1989. Status of Baiomys taylori in Texas, with new localities of record in the southern part of the state. Texas J. Sci., 41(l):85-88. Roberts, H. R., T. W. Jolley, L. L. Peppers, J. C. Cathey, R. Martinez, J. A. Peppers, A. L. Bates & R. D. Bradley. 1997. Noteworthy records of small mammals in Texas. Occas. Papers Mus., Texas Tech Univ., 172:1-7. Schmidly, D. J. 1983. Texas mammals east of the Balcones fault zone. Texas A&M Univ. Press, College Station, Texas xvii + 400 pp. Stangl, F. B., Jr. & W. W. Dalquest. 1986. Two noteworthy records of Oklahoma mammals. Southwestern Nat., 31(1): 123-124. Stangl, F. B. , Jr., B. F. Koop & C. S. Hood. 1983. Occurrence of Baiomys taylori (Rodentia: Cricetidae) on the Texas High Plains. Occas. Papers Mus., Texas Tech Univ., 85:1-4. Stuart, J. N. & N. J. Scott, Jr. 1992. Range extension of the northern pygmy mouse, Baiomys taylori , in New Mexico. Texas J. Sci., 44(4): 487-489. Thomas, O. 1887. Diagnosis of a new species of Hesperomys from North America. Ann. Mag. Nat. Hist., ser. 5, 19:66. Tumlison, R., V. R. McDaniel & J. G. Duffy. 1993. Further extension of the range of the northern pygmy mouse, Baiomys taylori , in southwestern Oklahoma. Southwestern Nat. , 38(3):285-286. RCD at: robert.dowler@angelo.edu THE TEXAS ACADEMY OF SCIENCE, 2002-2003 OFFICERS President: President Elect : Vice-President: Immediate Past President: Executive Secretary: Corresponding Secretary: Managing Editor: Manuscript Editor: Treasurer: AAAS Council Representative: DIRECTORS 2000 Bobby L. Wilson, Texas Southern University John P. Riola, Texaco Exploration 2001 David S. Marsh, Angelo State University Felipe Chavez-Ramirez, International Crane Foundation 2002 Sushma Krishnamurthy, Texas A&M International University Raymon D. Mathews, Jr., Texas Water Development Board SECTIONAL CHAIRPERSONS Anthropology: Roy B. Brown, Instituto Nacional de Antropologia y Histbria Biological Science: David S. Marsh, Angelo State University Botany: Cyndy Galloway, Texas A&M University-Kingsville Chemistry: Mary A. Kopecki-Fjetland, St. Edward’s University Computer Science: John T. Sieben, Texas Lutheran University Conservation and Management: Andrew C. Kasner, Texas A&M University Environmental Science: Cindy Contreras, Texas Parks & Wildlife Department Freshwater and Marine Science: Hudson DeYoe, University of Texas-Pan American Geology and Geography: Jeff Pittman, Lamar University Mathematics: Benjamin J. Sultenfuss, Stephen F. Austin State University Physics: Robert Hamilton, Angelo State University Science Education: Julie F. Westerlund, Southwest Texas State University Systematics and Evolutionary Biology: Allan Hook, St. Edward’s University Terrestrial Ecology: Monte Thies, Sam Houston State University Threatened or Endangered Species: Donald L. Koehler, Austin Parks and Recreation Dept. COUNSELORS Collegiate Academy: Jim Mills, St. Edward’s University Junior Academy: Vince Schielack, Texas A&M University Nancy Magnussen, Texas A&M University Larry D. McKinney, Texas Parks and Wildlife Department John T. Sieben, Texas Lutheran University John A. Ward, Brook Army Medical Center David R. Cecil, Texas A&M University-Kingsville Fred Stevens, Schreiner University Deborah D. Hettinger, Texas Lutheran University Ned E. Strenth, Angelo State University Patrick L. Odell, Baylor University James W. Westgate, Lamar University Sandra S. West, Southwest Texas State University SMITHSONIAN INSTITUTION LIBRARIES 3 9088 01402 2842 THE TEXAS JOURNAL OF SCIENCE PERIODICAL POSTAGE PrinTech, Box 43151 PAID AT LUBBOCK Lubbock, Texas 79409-3 1 5 1 TEXAS 79402 RETURN SERVICE REQUESTED 002AAU7904 ROOM 25 NHB SMITHSONIAN INSTITUTION LIBRARY-ACQUISITIONS *(SMIV) WASHINGTON, DC 20560-0154 o? -W |OH THE TEXAS JOURNAL OF SCIENCE Volume 54 Number 3 August 2002 PrinTech 8/02 pt345 1 GENERAL INFORMATION MEMBERSHIP.— Any person or member of any group engaged in scientific work or interested in the promotion of science is eligible for membership in The Texas Academy of Science. For more information regarding membership, student awards, section chairs and vice- chairs, the annual March meeting and author instructions, please access the Academy’s homepage at: www. texasacademyofscience. org Dues for regular members are $30.00 annually; supporting members, $60.00; sustaining members, $100.00; patron members, $150.00; associate (student) members, $15.00; family members, $35.00; affiliate members, $5.00; emeritus members, $10.00; corporate members, $250.00 annually. Library subscription rate is $50.00 annually. The Texas Journal of Science is a quarterly publication of The Texas Academy of Science and is sent to most members and all subscribers. Payment of dues, changes of address and inquiries regarding missing or back issues should be sent to: Dr. Fred Stevens, Executive Secretary The Texas Academy of Science CMB 5980 Schreiner University Kerrville, Texas 78028-5697 E-mail : FStevens@schreiner . edu AFFILIATED ORGANIZATIONS American Association for the Advancement of Science, Texas Council of Elementary Science Texas Section, American Association of Physics Teachers Texas Section, Mathematical Association of America Texas Section, National Association of Geology Teachers Texas Society of Mammalogists The Texas Journal of Science (ISSN 0040-4403) is published quarterly at Lubbock, Texas, U.S.A. Periodicals postage paid at San Angelo, Texas and additional mailing offices. POSTMASTER: Send address changes and returned copies to The Texas Journal of Science, Dr. Fred Stevens, CMB 5980, Schreiner University, Kerrville, Texas 78028-5697, U.S.A. The known office of publication for The Texas Journal of Science is the Department of Biology, Angelo State University, San Angelo, Texas 76909; Dr. Ned E. Strenth, Managing Editor. THE TEXAS JOURNAL OF SCIENCE Volume 54, No. 3 August, 2002 CONTENTS Response of Herbaceous Vegetation to Summer Fire in the Western South Texas Plains. By Donald C. Ruthven, III and David R. Synatzske . . . 195 Effects of Prescribed Burning on Vegetation and Fuel Loading in Three East Texas State Parks. By Sandra Rideout and Brian P. Oswald . 211 The Vascular Flora of Windham Prairie, Polk County, East Texas. By Larry E. Brown, Kate Hillhouse, Barbara R. MacRoberts and Michael H. MacRoberts . . . . . 227 Noteworthy Plants Associated with the Gulf Coastal Bend of Texas. By /. G. Negrete, C. Galloway and A. D. Nelson . 241 Mutagenic Activity of Idarubicin and Epirubicin in the Bacterium Salmonella typhimurium. By John M. Brumfield and William J. Mackay . . . 249 The Effects of Incubation Temperature on Locomotor Activity in Juvenile Hogna carolinesis (Araneae: Lycosidae). By Fred Punzo and Marie Chapla . . 261 Distributional Records of Mammals from the Permian Basin, Texas. By Joel G. Brant and Clyde Jones . . . . 269 General Notes Noteworthy Records of Bats from the Trans-Pecos Region of Texas. By Jana. L. Higginbotham, Robert S. DeBaca, Joel G. Brant and Clyde Jones . 277 Gastrointestinal Helminths of Gaige’s Tropical Night Lizard, Lepidophyma gaigeae (Sauria: Xantusiidae) from Hidalgo, Mexico. By Stephen R. Goldberg, Charles R. Bursey and Jose L. Camarillo-Rangel . 282 Acknowledgment . 286 Recognition of Member Support . . 287 Annual Meeting Notice for 2003 288 THE TEXAS JOURNAL OF SCIENCE EDITORIAL STAFF Managing Editor: Ned E. Strenth, Angelo State University Manuscript Editor: Robert J. Edwards, University of Texas-Pan American Associate Editor for Botany: Janis K. Bush, The University of Texas at San Antonio Associate Editor for Chemistry: John R. Villarreal, The University of Texas-Pan American Associate Editor for Computer Science: Nelson Passos, Midwestern State University Associate Editor for Environmental Science: Thomas LaPoint, University of North Texas Associate Editor for Geology: Ernest L. Lundelius, University of Texas at Austin Associate Editor for Mathematics and Statistics: E. Donice McCune, Stephen F. Austin State University Associate Editor for Physics: Charles W. Myles, Texas Tech University Manuscripts intended for publication in the Journal should be submitted in TRIPLICATE to: Dr. Robert J. Edwards TJS Manuscript Editor Department of Biology University of Texas-Pan American Edinburg, Texas 78541 redwards@panam.edu Scholarly papers reporting original research results in any field of science, technology or science education will be considered for publication in The Texas Journal of Science. Instructions to authors are published one or more times each year in the Journal on a space-available basis, and also are available from the Manuscript Editor at the above address. They are also available on the Academy’s homepage at: www . texasacademy ofscience . org The Texas Journal ofScience is published quarterly in February, May, August and November for $30 per year (regular membership) by The Texas Academy of Science. Periodical postage rates (ISSN 0040-4403) paid at Lubbock, Texas. Postmaster: Send address changes and returned copies to Dr. Fred Stevens, Executive Secretary, CMB 5980, Schreiner University, Kerrville, Texas 78028- 5697, U.S.A. TEXAS J. SCI. 54(3): 195-210 AUGUST, 2002 RESPONSE OF HERBACEOUS VEGETATION TO SUMMER FIRE IN THE WESTERN SOUTH TEXAS PLAINS Donald C. Rut liven, III and David R. Synatzske Texas Parks and Wildlife Department Chaparral Wildlife Management Area, P.O. Box 115 Artesia Wells, Texas 78001 Abstract. —With increases in wildlife related enterprises and ecological restoration efforts in southern Texas, there is an increased interest in utilizing summer fire to achieve management goals; yet, there is little data on the effects of summer burning on vegetation and wildlife. Herbaceous vegetation diversity, productivity, density and frequency were estimated on five summer burned and five nontreated sites utilizing 20 by 50 cm quadrats . Forb density and frequency was monitored for two growing-seasons postburn. Grass indices were measured three months postburn. Grass and forb yields were estimated in 0.25 m2 plots during the first growing-season postburn. Croton (Croton sp.) responded positively to summer burning during the first growing-season postburn for all indices measured. During the second growing season postburn, Croton densities were similar among treatments. Densities of erect dayflower (Commelina erecta) and beach groundcherry (Physalis cinerascens) were greatest on burned sites throughout the study. Silky evolvulus ( Evolvulus alsinoides) and hoary blackfoot (Melampodium cinereum) were more common on nontreated sites. Grass densities were lowest on burned sites three months postburn, and yields were similar between treatments by the middle of the first postburn growing season. Summer burning does not appear to provide any additional benefits in forb response over dormant-season burning. The long-term effect of a regimented burning regime on vegetation and influence of burn season on wildlife is not clearly understood and warrants further investigation. The Rio Grande Plains of south Texas is the southern-most extension of the Great Plains Grasslands. Fire, along with other climatic variables such as drought, presumably maintained the honey mesquite {Prosopis glandulosa) savannas and interspersed grasslands of pre- European settlement south Texas (Scifres & Hamilton 1993). Frequency of fire appeared to be highly variable and ranged from 5-30 years (Wright & Bailey 1982). Following European settlement, suppression of fire combined with heavy livestock grazing has lead to the current thorn woodlands common throughout southern Texas (Archer et al. 1988; Archer 1994). Beginning in the mid-twentieth century, south Texas landowners began to convert these thorn woodlands back to grasslands to enhance rangelands for livestock production. Mechanical treatments such as root plowing were commonly utilized methods for achieving this goal. Mechanical brush manipulation practices can significantly reduce woody plant cover while increasing herbaceous vegetation (Scifres et al. 1976; 196 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 3, 2002 Bozzo et al. 1992). However, once treated rangelands are re vegetated by woody species, woody plant diversity can be dramatically reduced (Fulbright & Beasom 1987; Ruthven et al. 1993), which may negatively impact diversity of wildlife species. Land ownership and land use practices in south Texas have changed in recent years. The size of individual landholdings has decreased and revenues derived from those properties have become increasingly dependent on wildlife rather than traditional livestock operations. Many wildlife management programs are directed towards game species such as white- tailed deer ( Odocoileus virgininaus) and northern bob white (Colinus virginianus) . In the southeastern United States, prescribed fire has long been utilized to manage habitat for northern bob white (Landers & Mueller 1992). Woody vegetation is a primary component of white¬ tailed deer diet in south Texas (Drawe 1968; Taylor et al. 1997), and prescribed burning in eastern portions of the Rio Grande Plains can reduce brush cover while maintaining woody plant diversity (Box & White 1969). Dormant-season prescribed burning in southern Texas has been shown to increase herbaceous vegetation preferred by wildlife (Hansmire et al. 1988; Ruthven et al. 2000). As a result of its reported benefits, south Texas rangeland managers are beginning to utilize prescribed fire to enhance wildlife habitat. In addition to the rise of wildlife related enterprises, there is growing interest in restoring ecosystems to pre- European settlement conditions. Many proponents of ecological restoration promote the use of summer burning to mimic the occurrence of natural fires. Most perennial grasses generally decrease following summer burns (Scifres & Duncan 1982; Engle et al. 1993; Engle et al. 1998). In Oklahoma, summer fire can increase forb productivity (Engle et al. 1998), while in southern Texas summer prescribed burns appeared to have little affect on forbs (Mayeux & Hamilton 1988). Although effects of summer fire are documented in many ecosystems, little information is available on the response of vegetation and wildlife to growing-season fire in the more xeric areas of the western Rio Grande Plains. The objective or this study was to determine the effects of summer prescribed fire on the diversity, density and productivity of herbaceous vegetation during the first and second growing-seasons post-treatment in the western Rio Grande Plains. It is hypothesized that prescribed burning south Texas rangelands during the growing season will result in enhanced germination and establishment of annual and perennial forbs and decreases of perennial grasses. RUTHVEN & SYNATZSKE 197 Figure 1 . Location of the Chaparral Wildlife Management Area within the South Texas Plains ecological region (stippled area) and the state of Texas. South Texas Plains ecological region boundaries were taken from Hatch et al. (1990). Materials and Methods The study area (Fig. 1) was located on the Chaparral Wildlife Management Area (28° 20’ N, 99° 25’ W) within the western South Texas Plains (Correll & Johnston 1979; Hatch et al. 1990) and northern portion of the Tamaulipan Biotic Province (Blair 1950). Climate is characterized by hot summers and mild winters with an average daily minimum winter (January) temperature of 5°C, an average daily maximum summer (July) temperature of 37° C, a growing season of 249 to 365 days, and average annual precipitation (1951-1978) of 55 cm (Stevens & Arriaga 1985). Average annual precipitation on the study site (1989-1999) was 54 cm. Precipitation patterns are bimodal with peaks occurring in late spring (May to June) and early fall (September to October). Five sites subjected to prescribed burns were paired with five nontreated sites utilizing a randomized block design. Study sites were approximately 2 ha in size. Burned sites were located within larger 198 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 3, 2002 areas that had been burned. Rangeland fire in southern Texas typically produces a mosaic of burned and nonburned areas as a result of uneven fuel loads (Box & White 1969). All study sites received 100% coverage by burns. Fire was applied to burn sites in August 1999. Relative humidity and air temperature, using a sling psychrometer, and surface wind speed, using the Bufort Scale, were estimated before ignition and at the completion of each fire. Weather conditions were relatively constant during all fires with a relative humidity of 32% , temperature of 39 °C and wind speed of 8 kph. Wind direction was variable. Soil moisture was not recorded. All burns were conducted 3 to 5 days following a 23 cm rainfall event (Hurricane Brett) and soil moisture was considered high. Because of variable wind speed and direction during burns and uneven fuel loads, rate of spread and flame height were highly variable and not recorded. Fuel loads appeared to vary within study sites. Adequate fuel loads for burning in western portions of south Texas are > 2,000 kg/ha and study sites met these levels based on visual estimations. All burns were ignited as head fires with drip torches. Soils were similar among sites and consisted of Duval fine sandy loam, gently undulating, Duval loamy fine sand, 0 to 5% slopes, and Dilley fine sandy loam, gently undulating (Stevens & Arriaga 1985; Gabriel et al. 1994). Duval series soils are fine-loamy, mixed, hyper¬ thermic Aridic Haplustalfs and belong to the Sandy Loam range site. Dilley series soils are loamy, mixed, hyperthermic shallow Ustalfic Haplargids and belong to the Shallow Sandy Loam range site. Topo¬ graphy was nearly level to gently sloping and elevation ranged between 177 and 186 m. Vegetation is characterized by a two-phase pattern of shrub clusters scattered throughout a grassland/savanna (Whittaker et al. 1979; Archer et al. 1988). Plant communities were characteristic of the P. glandulosa -granjeno ( Celtis pallida) association (McLendon 1991). Subdominant woody species include twisted acacia ( Acacia schaffneri ), brasil ( Condalia hookeri ) and hog-plum ( Coluhrina texana). Woody plant canopy cover was similar among all sites and averaged 40% (Gabor 1997). Prominent herbaceous species included Lehmann lovegrass ( Eragrostis lehmanniana) , fringed signalgrass (. Brachiaria cilliatissima) , hairy grama (. Bouteloua hirsuta ), croton ( Croton sp.), coreopsis ( Coreposis nuecensoides) , lazydaisy ( Aphanostephus sp.) and partridge pea ( Chamaecrista fasciculata ) (Ruthven et al. 2000; 2002). Pre¬ treatment sampling of herbaceous vegetation was not conducted; however, dominant herbaceous species are generally uniform in distribu- RUTHVEN & SYNATZSKE 199 tion across shrub clusters and the interspace (Whittaker et al. 1979; Ruthven 2001) and the assumption was made that all study sites were similar prior to application of burning treatments. Domestic livestock have grazed the study area since the 18th century (Lehmann 1969). Cattle were the major species of livestock since about 1870, prior to which sheep production dominated from about 1750 to 1870. Before 1969, grazing by cattle was continuous. From 1969 to 1984, livestock managers utilized a four-pasture rest rotation system of cattle grazing. Cattle were absent from the study site during 1984 to 1989. Since 1990, including the timeframe of this study, the study area has been grazed using stocker cattle under a high intensity, low fre¬ quency grazing system during the period October through April. Stock¬ ing rates were considered low to moderate and averaged one Animal Unit per 12 ha. Postburn grazing on the study site has little affect on forb response (Ruthven et al. 2000). Forb and grass density and frequency were estimated by counting individual plants in 50, 20 by 50 cm quadrats placed randomly in each study site. Forbs were sampled in fall (November) 1999 and spring (March- April) 2000 and 2001. Grass density and frequency were estimated in fall 1999. Frequency of occurrence data was used to estimate species richness, diversity and evenness. Forb and grass species diversity (FT) and evenness (J’) was quantified with Shannon’s Index (Chambers & Brown 1983). In June 2000, grass and forb yields were estimated in 10, 0.25 m2 plots randomly placed in each study site. Aboveground biomass was clipped in each plot. Current years’ growth was separated by species, air dried at 40 °C, and weighed to the nearest 0. 1 g. Scientific nomenclature and vernacular names of plants follow Hatch et al. (1990). Forb frequency and density estimates were analyzed with a two-way analysis of variance {AN OVA), with treatment and season as the main effects and a treatment by season interaction. Tukey’s HSD was used to make comparisons of seasonal means. Grass density and frequency and yield data were analyzed by a one-way ANOVA. All statistical comparisons were considered significant at P < 0.05. Results Forb richness and diversity was similar among treatments (Table 1). Richness and diversity varied by season with greatest values in spring 2001 than spring 2000 and fall 1999 and spring 2000 exhibiting greater values than fall 1999. An interaction existed for richness and evenness, 200 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 3, 2002 Table 1. Forb species richness, diversity and evenness by season on summer burned ( n = 5) and nontreated ( n = 5) sites on the Chaparral Wildlife Management Area, Dimmit and La Salle Counties, Texas, 1999-2001. Richness _ Diversity _ Evenness Season/treatment X SD X SD X SD Fall 1999 Burned 10.2 3.4 2.05 0.31 0.91 0.04 Nontreated 17.2 4.5 2.23 0.34 0.79 0.07 Spring 2000 Burned 21.0 3.6 2.44 0.2 0.80 0.04 Nontreated 22.6 5.1 2.67 0.31 0.86 0.02 Spring 2001 Burned 33.2 3.6 2.85 0.18 0.81 0.02 Nontreated 31.0 2.2 2.91 0.04 0.83 0.02 P-value Treatment Season Interaction 0.1408 <0.0001 0.0419 0.1024 <0.0001 0.7443 0.3780 0.2660 0.0001 with greater species on nontreated sites in fall 1999 and greater evenness on burned sites in fall 1999 and nontreated sites in spring 2000. Grass richness ( P = 0.1930), diversity ( P ft 0.6968) and evenness ( P = 0.3972) was similar between burned [12.0 ± 0.9 ( x ± SD), 2.11 ± 0.16 and 0.85 ± 0.04, respectively] and nontreated (13.4 ± 0.9, 2.15 ± 0.16, and 0.83 ± 0.02) sites. Total forb density was similar ( P = 0.6914) among burned (fall 1999, 11.5 ± 6.9 plants/m2; spring 2000, 78.7 ± 22.8; spring 2001, 222.7 + 86.8) and nontreated (fall 1999, 19.9 ± 8; spring 2000, 32 ± 8.9; spring 2001, 243.6 ± 33.8) sites and varied ( P < 0.0001) by season. Forb densities in fall 1999 were similar to spring 2000 with spring 2001 being greater than both fall 1999 and spring 2000. Densities of many dominant (frequency >5%) forbs utilized by livestock and wildlife (Everitt et al. 1999) varied by treatment and season (Table 2). Croton , erect day flower ( Commelina ere eta) and beach groundcherry ( Physalis cinerascens) densities were greatest on burned sites, whereas silky evolvulus ( Evolvulus alsinoides ) and hoary blackfoot ( Melampodium cinereum ) were greatest on nontreated sites. Aphanostephus sp., C. nuecensoides , rose ring Indian blanket ( Gallardia pulchella ), Dillens oxalis ( Oxalis dillenii ) and Hooker plantain ( Plantago hookeriana ) varied seasonally in the order: fall 1999 = spring 2000 < spring 2001. Croton varied by season in the order: spring 2000 > spring 2001 = fall 1999. Density of C. fas ciculata was similar between fall 1999 and spring 2000, spring 2001 was similar to spring 2000, and spring 2001 was greater than fall 1999. Density of E. alsinoides varied by season in the order: fall 1999 = spring 2000 > spring 2001. Density of Physalis cinerascens 3.6 2.2 0.4 0.7 5.0 2.5 1.9 0.7 6.1 2.0 0.9 0.9 <0.0001 0.1459 0.4162 Plantago hookeriana 0 0 0 0 6.2 7.4 1.3 1.1 39.5 39.8 55.2 34.2 0.6528 <0.0001 0.5473 Thelesperma burridgeanum 0 0 0.2 0.4 0.4 0.4 0.1 0.2 10.1 5.6 17.1 9.2 0.1645 <0.0001 0.1378 Thymophylla tenuiloha 0 0 0 0 0 0 0.1 0.2 1.0 0.9 3.2 5.4 0.3543 0.0741 0.4693 Verbena cloverae 0.2 0.2 0.2 0 0.3 0.4 0.1 0.2 1.0 1.1 1.1 1.1 0.8948 0.0137 0.8831 RUTHVEN & SYNATZSKE 201 o o o o o o o o o O o o O o © ►— o o b -* b bv b io o o o o o o o o o O o o p O o o o o o b to b b to SO b o o o o p ! . o p o O o 00 O o O o o as -* b bv b b io b io to bo o © o o p H- o p o © o Ov O o p O o 4 to b b b b •-* bv b io 4 p o o o p o p p to O 1 . w 4- © to co ■4 o Ov t . N> b b b b bv b b bo b b so b so b b O o o o i . o o p Ov © to 4^ O ■4 to o 00 _ io o o p Os to p h- _ o to oo SO to b b b b b io b b b b bo i . p o o , . o to o o o to to p © o co to ON b to b b b b b b b b SO 4 bv b 45- O o to OV © Ov to o as to o o O o O Ov Ov - CO O VO 4 b so 'as -* to bv b to bv b oo b '-4 o b oo tO p o to as o -4 w to , o o o , . O Ov -4 o ov Ov p <1 4 b so b to as b b b b io b b io b b b 4- co , Os co p , to to o u> p o © Ov O Ov 00 p OV CO to © b 4 b bo b b b b b b b b b bo b b bv to to , CO o p o oo o , . o p o CO o CO to © 45- co -4 4 'as b 4 b 4 b bo b to to so io b b p o © o o o O o o © p o p o o © o p © bo -4 4 b to 45- o o Ov OO oo s 8 to Ov £ 8 Ov Ov 4- to 8 OV Ov Ov b 4- -4 b Ov VO 8 to oo 8 8 Ov OV § o SO % as Ov 8 Ov Ov b Ov b to to A p o o A p © o O A o O A o © o o A o o o © A o p b o o b to 45- VO b © VO i U) b © Ov o Ov Ov 8 o b -4 b o o o vO Ov OO b o o to 8 Ov o 8 o o o o 00 8 o b to Ov to o o o b o CO 00 o © o o o o O p O o o o p o o p © p o as 8 to © b to to CO o Ov co o b Ov o 00 to 4 Ov OV to oo b b -4 b to to -4 Ov bv -4 OO 45- o VO OV b to -4 8 8 to OV b CO Ov b 00 to 4 b co 4- vO b VO OV to C/3 o * I X I Table 2. Density (plants/m2) of dominant (frequency >5%) forbs utilized by livestock and wildlife (Everitt et al. 1999) by season on summer burned (n = 5) and nontreated (n = 5) sites on the Chaparral Wildlife Management Area, Dimmit and La Salle counties, Texas, 1999-2001. Fall 1999 Spring 2000 Spring 2001 Burned Nontreated Burned Nontreated Burned Nontreated P-value P-value /’-value 202 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 3, 2002 Table 3. Density (plants/m2) and percent frequency (%) of dominant (frequency >5%) grasses and sedges on summer burned (n = 5) and nontreated (n = 5) sites on the Chaparral Wildlife Management Area, Dimmit and La Salle counties, Texas, November 1999. Density Frequency Burned Nontreated Burned Nontreated Class/species X SD X SD P-value X SD X SD P-value Aristida purpurea 0.6 0.4 0.8 0.4 0.4500 5 4 7 4 0.4034 Cyperus retroflexus 0.1 0.2 0.7 0.4 0.0278 1 2 6 4 0.0569 Brachiaria cilliatissima 1.8 1.6 5.2 2.2 0.0237 13 9 31 16 0.0417 Bouteloua hirsuta 9.0 8.0 21.7 10.1 0.0586 31 22 61 16 0.0456 Chloris cucullata 0.4 0.2 1.4 1.6 0.1950 3 2 10 11 0.2345 Digitaria cognata 4.2 1.1 5.7 1.3 0.0910 24 9 35 9 0.0826 Eragrostis lehmanniana 2.3 1.1 5.1 1.3 0.0071 20 9 30 7 0.0595 Eragrostis sessilispica 0.5 0.4 1.4 2.0 0.3576 4 4 10 16 0.4262 Eragrostis secundiflora 0.8 0.7 1.9 1.6 0.1866 5 4 12 7 0.0792 Paspalum setaceum 5.4 3.6 5.4 2.0 1.0000 28 13 26 4 0.6955 Setaria firmulum 1.0 1.3 0.9 0.7 0.8852 6 7 7 4 0.9205 silvery bladderpod ( Lesquerella argyaea ) was greatest in spring 2001 followed by spring 2000, which was greater than fall 1999. Croton and E. alsinoides exhibited an interaction, with Croton having greatest densities on burned sites in spring 2000 and E. alsinoides densities being greatest on nontreated sites in fall 1999. Grass density was greater (P = 0.0044) on nontreated (51.5 ± 7.4) than burned (28 ± 11.2) sites. Oneflower flatsedge (Cy perns retroflexus) , B. cilliatissima and E. lehmanniana densities were highest on nontreated sites (Table 3). Croton , C. erecta and P. cinerascens were most common on burned sites, while E. alsinoides and M. cinereum were more frequently encountered on nontreated sites (Table 4). Seasonal variations followed similar tends as density estimates. Treatment by season interactions for Croton and E. alsinoides were similar to density trends. Brachiaria cilliatissima and B. hirsuta were more commonly encountered on nontreated sites (Table 3). Grass (P = 0.4701) and forb (P = 0.1356) productivity was similar between burned (133.9 ± 46.3 g/m2; 27.6 ± 28.2, respectively) and nontreated (122.8 + 41.6; 6.4 ± 4.7, respectively) sites. Yield of grass RUTHVEN & SYNATZSKE N> © © © A A OU)OOOOO»-OOMO'-*OOO00 A A hJOOO-^O^OOOOOi— “OONJ-^J*— “OOONJNJO NJOt-*OU)O0\OO>-‘OOU>O>-*U)O^OO4^>-‘'~JO ►—ONJONJO-JOONJ-^-ONJObOJ^OS-^OO-^Ni-^-O >— h-» U) A jvj >— 00 >— N> UO^00U‘-‘N)OOWO-^'0'-ViM.(i^UUi(J'Ui00 _ ^_A ►— N> N> N> ^O^O\^N)WOOWOIv)U00*-'J'0UN)UO^O'J A *-* N> w •— .->h^0NO\N/->JN>©N>OOOO©00»-* oo >-* NO 4^ U> On OO A — H-* N> 4^- OO >— K> H* tv) A N u, ^ M ON 4*. 4^ o\4v>-.o4^>-‘W4^aio\'-wo\ooowtv)ouiuooa>oo\ _fsj _ A k- _ ui u> >-» s> ■vlKJooDiW^^AUff.-OOwONNlWWO-vltOMAOOO i — • <-/> on OO © I— Os 4± N> M — * — * oo p— M^OUlOO-fvO'COOOW ppppppppopoppoo k>L>baooook>u>o^— L’ ’ L _ - - - - ‘ ■ M 7| U OO N) OO A -* l/l '-4 — — , _ _ — . N> U) O 0'-jooosooou)oso -* Ui Iv O O |V O -4 ►■'• O OOOOOO^UOOt/\\OOOv J N) N) U) as ©►-000000 — 'OOOOOOOO^.k. •— 1 O N) N> Ui O v) Ui VO OO A A AA A AA AA A A AAA A pppppppppppppppppppppppo bbbbs)bbbbbwbbAbfjbbbbbbob 0NOOOU>ON>OOOU\OO4^OOOOOOO0sOO ON/OO0nO4^OOO'-JOO^0OOOK>OOO4^ — O .£»• no - >— © >— *— •— ■!— i— ^Ji-*^\Op— O'-*© — i-1 i- OO M i- oooooooooooooooooooo N> N> O OO © — N) N) U OUi^nJMmAW^Ui i— U>N> — ©U\© N)OiUvJ001/i0'00O5iO5' N)^^AN-40i-0 5\ W A N '-J O no O \0 O N) N> ON U> iO N) -* w o\ to poop 4^ Li 4^ © W /I N) OO 00 W \0 -4 GO — — K> X I __ NO i'Z NO 1 o NO 13 Si w C >1 P . a oo o-'o 2. & OTQ xi?i 3 O Si *\> p - rt> OO <1 *'o§ fet a p. SS JD < B sl § § S3 0> OO “T3 O i I & » g 8 i. 203 Table 4. Percent frequency (%) of dominant (frequency >5%) forbs utilized by livestock and wildlife (Everitt et al. 1999) by season on summer burned (n = 5) and nontreated (n = 5) sites on the Chaparral Wildlife Management Area, Dimmit and La Salle counties, Texas, 1999-2001. 204 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 3, 2002 Table 5. Yield (g/m2) of selected grasses and forbs season on summer burned ( n = 5) and nontreated ( n = 5) sites on the Chaparral Wildlife Management Area, Dimmit and La Salle counties, Texas, June 2000. Class/Species Burned * SD Nontreated * SD P-value Grass Aristida purpurea 14.3 12.5 13.4 9.6 0.9017 Brachiaria cilliatissima 0.5 0.7 8.4 11.0 0.1462 Bouteloua hirsuta 22.5 17.7 30 15.4 0.4949 Chloris cucullata 2.2 3.4 0.6 0.9 0.3329 Digitaria cognata 12.0 6.3 5.2 3.8 0.0716 Eragrostis lehmanniana 41.0 17.7 30.7 16.1 0.3635 Eragrostis sessilispica 5.6 5.8 4.7 6.5 0.8231 Eragrostis secundiflora 3.8 3.6 5.9 8.0 0.6085 Paspalum setaceum 11.7 5.1 7.0 8.5 0.3209 Forbs Chamaecrista fasciculata 3.9 4.5 0.4 0.7 0.1218 Croton sp. 7.5 5.8 0.2 0.2 0.0230 Commelina erecta 1.6 2.5 0 0 0.1839 Evolvulus sp. 1.9 2.0 2.2 1.3 0.7885 species was similar among treatments (Table 5). Croton yield was greatest on burned sites. Discussion The results of this study indicate that summer applied prescribed fire on south Texas rangelands can increase densities and productivity of important seed producing annuals such as Croton during the first growing season following burning. Croton , which is an important seed source for granivorous birds (Dillon 1961; Everitt et al. 1999), also responds positively to winter burning throughout the South Texas Plains (Box & White 1969; Hansmire et al. 1988; Ruthven et al. 2000). Annual sunflowers ( Helianthus sp.) are another important seed producer that also provide forage for O. virginianus and livestock (Everitt et al. 1999). Although not significantly affected by burning, prairie sunflower (. Helianthus petiolaris) was only encountered on burned sites. Chamae- crista fasciculata was one of the most common annual forbs encountered and the lack of treatment effect was similar to that resulting from winter burning (Ruthven et al. 2000; 2002). Chamaecrista fasciculata , which is important forage for C. virginianus , typically increases following winter and spring burns in the southeastern United States (Lewis & Harshbarger 1976). Cushwa et al. (1968) reported that moist heat greatly increases the germination of partridge pea. Although soil moisture conditions were considered high during the burns in this study, the lack of a response by C. fasciculata may have resulted from high temperatures associated with the fire and season of burn. Controlled RUTHVEN & SYNATZSKE 205 laboratory experiments simulating heat produced by prescribed fire have shown no increase in germination of C. fasciculata seed following exposure to heat of < 600 °F (Mitchell & Dabbert 2000). Additionally, much of C. fasciculata seed may be exposed on the soil surface during summer and consequently consumed by fires. Cool season annual forbs appeared unaffected by summer burning, which may be explained by abnormal precipitation patterns. In southern Texas, cool season annuals germinate in fall and reach peak flowering in early spring. Precipitation (7.5 cm) during this time period (October 1999 to February 2000) was 56% below normal, which may have negated any positive responses from burning. The varied response by perennials was similar to dormant-season fire (Hansmire et al. 1988; Ruthven et al. 2000). Commelina erecta and P. cinerascens slightly increased following winter burns in the western South Texas Plains (Ruthven et al. 2000), while in the transition zone between the South Texas Plains and the Gulf Coast Prairies both species were unaffected or slightly decreased following winter burning (Hansmire et al. 1988). Commelina erecta and P. cinerascens are important forages for a wide variety of wildlife. Commelina erecta is highly valued as forage for O. virginianus (cf. Everitt et al. 1999) and an important dietary component of the Texas tortoise (G op herns berlandieri ) (R. Kazmaier, pers. comm.), and the leaves and fruits of P. cinerascens are important foods of O. virginianus , javelina ( Tayassu tajacu) and wild turkey ( Meleagris gallopavo) (cf Everitt et al. 1999). Competition among herbaceous vegetation may explain the varied response of perennial forbs to burning. Declining perennials such as M. cinereum and E. alsinoides are found primarily in the interspace between shrub clumps where grass and annual forb abundance is greatest, whereas perennials such as C. erecta and P. cinerascens , which demonstrated postburn increases, are primarily found beneath P. glandulosa canopies in shrub clusters where grass and forb abundance are less than the interspace (Whittaker et al. 1979; Ruthven 2001). Increases in annual forbs in the interspace following burning may increase competition resulting in declines of M. cinereum and other perennials on burned sites. Although burning increased Croton during the first growing season postburn, this positive affect did not persist into the second growing season. This response was similar to winter burns conducted on the study site (Ruthven et al. 2002). It is unclear whether increases in forb densities in the first growing season result in an increase in availability of seeds for use by wildlife and future forb production. It does appear that additional disturbance is necessary to stimulate a significant increase 206 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 3, 2002 in germination of annual forbs. The extension of the positive response of some perennials into the second growing season on burned sites is similar to that observed on comparable study sites burned during winter. Increases in perennials may be explained by the release of nutrients into the soil (Scifres & Hamilton 1993). In part, seasonal variation of warm-season annuals can be explained by season of burn. Warm-season annual forbs would have been con¬ sumed by summer fires, explaining their absence during the fall 1999 sampling period. Seasonal variability and interactions are also a likely result of precipitation patterns. Atypical drought conditions persisted throughout late summer and early fall 2000, with 1.7 cm of precipitation being recorded between mid-June and mid-October on the study site. This lack of rainfall may have lead to poor seed production of warm- season annuals such as Croton and rough buttonweed ( Diodia teres), resulting in low numbers in spring 2001. The perennial forb E. alsinoides may have suffered mortality during this extended dry period, resulting in the decreased densities observed in spring 2001, while L. argyaea, which had greatest densities in spring 2001, appears more adapted to short-term periods of drought. In contrast to warm-season annual fords, densities of cool-season annuals were greatest in spring 2001. Fall and winter (October-February) rainfall (30.9 cm) during 2000-2001 was 183% above normal compared to below normal precipi¬ tation recorded during the same period in 1999-2000. The initial reduction of warm-season perennial grasses following summer burns was similar to that reported in other studies in the south central United States (Scifres & Duncan 1982; Mayeux & Hamilton 1988; Engle et al. 1998); however, productivity was similar among treatments midway through the first postburn growing season. Despite apparent grass mortality from fire, vigor of surviving plants appeared to be stimulated. This apparent increase in productivity may have resulted from soil moisture conditions at the time of burning. Soil moisture content in previous studies (Scifres & Duncan 1982; Mayeux & Hamilton 1988) was relatively low, while soil moisture was considered high in this study. Summer burns following significant precipitation events in northwestern Mexico were found to increase bufflegrass {Cenchrus ciliaris) productivity (Martin-R et al. 1999). Burning during the dormant and early growing season can also result in varied responses by grasses. Scifres & Duncan (1982) and Hansmire et al. (1988) reported overall increases in grass productivity following late winter and early spring burning, while Ruthven et al. (2002) reported that dormant-season burning had little affect on grass abundance. RUTHVEN & SYNATZSKE 207 Warm-season perennials such as tanglehead ( Heteropogon contortus), plains bristlegrass ( Setaria machrostachya) and multiflowered false rhodesgrass ( Chloris pluriflora) dominated the Sandy Loam and Shallow Sandy Loam range sites of south Texas under pre-European settlement conditions (Stevens & Arriaga 1985). These grasses are highly preferred by livestock and have decreased as a result of long-term overgrazing. The occurrence of natural fires in the South Texas Plains may have been less frequent than other grassland/savanna ecosystems (Wright & Bailey 1982), and little data is available on how these historically dominant warm-season perennial grasses respond to fire season and frequency. Invasion of sandy south Texas rangelands by E. lehmanniana is a growing concern. This introduced perennial is an aggressive invader that can displace native grasses and quickly become the predominant grass species (Anable et al. 1992). Although hot summer fires can kill E. lehmanniana (cf. Cable 1965), burning may increase germination of E. lehmanniana seed (Ruyle et al. 1988). Eragrostis lehmanniana was reduced following fires in this study, but productivity on burned sites equaled nontreated site midway through the first postburn growing season. Extended monitoring beyond the first postburn growing season may be necessary to determine the full effects of growing-season burning on this species. The grazing system employed on the study area appears to have little affect on the postburn response of forbs (Ruthven et al. 2000); however, it is unclear how grazing may have affected grasses following fire. High intensity low frequency grazing results in greater consumption of less-preferred forage species (Drawe 1988), which can lead to the more uniform grazing of herbaceous plants. The high intensity low frequency grazing system employed during the dormant-season on the study area appears to reduce selective grazing and minimize any effects postburn grazing pressure may have on grasses. If enhancing annual forbs is a primary goal, then conducting summer burns on a biennial schedule may be beneficial. However, grass production must be taken into consideration, as grasses are the primary fine fuels needed to conduct prescribed burns in much of southern Texas. Proper grazing management is crucial in producing adequate fuel loads for burning. The grazing strategy on the study site, which provides rest during the majority (May-October) of the growing season, appears to allow for ample fuel buildups necessary to conduct burns successfully. Although repeated summer burns on a biennial schedule can increase herbaceous plant diversity and density and increase 208 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 3, 2002 dominance of grasses in mesic environments (Lewis & Harshbarger 1976), it is unclear whether similar responses can be achieved in dryer climes. Short-term periods of drought and highly variable rainfall are typical of the South Texas Plains (Norwine & Bingham 1985). This coupled with the semiarid (annual precipitation < 54 cm) nature of western portions of the South Texas Plains may not provide adequate fuel loads on a regular basis to conduct burning on an alternating year schedule. Realistically, burning may be achieved on a three to four year cycle. This burning frequency may maintain benefits of burning on perennial forbs such as C. erecta and P. cinerascens. If prescribed burning activities are directed toward improving habitat for O. virginianus, it may be beneficial to limit the use of fire in areas dominated by highly preferred forage species such as M. cine ream, which respond negatively to fire. Forb response to summer fire was similar to that of dormant-season burning. Most land managers burn during the winter months when burning conditions are less volatile. Based on forb response, there appears to be no benefit in conducting prescribed burns during the summer months rather than the dormant season. In fact, late winter fire can increase grass productivity. Although burning during winter or early spring may be preferred because winter fires are easier to control, there is little data on the effects of prescribed fire on wildlife. Most herpetofauna hibernate during the winter months. In south Texas, the Texas horned lizard ( Phrynosotna comutum), a state threatened species, hibernates at shallow depths (Fair & Henke 1997), which may increase its susceptibility to direct mortality during dormant-season fires. Burning during summer, when herpetofauna are active, may lessen the probability of direct mortality. It is clear that climatic, edaphic and temporal factors can dramatically affect the impacts of prescribed burning. Response of forbs and grasses to fire can vary by burning date within a climatic season and among soil types (Hansmire et al. 1988) and care should be exercised when extrapolating the results of this study to other soil types in South Texas. Hiers et al. (2000) suggests that combinations of dormant- and growing-season burns may be necessary to promote species diversity. Further study into the effects of burning date within season, influence of pre- and postburn climatic variables, interactions between burning and herbivory, and long-term effects of multiple burns on vegetation, as well as wildlife, are necessary to better understand fire ecology in southern Texas. RUTHVEN & SYNATZSKE 209 Acknowledgments We thank Jim Gallagher with Texas Parks and Wildlife Department for assistance with statistical analyses. Rich Kazmaier with West Texas A&M University provided information on Texas tortoise ( Gopherus berlandieri) diet. Literature Cited Anable, M. E., M. P. McClaren & G. B. Ruyle. 1992. Spread of introduced Lehmann lovegrass Eragrostis lehmanniana Nees. in southern Arizona, USA. Biol. Conserv., 61:181-188. Archer, S., C. Scifres, C. R. Bassham & R. Maggio. 1988. Autogenic succession in a subtropical savanna: conversion of grassland to thorn woodland. Ecol. Monogr., 58:111-127. Archer, S. 1994. Woody plant encroachment into southwestern grasslands and savannas: rates, patterns and proximate causes. Pp. 13-68. in Ecological implications of livestock herbivory in the west (M. Vavra, W. A. Laycock, & R. D. Pieper, editors). Society for Range Management, 297 pp. Box, T. W., & R. S. White. 1969. Fall and winter burning of south Texas brush ranges. J. Range Manage., 22:373-376. Bozzo, J. A., S. L. Beasom & T. E. Fulbright. 1992. Vegetation responses to brush management practices in south Texas. J. Range Manage., 45:170-175. Blair, W. F. 1950. The biotic provinces of Texas. Tex. J. Sci., 2(1):93-117. Cable, D. R. 1965. Damage to mesquite, Lehmann lovegrass, and black grama by a hot June fire. J. Range Manage., 18:326-329. Chambers, J. C. & R. W. Brown. 1983. Methods for vegetation sampling and analysis on revegetated mined lands. USDA For. Serv. Gen. Tech. Rep., INT-151, 57 pp. Correll, D. S. & M. C. Johnson. 1979. Manual of the vascular plants of Texas. The University of Texas at Dallas, Richardson, xv + 1881 pp. Cushwa, C. T., R. E. Martin & R. L. Miller. 1968. The effects of fire on seed germination. J. Range Manage., 21:250-254. Dillon, O. W., Jr. 1961. Mourning dove foods in Texas during September and October. J. Wildl. Manage., 25:334-336. Drawe, D. L. 1968. Mid-summer diet of deer on the Welder Wildlife Refuge. J. Range Manage., 21:164-166. Drawe, L. D. 1988. Effects of three grazing treatments on vegetation, cattle production, and wildlife on the Welder Wildlife Foundation Refuge, 1974-1982. Welder Wildl. Foundation Contribution B-8, Sinton, Texas, 36 pp. Engle, D. M., J. F. Stritzske, T. G. Bidwell & P. L. Claypool. 1993. Late-summer fire and follow-up herbicide treatments in tallgrass prairie. J. Range Manage., 46:542-547. Engle, D. M., R. L. Mitchell & R. L. Stevens. 1998. Late growing-season fire effects in mid-successional tallgrass prairies. J. Range Manage., 51:115-121. Everitt, J. H., D. L. Drawe & R. I. Lonard. 1999. Field guide to the broad-leaved herbaceous plants of south Texas used by livestock and wildlife. Texas Tech Univ. Press, Lubbock, vii + 277 pp. Fair, W. S. & S. E. Henke. 1997. Effects of habitat manipulations on Texas horned lizards and their prey. J. Wildl. Manage., 61:1366-1370. Fulbright, T. E. & S. L. Beasom. 1987. Long-term effects of mechanical treatments on white- tailed deer browse. Wildl. Soc. Bull., 15:560-564. Gabor, T. 1997. Ecology and Interactions of sympatric collared peccaries and feral pigs. Ph.D. Dissertation, Oklahoma State Univ., Stillwater, 255 pp. Gabriel, W. J., D Arriaga & J. W. Stevens. 1994. Soil survey of LaSalle County, Texas., 210 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 3, 2002 U. S. Department of Agriculture, Washington, D.C., 183 pp. Hansmire, J. A., D. L. Drawe, D. B. Wester & C. M. Britton. 1988. Effects of winter burns on forbs and grasses of the Texas Coastal Prairie. Southwest. Nat., 33(3):333-338. Hatch, S. L., K. N. Gandhi & L. E. Brown. 1990. Checklist of the vascular plants of Texas. Tex. Agri. Exp. Stn. Misc. Pub. MP-1655, iv + 158 pp. Hiers, J. K., R. Wyatt & R. J. Mitchell. 2000. The effects of fire regime on legume reproduction in longleaf pine savannas: is season selective? Oecologia, 125:521-530. Landers, J. L. & B. S. Mueller. 1992. Bobwhite quail management: a habitat approach. Tall Timbers Res. Sta. Misc. Pub. No. 6, Tallahassee, Florida, 37 pp. Lehmann, V. W. 1969. Forgotten legions: sheep in the Rio Grande Plain of Texas. Texas Western Press, El Paso, xv + 226 pp. Lewis, C. E. & T. J. Harshbarger. 1976. Shrub and herbaceous vegetation after 20 years of prescribed burning in the South Carolina Coastal Plain. J. Range Manage. , 29:13-18. Martin-R, M., J. R. Cox, F. Ibarra-F, D. G. Alston, R. E. Banner & J. C. Malecheck. 1999. Spittlebug and bufflegrass response to summer fires in Mexico. J. Range Manage., 52:621-625. Mayeux, H. S., Jr. & W. T. Hamilton. 1988. Response of false broomweed and associated herbaceous species to fire. J. Range Manage., 41:2-6. McLendon, T. 1991 . Preliminary description of the vegetation of south Texas exclusive of coastal saline zones. Tex. J. Sci., 43(1): 13-32. Mitchell, R. & B. Dabbert. 2000. Potential fire effects on seed germination of four herbaceous species. Tex. J. Agr. Nat. Resour., 13:99-103. Norwine, J. & R. Bingham. 1985. Frequency and severity of drought in south Texas. Pp. 1-19 in Livestock and wildlife management during drought (R. Brown, editor). Caesar Kleberg Wildlife Research Institute, Kingsville, Texas, 84 pp. Ruthven, D. C., III. 2001. Herbaceous vegetation diversity and abundance beneath honey mesquite {Prosopis glandulosa) in the South Texas Plains. Tex. J. Sci., 53(2): 171-186. Ruthven, D. C., Ill, T. E. Fulbright, S. L. Beasom & E. C. Hellgren. 1993. Long-term effects of root plowing on vegetation in the eastern south Texas plains. J. Range Manage., 46:351-354. Ruthven, D. C., Ill, J. F. Gallagher & D. R. Synatzske. 2000. Effect of fire and grazing on forbs in the western south Texas plains. Southwest. Nat., 45(2):89-94. Ruthven, D. C., Ill, J. F. Gallagher & D. R. Synatzske. 2002. Response of herbaceous vegetation to winter burns in the western south Texas plains: an observation. Tex. J. Agr. Nat. Resour., 15: In press. Ruyle, G. B., B. A. Roundy & J. R. Cox. 1988. Effects of burning on germinability of Lehmann lovegrass. J. Range Manage., 41:404-406. Scifres, C. J. & K. W. Duncan. 1982. Brownseed paspalum response to season of burning. J. Range Manage., 35:251-253. Scifres, C. J. & W. T. Hamilton. 1993. Prescribed burning for brushland management: the south Texas example. Texas A&M University Press, College Station, xiii + 246 pp. Stevens, J. W. & D. Arriaga. 1985. Soil survey of Dimmit and Zavala Counties, Texas, U. S. Department of Agriculture, Washington, D.C., 161 pp. Taylor, R. B., J. Rutledge & J. G. Herrera. A field guide to common south Texas shrubs. Texas Parks and Wildlife Press, Austin, xiii -I- 106 pp. Whittaker, R. H., L. E. Gilbert & J. H. Connell. 1979. Analysis of two-phase pattern in a mesquite grassland, Texas. J. Ecol. 67:935-952. Wright, H. A. & A. W. Bailey. 1982. Fire Ecology. John Wiley and Sons, New York, xxi + 501 pp. DCR at: cwma@vsta.com TEXAS J. SCI. 54(3):21 1-226 AUGUST, 2002 EFFECTS OF PRESCRIBED BURNING ON VEGETATION AND FUEL LOADING IN THREE EAST TEXAS STATE PARKS Sandra Rideout and Brian P. Oswald USD A Forest Service , Forestry Sciences Laboratory 320 Green Street, Athens, Georgia 30602-2044 and Silviculture and Range Management, Stephen F. Austin State University Box 6109, SFA Station, Nacogdoches, Texas 75961-6109 Abstract.— This study was conducted to evaluate the initial effectiveness of prescribed burning in the ecological restoration of forests within selected parks in east Texas. Twenty -four permanent plots were installed to monitor fuel loads, overstory, sapling, seedling, shrub and herbaceous layers within burn and control units of Mission Tejas, Tyler and Village Creek state parks. Measurements were taken during the summers of 1999 and 2000. Prescribed burning was conducted between these sampling periods in early spring 2000. Results indicated that the current applications of prescribed burning do not significantly influence vegetation or fuels. Sustained drought, prior management practices and imposed local burn bans reduced the window within which prescribed burns could be applied, and limited the effectiveness of the burns. Historically, fire has played an important role in most terrestrial ecosystems. Fire has an influence in such ecosystem components as recycling of nutrients, regulating plant succession and wildlife habitat, maintaining biological diversity, reducing biomass, and controlling insect and disease populations (Mutch 1994). When conducted properly, prescribed fire undoubtedly alters the composition and structure of the understory vegetation within forests. Several subclimax communities and endangered species of Texas are dependent on fire. For example, fire is an essential element in the restoration and management of longleaf pine ( Pinus palustris Mill.) stands and pitcher plant ( Sarracenia alata Wood) wetland ecosystems. These and other communities benefit from an active prescribed burning program (Reeves & Corbin 1985). Prescribed burning is currently used as a management tool in several Texas state parks for the purposes of reducing forest fuels, improving wildlife habitat, altering the composition and structure of the understory vegetation and enhancing park appearances. This study was conducted to evaluate the initial effectiveness of prescribed burning in the ecological restoration of forests and consisted of monitoring pre- and 212 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 3, 2002 post- burn vegetative characteristics and fuel loads at three Texas state parks. At Mission Tejas State Historical Park, Tyler State Park and Village Creek State Park, 24 plots, eight in each park, were monitored in the summers of 1999 and 2000 to determine short-term ecological effects of pre-scribed burning on vegetation and fuel loads. Methodology The three parks surveyed in this study were all part of the Piney- woods Region of Texas Parks and Wildlife Department’s Parks and Historic Sites. Mission Tejas and Tyler State Parks had similar ecological characteristics. Typical overstory species within the burn units of these parks included shortleaf pine ( Pinus echinata Mill.), loblolly pine ( Pinus taeda L.), sweetgum ( Liquidambar styraciflua L.), water oak ( Quercus nigra L.), white oak ( Q . alba L.), mockernut hickory ( Cary a tomentosa (Poir.) Nutt.), white ash ( Fraxinus americana L.) and American holly ( Ilex opaca Ait.). Common understory species included yaupon (Ilex vomitoria Ait.), flowering dogwood (Comus florida L.), American beautyberry (Callicarpa americana L.), longleaf uniola ( Chasmanthium laxum var. sessiliflorum (L.) Yates), panicums (Panicum sp.) and various sedges (Texas Parks and Wildlife 2000a; Texas Parks and Wildlife 2000b). Average low temperatures in January range from 0 to 2°C, while July averages highs of 34 to 36°C. The first and last freezes typically occur around mid to late November and mid March to early April, respective¬ ly. Average rainfall exceeds 100 cm per year (Texas Parks and Wildlife 2000a; Texas Parks and Wildlife 2000b). Steep slopes abound in these parks, with elevation changes of 100 m within both parks (Texas Parks and Wildlife 2000a; Texas Parks and Wildlife 2000b; Robinson & Blair 1997). The historic fire return interval where these parks are located was 4 to 6 years. It is presently greater than 20 years (Jurney 2000) due to suppression, fragmentation and urbanization of the surrounding areas. Heavy fuel loads persist throughout the park due to decades of sporadic use of fire. Unlike the others, Village Creek State Park included cypress swamps, bottomland wetlands and blackwater sloughs in the flood plain of the Neches River. The burn unit was once a longleaf/little bluestem (Schizachyrium scoparium (Michx.) Nash.) stand. Due to fire exclusion it was being overtaken by broadleaf trees, such as water tupelo ( Nyssa aquatica L.), river birch ( Betula nigra L.), water oak and redbay RIDEOUT & OSWALD 213 ( Persea borbonia (L.) Spreng.), in addition to the invasive Chinese tallowtree ( Sapium sebiferum (L.) Roxb.). Common understory vegetative species included yaupon, flowering dogwood, American beautyberry, poison ivy ( Toxicodendron radicans (L.) Kuntze), little bluestem, panicums and various sedges. The park’s mean elevation was 7 m. January’s average low temperature was 3°C, while July’s average high was 34 °C (Texas Parks and Wildlife 2000c). Historic fire return interval in the area was 1 to 3 years. Now it is greater than 20 years (Jurney 2000). Methods for establishing plots, and sampling vegetation and fuel loads were as defined in the National Park Service Western Region Fire Monitoring Handbook (Western Region Prescribed and Natural Fire Monitoring Task Force 1992). Plot size and sampling locations varied for each monitoring variable. Consistent sample areas were used between plots for each variable. The entire 20 by 50 m rectangular plot was used for sampling overstory (Figure 1). Overstory trees were defined as all trees, living or dead, with dbh > 15 cm. Dbh (diameter at breast height) was defined as diameter outside bark at 1.4 m. Saplings were defined as standing living or dead trees with dbh >2.5 cm and < 15 cm. They were sampled only within Quarter 1. Seed¬ lings were defined as those living trees with dbh < 2.5 cm. Seedlings were monitored only in the 5 by 10 m medial section of Quarter 1. The point line- intercept method was used for sampling shrub and herbaceous layers. The point line- intercept transect ran along the Q4-Q1 50 m line delineating that outside long axis of the plot. Height of the tallest living or dead individual by species, and species from tallest to shortest intercepting the transect were recorded. To obtain shrub density, the Q4-Q1 transect was widened to a belt 0.5 m wide. A stem count of shrub species within the belt was recorded. To measure density of herbaceous plants, a 1 m2 frame was placed on the plot side of both outer 50 m transects every 10 meters. The total area sampled in each plot using this method was 10 m2. Herbaceous species and number of stems were recorded. Four transects extending 15.2 m in random directions from the centerline at the 10, 20, 30 and 40 m marks in each plot were used to measure fuel loads (Brown et al. 1982). One-, ten-, hundred- and 214 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 3, 2002 25m 25m 50m Overstory Sapling Seedling Figure 1. Sampling areas and transects for vegetation and fuel load monitoring (Western Region Prescribed and Natural Fire Monitoring Task Force 1992). thousand-hour fuels were sampled along these transects. Depth of Oj and Oe (litter) horizons combined was also measured, as well as, depth of Oa (duff) horizon. Samples of Oj and Oe horizons combined were collected and dried to determine litter weight. All vegetative and fuel load monitoring techniques were repeated during the same time of the year 2000. Texas Parks and Wildlife Department (TPWD) personnel produced the burn plans. Prescribed burns were conducted during late February to early March 2000 when weather and fuel moisture conditions allowed. To estimate the intensity of each burn, four tiles with heat-sensitive paint were attached to the center t-post of each plot. One tile each was placed 15 cm below ground, at ground level, 30 cm and 61 cm above ground. Tiles were removed immediately after the burn. Analyses of the tiles allowed an estimate within 38 °C of the fire temperature at plot origin. RIDEOUT & OSWALD 215 County burn bans prohibited burning in the parks until they were temporarily lifted following rain episodes. Because of the necessity to wait until a rain event, fuels were wet and resulting burns were weak and spotty. Firelines were monitored for two hours after each burn was completed. Park staff was responsible for monitoring the burn unit after that time. According to written burn plans (Sparks 1999a; Sparks 1999b; Robinson & Blair 1997), the primary objectives of the initial burns were to reintroduce the natural role of fire into the ecosystems and to reduce fuel loads. Other objectives mentioned included reducing risk of wild¬ fire, increasing species richness and diversity, increasing wildlife habitat for numerous species, encouraging longleaf pine seedlings at Village Creek State Park and beginning the first stage in restoration. Cool season burns were recommended every two years to reduce fuels sufficiently for growing season burns. Following three cool season burn cycles, burns would be conducted once every three years during the early to mid-growing season to increase mortality in under story hard¬ wood saplings. Fuel loading (Mg ha'1) was calculated using Excel software. ANOVA and paired /-tests were performed to test for significant differences in pre- and post-burn fuel loads and vegetation in SPSS Base 10.0 (SPSS Inc. 1999). Exploratory analysis was conducted on data in PC-ORD (McCune & Mefford 1999) using twinspan, Detrended Correspondence Analysis (DC A) and graphing the DC A. DC A was designed for ecological data sets. It is based on samples and species, and ordinates both simultaneously (McCune & Mefford 1999). Paired /-tests were conducted in Excel on overstory and sapling vegetation to determine differences in standing dead vegetation before and after the burns. Morisita’s index of similarities was conducted on seedling, shrub and herbaceous communities to determine differences in composition before and after the burns (Morisita 1959). Morisita’s index was formulated as follows: 216 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 3, 2002 Where: X . = Number of species i in community A Y j = Number of species i in community B na = EXi nb = EYi s _ Ml i)] A Na(Na-1) s _ E IX i(Y.i - 1)] B Nb(Nb-1) Results and Discussion Fuel loading results for all parks combined in 1999 (before burning) and 2000 (after burning), indicated a statistically significant reduction in one-hour fuels in burn plots in 2000; however, the actual difference was only 0.05 Mg ha'1. This is not ecologically significant. There was also a statistically significant reduction in ten-hour fuels in the control plots, while there was no change in the burn plots (Table 1). The only statistically significant difference in hundred- or thousand- hour fuels was an increase in thousand-hour fuels in control plots (Table 1). Larger fuels may have increased due to drought-stressed trees dying and falling. For all parks combined, Oj and Oe horizons’ combined weight decreased significantly ( t = 5.182, P < 0.001) in the burn plots while it did not in the control plots (Table 2). The actual decrease in the burn plots was 0.98 Mg ha'1. There was also a statistically significant decrease in depth of O; and Oe combined in the burn plots ( t = 2.074, P < 0.05), while there was a significant increase in the control plots ( t = (6.641, P < 0.001)(Table 2). Tiles recovered from the burns indicated weak burns at all parks, with Mission Tejas generally burning hotter than Tyler and Village Creek. Tiles showed no effect from the heat of the burns at the 61 cm (2 ft) level in any plot. One tile at Mission Tejas indicated 93 °C at the 30 cm RIDEOUT & OSWALD 217 Table 1. Mean fuel loads and paired Mest results for fuels in 1999 (pre-burn) and 2000 (post-burn) in Mission Tejas, Tyler and Village Creek State Parks combined. Plot type Measurement One- hour Ten- hour Hundred- hour Thousand- hour Total Burn 1999 fuel load ( Mg ha'1) 0.29 1.78 1.81 1.63 5.53 (n = 60, df= 59) 2000 fuel load (Mg ha1) 0.24 1.58 2.49 2.42 6.68 Mean difference 0.05 0.19 -0.68 -0.79 -1.15 SD 0.15 2.17 3.73 4.88 5.52 t 2.453 0.687 -1.406 -1.254 -1.608 Significance 0.017 0.495 0.165 0.215 0.113 Control 1999 fuel load (Mg ha'1) 0.31 2.25 1.74 2.55 6.84 (n — 36, df= 35) 2000 fuel load (Mg ha'1) 0.24 1.01 2.04 6.20 9.50 Mean difference 0.07 1.23 -0.30 -3.64 -2.50 SD 0.28 1.60 3.30 9.58 10.04 t 1.518 4.610 -0.553 -2.282 -1.584 Significance 0.138 <0.001 0.584 0.029 0.122 Table 2. Mean measurements in 1999 and 2000 and paired Mest results for O : and O e combined and O a horizons in Mission Tejas, Tyler and Village Creek State Parks combined. Plot Measurement O | and O e O j and O e oa depth type weight depth (cm) (Mg ha'1) (cm) Burn 1999 2.990 1.348 1.431 (n* = 60, 2000 2.015 1.203 1.353 df= 59) Mean difference 0.976 0.145 0.077 SD 1.409 0.542 0.550 t 5.182 2.074 1.084 Significance <0.001 0.042 0.283 Control 1999 3.716 1.492 1.571 {n = 36, 2000 3.480 2.196 1.600 df = 35) Mean difference 0.236 -0.703 -0.029 SD 1.664 0.636 0.742 t 0.850 -6.641 -0.234 Significance 0.401 <0.001 0.817 * n = 56 for O and O e weight in the burn plots, df = 55 for O ; and O e weight in the burn plots. (1 ft) level, while the others recorded no effect. At ground level, tiles indicated a range of intensities from 0°C to 538°C, with Mission Tejas averaging 293 °C, Tyler averaging 149°C, and Village Creek averaging 45°C. At the subground level Mission Tejas averaged 197°C and Tyler 218 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 3, 2002 averaged 13 °C, while tiles at Village Creek recorded no effect. This level of intensity could leave quite a bit of the O horizon and downed woody fuels unburned. After the fires, most surface fuels appeared charred but unconsumed. It appears the burns did not fully reach the objective of reducing fuel loads. The only ecologically important effects were the decreases in weight and depth of the Oj and Oe horizons in the burn plots. The loss in weight from 1999 to 2000 was 0.98 Mg ha"1, and the difference in depth between the burn and control plots in 2000 was 0.85 cm. These differences were possibly enough to affect the viability of seedlings or herbaceous plants. Vegetation Mission Tejas State Historical Park.— With Axis 1 of the DCA graph representing decreasing time since prior disturbance, one plot was separated to the far right of the other plots in most vegetation classes because it had been burned in the past. There were no records of how long ago the burn occurred. The authors estimated it to be between five and ten years. The plot was very thick with loblolly saplings ranging between one and three inches in diameter. In both 1999 and 2000, the overstory of Mission Tejas plots was dominated by shortleaf pine followed by sweetgum and loblolly pine. There was not a statistically significant change in number of dead standing overstory or sapling trees from 1999 to 2000. Saplings were dominated by shortleaf and loblolly pines, followed by white oak. Morisita’s similarity index showed relatively high similarity in composition of seedlings, 50 m shrub and herbaceous transects, shrub belts and herbaceous frames between burn and control plots in 1999 and 2000 (Table 3). They indicated little to no overall effect in these populations from the prescribed burn. Authors believe results would have indicated greater changes in composition had the burns been more severe. In the seedlings class, loblolly pine, white oak and Southern red oak (Quercus falcata Michx.) were common. Sassafras {Sassafras albidum (Nutt.) Nees) was absent from the burn plots in 1999, while it was present to either a moderate or heavy degree in 2000. RIDEOUT & OSWALD 219 Table 3. Morisita’s similarity index results for plot comparisons at Mission Tejas, Tyler and Village Creek State Parks pre- (1999) and post-burn (2000). Park Plots compared Seedlings 50 m shrub and herbaceous transects Shrub belts Herbaceous frames Mission Tejas Pre-burn: burn vs. control 0.93 0.61 0.76 0.69 Post-burn: burn vs. control 0.89 0.94 0.84 0.85 Burn plots: pre- vs. post-burn 1.00 0.95 0.88 0.99 Controls: pre- vs. post-burn 0.97 0.88 0.88 1.20 Tyler Pre-burn: burn vs. control 1.02 0.76 0.94 0.85 Post-burn: burn vs. control 0.92 0.99 0.99 0.95 Burn plots: pre- vs. post-burn 0.99 0.85 0.96 0.92 Controls: pre- vs. post-burn 1.02 0.98 0.90 0.96 Village Creek Pre-burn: burn vs. control 1.00 1.01 0.86 0.00 Post-burn: burn vs. control 1.00 0.00 0.60 0.00 Burn plots: pre- vs. post-burn 1.00 0.80 1.02 0.43 Controls: pre- vs. post-burn 1.01 0.00 0.41 0.00 For the 50 m shrub and herbaceous transect, litter was more common¬ ly intersected than all plant species combined. In the previously burned plot, the transect was dominated by a heavy ground cover of poison ivy, with little room for anything else. Smilax ( Smilax sp.), Virginia creeper ( Parthenocissus quinquefolia (L.) Planch.), poison ivy, muscadine grape (Vitis rotundifolia Michx.) and partridge-berry ( Mitchella repens L.) were commonly intersected in the other plots. The 0.5 m wide shrub belts in all plots at Mission Tejas were dominated by poison ivy, smilax and Virginia creeper, with moderate amounts of muscadine grape and American beautyberry. In the herba¬ ceous classification, the only obvious change from 1999 to 2000 was the heavy presence of goldenrod ( Solidago sp.) in two of the burn plots in 2000. Goldenrod is a common invader species after disturbance, and was not recorded at all in 1999. This burn was part of the fuel reduction phase described in the burn 220 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 3, 2002 plan (Robinson & Blair 1997). Killing or weakening understory shrubs and pine saplings was one goal of the fuel reduction phase. Results indicated no significant changes in overstory, sapling, seedling, shrub or herbaceous populations. Tyler State Park. — The overstory of plots at Tyler State Park was characterized by shortleaf pine and post oak ( Quercus stellata Wangenh.). There were no significant changes in dead standing over story trees from 1999 to 2000. When graphed in DC A, two plots were commonly placed on the right of the rest of the group. Axis 1 represented soil moisture, with decreasing soil moisture to the right of the graph. These two plots were higher in elevation and would have lower soil moisture than the others. T- tests indicated a significant increase in percent of dead saplings in 2000 in the burn plots ( t = 3.004, P = 0.003). In 1999, there were 7.9 percent dead saplings while there were 18.5 percent in 2000. The control plots indicated the opposite trend, although it was not significant statistically. Thus the increase in the burn plots was evidently due to the burn. Saplings were already suffering drought stress and the additional stress of the burn exterminated weaker individuals. Further /-tests indicated no significant differences in dbh or height class of saplings from 1999 to 2000, indicating that combined stresses affected saplings of all diameters and heights evenly. Morisita’s similarity index illustrated very high similarity between seedlings, 50 m shrub and herbaceous transects, shrub belts and herbaceous frames, from 1999 to 2000, even between burn and control plots (Table 3). In the seedlings class, sweetgum and sassafras were most common, followed by Southern red oak, winged elm ( Ultnus alata Michx.), red maple, flowering dogwood and American elm. Litter was most often recorded in the 50 m shrub and herbaceous transects. In 2000, twinspan separated plots based on the presence of bare ground. No bare ground was recorded in 1999. The presence of it in 2000 could have been a result of the prescribed burn removing the O horizon. There were some changes in shrub belt data from 1999 to 2000 in Tyler State Park. Muscadine grape, poison ivy and smilax were common. American beautyberry was absent in 1999, while there was a heavy presence of it in one plot in 2000 that had burned very hot, as RIDEOUT & OSWALD 221 evidenced by char height after the burn. Virginia creeper, which was heavily present in that plot in 1999, was absent in 2000. Longleaf uniola was common in the herbaceous frames. The 10.6 percent increase in dead saplings appears to be the only significant difference in vegetation. The burn plan (Sparks 1999a) called for increasing herbaceous species, reducing brush species and enhancing species diversity and richness. None of these objectives were reached. The burn was not hot enough to accomplish these goals. Village Creek State Park.— The overstory of Village Creek was characterized by longleaf pine, southern red oak, and sweetgum. Plots closest to the creek were separated from the others in twinspan because they contained river birch, commonly found in wet soils and stream- banks, and Southern magnolia ( Magnolia grandiflora L.), also common in moist valleys (Little 1980). They also contained lesser amounts of Southern red oak than did other plots, which is more commonly found in dry, sandy loams (Little 1980). When graphed, DC A Axis 1 represented increasing soil moisture in both years in most vegetation classes. T- tests indicated no significant changes in standing dead overstory trees. In saplings, yaupon and redbay were dominant. T- tests indicated a significant increase in the number of dead saplings in the burn plots from 1999 to 2000, 12.6 to 19.6 percent, respectively (t = 2.286, P = 0.023). There was only a slight increase in the control plots, from 12.8 to 13.9 percent. This illustrated a cumulative effect within the burn plots of the drought and the burn combined. There were no significant differences in dbh and height class between 1999 and 2000, illustrating that combined impacts of fire and drought affected all sizes evenly. Chinese tallowtree was becoming increasingly common in the sapling and seedling stages at Village Creek. It is a native species of China, which has been widely planted as an ornamental in the U.S., because of its vivid fall colors. Seedlings less than one foot tall were omnipresent in areas that were typically wet, but dry due to drought. Chinese tallowtree is hardy, common in sandy soils along streams and grows quickly into thickets (Little 1980). It has the potential to overtake natural vegetation in many areas of the park if left unmanaged. Morisita’s similarity index reflected nearly exact similarities in 222 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 3, 2002 seedling composition between all control and burn plots in both years (Table 3). The burn appeared to have no effect on composition of seedlings. This was not surprising considering the wet condition of the fuels during the burn. On the shrub and herbaceous transects, litter dominated intercepts on all plots. There were more species of vegetation, and vegetation occurred more often in 1999 than 2000. Although a burn could cause a reduction in shrub species, even herbaceous species, such as little bluestem and a carex sedge {Car ex joorii Bailey) were also reduced. This is more indicative of drought effects than those of prescribed burning. Morisita’s similarity index indicated a high degree of similarity between burn and control plots in 1999 (Table 3). However, in 2000, every hit along transects within control plots contacted no vegetation, only litter. This resulted in 0.00 similarity between burns and controls in 2000, and controls in 1999 and 2000. The lack of brush and herbaceous vegetation in the control plots was due to the sustained drought. Village Creek is the northern boundary of the park. The creek often floods in the winter and spring and cypress swamps are present near both the control and the burn units. Because of the drought, the yearly flooding had not occurred in 1999 or 2000; the swamps were dry, and vegetation severely affected. There were also decreases in the total number of shrub belt species and the numbers recorded within species from 1999 to 2000. The drought appeared to play an important factor from the first year to the next. Some species increased in certain plots while decreasing in other plots, with other species exhibiting opposite responses in those same plots. This is indicative of too few resources. The species with the firmer hold on an area won out. Morisita’s index also indicated a cumulative effect of the drought and the burn in Village Creek’s shrub belt composition (Table 3). Oddly, the highest rating (1.02) was received by the similarity in the burn plots between 1999 and 2000, indicating no effect on composition by the burn. The effect of prolonged drought was also evident in the herbaceous frames. In both years, the majority of herbaceous frames were empty RIDEOUT & OSWALD 223 in all plots. Morisita’s similarity index resulted in all comparisons receiving either 0.00 or a low rating (Table 3). This was due to the total lack of herbaceous vegetation in many of the frames in 2000. At Village Creek the only significant effect of the burn on vegetation was in the percent of dead saplings. The increase, seven percent, in the burn plots was six percent greater than in the control plots. The objectives of encouraging longleaf seedlings, herbaceous species, and increasing species richness and diversity were not met. Conclusions Compared to forests with long-interval, high-severity fire regimes, characterized by stand replacing fires, forests with low- to moderate-severity regimes, characterized by low- intensity surface fires may experience greater adverse effects from high intensity wildfires because they are not adapted to them. Generally, these forests adapted to low-intensity surface fires are more adversely affected by fire suppression and other human influences following European settlement. Active fire seasons occur at more frequent intervals than in long-interval types, due to longer fire seasons, higher average temperatures, and exposure to more potential ignitions during a given fire season. They have missed more fire cycles than longer interval fire regimes, and are generally in greater need of wildfire hazard reduction and restoration of ecological integrity. Wildfires in these areas not only cause more detrimental ecological effects, but they pose great risks to firefighters and property. It is anticipated with most prescribed burning programs, that the resulting post-fire landscape will have significantly reduced fuel loads and reduced risks of detrimental wildfires. If the post-fire landscapes are also attractive to those who influence policy, positive social benefits can be anticipated as well. The primary goal of each of these burns was to reintroduce or establish prescribed burning in these parks to further this mission. That objective was met. Park staffs were introduced to the duties, dangers and special considerations necessary with conducting prescribed burns. Each time they are performed by park staff, burns should become less stressful and more efficient. This short-term project has determined that future burns must be more 224 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 3, 2002 intense to meet the fuel loads and vegetation goals outlined in the burn plans. This will require a great deal of cooperation and preparedness from park staff. The window of opportunity to conduct a burn with the desired outcomes may be quite small in any given year. Fuel moisture, wind direction and speed, ambient temperature and capable staff availability must all be ideal to conduct a burn. Once the natural resources coordinator (NRC) has identified an area to be burned it is the responsibility of the park staff to prepare and maintain it in a ready condition. Initially, dormant season burns should be conducted every two years to reduce fuel loads sufficiently to initiate early to late spring burns. This will require at least two more cool season burns of greater intensity than the burns presently studied. Spring burns occurring every three years will establish a vegetation restoration phase. After a diverse herbaceous layer and open understory have been established, a maintenance phase of burning every five to eight years, depending on desired vegetation, can begin (DellaSala & Frost 2001; Manley et al. 2001). In years with inadequate prescribed fire windows due to extreme drought or flooding, prescribed burning should not be undertaken. It is too expensive and inefficient to extract employees from their normal duties, and use expensive tools, trucks and ATVs to accomplish so little ecologically. However, TPWD personnel must be willing to take risks based on the best available knowledge. Increasingly, scientific informa¬ tion points to the necessity of fire in maintaining sustainable, healthy forests in the Southeast. Being too cautious could be just as detrimental to the forest as an escaped prescribed fire. The risks of damage from wildfire, disease, insects and overcrowding are increased when pre¬ scribed fire is put off another year in hopes of better burning conditions. Fire exclusion will ultimately result in a shift from a nonlethal under¬ story fire regime to a stand-replacement regime accompanied by changes in composition and diversity. In Texas, county judges are responsible for issuing burn bans, even those with little ecological experience on which to rely. Ideally, a relationship should be fostered between the NRC and county judges issuing the bans. Judges are accustomed to making decisions based on facts and the good of the whole, rather than emotion. They should be capable of understanding the importance of fire on the landscape and the RIDEOUT & OSWALD 225 precautions taken to keep prescribed burns contained. These parks, particularly Village Creek, would have burned naturally during very dry periods. To be forced to adhere to burn bans during these times greatly reduces the restorative powers of prescribed burning. The judges have the authority to allow TPWD to burn for ecological reasons during a burn ban. In this instance, had TPWD not been bound by the burn bans, burns could have been conducted when fuels were more dry. The failure to reach the objective of reducing fuels in the parks was a direct result of waiting until after a rain event occurred to burn. Long-term interdisciplinary research projects are necessary to quantify the ecological effects, and economic and social trade-offs of prescribed burning. Only through long-term research may it be determined which natural fire functions can be emulated with prescribed burning, which are irreplaceable, and the implications for management. Acknowledgments The authors wish to thank Jeff Sparks of Texas Parks and Wildlife Department; student workers Stephanie Price, Kelly Scott and Terry Hanzak for data collection, and Drs. Michael H. Legg, Kenneth W. Farrish and Ray L. Darville from Stephen F. Austin State University. The authors also appreciate the constructive comments of two anonymous reviewers. Literature Cited Brown, J. K., R. D. Oberheau & C. M. Johnston. 1982. Handbook for Inventorying Surface Fuels and Biomass in the Interior West. USDA For. Serv. Gen. Tech. Rep. INT-129. Intm. For. and Range Exp. Stn., Ogden, Utah, 18 pp. DellaSala, D. A. & E. Frost. 2001. An Ecologically Based Strategy for Fire and Fuels Management in National Forest Roadless Areas. Fire Mgmt. Today, 61(2): 12-21. Jurney, D. 2000. Fire History of the West Gulf Coastal Plain. Presented at the Third Long leaf Alliance Regional Conference, October 16, 2000, Alexandria, Louisiana. Little, E. 1980. The Audubon Society Field Guide to North American Trees: Eastern Region. New York: Alfred A. Knopf, 639 pp. Manley, J., M. Keifer, N. Stephenson & W. Kaage. 2001. Restoring Fire to Wilderness: Sequoia and Kings Canyon National Parks. Fire Mgmt. Today, 61(2):24-28. McCune, B. & M. J. Mefford. 1999. PC-ORD. Multivariate Analysis of Ecological Data. Version 4. Gleneden Beach, Oregon: MJM Software Design. Morisita, M. 1959. Measuring of Interspecific Association and Similarity between Communities. Memoirs of the Faculty of Science Kyushu University Series E3:65-80. Mutch, R. W. 1994. Fighting Fire with Prescribed Fire: A Return to Ecosystem Health. J. of Forestry, 92(1 1):31-33. 226 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 3, 2002 Reeves, H. C. & D. Corbin. 1985. Fire and the Big Thicket. Texas Nat. Hist., l(2):23-26. Robinson, C. & K. Blair. 1997. Fire Management Plan: Mission Tejas State Historical Park. Texas Parks and Wildlife Department, 14 pp. SPSS Inc. 1999. SPSS Base 10.0. Chicago, Illinois: SPSS Inc. Sparks, J. C. 1999a. Fire Management Plan: Tyler State Park. Texas Parks and Wildlife Department, 22 pp. Sparks, J. C. 1999b. Prescribed Fire Burn Plan: Village Creek State Park. Texas Parks and Wildlife Department, 13 pp. Texas Parks and Wildlife. 2000a. Parks and Historic Sites: Mission Tejas State Historical Park. Last updated November 13, 2000. http : //www . tpwd . state . tx . us/park/ mission/mission . htm (12/1 8/00) . Texas Parks and Wildlife. 2000b. Parks and Historic Sites: Tyler State Park. Last updated November 14, 2000. http : //www . tpwd . state . tx . us/park/ty ler/ty ler . htm (12/1 8/00) . Texas Parks and Wildlife. 2000c. Parks and Historic Sites: Village Creek State Park. Last updated November 14, 2000. http : //www . tpwd . state. tx . us/ park/ village/ village. htm (12/1 8/00) . Western Region Prescribed and Natural Fire Monitoring Task Force. 1992. National Park Service Western Region Fire Monitoring Handbook. USDOI National Park Service, San Francisco, California, 261 pp. SR at: srideout@fs.fed.us TEXAS J. SCI. 54(3):227-240 AUGUST, 2002 THE VASCULAR FLORA OF WINDHAM PRAIRIE, POLK COUNTY, EAST TEXAS Larry E. Brown, Kate Hillhouse, Barbara R. MacRoberts* and Michael H. MacRoberts* Houston Community College, 1300 Holman, Houston, Texas 77004, RR 15, Box 1425, Livingston, Texas 77351 and * Herbarium , Museum of Life Sciences, Louisiana State University Shreveport, Louisiana 71115 Abstract.— -The flora and edaphic conditions are described from Windham Prairie (an isolated calcareous prairie) in Polk County of east Texas. Two hundred and forty-two species of vascular plants representing 182 genera are reported. The soils of this area are neutral to alkaline and very high in calcium. North American prairies are among the best studied and most endan¬ gered plant associations in the world (Diamond et al. 1987; Kucera 1992; Noss et al. 1995; Sims & Risser 2000). In Texas, studies have concentrated on remnants of the once vast prairies of the central and coastal regions of the state (Smeins & Diamond 1983; 1986; Smeins et al. 1992; Diamond & Smeins 1988; 1993; Diggs et al. 1999). However, there is a type of prairie in the east Texas piney woods region that has received virtually no ecological or floristic attention. These are the small, scattered or "isolated" calcareous prairies which are often only a few acres in size. Detailed floristic descriptions of isolated prairies have been made in Arkansas, Louisiana and states eastward (Foti 1989; MacRoberts & MacRoberts 1996; 1997; MacRoberts et al. in press; Moran et al. 1997; Leidolf & McDaniel 1998;), but not in Texas (Jordan 1973; 1977; Diamond et al. 1987; Carr 1993). The purpose of this study was to identify the vascular flora and determine the edaphic conditions of this east Texas isolated prairie. Study Site and Methods Windham Prairie is located at the intersection of Long Wolf and Windham roads about 8 km south of Livingston in Polk County, Texas. The prairie area measures between 2 to 3 ha within a property of approximately 22 ha. owned by Peebles Family Ltd Partnership. The prairie is a grassy opening with a few scattered shrubs and trees surrounded by typical east Texas pine forest. The topography is gently 228 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 3, 2002 rolling upland located on the Fleming Formation with exposed sandstone outcrops on lower slopes of the prairie. The elevation is 50 to 60 m above sea level. The soil is Wiergate-Burkeville calcareous clay that is gravelly, thin and well drained with a high shrink-swell potential and slow permeability (McEwen et al. 1987). Annual precipitation is about 125 cm and is fairly evenly distributed through the year. The climate is humid with a mean annual temperature of about 20 °C. The summers are long and hot with temperatures rising to 38°C; this, combined with short droughts, translates into very hot and dry conditions especially in open areas. Under drought conditions the soils dry, forming wide cracks. When wet, the soil is very sticky. Winters are short and mild with temperatures occasionally falling to freezing but there are few days of frost. Windham Prairie has been disturbed. Parts are currently grazed but the main portion has not been grazed for at least a decade. There is no record of it having been burned or plowed. Bothiochloa bladhii is persisting in one section where it was planted for erosion control. There is severe erosion, notably in the vicinity of a now abandoned oil well, with the black topsoil layer totally absent in places. There is a stock pond dug into a small portion of the upper prairie slope, which accounts for the presence of aquatic species on the checklist. Surveys were made monthly between March and December 2000, and again in February and April 2001. Voucher specimens were deposited in the Spring Branch Science Center Herbarium (SBSC). Soil samples from the upper 15 cm were collected from widely scattered sites within the prairie and analyzed by A & L Laboratories, Memphis, Tennessee. Results This study reports the presence of 242 species of vascular plants, representing 182 genera and 64 families, from Windham Prairie: FAMILY ACANTHACEAE Rue Ilia humilis Nutt. FAMILY ACERACEAE Acer saccharum Marsh, var. floridanum (Chapm.) Small & Heller FAMILY AGAVACEAE Manfreda virginica (L.) Salisb. ex Rose FAMILY ANACARDIACEAE Rhus copallinum L. Toxicodendron radicans (L.) Kuntze BROWN ET AL. 229 FAMILY APIACEAE Chaerophyllum tainturieri Hook. Cyclospermum leptophyllum (Pers.) Sprague ex Britt. & Wilson Daucus pusillus Michx. FAMILY AQUIFOLIACE Ilex decidua Walt. var. decidua Ilex opaca Aiton Ilex vomitoria Aiton FAMILY ARECACEAE Sabal minor (Jacq.) Pers. FAMILY ASCLEPIADACEAE Asclepias viridiflora Raf. Matelea gonocarpa (Walt.) Shinners FAMILY ASTERACEAE Ambrosia artemisiifolia L. Ambrosia trifida L. Amoglossum plantagineum Raf. SYN = Cacalia plantaginea (Raf.) Shinners Symphyotrichum drummondii (J. Lindley) Nesom var. texanum (Burgess) Nesom SYN — Aster drummondii Lindl. var. texanus (Burgess) A. Jones Symphyotrichum dumosum (L.) Nesom SYN = A. dumosus L. Symphyotrichum laeve (L.) A. & D. Love var. purpuratum (Nees) Nesom SYN = A. laevis L. var. purpuratus (Nees) A. G. Jones Symphyotrichum lateriflorum (L.) A. & D. Love SYN = A. lateriflorum (L.) Britton Symphyotrichum racemosum (S. Elliot) Nesom SYN = A . fragilis Willd. Symphyotrichum oolentangiense (Riddell) Nesom SYN = A. oolentangiensis Riddell Symphyotrichum subulatum (Michx.) Nesom SYN = A. subulatus Michx. Baccharis halimifolia L. Brickellia eupatorioides (L.) Shinners var. eupatorioides Evax vema Raf. Cirsium horridulum Michx. Conyza canadensis (L.) Cronq. Coreopsis lanceolata L. Coreopsis tinctoria Nutt. var. tinctoria 230 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 3, 2002 Croptilon divaricatum (Nutt.) Raf. Erigeron strigosus Muhl. ex Willd. Eupatorium capillifolium (Lam.) Small Eupatorium compos itifolium Walt. Gamochaeta falcata (Lam.) Cabrera SYN = Gnaphalium falcatum Lam. Grindelia lanceolata Nutt. var. lanceolata Helenium amarum (Raf.) H. Rock var. amarum Heterotheca subaxillaris (Lam.) Britt. & Rusby (including H . latifolia Buckl.) Iva angustifolia DC. lva annua L. Krigia cespitosa (Raf.) Chambers f. cespitosa Lactuca canadensis L. Liatris mucronata DC. Pluchea camphorata (L.) DC. Pyrrhopappus pauciflorus (D. Don) DC. SYN = P. multicaulis DC. Ratibida columnifera (Nutt.) Woot. & Standi. Rudbeckia hirta L. Rudbeckia missouriensis Boynt. & Beadle Solidago canadensis L. Xanthium strumarium L. FAMILY BETULACEAE Ostrya virginiana (Mill.) K. Koch FAMILY BIGNONIACEAE Bignonia capreolata L. Campsis radicans (L.) Seem, ex Bureau FAMILY BORAGINACEAE Heliotropium indicum L. Heliotropium procumbens Mill. Heliotropium tenellum (Nutt.) Torr. Onosmodium bejariense A. DC. var. bejariense FAMILY BUDDLEJACEAE Poly premum procumbens L. FAMILY CAMPANULACEAE Triodanis perfoliata (L.) Nieuw. var. biflora (R.& P.) Bradley Triodanis perfoliata (L.) Nieuw. var. perfoliata FAMILY CAPRIFOLIACEAE Lonicera japonica Thunb. BROWN ET AL. 231 Lonicera sempervirens L Viburnum rufidulum Raf. FAMILY CELASTRACEAE Euonymus americana I , FAMILY CONVOLVULACEAE Dichondra carolinensis Michx. Ipomoea cordatotriloba Dennstedt var. cordatotriloba FAMILY CORNACEAE Comus drummondii C. A. Mey. Comus florida L. FAMILY CUPRESSACEAE Juniperus virginiana L. var. virginiana FAMILY CUSCUTACEAE Cuscuta indecora Choisy var. longisepala Yuncker Cuscuta pentagona Engelm. var. pentagona FAMILY CYPERACEAE Car ex cherokeensis Schwein. Carex microdonta T. & H. Cyperus odoratus L. Cyperus virens Michx. Eleocharis montevidensis Kunth Fimbristylis autumnalis (L.) Roem. & Schult. Scleria oligantha Michx. FAMILY EBENACEAE Diospyros virginiana L. FAMILY EUPHORBIACEAE Croton monanthogynus Michx. Euphorbia bicolor Engelm. & Gray Euphorbia maculata L. syn = Chamaesyce maculata (L.) Small Euphorbia nutans Lag. SYN = C. nutans (Lag.) Small Euphorbia serpens Kunth SYN = C. serpens (Kunth) Small Euphorbia spathulata Lam. FAMILY FABACEAE Acacia angustissima (Miller) Kuntze var. hirta (Nutt.) B. L. Robinson Albizia julibrissin Durazzini Astragalus distortus T.& H. var. engelmannii (Sheld.) M. E. Jones 232 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 3, 2002 Cercis canadensis L. Chamae crista fasciculata (Michx.) Greene Crotalaria sagittalis L. Dalea compacta Sprengel var. compacta Dalea compacta Sprengel var. pubescens (A. Gray) Barneby Dalea multiflora (Nutt.) Shinners Desmanthus illinoensis (Michx.) MacM. Desmodium ciliare (Willd.) DC. Desmodium paniculatum (L.) DC. Erythrina herbacea L. Galactia volubilis (L.) Britt. Glottidium vesicarium (Jacq.) Harper SYN — Sesbania vesicaria (Jacq.) Ell. Indigofera miniata Ort. Lespedeza procumbens Michx. Lespedeza virginica (L.) Britt. Medicago lupulina L. Mimosa strigillosa T.& G. Neptunia pubescens Benth. Rhynchosia minima (L.) DC. Sesbania drummondii (Rydb.) Cory Sesbania herbacea (P. Mill) McVaugh SYN = Sesbania exaltata (Raf.) Cory Strophostyles umbellata (Willd.) Britt. Vicia ludoviciana Nutt. ssp. ludoviciana Vida sativa L. FAMILY FAGACEAE Quercus nigra L. Quercus shumardii Buckl. FAMILY GENTIANACEAE Centarium pulchellum (Sw.) Druce FAMILY GERANIACEAE Geranium carolinianum L. FAMILY HAMAMELIDACEAE Liquidambar styraciflua L. FAMILY HYPERICACEAE (CLUSIACEAE) Hypericum hypericoides (L.) Crantz. FAMILY IRIDACEAE Sisyrinchium rosulatum Bickn. (including the yellow-flowered S. exile Bickn.) BROWN ET AL. 233 FAMILY JUGLANDACEAE Cary a texana Buckl. Juglans nigra L. FAMILY JUNCACEAE J uncus marginatus Rostk. Juncus validus Cov. FAMILY LAMIACEAE Hedeoma hispidum Pursh Monarda citriodora Cerv. ex Lag. var. citriodora Monarda fi stulos a L. Prunella vulgaris L. Salvia azurea Lam. var. grandiflora Benth. Salvia lyrata L. Scutellaria cardiophylla Engelm. & Gray Scutellaria parvula Michx. var. parvula FAMILY LILIACEAE Allium ste llatum Nutt, ex Ker-Gawler Hypoxis sp. Nothoscordum bivalve (L.) Britt. FAMILY LOGANIACEAE Gelsemium sempervirens (L.) Ait. f. Mitreola petiolata (J. F. Gmel.) T. & G. FAMILY MYRICACEAE Myrica cerifera L. SYN = Morelia cerifera (L.) Small FAMILY NAJADACEAE Najas guadalupensis (Spreng.) Magnus FAMILY OLEACEAE Forestiera ligustrina (Michx.) Poir. Fraxinus americana L. Ligustrum lucidum Ait. f. L. sinense Lour. FAMILY ONAGRACEAE Ludwigia glandulosa Walt. Oenothera laciniata Hill Oenothera speciosa Nutt. ORCHIDACEAE Spiranthes tuberosa Raf. 234 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 3, 2002 FAMILY OXALIDACEAE Oxalis comiculata L. including O. dillenii Jacq. FAMILY PINACEAE Pinus echinata Mill. Firms taeda L. FAMILY PLANTAGINACEAE Plantago aristata Michx. Plantago rhodosperma Dene. Plantago virginica L. FAMILY PLATANACEAE Platanus occidentalis L. FAMILY POACEAE Agrostis hyemalis (Walt.) B.S.P. Andropogon gerardii Vitman Andropogon glomeratus (Walt.) B.S.P. Andropogon virginicus L. var. virginicus Aristida longespica Poir. var. geniculata (Raf.) Fern. Aristida oligantha Michx. Aristida purpurascens Poiret Bothriochloa bladhii (Retz.) S. T. Blake Bothriochloa ischaemum (L.) Keng. var. songarica (Fish. & Mey) Celarier & Harlan Bothriochloa longipaniculata (Gould) Allred & Gould Bouteloua curtipendula (Michx.) Torr. Bouteloua hirsuta Lag. Briza minor L. Bromus japonicus Thunb. ex. J. Murray Chasmanthium laxum (L.) Yates var. laxum Chasmanthium laxum (L.) Yates var. sessiliflorum (Poir.) Wipff & S. D. Jones Cynodon dactylon (L.) Pers. Dactyloctenium aegyptium (L.) Beauv. Dichanthelium acuminatum (Sw.) Gould & Clark var. acuminatum Digitaria ciliaris (Retz.) Koel. Elymus virginicus L. Eragrostis secundiflora Presl ssp. oxylepis (Torr.) S. D. Koch Lolium perenne L. Limnodea arkansana (Nutt.) L. H. Dewey Melica mutica Walt. Muhlenbergia capillaris (Lam.) Trin. BROWN ET AL. 235 Nas sella leucotricha (Trin. & Rupr.) Pohl SYN = Stipa leucotricha Trin. & Rupr. Panicum dichotomiflorum Michx. Panicum virgatum L. Paspalum dilatatum Poir. Paspalum langei (Fourn.) Nash Paspalum notatum Flugge Paspalum plicatulum Michx. Paspalum pubiflorum Rupr. ex Fourn. Paspalum urvillei Steud. Phalaris caroliniana Walt. Piptochaetium avenaceum (L.) Parodi Poa annua L. Setaria parviflora (Poiret) Kerguelen SYN = S. geniculata (Lam.) Beauv. Schizachyrium scoparium (Michx.) Nash So rg hast rum nutans (L.) Nash Sorghum halepense (L.) Pers. Sphenopholis obtusata (Michx.) Scribn. Sporobolus compositus (Poiret) Merrill var. compositus Trisetum interruptum Buckl. FAMILY POLYPODIACEAE Pleopeltis polypodioides (L.) Andrews & Windham var. michauxianum (Weath.) Andrews & Windham FAMILY PRIMULACEAE Anagallis arvensis L. FAMILY RANUNCULACEAE Anemone berlandieri Pritz. Delphinium carolinianum Walt. ssp. vimineum (D. Don) M. Warnock FAMILY RHAMNACEAE Berchemia scandens (Hill) K. Koch Rhamnus caroliniana Walt. SYN = Frangula caroliniana (Walt.) Gray FAMILY ROSACEAE Crataegus marshallii Egglston Crataegus spathulata Michx. Prunus angustifolia Marsh. Prunus caroliniana (Mill.) Aiton. Pyrus calleryana Dene Rubus argutus Link 236 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 3, 2002 SYN = R. louisianus Berger Rubus trivialis Michx. FAMILY RUBIACEAE Diodia virginiana L. Galium pilosum Aiton Galium virgatum Nutt. Houstonia pusilla Schoepf SYN = Hedyotis crassifolia Raf. Stenaria nigricans (Lam.) Terrell SYM = Hedyotis nigricans (Lam.) Frosb. RUTACEAE Zanthoxylum clava-herculis L. FAMILY SALICACEAE Populus deltoides Bart, ex Marsh. Salix nigra Marsh, var. nigra FAMILY SAPINDACEAE Sapindus saponaria L. var. drummondii (Hook. & Arn.) L. Benson FAMILY SAPOTACEAE Sideroxylon lanuginosum Michx. SYN = Bumelia lanuginosua (Michx.) Pers. FAMILY SCROPHULARIACEAE Agalinis purpurea (L.) Penn. Leucospora tnultifida (Michx.) Nutt. Mecardonia acuminatum (Walt.) Small Penstemon cobaea Nutt. Veronica arvensis L. FAMILY SMILACAEAE Smilax bona-nox L FAMILY TILIACEAE Tilia americana L. var. americana Tilia americana L. var. caroliniana (Mill.) Castig FAMILY TYPHACEAE Typha latifolia L. FAMILY ULMACEAE Ulmus alata Michx. Ulmus crassifolia Nutt. FAMILY VALERIANACEAE Valerianella radiata (L.) Dufr. f. parviflora (Dyal) Eggers BROWN ET AL. 237 FAMILY VERBENACEAE Callicarpa americana L. Verbena officinale L. ssp. halei (Small) Barber Verbena rigida Spreng. Verbena xutha Lehm. FAMILY VIOLACEAE Viola sororia Willd. FAMILY VITACEAE Ampelopsis arborea (L.) Koehne Parthenocissus quinquefolia (L.) Planch. Vitis cinerea (Engelm.) Engelm. ex Millardet Vitis mustangensis Buckl. Vitis rotundifolia Michx. Grasses, composites and legumes dominated accounting for 45 percent of the species. The soils are neutral to alkaline and very high in calcium (Table 1). They are very similar to prairies in Louisiana (MacRoberts & MacRoberts 1996). Discussion Windham Prairie is about 80 km east of the once extensive Blackland (Fayette) Prairie and about 60 km north of the Coastal prairies (Smeins & Diamond 1983; Diamond & Smeins 1984). Between Windham Prairie and the Blackland Prairie and the Coastal Prairie are other isolated prairies (Nesom & Brown 1998). On the basis of published (but incomplete) plant lists, Windham Prairie shows strong affinities to the upper clay/clay loam sections of the Fayette Prairie (Smeins & Diamond 1983; Diamond & Smeins 1988) and also to the upland section of the Coastal Prairie (Diamond & Smeins 1984; 1988). This is not surprising considering the close proximity of these prairies and the fact, as pointed out by Diamond & Smeins (1984), that the upland Coastal Prairie and upper clay Fayette Prairie are not distinct but represent a north-south continuum of prairie communities. Many species rare to east Texas (region I of Hatch et al. 1990) occur in Windham Prairie. They include Symphotrichum oolentangiense , Liatris mucronata , Grindelia lanceolata, Rudbeckia missouriensis , Carex microdonta , Acacia angustissima var. hirta, Dalea compacta var. compacta and D. compacta var. pubescens, Allium stellatum and Penstemon cobaea . Symphotrichum oolentangiense is mapped in 12 region I counties (Turner et al. in press). Polk county is a new county record. It occurs 238 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 3, 2002 Table 1 . Soil characteristics of Windham Prairie (OM = organic matter) . Exchangeable Ions (ppm) Sample PH P K Ca Mg OM% 1 7.1 20 153 6361 94 1.5 2 7.6 2 226 12502 190 2.3 3 7.9 7 137 11652 57 0.8 4 8.0 6 190 14553 93 2.5 widely in Louisiana prairies (Thomas & Allen 1993-1998; MacRoberts & MacRoberts 1996). Liatris rmcronata is reported abundant on the Edwards Plateau, the Plains Country, and north-central Texas but rare in east Texas. Polk is one of four counties mapped in region I (Turner et al. in press). Grindelia lanceolata was mapped only in Hardin and Walker counties by Nesom (1990). Turner et al. (in press) mapped two additional counties, San Jacinto and Polk. It is abundant here. Rudbeckia missouriensis is abundant on the site. Specimens of this taxon at SBSC and ASTC are from Polk County. At SB SC there is one additional record from Montgomery County. Turner et al. (in press) mapped it in Walker, Polk and Tyler counties. It also occurs in Louisiana prairies (Thomas & Allen 1993-1998). Car ex microdonta is one of the few Carex prairie species. Correll & Johnston (1970) consider this species rare in east Texas. Turner et al (in press) mapped it only in the region I county of San Jacinto. Polk is a new county record. It is common in Louisiana prairies (Thomas & Allen 1993-1998; MacRoberts & MacRoberts 1996). Acacia angustissima var. hirta is a small shrub that neither Turner (1959) nor Isely (1998) mapped for central-east Texas. Turner et al. (in press) mapped it in Cherokee and Montgomery counties of region I. This is a new county record. It is rare in Louisiana (Thomas & Allen 1993-1998). Dalea compacta var. comp acta and D. comp acta var. pubes cens are rare in east Texas. Dalea compacta var. compacta is mapped only in four region I counties including Polk (Turner et al. in press). The Polk Windham Prairie collection of D. compacta var. pubescens is a new county record. Dalea compacta var. pubescens is recorded for Louisiana (Thomas & Allen 1993-1998). BROWN ET AL. 239 Allium stellatum is mapped in five mostly north-central Texas counties (Turner et ah in press). This is the first record for east Texas. It is disjunct from Van Zandf the nearest mapped county. Penstemon cobaea is mapped in the region I counties of Montgomery, Walker, Houston, and Anderson (Turner et ah in press). The Windham Prairie record is one of the most eastern Texas stations. Acknowledgments Thanks to Billie L. Turner of the University of Texas at Austin (TEX/LL) for sending distribution maps from his soon to be published Atlas of the Texas Flora. These have helped us to a better understanding of the distribution of some significant Windham Prairie plants. The first author is thankful to the Houston Community College for a sabbatical leave during the spring semester of 2002 which aided the completion of this project. Literature Cited Carr, W. R. 1993. A botanical inventory of blackland prairie openings in the Sam Houston National Forest. Unpublished report. Texas Natural Heritage Program, Texas Parks and Wildlife Department, Austin, Texas, 60 pp. Correll, D. S. & M. C. Johnston. 1970. Manual of the vascular plants of Texas. Texas Research Foundation, Renner, Texas, 1881 pp. Diamond, D. D. & F. E. Smeins. 1984. Remnant grassland vegetation and ecological affinities of the upper coastal prairie of Texas. Southwestern Naturalist, 29:321-334. Diamond, D. D. & F. E. Smeins. 1988. Gradient analysis of remnant true and upper coastal prairie grasslands of North America. Canadian Journal of Botany, 66:2152-2161 . Diamond, D. D. & F. E. Smeins. 1993. The native plant communities of the Blackland Prairie. Pp. 66-81, in The Texas Blackland Prairie, land, history, and culture (M.R. Sharpless & J.C. Telderman, eds.), Baylor Univ. Program for Regional Studies, Waco, Texas, 202 pp. Diamond, D. D., D. H. Riskind & S. L. Orzell. 1987. A framework for plant community classification and conservation in Texas. Texas Journal of Science, 39:203-221. Diggs, G. M., B. L. Lipscomb & R. J. O’Kennon. 1999. Illustrated flora of north central Texas. Sida, Botanical Miscellany, 16:1-1626. Foti, T. L. 1989. Blackland prairies of southwestern Arkansas. Proceedings of the Arkansas Academy of Science, 43:23-28. Hatch, S. L., N. K. Gandhi & L. E. Brown. 1990. Checklist of the vascular plants of Texas. MP-1655. The Texas Agricultural Experiment Station, College Station, Texas, 158 pp. Isely, D. 1998. Native and naturalized Leguminosae (Fabaceae) of the United States (exclusive of Alaska and Hawaii). Brigham Young University Press, Provo, Utah, 1 107 PP- Jordan, T. G. 1973. Pioneer evaluation of vegetation in frontier Texas. Southwestern Historical Quarterly, 76:233-254. Jordan, T. G. 1977. Early northeast Texas and the evolution of western ranching. Annals of the Association of American Geographers, 67:66-87. 240 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 3, 2002 Kucera, C. L. 1992. Tail-grass prairie. Pp. 227-268, in Ecosystems of the World: natural grasslands (R.T. Coupland ed.), Vol. 8a. Elsevier, Amsterdam, 476 pp. Leidolf, A. & S. McDaniel. 1998. A floristic study of black prairie plant communities at sixteen section prairie, Oktibbeha County, Mississippi. Castanea, 63:51-62. MacRoberts, B. R. & M. H. MacRoberts. 1996. The floristics of calcareous prairies on the Kisatchie National Forest, Louisiana. Phytologia, 81:35-43. MacRoberts, M. H. & B. R. MacRoberts. 1997. Former distribution of prairies in northern Louisiana. Phytologia, 82:315-325. MacRoberts, M. H., B. R. MacRoberts & L. S. Jackson, in press. Louisiana prairies, in Blackland prairies of the gulf coastal plain: nature, culture, and sustainability (E. Peacock & T. Schauwacker, eds). Univ. of Alabama Press, Tuscaloosa. McEwen, H., K. Griffith & J. D. Deshotels. 1987. Soil survey of Polk and San Jacinto counties, Texas. U.S.D.A. Soil Conservation Service, Washington, D.C. Moran, L. P., D. E. Pettry, R. E. Switzer, S. T. McDaniel & R. G. Wieland. 1997. Soils on native prairie remnants in the Jackson prairie region of Mississippi. Bull. 1067. Mississippi Agricultural & Forestry Experiment Station, 13 pp. Nesom, G. L. 1990. Studies in the systematics of Mexican and Texan Grindelia (Asteraceae:Astereae). Phytologia, 68:303-332. Nesom, G. L. & L. E. Brown. 1998. Annotated checklist of the vascular plants of Walker, Montgomery, and San Jacinto counties, east Texas. Phytologia, 84:107-153. Noss, R. F., E. T. LaRoe & J. M. Scott. 1995. Endangered major ecosystems of the United States: A preliminary assessment of loss and degredation. Biological Report 28, U.S. Dept. Interior, Biological Service, Washington, D.C., 58 pp. Sims, P. L. & P. G. Risser. 2000. Grasslands. Pp. 324-356, in North American terrestrial vegetation (M.G. Barbour & W.D. Billings, eds.), Cambridge University Press, New York, 708 pp. Smeins, F. E. & D. D. Diamond. 1983. Remnant grasslands of the Fayette Prairie, Texas. American Midland Naturalist, 110:1-13. Smeins, F. E. & D. D. Diamond. 1986. Grasslands and savannahs of east central Texas: ecology, preservation status and management problems. Pp. 381-394, in Wilderness and natural areas in the eastern United States: a management challenge (D.L. Kulhavy & R.N. Conner eds.), Stephen F. Austin State University, Nacogdoches, Texas, 416 pp. Smeins, F. E., D. D. Diamond & C. W. Hanselka. 1992. Coastal prairie. Pp. 269-290, in Ecosystems of the World: natural grasslands. (R.T. Coupland ed.), Vol. 8a. Elsevier, Amsterdam, 476 pp. Thomas, R. D. & C. M. Allen. 1993-1998. Atlas of the vascular flora of Louisiana. Louisiana Department of Fish & Wildlife, Baton Rouge, Louisiana, 679 pp. Turner, B. L. 1959. The legumes of Texas. University of Texas Press, Austin, Texas, 284 pp. Turner, B. L., H. Nichols, O. Doron & G. C. Denny. In press 2002. Atlas of the Texas Flora. Vol. LDicots and Vol. 2 Monocots. Sida, Botanical Miscellany. LEB at: larry@theplantman.net TEXAS J. SCI. 54(3): 24 1-248 AUGUST, 2002 NOTEWORTHY PLANTS ASSOCIATED WITH THE GULF COASTAL BEND OF TEXAS. I. G. Negrete, C. Galloway and A. D. Nelson* Department of Biology, Box 158 Texas A&M University-Kingsville, Kingsville, Texas 78363 and * Department of Biological Sciences, Box T-0100 Tarleton State University, Stephenville, Texas 76402 Abstract.— Based on data from fieldwork on northern Padre and Mustang islands in the Coastal Bend region of Texas, 31 species of vascular plants in 19 families are reported as new distribution and occurrence records. Three species ( Cakile geniculata, Helianthus debilis subsp. cucumerifolius and Sisyrinchium sagittiferum ) represent new additions to the flora of the Coastal Bend region. In this study, distribution and occurrence of plant species from Mustang Island (Gillespie 1976), northern Padre Island (Negrete et al. 1999) and the Texas Coastal Bend (Jones 1982) were examined. These data were compared to known ranges as well as searches of the literature and local herbaria. Study Area and Methods Northern Padre Island, east of Corpus Christi, Texas, is the location of Padre Island National Seashore (PINS) and Mustang Island, located north of PINS, is the site of Mustang Island State Park. Both were formed by deposition of sand during the Pleistocene (McAlister & McAlister 1993) and are important barrier islands located in the Coastal Bend region of southeastern Texas. Annual temperatures increase and annual precipitation decreases from north to south down the Texas coast (McAlister & McAlister 1993). Typical habitats found on the islands include coppice dunes, foredunes, barrier flats and tidal flats (Nelson et al. 2000). Data on distribution and occurrence of species was obtained from fieldwork reported in Negrete et al. (1999) at PINS from 1996-1998 and 242 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 3, 2002 Gillespie (1976) at Mustang Island State Park from 1967-1968. This data was compared to ranges and occurrence of species in Jones (1982). Nomenclature of native or naturalized plants was standardized using Jones et al. (1997). Taxa are discussed alphabetically according to class and family. Voucher specimens reported in this study are deposited with the holdings of the Corpus Christi Museum (C.C. Museum), Padre Island National Seashore (PINS), Tarleton State University Herbarium (TAC) and Texas A&M University-Kingsville Herbarium (TAIC). Results and Discussion This analysis resulted in new distribution and occurrence records for 31 species from the Texas Coastal Bend in relation to information currently available in the Flora of the Coastal Bend (Jones 1982). Cakile geniculata, Helianthus debilis subsp. cucumerifolius and Sisyrinchium sagittiferum represent new additions to the flora of the Coastal Bend region. New distribution records for taxa are discussed individually. CLASS LILIOPSIDA FAMILY COMMELINACEAE Commelina erecta L. var. erecta is reported from Mustang Island (Gillespie 1976) and northern Padre Island (Negrete et al. 1999). It is considered common in prairies, openings, stream bottoms and in waste places along roads in the Coastal Bend (Jones 1982). It was found in waste places and barrier flats on northern Padre Island (TAIC N303). FAMILY CYPERACEAE Fuirena scirpoidea Michx. occurs in sandy depressions and marshes along the coast from north of Fulton, south of Ingelside, and the King Ranch (Jones 1982). Its range can now be extended to northern Padre Island by the voucher specimen (C.C. Museum 76D457) reported in Negrete et al. (1999). FAMILY IRIDACEAE Sisyrinchium sagittiferum Bickn. is found as a part of the flora of NEGRETE, GALLOWAY & NELSON 243 Mustang (Gillespie 1976) and northern Padre Island (Negrete et al. 1999). It is also found in low wet areas in eastern Texas (Correll & Johnston 1970) but has not been previously reported from the Coastal Bend region (Jones 1982). A voucher specimen (PINS 724) was located at the PINS Herbarium. FAMILY JUNCACEAE J uncus megacephalus M. A. Curtis has been reported from a marsh above the beach north of Fulton (Jones 1982) in the Coastal Bend region. The specimen was collected in the barrier flat of PINS (TAC N699) and its range should be extended to northern Padre Island (Negrete et al. 1999). Juncus roemerianus Scheele has been reported from sandy swales on the Aransas National Wildlife Refuge (Jones 1982). Its range can now be extended to northern Padre Island by the voucher specimen (PINS 2484) reported in Negrete et al. (1999). CLASS MAGNOLIOPSIDA FAMILY ACANTHACEAE Ruellia corzoi Tharp & Barkl. is frequent on dry sand and caliche from Mathis to Orange Grove, Alice and Premont (Jones 1982) in the Coastal Bend region. Its range can now be extended to northern Padre Island because of the voucher specimen (PINS 2532) reported in Negrete et al. (1999). FAMILY APIACEAE Hydrocotyle umbellata L. occurs on Mustang (Jones 1982) and northern Padre Island (Negrete et al. 1999). It is locally abundant on damp sands in swales, depressions and around lakes (Jones 1982) in the Coastal Bend region. It has also been reported from Aransas National Wildlife Refuge, north of Rockport, and Welder Wildlife Refuge (Jones 1982). Its range can now be extended to northern Padre Island due to the presence of a voucher specimen (PINS 1390) reported in Negrete et al. (1999). 244 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 3, 2002 FAMILY ASTERACEAE Croptilon divaricatum (Nutt.) Raf. is occasional on deep coastal sands and has been reported from south of Ingleside and at Goose Island State Park (Jones 1982). Its range can now be extended to northern Padre Island because of the voucher specimen (TAIC CSP 44) reported in Negrete et al. (1999). Helianthus debilis Nutt, subsp. cucumerifolius (Torr. & A. Gray) Heiser is found as part of the flora of northern Padre Island (Negrete et al. 1999) but was not reported as a part of the Coastal Bend flora (Jones 1982). A specimen was collected in the tidal flat of northern Padre Island (TAG N684). Helianthus praecox Engel m. & A. Gray subsp. runyonii (Heiser) Heiser occurs in coastal sands in a variety of habitats in the Coastal Bend region (Correll & Johnston 1970; Jones 1982). A voucher specimen (PINS 2364) located in the PINS Herbarium extends the range onto northern Padre Island. Palafoxia hookeriana Torr. & A. Gray occurs on northern Padre Island (Negrete et al. 1999) and has been reported from Rockport and Aransas Pass (Jones 1982). A voucher specimen (PINS 1383) was located in the PINS Herbarium that extends the range onto northern Padre Island. Thelesperrna nuecense Turner was reported by Jones (1982) to occur frequently along the mainland coast from Aransas National Wildlife Refuge to Baffin Bay. A voucher specimen (PINS 1348) located in the PINS Herbarium that extends the range onto northern Padre Island. FAMILY AVICENNIACEAE Avicennia germinans (L.) L. is occasional to locally abundant on moist sandy soils along beaches and on marshy islands at Packery Channel, Harbor Island and the mouth of the Aransas River (Jones 1982). Its range can now be extended to northern Padre Island by the presence of a voucher specimen (PINS 6668) reported in Negrete et al. (1999). FAMILY BRASSICACEAE Cakile geniculata (Robins.) Millsp. occurs on beaches and sandy NEGRETE, GALLOWAY & NELSON 245 places near the ocean (Correll & Johnston 1970) but was not included in the flora of the Coastal Bend region (Jones 1982). Its range can now be extended to the Coastal Bend region and northern Padre Island by the voucher specimen (TAIC 398) reported in Negrete et al. (1999). FAMILY CISTACEAE Helianthemum georgianum Chapm. is frequent in sandy oak woods along the coast and is known from woods and pastures on the Welder Wildlife Refuge and south of Refugio (Jones 1982). It has been reported from northern Padre Island (PINS 518) by Negrete et al. (1999) and this extends its range onto the island. FAMILY CLUSIACEAE Hypercium gentianoides (L.) Britton, Stems & Poggenb. is frequent on coastal sands in swales from Aransas National Wildlife Refuge to Flour Bluff and southward (Jones 1982). It has been reported from northern Padre Island (PINS 2430) by Negrete et al. (1999). Hypercium hypercoides (L.) Crantz subsp. hypercoides is frequent on deep coastal sands from Aransas National Wildlife Refuge to Flour Bluff and southward into the Coastal Bend region (Jones 1982). It has been reported from northern Padre Island (PINS 721) by Negrete et al. (1999). Hypericum pauciflorum H.B.K. is occasional on sandy soils in openings, prairies or lowlands of the barrier islands (Jones 1982). It has been reported from the adjacent mainland at the Texas A & M University-Kingsville Biological Station located on Baffin Bay (TAC 29). FAMILY CUSCUTACEAE Cuscuta indecora Choisy occurs abundantly in the Nueces River bottom near Calallen (Jones 1982). It was reported from northern Padre Island (PINS 885) by Negrete et al. (1999). FAMILY FABACEAE Clitoria mariana L. is occasional on deep coastal sands southwest of Aransas Pass and west of Flour Bluff (Jones 1982). It has been reported 246 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 3, 2002 from northern Padre Island (PINS 2400) by Negrete et al. (1999) and this extends its range onto the island. Crotolaria sagittalis L. was previously reported from the Aransas National Wildlife Refuge to Flour Bluff and southward into the Coastal bend region (Jones 1982). It has been reported from northern Padre Island (PINS 4092) by Negrete et al. (1999) thus extending its range onto the island. Dalea obovata (Torr. & A. Gray) Shinners is frequent on coastal sands from Aransas National Wildlife Refuge to Flour Bluff and southward into the Coastal Bend region (Jones 1982). It was reported from northern Padre Island (PINS 738) by Negrete et al. (1999) extending its range onto the island. Erythrina herbacea L. occurs on Mustang (Gillespie 1976) and northern Padre Island (Negrete et al. 1999). It is frequent on coastal sands from south of Refugio, Aransas National Wildlife Refuge, Flour Bluff and southward into the Coastal Bend region (Jones 1982). This species range can now be extended to northern Padre Island due to the presence of a voucher specimen (PINS 734) reported by Negrete et al. (1999). Glottidium vesicaria (Jacq.) R. M. Harper occurs on damp sands from south of Refugio, Aransas Pass and northeast of Orange Grove (Jones 1982). It has been reported from northern Padre Island (PINS 7748) by Negrete et al. (1999). FAMILY FAGACEAE Quercus minima (Sarg.) Small was reported as frequent on coastal sands from Aransas National Wildlife Refuge to Flour Bluff and southward into the Coastal Bend region (Jones 1982). It has been reported from northern Padre Island (CC Museum 7922) by Negrete et al. (1999). Quercus virginiana Mill. var. virginiana is frequent on coastal sands from Aransas National Wildlife Refuge to Baffin Bay (Jones 1982). Its range can now be extended to northern Padre Island (PINS 931) by Negrete et al. (1999). NEGRETE, GALLOWAY & NELSON 247 FAMILY MALVACEAE Kosteletzkya virginica (L.) K. Presl ex A. Gray occurs on moist soil from northeast of Tivoli, Aransas National Wildlife Refuge, north of Rockport, and on Mustang Island (Gillespie 1976; Jones 1982). Its range can now be extended to northern Padre Island (CC Museum 6912) by Negrete et al. (1999). FAMILY MYRICACEAE Morelia cerifera (L.) Small was reported as frequent on coastal sands from Aransas National Wildlife Refuge to Flour Bluff and southward into the Coastal Bend region (Jones 1982). It has been found on northern Padre Island (PINS 764) as reported by Negrete et al. (1999). FAMILY POLYGALACEAE Poly gala incamata L. was reported as occasional in moist coastal sands and sandy oak woods northwest of Refugio (Jones 1982). It has been reported from northern Padre Island (TAC N637) by Negrete et al. (1999). FAMILY POLYGONACEAE Polygonella polygama (Vent.) Engelm. and A. Gray was reported as infrequent in sandy low grounds along the coast from Aransas National Wildlife Refuge to Flour Bluff (Jones 1982). It has been reported from northern Padre Island (PINS 6698 and TAC N637) by Negrete et al. (1999). FAMILY RUTACEAE Thamnosma texana (A. Gray) Torr. was reported as frequent on dry sand and caliche north and west of Mathis and north of Orange Grove (Jones 1982). It has been reported south of these localities on a rocky hillside at Captain’s Pond in Kingsville, Texas (TAC N281). Acknowledgments The authors thank Paul Eubank and Darrell Echols at PINS, R. Nelson and M. Goetze, and students at TSU for field assistance. The authors appreciated support from Organized Faculty Research at TSU. The following herberia were helpful in providing access to their collections: CC Museum, PINS, TAC and TAIC. 248 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 3, 2002 Literature Cited Correll, D. S. & M.C. Johnston. 1970. Manual of the vascular plants of Texas. Texas Research Foundation. Renner, Texas., 1083 pp. Gillespie, T. S. 1976. The flowering plants of Mustang Island, Texas--An annotated checklist. Texas J. Sci., 27(1): 131-148. Jones, F. B. 1982. Flora of the Texas Coastal Bend, 3rd ed. Welder Wildlife Foundation, S inton, Texas, 267 pp. Jones, S. D., J. K. Wipff & P. M. Montgomery. 1997. Vascular Plants of Texas, A comprehensive checklist including synonymy, bibliography, and index. University of Texas Press, Austin, 404 pp. McAlister, W. H. & M. K. McAlister. 1993. A naturalist’s guide: Matagorda Island. University of Texas Press, Austin, Texas, 354 pp. Negrete I. G., A. D. Nelson, J. Goetze, L. Macke, T. Wilburn & A. Day. 1999. A Checklist for the vascular plants of Padre Island National Seashore. Sida, 18: 1241-1259. Nelson, A. D., J. R. Goetze, I. G. Negrete, V. E. French, M. P. Johnson & L. Macke. 2000. Vegetational analysis and floristics of four communities in the Big Ball Hill region of Padre Island National Seashore. Southwestern Naturalist 45 (4): 43 1-442. ADN at: nelson@tarleton.edu TEXAS J. SCI. 54(3): 249-260 AUGUST, 2002 MUTAGENIC ACTIVITY OF IDARUBICIN AND EPIRUBICIN IN THE BACTERIUM SALMONELLA TYPHIM URI UM John M. Brumfield and William J. Mackay Edinboro University of Pennsylvania Department of Biology & Health Services Edinboro, Pennsylvania 16444 Abstract. — Idarubicin (structural analogue of daunomycin) and epirubicin (epimer of adriamycin) are two "second-generation" anthracyclines that are widely used in chemotherapy against leukemias and metastatic breast tumors, respectively. This study shows that using the Salmonella Mutagenicity Assay that idarubicin, like daunomycin, can induce frameshift (TA98; 16.1 fold), GC to AT (TA7004; 8.7 fold), and AT to GC base-substitution transition mutations (TA7001; 2.8 fold). Epirubicin, like adriamycin, can also induce frameshift (19.7 fold) and GC to AT transition mutations (14.6 fold) in this assay. Interestingly, epirubicin, but not adriamycin, can induce AT to GC mutations (2.9 fold) in this assay. Anthracy cline antibiotics, primarily daunomycin and adriamycin, have been utilized in clinical practice since the 1960’s and represent one of the most commonly used classes of anticancer drugs against leukemias and solid tissue tumors (reviewed by Sinha & Politi 1990; Hande 1998; Gewirtz 1999; Ogura 2001; Felix 2001). However, these highly active chemotherapeutic agents are also associated with acute cardiotoxic effects and a dose-related cardiomyopathy (reviewed by Hortobagyi 1997; Keefe 2001). Extensive efforts since 1972 have resulted in the replacement of these parent compounds with less toxic "second-genera¬ tion" structural analogues (reviewed by Arcamone 1984; Carella et al. 1990; Fields & Koeller 1991; Hollingshead & Faulds 1991; Borchmann et al. 1997; Platel et al. 1999). Among these new compounds, 4-me- thoxy-daunorubicin (idarubicin), a structural analogue of daunomycin, was shown to be effective against acute nonlymphocytic leukemias with reduced cardiotoxic effects in clinical trials (Cersosimo 1992; Bogush & Robert 1996; Andersson et al. 1999; Lee et al. 2001). Similarly, 4’-epidoxorubicin (epirubicin), a 4’-epimer of adriamycin, is now widely used against early and metastatic breast cancers (Ganzina 1983; Weiss 1992; reviewed by Hortabagyi 2000; Razis & Fountzilas 2001; Trudeau & Pagani 2001). It is known that anthracyclines (particularly the sugar moiety of the compound) can interfere with a number of biochemical and biological 250 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 3, 2002 functions in the cell. Several studies strongly suggest that membrane binding and poisoning of topoisomerase II activity are possible modes of action for these anticancer antibiotics (Binaschi et al. 1998; Arcamone et al. 1999; Guano et al. 1999; Zunino at al. 2001). However, numerous studies also suggest that the biological activity of these drugs correlate with DNA binding with a preference to the GC bases (Fenick et al. 1997; Taatjes et al. 1997; Davies et al. 2000; Eaton et al. 2000; Qu et al. 2001). Several studies have demonstrated that anthracy dines are mutagenic in prokaryotic and eukaryotic cells (Marzin et al. 1983; Babudri et al. 1984; Olinski et al. 1997; El-Mahdy & Othman 2000; Mackay et al. 2000; Mackay & Phelps 2001). However, the mutageni¬ city of anthracy dines has been underestimated in the past, partly since these drugs are only "slightly" mutagenic in microbial assays (Kaldor et al. 1986; Tominaga 1986; Bokemeyer & Schmoll 1995). More recently, however, it has been suggested that the mutagenic properties of anthra¬ cy dines during tumor treatment may result in secondary cancers following chemotherapy (Olinski et al. 1997; 1998; Baguley & Ferguson 1998; Vakeva et al. 2000). Thus, it is very important to fully character¬ ize both the mutational spectrum and the mutagenic specificity of anthracy dines as a first step to better understand the antitumor and possible precarcinogenic effects of these compounds. Efforts in this laboratory have focused primarily in defining the mutagenic specificity of anthracyclines using prokaryotic genotoxicity assays. The use of bacterial mutation assays is now firmly established both for fundamental studies in mutagenesis and carcinogenesis, and for screening chemicals and environmental samples for genotoxic properties. The most used and validated bacterial reverse-mutation assay is the Salmonella Mutagenicity Assay (reviewed by Mortelmans & Zeiger 2000). The original Ames tester strains (i.e., TA98, TA100, etc.) identified mutagens which reverted point mutations in the his operon of Salmonella typhimurium. Although the Salmonella Mutagenicity Assay has been widely used to screen chemicals for potential genotoxicity, it was not originally designed to yield information about the precise nature of the his+ revertants that were obtained. For example, TA100 detects primarily GC to AT events, but this strain can also detect GC to TA and extragenic tRNA suppressor mutations (Koch et al. 1994). TA98 is a tester strain that detects compounds that induce frameshift mutations (Maron & Ames 1983). A new set of Salmonella strains was subse- BRUMFIELD & MACKAY 251 quently generated to identify specific base-substitution events (Gee et al. 1994; 1998). Since each strain can only revert by a single specific mutational event, it is not necessary to further classify or sequence the resulting revertants in order to know the mutation that has occurred. Previous studies in this laboratory have shown that daunomycin can induce both frameshift and base-substitution transition mutations in Salmonella typhimurium (Mackay et al. 2000), while adriamycin induces frameshift and GC to AT transition events (Mackay & Phelps 2001). Interestingly, although adriamycin is structurally very similar to daunomycin, it does not induce AT to GC mutations in this assay (Mackay & Phelps 2001). The present study was initiated to examine if the "second-generation" structural analogues of daunomycin and adriamycin, namely idarubicin and epirubicin, could induce frameshift and transition mutations. This report demonstrates that both compounds can induce frameshift, GC to AT, and AT to GC transition events in Salmonella typhimurium. Materials and Methods Bacterial strains.— The strains and their genotypes used in this study are listed in Table 1. TA98 detects frameshift mutations (Maron & Ames 1983). TA7001 and TA7004 are base-substitution specific strains, which carry a target missense mutation in the hisG gene. The latter two strains revert to a prototrophic his+ phenotype via a specific base- substitution event (TA7001, AT to GC, and TA7004, GC to AT) (Gee et al. 1994; 1998). Chemicals. — Adriamycin (doxorubicin hydrochloride), dimethyl sulfoxide (DMSO), ICR 191 acridine mutagen (6-chloro-9-[3-(2- chloroethylamino)propylamino]-2-methoxyacridine), N4-aminocytidine (N4AC), 4-nitroquinoline-N-oxide (4NQO) and streptonigrin (STN) were obtained from Sigma Chemical Co. (St. Louis). Idarubicin (Idamycin®) and epirubicin (Ellence™) were obtained from the Erie Cancer Research Center. Mutation assays. — The his reversion assays (triplicate assays were conducted for each strain) followed a modified version of the traditional Ames “plate-incorporation” test (Maron & Ames 1983) which utilized 252 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 3, 2002 Table 1. Ames Salmonella strains. Strain Genotype Mutation Detected TA7001 hisGl 775 Aara9 Achll004 (bio chlD uvrB chlA) galE503 rfal041/pKM\Q\ A:T to G:C TA7004 hisG9133 Aara9 Achll004 (bio chlD uvrB chlA) galE503 rfa 1 044 /pKMIO l G:C to A:T TA98 hisD3052 Aara9 A chll008 (bio chlD uvrB gal) rfa!004/ pKMIOl (-1 C) frameshift a pre- incubation step in order to increase the sensitivity of the strains to each anthracy cline compound and has been previously described (Mackay et ah 2000; Mackay & Phelps 2001). Very briefly, 110 /xL (approximately 2.0 x 108 cells) of a stationary phase S. typhimurium culture (TA98, TA7001, TA7004) was exposed to either daunomycin, idarubicin, adriamycin, or epirubicin (120 /xg/mL) for 30 minutes in a shaking incubator (250 rpm) at 37 °C. 100 /xL of this culture was plated onto minimal agar plates that contained 2% glucose, 0.05 mM L-histi- dine, and 0.005 mM biotin. These selective plates were incubated at 37 °C, and the numbers of his+ revertants were scored after 48 hr. For each strain, a "zero" control (culture that was not exposed to the experimental chemical) was included in order to estimate the number of spontaneous his+ revertants in each experiment. The total number of viable cells in each experiment was determined by plating serial dilutions onto nonselective plates (Luria-Bertani, DIFCO). Mutation frequency is expressed as the average number of his+ revertants on selective plates divided by the total number of viable cells (determined by the number of colonies on the non-selective plates). Positive control chemicals The Salmonella strains TA98, TA7001, and TA7004 were tested using positive control chemicals that are known to be mutagenic in this assay (Maron & Ames 1983; Gee et ah 1998; Christopher Sommers pers. comm.). ICR 191 was prepared in DMSO (4.0 /xg/ plate) and used as a positive control for TA98. STN, dissolved in DMSO (50 jxg/plate) , and N4AC, prepared in sterile deionized water (10 pcg/plate) were used as positive controls for TA7001. 4NQO, dissolved in DMSO (0.4 jug/ plate), and N4AC (10 /xg/plate) were used as positive controls for TA7004. BRUMFIELD & MACKAY 253 Results and Discussion Previous studies in this laboratory have shown that daunomucin and adriamycin can induce frameshift (i.e., TA98) and GC to AT base- substitution mutations (i.e., TA7004) in the bacterium Salmonella typ hi murium (Mackay et al. 2000; Mackay & Phelps 2001). Dauno- mycin also can induce AT to GC events (i.e., TA7001) in this assay (Mackay et al. 2000). However, adriamycin, a compound that is structurally very similar to daunomycin, did not induce AT to GC mutations in Salmonella (Mackay & Phelps 2001). This current study demonstrates that idarubicin and epirubicin, "second-generation" structural analogues of daunomycin and adriamycin, respectively, can induce frameshift and specific base-substitution transition events. The Ames tester strains are listed in Table 1 . Each Salmonella strain (i.e., TA98, TA7001, TA7004) was first verified using selected positive control chemicals, which have been shown to be mutagenic by Gee et al. (1994; 1998; Sommers, pers. comm.). Strains TA98, TA7001 and TA7004 were exposed to daunomycin, idarubicin, adriamycin, or epirubicin (120 /xg/ ml). The number of his+ revertants were monitored on selective minimal glucose plates and total number of cells on nonselective plates. Following the calculation of the total number of viable cells, it was possible to calculate a mutation fre¬ quency for each strain tested (in triplicate). Mutation frequency is expressed as the average number of his+ revertants on selective plates divided by the total number of viable cells. The results of this study are summarized in Tables 2 and 3. TA98 was highly mutable (mutation in¬ duction folds ranged from 15.4 to 44.5; x2' p< 0.005) when exposed to daunomycin, idarubicin, adriamycin, or epirubicin in this assay. These results demonstrate that the "second-generation" structural analogues, idarubicin and epirubicin, also induced frameshift mutations in this assay. The two base-substitution Ames strains (TA7001, AT to GC; TA7004, GC to AT) were mutable in the presence of daunomycin (mu¬ tation induction folds ranged from 7.3 to 7.6; x2« p< 0.005), while TA7004 was mutable in the presence of adriamycin (mutation induction fold 7.8; x2- p< 0.005). As expected from previous studies, adriamycin did not induce AT to GC (TA7001) mutations (mutation induction fold 1.1; x2- p>0. 10) in this assay. Mutation frequencies were also deter- 254 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 3, 2002 Table 2. Daunomycin/Idarubicin mutation frequencies of Ames Salmonella stains. Strain Spontaneous Mutation Frequency (S) Induced Mutation Frequency (I) (120 ng/m\ daunomycin) Induction Fold (daunomycin) (I/S) Induced Mutation Frequency (I) (120 /ig/ml idarubicin) Induced Fold (idarubicin) (I/S) TA7001 0.9 (± 0.1) X 10 8 6.6 (± 0.1) X 108 7.3 2.6 (± 0.1) X 108 2.8 TA7004 0.8 (± 0.1) X 107 6.1 (± 0.6) X 10 7 7.6 7.0 (± 0.1) X 10 7 8.7 TA98 1.1 (± 0.3) X 10 7 4.9 (+ 0.5) X 106 44.6 1.8 (± 0.3) X 10 6 16.1 mined for TA7001 and TA7004 in the presence of the "second-genera¬ tion" anthracy cline structural analogues, idarubicin and epirubicin. Both strains were mutable in the presence of either compound (mutation in¬ duction folds ranged from 2. 8 to 14.6; x2- p<0.01). Thus, while adria- mycin does not induce AT to GC mutations in Salmonella , the 4’-epimer of this compound, epirubicin, is mutagenic for this transition event. Several biochemical analyses suggest that the interaction(s) between anthracyclines, primarily daunomycin and adriamycin, and DNA are complex in nature (Davies et al. 2000; Eaton et al 2000; Qu et al. 2001). Anthracyclines can form DNA crosslinks in vivo (Skladanowski & Konopa 1994) and with GC base pairs, specifically a (GC)4 oligo¬ nucleotide, in vitro with formaldehyde (Taatjes et al. 1997; Fenick et al. 1997). The primary mode of action (antitumor effect) of anthracyclines appears to be the intercalation of the aglycone portion of the compound between adjacent DNA base pairs, and this activity results in topoiso- merase- induced DNA strand breaks (Liu 1989; Baguley & Ferguson 1998; Zunino et al. 2001). Alkylation of DNA and the production of reactive oxygen species have also been reported to cause DNA modifi¬ cations during anthracy cline chemotherapeutic treatments (Bokemeyer & Schmoll 1995; Olinski et al. 1997). This DNA damage has been found to possess premutagenic properties and, if not repaired, may contribute to carcinogenesis (Bokemeyer & Schmoll 1995; Olinski et al. 1998). The amino sugar is recognized to be a critical determinant of the antitumor activity of daunomycin and adriamycin. In an attempt to improve the pharmacological properties of these anticancer drugs, novel anthracyclines have been designed with altered amino sugars. Epirubi¬ cin differs from adriamycin by the epimerization of the OH group in position 4’ of the aminosugar moiety and has been shown to be less toxic during chemotherapeutic treatments against metastatic breast BRUMFIELD & MACKAY 255 Table 3. Adriamycin/Epirubicin mutation frequencies of Ames Salmonella stains. Strain Spontaneous Mutation Frequency (S) Induced Mutation Frequency (I) (120 /xg/ml adriamycin) Induction Fold (adriamycin) (I/S) Induced Mutation Frequency (I) (120 /xg/ml epirubicin) Induction Fold (epirubicin) (I/S) TA7001 0.9 (± 0.1) X 10* 1.0 (± 0.1) X 10 8 1.1 2.6 (+ 0.1) X 10 8 2.9 TA7004 0.8 (± 0.1) X 10 7 6.2 (± 0.7) X 107 7.8 1.2 (± 0.1) X 10 6 14.6 TA98 1.1 (± 0.3) X 10 7 1.7 (± 0.4) X IQ'6 15.4 2.2 (± 0.3) X 106 19.7 cancers (Ganzina 1983; reviewed by Hortabagyi 2000). Idarubicin, a 4-demthoxy- anthracy cline analogue of daunomycin, exhibits several features, which render this drug unique among anthracy dines against acute nonlymphocyctic leukemia. Its higher lipophilicity leads to faster accumulation within the nucleus, superior DNA-binding capacity, and consequently, a greater antitumor effect when compared to daunomycin (Borchmann et al. 1997). Furthermore, idarubicin can be administered orally at effective plasma concentrations and exhibits reduced cardiotoxi- city when administered at therapeutic doses (Cerosimo 1992; Weiss 1992). These mutational analyses also suggest that the interaction(s) between these anthracy dines and the DNA helix might indeed be very complex. Daunomycin and idarubicin exhibit similar mutagenic effects in Sal¬ monella (Table 2). However, the slight differences in structure between adriamycin and epirubicin give rise to different mutational spectra in this assay. Both compounds can induce GC to AT transition mutations (Mackay & Phelps 2001; this report). However, unlike epirubicin, which can also induce AT to GC mutations (Table 3), adriamycin is not mutagenic with TA7001 in the Salmonella Mutagenicity Assay (Mackay et al. 2000). These results suggest that there may exist small chemical differences between the binding of each antibiotic with DNA and also suggest that unique interactions of epirubicin and adriamycin with DNA might provide an explanation for the significantly different clinical activities of the two anticancer drugs. The incidence of many secondary cancers has been linked to high doses of chemotherapy (Kaldor et al. 1987; Swendlow et al. 1992; Olinski et al. 1997, 1998; Allen et al. 1998; Baguley & Ferguson 1998; Vakeva et al. 2000). In order to improve the clinical efficacy of anti- 256 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 3, 2002 neoplastic anthracycline compounds (i.e., daunomycin, adriamycin, epi- rubicin and idarubicin) during chemotherapy, it will be necessary to identify modulators of their activities that could potentially be exploited to sensitize target tissues to therapy or to protect nontarget tissues during therapy. Such modulators include metabolic activation pathways and DNA repair pathways. Anthracy dines can induce DNA crosslinks (Cullinane & van Rosmalen 1994; Cullinane et al. 2000). All cells have base excision repair mechanisms, which can recognize and remove cross-linking base adducts. For example, 3-methyladenine DNA glycosylase (Aag) recog¬ nizes and removes a variety of these DNA adducts (e.g. , groups that can attach at the N7 and N3 positions of the purine ring) (reviewed by Seeberg et al. 2000). Escherichia coli mutants (alkA tag), which lack Aag activity, are extremely sensitive to killing in the presence of monofunctional (for example, methyl methanesulfonate) and complex (for example, BCNU) alkylating agents (Evensen & Seeberg 1982; Clarke et al. 1984; Engel ward et al. 1996). Furthermore, addition of a mouse gene, which encodes Aag, to alkA tag cells, restores partial resistance to cell killing in the presence of several alkylating and DNA cross- linking agents (Engel ward et al. 1993). Future endeavors within this laboratory will determine if a base-excision repair pathway that includes Aag can recognize and repair anthracycline- induced DNA cross-links. Hopefully, these results may lead to a lessening of the detrimental effects of anthracycline compounds and/or increased antitumor efficacies of these drugs in future cancer chemotherapeutic treatments. These experiments are currently in progress. Acknowledgments The authors wish to thank Dr. Christopher Sommers for sharing his unpublished results and Dr. Susan Rosendahl for idarubicin and epiru- bicin. This research was partly supported by a Beta Beta Beta Founda¬ tion Research Scholarship to the senior author. Literature Cited Allen, J. M., B. P. Engelward, A. J. Dreslin, M. D. Wyatt, M. Tomasz & L. D. Samson. 1998. Mammalian 3-methyladenine DNA glycosylase protects against the toxicity and clastogenicity of certain chemotherapeutic DNA cross-linking agents. Cancer Res., 58(1 7): 3965-3973. BRUMFIELD & MACKAY 257 Andersson, B. S., S. Eksborg, R. F Vidal, M. Sundberg & M. Carl berg. 1999. Anthraquinone-induced cell injury: acute toxicity of carminomycin, epirubicin, idarubicin and mitoxantrone in isolated cardiomyocytes. Toxicology, 135(1): 1 1-20. Arcamone, F. 1984. Adriamycin and its analogs. Tumori, 70(2): 1 13-1 19. Arcamone, F., F. Animati, M. Bigioni, G. Capranico, C. Caserinin, A. Cipollone, M. De Cesare, A. Ettore, F. Guano, S. Manzini, E. Monteagudo, G. Pratesi, C. Salvatore, R. Supino & F. Zunino. 1999. Configurational requirements of the sugar moiety for the pharmalogical activity of anthracycline disaccharides. Biochem. Pharmacol., 57(10): 1 133-1 139. Babudri, N., B. Pani, M. Tamaro, C. Monti-Bragadin & F. Zunino. 1984. Mutagenic and cytotoxic activity of doxorubicin and daunorubicin derivatives on prokaryotic and eukaryotic cells. Br. J. Cancer, 50(l):91-96. Baguley, B. C. & L. R. Ferguson. 1998. Mutagenic properties of topoisomerase-targeted drugs. Biochim. Biophys. Acta., 1400( 1 -3) :2 13-222. Binaschi, M., R. Farinosi, C. A. Austin, L. M. Fisher, F. Zunino & G. Capranico. 1998. Human DNA topisomerase Ilalpha-dependent DNA cleavage and yeast cell killing by anthracycline analogues. Cancer Res., 58(9): 1886-1892. Bogush, T. & J. Robert. 1996. Comparative evaluation of the intracellular accumulation and DNA binding of idarubicin and daunorubicin in sensitive and multidrug-resistant human leukaemia K562 cells. Anticancer Res., 1 6( 1 ) : 365-368. Bokemeyer, C. & H. J. Schmoll. 1995. Treatment of testicular cancer and the development of secondary malignancies. J. Clin. Oncol., 13(l):283-292. Borchmann, P. K. Hyubel, R. Schnell & A. Engert. 1997. Idarubicin: a brief overview on pharmacology and clinical use. Int. J. Clin. Pharmacol. Ther., 35(2):80-83. Carella, A. M., E. Berman, M. P. Maraone & F. Ganzina. 1990. Idarubicin in the treatment of acute leukemias. An overview of preclinical and clinical studies. Haematologica, 75(2): 159-169. Cersosimo, R. J. 1992. Idarubicin: an anthracycline antineoplastic agent. Clin. Pharm., 11 (2): 152- 167. Clarke, N., M. Kvaal & E. Seeberg. 1984. Cloning of E. coli genes encoding 3-methyladenine DNA glycosylases I and II. Mol. Gen. Genet., 197(1): 368-378. Cullinane, C., & A. van Rosmalen. 1994. Does adriamycin induce interstrand cross-links in DNA? Biochemistry, 33(15):4632-4638. Cullinane, C., S. M. Cutts, C. Panousis & D. R. Phillips. 2000. Interstrand cross-linking by adriamycin in nuclear and mitochondrial DNA of MCF-7 cells. Nucleic Acids Res., 28(4): 1019-1025. Davies, D. B., R. J. Eaton, S. F. Baranovsky & A. N. Veselkov. 2000. NMR investigation of the complexation of daunomycin with deoxytetranucleotides of different base sequence in aqueous solution. J. Biomol. Struct. Dyn., 1 7(5): 887-90 1 . Eaton, R. J., D. A. Baranovskii, D. A. Veselkov, S. G. Osetrov, P. A. Bolotin, L. N. Dymant, V. I. Pakhomov, D. V. Davis & A. N. Veselkov. 2000. Study of the complex formation of daunomycin with deoxytetranucleotides with bases of differing sequence in an aqueous solution by 1H-NMR spectroscopy. Biofizika, 45(4): 586-599. El-Mahdy, S. & O. Othman. 2000. Cytogenetic effect of the anti-cancer drug epirubicin on Chinese hamster cell in vitro. Mutat. Res., 468(2): 109-1 15. Engelward, B. P., M. S. Boosalis, B. J. Chen, Z. Deng, M. J. Siciliano & L. D. Samson. 1993. Cloning and characterization of a mouse 3-methyladenine/7-methylguanine/3-methylguanine DNA glycosylase cDNA whose gene maps to chromosome 11. Carcinogenesis, 14(2): 175-181. 258 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 3, 2002 Engelward, B. P. , A. Dreslin, J. Christensen, D. Huszar, C. Kurahara & L. Samson. 1996. Repair-deficient 3-methyladenine DNA glycosylase homozygous mutant mouse cells have increased sensitivity to alkylation-induced chromosome damage and cell killing. EMBO J., 15:945-952. Evensen, G. & E. Seeberg. 1982. Adaptation to alkylation resistance involves the induction of a DNA glycosylase. Nature, 296(1 ):775-779. Felix, C. A. 2001. Leukemias related to treatment with DNA topisomerase II inhibitors. Med. Pediatr. Oncol., 36(5):525-535. Fenick, D. J., D. J. Taatjes & T. H. Koch. 1997. Doxoform and daunoform: Anthracycline-formaldehyde conjugates toxic to resistant tumor cells. J. Med. Chem., 40(1 6): 2452-2461. Fields, S. M. & J. M. Koeller. 1991. Idarubicin: a second-generation anthracycline. DICP, 25(5) :505-5 17. Ganzina, F. 1983. 4’-epi-doxorubicin, a new analogue of doxorubicin: a preliminary overview of preclinical and clinical data. Cancer Treat. Rev., 10(1): 1-22. Gee, P., D. Maron & B. N. Ames. 1994. Detection and classification of mutagens: A set of base-specific Salmonella tester strains. Proc. Natl. Acad. Sci. USA, 91(24): 11606-1 1610. Gee, P., C. H. Sommers, A. S. Melick, X. M. Gidrol, M. D. Todd, R. B. Burris, M. E. Nelson & E. Zeiger. 1998. Comparison of responses of base-specific Salmonella tester strains with the traditional strains for identifying mutagens: the results of a validation study . Mutation Res . , 4 1 2(2) : 1 1 5- 1 30 . Gewirtz, D. A. 1999. A critical evaluation of the mechanisms of action proposed for the antitumor effects of the anthracycline antibiotics adriamycin and daunorubicin. Biochem. Pharmacol., 57(7) : 727-74 1 . Guano, F., P. Pourquier, S. Tinelli, M. Binaschi, M. Bigioni, F. Animati, S. Manzini, F. Zunino, G. Kohlhagen, Y. Pommier & G. Capranico. 1999. Topoisomerase poisoning activity of novel disaccharide anthracyclines. Mol. Pharmacol., 56(l):77-84. Hande, K. R. 1998. Clinical applications of anticancer drugs targeted to topoisomerase II. Biochim. Biophys. Acta., 1400(1-3): 173-184. Hollingshead, L. M. & D. Faulds. 1991. Idarubicin. A review of its pharmacodynamic and pharmacokinetic properties, and therapeutic potential in the chemotherapy of cancer. Drugs, 42(4): 690-71 9. Hortobagyi, G. N. 1997. Anthracyclines in the treatment of cancer: An overview. Drugs, 54(S4): 1-7. Hortobagyi, G. N. 2000. Developments in chemotherapy of breast cancer. Cancer, 88(12 SUPPL):3073-3079. Kaldor, J. M., N. E. Day & S. Shiboski. 1986. Epidemiological studies of anticancer drug carcinogenicity. I ARC Sci. Publ., 78(1): 189-201. Kaldor, J. M., N. E. Day, P. Band, N. W. Choi, E. A. Clarke, M. P. Coleman, M. Hakama, M. Koch, F. Langmark, F. E. Neal, F. Pettersson, V. Pompe-Kim, P. Prior & H. H. Storm. 1987. Second malignancies following testicular cancer, ovarian cancer, and Hodgkin’s disease: an international collaborative study among cancer registries. Int. J. Cancer, 39(1): 571-585. Keefe, D. L. 2001. Anthracycline-induced cardiomyopathy. Semin. Oncol., 28(4 Suppl 12):2-7. Koch, W. H. , E. N. Henrikson, E. Kupchella & T. A. Cebula. 1994. Salmonella typhimurium strain TA100 differentiates several classes of carcinogens and mutagens by base substitution specificity. Carcinogenesis, 15(1 ) :79-88 . BRUMFIELD & MACKAY 259 Lee, S. T., L H. Jang, H. C. Suh, J. S. Hahn, Y. W. Ko & Y. H. Min. 2001. Idarubicin, cytarabine, and topotecan in patients with refractory or relapsed acute myelogenous leukemia and high-risk myelodysplastic syndrome. Am. J. Hematol., 68(4):237-245. Liu, L. F. 1989. DNA topoisomerase poisons as antitumor drugs. Annu. Rev. Biochem., 58:351-375. Mackay, W. J., L. A. Phelps, A. A. Cauchi & L. T. Weaver. 2000. The mutagenic effects of anthracyclines in the bacterium Salmonella typhimurium: Induction of transition mutations with daunomycin. Texas J. Sci., 52(3):223-229. Mackay, W. J. & L. A. Phelps. 2001. The induction of GC to AT transition mutations with adriamycin in the bacterium Salmonella typhimurium. Texas J. Sci., 53(3):239-246. Maron, D. M & B. N. Ames. 1983. Revised methods for the Salmonella mutagenicity test. Mutation Res., 1 13(3-4): 173-215. Marzin, D., C. Jasmin, R. Maral & G. Mathe. 1983. Mutagenicity of eight anthracycline derivatives in five strains of Salmonella typhimurium. Eur. J. Cancer Clin. Oncol., 19(5): 64 1-647. Mortelmans, K. & E. Zeiger. 2000. The Ames Salmonella/ microsome mutagenicity assay. Mutat. Res., 455(1 -2): 29-60. Ogura, M. 2001. Adriamycin (doxorubicin). Gan To Kagaku Ryoho, 28(10): 1331-1338. Olinski, R., P. Jaruga, M. Foksinski, K. Bialkowski & J. Tujakowski. 1997. Epirubicin-induced oxidative DNA damage and evidence for its repair in lymphocytes of cancer patients who are undergoing chemotherapy. Mol. Pharmacol., 52(5): 882-885. Olinski, R., P. Jaruga & T. H. Zastawny. 1998. Oxidative DNA base modifications as factors in carcinogenesis. Acta. Biochim. Pol., 45(2) :56 1-572. Platel, D., P. Pouna, S. Bonoron-Adyele & J. Robert. 1999. Comparative cardiotoxicity of idarubicin and doxorubicin using the isolated perfused rat heart model. Anticancer Drugs, 10(7):671-676. Qu, X., C. Wan, H. C. Becker, D. Zhong & A. H. Zewail. 2001. The anticancer drug-DNA complex: Femtosecond primary dynamics for anthracycline antibiotics function. Proc. Natl. Acad. Sci. USA, 98(25)14212-14217. Razis, E. D. & G. Fountzilas. 2001. Paclitaxel: epirubicin in metastatic breast cancer— a review. Ann. Oncol., 12(5):593-598. Seeberg, E., L. Luna, I. Morland, L. Eide, B. Johnsen, E. Larsen, I. Alseth, F. Dantzer, K. Baynton, R. Aamodt, K. I. Kristiansen, T. Rognes, A. Klungland & M. Bjoras. 2000. Base removers and strand scissors: Different strategies employed in base excision and strand incision at modified residues in DNA. Cold Spring Harbor Symp. Quant. Biol., 65(1): 135-142. Sinha, B. K. & P. M. Politi. 1990. Anthracyclines. Cancer Chemother. Biol. Response Modif., 11(1) :45-57. Skladanowski , A. & J. Konopa. 1994. Interstrand DNA crosslinking induced by anthracyclines in tumor cells. Biochem. Pharmacol., 47(12):2269-2278. Swendlow, A. J., A. J. Douglas & C. V. Hudson. 1992. Risk of second primary cancers after Hodgkin’s disease by type of treatment analysis of 2846 patients in the British National Lymphoma Investigation: relationships to host factors, histology, and stage of Hodgkin’s disease, and splenectomy. Br. Med. J., 304(1): 1137-1143. Taatjes, D. J., G. Gaudiano, K. Resing & T. H. Koch. 1997. Redox pathway leading to the alkylation of DNA by the anthracycline, antitumor drugs adriamycin and daunomycin. J. Med. Chem. 40(8): 1276-1286. Tominaga, S. 1986. Epidemiologic methods of estimation of a secondary cancer associated with cancer treatment. Gan To Kagaku Ryoho, 13(4 Pt 2): 1528-1533. 260 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 3, 2002 Trudeau, M. & O. Pagani. 2001. Epirubicin in combination with the taxanes. Semin. Oncol., 28(4 Suppl 12):41-50. Vakeva, L., E. Pukkala & A. Ranki. 2000. Increased risk of secondary cancers in patients with primary cutaneous T cell lymphoma. J. Invest. Dermatol., 1 15(1 ) :62-65 . Weiss, R. B. 1992. The anthracyclines: will we ever find a better doxorubicin? Semin. Oncol., 19(6): 670-686. Zunino, F., G. Pratesi & P. Perogo. 2001. Role of the sugar moiety in the pharmalogical activity of anthracyclines: development of a novel series of disaccharide analogs. Biochem. Pharmacol., 61(8):933-938. WJM at: bamackay@erie.net TEXAS J. SCI. 54(3):26 1-268 AUGUST, 2002 THE EFFECTS OF INCUBATION TEMPERATURE ON LOCOMOTOR ACTIVITY IN JUVENILE HOGNA CAROLINENSIS (ARANEAE: LYCOSIDAE) Fred Punzo and Marie Chapla Department of Biology, University of Tampa Tampa, Florida 33606 Abstract.— The effects of incubation temperature on locomotor activity in juveniles (third instar) of the wolf spider, Hogna carolinensis (Hentz) is reported. Egg sacs were incubated under low (24 - 26°C), medium (29 - 31 °C), and high (34 - 36°C) temperatures in an envi¬ ronmental chamber. Locomotor activities (distance, speed and frequency) were measured for 1 h at 30°C using digital activity recording equipment. The total distance travelled, frequency of activity, and minimum and maximum speeds were significantly different when compared to incubation temperature. In general, those spiders that hatched from egg sacs incubated at 29 - 31 °C travelled further and faster, and moved more frequently when compared to those incubated at lower or higher temperatures. This is the first reported effect of egg incubation temperature on locomotor activity of juvenile spiders. Although it is well known that ambient temperature affects many morphological, physiological and developmental processes in spiders (Pulz 1987; Punzo 1991; 2000), there is little information on the effects of incubation temperature on the behavior of juveniles or adults (Punzo & Henderson 1999). Previous research has shown that the thermal and hydric environment surrounding the developing embryos of ectotherms may have a profound influence on various phenotypic traits later in life (May 1985; Scheiner 1993). The wolf spider, Hogna carolinensis (Hentz) is the largest North American lycosid, and is widely distributed throughout the United States, southern Canada and northern Mexico (Dondale & Redner 1990). It is an ambush predator that is primarily nocturnal but may be seen occasionally during the day wandering over the surface or hiding beneath rocks (Gertsch 1979). Adult females can range in size (body length) from 22 - 35 mm, and males from 18-20 mm. In open, xeric habitats with sparse plant cover, these spiders frequently construct a burrow up to 30 cm in depth, with the entrance either unmodified or provided with a turret comprised of grasses, sticks or small stones depending on the location (Gertsch 1979). In the Chihuahuan and Sonoran deserts of the southwestern United States, adult females of H. carolinensis can be found carrying their egg 262 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 3, 2002 sacs (attached to the posterior region of the abdomen) from early June through mid-September (Shook 1978; Punzo 2000). After emergence, the spiderlings climb onto the back of their mother and remain there for 4-6 days before dropping to the ground and dispersing. This study investigated the effect of incubation temperature on locomotor activity in juveniles of H. carolinensis . To the author’s knowledge, this represents the first study to assess the effect of this parameter on the subsequent behavior of spiders. Materials and Methods General protocols and subjects.— During June and July, 1997, 54 adult females of Hogna carolinensis were collected from Madera Canyon in Big Bend Ranch State Park (Brewster County, Texas), which is located in the northern region of the Chihuahuan Desert. A detailed description of this site can be found in Punzo & Henderson (1999). Spiders were transported back to the laboratory and housed indi¬ vidually in plastic shoe boxes provided with a 1 : 1 v/v mixture of sand /peat moss substrate. They were maintained at 30° ± 0.5 °C, 65-70% relative humidity, and a 12: 12 h light:dark cycle in a Percival Model 85 environmental chamber (Boone, Iowa). Spiders were provided with water ad libitum and fed three times per week on a mixed diet of mealworms (Tenebrio molitor) , cockroaches (Periplaneta americana) and crickets (Acheta domestica ) . One adult of each prey species was fed on each of the three feeding days. Nine of the 54 females collected in the field produced fertilized egg sacs. These nine females with their egg sacs were randomly assigned to one of three incubation temperature groups: low (24 - 26 °C), medium (29 - 31°C), and high (34 - 35 °C). These temperature intervals were chosen on the basis of previous studies (Moeur & Eriksen 1972; Punzo & Jellies 1983) on the critical thermal minima and maxima and preferred temperatures of H. carolinensis , as well as on the observations that have been made over several years of captive breeding. Since previous work has shown that, depending on the species, there may or may not be any differences in the effects of diel cycling vs. constant incubation tempera¬ tures on various phenotypic traits in terrestrial arthropods (May 1985), and also on the basis of laboratory observations of H. carolinenesis , constant temperatures were chosen for this study. PUNZO & CHAPLA 263 Upon hatching, the first-instar nymphs (using the classification of Vachon 1957) climbed onto the back of their mother. Juveniles were collected immediately after leaving their maternal parent (4-5 days), housed separately in 120 mL plastic containers with 1.5 cm of moist peat moss substrate, and maintained at the same temperature, humidity and photoperiod regime as the adult females. The third-instar juveniles used in these experiments were between 9-10 days old ( X ± SE carapace width = 3.2 ± 0.4 mm), and were fed a mixed diet of apterous fruit flies ( Drosophila melanogaster ) and small cricket nymphs. All spiders were deprived of food for 24 h prior to testing. Measurement of locomotor performance Ten juveniles that had hatched from each of the nine egg sacs were randomly chosen for each incubation temperature treatment ( N = 30 for low, medium and high treatment groups) and tested them for locomotor performance at 30 °C. Based on our observations on the diel periodicity of H. carolinensis in the field, all locomotor activity experiments were conducted between 2100 and 0000 hours CST. These experiments took place in an envi¬ ronmental chamber at 20 and 30°C and 70% relative humidity. The chamber was illuminated with an infrared lamp since most spiders are not sensitive to light at this wavelength (Foelix 1996) and should behave as they would under conditions of darkness. Individual spiders were placed in a plastic arena (12 by 12 by 5 cm) with white strips of Whatman filter paper (Carolina Biological Supply, Burlington, NC) covering the floor. Three arenas were used simultane¬ ously, one from each incubation temperature treatment. The position of the treatment groups (right, middle, left) was alternated between trials to control for possible effects due to position bias (Walker et al. 1999). Each spider was given 30 min to acclimate to the arena prior to testing. The filter paper was changed and the arenas washed with a dampened soapy sponge between trials to eliminate any intraspecific odor cues. Locomotor activity was monitored for 1 h in each trial. Measure¬ ments were recorded for distance travelled (cm) and speed (cm/s) with a Sony infrared camera and an automated video digital-data collection system (Videomex-V, Columbus Instruments, Columbus, Ohio). This system converts video images to a background field of black and white pixels allowing one to continuously track a moving object (Boiteau 1997). The Videomax was set to track spiders in body length-incre¬ ments and to monitor distance travelled and speed at 2-min intervals for 264 600 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 3, 2002 Low Med High Incubation temperature (deg C) Figure 1 . Effect of incubation temperature on the mean ± SE distance travelled by juveniles of Hogna carolinensis over a one hour observation period at a test temperature of 30°C. Incubation temperatures: low (24 - 26°C), medium (29 - 31 °C), and high (34 - 35°C). 1 h. This is an extremely sensitive system capable of quantifying slight changes in the position of small animals (Boiteau 1997). The total distance travelled for each spider over the 1-h period was determined by summing the distances travelled for all 2-min intervals (N = 30). Data from all sampling periods were used to calculate average speed. Average speed was recorded from data on the mean of all speed measurements for all 2-min intervals during which spiders exhibited movement. The fastest speed exhibited for any of the 30 two-min inter¬ vals during each 1-h observation period was considered the maximum speed. Statistical analysis. — All statistical procedures followed those described by Sokal & Rohlf (1995). Because multiple measurements were taken on individual spiders over time, it was possible to use repeated measures ANOVA (between group factor is treatment tempera¬ ture, and within group factor is time). Bartlett’s test for homoscedasti- city showed that all behavioral data exhibited equality of variances (normality). Unplanned comparisions of differences between means were analyzed using Tukey’s least significant difference procedure (a = 0.05). PUNZO & CHAPLA 265 Low Med High Incubation temperature (deg C) Figure 2. Effect of incubation temperature on the mean ± SE frequency of bouts of activity of Hogna carolinensis at a test temperature of 30 °C. Incubation temperatures are the same as those listed in Fig. 1. Results and Discussion The total distance travelled if = 5.24, P < 0.0311; Fig. 1), fre¬ quency of bouts of activity (f2 = 4.89, P < 0.0284; Fig. 2) and minimum if = 11.6, P < 0.01; Fig. 3) and maximum if = 13.3, P < 0.01; Fig. 3) speed of locomotion, were significantly influenced by incubation temperature. In general, those juvenile spiders that hatched from egg sacs incubated at 29 - 31° travelled further and faster, and moved more frequently when compared to those incubated at 24 - 26° and 34- 36°. For each behavior, all pairwise comparisons between the medium incubation temperature vs. low and high temperature regimes were significant (P < 0.05). These data suggest, for the first time, that incubation temperature can exert a significant influence on locomotor activity of juvenile spiders. In this sense, the selection of an oviposition site, which can determine the thermal and hydric conditions to which the developing embryos are exposed, appears to be as important to the fitness of spiders as it is to reptiles. Unlike reptiles, female wolf spiders carry their eggs sacs and can therefore move them from less favorable to more favorable thermal 266 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 3, 2002 8 Low Med High Incubation temperature (deg C) Figure 3. Effect of incubation temperature on the mean ± SE minimum (unshaded bars) and maximum (shaded bars) speed of travel for juveniles of Hogna carolinensis at a test temperature of 30 °C. Incubation temperatures are the same as those listed in Fig. 1. conditions. Since incubation temperature can significantly affect subsequent postembryonic behavior, it may have acted as a strong selective agent in the evolution of thermoregulatory behavior in lycosids. After leaving their mothers, juvenile wolf spiders must disperse and find suitable prey and shelter. Owing to their small size, they are particularly vulnerable to predation and desiccation at this stage (Punzo & Jellies 1983; Pulz 1987; Punzo 2000). Spiders that move more frequently, and travel further and faster, should have an advantage in these respects. Enhanced locomotor activity may increase their con¬ sumption of prey resulting in faster growth rates and larger body size. As a result, attaining sexual maturity more rapidly may result in the production of multiple egg sacs or the insemination of more females (Morse 1994), and larger body size is often associated with increased fecundity (Marshall & Gittleman 1994; Punzo & Henderson 1999). In addition, faster maximum running speeds should enhance their ability to escape from cursorial predators. Future studies on spiders should investigate the effects of incubation temperature on reproductive success and morphological traits of off- PUNZO & CHAPLA 267 spring, as well as on additional behaviors including emergence, disper¬ sal, escape behavior, agonistic interactions, the ability to detect and capture prey, courtship, thermoregulation and burrow or web construc¬ tion. ACKNOWLEDGMENTS We are grateful to R. Shaw, L. Costa, C. Bradford and anonymous reviewers for helpful comments on an earlier draft of the manuscript, B. Garman for consultation on statistical procedures, H. Cummings for use of his Videomex-V digital data analysis system, and T. Punzo for assistance in maintaining animals in captivity. This research adhered to the Guidelines for the Use of Animals in Research of the Animal Behavior Society as well as approved protocols outlined by the University of Tampa (UT). A Faculty Development Grant to F. Punzo from UT provided financial support for much of this work. Literature Cited Boiteau, G. 1997. Behavioral effects of imidacloprid, a new nicotinyl insecticide, on the potato aphid. Can. Entomol., 1 29( 1 ) : 24 1 -249 . Dondale, C. D. & J. H. Redner. 1990. The insects and arachnids of Canada. Part 17. The wolf spiders, Nursery-web spiders, and lynx spiders of Canada and Alaska. Can. Dept. Agri. Publ. 1856:1-383. Foelix, R. F. 1996. The Biology of Spiders. 2nd edn. Oxford University Press, Oxford, 314 pp. Gertsch, W. J. 1979. American Spiders. 2nd edn. Van Nostrand, New York, 406 pp. Marshall, S. & J. L. Gittleman. 1994. Clutch size in spiders: is more better? Funct. Ecol., 8(1): 1 18-124. May, M. L. 1985. Thermoregulation. Pp. 507-552, in Comprehensive Insect Physiology, Biochemistry, and Pharmacology. Vol. 4. (G. A. Kerkut & L. Gilbert, eds.), Pergamon Press, New York, 787 pp. Moeur, J. E. & D. H. Eriksen. 1972. Metabolic responses to temperatures of a desert spider, Lycosa carolinensis (Lycosidae). Physiol. Zool., 45(2) : 290-30 1 . Morse, D. H. 1994. Numbers of broods produced by the crab spider Misumena vatia (Araneae, Thomisidae). J. Arachnol., 22(1): 195-199. Pulz, R. 1987. Thermal and water relations. Pp. 26-55, in Ecophysiology of Spiders. (W. Nentwig, ed.), Springer, Heidelberg, 314 pp. Punzo, F. 1991. Intraspecific variation to thermal stress in the tarantula Dugesiella echina Chamberlin (Orthognatha, Theraphosidae). Bull. Br. Arachnol. Soc., 8(2):277-283. Punzo, F. 2000. Desert Arthropods: Life History Variations. Springer, Heidelberg, 308 pp. Punzo, F. & L. Henderson. 1999. Aspects of the natural history and behavioral ecology of the tarantula spider Apho nop elma he ntzi (Girard 1854) (Araneae, Theraphosidae). Bull. Br. Arachnol. Soc., 1 1(1): 121-128. Punzo, F. & J. Jellies. 1983. Comparative water relations of araneid and lycosid spiderlings. Comp. Biochem. Physiol., 74A(7): 98 1-985. Scheiner, S. M. 1993. Genetics and the evolution of phenotypic plasticity. Annu. Rev. Ecol. System., 24:35-68. 268 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 3, 2002 Shook, R. S. 1978. Ecology of the wolf spider, Lycosa carolinensis Walckenaer (Araneae, Lycosidae) in a desert community. J. Arachnol., 6(l):53-64. Sokal, R. R. & F. J. Rohlf, F. J. 1995. Biometry. 3rd edn. W. H. Freeman, San Francisco, 818 pp. Walker, S. E., S. D. Marshall, A. L. Rypstra & D. H. Taylor. 1999. The effects of hunger on locomotory behaviour in two species of wolf spider (Araneae, Lycosidae). Anim. Behav., 58(3):515-520. FP at: fpunzo@ut.edu TEXAS J. SCI. 54(3):269-276 AUGUST, 2002 DISTRIBUTIONAL RECORDS OF MAMMALS FROM THE PERMIAN BASIN, TEXAS Joel G. Brant and Clyde Jones Department of Biological Sciences and the Museum of Texas Tech University Lubbock, Texas 79409 Abstract.— County records for mammals in the Permian Basin are reported based on fieldwork and museum surveys. Twenty-six records representing 20 species are reported from Crane, Ector, Loving, Ward and Winkler counties. The Permian Basin is composed of five Texas counties (Crane, Ector, Loving, Ward and Winkler) located east of the Pecos River. This area is an ecotonal zone at the junction of three major ecogeographic regions of Texas: the Edwards Plateau, the Llano Estacado and the Trans-Pecos (Blair 1950). The region is a roughly wedge shaped basin framed by the higher elevations of the Edwards Plateau and the Llano Estacado and slopes to the Pecos River. Although currently recognized as part of the Big Bend geographic region by the Texas Parks and Wildlife Department (Holt et al. 2000), the Permian Basin was not included in the extensive surveys of Trans- Pecos mammals (Schmidly 1977). Goetze (1998) reported on mammals from parts of Crane and Ector counties as part of his work on the mammals of the Edwards Plateau. Information on mammals from portions of Winkler and Ector counties was included in the surveys of the Llano Estacado (Choate 1997). The majority of the Permian Basin, however, has been overlooked in ecogeographic surveys of mammals. Davis & Schmidly (1994) predicted that the Permian Basin should have approximately 55-60 mammal species. Currently only 41 species have been reported from this area. Brant & Lee (1999) speculated that the reason for the paucity of information on the mammalian fauna is due to the remoteness of the area. In addition, most of the species currently reported for this area are the result of single night collecting efforts while traveling to a more distant location instead of a systematic survey of the region. Recent fieldwork conducted in the Permian Basin and surveys of specimens in museums have resulted in 26 county records 270 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 3, 2002 unreported previously for the region. Methods Fieldwork was conducted from October 2000 to August 2001 in Monahans Sandhills State Park and the surrounding areas of Ward and Winkler counties in conjunction with a mammal survey of the state park. Standard techniques were utilized to collect small and medium-sized mammals (Jones et al. 1996). Voucher specimens were prepared as standard museum study skins with skulls and frozen tissue. All materials were deposited in the Collection of Recent Mammals in the Natural Science Research Laboratory of the Museum of Texas Tech University. In addition, a survey of systematic collections was conducted for records of mammals from the Permian Basin. Specimen records were obtained from the following systematic collections: Collection of Recent Mammals, Midwestern State University (MWSU), Strecker Museum, Baylor University (SM), Vertebrate Collection, Sul Ross State Univer¬ sity (SRSU), Museum of Texas Tech University (TTU), National Museum of Natural History (USNM) and Centennial Museum, Univer¬ sity of Texas at El Paso (UTEP). Species Accounts The following accounts treat 20 species representing 26 county records for the Permian Basin. The arrangement of taxa and taxonomic nomenclature follows that of Manning & Jones (1998). Didelphis virginiana (Virginia opossum).— One female specimen (SRSU 1292) of this relatively uncommon species was collected in December 1968 from the Texas-New Mexico Tank Farm Bridge in Crane County. This represents the second record of this species from the Permian Basin with the only other voucher from Ward County (Davis & Schmidly 1994). Sylvilagus audubonii (desert cottontail).— A male desert cottontail (TTU 69433) was collected on 25 March 1995 from 12 miles north and five miles west of the city of Crane in Crane County. This species is not unexpected for the Permian Basin with records in every other county BRANT & JONES 271 (Davis & Schmidly 1994; Brant & Lee 1999). Ammospermophilus interpres (Texas antelope ground squirrel).— G. Donald of the U. S. Biological Survey collected a male specimen (USNM 118732) on 13 September 1902 from the Grand Falls region of the Castle Mountains in southern Ward County. This species is characteristic of xeric regions and reaches its northeastern distributional limit on the Permian Basin and Edwards Plateau with records from Crane and Reagan counties (Davis & Schmidly 1994). Spermophilus spilosoma (spotted ground squirrel).— A female specimen (SM 850) was collected on 8 April 1966 from 12 miles north of Mentone in Loving County. The locality of this specimen is well within the distributional range of the species with records from three counties in the Permian Basin (Davis & Schmidly 1994). Spermophilus variegatus (rock squirrel).— M. Cary of the U. S. Biological Survey collected a female rock squirrel (USNM 118601) on 12 September 1902 from the Castle Mountains of southern Ward County. This specimen represents the first record of S. variegatus in the Permian Basin. Cynomys ludovicianus (black-tailed prairie dog).— A single female specimen (UTEP 3545) was collected on 24 June 1972 from 3.75 miles north and 1.5 miles east of the county building in Crane County. The record of this specimen is well within the distributional limit of the species and records are known from two other Permian Basin counties (Davis & Schmidly 1994). Thomomys bottae (Botta’s pocket gopher).— M. Cary of the U. S. Biological Survey collected a female specimen (USNM 118582) on 12 September 1902 from the Castle Mountains in southern Ward County. This species is only known from the Trans-Pecos, Edwards Plateau and the southern portion of the Permian Basin (Davis & Schmidly 1994). Chaetodipus intermedius (rock pocket mouse).— On 27 July 1987, a male C. intermedius (TTU 47218) was collected from four miles south and two miles east of the city of Crane in Crane County. Five speci¬ mens of the rock pocket mouse (TTU 47221-47225) were collected from Ward County on 25 July 1987 from two miles west of Barstow. These 272 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 3, 2002 six records from Crane and Ward counties are at the eastern limit of the geographical distribution of this species, which only occurs east of the Pecos River in the Permian Basin (Davis & Schmidly 1994). Reithrodontomys megalotis (western harvest mouse). —A male western harvest mouse (TTU 45589) was collected on 19 April 1987 from two miles north and five miles east of Mentone in Loving County. Two additional specimens (TTU 45590-45591) were collected on 20 April 1987 from two miles west of Mentone. Two specimens of R. megalotis have been collected from Ward County. On 13 March 1969, a male (UTEP 4680) was collected from along the Pecos River near US High¬ way 80. Another specimen (TTU 82484) was collected from the Pump Jack Picnic Area (13R 706632E, 3502752N) of Monahans Sandhills State Park on 18 October 2000. These five records for Loving and Ward counties provide further insight into the range of this relatively uncommon species, which reaches its eastern distributional limits on the Edwards Plateau and the Llano Estacado (Davis & Schmidly 1994). Reithrodontomys montanus (plains harvest mouse). — On 8 March 1987, a female plains harvest mouse (MWSU 14511) was collected from nine miles west of Monahans in Ward County. This species is relatively rare in the Trans-Pecos (Schmidly 1977) and the Edwards Plateau (Goetze 1998). This species has been recorded from Ector and Winkler counties in the Permian Basin (Davis & Schmidly 1994). Peromyscus leucopus (white- footed mouse).— Several specimens of the white- footed mouse were collected from Loving and Ward counties. Four females and seven males were collected on 19-20 March 1987 from 1 mile west of Mentone (TTU 45571); two miles north and five miles west of Mentone (TTU 45572-45576); and two miles west of Mentone (TTU 45577-45581) in Loving County. A male (TTU 47267) was collected from three miles east of Mentone on 26 August 1987. Another specimen (TTU 69517) was collected eight miles north and eight miles east of Orla on 7 August 1994. Twenty-nine specimens (16 females and 13 males) have been collected from Ward County. A male (TTU 44210) was collected on 1 March 1986 from eight miles west- southwest of Monahans. Four specimens (TTU 53983-53986) were collected from nine miles west of Monahans on 7-8 March 1987. On 25 July 1987, three specimens (TTU 47276-47278) were collected from two miles west of Barstow. Eight P. leucopus (TTU 69518-69525) were BRANT & JONES 273 collected on 5 November 1994 from Monahans Sandhills State Park. Three specimens (TTU 69526-69528) were collected from two miles north of Monahans on 16 March 1995. Ten specimens (TTU 82481- 82482, 82485-82492) have been collected from Monahans Sandhills State Park from October 2000 to July 2001. Peromyscus maniculatus (deer mouse).— -Three deer mice (TTU 45582-45584) were collected on 19 March 1987 from one mile west of Mentone in Loving County. This relatively common species is distri¬ buted statewide with records for all of the surrounding counties (Davis & Schmidly 1994). Onychomys leucogaster (northern grasshopper mouse) . —Two northern grasshopper mice have been collected from Loving County. A female (TTU 45569) was collected on 19 March 1987 from two miles north and five miles east of Mentone. A male (TTU 58417) was collected on 13 July 1990 from two miles northeast of Mentone. This species is distributed throughout western Texas (Davis & Schmidly 1994) but is relatively rare in the Trans-Pecos (Schmidly 1977). Sigmodon hispidus (hispid cotton rat).— Several hispid cotton rats have been collected from Loving and Ward counties. Ten S. hispidus were collected from Loving County in 1987 from three localities: three females (TTU 45594-45596) collected on 19 March 1987 from one mile west of Mentone; five females and one male (TTU 45597-45602) collected on 20 March 1987 from two miles west of Mentone; and a female (TTU 47325) collected on 26 August 1987 from three miles east of Mentone. Four S. hispidus have been collected from Ward County. On 25 July 1987, a female (TTU 47334) was collected from two miles west of Barstow. A male and a female (TTU 69537-69538) were collected from Monahans Sandhills State Park on 1 1 September 1994. Another male (TTU 82483) was collected from the Pump Jack Picnic Area (13R 706632E, 3502752N) of Monahans Sandhills State Park on 18 October 2000. Urocyon cinereoargenteus (gray fox).— On 5 January 1969, a male gray fox (SRSU 1291) was collected from the Texas-New Mexico Tank Farm Bridge in Crane County. Another specimen (UTEP 2862) was collected from Crane County on 24 June 1972. This species is distri¬ buted throughout Texas, but has only been recorded from one other 274 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 3, 2002 county in the Permian Basin (Davis & Schmidly 1994). Procyon lotor (raccoon).— One specimen of the raccoon (TTU 82493) was collected on 10 July 2001 from Pump Jack Picnic Area (13R 706632E, 3502752N) of Monahans Sandhills State Park in Ward County. This relatively common species has only been recorded from one other Permian Basin county (Davis & Schmidly 1994) and is usually associated with watercourses in the Trans-Pecos (Schmidly 1977). Mustela frenata (long-tailed weasel).— On 24 June 1972, a single male long-tailed weasel (UTEP 3618) was collected from 3.75 miles north and 1 .5 miles east of the county building in Crane County. The records for this rare species in Texas are scattered throughout most of the state (Davis & Schmidly 1994). This is the first record for M. frenata from the Permian Basin. Taxidea taxus (American badger).— The American badger was collected from three counties in the Permian Basin. On 22 October 1987, a female (TTU 49077) was collected from 14 miles south and two miles east of Crane in Crane County. Another female (MWSU 18020) was collected on 28 March 1991 from 17.9 miles west-southwest of Midland in Ector County. In Ward County a female (MWSU 6045) was collected on 10 May 1968 from 35 miles southwest of Monahans. This species ranges throughout most of Texas but until now has not been recorded from the Permian Basin (Davis & Schmidly 1994). Mephitis mephitis (striped skunk).— M. Cary of the U. S. Biological Survey collected a male striped skunk (USNM 1 18618) on 12 September 1902 from the Castle Mountains in southern Ward County. This species is common throughout Texas, but has not been recorded from the Permian Basin until now (Davis & Schmidly 1994). Conclusions The 26 county records reported in this study help to fill in the gap in the understanding of Texas mammals of that geographical area known as the Permian Basin. Only four of the 20 species reported herein have never been reported from the Permian Basin; Mephitis mephitis, Taxidea taxus, Mustela frenata and Spermophilus variegatus. These four species increase the mammalian diversity of the area to 45 species, closer to the BRANT & JONES 275 predicted value of 55-60 (Davis & Schmidly 1994). Undoubtedly more elusive species will be encountered with further research and a systema¬ tic survey of this neglected area. Acknowledgments This study was funded by the Natural Resources Program (David H. Riskind, Director), Texas Parks and Wildlife Department (State Park Scientific Study Permit 55a-00). We appreciated the assistance of Kelly Bryan, Linda Hedges, Glen Korth and the staff of Monahans Sandhills State Park. John Bickham, Wesley Colvin, Charlene Cunningham, Carl Dick, Jerry Harclerode, D. Holbert, Robert Hollander, Stephen Kasper, Sonia MacKinder, Richard Manning, L. McLaughlin, Fransisca Mendez-Harcl erode, Marisol Salazar and J. Stone were associated with the field collection of specimens reported in this study. The following museums and museum personnel provided records of specimens from the Permian Basin and permitted their inclusion in this manuscript: Frederick B. Stangl, Jr., at the Collection of Recent Mammals, Midwestern State University (MWSU); David Lintz at the Strecker Museum, Baylor University (SM); James M. Mueller at the Vertebrate Collection at Sul Ross State University (SRSU); R. Richard Monk at The Museum of Texas Tech University (TTU); Robert Fisher at the United States National Museum of Natural History (USNM); and Arthur H. Harris at the Centennial Museum, University of Texas at El Paso (UTEP). Literature Cited Blair, W. F. 1950. The biotic provinces of Texas. Texas J. Sci., 2(2):93- 1 17. Brant, J. G. & T. E. Lee, Jr. 1999. New records of mammals for Loving County, Texas. Texas J. Sci., 51 (4): 347-350. Choate, L. L. 1997. The mammals of the Llano Estacado. Spec. Pubs. Mus., Texas Tech Univ., 40:1-240. Davis, W. B. & D. J. Schmidly. 1994. The mammals of Texas. Texas Parks and Wildlife Press, Austin, 338 pp. Goetze, J. R. 1998. The mammals of the Edwards Plateau, Texas. Spec. Pubs. Mus., Texas Tech Univ., 41:1-263. Holt, E. A., K. E. Allen, N. C. Parker & R. J. Baker. 2000. Ecotourism and conservation: richness of terrestrial vertebrates across Texas. Occas. Papers Mus., Texas Tech Univ., 201:1-16. Jones, C., W. J. McShea, M. J. Conroy & T. H. Kunz. 1996. Capturing mammals. Pp. 115-155, in Measuring and monitoring biological diversity: standard methods for 276 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 3, 2002 mammals. Wilson, D. E., F. R. Cole, J. D. Nichols, R. Rudran and M. S. Foster, eds. Smithsonian Institution Press, Washington, D. C. 409 pp. Manning, R. W. & C. Jones. 1998. Annotated checklist of Recent land mammals of Texas, 1998. Occas. Papers Mus., Texas Tech Univ., 182:1-20. Schmidly, D. J. 1977. The mammals of Trans-Pecos Texas. Texas A&M University Press: College Station, 225 pp. JGB at: jbrant@ttacs.ttu.edu TEXAS J. SCI. 54(3), AUGUST, 2002 277 GENERAL NOTES NOTEWORTHY RECORDS OF BATS FROM THE TRANS-PECOS REGION OF TEXAS Jana. L. Higginbotham, Robert S. DeBaca, Joel G. Brant and Clyde Jones Department of Biological Sciences and the Museum Texas Tech University, Lubbock, Texas 79409 Recent investigations of bat communities in the Trans- Pecos region of Texas provided noteworthy records for six species ( Myotis calif omicus , My otis volans , Lasiurus borealis , Nyctinomops femorosaccus , Nyctino- mops macrotis and Eumops perotis). All voucher specimens reported herein (skins and skulls), and their associated tissues, are deposited in the Collection of Recent Mammals in the Natural Science Research Laboratory of the Museum of Texas Tech University (TTU). Myotis calif omicus . —The California myotis is an established resident of the Trans- Pecos area and has been documented in El Paso, Culber¬ son, Jeff Davis, Presidio and Brewster counties, yet records of this species from Hudspeth County are lacking (Schmidly 1991; Davis & Schmidly 1994). Two females (TTU 82465, 82467) were obtained with mist nets set across a pool of a desert spring on 18 June 2001 at the Indio Mountains Research Station (University of El Paso), Hudspeth County (UTM 13R 498907E 3407109N, elevation 1280m). The edges of the pool were overgrown with cattail ( Typha sp.) and a rock ridge bordered the northeast side. The surrounding vegetation consisted of thorny desert scrub, including acacia ( Acacia sp.), yucca ( Yucca sp.) and prickly pear ( Opuntia sp.). The discovery of M. califomicus in Hudspeth County completes a contiguous distribution of this species across the western Trans-Pecos region. Associated species of bats encountered at this site on 18 June include Pipistrellus hespems , Antrozous pallidus , Myotis ve lifer and Myotis volans . Myotis volans.— The long-legged myotis is considered relatively rare in Texas, yet it is common in the higher elevations of extreme western Texas. In the Trans-Pecos area, this species has been documented in the mountainous regions of Culberson, Jeff Davis, Presidio and Brewster counties, with an addition record from Knox County in the Rolling 278 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 3, 2002 Plains (Schmidly 1991; Davis & Schmidly 1994). On 18 June 2001, one female M. volans (TTU 82466) was captured at the Indio Mountains Research Station (locality same as above). The elevation of the capture site (1280m) is atypical of that associated with M. volans in west Texas, as it is usually found between 2000 to 3000m (Warner & Czaplewski 1984). Other bats captured in association with this species on 18 June were M. ve lifer, M. calif omicus , P. he sp eras and A. pallidus. Lasiurus borealis.— The eastern red bat displays a spotty distribution in the Trans-Pecos region, where it typically is found in mountainous terrain (Schmidly 1991; Davis & Schmidly 1994). Three pregnant females (TTU 82478, 82479, 82480), each having three embryos with crown rump lengths ranging from 12 to 20mm, were captured in mist nets on 22 May 2001 at Big Bend Ranch State Park (BBRSP), Presidio County (UTM 13R 589266E 3269761N, elevation 1012m). In Presidio County, L. borealis is known from the Chinati Mountains (Schmidly 1991) and the Sierra Vieja (Jones & Bradley 1999). Yancey’s (1997) comprehensive survey of the mammals of BBRSP produced no record of L. borealis , yet he listed it as a probable summer migrant of the surrounding riparian woodland habitats. The capture site was characterized by a dense growth of cottonwoods ( Populus sp.) comprising a riparian gallery forest spanning more than three kilometers in length. The site was spring-fed and water levels along this intermittent stream varied little between visits throughout the spring and summer in 2001. This locality is atypical of Chihuahuan Desert habitat, and it likely plays an important role in providing suitable roosting and foraging grounds for bats. The embryos obtained were approaching full- term developmentally, suggesting the adults were to give birth in the area. Forty-one bats were captured with L. borealis on 22 May, including Mormoops me galop hylla, M. calif omicus , P. hespems , Lasiums cine reus , A. pallidus and Tadarida brasiliensis . Nyctinomops femorosaccus . — The distribution of the pocketed free- tailed bat in Texas is quite restricted, and previously has been documented only from Big Bend National Park (BBNP), Brewster County (Easterla 1968; Schmidly 1991; Higginbotham & Ammerman 2002). One lactating female (TTU 82477) was collected in a mist net over a stock tank on 6 July 2001 at BBRSP, Presidio County (UTM 13R 602303E 3258748N, elevation 1300m) in desert scrub habitat dominated by creosote bush (Larrea tridentata). This record expands the known TEXAS J. SCI. 54(3), AUGUST, 2002 279 distribution of this species in Texas 80km west of the previously known range and represents the first account for this species from Presidio County. Yancey (1997) reported N . femorosaccus as a likely occurrence in BBRSP, yet he did not encounter any during his extensive survey of the mammals there. During a recent study in nearby Brewster County at BBNP, N. femorosaccus was encountered only in nets stretched over open water with sizeable surface area (Higginbotham & Ammerman 2002). The ephemeral nature of full stock tanks and other sizeable bodies of water at BBRSP may explain the species’ absence from past studies in the area. The pocketed free-tailed bat roosts in crevices of high rocky canyons in BBNP (Higginbotham & Ammerman 2002). Deep canyons with suitable crevices are numerous at BBRSP, and it is likely that roosts of N. femorosaccus occur within them. Eptesicus juscus , T. brasiliensis , Nyctinomops macrotis and Eutnops perotis were captured on 6 July in association with this species. Nyctinomops macrotis.— Reports of the big free-tailed bat from the Trans-Pecos area are scattered sparsely across both upland and lowland habitats in Brewster, Presidio, Jeff Davis, Reeves, Culberson and El Paso counties (Schmidly 1991; Davis & Schmidly 1994). An adult female (TTU 82469) was captured on 1 July 2001 at an upland site with¬ in coniferous forest at the Davis Mountains Preserve (The Nature Con¬ servancy of Texas), Jeff Davis County (UTM 13R 584126E 3394303N, elevation 1860m). The rare occurrence of this species in the Davis Mountains uplands is evidenced by systematic studies of mammals in the vicinity since 1998, in which only a single specimen of N. macrotis has been encountered at the Davis Mountains Preserve (Bradley et al. 1999). Additionally, one pregnant female N. macrotis (TTU 82468; embryo crown- rump length = 27.5mm) was collected on 29 May 2001 at the same locality in BBRSP as the previously reported L. borealis. Eight lactating females (TTU 82469, 82470, 82471, 82472, 82473, 82474, 82475) were taken on 6 July 2001 in mist nets at BBRSP at the same site as the N. femorosaccus discussed above. These two localities represent vastly different habitats within BBRSP. The former was a riparian gallery forest bordering a narrow, intermittent stream, and the latter, a stock tank (surface area 15 by 20m) within desert scrub habitat. A single specimen of N. macrotis was obtained at BBRSP prior to the work reported herein (Yancey 1997). In Texas, encounters of significant numbers of N. macrotis 280 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 3, 2002 historically have been documented only from BBNP in Brewster County (Borell & Bryant 1942; Easterla 1973; Higginbotham & Ammerman 2002). However, recent investigations of mammals at the Davis Mountains State Park, Jeff Davis County, have produced several captures of N. macrotis at a lowland site. These accounts suggest that N. macrotis is perhaps more common in the Trans-Pecos region than was thought previously. Eumops perotis.— Reports of the western mastiff bat from Texas are few, and all originate from the Trans-Pecos region in proximity to the Rio Grande corridor (Schmidly 1991). This work reports an additional record of one adult male (TTU 82476) collected at BBRSP on 6 July 2001 from the same locality as the aforementioned molossids ( N . femorosaccus and N. macrotis ) . Three reports exist for E. perotis from Presidio County over the past 45 years (Eads et al. 1957; Ohlendorf 1972; Scudday 1976). Scudday (1976) mentions E. perotis as fairly common at Arroyo Segundo (BBRSP), a site also netted by Yancey (1997). This same locality was surveyed several times during summer 2001, yet no western mastiff bats were encountered. Maneuvering ability is apparently compromised in the western mastiff bat due to its wing morphology and large size (Findley et al. 1971), and therefore it probably avoids smaller bodies of water, or enclosed areas. Eumops perotis is known to roosts in crevices within high canyon walls typical of the rocky, rugged terrain of the Big Bend region, therefore, they may be more common in the area than the historical capture data suggests, yet more difficult to capture where large bodies of water are scarce. Eumops perotis , N. femorosaccus and N. macrotis were encountered together at a single locality although extensive sampling was conducted throughout BBRSP during this survey and by Yancey (1997). A distin¬ guishing feature of the site where all three species were captured was the size of the tank (15 by 20 m). When nets spanned the center of the tank, 93 % of mist net captures consisted of the four species of molossids known to inhabit the area, including Tadarida brasiliensis . During sampling periods when mist nets were set only along the perimeter of the tank, molossids were not encountered in the nets, although audible echolocation calls associated with both E. perotis and N. macrotis frequently were heard throughout the evening. This suggests that the larger molossid species inhabiting the Trans-Pecos region of Texas are TEXAS J. SCI. 54(3), AUGUST, 2002 281 obligates of large, open water sources for drinking. Therefore, the rarity with which these bats are encountered in the Trans-Pecos area may be explained in part by the significant fluctuation in surface area of sizeable, reliable water sources in the region. Consequently, the adjacent Rio Grande is an important resource for these species, as it is a stable source of water relative to other, often ephemeral, sources of water in the region. Acknowledgments The fieldwork and collection of specimens were conducted in accordance with scientific permits issued by the Texas Parks and Wildlife Department (SPR-0790- 1 89 and 55A-00). Financial assistance was facilitated by research assistantships granted to J. L. Higginbotham, R. S. DeBaca and J. G. Brant during the summer of 2001 by the Department of Biological Sciences at Texas Tech University. Access to the Indio Mountains Research Station was provided by J. D. Johnson, University of El Paso. Access and logistic support were provided by the Nature Conservancy of Texas and by the personnel of the Big Bend Ranch State Park (L. Armendariz, Superintendent). M. Revelez, M. A. Abbey, A. Matthews and B. Reece assisted in the collection and preparation of specimens. Literature Cited Borrell, A. & M. D. Bryant. 1942. Mammals of the Big Bend area of Texas. Univ. California. Publ. in Zool., 48:1=62. Bradley, R. D., D. S. Carroll, M. L. Clary, C. W. Edwards, I. Tiemann-Boege, M. J. Hamilton, R. A. Van Den Bussche & C. Jones. 1999. Comments on some small mammals from the Big Bend and Trans-Pecos regions of Texas. Occas. Pap. Mus., Texas Tech Univ., 193:1-6. Davis, W. B. & D. J. Schmidly. 1994, The mammals of Texas. Texas Parks and Wildlife Dept., Austin, Texas, x + 338 pp. Eads, R. B., J. E. Grimes & A. Conklin. 1957. Additional Texas bat records. J. Mammal., 38:514. Easterla, D. A. 1968. First records of the pocketed free-tailed bat for Texas. J. Mammal., 49:515-516. Easterla, D. A. 1973. Ecology of the 18 species of Chiroptera at Big Bend National Park, Texas. Northwest Missouri St. Univ. Stud., 34:1-165. Findley, J. S, E. H. Studier & D. E. Wilson. 1971. Morphological properties of bat wings. J. Mammal., 53:429-444. Higginbotham, J. L. & L. K. Ammerman. 2002. Chiropteran community structure and seasonal dynamics in Big Bend National Park, Texas. Spec. Publ. Mus., Texas Tech Univ., 44:1-44. Jones, C. & R. D. Bradley. 1999. Notes on red bats, Lasiurus (Chiroptera: Vespertilioni- dae), of the Davis Mountains and vicinity, Texas. Texas J. Sci., (51)4:341-344. 282 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 3, 2002 Ohlendorf, H. M. 1972. Observations on a colony of Eumops perotis (Molossidae). Southwestern Nat., 17:297-300. Schmidly, D. J. 1991. The bats of Texas. Texas A&M Univ. Press, College Station, xvii + 188 pp. Scudday, J. F. 1976. Vertebrate fauna of the Fresno Canyon area. Pp. 97-110, in Fresno Canyon a natural area survey no 10. Lyndon B. Johnson School of Publ. Aff., The Univ. of Texas at Austin, 144 pp. Warner, R. M. & N. J. Czaplewski. 1984. Myotis volans. Mammalian Species, 224:1-4. Yancey, F. D., II. 1997. The mammals of Big Bend Ranch State Park, Texas. Spec. Publ. Mus., Texas Tech Univ., 39:1-210. JLH at: jana@packrat.musm.tu.edu 9j« S(C S}S 9|C GASTROINTESTINAL HELMINTHS OF GAIGE’S TROPICAL NIGHT LIZARD, LEPIDOPHYMA GAIGEAE (SAURIA: XANTUSIIDAE) FROM HIDALGO, MEXICO Stephen R. Goldberg, Charles R. Bursey and Jose L. Camarillo-Rangel Department of Biology, Whittier College Whittier, California 90608 Pennsylvania State University, Shenango Campus Department of Biology, Sharon, Pennsylvania 16146 and Laboratorio y Coleccion de Herpetologia Escuela Nacional de Estudios Profesionales Iztacala Universidad Nacional Autonoma de Mexico, A.P. 314, Tlalnepantla Estado de Mexico, Mexico Gaige’s tropical night lizard, Lepidophyma gaigeae Mosauer, occurs in limestone crevices within pine-oak woodlands in the Mexican states of Hidalgo and Queretaro (Bezy 1984). To the author’s knowledge, there are no reports of helminths from this lizard. Fifty-five Lepidophyma gaigeae (mean snout- vent length, SVL = 54 mm ± 4.4 SD, range = 42-63 mm; 22 females, SVL = 53 mm ±5.1, range = 42-63 mm; 33 males, SVL = 54 mm ±3.8, range 45-62 mm) were collected at Durango, Hidalgo (20°54’N, 99°14’W) March 1999 to February 2000. Lizards were fixed in 10% formalin, preserved in alcohol and deposited in the herpetology collection of the Escuela Nacional de Estudios Profesionales Iztacala, Universidad Nacional Autonoma de Mexico: ENEPI 6978-6981, 6983-6985, 7003, 7062, 7116- TEXAS J. SCI. 54(3), AUGUST, 2002 283 7120, 7171, 7173, 7176-7180, 7205, 7207-7211, 7267-7269, 7271, 7298- 7300, 7302, 7535, 7540-7543, 7547, 7548, 7572, 7574, 7577-7581, 7608, 7610-7612, 7614, 7617. The abdominal cavity of each lizard was opened and the gastro¬ intestinal tract was excised by cutting across the esophagus and rectum. Each tract was slit longitudinally and examined under a dissecting microscope for helminths. When found, helminths were removed to a drop of undiluted glycerol on a glass slide for initial study. One species of Cestoda, Bitegmen gerrhonoti (Telford 1965) and two species of Nematoda, gravid individuals of Spauligodon giganticus (Read & Amrein 1953) and larvae of Ascaridia sp. were found. Selected specimens were placed in vials of 70% ethanol and deposited in the United States National Parasite Collection (USNPC), Beltsville, Maryland (Table 1). Because there was no statistical difference for SVL between male and female lizards in the sample (Kruskal Wallis = 1.34, 1 df,P> 0.05) and because there was no statistical difference for infection by S. giganticus and Ascaridia sp. (only 1 male lizard infected with B. gerrhonoti ) between male and female lizards (x2 = 1.35, 0.01, 1 df, P > 0.05, respectively), results are presented as a single data set. Prevalence, mean intensity, range and abundance as defined by Bush et al. (1997) are given in Table 1 for each helminth species. Bitegmen gerrhonoti , originally described as Baerietta gerrhonoti by Telford (1965) from the southern alligator lizard, Elgaria multicarinata (= Gerrhonotus multicarinatus webbi) , was reassigned to its current taxonomic position by Jones (1987). This is the second report of B. gerrhonoti in lizards; it is also known from the salamander Ensatina eschscholtzii from southern California (Goldberg et al. 1998). The life cycle of B. gerrhonoti is unknown, but Joyeux (1927) regards the life history of nematotaeniid cestoides to be direct; infection of a new host occurs through ingestion of eggs. Lepidophyma gaigeae is a new host record; Mexico is a new locality record. Spauligodon giganticus is a common intestinal helminth of North American lizards, especially sceloporine lizards; additions to the host list provided by Bursey & Goldberg (1992) are presented in Goldberg et al. (2003). The life cycle of S. giganticus has not been studied; but the life cycles of other oxyurids are direct and infection is by an oral route (Anderson 2000). Goldberg & Bursey (1992) found eggs of S. giganti¬ cus in the digestive tracts of neonatal Sceloporus jarrovii indicating that 284 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 3, 2002 Table 1. Accession number, prevalence, mean intensity, range and abundance for helminth species from Lepidophyma gaigeae from Hidalgo, Mexico. Helminth USNPC tt Prevalence Mean intensity Range Abundance ± SD Cestoda Bitegmen gerrhonoti 91755 2% 3 — 0.05 Nematoda Spauligodon giganticus 91756 76% 8.1 -1- 5.9 1-26 6.20 Ascaridia sp. 91757 24% 8.3 ± 13.6 1-51 1.96 infection takes place soon after birth; young lizards presumably acquire eggs by ingesting substrate. Lepidophyma gaigeae is a new host record and the first xantusiid known to harbor S. giganticus. Larvae of Ascaridia sp. were found encysted on the outer wall of the digestive tract. Moravec & Kaiser (1995) reported encystment by larvae of similar description in species of Eleutherodactylus collected in Dominica and Tobago, West Indies. Species of Ascaridia are common parasites of gallinaceous birds; eggs and larvae can be harbored by earthworms (Anderson 2000). This is the first report of Ascaridia sp. in a lizard species and may represent an accidental infection which could occur in any vermivore. Literature Cited Anderson, R. C. 2000. Nematode parasites of vertebrates: their development and transmission., 2nd edit., CABI, Wallingford, Oxon, United Kingdom, xx + 650 pp. Bezy, R. L. 1984. Systematics of xantusiid lizards of the genus Lepidophyma in northeastern Mexico. Cont. Sci. Nat. Hist. Mus., Los Angeles County, 349:1-16. Bursey, C. R. & S. R. Goldberg. 1992. Monthly prevalences of Spauligodon giganticus (Nematoda, Pharyngodonidae) in naturally infected Yarrow’s spiny lizard Sceloporus jarrovii jarrovii (Iguanidae). Am. Midi. Nat., 127(l):204-207. Bush, A. O., K. D. Lafferty, J. M. Lotz & A. W. Shostak. 1997. Parasitology meets ecology on its own terms: Margolis et al. revisited. J. Parasitol., 83(4):575-583. Goldberg, S. R. & C. R. Bursey. 1992. Prevalence of the nematode Spauligodon giganticus (Oxyurida: Pharyngodonidae) in neonatal Yarrow’s spiny lizards, Sceloporus jarrovii (Sauria: Iguanidae). J. Parasitol., 78(3):539-541 . Goldberg, S. R., C. R. Bursey & J. L. Camarillo-Rangel. 2003. Gastrointestinal helminths of seven species of sceloporine lizards from Mexico. Southwest. Nat., (in press). Goldberg, S. R., C. R. Bursey & H. Cheam. 1998. Composition and structure of helminth communities of the salamanders, Aneides lugubris, Batrachoseps nigriventris , Ensatina eschscholtzii (Plethodontidae), and Taricha torosa (Salamandridae) from California. J. Parasitol., 84(2):248-251 . Jones, M. K. 1987. A taxonomic revision of the Nematotaeniidae Liihe, 1910 (Cestoda: Cyclophyllidea). Syst. Parasitol., 10(3): 165-245. Joyeux, C. 1927. Recherches sur la faune helminthologique Algerienne (cestodes et TEXAS J. SCI. 54(3), AUGUST, 2002 285 trematodes). Arch. Inst. Pasteur Algerie, 5:509-528. Moravec, F., & H. Kaiser. 1995. Helminth parasites from West Indian frogs, with descriptions of two new species. Carib. J. Sci., 31(3-4):252-268. Telford, S. R., Jr. 1965. A new nematotaeniid cestode from California lizards. Jap. J. Exp. Med., 35(4): 30 1-303. SRG at: sgoldberg@whittier.edu 286 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 3, 2002 On behalf of the Membership and the Board of Directors, the Editorial Staff wishes to extend our thanks and appreciation to DR. PATRICK L. ODELL Baylor University for serving as Manuscript Editor of the Texas Journal of Science from 2001 to 2002. THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 3, 2002 287 IN RECOGNITION OF THEIR ADDITIONAL SUPPORT OF THE TEXAS ACADEMY OF SCIENCE DURING 2001 Patron Members Ali R. Amir-Moez Deborah D. Hettinger Don W. Killebrew David S. Marsh Patrick L. Odell John Sieben Ned E. Strenth Charles H. Swift Sustaining Members James Collins Dovalee Dorsett Stephen R. Goldberg Norman V. Horner Michael Looney Judith A. Schiebout Fred Stevens Supporting Members David A. Brock Frances Bryant Edens Donald E. Harper, Jr. Paul D. Mangum George D. McClung Jimmy T. Mills Jim Neal Nancy Ellen Partlow Paul T. Price Sammy M. Ray John Riola Lynn Simpson William F. Thomann Milton W. Weller 288 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 3, 2002 Plan Now for the 106th Annual Meeting of the Texas Academy of Science February 27 - March 1, 2003 Stephen F. Austin State University Program Chair John T. Sieben Dean of Natural Sciences & Mathematics Texas Lutheran University 1000 W. Court Street Seguin, Texas 78155 Phone: 830.372.6005 FAX: 830.372.6095 E-mail: jsieben@tlu.edu Local Host William D. Clark Dept, of Mathematics & Statistics Stephen F. Austin State University P.O. Box 10340 SFA Sta. Nacogdoches, Texas 75962 Phone: 936.468.3805 FAX: 936.468.1669 E-mail: clark@sfasu.edu For additional information relative to the Annual Meeting, please access the Academy homepage at: www . texasacademyofscience . org Future Academy Meetings 2004-Schreiner University THE TEXAS ACADEMY OF SCIENCE, 2002-2003 OFFICERS President : President Elect : Vice-President'. Immediate Past President : Executive Secretary : Corresponding Secretary '. Managing Editor: Manuscript Editor: Treasurer: AAAS Council Representative: Larry D. McKinney, Texas Parks and Wildlife Department John T. Sieben, Texas Lutheran University John A. Ward, Brook Army Medical Center David R. Cecil, Texas A&M University-Kingsville Fred Stevens, Schreiner University Deborah D. Hettinger, Texas Lutheran University Ned E. Strenth, Angelo State University Robert J. Edwards, University of Texas-Pan American James W. Westgate, Lamar University Sandra S. West, Southwest Texas State University DIRECTORS 2000 Bobby L. Wilson, Texas Southern University John P. Riola, Texaco Exploration 2001 David S. Marsh, Angelo State University Felipe Chavez-Ramirez, International Crane Foundation 2002 Sushma Krishnamurthy, Texas A&M International University Raymon D. Mathews, Jr., Texas Water Development Board SECTIONAL CHAIRPERSONS Anthropology: Roy B. Brown, Instituto Nacional de Antropologia y Historia Biological Science: David S. Marsh, Angelo State University Botany: Cyndy Galloway, Texas A&M University-Kingsville Chemistry: Mary A. Kopecki-Fjetland, St. Edward’s University Computer Science: John T. Sieben, Texas Lutheran University Conservation and Management: Andrew C. Kasner, Texas A&M University Environmental Science: Cindy Contreras, Texas Parks & Wildlife Department Freshwater and Marine Science: Hudson DeYoe, University of Texas-Pan American Geology and Geography: Jeff Pittman, Lamar University Mathematics: Benjamin J. Sultenfuss, Stephen F. Austin State University Physics: Robert Hamilton, Angelo State University Science Education: Julie F. Westerlund, Southwest Texas State University Systematics and Evolutionary Biology: Allan Hook, St. Edward’s University Terrestrial Ecology: Monte Thies, Sam Houston State University Threatened or Endangered Species: Donald L. Koehler, Austin Parks and Recreation Dept. COUNSELORS Collegiate Academy: Jim Mills, St. Edward’s University Junior Academy: Vince Schielack, Texas A&M University Nancy Magnussen, Texas A&M University PERIODICAL POSTAGE PAID AT LUBBOCK TEXAS 79402 RETURN SERVICE REQUESTED 002AAU7904 ROOM 25 NHB SMITHSONIAN INSTITUTION LIBRARY-ACQUISITIONS *(SMIV) WASHINGTON, DC 20560-0154 THE TEXAS JOURNAL OF SCIENCE PrinTech, Box 43 1 5 1 Lubbock, Texas 79409-3151 -T4X THE TEXAS JOURNAL OF SCIENCE Volume 54 Number 4 November 20ft GENERAL INFORMATION MEMBERSHIP.— Any person or member of any group engaged in scientific work or interested in the promotion of science is eligible for membership in The Texas Academy of Science. For more information regarding membership, student awards, section chairs and vice-chairs, the annual March meeting and author instructions, please access the Academy’s homepage at: www . texasacademy ofscience . org Dues for regular members are $30.00 annually; supporting members, $60.00; sustaining members, $100.00; patron members, $150.00; associate (student) members, $15.00; family members, $35.00; affiliate members, $5.00; emeritus members, $10.00; corporate members, $250.00 annually. Library subscription rate is $50.00 annually. The Texas Journal ofScience is a quarterly publication of The Texas Academy of Science and is sent to most members and all subscribers. Payment of dues, changes of address and inquiries regarding missing or back issues should be sent to: Dr. Fred Stevens, Executive Secretary The Texas Academy of Science CMB 5980 Schreiner University Kerrville, Texas 78028-5697 E-mail: FStevens@schreiner.edu AFFILIATED ORGANIZATIONS American Association for the Advancement of Science, Texas Council of Elementary Science Texas Section, American Association of Physics Teachers Texas Section, Mathematical Association of America Texas Section, National Association of Geology Teachers Texas Society of Mammalogists The Texas Journal ofScience (ISSN 0040-4403) is published quarterly at Lubbock, Texas, U.S. A. Periodicals postage paid at San Angelo, Texas and additional mailing offices. POSTMASTER: Send address changes and returned copies to The Texas Journal of Science, Dr. Fred Stevens, CMB 5980, Schreiner University, Kerrville, Texas 78028-5697, U.S. A. The known office of publication for The Texas Journal of Science is the Department of Biology, Angelo State University, San Angelo, Texas 76909; Dr. Ned E. Strenth, Managing Editor. THE TEXAS JOURNAL OF SCIENCE Volume 54, No. 4 November, 2002 CONTENTS DesCartes’ and Hessian Foliums. By A. R. Amir-Moez and J. A. Chavoshi . 291 New Record of Anthracotheriidae (Artiodactyla: Mammalia) from the Middle Eocene Yegua Formation (Claiborne Group), Houston County, Texas. By Patricia A. Holroyd . 301 Ichnology, Stratigraphy and Paleoenvironment of the Boerne Lake Spillway Dinosaur Tracksite, South-Central Texas. By J. Michael Hawthorne, Rena M. Bonem, James O. Farlow and James O. Jones . 309 Modeling Survival for Unthinned Slash Pine Plantations in East Texas Under the Influence of Non-planted Tree Basal Area and Incidence of Fusiform Rust. By Young-Jin Lee and Dean W. Coble . 325 Mass Capture of Insects by the Pitcher Plant Sarracenia alata (Sarraceniaceae) in Southwest Louisiana and Southeast Texas. By Robert E. Evans, Barbara R. MacRoberts , Thomas C. Gibson and Michael H. MacRoberts . 339 Female Reproduction in the Western Diamond-Backed Rattlesnake, Crotalus atrox (Serpentes: Viperidae), from Arizona. By Philip C. Rosen and Stephen R. Goldberg . . 347 New Distribution Record and Ecological Notes of the Freshwater Hydrozoan Craspedacusta sowerbii in Southeast Texas. By Richard C. Harr el . 357 General Notes A Scientific Comparison of Centrifugally Cast Fiberglass Reinforced Polymer Pipe and Bar Wrapped Concrete Cylinder Pipe using Finite Element Analysis. By M. Faruqi and M. Jao . 363 Noteworthy Records of Mammals from the Rolling Plains of Texas. By Chad A. Campbell, Thomas E. Lee, Jr. and Allan J. Landwer . 365 Recent Records of Bats from the Lower Canyons of the Rio Grande River of West Texas. By Loren K. Ammerman, Rogelio M. Rodriguez, Jana L. Higginbotham and Amanda K. Matthews . 369 Recognition of Member Support . 375 Annual Meeting Notice for 2003 . 376 Index to Volume 54 (Subject, Authors & Reviewers) . 377 Postal Notice . 383 THE TEXAS JOURNAL OF SCIENCE EDITORIAL STAFF Managing Editor: Ned E. Strenth, Angelo State University Manuscript Editor: Robert J. Edwards, University of Texas- Pan American Associate Editor for Botany: Janis K. Bush, The University of Texas at San Antonio Associate Editor for Chemistry: John R. Villarreal, The University of Texas-Pan American Associate Editor for Computer Science: Nelson Passos, Midwestern State University Associate Editor for Environmental Science: Thomas LaPoint, University of North Texas Associate Editor for Geology: Ernest L. Lundelius, University of Texas at Austin Associate Editor for Mathematics and Statistics: E. Donice McCune, Stephen F. Austin State University Associate Editor for Physics: Charles W. Myles, Texas Tech University Manuscripts intended for publication in the Journal should be submitted in TRIPLICATE to: Dr. Robert J. Edwards TJS Manuscript Editor Department of Biology University of Texas-Pan American Edinburg, Texas 78541 red wards@panam . edu Scholarly papers reporting original research results in any field of science, technology or science education will be considered for publication in The Texas Journal of Science. Instructions to authors are published one or more times each year in the Journal on a space-available basis, and also are available from the Manuscript Editor at the above address. They are also available on the Academy’s homepage at: www . texasacademy ofscience . org The Texas Journal of Science is published quarterly in February, May, August and November for $30 per year (regular membership) by The Texas Academy of Science. Periodical postage rates (ISSN 0040-4403) paid at Lubbock, Texas. Postmaster: Send address changes and returned copies to Dr. Fred Stevens, Executive Secretary, CMB 5980, Schreiner University, Kerrville, Texas 78028- 5697, U.S.A. TEXAS J. SCI. 54(4):291-300 NOVEMBER, 2002 DESCARTES’ AND HESSIAN FOLIUMS A. R. Amir-Moez and J. A. Chavoshi Department of Mathematics , Texas Tech University Lubbock, Texas 79409-1042 and 93310, Le Pre, St. Gernais, France Abstract. — The conic polars ofx3 + y3 = 3axy are called the Hessian Foliums. The set of areas of the loops of this set becomes a convergent infinite series. Also the set of lengths of these loops gives another convergent series (Bagchi 1939). 1 . The Folium The equation of the DesCartes’ folium is x 3 + y3 = 3 axy. (1) For simplicity let a = 1 . Then one obtains a set of parometric equations for (1) by the following set of equations x3 + y3 = 3xy (2) This set of equations will give jc3(1 + t 3) = 3jc 2t. (3) From (2) and (3) one obtains x 1 + t 3 3 12 1 + t3‘ (4) y = For the points at infinity (oo) one lets 1 -ft3 approach zero, that is, 1 + t3 = (1 + 0(1 - t + t2) ->0. (5) 292 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 4, 2002 One observes that 1 — t + t2 ^ 0. So in order to obtain (5) one must let t approach — 1 . This implies that the asymptote of the curve is parallel to y = — x or x + y = 0. Let the asymptote be y = - Jt + b. (6) Then t 2 -\-t b = lim (y + x) = 3 lim . (7) A'- *00 v t-+ - 1 i . 3 t~* - 1 1 + t Employing L’ Hospital’s rule, one gets b = 3 lim, - - 1 = -1 (8) ,— 1 3/ 2 So (6) and (8) imply that the asymptote is y=-x-l. (9) Now change (1) to the polar coordinates. So one gets r3(cos3 6 + sin30) = 3r2 sin 6 cos 0 , which yields 3 sin 6 cos 6 cos3 6+ sin3 & (10) Clearly the graph is symmetrical with respect to y = x. Therefore for the graph one only considers < 6 < ^ which, along with the y = x symmetry, allows one to sketch the graph. Or with the use of a computer, one can obtain a good graph (Figure 1). Figure 1. AMIR-MOEZ & CHAVOSHI 293 One may observe that (1) in polar coordinates is 3asmdcos6 r = - . (11) cos3 6 + sin3 6 So in general other foliums can be obtained from Figure 1 by homothetic transformations. 2. Homogeneous Coordinates X Y If one lets x = — and y = — in (1), one obtains One can write (12) as X3 +Y3 - 3aXYZ = 0. (12) (13) Note that Z -> 0 corresponds to the points at oo . One may obtain the asymptote by letting Z 0 in (13). Since the asymptote has already been obtained, one shall not pursue this method. 3. The Conic Polar Let A(p,q) be a point in the plane of a plane curve of a third degree equation. Let f(x,y,z) = 0 be the homogeneous equation of the curve. (Note that small letters have been used instead of X, Y, Z). Let (p,q,r ) be the set of homogeneous coordinates of A. Then df df df A dx dy dz is defined to be the conic polar of A with respect to f(x,y) = Now one shall apply (14) to the folium x 3 + y3 - 3axyz one obtains p(3x2 — 3 ayz) + q{3y 2 — 3 axz) + r{ — 3axy) — 0. (14) 0. = 0. So (15) By setting z — 1 and r = 1, this equation in ordinary Cartesian coordinates will be p(x2 - ay) + q(y2 - ax) - axy = 0. (16) Simplifying (16) one gets px 2 — axy + qy2 — aqx — apy = 0. (17) 294 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 Note that (17) is the equation of a conic section. 4. Types of Conic Sections Consider the second degree equation ax 2 + bxy + cy2 + px + qy + d = 0. (18) Besides real and imaginary circles, there are nine cases to be considered. The following is a list: I. The real ellipse, VI. The parabola, II. The imaginary ellipse, VII. Imaginary parallel lines, III. Imaginary intersecting VIII. Real parallel lines, lines, IX. Two coincidental lines. IV. The hyperbola, V. Real intersecting lines, Essentially one puts all these into three categories: (0 Ellipse, where b2 — 4 ac < 0, (ii) Parabola, when b2 — 4 ac = 0, (iii) Hyperbola, when b2 — 4 ac > 0. Degenerate cases are parts of these categories, for example, two intersecting lines is a special case of hyperbola and satisfies (iii). For this study, one shall study the necessary and sufficient condition for which (18) is two lines. Let one write (18) as a quadratic equation in x ; that is, ax 2 + (by + p)x + cy2 + qy + d = 0. (19) For (19) to become two straight lines, one must have its discriminant to be a perfect square. The discriminant of (19) is (by + p) 2 — 4a(cy 2 + qy + d). (20) This is a polynomial of degree two in y. For (20) to be a perfect square, its discriminant has to be zero. One shall apply this last part of this section to (17). AMIR-MOEZ & CHAVOSHI 295 5. Hessian Foliums One can write (17) as px 2 —a(y + q)x + qy~ — apy = 0. The discriminant of (21) is a = a 2 (y + q) 2 - 4 p(qy 2 - apy). Now write A as a second degree polynomial in y, that is (a2 - \pq)y 2 + 2 (a2 q + lap 2)y + a2 q2 So setting the discriminent of (23) equal zero, one obtains {a 2 q + lap 2 ) 2 - a2 q2 {a2 - 4pq) = 0. Simplifying one gets q 3 + apq = 0. (21) (22) (23) (24) (25) This is quite interesting, because it means the locus of A{p,q) whose polar with respect to the folium x 3 + y 3 = 3ary is a pair of straight lines is another folium; that is jt 3 + y 3 = ax y. (26) Bagchi (1939) calls this the Hessian folium of (1). Note that (26) is homothetic of (1) with center 0, the origin and ratio — • 3 ' 6. The Set of Hessian Foliums If one applies to (26) what has been done for (1) one obtains another folium which is homothetic of (26) with ratio — — . This way one ob¬ tains 3 I 3 1 x* + ? = —axy. (27) Without loss of generality one can let a > 0. Then the set of so-called Hessian foliums are obtained from (1) by homothetic transformations of center (0,0) a ratio . Call these A,,A2,A3,... as follows: 296 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 A,: r’+y3 = xy, A2: x3+/ = - a-xy, (28) 7. Areas of Hessian Foliums First one obtains the area of the loop of (1). One gives the details. Let the area be A. Then A=±[ 2r2dd, where 3a sin 6 cos 6 r= - . sin3 6 + cos3 6 One observes that one can write (29) (30) 3tftan0sec 6 „ n r= - ,0 ^ 6. 1 +tan3^ So (31) tan2 6 sec2 6 (1+tan3 0)2 dd. (32) Let tan 6 — t. Then sec 2 Odd = dt , and 0 < t < 1 . Therefore (32) will be t2dt o+f 3f (33) Again one changes the variable. Let t3 = u. Then 3 r2dt = du and 0 < u < 1 . Finally one has du 0 +«)2 3a2 2 ' (34) AMIR-MOEZ & CHAVOSHI 297 Let An be the area of the loop of A„, n = 1,2,.... Then (35) So the set {A {,A „,...} is a geometric progression of ratio . Therefore -A. 8 (36) So the area of the loop of (1) is eight times the sum of the areas of the loops of its Hessian foliums. 8. The Parabolic Case Let one look at (17) again, that is the conic polar of A(p,q) with respect to (1). If (17) is a parabola, then one must have a2 — 4 pq = 0. (37) This means that the locus of A for which the polar is a parabola is the equilateral hyperbola _ <32 ^ = T (38) This idea could be applied to all A„’s, but it is not interesting. For a fixed a > 0, one can give the graph of this hyperbola. 9. Elliptic and Hyperbolic Cases Again for (17) to be an ellipse one must have a 2 - 4 pq < 0 (39) This implies that the locus of A(p,q) is a region of the plane which satisfies (40) For a = 2, one gets the shaded region of Figure 2. 298 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 Similarly, for the polar to be hyperbola one must have a2 -4pq> 0. So the locus of A(p,q ) is the region which is the unshaded region of Figure 2. Figure 2. (41) (42) 10. The Lengths of Loops Let one consider the set of loops of {A, A ,,A2,...}. Suppose the length of the loop of A; that is, the original folium is s. Then the length of the loops of A1,A2,A3,...,A„,... will be respectively 1 1 -5, -5, 3 9 (43) This is quite easy to prove, even though obtaining s is quite difficult. Let one call the lengths of these loops respectively l • Then eo oo ^ .a E/ =sE — = -s. «.l " n-1 3" 2 One shall write s as an integral. (44) = 6 a f Jo i {( l-2E)2+(2f-f*)2 (1 +t*) dt. (45) AMIR-MOEZ & CHAVOSHI 299 Computing s will be left as an exercise. Note that this section 10 was not studied in Bagchi’s paper. 1 1 . The Curvature Let one look at the set of parametric equations of (1); that is, x = 3a * 1 H3 q r2 y = 3a - \H3 (46) The curvature of the curve is obtained from k = dx # d2y dy # d2x dt dt 2 dt dt2 3 2 (47) One shall study the curvature at the node of the curve. This point corresponds to t = 0 and t -> oo . Since the curve is symmetrical with respect to the angle bisector of the first quadrant, one only needs to obtain the curvature at t = 0. Note that 3 a dx dt d2 x dt 2 — =3a dt d2y_ 1-2 1 = -18 a (l+f5)2j 2t2 - (l+f5)3 2 t-t4 = 6 a dt (i +t3)2 ■ i -it3+t6 3\ 3 (l+0 (48) So for t = 0, one has — =3a,— =0,^=0. dt dt2 dt d^y dt2 = 6 a. (49) 300 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 4, 2002 2 This implies that at the node the curvature is k = the radius of the curvature is and the center of the curvature is at either (0, or (y-, 0). Note that there are two circles of curvature at the node. One of them is or (50) x2 + y2 - 3 ay = 0 (51) The case a = 1 is shown in Figure 3. The other circle of curvature is x2 +y2 -3ax = 0. (52) This should explain interesting parts of Bagchi’s paper. Figure 3. Acknowledgments Appreciation is extended to Dr. Philip Morey of TAMU-Kingsville for his suggested improvements to the manuscript as well as providing a final typed copy suitable for use by the Editorial Staff. Literature Cited Bagchi, Amal Kumar. 1939. Note on the Folium of DesCartes. J. of the Science Book Club, Vol. 1, No. 1. TEXAS J. SCI. 54(4):301-308 NOVEMBER, 2002 NEW RECORD OF ANTHRACOTHERIIDAE (ARTIOD ACT YLA : MAMMALIA) FROM THE MIDDLE EOCENE YEGUA FORMATION (CLAIBORNE GROUP), HOUSTON COUNTY, TEXAS Patricia A. Holroyd Museum of Paleontology , University of California Berkeley , California 94720 Abstract.— A small new species of the anthracotheriid artiodactyl genus Heptacodon is described from the middle Eocene Yegua Formation (Claiborne Group) from a site near Lovelady, Houston County, Texas. These specimens represent the first record of Eocene anthracotheres in Texas and are the southernmost and easternmost occurrence of the genus. The new species appears to be the most primitive of the four species of Heptacodon and provides an opportunity to emend the generic diagnosis of this exclusively North American taxon. The remains of Eocene land mammals from the Gulf Coast Plain are exceedingly rare (see review in Westgate 2001). Although they are known from other formations within the middle Eocene Claiborne Group, land mammals have not previously been reported from the Yegua Formation. The purpose of this report is to describe an unusual, new occurrence of the anthracotheriid artiodactyl Heptacodon from the Yegua Formation in Houston County (100 miles north of the city of Houston and 100 miles east of Waco), Texas, and to focus new attention on this genus, the rarest among North American anthracotheriids. Anthracotheriids are a family of extinct suiform, bunoselenodont artiodactyl s that range in age from middle Eocene to Miocene and occurred throughout the Old World as well as in North America. Heptacodon is an exclusively North American genus that was first described from an isolated upper molar by Marsh (1894), and additional species were described by Troxell (1921) and Scott (1940), all from the White River Group of the northern Great Plains. MacDonald (1956) last reviewed the alpha taxonomy of the genus and recognized three species: H. curtus in the early Oligocene Upper Brule Formation, and H. occidentals and H. quadratus from the late Eocene-early Oligocene Lower Brule Formation, although he admitted that the latter species might fall within the range of variation of the former. MacDonald also reported the first occurrences of the genus outside the White River Group, from the Chadron Formation in Wyoming and South Dakota, 302 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 4, 2002 further confirming the presence of the genus during the late Eocene. More recently, yet earlier records of Heptacodon have appeared. Storer (1983) described a small species, H. pellionis , from the Lac Pelletier faunas of Saskatchewan which are placed within the late middle Eocene Duchesnean North American Land Mammal Age (NALMA) (Storer 1987; 1996). Fragmentary remains of Heptacodon have also been reported in the late middle Eocene (Duchesnean NALMA) Hancock Quarry fauna of the Clarno Formation in Oregon (Hanson 1996) and from the Duchesnean Claron Formation of central Utah (Eaton et al. 1999). Unfortunately, the fossils from Oregon or Utah are not adequate to diagnose to species. Thus, three previously-described species are recognized: the Whitneyan type species H. curtus ; the Orellan H. occidentalis’, and the Duchesnean H. pellionus. Heptacodon quadratics, as noted by MacDonald, appears to fall within the range of variation of H . occidentalis. Kron & Manning (1998) noted an undescribed Heptacodon from the Gulf Coast of Texas in their overview of North American anthracotheriid distribution, and this important record is described below. Material examined during the course of this study are deposited with the Frick Collection (F:AM) of the American Museum of Natural History (AMNH) in New York. Systematic Paleontology Order Artiodactyla Family Anthracotheriidae Genus Heptacodon Marsh 1894 Synonymies .—Heptacodon Marsh 1894, Anthracotherium Osborn & Wortman 1894, Octacodon Troxell 1921 (in part). Type species .—Heptacodon curtus Marsh 1894. Included species. — type, H. occidentalis , H. pellionus, H. yeguaensis, new species. Occurrences.— Duchesnean of Saskatchewan, Oregon, Texas and Utah; Chadronian of Wyoming, South Dakota and Colorado; Orellan and Whitneyan of South Dakota. Emended diagnosis .—Heptacodon differs from other Paleogene anthracotheriids in having a fused mandibular symphysis without trace of suture (unfused in most anthracotheriids), P/2 postprotocristid more buccally positioned and a slight central swelling along molar cristid HOLROYD 303 obliqua. Differs from the Asian genera Anthracothema , Anthracokeryx, Siamotherium and Anthracosenex in possessing a postentocristid, having a broken hypolophid and lacking an anterior protolophid. Differs from Anthracokeryx , North American Bothriodon, Aepinacodon and Arreto- therium , and Euro- American Elomeryx and African Bothriogenys in possessing a strong postprotocristid, tooth rows without significant diastemata between canine and P/1 and/or P/l-P/2, relatively simple P/2-P/3 with posterior cingulid slight, P/4 only slightly elaborated by a strong protocristid and lacking posterior cingulid, premolar lingual cingula absent, molar paracristid ending near base of metaconid and unconnected to anterior cingulum, and molar postentocristid weak (H. curtus ) to absent (other sp.). Further differs from Elomeryx , Bothrio¬ don , Aepinacodon and Arretotherium in having a compressed (rather than open) mesostyle and lower crown height. Heptacodon yeguaensis , new species Figure lb,c,f Holotype.- F:AM 42984, left M2/ (Fig. 1c). Paratype.— F:AM 42985, right M/3 (Fig. lb and f). Type Locality. — "Loc. 3, Lovelady, Houston County, Texas, Yegua Formation" (data from specimen tag). Type Horizon . —stratigraphic position unknown, middle Eocene Yegua Formation, Claiborne Group. Diagnosis. —Differs from all other Heptacodon (where known) in its smaller size and in having a moderately-developed mesiobuccal cingulum on the upper molar parastyle and relatively greater buccal projection of parastyle. Further differs from H. curtus and is similar to H. occidentalis and H. pellionus in retaining a relatively stronger hypolophid on the molars and having a weakly-developed preentocristid. Description . — F : AM 42984, a left M2/ or possibly M3/, is a low- crowned five-cusped tooth (Figure lc). It measures 15.1 mm in maximum length, 14.5 mm long at the midline and 19.8 mm in maxi¬ mum width. Assignment as an M2/ is most likely. The metastyle is poorly developed. In most anthracotheriids the metastyle is at least moderately developed on M3/ in order to occlude with the posteriorly- extended M/3 hypoconulid (e.g., as in AMNH 1039, Fig. la). How¬ ever, since a posterior wear facet is lacking on this tooth and M2/ and M3/ are not markedly different in size in known Heptacodon sp., it is 304 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 4, 2002 Figure 1 . Heptacodon sp., shown to demonstrate differences among species of Heptacodon. All specimens except lb are shown coated with ammonium chloride, and all scale bars = 5 mm. (a) H. occidentalism AMNH 1039, a left maxilla with P2/-M3/, from the Orellan Scenic Member of the Brule Formation, South Dakota; (b) H. yeguaensis , F:AM 42985, right M/3 in medial (lingual) view; (c) H. yeguaensis, F:AM 42984, left M2/; (d) H. curtus , F:AM 105170, right M/3, from Whitneyan Poleslide Member of the Brule Formation, South Dakota; (e) H. occidentalism AMNH 1360, right M/3, from Orellan Scenic Member of the Brule Formation, South Dakota; (f) H. yeguaensis , F:AM 42985, right M/3 in occlusal view. HOLROYD 305 not possible to exclude the possibility that F: AM 42984 is an M3/ of this species. If so, H. yeguaensis would also be characterized by an unusually small M3/ metastyle. The paracone and metacone are subequal in size with metacone positioned slightly toward the midline (linguad). Both cusps bear strong buccal ridges, better developed on the paracone than the metacone. The metastyle is tall, well-developed and cuspidate. It projects buccally beyond the bases of buccal cusps. The parastyle is larger, projecting buccally and distally with a slight crest atop it having a principally distal (rather than buccal) orientation. A moderately-developed cingulum is present on the buccal surface of the parastyle. This cingulum extends mesiad and sharply "ascends" to terminate near the occlusal surface. A well-developed and beaded (where unworn) anterior cingulum is present, extending lingually from near the midline and terminating near the base of the protocone. The protoconule is moderately developed and is approximately one-half the size of the major cusps. It is placed equi¬ distant between the paracone and protocone. The metaconule is pyramidal in shape with a moderately developed premetacrista that is mesiolingually directed to join the slight lingual cingulum between protocone and metaconule. A posterior cingulum extends from the base of the metaconule to the base of the metacone. Wear is strongest on the mesial faces of the cusps and crests, render¬ ing the paraconule confluent with the protocone, and the postprotoconule crista is nearly obliterated. The buccally-oriented pre- and postmeta- conule cristae are worn. F:AM 42985, a right M/3, is a five-cusped tooth (Figure lb, f), measuring 21 .7 mm in maximum length, 11.75 mm maximum width and with the hypoconulid alone measuring 6.65 mm in length and 7.33 mm in width, which is 18% shorter and 11% narrower than Heptacodon pellionus , the next smallest species of the genus. Enamel is missing on the posterior face of the protoconid and mesiobuccal corner of the protoconid, and the posterior half of the hypoconid is missing. The trigonid is formed by subequal protoconid and metaconid with the protoconid slightly mesiad of the metaconid. The paracristid is strong, descending the face of the paraconid in a shallow arc, terminating near the middle of the metaconid’ s base and a few millimeters above the slight anterior cingulum. The metaconid and protoconid are joined by a moderately straight protolophid, and there is a strong postmetacristid. 306 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 The trigonid is slightly wider than the talonid. The cristid obliqua is strong and high, taking a sinuous course across the talonid basin to ascend the posterior trigonid wall just lingual of the midline. A slight swelling is present near the center of the crest. The entoconid and hypoconid are joined by a V-shaped posterior hypocristid and posterior entocristid. The hypolophid is discontinuous and is better developed in its buccal half. The "heel" of M/3 is formed by a well-developed hypoconulid. The "loop" begins at the midline and terminates just posterior to the entoconid’s base, leaving the hypoconulid basin lingually open. Very slight buccal cingulids are present between hypoconulid and hypoconid and between hypoconid and protoconid. Compared with Heptacodon curtus (Figure Id) and H. Occident ali s . (Figure le), the M/3 is relatively narrower with respect to length, and the crests are more weakly developed. In most respects, the M/3 of H. yeguaensis appears to be a scaled-down version of its much larger relatives. However, like H. occidentalis and H. pellionus (not figured) it still retains a hypolophid , which is largely lost in H. curtus , and has a weak preentocristid, which is strongly developed in H. curtus. Etymology This species is named for the Yegua Formation from which it was collected. Discussion Although poorly known, Heptacodon yeguaensis appears to be the most primitive of the known species of Heptacodon. The retention of a lingual portion of the hypolophid on M/3, a lower molar preento¬ cristid, and a buccal cingulum on the upper molar are primitive features that are shared with other, earlier Paleogene anthracotheriids. These features are also more primitive than the condition observed in Heptacodon specimens from known Duchesnean and Chadronian sites (where comparable). This record is the first for this genus in the southern United States as well as its easternmost occurrence. This species may also represent the oldest occurrence of the genus in North America, but its precise relative age is difficult to determine for several reasons. The Yegua Formation is well constrained through micropaleontological analyses to approxi¬ mately two million years of the late middle Eocene, spanning the entirety of planktonic foraminiferal zone P14 and part of P15 (Meckel and Galloway 1996). Based on correlation of these zones to the HOLROYD 307 Geomagnetic Reversal Time Scale (Aubry et al. 1988), these zones correspond to Chrons 18n and the lower part of Chron 17r. Based on the latest attempts to determine the placement of the Uintan-Duchesnean boundary (Prothero 1996; Prothero et al. 1996), the interval spanned by the Yegua Formation would correspond to the latest Uintan and much of the Duchesnean. However, the precise stratigraphic position of the type locality within this formation has not yet been determined. The type and referred specimen were collected in 1936 by Claude Riley from a site near Lovelady, Houston County, Texas, but its precise location and stratigraphic position within the formation is not known. Other fossils collected with the mammal teeth include a dermatemydid turtle and an ariid catfish, which do not provide any additional evidence for the age of the locality. This limited evidence, combined with the Duchesnean age of faunas from the overlying Jackson Group, is sugges¬ tive of an early Duchesnean age assignment for the Lovelady fossils. However, an older or slightly younger age cannot be eliminated on available evidence. Heptacodon yeguaensis also provides a frustratingly incomplete addition to the knowledge of Eocene mammals on the Gulf Coastal Plain. As recently reviewed by Westgate (1986; 1990; 2001) the knowledge of Eocene mammal evolution on the Gulf Coast Plain is principally from the Uintan Casa Blanca local fauna of Texas. The remainder of sites and faunas, like that described here from near Lovelady, represent only isolated occurrences, hinting at the tropical faunas that once thrived along the ancient Texas shore. Acknowledgments Earl Manning first recognized the anthracotheriid affinities of the material described here, and I thank Malcolm McKenna for suggesting that I examine these fossils. Funding for museum work was provided from the Annie M. Alexander Endowment. Jon Baskin and James Westgate provided helpful reviews. This is University of California Museum of Paleontology contribution #1790. Literature Cited Eaton, J. G., J. H. Hutchison, P. A. Holroyd, W. W. Korth & P. Goldstrand. 1999. Vertebrates of the Turtle Basin Local Fauna from middle Eocene rocks of the Sevier Plateau, south-central Utah. Pp. 463-469, in Vertebrate Paleontology in Utah, Utah State Geol. Survey Misc. Publ. 99-1 (D. D. Gillette, ed.), xiv + 554 pp. Hanson, C. B. 1996. Stratigraphy and vertebrate faunas of the Bridgerian-Duchesnean 308 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 Clamo Formation, north-central Oregon. Pp. 206-239, in The Terrestrial Eocene-Oligocene Transition in North America. Cambridge University Press, New York (D. R. Prothero and R. J. Emry, eds.), xii + 688 pp. Kron, D. G. & E. Manning. 1998. Anthracotheriidae. Pp. 381-388, in C. M. Janis, K.M. Scott, and L.L. Jacobs (eds.), Evolution of Tertiary Mammals of North America, Volume I: Terrestrial Carnivores, Ungulates, and Ungulate-like Mammals, xii + 691 pp. MacDonald, J. R. 1956. The North American anthracotheres. J. Paleontol., 30(3):615-645. Marsh, O. C. 1894. A new Miocene mammal. Amer. Journ. Sci., 3rd Ser., 48:409. Meckel, L. D. Ill & W. E. Galloway. 1996. Formation of high-frequency sequences and their bounding surfaces: case study of the Eocene Yegua Formation, Texas Gulf Coast, USA. Sed. Geol., 102:155-186. Osborn, H. F. & J. L. Wortman. 1894. Fossil mammals of the lower Miocene White River Beds. Collection of 1892. Bull. Amer. Mus. Nat. Hist. 6(7): 199-228. Prothero, D. R. 1996. Magnetostratigraphy of the Eocene-Oligocene transition in Trans-Pecos Texas. Pp. 189-198, in The Terrestrial Eocene-Oligocene Transition in North America. Cambridge University Press, New York (D. R. Prothero and R. J. Emry, eds.), xii + 688 pp. Prothero, D. R., J. L. Howard & T. H. H. Dozier. 1996. Stratigraphy and paleomagnetism of the upper middle Eocene to lower Miocene (Uintan to Arikareean) Sespe Formation, Ventura County, California. Pp. 171-188 in The Terrestrial Eocene-Oligocene Transition in North America. Cambridge University Press, New York (D. R. Prothero and R. J. Emry, eds.), xii -I- 688 pp. Scott, W. B. 1940. The mammalian fauna of the White River Oligocene. Part IV. Artiodactyla. Trans. Amer. Phil. Soc., 28:363-746. Storer, J. E. 1983. A new species of the artiodactyl Heptacodon from the Cypress Hills Formation, Lac Pelletier, Saskatchewan. Canadian J. Earth Sci., 20:1344-1347. Storer, J. E. 1987. Dental evolution and radiation of Eocene and early Oligocene Eomyidae (Mammalia, Rodentia) of North America, with new material from the Duchesnean of Saskatchewan. Dakoterra, 3:108-117. Storer, J. E. 1996. Eocene-Oligocene faunas of the Cypress Hills Formation, Saskatchewan. Pp. 240-261, in The Terrestrial Eocene-Oligocene Transition in North America. Cambridge University Press, New York (D. R. Prothero and R. J. Emry, eds.), xii 4- 688 pp. Troxell, E. L. 1921. The American Bothriodonts. Am. Journ. Sci. Ser 5., 1:325-339. Westgate, J. W. 1986. Stratigraphic occurrence and correlation of early Tertiary vertebrate faunas, Trans-Pecos, Texas: Agua Fria-Green Valley Areas. J. Vert. Paleont., 6:350-373. Westgate, J. W. 1990. Uintan land mammals (excluding rodents) from an estuarine facies of the Laredo Formation (middle Eocene, Claiborne Group) of Webb County, Texas. J. Paleont., 64(3) :454-468. Westgate, J. W. 2001. Paleoecology and biostratigraphy of marginal marine Gulf Coast Eocene vertebrate localities. Pp. 263-297, in Eocene Biodiversity: Unusual Occurrences and Rarely Sampled Habitats (G. F. Gunnell, ed.),. Kluwer Academic/Plenum Publishers, New York, xxi -h 1 -433 . PAH at: pholroyd@uclink4.berkeley.edu TEXAS J. SCI. 54(4):309-324 NOVEMBER, 2002 ICHNOLOGY, STRATIGRAPHY AND PALEOENVIRONMENT OF THE BOERNE LAKE SPILLWAY DINOSAUR TRACKSITE, SOUTH-CENTRAL TEXAS J. Michael Hawthorne, Rena M. Bonem, James O. Farlow and James O. Jones* Hy4 Environmental, Ltd., 500 N. Carroll Ave. , Suite 120, Southlake, Texas 76092; Geology Department, Baylor University, Waco, Texas 76798-7354 and Department of Geosciences, Indiana University-Purdue University Fort Wayne 2101 Coliseum Boulevard East, Fort Wayne, Indiana 46805 * Deceased Abstract. — Record flooding in late June 1997 in south central Texas exposed dinosaur footprints on the Boerne Lake spillway near Boerne, Texas. At least three trackways are present in the upper portion of Unit No. 3 of the Lower Cretaceous Glen Rose Formation. This sequence represents a period of high-frequency depositional cyclicity on a very shallow and partly restricted inner shelf. The track-bearing layer exhibits features characteristic of sabkha-like tidal flats subject to subaerial exposure. It is overlain and underlain by red clay horizons of secondary origin. Those dinosaur prints exhibiting sufficient morphologic preservation to be identifiable appear to have been made by a theropod (carnivorous) dinosaur. Other footprints were too poorly preserved to permit identification of the trackmakers more specifically than as bipedal dinosaurs. Record flooding in south central Texas resulted in four deaths and several million dollars in property damage in late June of 1997. When the flood waters receded, erosion had removed vegetation, soil and bedrock, necessitating repair to the emergency spillway at Boerne Lake near Boerne, Texas. When Natural Resources Conservation Service officials surveyed the damaged area, they discovered newly exposed dinosaur tracks on the floor of the spillway. Geologists from Baylor University, the University of Texas at San Antonio and Indiana University-Purdue University Fort Wayne investi¬ gated the site. In cooperation with city, county and federal officials, they attempted to keep the site location concealed until the tracks could be documented. However, the news media learned of the site and turned the discovery into a TV and newspaper mass media event with national news coverage. Location Boerne Lake is located approximately 0.5 miles west of Interstate Highway 10 at the Ranger Creek exit just northwest of Boerne, Texas (Fig. 1). The spillway is located along the southeast corner of this lake, 310 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 4, 2002 Figure 1 . Location of the Boerne Lake Spillway dinosaur tracksite. which provides potable water to the City of Boerne. The tracks lie in thin-bedded limestone in the upper third of the spillway. Methods Following an initial evaluation of the site, flood detritus and remnants of the overlying red clay were removed using small hand tools. Photo¬ documentation included oblique photos of the three well-preserved trackways from multiple perspectives. Each individual footprint was also photographed from directly overhead to capture the track shape as preserved. A detailed footprint map was created by laying out a chalk line grid on the track layer at five foot intervals with the axes corresponding to magnetic N/S and E/W directions as determined in the field with a Brun- ton compass/clinometer. Individual footprint centers were measured relative to this grid, yielding north-south and east-west coordinates for HAWTHORNE ET AL. 311 the center of each print. These coordinates were input into a CADD program to develop the base map. Individual footprint bearings as measured with the Brunton were utilized to orient the tracks on the map relative to magnetic north. Footprint measurements were made in accordance with standard pro¬ tools (e.g., Leonardi 1987; Thulborn 1990; Lockley 1991; Farlow & Galton 2002). Footprint lengths were measured from the rear margin ("heel" - actually the back of the toe region of these digitigrade animals) of the print to the tip of the middle toe (digit III). Footprint widths were measured across the tips of the inner and outer toes. Maximum foot¬ print depths were measured both in the toe region and in the rear ("heel") portion of the print. Individual footprint compass bearings were sighted from the rear margin of the print to the tip of digit III. The averages of the bearings of a particular footprint and the print preceding it, and also of the footprint and the print following it, were taken to indicate the overall direction of travel of the dinosaur during the two steps. The bearing of a given footprint could then be compared with these two estimates of the animal’s direction of travel. Footprint rotation with respect to the preceding print compares the inward (toward the trackway midline) or outward (away from the trackway midline) orientation of a particular footprint relative to an overall direction of travel estimated as the average of the bearings of the print in question and the preceding footprint. Footprint rotation with respect to the following print uses an overall direction of travel estimated from the print in question and the following footprint. By convention (Leonardi 1987) negative values indicate that the footprint toes inward, and positive values indicate outward footprint rotation. A zero value indicates that the footprint bearing coincides with the dinosaur’s direction of travel. Trackway paces denote the distance from a given print to the next print made by the opposite foot, and strides refer to the distance from a given print to the next print made by the same foot. When possible, paces and strides were measured using the tip of a footprint’s digit III as the reference point; where this was not possible, the centers of foot¬ prints were utilized. From the pace ending in a particular footprint, the pace beginning with that footprint, and the stride opposite the footprint (made as the opposite foot was brought forward), the pace angulation around the footprint was calculated using the law of cosines. This 312 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 indicates the 1 inear ity/narrowness of the trackway; a pace angulation of 180° means that footprints of the left and right side of the body fall on a single line, a very narrow trackway. With increasingly narrow track¬ ways the value of the pace angulation is sensitive to slight errors of measurement. If the measured stride is greater than the sum of the two paces opposite it, the pace angulation cannot be calculated. In such cases the pace angulation was estimated as ~ 180°. Casts of selected footprints were made and topographically digitized to evaluate print morphology. Utilizing this information, contour maps of the footprints were created. Stratigraphic field investigative methods included measuring and describing the section at the site, field correlation with local marker beds of the upper Glen Rose Formation exposed at adjacent cliff outcrops, photo-documenting stratigraphic features of the site and surrounding area, and sampling select horizons for further analysis. Field correla¬ tions to adjacent outcrops were performed visually. The flat-lying nature of the beds of the Glen Rose Formation facilitated gross correla¬ tions between outcrops, as did the appropriate topographic and geologic maps (U.S.G.S. 1964; Texas Bureau of Economic Geology 1982). Geologic Setting Structurally the site lies south of the Llano Uplift and the associated San Marcos Arch, both of which were positive structural features during the Cretaceous Period (Adkins 1932). These two features appear to have been primary controls of the depositional and structural setting of the site, creating a broad area of shallow-water and shoreline deposition in the encroaching Trinity Sea (Stricklin et al. 1971). The shallow shelf of central Texas during early Cretaceous time exhibited a cyclically transitional shoreline and shallow marine environment along an epeiric sea. Typical environments included streams, marshes, tidal flats, lagoons and shallow subtidal marine environments. Deposition occurred primarily in shallow water, with frequent subaerial exposure. Isolated rudist and coral patch reefs occurred commonly in the shallow waters, with shelf margin reefal facies occurring farther southeast along the developing Stuart City Reef Trend (Stricklin et al. 1971). HAWTHORNE ET AL. 313 Description of the Boerne. Lake Spillway Tracksite The site occurs on the gently sloping spillway formed of thinly bedded, horizontal limestone layers. Much of the face is covered with flood debris. The active portion of the spillway during the flood was denuded of vegetation. Dinosaur prints exposed on the spillway of Boerne Lake occur in thinly bedded limestones in the upper Glen Rose Formation of the Lower Cretaceous Trinity Group, about 40 m above the "Corbula bed" that marks the top of the lower Glen Rose (Fig. 2). This stratigraphic position corresponds to the upper part of Unit 3 of the upper Glen Rose (Stricklin et at. 1971). This is considerably higher in the Glen Rose Formation than the common track-bearing layers, which lie a few meters below the Corbula zone (Stricklin et at. 1971; Stricklin & Amsbury 1974). The Boerne Lake Spillway tracksite is within the upper third of a 4.4-m-thick cyclic interval of thinly bedded wackestone and nodular clayey wackestone, which represents shallow-subtidal to tidal-flat deposition. Fossils in this unit include abundant thin-shelled oysters, other small bivalves, thin, high-spired gastropods, orbitolinids, miliolids and ostracodes. Cyanobacterial stromatolites, mud-cracked layers, and horizontal dissolution cavities (probably dissolved evaporite layers) are associated with the dinosaur prints. The upper meter contains two bored hardground surfaces. The track-bearing lithologic unit represents a period of high-frequency depositional cyclicity on a very shallow and partly restricted inner shelf. During a period of lowest sea level, when the sabkha-like tidal flats prograded farthest seaward, dinosaurs were able to move through the area. While not necessarily diagnostic of tidal flat deposition, it is nonetheless characteristic of many Lower Cretaceous dinosaur footprints in Texas to occur in such environments. This upper interval of Unit 3 rests on a well-defined bored and oyster-encrusted hardground at the top of a 2-m-thick burrowed wacke¬ stone containing some requinid and monopleurid rudists (probably equivalent to the "Massive" marker bed of Unit 3 of Stricklin et al. 1971). Just below this, the upper Glen Rose is predominantly clay with abundant orbitolinids, probably representing a slightly deeper and open inner shelf. 314 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 Joshua Creek Kendall County, Texas Stricklin et al. 1971 Boerne Lake Spillway Kendall County, Texas Meters Feet EDWARDS 125-r400 GROUP Exogyra Bed - miliolids, bivalves, echinoid fragments slightly glauconitic LOWER GLEN ROSE FORMATION UNIT 4 UNITS Porocystis, echinoids, mollusks burrowed oysters bored hardground encrusted with oysters Orbitolina miliolids nodular, burrowed mollusks bored hardground miliolids, ostracodes, mollusks Orbitolina oysters stromatolite DINOSAUR TRACKS mudcracks ripple marks Orbitolina miliolids, ostracodes, mollusks nodular, burrowed - - thin-shelled oysters nodular, burrowed miliolids abundant Orbitolina horizontal dissolution cavity with red clay infilling bivalves abundant Orbitolina bored hardground encrusted with oysters Toucasia and Monopleura rudists gastropods Orbitolina burrowed miliolids, ostracodes, mollusks bivalves burrowed abundant Orbitolina a E3 EXPLANATION C = calcareous clay cW = clayey wackestone W = wackestone P = packstone ■ i G = grainstone Figure 2. Stratigraphic position of the Boerne Lake Spillway dinosaur tracksite. The approximate location of the contact between units 3 and 4 of Stricklin et al. (1971) in the Boerne Lake Spillway section is indicated. HAWTHORNE ET AL. 315 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0 -5 -20 -15 -10 -5 0 5 10 15 20 25 30 35 DINOSAUR TRACKS NORTH Figure 3. Diagrammatic map showing disposition of the better-preserved dinosaur footprints at the Boerne Lake Spillway tracksite (some of the footprints of the trackways for which measurements are given in Tables 1 and 2 are not depicted due to poor preservation). The interval between each tic mark along the horizontal and vertical margins of the map is one foot; mesh size of the lines in the grid is five feet. Letter labels designate trackways, and numbers individual footprints within those trackways. The OA sequence is the best-preserved trackway. The OD footprints were initially thought to constitute a single trail, but are presently interpreted as made by two animals (see Table 1 for assignment of prints to the two trackways). The OB and OC prints may represent trackways (with OB a possible continuation of OA), but were not well enough preserved to warrant measuring. 316 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 Immediately overlying the tidal-flat and hardground layers of upper Unit 3 is about a meter’s thickness of clay containing echinoids, bivalves, gastropods and green algae, representing a return to more open, normal-marine conditions. This fossiliferous clayey interval is probably the basal layer of Unit 4 as defined by Stricklin et al. (1971). Dinosaur Footprints Although the track-bearing surface is covered with numerous depres¬ sions that might have been footprints, most of these are so poorly preserved that nothing can be said about the number and kinds of animals responsible for them. Three sets of footprints can unambiguous¬ ly be associated in trackways, however (Fig. 3). A large bipedal dinosaur made Trackway OA. None of the footprints in the trackway is well preserved. The average footprint length in this trail is about 49 cm, and the average width about 43 cm (Table 1). Footprints of greater length than width are usually interpreted as having been made by theropod dinosaurs (Moratalla et al. 1988; Thulborn 1990; Lockley 1991). Although some of the prints have relatively short, thick toes (Figs. 4, 5), a feature typical of prints attributed to ornithopods (Thulborn 1990; Farlow & Chapman 1997), the best preserved footprint in the trackway shows rather longer, narrower toes, giving it a more theropod-like appearance (Farlow 1987; Pittman 1989; Hawthorne 1990). Consequently the ornithopod-like gestalt of most of the prints in the trackway is likely an artifact of preservation, and the trackmaker is identified as a large theropod. The size of the prints is comparable to those of footprints attributed to large theropods at other Glen Rose tracksites (Farlow 1987; 2001; Pittman 1989). The maker of trackway OA moved in a northerly direction across the site, making sharp changes in direction twice (Fig. 3). Over the interval of prints OA5 through OA7 the dinosaur abruptly turned to the right, and after making print OA17 it turned less sharply to the left. The pace angulation is generally high over the length of the trail (Table 2), as is typical for trackways of bipedal dinosaurs (Farlow & Chapman 1997), but the two abrupt changes of direction resulted in abnormally low values of the pace angulation. The dinosaur’s footprints usually toe slightly inward (slightly negative footprint rotation), as is typical of Comanchean tridactyl footprints attributed to theropods (Farlow 1987). HAWTHORNE ET AL. 317 Table 1. Measurements of individual Boerne Lake Spillway dinosaur footprints. Track- Foot- Symmetry Footprint Footprint Toe "Heel" way print (Left or Length Width Depth Depth Right) (cm) (cm) (cm) (cm) OA OA1 Right 46.3 42.1 13.1 9.1 OA2 Left 56.4 38.7 10.7 8.2 OA3 Right 47.2 50.3 12.2 10.1 OA4 Left 53.9 45.7 10.1 11.0 OA5 Right 43.9 43.0 10.7 10.7 OA6 Left 48.8 40.5 15.2 14.9 OA7 Right 45.7 40.5 11.3 6.7 OA8 Left 43.3 36.0 5.5 2.4 OA9 Right 47.9 40.2 11.6 7.3 OA10 Left 48.2 40.8 8.8 6.4 OA1 1 Right 46.3 48.8 11.3 9.1 OA12 Left 46.6 49.4 10.7 9.1 OA13 Right 46.3 44.2 6.7 5.5 OA14 Left 51.8 44.2 10.1 8.8 OA15 Right 54.9 42.1 8.5 7.6 OA16 Left 48.8 45.7 12.2 7.9 OA17 Right 57.9 48.8 10.7 4.6 OA18 Left 54.9 39.6 7.0 7.6 OD1 OD8 (1st print) ? ? ? ? ? OD6 (2nd print) ? 30.5 30.5 7.6 7.6 OD2 (4th print?) ? 19.8 18.9 4.6 4.0 ODO (5th print?) ? 25.9 21.3 4.6 7.9 OD-2 (6th print?) ? 27.4 25.9 7.6 6.7 OD2 OD7 (1st print) Left? 9 ? ? ? OD5 (2nd print) Right? 33.5 30.5 9.1 9.1 OD3 (3rd print) Left? 33.5 30.5 6.1 6.1 OD1 (4th print) Right? 27.4 27.4 7.6 7.6 OD-1 (5th print) Left? ? ? ? 9 OD-3 (6th print) Right? 24.4 25.9 4.6 2.1 Trackmaker OA moved in a leisurely fashion. The average stride is about 242 cm, roughly five times the length of individual footprints in the trail. This is toward the low end of stride/footprint length ratios observed in Glen Rose theropod trackways (Farlow 1987). What are interpreted as trackways (OD1 , OD2) of two smaller bipedal dinosaurs cut across trail OA at the northern end of the latter trail (Table 2; Figs. 3, 4d). Conceivably OD1 and OD2 constitute the trackway of a single animal with a rather broad straddle and a very low pace angula¬ tion; " wide-gauge" trackways attributed to theropod dinosaurs are known from the Jurassic (Lockley et al. 1996; Day et al. 2002). However, unambiguous wide-gauge trackways of bipedal dinosaurs have yet to be found in the Glen Rose Formation, where narrower trackways are common (Farlow 1987), and so it appears more likely that these are 318 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 Figure 4. Photographs of dinosaur footprints and trackways of the Boerne Lake Spillway tracksite. A-C: Trackway OA. (A) Oblique view of a portion of the trackway, beginning with footprint OA7; the dinosaur was moving away from the viewer. (B) Footprint OA7 (a right). (C) Footprint OA15 (a right). (D) Oblique view of the two OD trackways; the dinosaurs were moving toward the viewer. Footprints OD2 and ODO of one of the OD trackmakers are to the left of the meter stick, and print OD1 of the other OD trackmaker is to the right of the meter stick. The trackway of the large OA dinosaur crosses the trails of the two smaller animals, moving from right to left; print OA16 of the big dinosaur is at the end of the meter stick toward the viewer. HAWTHORNE ET AL. 319 Footprint 0A7 Footprint 0A15 B Figure 5. Topographic maps of footprints from trackway OA. Sizes of footprints are indicated by centimeter scales along the vertical and horizontal margins of the maps. (A) Footprint OA7 (a right). The topographic map was made from a cast (negative copy) of the footprint, and so left-right symmetry and topography are reversed from the actual footprint. Although the footprint has a shape reminiscent of prints attributed to ornithopods, note the suggestion of a clawmark on digit IV (the leftmost toemark in the topographic map). (B) Footprint OA15 (a right). In this case the topographic map was made from a positive copy of the footprint, and so topography and symmetry are the same as in the actual footprint. This footprint suggests longer and narrower, more typically theropod-like toes. Table 2. Measurements of Boeme Lake Spillway dinosaur trackways. Track- Reference Footprint Footprint Rotation (degrees) Pace Length Stride Opposite Pace way Footprint Bearing with Respect to: the Footprint Angulation (degrees; 0° =North) Preceding Following Ending in Beginning with (degrees) 320 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 4, 2002 I I CM Os i vo'O'ort-oooinTj-in^-'nmvo i I I I I o r- OO SO I ONOOomTt | I so vo oo m t"- I 8 8 8 Tth-n^oori'^t^O oo o — < O oo |odr^(Nr^vdrnod^j:o\’- o c4 o'Ot^oo — < >— 1 i i c QQQQQQQ ooooooo < o Q O Nu "^J1=7N j —(p u +)N u, (2) where: p i = instantaneous mortality rate for infected trees, p u — instantaneous mortality for uninfected trees and, <|> = instantaneous rate of uninfected trees becoming infected, and 7 = instantaneous rate of infected trees becoming uninfected = 0. After a period of time (dA), the numbers in each group will change (dN j and dNu). The number of trees in the infected group will de¬ crease due to mortality at the rate p but will gain the number of uninfected trees that become infected during this time at the rate (J). Mortality and a change in uninfected status at the rates pu and (J), respec¬ tively, will both decrease the number in the uninfected component. The LEE & COBLE 329 parameter 7 = 0 because there is no possibility of infected trees becom¬ ing uninfected. Adams’ equations can be solved (Lee & Coble 2002) via the Method of Determinants (Grossman & Derrick 1988). The resulting equations are expressed as the change in numbers of slash pine trees between two time periods, and A2: Ni2=0Nul e-a(A2-Al) + (Nil-/5Nul)e-pi(A2"Al) (3) N =N P_a(A2_A') 1Nu2 1Nul e > where: A2 = projection age (years), A! = initial age (years), Ni2 = number of surviving infected trees per hectare at A2, N„ = number of surviving infected trees per hectare at A,, Nu2 = number of surviving uninfected trees per hectare at A2, Nul = number of surviving uninfected trees per hectare at A1? and a,/5,Pi = parameters to be estimated. Equations (3) provide for separate estimates of mortality rates for infected and uninfected slash pine trees, as well as the possible transition from an uninfected to infected status. The parameter a = (pu + (j>) is the rate at which trees are lost from the uninfected class. The parameter /3 = - represents the proportion of unifected trees that become in¬ fected, some of which are lost at the rate, a. Behavior of this model is consistent with the desired properties of path invariance and conver¬ gence; the number surviving planted pine trees converge to zero as age goes to positive infinity. Adams (1989) and Adams et al. (1996) reported that pine survival in their studies decreased as site productivity (as measured by site index) 330 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 4, 2002 increased. The faster rate of stand development in plantations (and natural forests) of higher site productivity triggers competition- induced mortality at earlier ages than in plantations of lower site productivity (Oliver and Larson 1996). Thus, the number of surviving trees at any plantation age will be lower in plantations of higher site productivity versus plantations of lower site productivity. As mentioned earlier, the competitive effects of non-planted trees may also influence pine survival. So, a combined variable for the ratio of non-planted tree basal area to total basal area per hectare (RNTB) and site index (S) in meters (base age = 25 years) was incorporated into the differential equations (2) of Adams (1989) and solved in a similar manner as before (subject to the assumption that RNTB is constant with respect to age; see discussion below): Ni2=j3Nule-^*^^-A^ + (Nil-j8Nul)e-^^(^-^ Nu2 = Nule — <*(S*RNTB)(A2— A,) (4) where all other variables and parameters defined as before. The introduction of the two variables, S and RNTB, had the potential to alter the solution of equations (2) if they were not constant terms; i.e., if S and/or RNTB were functions of plantation age, then the solu¬ tions in equations (4) do not follow from the differential equations. The following hypotheses were tested via simple linear regression to deter¬ mine if S and RNTB were constant terms before the equations (4) were fit to the data: Hoi: S is constant across plantation age, Ho2: RNTB is constant across plantation age, and Ho3: S*RNTB is constant across plantation age. None of the three hypotheses were rejected at the a = 0.01 probability level (P = 0.6393, P = 0.0117, P = 0.0706, respectively), so S and RNTB were assumed to be constant across the range of plantation ages in this study. LEE & COBLE 331 After S and RNTB were found to be constants, equations (4) were fit to the data. Preliminary analyses (not presented) showed that survival, site index, and RNTB were significantly correlated (P < 0.05). A fitting procedure described by Borders (1989) was used to account for the presence of this cross-equation error correlation. As a result, equations (4) were fit to the 197 observations in a simultaneous manner using the SYSNLIN procedure in SAS (1985). Model evaluation.— The statistical measures used in this study for model evaluation were the coefficient of determination (R2), root mean square error (RMSE), mean percent bias (described below), and a simple linear regression analysis of observed versus predicted total surviving trees per hectare (described next). Simple linear regression (Zar 1999) was used to compare observed and predicted total surviving trees per hectare. Observed and predicted values were related according to the following simple linear model: Predicted TPH = b0 + b, * Observed TPH. If the survival prediction models correctly estimated the number of surviving trees per hectare, then the intercept (bG) would not be significantly different from zero and the slope (bj) would be not be significantly different from one. A simultaneous /-test (Neter et al. 1985 :p. 147) was used to evaluate the hypothesis: Ho: (ft, ft) = (0,1), Ha: (ft, ft) 5* (0,1). Reynolds (1984) developed estimation procedures to test the accuracy of models. His procedures test both bias and precision rather than over¬ all prediction accuracy. These procedures were converted to a BASIC program (Rauscher 1986), then later to a SAS program (SASATEST; Gribko & Wiant 1992). SASATEST was used in this study to further examine the performance of the survival prediction models. SASATEST examines both bias and precision on an absolute or percentage basis. In SASATEST, percent bias is calculated as a percentage of the observed surviving trees per hectare: Y-Y BIAS = 100 y-1-, where: Y = predicted surviving trees per hectare and Y = observed surviving trees per hectare. In this study, precision is expressed as the standard 332 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 deviation of percent bias, which is also calculated by SASATEST. SASATEST then uses the mean percent bias (measure of bias) and the standard deviation (measure of precision) to calculate a 95 % confidence interval. If this confidence interval does not contain zero, then the bias is significant at the a = 0.05 level. SASATEST also checks the errors between predicted and observed values for departures from normality. If non-normality is detected, a 10% trimmed mean and jackknife stand¬ ard deviation were used to provide more robust confidence intervals. Results and Discussion Survival prediction models.— The following model developed from the slash pine plantation survival data provides separate estimates of the surviving number of slash pine trees: N|2=(Nil — 0.424429Nul)e(_0 021002((SH'RNTB)(A2_A,)) +0.424429N , ^-o ws^^swtbxaj-a,)) ^ AT -W p (— 0.00541647(SH‘RNTB)(A2— A,)) Iyu2~ Iyu\e > where, all variables are defined as before. A. The asymptotic standard errors for coefficients p{, a , B are 0.0030458, 0.0012421 and 0.12514, respectively. All parameters were significantly different from zero ( P < 0.05). The uninfected component in equation (5) explained about 92% of the variation in the surviving number of trees per hectare, while the infected component in equation (5) explained about 56% of the variation in the surviving number of trees per hectare (Table 2). Thus, the uninfected component was more accurately predicted than the number of surviving infected trees. Residual plots (not shown) revealed a random pattern around zero with no detectable trends. Fit statistics for equation (5) based on the data from 194 evaluation subplots are presented in Table 2. The survival prediction model (5) for uninfected slash pine trees over-estimated the number of surviving trees per hectare by 2.42%, though this value was not significant ( P > 0.05; Table 2). The survival prediction model (5) for infected slash pine trees significantly (P < 0.05) over-estimated the number of surviving trees per hectare by 7.46% (Table 2). This large value can be explained by the large amount of variability in the percent bias values (note the large confidence interval LEE & COBLE 333 Table 2. Fit statistics for performance evaluation of east Texas unthinned slash pine plantation survival model. Nu2 = number of surviving trees per hectare without a fusiform rust stem gall at Age 2, Ni2 = number of surviving trees per hectare with a fusiform rust stem gall at Age 2, and ** = significant (P < 0.05). Equation R2 Root Mean SquareError Mean Percent Bias 95 % Confidence Interval for Percent Bias nu2 0.92 100.82 2.42 2.42 ± 3.12 Ni2 0.56 106.23 7.46** 7.46 ± 5.59 in Table 2). This result was not unexpected since less variability in the predicted number of surviving trees was explained by the model for infected trees (R2 = 56%; Table 2) versus uninfected trees (R2 = 92%; Table 2). Adams (1989) and Adams et al. (1996) also found a larger variability in predicting infected fusiform rust incidence versus an uninfected incidence. The survival prediction equations (5) significantly (P < 0.05) over¬ estimated the total number of surviving trees by 3.25 % (95 % confidence interval for overall model bias = 3.25% ( 1.19%) across the range of observed stand densities. The simultaneous /-test also revealed that the total estimated number of surviving trees per hectare was significantly different (P < 0.0001; /-statistic = 33) from the total observed number of surviving trees per hectare (Figure 1). The total number of surviving trees per hectare is over-estimated to a greater magnitude for densities > 1000 trees per hectare than at densities < 1000 trees per hectare (Figure 1). This result is not unexpected considering that fewer, high- density plots were available for model fitting. However, this bias is not a practical concern because tree densities in operational east Texas slash pine plantations typically do not exceed 1000 trees per hectare. In this study, the null hypothesis, Ho2: RNTB is constant across plantation age, would have been rejected at the a = 0.05 probability level (P — 0.0117). This implies that RNTB is not strongly disasso¬ ciated from plantation age, which may be a problem because the solution to the differential equation does not follow as stated in this study if RNTB is a function of age (note that S*RNTB was used in [5], and it was not significantly [P = 0.0706] associated with age). We did not find a similar result when this survival model was fit to data for loblolly pine plantations in east Texas (Lee & Coble 2002). No clear explana¬ tion can be provided as to the different results. One possible explana¬ tion could be that slash pine was more likely to become infected and die 334 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 Figure 1 . Comparison between observed and predicted total surviving slash pine trees per hectare (TPH). than loblolly pine. So, more growing space might have been available to the non-planted trees as the plantation aged, thereby increasing RNTB as time increased. Another possible explanation could be that a larger dataset was available to fit the loblolly survival equations, thereby better capturing the effects of non-planted tree competition on planted pine survival. In any case, a survival model that incorporates non-planted tree competition as a function of age would be ideal. Illustrations of Survival Projections The predicted numbers of surviving slash pine trees (both infected and uninfected) decreases as the percent of non-planted tree basal area increases (Figure 2). In Figure 2, the percent of non-planted tree basal area to total basal area per hectare ranges from 10% to 60%, site index = 21 meters, and stem fusiform rust incidence at year 5 = 10%. The total number of survivors can also be divided into the number of slash pine trees infected or uninfected by fusiform rust (Figure 3). In Figure 3, the numbers of uninfected and infected slash pine trees are displayed for the 15% of non-planted tree basal area to total basal area per hectare (RNTB = 0.15), site index = 21 meters, and stem fusiform rust incidence at year 5 = 10%. The predicted numbers of surviving slash pine trees (both infected and uninfected) also decrease as site index increases (Figure 4). In Figure 4, site index ranges from 15 to 30 meters, the ratio of non-planted tree LEE & COBLE 335 1200 5 200 RNTB - 0.1 (10%) RNTB - 0.2 (20%) RNTB - 0.3 (30%) • RNTB - 0.4 (40%) RNTB -0.5 (50%) RNTB -0.6 (60%) 10 15 20 Plantation Age (years) Figure 2. Predicted surviving planted slash pine trees by plantation age and the percent of non-planted tree basal area to total basal area per hectare classes. Site index = 21 meters (base age = 25 years); stem fusiform rust incidence = 10% at 5 years. Figure 3. Numbers of surviving slash pine trees infected and not infected by fusiform rust by plantation age. The ratio of non-planted tree basal area to total basal area per hectare = 15%; site index = 21 meters (base age = 25 years); stem fusiform rust incidence 10% at 5 years. basal area to total basal area per hectare was 15% (RNTB = 0.15), and stem fusiform rust incidence at year 5 was 10%. These results corrobo¬ rate those of Adams (1989) and Adams et al. (1996). As explained 336 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 4, 2002 Figure 4. Number of surviving slash pine trees, infected and not infected by fusiform rust, by plantation age and site index in meters (base age = 25 years). The ratio of non-planted tree basal area to total basal area per hectare = 15%; stem fusiform rust incidence = 10% at 5 years. earlier, mortality occurs at a faster rate on more productive sites. Thus, more productive sites have fewer trees at a given age than less pro¬ ductive sites. Conclusions The results of this study show that the number of surviving slash pine trees both infected and uninfected by fusiform rust can be accurately predicted for a range of site qualities and levels of non-planted tree competition. The survival model (equation 5) depicts the decreasing number of surviving slash pine as non-planted tree competition and fusiform rust incidence increases. Though the model significantly over-estimates the total number of surviving trees at densities > 1000 trees per hectare, this is of no practical concern since operational slash plantations in east Texas typically do not exceed 1000 trees per hectare. Management activities that reduce the number of non-planted trees early in the life of the plantation are beneficial to increasing the survival of the planted slash pine trees. This reduction is also an important considera¬ tion on the sites with higher productivity since the total number of surviving trees decreases as site index increases. LEE & COBLE 337 Acknowledgements The authors are indebted to J. David Lenhart for his foresight in creating the East Texas Pine Plantation Project. The authors also thank two reviewers for their helpful suggestions. The authors also wish to thank the forest product companies in the ETPPRP: International Paper Company, Louisiana- Pacific Corporation, Resource Management Serv¬ ices, Inc. and Temple-Inland Forest Products Corporation. The authors also thank the Arthur Temple College of Forestry, Stephen F. Austin State University for its continued support of the ETPPRP. Literature Cited Adams, D. E. 1989. A whole stand survival model from a system of differential equations for pine plantations infested with fusiform rust. MS thesis. University of Georgia, 64 pp. Adams, D. E., J. D. Lenhart, A. B. Vaughn & J. Lapongan. 1996. Estimating survival of east Texas loblolly and slash pine plantations infected with fusiform rust. South. J. Appl. For., 20( 1 ) : 30-35 . Arabatzis, A. A., T. G. Gregoire & J. D. Lenhart. 1991. Fusiform rust incidence in loblolly and slash pine plantations in east Texas. South. J. Appl. For., 15(2) : 79-84. Bailey, R. L., B. E. Borders, K. D. Ware & E. P. Jones, Jr. 1985. A Compatible model relating slash pine plantation survival to density, age, site index and type and intensity of thinning. For. Sci., 31(1): 180-189. Borders, B. E. 1989. Systems of equations in forest stand modeling. For. Sci., 35(2):548-556. Burkhart, H. E. & P. T. Sprinz. 1984. A model for assessing hardwood competition effects on yields of loblolly pine plantations. Publication No. FWS-3-84. VPI&SU, 55 pp. Burkhart, H. E., K. D. Farrar, R. L. Amateis & R. F. Daniels. 1987. Simulation of individual tree growth and stand development in loblolly pine plantations on cutover, site-prepared areas. Publication No. FWS-1-82. VPI&SU, 62 pp. Clutter, J. L. & E. P. Jones. 1980. Prediction of growth after thinning in old-field slash pine plantations. USDA For. Serv. Research Pap. SE-217, 14 pp. Clutter, J. L., W. R. Harms, G. H. Brister & J. W. Rheney. 1984. Stand structure and yields of site-prepared loblolly pine plantations in the lower coastal plain of the Carolinas, Georgia, and north Florida. USDA, For. Serv. Southern Forest Experiment Station. Asheville, NC. Gen. Tech. Rep. SE-27, 173 pp. Devine, O. J. & J. L. Clutter. 1985. Prediction of survival in slash pine plantations infected with fusiform rust. For. Sci., 31(l):88-94. Fortson, J. C., B. D. Barry & L. Shackelford. 1996. Removal of competing vegetation from established loblolly pine plantations increases growth on Piedmont and Upper Coastal plain sites. South. J. Appl. For., 20(4): 188-192. Glover, G. R. & B. R. Zutter. 1993. Loblolly pine and mixed hardwood stand dynamics for 27 years following chemical, mechanical, and manual site preparation. Can. J. For. Res., 23(10):329-334. Gribko, L. S. & H. V. Wiant, Jr. 1992. A SAS template program for the accuracy test. The Compiler, 10(1):48 - 51. Grossman, S. I. & W. R. Derrick. 1988. Advanced Engineering Mathematics. Harper Collins Publishers, Inc. New York, 1089 pp. Haywood, J. D. & A. E. Tiarks. 1990. Eleventh year results of fertilization, herbaceous, 338 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 and woody control in a loblolly pine plantation. South. J. Appl. For., 14(4): 173-177. Lee, Young-Jin. 1998. Yield prediction models for unthinned loblolly and slash pine plantations in east Texas. Ph.D. Dissertation. Stephen F. Austin State Univ., 152 pp. Lee, Young-Jin & D. W. Coble. 2002. A survival model for unthinned loblolly pine plantations that incorporates non-planted tree competition, site quality, and incidence of fusiform rust. Bioresource Technology, 85(3):301-308. Lenhart, J. D. 1972. Predicting survival of unthinned old-field loblolly pine plantations. J. Forestry, 70(12):754-755. Lenhart, J. D., E. V. Hunt, Jr. & J. A. Blackard. 1985. Establishment of permanent growth and yield plots in loblolly and slash pine plantations in east Texas. Pp. 436 - 437, Proc. Third Bienn. South. Silvic. Res. Conf. (E. Shoulders, ed.), USDA For. Serv. Gen. Tech. Rep. SO-54, 589 pp. Lenhart, J. D., T. G. Gregoire, G. D. Kronrad & A. G. Holley. 1994. Characterizing fusiform rust incidence and distribution in east Texas. South. J. Appl. For., 18(l):29-34. Neter, J., W. Wasserman & M. H. Kutner. 1985. Applied Linear Statistical Models, 2nd edition. R.D. Irwin, Inc. Homewood, Illinois, 1127 pp. Oliver, C. D. & B. C. Larson. 1996. Forest Stand Dynamics. John Wiley and Sons, Inc., New York, New York, 520 pp. Rauscher, H. M. 1986. Testing prediction accuracy. USDA For. Serv. Gen. Tech. Rep. NC-107, 19 pp. Reynolds, M. R. 1984. Estimating the error in model prediction. For. Sci., 30(2): 454-469. SAS. 1985. S AS Users Guide ETS. Version 5 ed. SAS Institute Inc., Cary, NC. Shapiro, A. 1946. The kinetics of growth and mutation in bacteria. Pp. 228-235, in Symp. Quant. Biol. Cold Spring Harbor Laboratory, New York, 314 pp. Shiver, B. D., J. W. Rheney & M. J. Oppenheimer. 1990. Site-preparation method and early cultural treatment affect growth of flatwoods slash pine plantations. South. J. Appl. For., 14(4): 183-188. Somers, G. L., R. G. Oderwald, W. R. Harms & G. O. Langdon. 1980. Predicting mortality with a Weibull distribution. For. Sci., 26(2): 29 1-300. Stewart, R. E., L. L. Gross & B. H. Honkala. 1984. Effects of competing vegetation on forest trees - a bibliography with abstract. USDA For. Serv. GTR. WO-43, 212 pp. Zar, J. H. 1999. Biostatistical Analysis, 4th ed. Prentice-Hall, Inc., New Jersey, 663 pp. DWC at: dcoble@sfasu.edu TEXAS J. SCI. 54(4):339-346 NOVEMBER, 2002 MASS CAPTURE OF INSECTS BY THE PITCHER PLANT SARRACENIA ALATA (SARRACENIACEAE) IN SOUTHWEST LOUISIANA AND SOUTHEAST TEXAS Robert E. Evans, Barbara R. MacRoberts, Thomas C. Gibson and Michael H. MacRoberts NatureServe , 6114 Fayetteville Road, Durham, North Carolina 27713; Museum of Life Sciences, Louisiana State University, Shreveport, Louisiana 71115; Department of Botany, University of Wisconsin, Madison, Wisconsin 53706 and Bog Research, Shreveport, Louisiana 71104; Abstract. — This study documents the mass capture by Sarracenia alata (pitcher plant) of Plecia nearctica (lovebug) during an emergence that occurred in southwest Louisiana and southeast Texas during the fall of 1996. An estimated 7700 pitcher plants in a 0.4 ha bog area were found to be capable of harvesting approximately two million specimens of P. nearctica. Several proposals relative to the understanding of mass insect captures by pitcher plants are also discussed. Published observations on mass capture of insects by carnivorous plants are rare. Oliver’s (1944) account of a 191 1 emergence of cabbage white butterflies (Pieris raped), in which six million butterflies were caught on a two acre patch of sundews (Drosera anglica Huds.) in only nine hours, is a spectacular example of the episodic nature of insect capture by a carnivorous plant species. Although Juniper et al. (1989) comment that all carnivorous plants, particularly those living near patches of open water or on the migration routes of insects, are likely to experience episodic arrivals of one or a few species, documented observations of carnivorous plant and episodic insect interactions are rare; indeed, studies of insect capture processes in carnivorous plants are rare. Studies of normal capture and feeding rates for carnivorous species show wide variations. Some species appear to capture very little, which, along with laboratory experiments where carnivorous species are grown without supplemental food, suggests that insects may not be required for carnivorous plant growth and reproduction (Adamec 1997). For example, Newell & Nastase (1998) found that New Jersey popula¬ tions of Sarracenia purpurea L. captured 2.1% of potential prey, 340 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 4, 2002 resulting in only about one insect every 15 days. Heard (1998) found that S. purpurea in Newfoundland captured a paltry 1 1 mg (range 0-67 mg) dry mass over its lifetime. In another study, Dixon et al. (1980) found that Drosera erythrorhiza Lindl. in Australia averaged only 2.5 organisms per plant in one year and 6.9 in another. However, others have found much higher capture rates and have found that captured insects contribute significantly to growth and reproduction (Gibson 1991; Adamec 1997). Gibson (1983:105, Appendix 3), for example, reported much higher capture rates for such carnivorous plants as Sarracenia leucophylla Raf. and Drosera in the southeastern United States. He found up to ten prey captured per day and in S. leucophylla about 0.50 g dry insect biomass per trap during a season. Folkerts (1992) found the mean dry weight of arthropod prey captured in Sarracenia alata Wood study plots was between 5 and 60 mg; she also briefly mentioned Bibionid (Diptera: Bibionidae) outbreaks but that rarely were pitchers filled to capacity. On the other hand, Folkerts & Folkerts (1995:7) stated in a brief description of S. leucophylla : "When plagues of lovebugs ... occur, the pitchers fill to the brim, some containing more than 2,000 love bugs." The overall conclusion is that, while carnivorous species lie along a gradient from near independence to almost total dependence on prey for their growth, for reproduction and spread in natural habitats, as opposed to greenhouse settings, prey are necessary (Givnish 1989; Juniper et al. 1989; Adamec 1997). Each September, mating flights of lovebugs (Diptera: Plecia nearctica Hardy) occur throughout the southeast (Hetrick 1970; Denmark & Mead 1992). These flights begin with swarms of males and eventually consist largely of copulating pairs; the females are gravid (Hetrick 1970; Denmark & Mead 1992). The intensity of flights varies annually, with an occasional emergence (large magnitude flights) taking place. Emer¬ gences are spectacular events of short duration, which attract much local attention; many thousands of individuals may be smashed by vehicles traveling only a few miles. Between 1991 and 1998 two large emergences were observed and prompted these observations made in September 1996 on mass capture in S . alata in southeastern Texas and southwestern Louisiana where pitcher plants are common (MacRoberts & MacRoberts 2001). EVANS ET AL. 341 Study Sites Observations were made on rate of capture and distribution of the phenomenon in two pitcher plant bogs on the Angelina National Forest, Angelina and Jasper counties, Texas, and in one bog on the Vernon Ranger District of the Kisatchie National Forest in Vernon Parish, Louisiana. General observations on the emergence of lovebugs were made across southeast Texas and southwest Louisiana. Materials and Methods Late summer pitchers over 20 cm tall were counted in a 0.4 ha hill¬ side bog in east Texas. The bog had approximately 7700 pitchers all of which were full of insects. In addition, the insect contents of 35 pitchers were extracted and counted, and from these data lovebug capture rates were extrapolated. A sample of 1000 lovebugs was collected, which had a dry weight of 2.4 g. Since lovebugs are poten¬ tially important sources of minerals, including several that have been considered limiting in carnivorous plant habitats (Gibson 1983), love¬ bugs and pitcher plant leaves were chemically analyzed by A & L Laboratories, Memphis, Tennessee. Results Plants of Sarracenia alata in hillside bog communities produce new traps in both the spring and late summer (MacRoberts & MacRoberts 2001). In September 1996, new summer pitchers quickly filled with lovebugs; whereas old spring pitchers remained empty. In each of several bogs examined during this period, all new pitchers over approxi¬ mately 20 cm in height were filled almost to the pitcher opening. Approximately 2 to 4 cm of the uppermost lovebugs were alive; most were trapped, except those on the upper surface layer that were able to crawl freely in and out of the pitchers. Slightly farther down the pitcher, lovebugs were dead, in various stages of digestion and decom¬ position. For the 35 pitchers examined for insect contents, with the exception of an occasional moth, pitchers were invariably filled with lovebugs (Fig. 1). The largest pitchers, measuring up to 70 cm in height with 5 342 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 4, 2002 Figure 1 . Number of lovebugs caught as a function of pitcher plant height. cm diameter openings, had captured almost 600 lovebugs. Smaller pitchers (20 cm or less) trapped primarily ants, but these were not filled; only one of 12 pitchers in this size class had captured lovebugs. At observed capture rates of approximately 250 lovebugs/pitcher, this bog was harvesting approximately two million lovebugs. The largest pitchers in our sample were capturing approximately 1.4 g (dry weight) of lovebugs in a short period, and the bog community was capturing between 4 and 5 kg of material (dry weight). Chemical analysis of lovebugs and pitcher plant leaves is shown in Table 1. Although gorging appears to provide access to significant nutrient supplies, mass capture of this magnitude may have some drawbacks. Filled pitchers showed signs of rotting. In a sample of 23 filled pitcher plants over 20 cm tall, leaf tissue on 21 was rotting. Rot generally EVANS ET AL. 343 Table 1 . Element composition in lovebugs and plant leaf. Element Lovebugs Plant leaf Nitrogen 13.04% 2.24% Sulphur 0.94% 0.16% Phosphorus 1.20% 0.24% Potassium 1.36% 0.60% Magnesium 0.22% 0.20% Calcium 0.31% 0.10% Sodium 0.24% 0.10% Boron 8 ppm 7 ppm Zinc 293 ppm 29 ppm Manganese 28 ppm 19 ppm Iron 263 ppm 70 ppm Copper 31 ppm 10 ppm Aluminum 1 15 ppm 43 ppm began opposite the ala of the pitcher between 60 and 90 percent of the pitcher height (Figure 2). Neither small pitchers (lacking lovebugs) nor spring pitchers appeared to have rotted at all, and fall pitchers in non-emergent years do not show rot. Similar observations of rotting due to gorging have been made for S. leucophylla by the third author (T.C.G.) in Florida on plants filled with lovebugs. As a result of gorging and then rotting, the plant may be losing photosynthetic ability because of tissue loss. Discussion It has been hypothesized that insect resources are particularly important because they allow plants to accumulate nutrient reserves that permit greater flowering, increased seed production, and recovery from disturbance (Gibson 1983; 1991). If it is assumed that normal feeding provides these reserves, as most studies of carnivory suggest, and the chemical analysis presented here suggests, then episodic capture may provide an additional accumulation. 344 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 4, 2002 Figure 2. Rot in pitcher plant. Emergences such as those described here far exceed normal insect numbers, and consequently it is likely that pitchers would catch more of these species than other less common insects. The question is: Are the emergences of lovebugs a repeating cycle to which pitcher plants have adapted? On the other hand, it is also possible that since lovebugs are an expanding species that has only recently been noted to have massive emergences (Hetrick 1970; Denmark & Mead 1992), they may be co- evolutionarily naive to pitcher plants; hence, there has not yet been selection for avoidance of the traps. In support of this hypothesis, the third author (T.C.G.) has found that fire ants, new to Florida, some¬ times get trapped in large numbers. However, it is also possible that lovebugs have pheromones that attract other lovebugs and that once a few are trapped in a pitcher, more and more become attracted to it by EVANS ET AL. 345 the sexual odor. Capture may, therefore, have nothing to do with naivete. Although this type of episodic capture has been witnessed in many places, it is very poorly reported. The present findings supplement those of Juniper et al. (1989) by expanding the range of episodic interactions; the present observations were not of migratory insect flights and the sites were not adjacent to open water. Clearly, further observa¬ tions, even anecdotal ones, would be welcome. Detailed study of such events, however, may be hampered by their unpredictability, infre¬ quency and short duration. Acknowledgments Guy Nesom, Botanical Research Institute of Texas, Fort Worth, Texas; Philip Sheridan, Meadowview Biological Research Station, Woodford, Virginia; Robert R. Fleet, Stephen F. Austin State University, Nacogdoches, Texas and an anonymous reviewer made comments on an earlier version of this paper. Robert Kal insky, Louisiana State University in Shreveport, aided with Figure 1 . Literature Cited Adamec, L. 1997. Mineral nutrition of carnivorous plants: a review. Bot. Rev., 63:273-289. Denmark, H. A. & F.W. Mead. 1992. Lovebug, Plecia nearatica Hardy (Diptera: Bibionidae). Entomology Circular 350, Florida Department of Agriculture and Consumer Services. Gainesville, Florida. Dixon, K. W., J. S. Pate & W. J. Bailey. 1980. Nitrogen nutrition of the tuberous sundew Drosera erythrorhiza Lindl. with special reference to catch of arthropod fauna by its glandular leaves. Austr. J. Bot., 28:283-297. Folkerts, D. R. 1992. Interactions of pitcher plants {Sarracenia: Sarraceniaceae) with their arthropod prey in the southeastern United States. Ph.D. dissertation, University of Georgia, Athens, 214 pp. Folkerts, G. W. & D. R. Folkerts. 1995. Carnivorous plants of Conecuh National Forest. U.S.D.A. Forest Service, Forestry Report R8-FR 49, Washington D.C., 24 pp. Gibson, T. C. 1983. Competition, disturbance, and the carnivorous plant community in the southeastern United States. Ph.D. Dissertation, University of Utah, 260 pp. Gibson, T. C. 1991. Competition among threadleaf sundews for limited insect resources. Amer. Nat., 138:785-789. Givnish, T. J. 1989. Ecology and evolution of carnivorous plants. Pp. 243-290, in Plant-animal interactions (W.G. Abrahamson, ed.). McGraw Hill, New York, 480 pp. Heard, S. B. 1998. Capture rates of invertebrate prey by the pitcher plant, Sarracenia purpurea L. Amer. Midi. Nat., 139:79-89. 346 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 4, 2002 Hertick, L. A. 1970. Biology of the "love-bug", Plecia neartica (Diptera: Bibionidae). Florida Entomologist, 53:23-26. Juniper, B. E., R. J. Robins & D. M. Joel. 1989. The carnivorous plants. Academic Press, London, 353 pp. MacRoberts, M. H. & B. R. MacRoberts. 2001. Bog communities of the West Gulf Coastal Plain: a profile. Bog Research Papers in Botany and Ecology, 1:1-151. Newell, S. J. & A. J. Nastase. 1998. Efficiency of insect capture by Sarracenia purpurea (Sarraceniaceae), the northern pitcher plant. Amer. J. Bot., 85:88-91. Oliver, F. W. 1944. A mass catch of cabbage whites by sundews. Proc. Royal Entomol. Soc. Lond. Series A. 19:5. MHM at: bogresch@softdisk.com TEXAS J. SCI. 54(4):347-356 NOVEMBER, 2002 FEMALE REPRODUCTION IN THE WESTERN DIAMOND-BACKED RATTLESNAKE, CROTALUS ATROX (SERPENTES: VIPERIDAE), FROM ARIZONA Philip C. Rosen and Stephen R. Goldberg School of Renewable Natural Resources, University of Arizona Tucson, Arizona 85721 and Department of Biology, Whittier College Whittier, California 90608 Abstract.— Reproductive tissue was examined from 46 sexually mature female Crotalus atrox specimens from Arizona, and additional data were obtained from gravid females held in the laboratory and from palpation of live snakes in the field. Mean litter size for 19 females was 8.3 ± 2.6 SD range = 5-15 based on enlarged (> 12 mm) ovarian follicles, but was significantly less in 10 records of oviductal ova or whole litters, at 5.6 ± 1.5 SD, range = 4-9, with an overall mean of 7.3 ± 2.6 for adult females averaging 797 mm SVL, range = 648-1010. These values are lower than estimates in the literature from elsewhere, but seem consistent with the smaller observed size of adult females in the desert region where the samples originated. The number of enlarged follicles correlated positively with female SVL, but oviductal eggs and neonates did not. Four values for RCM ([clutch mass] / [gravid female mass]) averaged 0.303 ± 0.099 SD, range = 0.222 - 0.438. Twenty-one neonates born to wild-caught females averaged 319.7 mm total length and 19.6 g body mass. Enlarged ovarian follicles (> 12 mm length) were found February-June (ovulation during present year) and September-November (ovulation next year), and young are born in late July to September. The earliest field dates for appearance of neonates ( n = 75 observed) were 8-11 August. Yolk deposition is normally completed over two activity seasons, although there was some indication that during times of great abundance for rodent prey, individual females might reproduce in successive years. On average, approximately half of all females reproduce in a given year, but significantly higher proportions (73%) were gravid in years of higher compared to lower rodent abundance (28% gravid). The western diamond-backed rattlesnake, Crotalus atrox , ranges from southeast California to Arkansas and eastern Texas, Arizona, New Mexico and Oklahoma, south to northern Sinaloa and San Luis Potosi, Mexico (Stebbins 1985). There is little information available on the seasonal ovarian cycle of this species. Reports on female C. atrox reproduction are in Tinkle (1962), Klauber (1972), Tennant (1984), Fitch (1985), Lowe et al. (1986), Ernst (1992), Fitch & Pisani (1993), Price (1998) and Werler & Dixon (2000). The purpose of this paper is to provide information on the ovarian cycle of C. atrox from Arizona and to present the first seasonal reproductive data for females of this species. These data will be useful for making comparisons of the reproductive output of Arizona C. atrox with populations elsewhere in 348 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 its geographic range. Materials and Methods Forty- three female C. atrox from Arizona were examined from the herpetology collection of the University of Arizona, Tucson (UAZ) and three additional dissections were performed in the field ( n = 46, mean Snout-Vent Length, SVL = 792 mm ± 86 SD, range = 636-1010 mm). Snakes were collected 1949 to 1998. The left ovary was removed for histological examination. Histological sections of ovary were examined for the presence of yolk deposition (secondary vitellogenesis sensu Aldridge (1979). No histology was done on enlarged follicles (> 12 mm length) or oviductal eggs, both of which were counted. In addition, in the course of field studies, five adult females found gravid in the field during July- August were held until parturition, and weights and lengths of the neonates were recorded prior to their release at study areas. During field work between Tucson and Organ Pipe Cactus National Monument (ORPI), Pima County, 59 adult females were palpated during May-July, when reproductive status (gravid versus non-gravid) for the year could be determined, and counts of ovarian or oviductal ova were made on five of these. The relationship between SVL and litter size was investigated by regression analysis. T tests were used to compare mean litter sizes, and a corrected chi-square test to compare gravid functions among yearly samples. Material examined.— The following specimens of Arizona C. atrox females were examined: MARICOPA COUNTY, 2 specimens (UAZ 46426, 50737); MOHAVE COUNTY, 2 specimens (UAZ 27112, 37052); PIMA COUNTY, 31 specimens (UAZ 13556, 13564, 13565,27093,27102, 27120, 27165, 27170, 27172, 27179, 27180, 27209, 27218, 27238, 27293, 27294, 27298, 27318, 27320, 27323, 27333, 42486, 44076, 44939, 46412, 47966, 48853, 48854, 51488, 54048, 54049); PINAL COUNTY, 5 specimens (UAZ 27094, 27162, 27291, 27732, 27334); SANTA CRUZ COUNTY, 1 specimen (UAZ 27121); YUMA COUNTY, 2 specimens (UAZ 27215, 35868). Results and Discussion Females with enlarged follicles (> 12 mm length) or oviductal eggs were found February-June and September-November (Table 1). Those females from February-June would likely have ovulated and produced ROSEN & GOLDBERG 349 Table 1 . Monthly distribution of conditions in seasonal ovarian cycle of 46 Crotalus atrox. Values shown are the number of females exhibiting each of the four conditions. Month n Inactive Early yolk deposition Enlarged follicles > 12 mm length Oviductal eggs February 1 0 0 1 0 March 3 2 0 1 0 April 7 0 1 6 0 May 11 3 0 7 a 1 June 4 1 1 2 0 July 3 3 0 0 0 August 3 3 0 0 0 September 6 5 0 1 0 October 4 2 0 2 0 November 4 1 0 3 0 a Includes one female with enlarged squashed follicles that could not be counted. young later that year. The eggs in the only oviductal female (UAZ 27170) found in a museum collection specimen, collected 21 May were squashed making it impossible to count them. The females with enlarged eggs from September-November would likely have ovulated the next spring and given birth later that summer indicating that yolk deposition is completed over two activity seasons. This also occurs in other North American rattlesnakes (Goldberg 1999a; 1999b; 1999c; 2000a). The smallest reproductively active female (UAZ 27320) (enlarged follicles > 12 mm) measured 648 mm SVL, and the mean size of gravid females in Table 2 was 797 + 75 mm SVL. The mean litter size based on ovarian follicles (> 12 mm) for 19 C. atrox females from Arizona was 8.3 ± 2.6 SD, range = 5-15. This value may be higher than what actually occurs since not all enlarged follicles may complete development. Indeed, mean litter size in 10 females based on oviductal eggs or young born was significantly lower ( t — 3.36, df = 24, P < 0.01), at a mean of 5.6 + 1.5 SD , range = 4-9. Overall mean litter size was 7.3 ± 2.6 for adult females averaging 797 mm SVL, range = 648-1010 mm. Relative clutch mass (RCM) for four litters averaged 350 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 Table 2. Litter sizes for 29 Crotalus atrox females from Arizona. All museum specimens represent counts of enlarged follicles > 12 mm length in dissected snakes. Snakes that were palpated or gave birth in captivity were released as part of a marking study. Those held for live-bearing were captured in July or August. Date SVL (mm) Litter size County Source 22 February 1965 866 9 Pima UAZ 27120 28 March 1971 863 11 Pima UAZ 51488 1 April 1967 727 9 Pinal UAZ 27162 1 April 1967 770 9 Pima UAZ 27293 1 April 1967 807 7 Pinal UAZ 27291 20 April 1985 685 8 Maricopa UAZ 46426 20 April 1985 763 6 Maricopa UAZ 50737 23 April 1993 724 6 Pima palpation 2 May 1964 773 7 Pima UAZ 13564 2 May 1964 790 7 Pima UAZ 13565 6 May 1964 648 6 Pima UAZ 27320 9 May 1965 896 9 Pinal UAZ 27732 13 May 1964 1010 13 Pinal UAZ 27334 4 June 1992 730 9 Pima palpation 7 June 1995 812 5 Pima UAZ 54048 13 June 1973 882 7 Yuma UAZ 35868 1 July 1989 819 5 a Pima palpation 15 August 1994 817 5 Pima palpation 21 August 1998 812 4 Pima palpation 27 August 1987 732 5 Yavapai live birth 28 August 1993 no data 6 Pima live birth 3 September 1973 780 6 Mohave UAZ 37052 10 September 1997 747 4 b Pima UAZ 54049 14 September 1990 801 6 Pima live birth 29 September 1989 792 5 Pima live birth ROSEN & GOLDBERG 351 Table 2 cont. Date SVL (mm) Litter size County Source 1 October 1966 770 15 Pima UAZ 27323 26 October 1970 826 8 Pima UAZ 46412 15 November 1990 753 6 Pima UAZ 48854 16 November 1966 920 10 Pima UAZ 27093 a Produced infertile eggs, 3 August 1989 b Postpartum; 4 large corpora lutea and vascularized oviductal sites 0.303 ± 0.099 SD, range = 0.222 - 0.438 based upon (offspring mass) /(gravid female mass), and was 0.484 ± 0.167 SD, range = 0.370 - 0.637 based upon (offspring mass)/(postpartum female mass). These values are close to the mean value for other viviparous snakes, including viperids, given by Seigel & Fitch (1984). The litter size value based on enlarged ovarian follicles is within the range, but slightly lower than in other reports: 6-19 in Tinkle (1962), 2-24 in Lowe et al (1986), 6-25 in Armstrong & Murphy (1979), 4-24 in Fitch & Pisani (1993), and 4-24 in Klauber (1972) for C. atrox. Estimates of litter size in this study were significantly lower than that based on Klauber’ s (1972) data (10.2 ± 5.3, n = 36), whether based on ovarian follicle results ( t = 1 .78, df = 53, P < 0.05) or on all data ( t = 2.78, df = 63, P < 0.004). Klauber’s (1972) sample may have been heterogeneous, and Fitch & Pisani’ s (1993) estimate based on enlarged ovarian follicles from an Oklahoma sample was even higher at 13 + 4.4. The mean size of reproductive females in the current study (797 mm SVL) is lower than the minimum SVL for maturity of about 830 mm in Fitch & Pisani (1993) and 800 mm in Tinkle (1962), although Klauber (1937) reported a considerably lower minimum size of 742 mm (apparently total length) for gravid female C. atrox. However, this estimated litter size is more or less consistent based on extrapolation to the smaller size, with the size-fecundity graph for Oklahoma (Fig. 6 in Fitch & Pisani 1993). It is possible that during this C. atrox study in the Sonoran Desert, Pima County, realized litter size was lower than potential litter size that would be inferred based on enlarging ovarian follicles. Female C. atrox 352 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 at ORPI at times appeared highly stressed (underweight, weak and not robust), particularly when reproduction coincided with drought. Among the captive females, one litter of five consisted of undeveloped eggs, and 1 of 21 other births was a partially developed embryo with yolk attached. Births were recorded in the laboratory on 3 August (infertile eggs), and 27 August, 28 August, 14 September and 29 September. Captivity may have delayed birth somewhat. Recently postpartum females were found at ORPI on 12 August- 18 September, and the latest capture of a gravid female was 24 August. Copulation was observed in the field on 25 August 2001 at the old Esmond Railroad Station near Tucson and at a den a few miles east of there on 12 March 1997. Neonates born to wild-caught females ( n =21; 10 males, 11 females) averaged 319.7 ± 14.8 SD mm total length, range = 294-345, and 19.55 ± 2.81 SD g total body mass, range 16.9-24.7. Neonate males and females were closely similar in mass and total length, although in females the tail was shorter (7.6 % of SVL) than in males (10.8 % of SVL). The earliest appearances of neonates in the field ( n = 75 observed) were 8 August in the grassland of southeastern Arizona, 10 August near Tucson and 1 1 August at ORPI; there is some indication that birth, or at least the appearance of young, is later at ORPI than in southeastern Arizona (Fig 1). Since neonatal rattlesnakes, including C. atrox appear to remain aggregated and with their mother for about 10 days (see Price 1988), their detection in the field portrayed in Fig. 1 is probably delayed relative to birth. Thus, these results generally support a late July to early or mid-September parturition period for C. atrox in southern Arizona. Regression analysis revealed a significant positive correlation between In (Litter Size) and In (SVL) for 28 litters of C. atrox in Table 2: (In (Litter Size) = -6.41 + 1.25 In (SVL); r = 0.12; P = 0.05. Back transformed this regression equation describes the allometric relationship between the variables via a power function: Litter Size = e'6 41 SVL1 25 . However, the entire positive relationship was based on counts of enlarged ovarian follicles; no relationship was found between female SVL and litter size based on embryos or live-born litters (Fig. 2). The presence of inactive females, comprising nearly half (48%) of the dissection sample, shows that not all females reproduce each year. This ROSEN & GOLDBERG 353 1-15 16-31 1-15 16-30 1-15 16-31 1-15 Aug. Aug. Sept. Sept. Oct. Oct. Nov. Figure 1 . Seasonal distribution of observed young of the year with a single rattle segment (YoY) for Crotalus atrox in southern Arizona, 1985 - 2001. Field work was carried out throughout March to early November at or near Organ Pipe Cactus National Monument in western Pima County and near Tucson, primarily in Arizona Upland Sonoran desertscrub, and in the desert grassland of Sulphur Springs, San Bernardino, and Altar valleys of southeastern Arizona. By mid-October, most YoY had 2 or 3 rattle segments, and were not included in this histogram. is also born out in the palpation results (Table 3), in which 51 % of the sample was found to be non-gravid during the gravid season. Biennial, triennial or less frequent reproduction appears to be common among species of North American rattlesnakes (Goldberg 1999a; 1999b; 2000b; Goldberg & Holycross 1999; Holycross & Goldberg 2001). Tinkle (1962) reported female C. atrox from Texas to reproduce biennially, although Fitch & Pisani (1993) found C. atrox from Oklahoma to reproduce annually. Werler & Dixon (2000) reported that C. atrox from northern Texas where the winters are severe reproduce biannually, whereas those living in the warmer southern part of the state reproduce annually. Goldberg & Rosen (2000) reported that yearly percentages of gravid Crotalus scutulatus appeared related to food abundance. This trend, first reported for C. scutulatus , was also found to be highly significant for C. atrox in southern Arizona. In years in which rodent monitoring indicated low food availability for the snakes (1989-90, 354 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 18 14 Q) N 55 10 o 6 - Enlarged Ovarian Follicles: Litter = - 3.9 + 0.01 5(SVL) R = 0.50, n = 19, P < 0.03 _ ... - -Or O '6 ' O o o o o • _o oo o 9 9 09 9 9 Oviductal Eggs, Neonates: n = 9, n.s. 600 700 800 900 1000 SVL (mm) Figure 2. Linear regressions of number of enlarged ovarian follicles ( > 12 mm length), and of oviductal eggs and neonates born, on snout- vent length (mm) of the mother for 28 Crotalus atrox females from Arizona. Open circles represent enlarged ovarian follicles; solid circles represent oviductal eggs or neonates. 1996-98), the gravid fraction (28%) was significantly lower than in years of "booming" rodent populations (1991-1995; 73%; corrected chi- square = 8.64, df = 1, P < 0.005). The run of years with high clutch frequency (Table 3) suggests that female C. atrox may sometimes reproduce in successive years in the Sonoran Desert. This hypothesis remains to be demonstrated rigorously, but is consistent with geographic variation in reproductive frequency found in the literature cited above. Fitch (1985) suggested litter sizes of C. atrox increased southward and decreased westward. Results of this current study support decreased litter size in the west, and suggest that various stresses associated with the increasingly arid environment may be contributory. Examinations of C. atrox females from various parts of its range will be required before the degree of variation in the ovarian cycle of this wide-ranging species can be known in detail. While some information on female reproduction can be obtained from examination of museum specimens, additional field studies will be essential to ascertain the frequency of female C. atrox reproduction. ROSEN & GOLDBERG 355 Table 3. Yearly percentages of gravid versus non-gravid female Crotalus atrox captured May-August in the Sonoran Desert of southern Arizona, Pima County, 1987-1998. Year 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 Total n 2 6 3 7 5 9 8 3 1 6 4 5 59 Gravid 2 1 3 1 2 7 7 2 1 1 1 1 29 Non-gravid 0 5 0 6 3 2 1 1 0 5 3 4 30 % Gravid 100 17 100 14 40 78 88 67 100 17 25 20 49.2 Acknowledgments We thank Charles H. Lowe (University of Arizona) for permission to examine C. atrox and George L. Bradley, Peter A. Holm, David A. Parizek Jr., Shawn S. Sartor ius, Elizabeth B. Wirt and many others for important field and laboratory assistance. Literature Cited Aldridge, R. D. 1979. Female reproductive cycles of the snakes Arizona elegans and Crotalus viridis. Herpetologica, 35(3) :256-261 . Armstrong, B. L. & J.B. Murphy. 1979. The natural history of Mexican rattlesnakes. Univ. Kansas Mus. Nat. Hist., Special Publ. No. 5, vii -I- 88 pp. Ernst, C. H. 1992. Venomous reptiles of North America. Smithsonian Institution Press, Washington, ix + 236 pp. Fitch, H. S. 1985. Variation in clutch and litter size in New World reptiles. Univ. Kansas Mus. Nat. Hist. Misc. Publ., 76:1-76. Fitch, H. S. & G. R. Pisani. 1993. Life history traits of the western diamondback rattlesnake ( Crotalus atrox) studied from roundup samples in Oklahoma. Occas. Pap. Mus. Nat. Hist. Univ. Kansas, 156:1-24. Goldberg, S. R. 1999a. Reproduction in the tiger rattlesnake, Crotalus tigris (Serpentes: Viperidae). Texas J. Sci., 5 1 ( 1 ) :3 1 -36. Goldberg, S. R. 1999b. Reproduction in the blacktail rattlesnake, Crotalus molossus (Serpentes: Viperidae). Texas J. Sci., 5 1(4): 323-328. Goldberg, S. R. 1999c. Reproduction in the red diamond rattlesnake in California. Calif. Fish & Game, 85(4): 177-180. Goldberg, S. R. 2000a. Reproduction in the rock rattlesnake, Crotalus lepidus (Serpentes: Viperidae). Herpetological Nat. Hist., 7(l):83-86. Goldberg, S. R. 2000b. Reproduction in the speckled rattlesnake, Crotalus mitchellii (Serpentes: Viperidae). Bull. Southern California Acad. Sci., 99(2): 101-104. Goldberg, S. R. & A. T. Holycross. 1999. Reproduction in the desert massasauga, Sistrurus catenatus edwardsii, in Arizona and Colorado. Southwest. Nat., 44(4):531-535. Goldberg, S. R. & P. C. Rosen. 2000. Reproduction in the Mojave rattlesnake, Crotalus scutulatus (Serpentes: Viperidae). Texas J. Sci., 52(2): 101-109. Holycross, A. T. & S. R. Goldberg. 2001. Reproduction in northern populations of the ridgenose rattlesnake, Crotalus xvillardi (Serpentes: Viperidae). Copeia, 2001(2):473-481 . 356 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 Klauber, L. M. 1937. A statistical study of the rattlesnakes. IV. The growth of the rattlesnake. Occas. Pap. San Diego Soc. Nat. Hist., 3:1-56. Klauber, L. M. 1972. Rattlesnakes. Their habits, life histories, and influence on mankind. 2nd ed. Vol. 1. Univ. California Press, Berkeley, xlvi + 740 pp. Lowe, C. H., C. R. Schwalbe & T. B. Johnson. 1986. The venomous reptiles of Arizona. Arizona Game & Fish Department, Phoenix, ix + 115 pp. Price, A. H. 1988. Observations of maternal behavior and neonate aggregation in the western diamondback rattlesnake, Crotalus atrox (Crotalidae). Southwest. Nat., 33(3):370-373. Price, A. H. 1998. Poisonous snakes of Texas. Texas Parks and Wildlife Press, Austin, 112 pp. Seigel, R, A. & H. S. Fitch. 1984. Ecological patterns of relative clutch mass in snakes. Oecologia, 6 1 (3) :293-30 1 . Stebbins, R. C. 1985. A field guide to western reptiles and amphibians. Houghton Mifflin Company, Boston, xiv + 336 pp. Tennant, A. 1984. The snakes of Texas. Texas Monthly Press, Austin, 561 pp. Tinkle, D. W. 1962. Reproductive potential and cycles in female Crotalus atrox from northwestern Texas. Copeia, 1962(2):306-313. Werler, J. E. & J. R. Dixon. 2000. Texas snakes. Identification, distribution and natural history. University of Texas Press, Austin, xv 4- 437 pp. PCR at: pcrosen@u. arizona.edu TEXAS J. SCI. 54(4):357-362 NOVEMBER, 2002 NEW DISTRIBUTION RECORD AND ECOLOGICAL NOTES OF THE FRESHWATER HYDROZOAN CRASPEDACUSTA SOWERBII IN SOUTHEAST TEXAS Richard C. Harrel Department of Biology, Lamar University Beaumont, Texas 77710 Abstract. — Craspedacusta sowerbii, the only species of freshwater cnidarian exhibiting a medusa stage in it’s life cycle, was collected from an excavated pond in Jefferson County, Texas from 6 July to 12 November 2001. Density was patchy and mean density per collection date varied from 0.59/m3 to 34.43/m3. The highest density from a single plankton tow was 121.32/m3. Five polyps were collected from artificial substrates suspended in the water column, indicating that sexual reproduction had occurred. The medusae disappeared when the temperature decreased to 21 °C even though the zooplankton food supply was still abundant. Craspedacusta sowerbii Lankester (Cnidaria: Hydrozoa) is the only species of freshwater cnidarian in North America exhibiting a medusa stage in it’s life cycle. It has been reported from at least 40 states and over 100 locations (Acker & Muscat 1976; Devries 1992; http://www. iup.edu/-tpeard/jellyfish.htmlx >). However, it’s occurrence is sporadic and each new sighting is considered noteworthy. Previously published Texas records of C. sowerbii were by Cheatum (1934), Schmitt (1939), Jurgens (1957), all from central Texas, and Baker (1960) and McCullough et al. (1981) from east Texas. During July of 2001 a population of C. sowerbii was discovered in Kaiser Pond located in northwest Beaumont, Jefferson County, Texas. It is located about 200 m south of Spurlock Road between Major Drive and Keith Road at latitude 30° 08.34’ north, longitude 90° 12.37’ west. This report documents ecological observations of this population and limnological conditions in Kaiser Pond from 6 July to 12 November 2001 when medusae were present. Description of the Pond Kaiser Pond basin was constructed in 1985 by excavation of Beaumont clay soil. The pond is 48.5 m wide and 88 m long with a surface area of 0.43 ha (1 .2 acres). The margins have a steep slope and most of the pond is about 4 m deep. However, a 7 to 8 m deep trench is located at the south end of the pond. The water level was fairly constant and never varied more than 0.7 m. Two piers extend out into 358 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 4, 2002 the pond. The west pier is 10 m long and the east pier is about 25 m long and was used to collect physical/chemical data and to take plankton tows. The pond receives very little surface runoff due to elevated surface soil deposited during construction surrounding the pond. About 44 m of the east bank has a concrete apron that extends out into the water about 5 or 6 m to a depth of about 2 m. No aquatic vascular plants were present and filamentous algae were absent in the water column, but Oscillatoria coated the concrete apron and the pilings of the piers. The pond has a large fish community that was stocked from various sources. This stocking appears to be the probable means by which C. sowerbii was introduced into the pond. The pond was built for recreation purposes and is frequently visited by the owner throughout the year. This was the first year that the medusae were observed. Materials and Methods Visual observations and measurements of physical/chemical conditions and/or medusa densities were taken on 27 different dates from 6 July to 19 December 2001. Observations of medusae were taken from both piers on each date and while snorkeling on 12 July, 2 August, 4 August and 7 September. Density of the medqsae was determined by towing a No. 20 plankton net, with a 30 cm opening, 10 to 20 m from a depth of about two meters to the surface from the east pier. Some of the water from the plankton net bucket was preserved and used to determine the types of plankton present. Physical/chemical measurements were taken from the end of the east pier where the water depth varied from 3.3 to 4 m. Water temperature, dissolved oxygen concentration, oxygen percent saturation, pH and specific conductivity were measured at meter depth intervals using a Hydrolab Surveyor 3 and H20 Multiprobe. Water depth was determined using the Hydrolab and a hand held Speedtech Depthmate portable sounder. Alkalinity was determined by titration (APHA 1989). Water transparency was determined with a Secchi disk and euphotic depth with a submarine photometer. Artificial substrates of the Parsons & Tatum (1974) design (0.09 m2 surface area) were suspended at depths of 0.5, 1.0 and 1.5 meters and removed after seven weeks and 12 weeks in an attempt to collect the polyp life cycle stage. Results and Discussion Kaiser Pond exhibited no thermal stratification and the maximum difference between the surface and the 4 m depth was 2°C. Water HARREL 359 temperature ranged from 33 to 21 °C when the medusae were present (Table 1). Dissolved oxygen concentrations varied from 9.4 mg/L (133 % saturation) at 3 m to 0.2 mg/L (3.3 % saturation) at 3.8 m. During most collections oxygen concentrations were similar and >60% satura¬ tion at all depths, except at the bottom. The pH ranged from 6.5 to 7.8 and all values <7.0 occurred directly after heavy rains. Surface alkalinity ranged from 109 to 115 mg/L. Specific conductance ranged from 249 to 269 /xS/cm. The Secchi disk depth varied from 1.8 to 3.8 m. The wide variability was due to time of day, cloud cover and surface water movements when the measurements were taken. The euphotic depth varied from 3.3 to >4 m. The pH, alkalinity, specific conductance and water transparency were all higher than most local surface waters. Other ponds in this area constructed on Beaumont clay are always very turbid due to colloidal clay particles. All water quality parameters were within the ranges where C. sowerbii had been reported by other investigators. The medusae were first observed by the pond’s owner during early June 2001. The density was already very high at this time; the owner collected five specimens by dipping a narrow mouth bottle ( ~ 500 mL) into the water. These were taken to the county extension agent for identification. They were then brought to this investigator. The first observation by the author was on 6 July 2001 and about 50 specimens were collected off the west pier with five or six sweeps of a dip net. These specimens were maintained in the laboratory in aerated pond water until 2 August, when the last specimen died. When collected they varied from 15 to 22 mm in diameter. The longer they were held in the laboratory the smaller they became and were between 5 and 10 mm in diameter when they died. Other attempts to maintain the medusae in the laboratory had similar results. Qualitative observations of the medusae were made on 22 dates between 6 July and 7 November, and by snorkeling on 12 July, 2 August, 4 August and 7 September. Medusae were observed throughout the pond, however none were ever observed closer than about five meters from the shore. While swimming, their tentacles were extended aborally in the direction of movement. Many specimens would swim up in the water column until the longer tentacles touched the surface, they would then flip over and swim sideways or down. While snorkeling, many specimens brushed against the swimmers exposed skin and only a slight tingle was detected when they touched the more sensitive skin areas, such as around the lips or under the arms. No redness resulted. 360 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 4, 2002 Table 1 . Extremes of physical/chemical parameters at different depths in Kaiser Pond when Craspedacusta sowerbii was present in 2001. Depth m Temp. °C Oxygen mg/L PH Conductivity pS/cm Alkalinity mg/L Surf 21 - 33 5.6 - 8.1 6.5 - 7.8 245 - 267 109 - 115 1 21 - 33 5.4 - 8.6 6.8 - 7.8 247 - 267 2 21 - 31 2.6 - 9.1 6.9 - 7.6 248 - 268 3 21 - 31 0.7 - 9.4 7.0 - 7.6 249 - 271 3.8 21 - 31 0.2 - 6.9 6.9 - 7.4 251 - 271 Quantitative samples of the medusae were collected on 17 dates between 8 August through 7 November, when the last specimens were found. Distribution was patchy and density for individual plankton tows varied from zero on 7 September, 22 October, 5 November and 7 November to 121.32/m3 on 9 September (Table 2). Mean density for each collection date varied from 0.59/m3 on 7 September and 7 November to 34.43/m3 on 22 August (Table 2). On 15 October, 11 different plankton tows were taken at 15 to 30 minute time intervals from the east pier (Table 3). Collection began at 09:30 a.m. and ended at 12: 15 p.m. During the first five tows the sky was clear and no cloud cover occurred. By the sixth tow cloud cover began to move in and by the 11th tow 100 percent cloud cover existed and rainfall occurred. The density of C. sowerbii was significantly lower when no cloud cover existed than during the later six tows when cloud cover was increasing (Man-Whitney U test, t= 15, P^O.004). The Spearman rank order correlation coefficient between density and percent cloud cover was r=0.744, P= 0.0068. The positive relationship between density and percent cloud cover may have been a phototactic response to decreasing light conditions. It also could have been due to an upward migration following their zooplankton food source in the water column due to changing light conditions. These data (Table 3) also suggest that the other density measurements (Table 2), which were collected at different times between 08:00 and 13:30 hours under various light conditions, are not representative of the density at all depths. One specimen of a newly budded medusa, 1 mm in diameter with only four tentacles, was observed in a plankton sample collected 5 October. Five polyps were found on one of the artificial substrates that had been out for seven weeks at one meter depth. No polys were found on the other substrates. The size of the polyps ranged from 0.6 to 1.1 mm long and the entire body, except for the area around the oral HARREL 361 Table 2. Mean and extreme densities (No./m3) of Crcispedacusta sowerbii in Kaiser Pond for each date during 2001. n = number of tows. Date n Means Extremes Aug. 8 3 21.23 8.49 - 41.05 Aug. 10 3 1.78 1.41 - 2.83 Aug. 12 4 6.70 2.83 - 11.32 Aug. 18 4 5.98 4.25 - 7.00 Aug. 22 4 34.43 19.30 - 59.20 Sep. 2 4 27.99 14.15 - 50.19 Sep. 7 4 0.59 0 - 1.79 Sep. 9 5 28.66 1.29 - 121.32 Sep. 14 4 33.78 14.15 - 59.19 Sep. 22 4 15.76 10.29 - 19.30 Sep. 28 4 13.62 5.15 - 21.23 Oct. 5 6 15.38 5.90 - 31.14 Oct. 15 11 20.59 2.36 - 96.72 Oct. 22 6 4.33 0 - 8.26 Oct. 29 8 10.47 7.08 - 15.33 Nov. 5 6 0.79 0 - 2.36 Nov. 7 4 0.59 0 - 1.18 Table 3. Density of Craspedacusta sowerbii in individual plankton tows collected at 15 to 30 minute intervals under different cloud cover (light) conditions on October 15, 2001. Tow No. Time No./m3 % Cloud Cover 1 09:15 a.m. 4.72 0 2 09:30 3.54 0 3 09:45 4.72 0 4 10:00 2.36 0 5 10:15 5.90 0 6 10:30 10.60 20 7 10:45 3.54 30 8 11:00 30.67 50 9 11:20 55.44 50 10 11:45 8.26 75 11 12:15 p.m. 96.72 100 (rain) opening that contained nematocycts, was covered with detritus and difficult to see. One polyp specimen consisted of two zooids and the others were single zooids. The presence of the polyps on the suspended substrate indicates that sexual reproduction had occurred. Phytoplankton were always very sparse in the water column, except on 5 October, when a bloom of the yellow-green algae Dinobryon occurred. Zooplankton, including cladocerans, copepods and the rotifer Keratella , were present in high numbers throughout the study and were still abundant when the medusae disappeared. This suggests that food was not a limiting factor. Acker & Muscat (1976) reported that water 362 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 4, 2002 temperature affected the occurrence and survival of the different life cycle stages of C. sowerbii. They listed the minimum and maximum temperatures for the medusae as 15 and 30 °C, respectively. In Kaiser Pond the temperature extremes when the medusae were present varied from 21 to 33 °C. Acknowledgments I thank James and Francis Kaiser for access to their pond. Stephanie Bennie, Vickie Bordelon, Ana Christensen, David Hicks, Steven Lewis and Becky Wolff assisted in fieldwork. Paul Nicoletto criticized the manuscript. Literature Cited APHA - American Public Health Association. 1989. Standard methods for examination of water and wastewater. 17th ed. America Public Health Association, Washington DC, pp. 2-35 to 2-39. Acker, T. S. & A. M. Muscat. 1976. The ecology of Craspedacusta sowerbii Lankester, a freshwater hydrozoan. Am. Midi. Nat., 95:323-336. Baker, S. 1960. What’s new in the science department. Tyler County Booster, Woodville, Texas. Tuesday, September 15, p. 2. Cheatum, E. P. 1934. New distribution record for the medusa Craspedacusta. Science, 80:528. DeVries, D. R. 1992. The freshwater jellyfish Craspedacusta sowerbyi: A summary of it’s life history, ecology, and distribution. J. Freshwater Ecol., 7:7-16. Jugens, K. 1967. Strange creatures. Texas Game and Fish. July, p. 8. McCullough, J. D., M. F. Taylor & J. L. Jones. 1981. The occurrence of the freshwater medusa Craspedacusta sowerbyi Lankester in Nacogdoches Reservoir, Texas and associated physical-chemical conditions. Tx. J. Sci., 33(1): 17-23. Parsons, D. S. & J. W. Tatum. 1974. A new shallow water multiple-plate sampler. Progressive Fish-Cult., 36:179-180. TEXAS J. SCI. 54(4), NOVEMBER, 2002 363 GENERAL NOTES A SCIENTIFIC COMPARISON OF CENTRIFUGALLY CAST FIBERGLASS REINFORCED POLYMER PIPE AND BAR WRAPPED CONCRETE CYLINDER PIPE USING FINITE ELEMENT ANALYSIS M. Faruqi and M. Jao Department of Civil Engineering Texas A&M University-Kingsville MSC 194, Kingsville, Texas 78363 and Department of Civil Engineering, Lamar University Beaumont, Texas 77710 Composite materials have gained popularity in industry over the past fifteen years. They are being used in rehabilitation of existing structures and the design of new ones. Much research has been done on the use of fiber reinforced polymers (FRP) to rehabilitate and strengthen concrete columns, piers, bridge decks and to protect dock fenders. Until recently, very little research has been done pertaining to the use of FRP in piping applications, particularly for pressure-flow applications. Two kinds of FRP are commonly used; carbon-based and fiberglass- based composites. This paper focuses on centrifugally cast fiberglass reinforced polymer (CCFRPM) pipe and looks closely at large diameter (64-inch) pipe subjected to pressure-flow applications. A general comparison is given in Table 1 . A typical application is presented using finite element analysis to compare CCFRPM pipe to concrete bar wrapped cylinder pipe (B-303 Pipe) (Table 2). The simulation models the behavior of large diameter pipe under pressure-flow conditions on an elastic foundation. Resulting stresses and deflection of the CCFRPM pipe are compared to those of the B-303 Pipe. The following example is used to illustrate key differences between B-303 and CCFRPM pipe. Consider the design of a waterline main with a 64-inch inside diameter subjected to a 150 psi working pressure and a 75 psi surge pressure. The system is assumed to have soil cover conditions varying from 6 feet to 14 feet and is subjected to American Association of State Highways and Transportation Officials (AASHTO) HS-20 highway loading. The overburden soil has a unit weight of 120 pcf. The cross-sectional area of the B-303 pipe is designed in 364 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 Table 1. General comparison of B-303 and CCFRPM pipes. B-303 Pipe CCFRPM Pipe Inside Diameter 64 inch 64 inch Wall Thickness 2.5 inch 1 .5 inch Young’s Modulus (E) 2,350 ksi 4,000 ksi Maximum Allowable Deflection of Cross-section 1.1 inch 3.2 inch Maximum Allowable Externally Applied Load 14,700 lb/LF 14,700 lb/LF Table 2. Finite element analysis therefore yields the following results. B-303 Pipe CCFRPM Pipe Maximum Bending Stress 36.27 ksi 27.19 ksi Maximum Shearing Stress 14.85 ksi 10.73 ksi Maximum Deflection 1.323 inch 2.68 inches accordance with the method outlined in American Water Works Associa¬ tion (AWWA 1995) M9 standard C-303 for concrete bar wrapped cylinder pipe. Design yielded a 2.5 inch wall thickness. The inner steel cylinder is 0.353 inch thick, and the steel helical wrap was 5/8 inches in diameter on a 1 2/3 inch center to center spacing. The cross section of the CCFRPM pipe was designed according to AWWA specifications and manufacturer’s guidelines. The wall thickness is 1.5 inches and the Young’s Modulus is 2,350 ksi. STAAD-III computer software is used to model a 20 foot long pipe section with a 64 inch inside diameter. This version of STAAD (STAAD-III, 1995) does not support circular or radial elements. There fore, the circular pipe cross-section was approximated using a 32 sided polygon. A three-dimensional mesh surface was generated having 6.28 inch by 12 inch elements. It is assumed that soil bedding conditions offered support over the bottom half of the pipe. Soil spring coefficients are calculated using a soil subgrade modulus of 250 kip/sf/foot of deflection and projecting element areas in both the x and y directions. This simulated bedding conditions act in the radial direction on the bottom half of the pipe. External loads due to soil overburden are next calculated for both six feet and fourteen feet of cover. The model is then run using several loading combinations of dead and live loads to ensure those worst-case scenarios are considered and that maximum stress and deflection are obtained within the system. TEXAS J. SCI. 54(4), NOVEMBER, 2002 365 Note that the deflection of the B-303 pipe exceeded allowable deflections. This indicates that the soil stiffness modeled in STAAD is not adequate to prevent the pipe from deflecting, and pipe bedding conditions require re-evaluation if this pipe were to be installed. Conclusions The following conclusions can be drawn from the above finite element analysis and supporting background information. 1. CCFRPM Pipe is a less dense and less rigid system than B-303 pipe. 2. CCFRPM Pipe yielded lower stresses and higher deflections in finite element analysis than did the B-303 pipe with the same inside diameter subjected to the same internal pressure and external loading conditions. 3. CCFRPM Pipe is a viable alternative for large diameter pipe under pressure applications 4. Additional research is required before CCFRPM pipe will become is widely used in industry. Much work is required to develop design standards and guidelines to ensure its safe and economical use. Literature Cited American Water Works Association Manual of Water Supply Practices: Concrete Pressure Pipe, 1995, 295 pp. STAAD-III For Windows Reference Manual, 1995, 230 pp. MF at: M-Faruqi@tamuk.edu NOTEWORTHY RECORDS OF MAMMALS FROM THE ROLLING PLAINS OF TEXAS Chad A. Campbell, Thomas E. Lee, Jr. and Allan J. Landwer* Department of Biology, Box 27868 Abilene Christian University, Abilene, Texas 79699 * Department of Biology, Hardin-Simmons University, Abilene, Texas 79698 The southern Rolling Plains of north central Texas is an area of transition from eastern forest to western grassland mammalian species. This region is characterized by level to rolling plains with extensions of 366 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 4, 2002 eastern riparian habitats and open stands of mesquite (Blair 1950). There are isolated outcroppings of Edwards Plateau limestone in the southern portion of the Rolling Plains (Matthews 1960). Specimens were collected in habitats consistent with those of central Texas as described by Hanson et al. (1998). All specimens documented within this report represent county records or minor range extensions based on data from Davis & Schmidly (1994) and subsequent publications. Study skins and skeletal material are deposited in the Abilene Christian University Natural History Collection (ACUNHC) and the Hardin- Simmons University, Collection of Vertebrates (HSUCV). Cryptotis parva.— The least shrew occurs throughout the eastern and central portions of Texas and its range extends west in the Panhandle into New Mexico and south along the Rio Grande (Davis & Schmidly 1994). Revelez & Dowler (2001) report range extensions into the Edwards Plateau in Tom Green and Concho counties. Cryptotis parva inhabits grassland and seldom occurs in forests (Davis & Schmidly, 1994). The specimens reported here (ACUNHC 193 and 224) were salvaged from a cat (which is consistent with other reported records) in Albany, and represent the first from Shakelford County. Notiosorex crawfordi .—The distribution of the desert shrew includes the more arid, western and southern regions of the state in areas with available cover of Opuntia cactus (Davis & Schmidly 1994). Records in Texas are spotty, possibly due to standard methods of trapping which are not often conducive to catching shrews (Goetze 1998). Collection of a single specimen occurred using pitfall traps in an area of mesquite ( Prosopis sp.) and mixed grass vegetation in proximity to natural and man-made cover. Thornton & Lee (1996) reported extension of the range of N. crawfordi (ACUNHC 90) east into Callahan County. The collections of this species in southern Taylor County, 3 miles east of Bradshaw (ACUNHC 783) supports a more widespread occurrence in the southern Rolling Plains. Lasiurus borealis.— The eastern red bat is found in woodlands and riparian habitats (Goetze 1998). It is considered a year-round resident of east Texas; however, it is highly migratory and may be only a summer migrant to the western half of the state (Schmidly 1991). The specimen (ACUNHC 555) was hit by a U.S. mail truck in Brown County 10 miles north of Brownwood. Brown County is located along the ecotone of the Edwards Plateau and the Rolling Plains (Blair 1950). TEXAS J. SCI. 54(4), NOVEMBER, 2002 367 Sylvilagus floridanus . —The eastern cottontail is found throughout Texas. It is an inhabitant of brushy agriculture regions and riparian habitats usually not far from water (Davis & Schmidly 1994; Goetze 1998). Taylor County is a mosaic of agriculture, grassland and brush providing ample habitat for this species. The specimen documented here (ACUNHC 127) was collected from Taylor County (no specific locality) in 1969. Lepus califomicus .—The black-tailed jackrabbit is common statewide except for the Big Thicket region (Davis & Schmidly 1994). This species is common on the Rolling Plains, but not well documented. This report documents county records for L. califomicus collected in 1969 from Taylor County no specific locality (ACUNHC 126) and collected in 1996 from Callahan County; 3 miles south of Putnam (ACUNHC 397). Perognathus merriami .—The distribution of Merriam’s pocket mouse is known from the western two- thirds of the state, but it is absent from the extreme northern Panhandle and western Trans-Pecos due to the presence of Perognathus flavus (Lee & Engstrom 1991; Davis & Schmidly 1994). It is common in rocky habitats with sparse ground cover and short to mid-height grasses (Goetze 1998). Collection of this specimen from 2 miles north Lake Abilene (ACUNHC 780) was in a juniper-dominated community with mid height grasses and rock outcrop¬ pings of Triassic origin. Taylor County is within the range of P. merriami and parts of the county are consistent with its typical habitat. This specimen, however, represents the first record for Taylor County. Furthermore, we also report on a specimen of P. merriami from Jones County; Dub Wofford Ranch, 32° 40’ N, 99° 39’W (HSUCV 173) collected in 1970. Sigmodon hispidus— The hispid cotton rat is widespread in Texas (Davis & Schmidly 1994). Specimens collected in 2001 from Hawley (ACUNHC 785, 786) and in 1970 from Dub Wofford Ranch, 32° 40’ N, 99° 39’ W (HSUCV 13) represents the first record for Jones County. The habitat from which the Hawley specimen was taken was highly disturbed by mowing and the planting of pine trees. Peromyscus pectoralis. — The white-ankled mouse was collected in Jones County, Hawley (ACUNHC 790). This specimen was found at the northwestern limits of the species’ range in Texas (Davis & Schmidly 368 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 1994) and was taken from the same trap line as the above mentioned S. hispidus near Hawley. Neotoma micropus The southern plains woodrat prefers open grass and brush habitat that is common in Jones County. The first specimens (HSUCV 165, 271) collected in this county were from the Dub Wofford Ranch, 32° 40’ N, 99° 39’W . Baiomys taylori. — The northern pygmy mouse has been documented in the southern Rolling Plains (Davis & Schmidly 1994; Hanson et al. 1998; Revelez & Dowler 2001). The undocumented presence of this common species indicates how poorly the mammal fauna of Jones County is known. The specimen (HSUCV 334) was collected in 1970, Jones County; Dub Wofford Ranch, 32° 40’ N, 99° 39’W . Procyon lotor.— The common raccoon is present throughout the state (Davis & Schmidly 1994). It is common in riparian habitats or wooded areas, abandoned farmlands, and often around humans (Goetze 1998). Distribution is influenced more by the presence of water than by the type of vegetation (Davis & Schmidly 1994). A specimen representing a county record (ACUNHC 584) was collected 4 miles south Putnam in an oak-juniper-dominated community of Callahan County. Literature Cited Blair, F. W. 1950. The biotic provinces of Texas. Texas J. Sci., 2( 1 ) :93- 1 17. Davis, W. B. & D. J. Schmidly. 1994. The mammals of Texas. Texas Parks and Wildlife Press, Austin, x + 338 pp. Goetze, J. R. 1998. The mammals of the Edwards Plateau. Special Publications Museum of Texas Tech Univ., 41:1-263. Hanson, J. D., C. E. Peden & T. E. Lee, Jr. 1998. Records of species and range extensions of mammals in Taylor County, Texas. Texas J. Sci., 50(3):251-255. Lee, T. E., Jr. & M. D. Engstrom. 1991. Genetic variation in the silky pocket mouse ( Perognathus flavus ) in Texas and New Mexico. J. Mamm., 72(2):273-285. Matthews, W. H. 1960. Texas fossils, an amateur collectors handbook. Bureau of Economic Geology. University of Texas Press, Austin, viii + 123pp. Revelez, M. A. & R. C. Dowler. 2001. Records of Texas mammals housed in the Angelo State Natural History Collections, Angelo State University. Texas J. Sci., 53(3):273-284. Schmidly, D. J. 1991. The bats of Texas. Texas A&M Univ. Press, xvii + 188pp. Thornton, M. L. & T. E. Lee, Jr. 1996. Distributional records of three mammals from the Rolling Plains of central Texas. Texas J. Sci., 48(4) :33 1-332. TEL at: lee@biology.acu.edu TEXAS J. SCI. 54(4), NOVEMBER, 2002 369 RECENT RECORDS OF BATS FROM THE LOWER CANYONS OF THE RIO GRANDE RIVER OF WEST TEXAS Loren K. Ammerman, Rogelio M. Rodriguez, Jana L. Higginbotham* and Amanda K. Matthews Department of Biology, Angelo State University San Angelo, Texas 76909 * Department of Biological Sciences and the Museum of Texas Tech University Lubbock, Texas 79409 Past surveys of mammals in Brewster and Terrell counties, Texas, focused primarily on the mountains and surrounding lowland desert habitats, the tributaries to the Rio Grande and their associated canyons, and sections of the Rio Grande within the boundaries of Big Bend National Park (Bailey 1905; Borrell & Bryant 1942; Schmidly et al. 1976; Schmidly & Ditton 1979; Hollander et al. 1990; Higginbotham & Ammerman 2002). These investigations did not include the Lower Canyons section of the Rio Grande. The Lower Canyons region of the Rio Grande stretches for approxi¬ mately 100 km from Reagan Canyon, just east of La Linda, Brewster County, and continues east- northeast to the Dryden crossing takeout, east of San Francisco Canyon, Terrell County. This section of the Rio Grande marks a gradual transition from the Chihuahuan Biotic Province towards a landscape that becomes increasingly characteristic of the convergence zone between the Balconian and Tamaulipan biotic prov¬ inces described by Blair (1940; 1950) and Dice (1943). The combina¬ tion of these different ecosystems has the potential to support high species diversity. Additionally, along the banks of the river, the native riparian vegetation has been impacted by recent invasions of exotic plant species, such as salt cedar (Tamarix sp.) and giant cane (. Arundo donax), which also contribute to plant diversity along the Rio Grande. The Rio Grande Wild and Scenic River historically served as an important biogeographic barrier along the U.S. -Mexican border (Schmidly 1977), but recent changes due to impoundments, irrigation, invasion of non-native plants, and pollution potentially have altered this riparian ecosystem and the occurrence of animal and plant species (Schmidly & Ditton 1979). There are no published accounts of bats for the Lower Canyons 370 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 stretch of the Rio Grande. However, several investigators recently reported unexpected bat species from nearby (Yancey et al. 1995; Dowler et al. 1999; Higginbotham et al. 1999; Jones & Bradley 1999; Jones et al. 1999). These findings suggest that bat distributions are changing in Texas. However, these distributional shifts are still not understood and may simply reflect an increase in the knowledge of bat distributions as a result of increased survey efforts in these areas. This report notes the occurrence of 13 bat species in the Lower Canyons of the Rio Grande, including two new records from Terrell County. A series of three surveys, 30 October to 4 November 1999, 19-22 March 2001 and 7-12 October 2001, were conducted with the assistance of the National Park Service. Mist- nets were placed at a total of eight sites over or along the banks of the river, at mouths of canyons, and/or over shallow pools. Captured bats were identified to species, sexed, aged, and measured using standard procedures (Handley 1988). Voucher specimens were deposited in the Angelo State Natural History Collection (ASNHC), Angelo State University. From 30 October to 4 November. 1999, a total of 18 bats was captured. From 19-22 March 2001 there was a total of 44 captures. The period of 7-12 October 2001 produced a total of 38 captures. Species caught (in order of decreasing abundance) were Myotis yumanensis, Antrozous pallidus, Tadarida brasiliensis , Mormoops megalophylla, Cory norhinus towns endii, Las iurus cine reus, Nyctinomops femorosaccus, Nyctinomops macrotis, Pipistrellus hesperus, Myotis califomicus, Myotis thy s anodes, Myotis ve lifer and Lasionycteris noctivagans . Details of the occurrence of M. megalophylla, N. femorosaccus, M. thysanodes, C. townsendii, M. califomicus and L. noctivagans, are described below. Terrell County records include M. megalophylla and N. femorosaccus. Mormoops megalophylla. — Ghost- faced bats are known in Texas from the southern Trans- Pecos region, the southern edge of the Edwards Plateau, and extreme south Texas (Davis & Schmidly 1994). Two adult females were captured, one on 11 October 2001 (River Mile 710.5; 29°52’47”N, 102°19’12”W) and another (ASNHC 11581) on 12 October 2001 (River Mile 698; 29°50,51.3”N, 102° 1 1’0.7”W). Al¬ though records have been reported for surrounding counties, Brewster and Val Verde, these captures represent the first reported for Terrell TEXAS J. SCI. 54(4), NOVEMBER, 2002 371 County. Other bats captured in the same night included M. yumanensis , A. pallidus, T. brosiliensis , N. femorosaccus and P. hesperus. Nyctinomops femorosaccus Within the United States, pocketed free-tailed bats have only been reported from Big Bend National Park, Brewster County, Texas, along with localities in southern California, southern Arizona and southeastern New Mexico (Schmidly 1991; Davis & Schmidly 1994). Five bats (2 subadult males, 2 subadult females and 1 adult female) were caught on 11 October 2001 (River Mile 710.5; 29°52’47”N, 102°19,12”W). One subadult male was collected (ASNHC 1 1587). This specimen represents the first reported for Terrell County. These captures also represent the most eastern records of the species in the United States. Other bats captured in the same night were M. yumanensis , A. pallidus , T. brasiliensis, M. megalophylla and P. hesperus. My otis thys anodes.— The fringed myotis is a migratory species that generally is known in the Trans-Pecos to arrive in April and depart in October (Easterla 1973; O’Farrell & Studier 1980). On 20 March 2001 , one adult male (ASNHC 11517) was collected (River Mile 723.2; 29°46’17”N, 102°23’54”W). This record extends the seasonal occur¬ rence of M. thysanodes in the Trans-pecos region. Corynorhinus townsendii. — The genus Corynorhinus is used (instead of Plecotus) due to recent taxonomic revision (Bogdanowicz et al. 1998; Hoofer & Van Den Bussche 2001). In Texas, Townsend’s big-eared bat is known from the Trans-Pecos and Plains regions (Davis & Schmidly 1994). An adult female (ASNHC 1 1525) was taken in Brewster County within 300 meters of the Brewster-Terrell County line (River Mile 710.8; 29°53’17.2”N, 102°20’ 16.2”W) on 3 November 1999. This specimen represents the easternmost record of this species within Brewster County and suggests its range extends into Terrell County. Other bats caught in the same night were T. brasiliensis, N. femorosaccus , and M. calif omicus. Myotis califomicus. — Within Texas, the California myotis is found mainly in the Trans-Pecos region with one record from the Panhandle (Davis & Schmidly 1994). An adult male (ASNHC 11512) was collected within 300 meters from the Brewster-Terrell County line 372 THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 4, 2002 (River Mile 710.8; 29°53’17.2”N, 102°20’16.2”W) on 3 November 1999. This specimen represents the eastern-most record of this species within Brewster County and suggests a possible occurrence in Terrell County. Other bats caught in the same night were T. brasiliensis , N. femorosaccus and C. townsendii. Lasionycteris noctivagans .-Silver-haired bats are thought to have a wide distribution in Texas based on scattered localities across the state (Schmidly 1991; Davis & Schmidly 1994). Dowler et al. (1992) reported two males northeast of Dryden within Terrell County. One adult male (ASNHC 11505) (River Mile 698; 29°50’54”N, 102° ll’Or’W) was collected in Terrell County on 22 March 2001. This capture represents the third record reported for the county. Other bats caught in the same night were T. brasiliensis , M. yumanensis, A. pallidas and M. velifer. This investigation contributes to the current understanding of the distributions of bat species along the Lower Canyons region of the Rio Grande Wild and Scenic River in Texas. The Rio Grande has experi¬ enced substantial change throughout the last century; changes that could severely impact the river’s water flow, wildlife potential, and integrity of the native vegetation. As the vegetation continues to change due to invasion of non-native species like salt cedar ( Tamarix sp.) and giant cane (Arundo donax ), distributions of bats (and other organisms) are likely to be affected. Acknowledgements We thank the National Park Service personnel, especially Michael Ryan and Stephen McAllister, at Big Bend National Park for their assistance, loan of equipment and for guiding us safely down the river. Raymond Skiles and Marcos Paredes provided logistical support. These surveys were conducted in cooperation with the National Park Service under a resource activity permit. Collection permits were issued by the Texas Parks and Wildlife Department (# SPR-0994-703). Scott Burt, Michael Dixon, David Long, Michael Moreno and Bryan Reece provided valuable field assistance. We also thank Robert Bradley, Robert Dowler and Frank Yancey for providing comments on a previous version of this manuscript. TEXAS J. SCI. 54(4), NOVEMBER, 2002 373 Literature Cited Bailey, V. 1905. Biological survey of Texas. N. Am. Fauna, vol. 25. Washington, D.C. : Dept, of Agriculture, Bureau of Biological Survey, 222 pp. Blair, W. F. 1940. A contribution to the ecology and faunal relationships of the mammals of the Davis Mountain region, southwestern Texas. Misc. Publ. Univ. Michigan Mus. Zool., Ann Arbor, 46:1-39. Blair, W. F. 1950. The biotic provinces of Texas. Texas J. Sci., 2:93-117. Bogdanowicz, W. , S. Kasper & R. D. Owen. 1998. Phylogeny of plecotine bats: Reevaluation of morphological and chromosomal data. J. Mamm., 79(l):78-90. Borrell, A. E. & M. D. Bryant. 1942. Mammals of the Big Bend area of Texas. Univ. California. Publ. Zool., 48:1-62. Davis, W. B. & D. J. Schmidly. 1994. The mammals of Texas. Texas Parks and Wildlife Department, Austin, Texas, x-f-338 pp. Dice, L. R. 1943. The biotic provinces of North America. Univ. Michigan Press, Ann Arbor, 78 pp. Dowler, R. C., T. C. Maxwell & D. S. Marsh. 1992. Noteworthy records of bats from Texas. Texas J. Sci., 44(1): 121-123. Dowler, R. C., R. C. Dawkins & T. C. Maxwell. 1999. Range extensions for the evening bat ( Nycticeius humeralis) in west Texas. Texas J. Sci., 51(2): 193-195. Easterla, D. A. 1973. Ecology of the 18 species of Chiroptera at Big Bend National Park, Texas. Northwest Missouri State Univ. Studies, 34:1-165. Handley, C. O., Jr. 1988. Specimen preparation. Pp. 437-457, in Ecological and behavioral methods for the study of bats (T. H. Kunz, ed.) Smithsonian Institution Press, Washington D.C., xxii + 533 pp. Higginbotham, J. L. & L. K. Ammerman. 2002. Chiropteran community structure and seasonal dynamics in Big Bend National Park. Spec. Pub., Mus. of Texas Tech Univ., 44:1-44. Higginbotham, J. L., L. K. Ammerman & M. T. Dixon. 1999. First record of Lasiurus xanthinus (Chiroptera: Vespertilionidae) in Texas. Southwestern Nat., 44(3):343-347. Hollander, R. R., C. Jones, J. K. Jones, Jr. & R. W. Manning. 1990. Preliminary analysis of the effects of the Pecos River on geographic distribution of small mammals in western Texas. J. Big Bend Studies, 2:97-107. Hoofer, S. R. & R. A. Van Den Bussche. 2001. Phylogenetic relationships of plecotine bats and allies based on mitochondrial ribosomal sequences. J. Mamm., 82(1): 131-137. Jones, C. & R. D. Bradley. 1999. Notes on red bats, Lasiurus (Chiroptera: Vespertilionidae), of the Davis Mountains and vicinity, Texas. Texas J. Sci., 51(4):341-344. Jones, C., L. Hedges & K. Bryan. 1999. The western yellow bat, Lasiurus xanthinus (Chiroptera: Vespertilionidae), from the Davis Mountains, Texas. Texas J. Sci., 51(3):267-269. O’Farrell, M. J. & E. H. Studier. 1980. Myotis thysanodes. Mammalian Species, 137:1-5. Schmidly, D. J. 1977. Factors governing the distribution of mammals in the Chihuahuan Desert region. Pp. 163-192, in Transactions of the symposium on the biological resources of the Chihuahuan region, United States and Mexico. (R. Wauer and D. Riskind, eds.). National Park Service Transactions and Proceedings Series no. 3, pp. 163-192. Schmidly, D. J. 1991. The Bats of Texas. Texas A&M University Press, College Station, xviii+ 188pp. 374 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 4, 2002 Schmidly, D. J. & R. B. Ditton. 1979. Relating human activities and biological resources in riparian habitats of western Texas. Pp. 107-1 16 in Recreational Impact on Wildlands Conference Proceedings (R. Ittner, D. R. Potter, J. K. Agee, S. Anschnell eds.). U.S. Forest Service, Pacific Northwest Region No. R-6-01 1-1979, 333 pp. Schmidly, D. J., R. B. Ditton, W. J. Boeer & A. R. Graefe. 1976. Inter-relationships among visitor usage, human impact, and the biotic resources of the riparian ecosystem in Big Bend National Park. Paper presented at the First Conference on Scientific Research in the National Parks, November 9, 1976, New Orleans, Louisiana. Yancey, F. D., II., C. Jones & R. W. Manning. 1995. The eastern pipistrelle, Pipistrellus subflavus (Chiroptera: Vespertilionidae), from the Big Bend Region of Texas. Texas J. Sci., 47:229-231. LKA at: loren.ammerman@angelo.edu THE TEXAS JOURNAL OF SCIENCE— VOL. 54, NO. 4, 2002 375 IN RECOGNITION OF THEIR ADDITIONAL SUPPORT OF THE TEXAS ACADEMY OF SCIENCE DURING 2002 Patron Members Ali R. Amir-Moez Deborah D. Hettinger Don W. Killebrew David S. Marsh Patrick L. Odell John Sieben Ned E. Strenth Charles H. Swift Sustaining Members James Collins Dovalee Dorsett Stephen R. Goldberg Norman V. Horner Michael Looney Judith A. Schiebout Fred Stevens Supporting Members David A. Brock Frances Bryant Edens Donald E. Harper, Jr. Paul D. Mangum George D. McClung Jimmy T. Mills Jim Neal Nancy Ellen Partlow Paul T. Price Sammy M. Ray John Riola Lynn Simpson William F. Thomann Milton W. Weller 376 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 Plan Now for the 106th Annual Meeting of the Texas Academy of Science February 27 - March 1, 2003 Stephen F. Austin State University Program Chair John T. Sieben Dean of Natural Sciences & Mathematics Texas Lutheran University 1000 W. Court Street Seguin, Texas 78155 Phone: 830.372.6005 FAX: 830.372.6095 E-mail: jsieben@tlu.edu Local Host William D. Clark Dept, of Mathematics & Statistics Stephen F. Austin State University P.O. Box 10340 SFA Sta. Nacogdoches, Texas 75962 Phone: 936.468.3805 FAX: 936.468.1669 E-mail: clark@sfasu.edu For additional information relative to the Annual Meeting, please access the Academy homepage at: www . texasacademyofscience.org Future Academy Meetings 2004-Schreiner University TEXAS J. SCI. 54(4):377-382 NOVEMBER, 2002 INDEX TO VOLUME 54 (2002) THE TEXAS JOURNAL OF SCIENCE Ned E. Strenth Department of Biology, Angelo State University San Angelo, Texas 76909 This index has separate subject and author sections. Words, phrases, locations, proper names and the scientific names of organisms are followed by the initial page number of the article in which they appeared. The author index includes the names of all authors followed by the initial page number of their respective article(s). SUBJECT INDEX A Activity, mutagenic 249 Albian 133 Ammonia 3 A no l is 51 Anthracotheriidae 301 Anthracy dines 249 Araneae: Lycosidae 261 Arizona 143, 347 Artiodactyla: Mammalia 301 Automated storm sampling 177 B Bacterium 249 Baiomys taylori 1 89 Bats 277 , 369 Beaumont 357 Bipedal dinosaurs 309 Black bass 125 Boerne Lake Spillway 309 C Caddo Lake (fishes) 1 1 1 Checklist (fishes) 1 1 1 Chinese tallow 63 Chiroptera 89 Chrysophytes 27 Clairborne Group 301 Cnidaria: Hydrozoa 357 Coachwhip 143 Coastal Bend 241 Collared lizard 151 Coins laevigata 63 Coluber constrictor 59 Concrete pipe 363 Conic polars 291 Convergent infinite series 291 Cretaceous, Lower 133, 309 Cronartium quercuum 325 Cro talus atrox 347 Crotaphytus collaris 1 5 1 378 THE TEXAS JOURNAL OF SCIENCE- VOL. 54, NO. 4, 2002 D Davis Mountains 89 DesCartes’ foliums 291 Dinosaur trackways 133 Distributional records 269 DNA, Mitochondrial 151 Downstream effects 69 E Ecological notes 357 Effects of incubation temperature 261 Effects of prescribed burning 2 1 1 Effects of temperature and light 63 Electron microscopy 27 Elphidium 3 Enzymatic variation 37 Eocene, Middle 301 Equus 17 Euglandina texasiana 37 F Facilitated succession 163 Fiberglass 363 Finite element analysis 363 Fire 195,211 Fishing tournaments 125 Foliums 291 Foraminifera 3 Fuel leading 211 Fusiform rust 325 G Gaige’s tropical night lizard 282 Gastrointestinal 282 Gastropoda: Pulmonata 37 Gene flow 151 Ghost-faced bat 89 Glen Rose Formation 133, 309 H Habitat utilization 59 Helminths 282 Heptacodon 301 Herbaceous vegetation 195 Hessian foliums 291 Hidalgo, Mexico 282 Hogna carolinensis 261 Horse 17 Hydrozoan 357 Ichnology 309 Idarubicin 249 Incubation temperature 261 Iguanidae 45, 51 J Jamaica 51 L Lake Amistad 125 Land snail 37 Las Palomas 163 Learning, spatial association 45 Lepidophyma gaigeae 282 Lizard 45, 151, 282 Locomotor activity 261 Louisiana 339 Lovebug 339 Lovelady 301 Lower Canyons, Rio Grande 369 INDEX 379 M Mallomonas multisetigera 27 Mammals (Permian Basin) 269 Mammals (Rolling Plains) Mass capture of insects 339 Masticophis flagellum 143 Medusa stage 357 Metapod ials 17 Mexico 37, 282 Micropterus sp. 125 Microscopy, electron 27 Minimum flow 177 Mitochondrial DNA 151 Modeling 325 Mormoopidae 89 Mormoops megalophylla 89 Mortality (fish) 125 Mouse, northern pygmy 1 89 Mustang Island 241 Mutagenic activity 249 N Northern pygmy mouse 1 89 Nueces Valley 17 O Ovulation 347 Ozone 99 P Padre Island 241 Paleoenvironment 309 Permian Basin 269 Pine, slash 325 Pirns elliottii 325 Pitcher plant 339 Plants (Coastal Bend) 241 Plecia nearctica 339 Pleistocene, Late 17 Pollution 99 Polymer pipe 363 Postal notice 383 Pulmonata 37 R Range 1 89 Rattlesnake, Western Diamond- backed 347 Reproduction 143 Reproduction, Crotalus atrox 347 Response to fire 195 Re-vegetation 163 Rio Grande River 369 Rodentia: Muridae 189 Rolling Plains 365 S Salmonella typhimurium 249 San Marcos River 69 Sapium sebiferum 63 Sarracenia alata 339 Sarraceniaceae 339 Sauria: Iguanidae 45 Sauria: Xantusiidae 282 Sceloporus poinsettii 45 Seed germination 63 Sensors (ozone) 99 Serpentes: Colubridae 143 Serpentes: Viperidae 347 Silica-scaled 27 Slash pine 325 Snail 37 Stratigraphy 309 Structural habitat 51 Succession, facilitated 163 Survival modeling 325 Synurophytes 27 380 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 T U Texas 17, 37, 111, 125, 133, 151, 163, 189, 195, 227, 241, 269, 277, 301, 309, 339, 365, 369 Texas Counties Brewster 369 Cameron 163 Crane 269 Dallas 59 Ector 269 Houston 301 Jefferson 357 Kinney 133 Loving 269 Polk 227 Terrell 369 Ward 269 Winkler 269 Texas Sugarberry 63 Theropod (trackways) 133 Tournaments, fishing 125 Trackways, dinosaur 133 Trans-Pecos 277 Tree basal area 325 Upstream changes 69 V Vascular flora 227 Vegetation, herbaceous 195 W Watershed 177 Windham Prairie 227 Y Yegua Formation 301 Yellowbelly racers 59 Yolk deposition 347 Z Zooplankton 357 INDEX 381 AUTHOR INDEX Amir-Moez, A. R. 291 Ammerman, L. K. 369 Baskin, J. A. 17 Bonem, R. M. 309 Bowen, C. J. 3 Brant, J. G. 189, 269, 277 Brown, L. E. 227 Brumfield, J. M. 249 Bursey, C. R. 282 Buzas-Stephens, P. 3 Camarillo-Rangel, J. L. 282 Campbell, C. A. 365 Campbell, J. H. 151 Chapla, M. 261 Chavoshi, J. A. 291 Coble, D. W. 325 DeBaca, R. S. 89, 277 Dorsett, D. 99 Dowler, R. C. 189 Earl, R. A. 69 Evans, R. E. 339 Farlow, J. O. 309 Faruqi, M. 363 Ferguson, G. W. 51 Galloway, C. 241 Gehrmann, W. H. 59 Gibson, T. C. 339 Goldberg, S. R. 143, 282, 347 Harmel, R. D. 177 Harrel, R. C. 357 Hawthorne, J. M. 309 Higginbotham, J. L. 277, 369 Hillhouse, K. 227 Holroyd, P. A. 301 Hubbs, C. Ill Jao, M. 363 Jones, C. 89, 269, 277 Jones, J. O. 309 Judd, F. W. 163 King, K. W. 177 Landwer, A. J. 51, 365 Lankau, R. A. 63 Larson, D. H. 125 Lee, Jr., T. E. 365 Lee, Y. 325 Lonard, R. I. 163 Mackay, W. J. 249 MacRoberts, B. R. 227, 339 MacRoberts, M. H. 227, 339 Matthews, A. K. 369 McCoy, J. K. 151 Miserendino, N. 99 Mosqueda, A. E. 17 Negrete, I. G. 241 Nelson, A. D. 241 Nijjer, S. 63 Oswald, B. P. 211 Perez, K. E. 37 Pessagno, Jr., E. A. 3 Punzo, F. 45, 261 Reams, R. D. 59 Redell, W. H. 125 Rideout, S. 211 Rodriguez, R. M. 369 Rogers, II, J. V. 133 Rogers, W. E. 63 Rosen, P. C. 347 Ruthven, III, D. C. 195 Siemann, E. 63 Strenth, N. E. 37 Synatzske, D. R. 195 Torbert, H. A. 177 Van Kley, J. E. 27 Waggerman, G. L. 163 Wee, J. L. 27 Wilde, G. R. 125 Wilde, III, G. R. 125 Wolfe, J. E. 177 Wood, C. R. 69 Wujek, D. E. 27 382 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 REVIEWERS The Editorial staff wishes to acknowledge the following individuals for serving as reviewers for those manuscripts considered for publication in Volume 54. Without your assistance it would not be possible to maintain the quality of research results published in this volume of the Texas Journal of Science. Baker, R. Gregory, P. Morey, P Banihatti, N. Gutberlet, R. Mueller, J. Barry, D. Haywood, J. Murray, H. Baskin, J. Holmes, W. Nesom, G. Bradley, R. Horner, N. Novikov, A. Cecil, D. Howells, R. Pierce, B. Chapman, B. Hubbs, C. Pittman, J. Cobb. V. Jones, C. Reynolds, E. Diamond, D. Judd, F. Sommers, C. Dixon, J. Keeley, B. Stangl, F. Dowler, R. Kelley, B. Sunter, G. Doyle, R. Ki Hebrew, D. Svensson, J. Drawe, L. Kritsky, D. Taylow, M. Duraid, Z. Lancaster, D. Van Auken, W Echelle, T. Lonard, R. Vaughan, K. Edwards, R. Lundelius, E. Webb, R. Ernst, E. McAllister, C. Westgate, J. Fitch, H. McBee, K. Wilde, G. Garrett, G. McCullough, J. Wilkins, K. Gelwick, F. McCune, E. Wilson, L. Goldberg, S. McHugh, D. Yancey, F. THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 383 United States Postal Service Statement of Ownership, Management, and Circulation 1 . Publication Title The Texas Journal of Science 2. Publication Number 3. Filing Date 1 October 2002 0 0 A 0 _ 4 A 03 4. Issue Frequency Quarterly 5. Number of Issues Published Annually A 6. Annual Subscription Price $30 Membership $50 Subscription 7. Complete Mailing Address of Known Office of Publication (Not printer) (Street, city, county, state, andZlP+4) Biology Department, Angelo State University 2601 West Avenue N, Tom Green County, San Angelo, TX 76909 Contact Person N.E. Strenth Telephone 915-942-2189 ext. 247 8. Complete Mailing Address of Headquarters or General Business Office of Publisher (Not printer) Dr. Ned E. Strenth, Biology Department Angelo State University San Angelo, TX 76909-5069 _ 9. Full Names and Complete Mailing Addresses of Publisher. Editor, and Managing Editor (Do not leave blank) _ Publisher (Name and complete mailing address) Dr. Ned E. Strenth, Biology Department Angelo State University San Angelo, IX 76909-5069 _ Editor (Name and complete mailing address) Dr. Ned E. Strenth, Biology Department Angelo State University San Angelo, IX 76909-5069 _ Managing Editor (Name and complete mailing address) Dr. Ned E. Strenth, Biology Department Angelo State University San Angelo, TX 76909-5069 _ _ 10. Owner (Do not leave blank. If the publication is owned by a corporation, give the name and address of the corporation immediately followed by the names and addresses of all stockholders owning or holding 1 percent or more of the total amount of stock If not owned by a corporation, give the names and addresses of the individual owners. If owned by a partnership or other unincorporated firm, give its name and address as well as those of each individual owner. If the publication is published by a nonprofit organization, give its name and address.) Full Name Complete Mailing Address Texas Academy of Science Angelo State University Department of Biology 2601 West Avenue N San Angelo, TX 76909-5069 11. Known Bondholders, Mortgagees, and Other Security Holders Owning or Holding 1 Percent or More of Total Amount of Bonds, Mortgages, or Other Securities. If none, check box ■ ► None Full Name Complete Mailing Address 1 2. Tax Status (For completion by nonprofit organizations authorized to mail at nonprofit rates) (Check one) The purpose, function, and nonprofit status of this organization and the exempt status for federal income tax purposes: □(Has Not Changed During Preceding 12 Months □ Has Changed During Preceding 12 Months (Publisher must submit explanation of change with this statement) PS Form 3526, October 1999 (See Instructions on Reverse) 384 THE TEXAS JOURNAL OF SCIENCE-VOL. 54, NO. 4, 2002 13. Publication Title The Texas Journal of Science 14. Issue Date for Circulation Data Below November 2002 15. Extent and Nature of Circulation Average No. Copies Each issue During Preceding 12 Months No. Copies of Single Issue Published Nearest to Filing Date a. Total Number of Copies (Net press run) 1100 1100 0) Paid/Requested Outside-County Mail Subscriptions Stated on Form 3541. (Include advertiser's proof and exchange copies) 808 965 b. Paid and/or Requested Circulation (2) Paid in-County Subscriptions Stated on Form 3541 (Include advertiser's proof and exchange copies) 15 15 (3) Sales Through Dealers and Carriers, Street Vendors, Counter Sales, and Other Non-USPS Paid Distribution 157 157 (4) Other Classes Mailed Through the USPS 0 0 c- Total Paid and/or Requested Circulation [Sum of 15b. (1). (2),(3),and (4)J ► 980 980 dFree Distribution (1) Outside-County as Stated on Form 3541 0 0 by Mail (Samples, compliment ary, and other free) (2) In-County as Stated on Form 3541 0 0 (3) Other Classes Mailed Through the USPS 0 0 e- Free Distribution Outside the Mail (Carriers or other means) 0 0 Total Free Distribution (Sum of 15d. and 15e.) ► 0 0 g. Total Distribution (Sum of 15a and 15f) ► 980 980 h. Copies not Distributed 120 120 i. Total (Sum of ISg. and h.) > , 1100 1100 i- Percent Paid and/or Requested Circulation (15c. divided by 15g. times 100) 89% 89% 16. Publication of Statement of Ownership Vol. 54, #4 . issue of this publication. □ Publication not required. 17. Signature and Title of Editor, Pubtteh eCBusiness Manager, or Owner Date 27 Sept. 2002 I certify that ail information furnished on this form Is true and complete. I understand that anyone who furnishes false or misleading information on this form or who omits material or information requested on the form may be subject to criminal sanctions (including fines and imprisonment) and/or civil sanctions (including civil penalties). Instructions to Publishers 1 . Complete and file one copy of this form with your postmaster annually on or before October 1 . Keep a copy of the completed form for your records. 2. In cases where the stockholder or security holder is a trustee, include In Hems 10 and 11 the name of the person or corporation for whom the trustee is acting. Also include the names and addresses of individuals who are stockholders who own or hold 1 percent or more of the total amount of bonds, mortgages, or other securities of the publishing corporation. In item 11, if none, check the box. Use blank sheets if more space is required. 3. Be sure to furnish all circulation information called for in item 15. Free circulation must be shown in items 15d, e, and f. 4. Item 15h., Copies not Distributed, must include (1) newsstand copies originally stated on Form 3541 , and returned to the publisher, (2) estimated returns from news agents, and (3), copies for office use, leftovers, spoiled, and all other copies not distributed. 5. If the publication had Periodicals authorization as a general or requester publication, this Statement of Ownership, Management, and Circulation must be published; it must be printed in any issue in October or, if the publication is not published during October, the first issue printed after October. 6. In item 1 6. indicate the date of the issue in which this Statement of Ownership will be published. 7. Item 17 must be signed. Failure to file or publish a statement of ownership may lead to suspension of Periodicals authorization. PS Form 3526, October 1999 (Reverse) THE TEXAS ACADEMY OF SCIENCE, 2002-2003 OFFICERS President : President Elect : Vice-President: Immediate Past President : Executive Secretary : Corresponding Secretary: Managing Editor : Manuscript Editor: Treasurer: AAAS Council Representative: Larry D. McKinney, Texas Parks and Wildlife Department John T. Sieben, Texas Lutheran University John A. Ward, Brook Army Medical Center David R. Cecil, Texas A&M University-Kingsville Fred Stevens, Schreiner University Deborah D. Hettinger, Texas Lutheran University Ned E. Strenth, Angelo State University Robert J. Edwards, University of Texas-Pan American James W. Westgate, Lamar University Sandra S. West, Southwest Texas State University DIRECTORS 2000 Bobby L. Wilson, Texas Southern University John P. Riola, Texaco Exploration 2001 David S. Marsh, Angelo State University Felipe Chavez-Ramirez, International Crane Foundation 2002 Sushma Krishnamurthy, Texas A&M International University Raymond D. Mathews, Jr., Texas Water Development Board SECTIONAL CHAIRPERSONS Anthropology: Roy B. Brown, Instituto Nacional de Antropologia y Historia Biological Science: David S. Marsh, Angelo State University Botany: Cyndy Galloway, Texas A&M University-Kingsville Chemistry: Mary A. Kopecki-Fjetland, St. Edward’s University Computer Science: John T. Sieben, Texas Lutheran University Conservation and Management: Andrew C. Kasner, Texas A&M University Environmental Science: Cindy Contreras, Texas Parks & Wildlife Department Freshwater and Marine Science: Hudson DeYoe, University of Texas-Pan American Geology and Geography: Jeff Pittman, Lamar University Mathematics: Benjamin J. Sultenfuss, Stephen F. Austin State University Physics: Robert Hamilton, Angelo State University Science Education: Julie F. Westerlund, Southwest Texas State University Systematics and Evolutionary Biology: Allan Hook, St. Edward’s University Terrestrial Ecology: Monte Thies, Sam Houston State University Threatened or Endangered Species: Donald L. Koehler, Austin Parks and Recreation Dept. COUNSELORS Collegiate Academy: Jim Mills, St. Edward’s University Junior Academy: Vince Schielack, Texas A&M University Nancy Magnussen, Texas A&M University SMITHSONIAN INSTITUTION LIBRARIES 3 9083 01402 2867 PERIODICAL POSTAGE PAID AT LUBBOCK TEXAS 79402 RETURN SERVICE REQUESTED THE TEXAS JOURNAL OF SCIENCE PrinTech, Box 43151 Lubbock, Texas 79409-3151 002AAU7904 ROOM 25 NHB SMITHSONIAN INSTITUTION LIBRARY -ACQUISITION S *(SMIV) WASHINGTON, DC 20560-0154