SIDA Sis" VOLUME 8 NUMBER 1 JUNE 1979 CONTENTS Charles Mohr and plant life of Alabama. L. J. Davenport. ] Taxonomy and distribution of Gentiana (Gentianaceae) in Mexico and Central America. Il. Sect. Chondrophyllae. James S, Pringle. 14 Revision of North and Central American Najas (Najadaceae). Robert R. Haynes. 34 Bassia (Chenopodiaceae) in North America. Scott L. Collins and W7ll H. Blackwell, Jr. 57 A new species of Erigeron (Compositae) from Coahuila. Guy L. Nesom. 65 Vascular flora of Cedar Gap Lake and environs, Webster and Wright counties, Missouri. Ralph L. Thompson. 71 Taxonomic notes and new sections of Clitoria subgenus Bractearia (Leguminosae). Paul R. Fantz, 90 New species of Clitoria subgenus Bractearia section Bractearia (Leguminosae) from Peru and Ecuador. Paul R. Fantz. 25 Floristic and geologic aspects of indian middens in salt marshes of Hancock County, Mississippi. Lionel N. Eleuterius and Ervin G, Otvos, Jr. 102 NOTES. Notes on three species of ee ssum from North eee 113—New state records for Ranunculaceae in the southeastern United States. 114— Cuphea carthagenensis (Jacquin) MecbHaee correct orthography 114—Zamia (Cycadaceae) new for Georgia. 115—Galium species new to the southeastern United States. 116—Notes on Mitracarpus hirtus (M. villosus) and Scirpus Pai senitus in Mississippi. 117. Documented plant chromosome numbers 1979:1. 119 NOTICE US ISSN 0036-1488 SIDA, CONTRIBUTIONS TO BOTANY Founded by Lloyd H. Shinners, 1962 Editor & Publisher Wm. F. Mahler SMU Herbarium Dallas, Texas, 75275 Associate Editor Assistant Editor John W. Thieret Barney L. Lipscomb Northern Kentucky University SMU Herbarium Highland Heights, Kentucky, 41076 Dallas, Texas, 75275 Guidelines for contributors are available upon request beginning with volume 8. Subscription: $6:00 (U.S.) per year; numbers issued twice a year. Sida, Contributions to Botany, Volume 8, Number 1, pages 1-121. Copyright 1979 by Wm. F. Mahler CHARLES MOHR AND PLANT LIFE OF ALABAMA’ L. J. DAVENPORT Ecology and Systematics Section, Debartment of Biology, The University of Alabama, University, Alabama 35486 Charles Mohr (1824-1901) was one of the foremost botanists of the southern United States during the late nineteenth century. His stature as an early southern botanist is overshadowed only by that of A. W. Chapman, author of numerous editions of the Flora of the Southern States. Like Chap- man, Mohr was both a resident botanist and, for the most part, an amateur botanist. His 40 years of study centered in Mobile, Alabama. These studies culminated in the publication shortly after his death of Plant Life of Alabama, which remains as the only published attempt to describe the flora of the entire state. During an investigation of the contents of the herbarium of the Geological Survey of Alabama (Mohr Herbarium), I realized that very little was known about either Mohr’s life or his scientific works. It is the purpose of this paper to examine the life of Mohr, his scientific works, and the circumstances lead- ing to the publication of Plant Life of Alabama. Facts contained in the following biographical sketch were drawn largely from Hoffmann (1887), Scribner (1893), Smith (1901b, c), and Tracy (1901). Supplementary facts were garnered from less extensive works and from letters exchanged between Mohr and Eugene Allen Smith, then State Geologist of Alabama. These letters are stored in the Special Collections of the Amelia Gayle Gorgas Library, The University of Alabama. The sources for other letters cited are included with the references. SKETCH OF THE LIFE OF CHARLES MOHR Charles Theodore Mohr (originally, Karl Theodor Mohr) was born De- cember 28, 1824 in Esslingen, Wiirttemberg, Germany, the son of Louis M. and Dorothea Mohr. He spent his youth in Esslingen and at the Cloister Denkendorf, where his father had obtained an interest in a chemical factory. His early interest in plants is linked to his great uncle, who was a pensioned forester of the district, and to a local pastor who was interested in nature studies. In the fall of 1842 Mohr entered the polytechnic school of Stuttgart, where his studies concentrated on chemistry and natural sciences. These studies were supplemented by weekend geological and botanical excursions, during ‘This paper is adapted from a thesis submitted to The University of - labama in partial fulfillment of the requirements for the degree of Master of Scienc SIDA 8(1): 1-13. 1979. (Ae ced « foc Ae ; oo pa Fig. 1. Portrait of Charles Mohr, from Smith (190Ic). 3 which Mohr made notations of the geological formations and plant associa- tions observed (Hoffmann 1887). During the summer of 1845 Mohr was assigned to arrange the herbarium duplicates of the institute and to distribute these specimens to the secondary schools in the area. While pursuing this work he met August Kappler, who was engaged in the collection of natural history specimens from Dutch Guiana (Surinam). In November, 1845, he left with Kappler on an expedi- tion to explore the headwaters of the Surinam River. The party arrived at the coast of Guiana near the middle of March, 1846. Mohr was only able to spend seven months in South America, however, as an attack of dysentery and fever confined him to camp—the first attack of the ill health that would always plague him. According to Robinson (1903), Kappler “heartlessly left him behind, to live or die, while he [Kappler} pushed onward in his ventures.” In March, 1847, Mohr returned to Europe and soon found employment as a chemist in Brunn, Austria. However, the political turmoil of 1848 forced the business to close, while it also eliminated Mohr’s hopes for a higher education (Mohr 1879). At this time Mohr and his older brother immigrated to the United States, arriving in New York in October, 1848. After a short stay in Philadelphia, they moved on to Cincinnati, where Mohr again found employment as a chemist. In March, 1849, Mohr set off with about 50 other young men as a member of the “Cincinnati Gold Mining and Trading Company” to seek his fortune in the gold fields of California. After outfitting in Independence, Missouri, the group headed westward in mid-April with 10 wagons. The journey be- came increasingly difficult, and at Fort Laramie many belongings had to be abandoned. Ac this point Mohr traded his gun to a Sioux Indian in exchange for a pack horse to carry his books and plant collection (Hoffmann 1887). After splitting into parties of 10, the group reached Fort Hall in July. It was here that the wagons and baggage had to be given up, and Mohr dumped his books and plant collection into the Snake River. The parties reunited in Sacramento near the middle of August, after a trip totaling 110 days. Mohr and his party met with some success at placer mining in California. However, again, in the spring of 1850 Mohr fell ill to mountain fever and decided to return east. He and some others left the mines in September, 1850, for San Francisco, where they booked passage for Panama. While traveling across the Isthmus, the party’s pack mules—along with Mohr’s collection of California plants—were stolen, and Mohr again fell ill to fever. He finally reached New Orleans and traveled to Cincinnati, arriving toward the end of December, 1850. He and his brother purchased a farm in Clark County, Indiana, and he married Sophie Roemer on March 12, 1852. Various attacks of rheumatism soon convinced Mohr that he was not suited for a farmer's life, so he established himself in the drug business in Louisville, 4 Kentucky. Here he became acquainted with the bryologist Leo Lesquereux and resumed his botanical studies, concentrating on the mosses. However, again his health failed, and Mohr was advised by his physician to move to a warmer climate. He moved first to Louisiana, then to Vera Cruz, Mexico, and in early 1857 established himself in the highlands of Mexico between Cor- doba and Orizaba. Again, political turmoil forced him to move on, and he arrived in Mobile, Alabama, in the autumn of 1857 Mohr remained in Mobile for the next 40 years, where he founded a suc- cessful drug business. This business suffered much during the Civil War, during which Mohr was employed by the Confederate government in the manufacture of drugs from native resources and the testing of medicines smuggled in from Europe. He remained in the drug business until 1892, when he turned the management over to his son, Charles A. Mohr. Mohr lived in Mobile for eight years following his retirement, and Schenck (1955 ) describes him during this time as “an old and broken man, who spoke Eng- lish with a Wurcemberg ian accent, although he had lived in the United States for forty years. In March, 1900, Mohr and his wife moved to Asheville, North Carolina, both to provide a change in climate and to give him the opporunty to work in the Biltmore Herbarium. He died in Asheville on July 17, 1901 — SCIENTIFIC WORKS OF CHARLES MOHR The scientific works of Charles Mohr can be summarized as being scien- tifically accurate but leaning toward the practical and economic aspects of botany, agriculture, and geology. In his biographical sketch, Smith (1901c) states: From about 1878 the results of these [Mohr’ - scientific investigations began to be made public in a series of articles, at first practical and economic only, afterwards more strictly scientific and specialized, but a directed toward the imparting of useful knowledge to his fellowmen. As with his collections, primarily intended t illustrate some feature of our natural resources, they grew in breadth and comp ness until they became illustrative of monographs. Oo = fod Bi These interests and goals adequately sum up the scientific works of Charles ohr. Tracy (1901) states that Mohr’s first scientific paper was a lecture pre- sented in Vienna in 1847 on the geology of the Surinam region. Like most of his peers, Mohr was interested in other aspects of the natural sciences besides forest botany, and he included mineralogy, geology, and conchology among his interests. He contributed a short paper concerned with both geology and anthropology to the 1881 Smithsonian Institution Annual Report (Mohr 1883a Smith (1901b) states that during the summer of 1876 Mohr examined the gold reserves of the metamorphic region of Alabama, and that the floristic notes that he took during this study were published in Berney’s Hand Book. Smith further states that the collection of minerals from this trip was ex- hibited in Mobile in 1876 and in Atlanta in 1881, and that the ieee was eventually sent to the U. S$. Department of eis A report by Mohr on the economic geology of Alabama was issued in 1878. Mohr’s earliest contributions to North American botany involved collec- tions of mosses and ferns. As mentioned previously, Mohr met the bryologist Leo Lesquereux while at Louisville, and Scribner (1893) states that Mohr “greatly assisted Lesquereux and James by furnishing material for their work on the mosses of North America.” Smith (1901b, c) adds that Mohr’s col- lection of ferns of southern Alabama was sent to Daniel C. Eaton and incor- porated into his Ferns of North America. Mohr also published ae short papers on bryology in the Bulletin of the Torrey Botanical Club in 1874. Although he published no more on the subject, Mohr’s interest in ena must have continued throughout his life, as F. S. Earle read a paper by Mohr on the mosses of Alabama to the 1899 Columbus botanical meetings (Anony- mous 1899). Mohr’s interests always leaned toward the practical side of science, and these interests brought him to perform numerous investigations for the Grange on methods of improving the soils of Alabama. These investigations included the assessment of the value of various types of wood ashes, pine straw, leaves, and cotton seed hulls as soil builders. The results of these experiments were never published. Another primary interest of Mohr’s was the examination of plants intro- duced to North America on ships’ ballast. By living in the international port of Mobile, Mohr was able to comb the ballast grounds for newly introduced species. His preliminary findings were published in the Botanical Gazette in 1878, and many detailed notations of plant introductions and dispersals are found in Plant Life of Alabama. Many plants collected by Mohr and de- posited in the Mohr Herbarium are simply marked “ballast ground” and include the date of their first introduction. Botany, however, was a recreational activity for Mohr, as his training was that of a chemist and his business was based on drugs. He combined his vocation and avocation in a number of articles in the Proceedings of the American Pharmaceutical Association, as well as in the German Pharma- ceutische Rundschau. His “Medicinal Plants of Alabama” was published both in English and in German (Mohr 1890a, b). His knowledge of medic- inal plants was well known, and in 1897 Mohr was appointed to the subcommission of the Pan-American Medical Congress to study the Kener can medicinal flora (Anonymous 1897). A quick glance at Mohr’s publications indicates that his interests lay pri- marily in forestry and forest products. These interests led to his many popular articles published in Garden and Forest and in the Rundschau, as well as dresses to various meetings, such as the American Cotton Planters Asso- ciation (Mohr 1883c). He was in charge of several natural history exhibi- 6 tions at regional expositions, including the New Orleans exposition of 1884. Mohr (1883b) published a list of the natural resources displayed at the Louisville exhibition for the Louisville and Nashville Railroad which Scrib- ner (1893) described as “one of the few papers of its kind which possesses real scientific merit.” In 1880, Mohr was contacted by Charles Sprague Sargent to investigate the forests of the Gulf states. Although he expressed concern about receiving credit for his work (Mohr 1882a), Mohr’s results were published in the ninth volume of the Tenth United States Census. This work marked the be- ginning of a long association with the Department of Agriculture, which culminated in the writing of a series of extensive monographs on southern trees of economic importance. One such report (Mohr 1896a) treated the southern pines and included a study on wood anatomy by another author. “Notes on the Red Cedar’ was published posthumously. Smith (1901b) reported that at the time of Mohr’s death the monographs on Bald Cypress and Juniper were also in press, and that one on the oaks had been completed, but none of these papers has ever been published. Mohr was a member of many pharmaceutical and scientific associations, and he played an active role in most of them. As enumerated by Tracy (1901), Mohr was an honorary member of the state pharmaceutical asso- ciations of both Ohio and Louisiana, and ts listed as a member of the Ameri- can Pharmaceutical Association from 1871 onward (Anonymous 1885). Mohr's letters indicate his participation in two revisions of the Pharma- copeia (Mohr 1890c, 1900a). As for botanical associations, Mohr was a corresponding member of the Philadelphia Academy of Natural Sciences, the Massachusetts Horticultural Society, and the Torrey Botanical Club, and a fellow of the American Association for the Advancement of Science. He is known to have presented papers to the 1891 and 1892 meetings of the latter Association (Anonymous 1891, 189 Mohr’s most active participation, however, was in the field of forestry, and especially as a member of the American Forestry Congress. At the Cincinnati meeting in April, 1882, he presented a paper, “The Southern Pine,” and was appointed to chair a committee “to report upon Forest Fires, and the Injuries to Forests by Cattle’ (Anonymous 1882a). At the Montreal meeting in August, 1882, he was appointed temporary treasurer and a mem- ber of three committees, and also presented a paper (Anonymous 1882b). Through his many activities, Mohr was able to meet and maintain contact with many of the prominent scientists of his day. He corresponded frequently with N. L. Britton, largely concerning questions of nomenclature and the identification of sedges, and these letters can be found in the archives of the New York Botanical Garden. The Smithsonian Institution Archives contain his many letters to G. R. Vasey, J. N. Rose, S. M. Tracy, and C. L. Pollard. In letters to Smith, Mohr (1882b, c) mentioned meeting A. W. Chapman while working in Washington and meeting George Engelmann 5 while staying with Sargent in Massachusetts. His correspondence and con- sultation with Chapman continued until the latter’s death and is related in a biographical sketch of Chapman (Mohr 1899a). In a different letter, Mohr (1898) mentioned that Gifford Pinchot hoped to meet him, and his association with another prominent forester, George B. Sudworth, is described in an earlier publication (Mohr 18922). PUBLISHED WORKS OF CHARLES MOHR The following is a list of the published works of Charles Mohr. Fifty-five entries are listed, and, although it is hoped to be complete, additional works may be found, especially in German journals. This list does not include re- printings or re-editions, nor does it include the many abstracts listed by Harper (1928). Journal title abbreviations follow Lawrence et al. (1968). An asterisk (*) indicates a work that I have not seen. mdeinons to the bryology of the ae States, I-V. 1874. Bull. Torrey Bot. Club -8, 12-13, 34-35, 49-50, 50- On Pynanthemum lintfolinm ae its ial constituents. 1877. Proc. Amer. Pharm. Assoc. 24:5 Foreign ies edad into the Gulf states. 1878. Bot. Gaz. (Crawfordsville) 3 :42-46. *Economic geology of Alabama: report on the minerals of Alabama, exhibited at the se annual fair of the Mobile agricultural, horticultural, and mechanical fair . 1878. Gulf ee 1: 1-537, The rte of Alabama, and their products. 1878. Pages 221-235 im Saffold Berney, ed. Hand book of Aiea. va Register Press, Mobile, Ala. 338 The grasses, and other forage plants of Alabama: i een naturalized, nd culti- vated. 1878. Pages 236-247 in Saffold Berney, ed. Hand book of Alabama. Mobile Register Press, Mobile, Ala. 338 pp Preliminary list of the plants growing without cultivation in Alabama, from the collections made by Eugene A. Smith, Tuscaloosa, and Chas. Mohr, Mobile, Ala. Analysis of Eriodictyon See £880: Proc. Amer. Pharm. Assoc. 27:736-740. Rhus cotinoides, Nutt. 1882. Proc. Acad. Nat. Sci. Phil ae 1882:217-220. On the distribution of the more ee forest trees in the Gulf region. 1882. Amer. J. Forest. 1:78-88, 120-126. On distribution of the more oo forest trees in the Gulf region. 1883. J. Forest. 1:179-184, 200-2 ees lt ee ay and aan ov in Alabama. 1883. Annual Rep. Smith- sonian Inst., 1881:617 List of trees ee plants characters of each region of the state. 1883. Rep. Alabama Geol. Surv., 1881-82 :291-2 The natural resources of hn displayed in the exhibit of the Louisville and Nashville Railroad Company, at the ee Exposition, Louisville, Kentucky. 1883. Daily Register Press, Mobile, Ala. 22 On Quercus durandii, Buckley. 1883. Proc. ee ar Sci. Philadelphia, 1883 :37-38. The lands of the Louisville and Nashville Railroad in Alabama the scttler. 1884. Roberts and Son, Birmingham, Ala. 16 pp. *Rare and little known trees and shrubs of Alabama. 1884. Trans. Mississippi Valley ort. Soc. 2:216-219. *Ueber die eee der terpentin liefernden pinusarten im suden der vereinigten as homesteads for staaten und ueber die gewinnung und Gece ue terpentin. 1884. Pharm. Rundschau (Berlin & New York) 2:163-166, 187- The forests in the vicinity of Mobile. 1885. Pages ns a in Sub Rosa [pseudonym for Paul Ravesies}, ed. Scenes and settlers of Alabama. No publisher. 120 ae of forest ae — exas, Louisiana, Mississippi, Alabama, and Floridal: 885. Pages 66-67 . V. Riley, ed. Fourth report of the United States Ento- eens jeans *Ueber drei vercinzelte burger ie — see der nordamerikanischen sudstaaten. 1887. Pharm. Rundschau (Berlin & New York) 5:8- *Untersuchung der blatter von See triacanthos, L. 1887. Pharm. Rundschau Berlin & New York) 5:250. fae of the United States. 1887. Pap. New Orleans Acad. Sci. 1:19-39. Winter in Mobile. 1888. Gard. & Forest 1:16. The ea forests of the South. 1888. Gard. & Forest 1:34-35. Spring in Mobile. 1888. Gard. & Forest 1:88. The long-leaved pine. 1888. Gard. & ae 1:261- *Ueber die verbreitung der pflanzen durch thiere. ee Pharm. Rundschau (Berlin & New ae 6:177-181. The pecan tree (Hickoria pecan). 1889. Gard. & Forest 2:569-570 The lz vTee}st aaduce to the shrubs of eastern North America. 1889. Gard. & Forest The medicinal plants of Alabama: systematic list of the medicinal plants occurring within the limits of the state, with notes on their distribution and oe ume of collecting the parts used. 1890. Mo - pe Press, Mobile, Ala. 17 Vegetation in southern Alabama. 1890. . & Forest 3:140, 212 Pinus glabra. 1890. Gard. & Forest 3.295. The Florida spruc e pine, 1890. Gard. & Forest 3:402-403. * Hickoria olivaeformis. 1890. Pharm. Rundschau (Berlin & New York) 8:56-59. *Die medicinischen pflanzen von Alabama. 1890. Pharm. cere (Berlin & New York) 8:240-243, 257-262. *Vegetation of Louisiana and adjoining regions, and its products, in relation to een and allied siddnstiies, 1891. Pharm. Rundschau (Berlin & New York) 9:132-135, O= The general features of the vegetation of Louisiana and adjoining region, and its products in relation to pharmacy and allied industries. 1892. Proc. Amer. Pharm. Assoc. 39:76-84 Our remaining hard-wood resources. 1892. Engineering Mag. 4:378-385. Variation in the leaves of Clematis reticulata and other notes. 1892. Bull. Torrey Bot. Club 19:308-3 *Die gebirgsflo ora Mabe 1892. Pharm. Rundschau (Berlin & New York) 10:253. The mountain flora of Alabama. 1892. Gard. & Forest 5:507-508 The distribution of some forest trees in the southern states. 1893. Gard. & Forest 6:372-373. *Die walder des sudlichen Alabamas. 1894. Pharm. Rundschau (Berlin & New York) 21 *Die walder det alluvial region des Mississippi in den staaten eee Lo und Arkansas. 1895. Pharm. Rundschau (Berlin & New York) 13:1 *Die walder des sudlichen Alabamas. 1895. Pharm. Rundschau (Berlin & eae York) jon *Ueber das vorkommen des balsams von Liguidambar styraciflua L. 1895. Pharm. The timber pines of the southern United States: with an introduction by B. Fernow, and a discussion of the structure of their wood by Filiberr Roth. U.S.D.A. Div. Forest. Bull. 13. 160 pp. y Notes on some undescribed and little known plants of the Alabama flora. 1897. Bull. Torrey Bot. Club 24:19-28. Report on the forests of Sand Mountain: 1898. The Forester 4:211-215. Notes on some ae ee little known plants of the Alabama flora. 1899. Bull. Torrey Bot. Club 26:11 Alvin Wentworth ean 1899. Bot. Gaz. (Crawfordsville) 27:473-478 Plant life of Alabama: an account of the distribution, modes of association, and adaptations of the flora of Alabama, together with a systematic catalogue of the #12921. Plant life of Alabama. 1901. Alabama Geol. Surv. Monogr. 5. 921 pp. *Notes on the red cedar. 1901. U.S.D.A. Div. Forest. Bull. 31. 37 pp. PLANT LIFE OF ALABAMA Mohr's major scientific work was Plant Life of Alabama, published July 31, 1901—just two weeks after his death. This work was pro Hee with the assistance of the Geological Survey of Alabama and its head, Eugene Allen Smith, and represents an elaboration of the “Preliminary List’ from the collections of Mohr and Smith (Mohr 1880). Letters exchanged between Mohr and Smith fail to indicate when the deci- sion was made to begin work on a complete flora of Alabama. However, in 1882, after completion of his field work for the Census, Mohr (1882d) wanted “to be left quietly at home’ until he had completed “our Alabama Flora.” Four years later, Mohr (1886) mentioned working on the Flora, and by 1889 he was ready to add the lower plants to it (Mohr 1889). Concentrated work on the project was not begun until 1891, when ap- proval of funding was granted by the governor of Alabama (Mohr 1891). Under the terms agreed upon, Mohr received $50 per month as a retainer, with the total amounting to $5 per finished page "eh: 1892b, 1896b). Mohr was able to devote most of his time to the Flora after his retirement from the pharmacy business in 1892, although commitments to the U.S.D.A. had to be met also. In 1896, Mohr asked for and was granted a temporary reprieve from his Department of Agriculture duties (Mohr 189Ge). Work on the Flora accelerated in 1896. In that year a botanist at Auburn, Alabama—P. H. Mell—published a bulletin describing the plants of Alabama (Mell 1896). In this work, Mell utilized data obtained from Mohr. The Botanical Gazette (Anonymous 1896a) accused Mell of undermining Mohr’s work, stating that “we cannot believe that this extensive information was obtained from our good friend with the expressed intention of anticipating his own flora. In other words, Dr. Mohr must have granted a favor that has been abused.” Although a later article (Anonymous 1896b) in that journal exonerated Mell from blame, a series of letters exchanged between Mohr and Smith indicate that suspicions of Mell’s abuse were valid (Mohr 1896c; Smith 1896b). Mohr soon after decided to treat the entire Mell affair with “silent contempt” (Mohr 1896d). Later in 1896, Mohr indicated additional pressure to complete and publish his work from such botanists as John K. Small and Thomas H. Kearney. 10 He stated (Mohr 1896f): “By the activity with which of late the younger northern botanists begin to explore these parts, they are apt to take the wind out of my sails on many points.” For this reason he decided to quickly publish his “novelties” in the Bulletin of the Torrey Botanical Club. He was further dismayed by the publication of Britton and Brown's Illustrated Flora, as he felt that it embraced a large number of southern plants (Mohr 189G6h). Although Mohr relied heavily upon his own experience, knowledge, and herbarium in writing Plant Life of Alabama, many other experts and col- lections were consulted. An updated catalogue of the plants contained in the Geological Survey collection was used as a working model, and this catalogue was sent to N. L. Britton for revision to fit the Rochester rules of nomencla- ture (Mohr 1893a). Mrs. N. L. Britton asked to examine the portions of the manuscript concerning the mosses (Mohr 1899b). Specimens of the genus Aster were sent to Prof. Burgess in New York, while the grasses and Silphium were shipped to Columbia University for verification (Mohr 1897a, b). Mohr conversed with C. H. Merriam on plant distributions within the state (Mohr 1894). The Engelmann Herbarium at St. Louis was consulted, as well as the herbarium and libraries at Cambridge, Massachusetts (Mohr, 1893b, 1895). Mohr constantly incorporated information from the most recent publica- tions, even those published while Plant Life was in press (Mohr 1901a). In a letter to Tracy, Mohr ( listed his last-minute revisions and consultation of the type material at the National | Herbarium, with all changes made at “no matter what sacrifice of time and trouble. Plant Life of Alabama was originally planned to be one of the publications of the Geological Survey of Alabama, and Mohr was partially supported by Survey funds from 1892 on. However, it became apparent to Smith that publication by the Survey would be far too costly and might necessitate the elimination of large parts of the work. He suggested that a publication arrangement be made with the Smithsonian Institution or with the Depart- ment of Agriculture, and that an extra edition with an altered title page be printed as an Alabama state document (Smith 1896a, c). Such an agreement was reached to print the work as a Contribution from the U. S. National Herbarium (Mohr 1896g). Publication by the U. S. National Herbarium, however, resulted in a great number of problems. In Smith’s opinion, the cost of publication became exorbitant, and he expressed regret at having the work published in Wash- ington (Smith 1900). Although Mohr was informed in July, 1897, that the Printing Office was ready for his manuscript, the work was not published until four years—as well as Mohr’s life—had elapsed. Mohr (1900c) could rightfully claim “to have been among the first (if not te first) to treat the flora east of the Rockies in respect to its ecological relations,” but delays in printing his work allowed others to assume this distinction. ll In its final form, Plant Life of Alabama is an expanded checklist of all plants known to occur in the state. Each species or variety of plant, from slime mold to angiosperm, is listed with a bibliographic citation, important synonyms, distribution, occurrence in the state, type locality, and the loca- tion of representative specimens. A tabular summary of the plants follows the checklist, as well as a list of cultivated species. Plant Life of Alabama was lauded as “the most One and philosophical local flora ever published...” (Tracy 1901). It is more than a mere check- list of plants, because the first 137 pages are oe These pages include accounts of the history of Alabama botany, physiographic features of the state, plant distributions and associations, the relation of the Alabama flora to other floras, and a discussion of introduced plants and their influences. The work thus includes sections covering most of Mohr’s interests. Plant Life of Alabama was planned to be the first of a two-volume work on the botanical resources of Alabama, with the second volume dedicated to the economic plants (Smith 1901la). While the economic volume was viewed as being more immediately valuable to the populace, it was felt that the systematic volume was needed first (Smith 1896a). The earliest mention of an Economic Flora of Alabama was made in 1887, while the first remittance for such work was forwarded in 1899 (Mohr 1887; Smith 1899). The first 350 pages of the work were to be devoted to trees and shrubs, with a sketch of forest botany, while short descriptions of plant families would be included to allow the work to be used as a systematic botany textbook (Mohr 1901b, c It is generally believed that Mohr’s work on an Economic Flora was prog- ressing well at the time of his death, although a search of Mohtr’s belongings by his son, Herman B. Mohr, failed to locate the manuscript. His son later implied that, due to his father’s failing mental faculties, actual work on the Economic Flora may never have begun (H. B. Mohr 1901a, b). If a manu- script was produced, it is possible that Harper used it as a basis for his very similar work (Harper 1928). However, an examination of the extensive Harper material at The University of Alabama, Gorgas Library, Special Col- lections failed to turn up Mohr’s manuscript. ACKNOWLEDGMENTS I wish to thank Dr. C. Earle Smith, Jr. and Dr. Robert R. Haynes for their help and encouragement in the completion of this project. REFERENCES a hela 1882a. The American Forestry Congress: Cincinnati meeting. Amer. J. Forest. 1:31- 1882b. The American Forestry Congress: Montreal meeting. Amer. J. ~ Forest. 1:64-68. 1885. Roll of aN — a Pharm. Assoc. 32:561—-583. 1891. Botanical club o . S. Bot. Gaz. (Crawfordsville) 16: a 1892. Papers presented to the botanical club of the A. A. A. S. Bot. Gaz. (Crawfordsville) 17: ak Rg _, «1896a. New t. Gaz ules her 29 E ; 1896b. Onn ene Bot. (Cr forse) 22 417-421. _ «1897. News. Bot. Gaz. ek omen 24:13 44, - 1899, A synopsis . proceedings of the bocanical organizations meet- ing | at Columbus, Ohio, August 17- 25, 1899. Pall tere Bot. Club 26:500-505. HARPER, R. M. 1928. fie botany of Alabama, part 2: catalogue of the trees, shrubs, and vines of Alabama, ise Ae economic neat and local distribu- tion. Alabama Geol. Surv. Monc Q.. 4 HOFFMANN, F. 1887. Karl ae Mohe: ‘eine biographische skizze. Pharm. Rundschau (Berlin ae oe 5 :4-— LAWRENCE, G. H. SGC ERe G. S. DANIELS, and H. DOLEZAL, eds. ane Boni perce Huntianum. Hunt Botanical Library, Pittsburgh, Penna. )63 p MELL, P. 1896. The flora of Alabama, part 5. Alabama Agric. Exp. Stat. Agric. MOHR, C. T. 1879. er to E. A. Smith nas Mobile, Ala., 21 May - 1880. ice list of the plants growing ae een in Ala- bama, from the collections made by Eu ae A. aes Tuscaloosa, and Chas. Mohr, Mobile — No publisher a _.. 1882a. fe to E. A. Smith from Mobile, Ala., 30 ae _____. 1882b. Letter to 5 A. Smith from a D.C. 5 Jul. as . 1882c. Letter to E. A ree ith from Geor n, D. S 11 re —_.. 1882d. Letter to E. A. Smith fro Pave in omery, Ala., 21 O t: 1883a. Aenea soapstone quarry and shell-heaps ‘in Nise Annual Rep. Smithsonian eae 1881 :617- 83b. The natural resources ‘of Alabama, displayed in the exhibit of the Louisville ma Nashville eee See at oe Southern Exposition, Louisville, ucky. Daily Register Press, Mobile eeene eT 1883c. Letter to E. A Smith fom ‘Mobile, Ala, 19 a = __. 1886. Letter to E. A. Smith fro Mobile, Ala., 26 S - —. 1887. Letter to E. A. Smith oan Mobile, Ala., 26 May _ . 1889. Letter to E. A. Smith from Mobile, Ala. 1890a. The medicinal plans of pagers Die list of the medicinal plants occurring within the limits of the state, with es on their distribution proper ime of collecting the nee used. Mobile bovine Press, Mobile, Ala. 17 _ ie medicinischen pflanzen von Alabama. Pharm. Rundschau (Be. “din & New See 8:240-243, 257-262. _ . 1890c. oe a E. A. ae from Mobile, 27 a : . ey Let o E. A. Smith from Mobile, iG 19 O 7 892a. Variation in the ae oe Clematis reticulata a other notes. Bull. Torrey | on Club 19 :308—30 - 1892b. Letter to E. A. van se Mobile, oe 2 May. a . 1893a. nes to E. A. Sm Mobile 28 May. a _. 1893b. Letter to E. A. Sm | om event Sea 6 Oct. 1894, ee to E. . Smith from Mobile, ‘Ala., 2 Oct. ________. 1895. Letter to E. A. Smith from Mobile, Ala., 22 Aug. = 1896a. The timber oe es of the southern United States: with an int duction ‘by B. E. Fernow, and a ere of the structure of their woods by Fili bert Div. Forest. Bull. 13. O pp. —_. 1896b. Letter to E. A. Sm th from Mobile, Ala., 4 Jan. 189G6c. Letter to E. A. Smith from Mobile, Ala, 17 Apr 1896d. Letter to E. A. Smith from Mobile, Ala., 23 Apr _, «189G6e. Letter to E. A. Smith fro obile, , 1O May _ 1896f. Letter to E. A. Smith from Mobile, Ala., 27 Oct _.. 1896g. Letter to E. A. Smith from Mobile, Ala., 19 Dec. «18 96h. te - E. A. Sm ie from Mobile, Ala., ur . 1897a. L A. Sm bile, oe la., anes _ o E. from Mo 2 a . 1897b. ee . E. A. Smith ane Mobile, Ala., 6 TL 13 ________. 1898. Letter to E. A. Smith from Mobile, Ala., 28 Jul. —_____. 1899a. Alvin Wentworth Chapman. Bot. Gaz. (Crawfordsville) 27:473- 478. Seo 1899b. Letter to Mrs. N. L. Britton from Mobile, Ala., 27 Jan. Archives, New y York Botanical Garden. __,s«1900a. Letter to E. ‘A. Smith — Asheville, N. C., 25 Apr —_____. 1900b. Letter to S$. M. Tracy from Asheville, N. ve Jun. Archives, Smithsonian Aon ae Mills ag Papers, 1895-1914 _.,s«21900c. Letter to E. A. Smith from ene N. C., 7 Oct __. «1901a. Letter to E. A. ae from Asheville, N. C., 22 Jan. ___. 1901b. Letter to E. A. Smith from As Reville , N. C., 1 Mar. . 1901c. Letter to E. A. ei ne Asheville. N. C., 8 Mar MOHR, iL 1901a. ma to " A. Smith from Asheville, N. C 20 Ju. : Olb. Let o E. Smith from Asheville, N. 1903. Dr. sae Mohr, the South’s oa Ties Democrat an. Fan Onae a Vs 1055. The Biltmore story. Minnesota Historical Society, St. Paul. 22 SCRIBNER, F. L. 1893. Southern botanists. Bull. Torrey Bot. Club 20:315—-334. SMITH, E. A. 1896a. Letter to C. T. Mohr from University, Ala., 2 : _______. 1896b. Letter to C. T. Mo . ae Univ ersity, Ala., 19 Apr. ate . Moh m Uni Jul. ________. 1899. Letter to C. 'T. Mohr fees Sas Ala., 15 Nov. «1900. Letter to C. T. Mohr from University, Ala., 5 Apr 1901a. Letter of aie Pages iii-iv in C. T. MGRE: Dian life of Ala- ~ bama. Alabama Geol. Surv. Monogr. 5. 921 pp. Se ee Olb. Charles Theodor ore eee Pages v-xii 7 C. T. Mohr. Plant life of Alabama. Barca Geol. Surv. Mon 292 = a Ae C. Biosenial oe “Of ‘Dr. Charles Mohr. Bull. Torrey Bot. Club 28:59 604. oe TRACY, S. M. 1901. Dr. Charles Mohr. Pl. World 4:167-170. TAXONOMY AND DISTRIBUTION OF GENTIANA (GENTIANACEAE) IN MEXICO AND CENTRAL AMERICA. I. SECT CHONDROPHYLLAE JAMES S. PRINGLE Royal Botanical Gardens, Box 399, Hamilton, Ontario, Canada L8N 3H8 This paper is the second of two monographs on Gentiana L. in Mexico and Central America. Delineation of the genus, terms pertaining to its morphology, and other introductory topics were discussed in the first paper (Pringle 1977), which also included a key to the sections of Gentiana in these regions. GENTIANA {sect.} CHONDROPHYLLAE Bunge, Nouv. Mém. Soc. Imp. Nat- uralistes Moscou 1:207. 1829. Lectotype species (Pringle 1978): Gentiana aquatica L. Sage Philippi, Florula Atacamensis p. 35. 1860. Type species (only species ted): Gentiana podocarpa Vee ae Griseb., as Varasia podocarpa Philippi Chondrophl (Bunge) A. Nels., Bull. Torrey Bot. Club 31:245. 1904 ar n.). Holubla Live & Live, Anales Inst. . Cavanilles 32:226. 1975. Type species: Sb pyrenaica L., as Holi ee (L.) Love & Léve. Non Holubia Hooker’s Icon. Pl. 15:59. 1884. Poesia: Live & Léve, Bot. se 131:385. 1978. Based on Holubia Live & Live, non Oliv © Plants monocarpic or perennial. Roots slender, branched; taproot persistent in annual species. Stems solitary or clustered, usually branched except in minute plants, decumbent to erect, slender, in many species less than 15 cm long, up to 30 cm in others. Basal and lower cauline leaves of most mono- carpic and some perennial species persistent, crowded, and broader than me- dian and upper cauline leaves. Leaf bases strongly connate-sheathing. Blades usually less than 1.5 cm long, scalelike to oblanceolate, oblong, or orbicular, usually carinate. Margins obscurely to prominently white-cartilaginous, en- tire or minutely denticulate. Flowers solitary or in nee condensed cymes, without involucral bracts, 5-30 mm long. Calyx tu or 15-veined, cylindric to funnelform, uncleft. Calyx lobes deltoid to oblong often carinate, with cartilaginous margins. Intracalycular membrane discontinuous or not developed. Corollas usually pale to deep blue, sometimes white, pale yellow, or violet, usually suffused externally with green and/or deep violet, often ‘Contribution No. 31 from the Royal Botanical Gardens, Hamilton, Ontario, Canada ~ SIDA 8(1): 14-33. 1979. ihe; photo-, thigmo-, and seismonastically closing. Corolla tubes cylindric to funnelform. Corolla lobes abruptly wide-spreading, shorter than tube, deltoid to ovate or oblong-ovate. Free portions of corolla appendages nearly sym- metrical, uncleft to shallowly or deeply bifid, or erose to fimbriate. Stamens equal, with straight filaments. Anthers free or loosely connate. Pollen grains oblate-spheroidal to prolate. Sexine striate, striato-reticulate, or nearly smooth. Ovaries compressed-ellipsoid, usually widest near middle, abruptly contracted at summit. Stigmas short, in most species sessile or subsessile, in some ele- vated by a well-developed style. Gynophores of some species greatly elongat- ing in fruit, remaining short in others. Carpels each with 1 dorsal and ventral veins. Ovules in 4 parietal zones corresponding to positions of ventral carpel veins. Seens ovoid to ellipsoid, often apiculate, wingless or with nar- row, incomplete wings. Testa striato-reticulate. Sectional description com- piled from Kusnezow (1895), Lindsey (1940), Nilsson (1967), Toyokuni (1963), and my own observations. Section Chondrophyllae occurs in the Pyrenees, the Alps, and the Carpa- thians; in the Caucasus, the Himalayas, Siberia, China, Japan, and Southeast Asia; in Indonesia, New Guinea, and the Philippines; from the Aleutians and the Brooks Range in Alaska through the Rocky Mountains to Arizona; in Mexico, Guatemala, and Costa Rica; and in the Andes from Venezuela to Tierra del Fuego. Throughout its range, this section is confined largely to mountainous regions, with many of its species occurring above treeline. The pollination of G. prostrata Haenke in Europe has been discussed by Kerner von Marilaun (1891), who concluded that although outcrossing was promoted by protandry, autogamy could readily occur within corollas prevented from opening by prolonged cloudy periods. Similarities in corolla morphology and in pistil and anther development indicate that these con- clustons are generally applicable to the species discussed in the present paper. The widespread phenomenon of thigmonastic corolla-closing in this section is usually assumed to be related to pollination, but its specific role has not been investigated. From the field aspect of open capsules of G. perpusilla Brandegee, i appears that the fruits of this and some other species of sect. Chondrophyllae function as splash-cups. In species with more elongate gynophores, seeds are probably dispersed as wind shakes the fruits (censer mechanism). Many species grow where rill action can further disperse the seeds after they have fallen. Long-distance dispersal evidently has also occurred, since the alpine habitats of the Mexican and Central American species are widely separated, are mostly of late Tertiary or See origin, and have no history of con- nection by continuous alpine vegetation Since its establishment by Bunge, sect. Chondrophyllae has been accepted as a section of Gentiana by most authors, except for those who restricted Gentiana to the five species usually Gare as its nominate section. Varasia Philippi and Chondrophylla (Bunge) A. Nels. were ae ae tO: SeGr. 16 Chondrophyllae only. Neither of these names has been used recently. Love & Léve (1975) divided sect. Chondrophyllae between two segre- gate genera. The “annual or biennial species with x = 9” were placed, along with sect. Crmnalis (Adans.) Dumort. (which consists of perennial species), in the genus Ciminalis Adans. emend. Holub (type species: Gentiana acaulis L., in sect. Ciminalis, as Ciminalis [sp.]). The “perennial species... with ...x = 13,” believed by Love & Love to comprise “about ten species of Eurasia, North and South iar ” formed Holubia Love & Love, later renamed Holucogentia Love Love. Weaver & Riidenberg 1975), however, reported 2 = 20 1 Dee called G. sedifolia, but prob- ably an unnamed, closely related species, in Venezuela. In Love & Love's generic concept, this species, perhaps with certain Asiatic species having n = 10, will presumably be treated as a third genus. Little can now be postulated about relationships within sect. Chondro- phyllae in the Western Hemisphere, because its diversity has largely remained unrecognized, Existing information on chromosome numbers and_ pollen morphology indicates that phylogenetic studies will require more such data. Subdivisions of sect. Chondrophyllae are therefore not considered here. In this study, I examined specimens in the following herbaria: BM, DS, ENCB, F, GH, HAM, K, MEXU, MICH, MO, MSC, NY, TENN, US, and WIS (abbreviations from Holmgren & Keuken 1974). Among these are collections each comprising several to many plants, representing well the variation that may occur within a population. Specimens from South America, northern North America, and Eurasia were among those studied, and G. perpusilla and G. prostrata were observed in the field. Consequently, it has een possible to compare the Mexican and Central American plants with those in other regions, and to distinguish, with reasonable confidence, taxo- nomically significant differences among populations from minor variations among individuals. All of the species discussed here have sometimes been equated with G. prostrata. Comparisons with specimens of G. prostrata from Colorado (Figs. 7, 13), Alaska (Fig. 19), and the type region in Austria, however, indicate that all of the Mexican and Central American plants are distinct from this species. Gentiana fieldiana, which resembles G. prostrata in its leaf shape and in its apiculate corolla lobes, differs in its wider white leaf margins, in ie more oblong shape and strict apical symmetry of its corolla lobes, in e greater adnation of its stamens, and in its usually cymose aa ee sedifolia resembles G. prostrata in its narrow leaf ma and in the long free portions of its filaments, but differs in its tome much narrower leaves, in its blunt corolla lobes, in its more copiously and more irregularly erose-serrate corolla appendages, and in its coarser sexine sculp- turing without well-developed longitudinal striations. The other species differ even more sharply from G. prostrata: G. beamanii in its narrower leaves and in the much shorter free portions of its stamens, G. perpusilla i? in its proportionately small corolla lobes and in its uncleft, nearly entire corolla appendages, and G. pumilio in its proportionately much broader eaves, in its wide and prominently denticulate leaf margins, and in its spheroidal pollen grains. KEY TO MEXICAN AND CENTRAL AMERICAN SPECIES la. moe pgs sonal a a spatulate to orbicular, many with 5 or more prominent primary veins 2-4 lesser ones; white leaf margins conspicuous, ca. 03 mm wide, those a upper leaves denticulate from below middle to apex; flowers often sessile in small clusters; calyx tube largely concealed by leaves; pollen grains spheroidal 4. G. lb. Leaves, except sometimes near base of stem, less densely crowded and_ propor- tionately narrower, usually with only 1 or 3 promin nent primary veins; white leaf solitary at end of main stems and branches; calyx tube not concealed; pollen grains subprolate to prolate. 2a. Leaves linear-oblong, usually more than 3 times as long as ee cue! to strongly divergent from near base; stamens becoming free in | 2/5 of corolla, with free portions, including anthers, 3-7 mm long; throat “of corolla with race color demarcation sedifolia 2b. Leaves oblanceolate to obovate, usually less than 3 times as nie as aes if proportionately longer then proximally appressed-ascending, only the distal portion spreading; stamens becoming ae near or above mide or corolla, with free portions up to 3 mm long, if diverging lower, then with free portions up to 2 mm long and with ae appendages cee throat of corolla with or without color differentiation. 3a. Stems erect except at base, with strongly ascending branches; internodes, except near base of stem, mostly as long as or longer than subtending leaves; leaf blades mostly over 5 mm long; throat of corolla with distinct tates 2. G. beamanii . Stem and branches variously spreading; internodes mostly much shorter than subtending leaves; many leaf blades less than 5 mm long; corolla throat Ww i 4a, White leaf margins and keel well defined, those of upper aoe distinctly denticulate; corolla lobes oblong- eaneualar, abruptly rounded to apiculate tip; free portions of Ae appendages erose-serrate; stamens cee in free near middle of corolla woo... cccccccccececcceeeecces sevens veeees G. freldiana dX > . White leaf margins aad ke el “obscure, generally entire or ee Howly un- dulate; corolla lobes ovate-triangular, tapering gradually to subacute - free portions of corolla appendages entire or nearly so; stamens becom free below middle of corolla 5. G. See TAXONOMIC TREATMENT lis Sanit SEDIFOLIA H.B.K., Nov. Gen. Sp. - ae ae need ext). Type: without locality [probably ECUADOR: Napo or Pichincha}: Humboldt & Bonpland s.n., no date [probably ne Jun 1802] ee P, photo MSC! isotype K! ). Figs. 1, 8, 14. Gentiana cespitosa Willd. ex Schultes, Syst. Veg. 6:185. 1820. Type: Duplicate Humboldt et al., cited above (B-herb. Willd.). (Ex char. and fide Kunth in Humboldt et al., 1823.) Gentiana ebineb opacenars Willd. ex Schultes, Veg. 6:185. 1820. Type: SUADOR: Chimborazo: Chimborazo, Seen et al, s.m., no date (B-herb. Willd.). (Ex char.) Ericala sedifolia (H.B.K.) G. Don, Gen. Hist. A: ne “1838” [1837]. As Seak edifolta var. casapaltensis J. Ball, J. L Soc., Bot. 22:49. 1885 PE ima: supra Casapalta [Casapalca], Ball m,, 22 Apr 1882 (holotype K! isotype GH!). Gentiana sedifolia var. [3 grandiflora Kusn., Trudy S.-Peterburgsk. Obshch. Fstestvoisp., Otd. Bot. 24(2):217. 1894. Syntypes: ECUADOR: Napo_ or Pichincha: Antisana, Lehmann, no. and date not cited (LE) and COLOMBIA: Magdalena: Santa Marta, Lehmann, no. and date not cited (LE); probable isosyntype, “Cordillera of Ecuador and Colombia,” Lehmann 623, 22 Dec 1880 (K! ) Plants usually 1.3—4 cm tall, occasionally up to 6 cm. Stems 1-many. Well- developed stems decumbent at least at base and often for much of length; longer stems usually much branched. Leaves linear-oblong, 5-10 mm long, 1.5-2.5(-3) mm wide, apiculate. White leaf margins inconspicuous and sometimes not sharply defined, mostly 0.02-0.06 mm wide distally on lower leaves, ca. 0.1 mm wide on upper leaves, entire or irregularly and minutely denticulate toward apex. Basal leaves persistent, crowded, less conduplicate than cauline leaves. Cauline leaves increasingly conduplicate and sometimes more or less arcuate On upper part of stem (except on very small plants), with midrib keels more distinctly developed (white portions up to 0.05 mm wide). Median and upper internodes often scarcely longer than sheathing leaf bases, sometimes up to 2—6 mm long. Calyx tube 5.5-8 mm long, ridged distally along median sepal veins. Calyx lobes ovate-triangular, erect proximally, sometimes more or less arcuate distally, conduplicate and nar- rowly carinate toward apex, (1.2—)1.8—4 mm long, acute, with white margins ca. 0.2 mm wide, widening near apex, and keel up to 0.05 mm _ wide; margins and keel entire. Corollas 9-20 mm long when closed; expanded limb 6-14 mm across. Corolla tube 6.5-17 mm long. Lobes broadly ovate- triangular, 1.8-4 mm long, as wide or slightly wider than long, obtuse to subacute. Free portions of appendages ca. 0.75 times as long and about as wide as lobes, broadly rounded to triangular, shallowly to rather deeply erose-serrate or shallowly several-cleft near apex, not distinctly bifid. Corolla tube yellowish-white; upper portion suffused with deeper yellow and often bordered and spotted with purplish-brown. Uppermost part of fused portion of corolla, lobes, and free portion of appendages pale to deep blue, with a narrow whitish zone next to yellowish eye, or occasionally de white throughout; exterior surface of lobes, except near inner margin, and adjacent portions of tube suffused with deep violet (scarcely or not at all in white- flowered forms) and also with green, especially near tips of lobes. Stamens becoming free at 1/4—2/5 total height of corolla; free portions (including anthers) 3-7 mm long. Pollen grains prolate. Sexine relatively coarsely striato-reticulate. Pistil tapering into style ca. 1 mm long. Capsule 3-5 mm on) ee i 2 eee arium specimens of oe spp. Ecuador). Fig. 2. G. beamanti aes rapes eo 3071, MSC, a ae large plant). G. sedifolia Cem io 3899, MSC). Fig G. 20 long, elevated well above marcescent corolla by gynophore 15-18 mm long ac maturity. Seeds ellipsoid, light brown, striato-reticulate, ca. 0.85 mm long, mm in diameter. Central American specimens examined: COSTA RICA: Cartago (all from Cordillera de Talamanca, near Cerro de la Muerte): not more specifically cua Carlson 3630 (F); Ciénega “3 de Junio,” at km 73 SE of San José on Carretera Panamericana, Jiménez M. 1993 (F, NY), Jiménez M. 3372 (F), Lent 405 (NY), and Anderson 7 Mori 231 (BM, DS, F, WIS); near Ojo de Agua, Williams et al. 28281 (BM, F, WIS); La Trinidad, be- tween oe 60 and 77, Molina R. et al. 17855 (F, ed Border of aed and San José: Cerro de la Muerte, Hetthaus 318 (MO). San José: Cerro de Buena Vista, Tonduz (Pinter exped.) 3498 (US); Cerro de las Vueltas, Standley & Valerio 43640 (F, US). PANAMA: Chiriqui: between Cerro Bine and IJtamut, Weston 10175 (MO; first report of Gentiana in Panama). In Central America, G. sedifolia is known only from high elevations in the Cordillera de Talamanca in Costa Rica and the adjacent Cordillera Cen- tral in Panama (Fig. 20). It is widely distributed in the Andes of South America, from Col ne and Ecuador south at least to the Cordillera Central of southern Peru (Fi Herbarium data eae that G. sedifolia in Costa Rica and Panama grows in boggy depressions and similar wet sites in meadows from 2275 to 3335 m elevation. Plants have been collected in flower from December through uly. In their original description of G. sedifolia, Humboldt et al. (1819) reported having seen this species “in montibus ignivomis Antisanae {Ecua- dor}, Puracé, Paramo de Almaguer [Colombia], etc., alt. 1600-1800 hex. [= 3072-3456 m}.” On the basis of minor geographic variations in this species, discussed below, it seems most likely that the type was collected on or near Antisana. Distinctive vegetative features of G. sedifolia throughout its range include the proportionately long-tapering leaves and the very narrow white margins. Distinctive floral features include the funnelform corolla tube, which widens more conspicuously from base to summit than the nearly cylindric corolla tubes of the other species described here, and the relatively low divergence and long free portions of the filaments. Pollen grains of G. sedifolia from Costa Rica closely resemble those of Ecuadorean specimens, the sexine in both being relatively coarsely reticulate, with poorly developed longitudinal striations. The two later species names cited above have generally been accepted as synonyms of G. sedrfolia, at least the first having been based on the same collection (Humboldt et al., 1823). Ten taxa, with 11 epithets, were de- scribed as varieties, subyacieries or forms of G. sedifolia during the 19th Century, but, except for the two cited above, their names seem to have been based. on specimens of other species. Because all of these names were based on South American plants, a thorough investigation of their status has not Zu been included in this study. More recent collections, however, have indicated that much of the variation in G. sedifolia occurs within populations. Con- spicuous variation exists in stem and internode length, but extremes can be found within a single population and even within a single collection, e.g., Couthouy in 1855 (GH), from Ecuador, and Pennell & Hazen 9996 (GH), from Colombia; some variation may occur among the stems of an individual plant. This variation, appearing to represent responses to seasonal factors and differences in microhabitat, is not recognized taxonomically here. Max1- mum flower size is greater in Ecuador than farther north, but there is considerable variation in all regions and no discontinuities are evident. Corolla color likewise varies within populations. Plants of G. sedifolia from the region of Antisana and Cotopaxi in Ecua- dor often have relatively long, much-branched prostrate stem bases, which appear brownish-black from the remains of old leaf bases, and from which adventitious roots arise. Such plants resemble the type collection and appear to be long-lived and probably recurrently flowering, corresponding to the description of G. sedifolia as perennial by Humboldt et al. (1819). Plants from other parts of the range of G. sedifolia generally have less well de- veloped prostrate stems, without blackish leaf remnants, and root systems comprising only the primary root and its branches. These plants appear to be relatively short-lived and monocarpic. Longevity and life cycles are, how- ever, difficult to determine in specimens from the tropics, and no year-round observations have been made on populations of this species. Therefore it is not certain whether these differences represent genetic differentiation, or whether climatic conditions near the Equator permit greater longevity of individual plants. Weberbauer (in Gilg, 1906), however, commented on the inadequacy of plant size and apparent longevity as bases for dividing G. sedifolia, noting that at a single locality in Peru plants of this species might exhibit all stages of intergradation from minute, unbranched, one-flowered forms with a single filiform root to large, much-branched, patch-forming plants. Therefore no variants based on apparent longevity are recognized here. 2. GENTIANA beamanii Pringle, sp. nov. Figs. 2, 9, 15. plus minusve erecti, 3.5-16 cm alti, internodis vulgo longioribus quam Caules foliis. Folia oblanceolata vel anguste spachulata: 5-10 mm longa, 1.5-2 mm lata, proximale fere ewan et distale patentia, marginibus albis usque ad 0.05 mm latis, apicem versus minute denticulatis. Tubus calycis 5-8 mm longus. Lobi calycis triangulares, 1.8—3 Pion. acuti, marginibus albis circa 0.15 mm _ latis. Tubus corollae cylindricus, 8. 61 m longus. Lobi corollae late ovato-triangulares 1.5—3 emarginatae, breviores quam lobi. Filamenta staminum sese liberantia aliquantum supra medium corollae totae, partibus libris antheribus inclusis circa 2.5 mm_longis. Pollinis grana subprolata vel prolata, sexinio striato-reticulato. Capsula 4-6 mm longa, in parte exserta. Type: GUATEMALA: aac: Sierra - los Cuchumatanes, aay ees N of Tojiah [Tojquia} at km 322 on Ruta Nacional 9N, Beaman 3899, 1 Aug 1960 (holotype MSC, isotypes GH, US). Paratypes: GUATEM MALA: Huehuetenango: pas vicinity of Chémal [Xémal], Sierra de los Cuchumatanes, Steyermark 50287, 8 Aug 19a) (i, GH). Plants 3.5-16 cm tall. Stems more or less erect throughout or basally decumbent, usually 1-6, each usually with 1-3 branches arising in basal half and reaching about equal height, sometimes with additional branches from upper axils. Leaves usually broader and more closely spaced near stem base; lowest leaves (sometimes shriveled at flowering time) elliptic-oblong or oblanceolate, 5-15 mm long, 1.8-4 mm wide, spreading or ascending, scarcely conduplicate. Median and upper cauline leaves oblanceolate to narrowly spatulate, 5-10 mm long, 15-2 mm wide, increasingly conduplicate up- ward; proximal portions (ca. 1/3 near middle of stem, ca. 2/3 near summit ) appressed-ascending, distal portions more or less arcuate-spreading. Leaf apices abruptly acuminate, apiculate. White leaf margins almost obsolete proximally, widening distally to ca. 0.08 mm, entire or sparsely and minutely undulate-denticulate above middle. Midrib keel of similar width, minutely denticulate above middle. Internodes distinct, all but lowest few usually 1-2.5 times as long as subtending leaves. Calyx tube 5-8 mm long, slightly ridged along midveins. Calyx lobes triangular, 18-3 mm long, acute to acuminate, scarcely to moderately conduplicate, erect or slightly arcuate- spreading distally, with white margins ca. 0.15 mm wide, entire or nearly so, and keel up to ca. 0.08 mm wide, entire or sparingly denticulate. Corollas 10-17 mm long when closed; expanded limb 6.5—12 mm across. Corolla tube 8.5-14 mm long. Lobes broadly ovate-triangular, 1.5-3 mm long, slightly wider than long, subacute to acute. Free portions of appendages ca. 0.8 times as long and about as wide as lobes, triangular, sparingly erose-serrate, usually emarginate to shallowly bifid. Corolla tube mostly whitish, blue above, with dark purple streaks on interior surface of throat, in 2 lines in each petal roper and 1 in each appendage; lobes and free portions of appendages medium blue; exterior surface with deep violet pigment between relatively pale margin and overlapping = (when c ne of next lobe, giving striped — _ — orollas fred portions eer anthers) ca. 25 mm long. Pollen grains subprolate to prolate. Sexine relatively finely striato-reticulate. Pistil tapering into style ca. 1 mm long. Capsule 4-6 mm long, about 2/3 exserted from marcescent corolla by gynophore 7-10 mm long at maturity. Seeds not seen. Gentiana beamanti is known only from a small area in Los Altos Cuchumatanes, Guatemala (Fig. 23). Both collections cited were made along small streams in meadows at ae of 3200-3750 m. iis species is named for Dr. n H. Beaman, author of several papers on the alpine flora of Mexico and See whose notes and comments on differences among the Central American species of sect. Chondrophyllae have been very helpful in this study. 1e elongate, erect stems and widely separated leaves of G. beamanii give this species an aspect unique among Central American species of sect. 23 Chondrophyllae. Its distinctness from plants treated here as G. fieldiana was first noted by Steyermark (in sched., Steyermark 50287), as follows: Corolla larger, broader, and with longer tube . . . corolla-lobes with deeper purple in outer lobes [sic] . . . also stems with more branching and with more flowers than in other smaller type.” Beaman (unpublished notes) like- wise contrasted these plants, noting that those here called G. beamanii had narrower leaves with less conspicuous white margins. Of the three Guatemalan species, G. beamantzi is the closest to G. sedifolia in floral morphology. It differs conspicuously, however, in its cylindric, abruptly flaring corolla tube and in the much greater adnation of its stamens. Figs. 4-7. Herbarium specimens — ems spp. Fig G. fieldiana ene 3071, MSC, a plant ca. average size). 5. G. pu Seite wie 2868, M) Fig. 6. G. perpusilla (Pringle aes Lae Fig. 7. G. prostrata ae Zoe MICH, from Colorado). Figs. 1-7 to same scale. 24 3. GENTIANA fieldiana Pringle, sp. nov. Figs. 3, 4, 10, 16. Caules ascendentes vel erecti, 1-4(—6.5) cm alti, vulgo pauciramosi. mae congesta. Folia basalia oblongo-oblanceolata vel elliptico-obovata, 4— = mm longa et 2-4 mm n saree marginibus albis circa 0.05 mm latis, integri . Tubus corollae cylindricus, 4.5-10 mm longus. Lobi corollae oblongo-triangulares, ee 2 mm longi, apiculata, azurei aut albi. Partes librae plicarum triangulares, eroso-serratae et vulgo emarginatae, breviores quam lobi. Filamenta staminum sese liberantia circa medium corollae totae, partibus libris antheribus oS circa 2 mm longis. Pollinis grana prolata, sexinio striato-reticulato. Capsula 3—5 mm longa, in parte vel in toto exserta Type: GUATEMALA: beta etenango: Chéemal [Xémal], at km 318 on Ruta Nacional 9N, Beaman 3071, 4 Aug 1959 (holotype MSC). gee : GUATEMALA; Huchuetenango (all from Los Altos Cuchumatanes): vicinity of Tunima [Tuinima}], Steyermark 48368, 7 Jul 1942 (F, GH); 2.5 mi E of San ae Ixtatan, Steyermark 49875, 31 Jul 1942 (F); near Paquix, Sharp 451025, 24 Dec 1945 (F, TENN); E of Tocquin {Tojquia}], Holdridge 2345, 27 Apr 1948 (US); Llano de Tierra Blanca, near oo to Todos Santos from Hane de San Miguel, near Chémal [Xémal], ca. 5 km W of km 311 on Ruta Nacional 9N, Beaman 3118, 5 Aug 1959 (MSC); 3 km § of road ae Llano de San Miguel and Todos Santos, from a point 2.5 mi W of Llano de San Miguel, Beaman 3975, 2 Aug 1960 (MSC); 7 mi N of Santa Eulalia along road to San Mateo Ixtatan, Breedlove 8596, 5 Feb 1965 (F); along road to San Pedro Soloma, 3 mi SW of San Mateo Ixtatan, Breedlove _ 6 Sep 1965 (F, US). Totonicapan: near Cumbre del Aire, on road between Huehu tenango and Siga, Standley 65939, 30 Feb 1939 (F); Tecum Uman Ridge at me 154 on Ruta Nacional No. 1, ca. 20 km of Totonicapan, Beaman 4205, 14 Aug 1960 (MSC). Plants 1-4(-6.5) cm tall. Stems 1-12, ascending to erect, those of larger plants usually with 1-4 branches at various levels and further branched near summit, bearing flowers in small, leafy cymes. Basal leaves (sometimes more or less marcescent at flowering time) spreading, scarcely conduplicate, with midrib keel well developed only near apex, broadly oblong-oblanceolate to elliptic-obovate, usually 4-9 mm long, 2-4 mm wide. Cauline leaves some- what narrower, increasingly conduplicate and carinate upward; upper ones carinate for most of length. Leaf apices acute, apiculate. White leaf margins and keel sharply defined, both 0.05—0.13(-0.2) mm wide, or slightly wider toward apex, at least upper leaves with keel and usually margins minutely denticulate. Leaves of most plants densely crowded throughout; those of taller plants often more distant on lower part of stem, with lower internodes up to twice as long as subtending leaves, but crowded on upper part. Calyx tube 5-8 mm long. Calyx lobes triangular, 1-1.5 mm long, erect, scarcely conduplicate, acute to acuminate, with white margins ca. 0.05 mm wide, entire, and keels scarcely developed or up to 0.04 mm wide. Corollas 6-12 mm long when closed; expanded limb 5—8 mm across. Corolla tube 4.5-10 mm long. Lobes oblong-triangular, 15-2 mm long, apiculate. Free portions of appendages ca. 0.75 times as long and about as wide as lobes, triangular, erose-serrate, sometimes emarginate. Corolla tube mostly whitish; lobes, free Figs. 8-9 a-c, 10-13 b Cc b Figs. 8-9. Gentiana spp. G. sedifolia. Fig. 9. G. beamanii. a, interior surface of corolla, slit oe and pressed; ae exterior surface of calyx; c, pistil; d, intact open flower from above; e, closed flow 0-13. Gentiana spp. Fig. G. fieldiana. Fig. 11. G. pumilio. Fig. 12. “perp Fig. 13. G. prostra ” dee awn from a specimen from Colorado, where species normally bears tetramerous flowers). a, interior surface of corolla, slit ia and pressed; b, exterior surface of calyx; c, pistil. ay portions of appendages, and uppermost portion of tube white to medium blue; exterior surface of lobes, except near margins, suffused with green, and occasionally also with deep violet-blue except near inner margin. Stamens becoming free at ca. 1/2 total length of corolla; free portions (including anthers) ca. 2 mm long. Pollen grains prolate. Sexine relatively coarsely striato-reticulate. Pistil tapering into style ca. 1 mm long. Capsule 3-5 mm long, 2/3 to entire length and up to 2 mm of gynophore exserted from marcescent corolla at maturity. Seeds ellipsoid, light brown, striato-reticulate, ca. 0.7 mm long, 0.35 mm in diameter. Gentiana fieldiana is native to the Sierra Madre and Los Altos Cuchu- matanes in Guatemala (Fig. 22) where it grows in moist sites in subalpine meadows (llanos) and openings in pine forests, from 2500 to 3500 m. Plants have been collected in flower from December through July. Its name honors the contributions of botanists now and formerly of the Field Museum of Natural History to the knowledge of the flora of Guatemala, especially Dr. Julian A. Steyermark and the late Mr. Paul C. Standley, who first collected this species. A conspicuous feature of all but the smallest plants of G. freldiana is the presence of 1-several short, leafy branches near the summit of the stem, with the flowers thus being borne in small cymes. In this trait, G. freldiana differs from G. sedifolia, G. beamanii, and G. prostrata, in all of which solitary flowers generally terminate longer branches or unbranched stems (one such branch occasionally present in G. sedifolia). Also, G. fieldiana differs from the other species discussed here in that its corolla lobes taper abruptly to an apiculate tip, rather than gradually from near the base. Its pollen is similar to that of G. sedifolia, but the sexine pattern is more dis- tinctly striato-reticulate, rather than simply reticulate, and the lirae are finer. Photographs of living plants of G. fieldiana were not available for this study, but Breedlove 8637 and Sharp 451025 contain some corollas pressed while open, with the colors well retained. From these specimens it appears that no guidelines or sharply contrasting corolla-throat patterns are present in this species. Some specimens of G. fieldiana were previously identified as G. pumulio. Both of these species, as well as G. beamanii, have been collected near Xémal, but even in this area they maintain their distinctness. Gentiana fieldiana is readily distinguishable from G. pumilio by its narrower, less densely crowded leaves with narrower, less prominently denticulate margins, and by the shape of its corolla lobes. Breedlove 8596 consists of usually large plants with relatively narrow leaves for this species. The general aspect of these plants consequently re- sembles that of G. sedifolia, but their well-developed leaf margins and keels and the position of their stamens clearly identify them as G. fieldiana. 4. GENTIANA PUMILIO Standl. & Steyerm., Publ. Field Mus. Nat. Hist., Bot. Ser. 23:76. 1944. Type: GUATEMALA: San Marcos: between San Sebas- 28 tian and summit of Volcén de Tajumulco, rel mark 35489, Feb 1940 (holotype F!, isotypes MSC!). Figs. 5, 11, AL csasstinissestnsssssssisnsninasied 0.0] mm Figs. tee Scanning-electron eee a a of pollen grains of Gentiana spp. Fig. an y. sedifolia (Mulroy 1099, HAM). ae a — beam eg (Beaman 3899, MSC) 16. aA sags saa (Beamon 3071, 17 ee aad erage 2868, MSE). ae perpusilla ani ie cs “MICH Fig. prostrata (Sharp s ae from Alaska). Figs. 14-19 to same ree 22 Plants 0.7-4 cm tall. Stems usually 1-10, sometimes much branched form- ing dense tufts. Basal and lower cauline leaves persistent and green or the lowest pairs marcescent. Leaves ascending, densely imbricate-crowded on whole length of stem, broadly spatulate to orbicular, either tapering grad- ually to base or at least upper leaves more commonly abruptly contracted below middle. Larger, lower leaves 2.5-15 mm long, 0.3-1.2 times as wide as long; upper leaves smaller and more strongly ascending but similar in proportions. Apices abruptly acuminate-apiculate. All leaves prominently white-margined and keeled but not strongly conduplicate; margins and keels of upper leaves mostly 0.25-0.35 mm wide, minutely but copiously denticu- late at least above middle. Flowers solitary or in clusters of 2-3. Calyx tube largely concealed by leaves, 3-5.5 mm long, carinate distally. Calyx lobes ovate-triangular, erect, strongly carinate-conduplicate, 1-2. mm long, 0.5- 1.5 mm wide, obtuse, with prominent white margins and keels ca. 0.2 mm wide, denticulate to base. Corollas 5-11 mm long when closed; expanded limb 4-7 mm across. Corolla tube 4-9 mm long. Lobes ovate-triangular, 1-2 mm long, ca. 1.2 times as wide as long, obtuse. Free portions of appen- dages triangular, ca. 0.6 times as long and 0.8 times as wide as lobes, acute, entire or shallowly undulate. Corolla tube mostly greenish-white; lobes, free portions of appendages, and upper part of tube white to light blue; exterior surface of petals proper moderately to strongly suffused with green, and in blue corollas sometimes with deep violet, the suffusion strongest toward center. Stamens becoming free at ca. 3/5 total height of corolla; free por- tions (including anthers) ca. 2.2 mm long. Pollen grains spheroidal. Sexine relatively fine striato-reticulate. Style ca. 3 mm long, spiralling in fruit. Capsule 5-7 mm Jong, 1/2 to almost completely exserted from marcescent corolla. Seeds cylindric-ellipsoid, medium brown, striato-reticulate, ca. 0.9 mm long and 0.3 mm in diameter. Additional specimens examined: GUATEMALA: Huehuetenango (all from Los Altos Cuchumantanes): region of Chémal [Xémal]}, Standley 81107 (F); cumbre de la Sierra de los Cuchumatanes, between the first cumbre and La Pradera, Standley 81172 (F); near Calaveras, Williams et al. 21954 (F) and 21955 (F, NY, US); 3-15 km N of Chémal [Xémal], Williams et al. 22208 (F).San Marcos: between Sibinal and summit of Volcan Tacana, Steyermark 36156 (F); Volcan Tajumulco, Bunting 337 (F); between Comitancito {[Comitancillo} and Santa Rosa junction, Sierra Madre, Williams et al. 27058 (F); lower slopes of Volcan Tajumulco near San Paaeieee Plowman 3064 (GH). Solola: near Maria Tectn, Sierra Madre, about 10— cm NW of Los Encuentros, Williams et al, 27325 (F). Totonicapan : region of Esra Standley 62735 (F); region of Salvachan, mountains above Totonicapan, just before reaching Desconsuelo, Standley 84489 (F); slopes of Maria Tecin above Totonicapan, Williams & Williams 18557 (GH, US); summit of Cerro Maria Tectin, about 12 km SW of Totonicapan, Williams et al. 25457 (F); Maria Tecin, Molina R. et al. 16376 (F, NY, US). MEXICO: Chiapas: Volcan de Tacana pico, Matuda 2868 (F, GH, HAM, K, MEXU, MICH, MSC, NY, US). Gentiana pumilio is native to the Sierra Madre of southwestern Guatemala T q ICARAGUA GENTLANA SEDIF OLIA PANAMA ch - , ’Matanes a GENTIANA FIELDLANA SALVADOR 9 eee ) L | A GUATEMALA GENTIANA PUMIL1O Figs. 20-24. Documented distribution of Gentiana spp. at and adjacent Chiapas, Mexico, and to Los Altos Cuchumantanes in Guatemala. All collections of this species are from open habitats, mostly in subalpine or alpine zones, from ca. 2500 to 4062 m. Most collections are from moist sites, but some are from dry slopes and pastures. Plants have been collected in flower from November through May. Because of its proportionately wide leaves and the wide and conspicuous white margins and keels of its leaves and calyx lobes, G. pumilio is readily distinguishable from the other species described here. Its relatively long style and spheroidal pollen grains are also distinctive. Gentiana pumilio further differs from the three preceding species in that the flowers are often borne in clusters of two or three, rather than singly at the ends of branches. 5. GENTIANA PERPUSILLA Brandegee, Zoe 5:181. 1904. Type: MEXICO: México: Ixtaccthuatl, Purpws 302, Mar-Jul 1903 (holotype US!, isotypes F!, GH!, MO!, MSC!, UC). Figs. 6, 12, Plants 0.8—3 cm tall. Stems 1-10, ascending, not branched except occa- sionally near base. Leaves short-acuminate, apiculate. Basal leaves mostly in good condition at flowering time, obovate to broadly spatulate, moderately ascending, scarcely to moderately conduplicate, 4-10 mm long, 1.8-3.5 mm wide. Cauline leaves densely crowded, increasingly conduplicate and more strongly ascending upward, those near summit of stem oblanceolate, 3-6 mm long, 1-2.5 mm wide. White leaf margins and midrib keel inconspicuous, ca. 0.05 mm wide, entire or with a few shallow projections near apex. Calyx tube 4-6.5 mm long, carinate distally. Calyx lobes triangular, erect, 1—1.2 mm long, 0.7-1 mm wide, acute, with white margins and keels ca. 0.08 mm wide, entire. Corollas 5-9.5 mm long when closed; expanded limb 3.5— 4.5 mm across. Corolla tube 4-8.5 mm long. Lobes ovate-triangular, 1-2 mm long, about as wide as long, subacute. Free portions of appendages triangular, ca. 0.67 times as long and about as wide as lobes, subacute to acute, entire, merely undulate, or shallowly few-toothed. Corolla tube mostly greenish- white; lobes and free portions of appendages pale to medium blue; exterior surface of lobes, except near margins, and adjacent portions of tube strongly suffused with green. Stamens becoming free at ca. 2/3 total height of corolla; free portions (including anthers) 1.5—-2 mm long. Pollen grains prolate. Sexine relatively finely striato-reticulate. Style almost obsolete. Capsule 2.5— 3.5 mm long, 1/2 to entire length and up to 3 mm of gynophore exserted from marcescent corolla. Seeds ovoid-ellipsoid, brownish-white, striato- reticulate, ca. 0.75 mm long, 0.5 mm in diameter. Additional specimens examined: MEXICO: México (all from the SW side of Ixtaccihuatl): Canada de Alcalican (Tlaltapitongo) below La Joya, Beaman 3534 (MSC); La Joya, Canada de Alcalican, Rzedowski 21594 (ENCE) and 25681 (ENCB, na MSC); in canyon below end of auto road, ca. 2.4 km N of microwave tower, J.S. Pringle 1749 (HAM). Veracruz: Cofre de han NW side of mtn., Bean 5197 (GH, MSC). es) i) | a ae ig =, HIDALGO } = | op — ; LS eee ; od ee ( fr» tiene) oe Me # \ A A, ‘a \ ( m TLAXCALA be. ee 2 Ae Drak, * acinar é 1 La 7 / e VERACRUZ |. f ~~ MORELOS < J ne P PUEBLA i we ae Is) Leo ? _ J — 25 - 7) a 4 _ : oe ee GUERRERO f \ eo OAXACA ‘ - ph 9 | Fig. 25. Documented distribution of Gentiana perpusilla. The two localities known for G. perpusilla are on peaks in the Transverse Volcanic Zone of Mexico (Fig. 25). It grows in the alpine zone at ca. 3850 m on Ixtaccihuatl and 3590 on Nauhcampatépetl (Cofre de Perote), in microhabitats that are wet at least in summer. Among the species described here, G. perpusilla has by far the smallest corollas, and its pollination appears probably to be largely hygrocleistogamous. It grows where fog and drizzle are nearly continuous during its flowering season, and consequently its photonastic corollas are infrequently and briefly open. Observation of pollen in preserved flowers from Ixtaccihuatl indicated that dehiscence of the anthers into the small space within a closed corolla can readily bring pollen into contact with the stigmatic lobes. In one corolla, however, the presence of a few alien grains indicated that some chasmogam- ous pollination does occur. ACKNOWLEDGEMENTS I express my sincere thanks to: Dr. John H. Beaman, for his comments and notes on the species described here, and for his review of the manu- script; Dr. Juliana C. Mulroy, for suggestions for the improvement of an earlier version, and for collections of G. sedifolia and related species; Mrs. Margaret Williams, for additional collections; Dr. Thomas Duncan, for his 2) cooperation and assistance with field work; Messrs. James Darley and ouglas Brown, for SEM photographs of pollen grains, using facilities at McMaster University; the curators and staff of the herbaria listed in the introduction of this paper, for making specimens available for study; the Consejo Nacional de Ciencia y Tecnologia, for permission to collect speci- mens of Gentiana in Mexico; and the Board of the Royal Botanical Gardens, for its support of the publication of this paper. REFERENCES GILG, E. 1906. Beitrage zur Kenntnis der Gentianaceae III. Gentianaceae andinae. Repert. Spec. Nov. Regni Veg. 2:33-56. HOLMGREN, P.K., & W. KEUKEN ae Index Herbariorum. Part I: The herbaria of the world, ed. 6. Regnum Veg. 92. pp. HUMBOLDT, [F.H.} A. VON, A. [J.] BONPLA ND, & CS. KUNTH. 1815-“1823” {-1825]. Voyage aux Régiones Equinoctiales du Nouveau Continent, Fait en 1799- 1804. Sixiéme Partie. Bota anique. Sect. 3. Nova Genera et Species Plantarum Libraria Graeco-Latino-Germanica, cee 7 vols. (Gentiana in vol. 3(10). 1819; supplementary note in vol, 5(23). 3, KERNER VON MARILAUN, ae 887-1891. me Bibliographisches Institut, Leipzig. 2 vols. (Gentiana prostrata in vol. 2 91.) ae translation by F.W. OLIVER ecu Henry Holt and canes New KUSNEZOW, N.J. 1895. Gentiana Tournef. Iz: ENGLER, [H.G.] i “& [A.E.] P L. Die aes Pflanzenfamilien . . . Verlag von Wilhelm foi Leipzig. 4(2) :8 LINDSEY, me 1940 a anatomy in the Gentianaceae. Amer. J. Bot. 27 :640-652. LOVE OVE. 1975. The Spanish gentians. Anales Inst. Bot. Cavanilles 32:22] "932. NILSSON, S. 1967. Pollen morphological studies in the Gentianaceae-Gentianinae. 145. PRINGLE, J.S. 1977. Taxonomy of Gentiana aguas in Mexico and Central America. I. Sect. Bee une. Sida 7:174-217. __. 1978. Sectional and subgeneric names in Gentiana (Gentianaceae). Sida 47. TOYOKUNI. H. 1963. Conspectus Sg eee Re re eee a general view of the Gentianaceae aa to Japan. J. Fac. . Hokkaido Imp. Univ., Ser. 5, Bot. 7:137-—259, pl. LIV. EAVER, R. - Tie ary RES. 1975. Cytotaxonomic notes on some Gentianaceae. a Arnold Arbor. 56:211 REVISION OF NORTH AND CENTRAL AMERICAN NAJAS (NAJADACEAE) ROBERT R. HAYNES Ecology and Systematics Section, Department of Biology, The University of Alabama, University, Alabama 35486 Najas is a cosmopolitan genus of submerged aquatic plants with its greatest diversity in tropical and subtropical regions. In North America, individuals of the genus can be separated from other aquatic genera by the plants of Najas possessing sub-opposite serrulate leaves and axillary see flowers. Although floristic treatments have been prepared for certain eastern United States (Clausen 1936) and Panama (Wentz and ie 1973 )—at no time since Rendle (1901) have all the species, either world- wide or regional, been included in one revisionary treatment. Morphological variability within the group has been poorly understood. Because of the ack of a thorough revisionary treatment and a poor understanding of the taxa, the genus is, I believe, in need of revision. The genus Najas has been divided into two subgenera, Najas and Caulinia, by Ascherson (1864) with Najas being dioecious and Caulinia_ being monoecious. Magnus (1889) separated subgenus Caslinia into two sections, Americanae with sloping leaf sheaths and Ewvvaginatae with truncate to auriculate leaf sheaths. Rendle (1899) divided the subgenus into two more sections, Spathaceae with staminate and carpellate flowers enclosed in floral envelopes and Nwdae with no flowers enclosed in floral envelopes. According to Rendle (1899), the sections Americanae and Euvaginatae were composed of plants with staminate flowers enclosed in floral envelopes and carpellate flowers without such envelopes. Wilde (1961), in an investigation of Asiatic and Malaysian species of Najas, determined the presence of the envelopes to be too inconsistent to segregate sections. He thus proposed the abolition of Rendle’s sections. Variation in vegetative parts (plasticity) within individual taxa of aquatic plants has been recognized in Callitriche (Fassett 1951), Halodule (Phillips 1967), Nymphaea (Williams 1970), Najas minor (Wentz & Stuckey 1971), and Potamogeton (Haynes 1974). These workers concluded that because of this plasticity, the taxonomy of the genera in question should rest_pri- marily on reproductive characters. As a result of my field work and the examination of several thousand herbarium specimens, I have found consider- able plasticity within the leaf sheaths of Najas. 1 have concluded, therefore, that the separation of sections based upon the shape of the auricles is super- "Paper No. 7 from the Aquatic Biology Program, The University of Alabama. SIDA 8(1): 34-56. 1979. —_~ 3 ficial. I agree with Wilde that four sections are unwarranted. Wilde (1961) also suggested that the two subgenera be reduced to sections because of the “great resemblance of N. marina to other species of Najas ....” I agree that N. marina resembles the other species. However, with N. marina being dioecious, having a testa several cell layers thick, and having prickly inter- have a testa three cell layers thick, and do not have prickly internodes and dorsal surfaces of the laminae—I believe the subgeneric category to be warranted. The nomenclature of Najas is difficult. Many of the taxa in North and Central America were named by Sprengel (1825), Braun (1864), and Magnus (1870, 1894). The specimens examined by these workers were de- posited at Berlin and later were destroyed during World War I. Gerloff (personal communication) informs me that no pre-World War II specimens of Najas (including the Willdenow collection) remain at Berlin. Therefore, I have had to resort to published descriptions and accepted usage of many names. In some instances (see Haynes and Wentz 1974) when only one or two taxa of Najas occur in an area, there is good evidence for my decision. In others, however, I have only the accepted usage and description to follow. The morphology, systematic relationships, and economic value of Najas were discussed by Haynes (1977). Evidence from field work has indicated that Najas is just as morphologically plastic as Callitriche (Fassett 1951), Potamogeton (Haynes 1974), and Ranunculus (Bostrack and Millington 1962). Just as workers in these other genera have resorted to reproductive characters as a basis of a classification, I believe that a classification based mostly on reproductive structures is the only one workable for Najas. One should always attempt to collect specimens of Najas with seeds (Fig. 8). Just as identification of a sterile Aster may be nearly impossible, it is diffi- cult at best to determine a sterile Najas. The treatment that follows is based on field study and an examination of approximately 5000 herbarium specimens from 35 North American and European herbaria: A, ALU, AUA, B, BH, CAN, DAO, F, FSU, GA, GH, JEPS, K, KANU, KNK, LAF, LSUS, MEXU, MICH, MIN, MO, NCSC, NY, OS, PH, TENN, UC, UNA, UNC, UMBS, US, USF, UTC, VDB, and WIS (abbreviations according to Holmgren and Keuken 1974). Keys are based upon materials containing seeds; dimensions of leaves are taken from the fully expanded; longest leaves of a specimen; measurements of widths are taken approximately at the widest point of the leaf; and descriptions of the seeds are taken strictly from mature structures (Fig. 8). NAJAS Linnaeus, Sp. Pl. 2: 1015. 1753; Gen. PI. a 5. 449547 54. Caroliana Raf., J. Phys. Chim. Hist. Nat. Arts 89:259. 1819. Caulinia Willd., Mem. Acad. Roy. Sci. Hist. (Berlin) 1798. Fluvialis Adans., Fam. Pl. 2:472. 1763 Hyas Dumort., Anal. Fam. Pl. 61. 1829. Ittnera C. C. Gmel., Fl. Bad. 3:590. 1808. 36 Plants glabrous, herbaceous, annual or rarely perennial, aquatic, submersed in fresh or brackish waters, monoecious or dioecious. Stems slender, much branched, rooting at the lower nodes, sometimes armed with prickles on the internodes. Leaves subopposite or appearing whorled due to reduced inter- and flattened, l—-nerved, sometimes dorsally armed with prickles on the midrib; margins usually serrulate with 5-100 teeth per side; apex acute to acuminate, with 1-3 teeth per side; teeth multicellular, formed by layers of cells decreasing in cell number outward terminated by a large, sharp-tipped cell, or unicellular. Sheaths variously shaped, each enclosing a pair of tiny hyaline scales; margins usually toothed with 1-15 teeth per side. Flowers pee axillary, sessile or short pedunculate, solitary or clustered, often subtended by an involucre; involucre clear, bronze, brown, light green, purple, or red-purple. Staminate flowers subtended by a membranous involucre or the involucre rarely absent, each consisting of a single stamen; peduncle at first short, elongating at anthesis, pushing the anther through the involucre; anther sessile, 1- or 4-loculed, dehiscing irregularly; pollen 3-celled, glob- ular or ellipsoid, densely filled with starch, monocol pate, the wall with shallow reticulations, thin, not divided into exine and entine. Carpellate flowers sessile, 1-loculed, 1-ovuled; involucre absent or rarely present; ovule basal, anatropous, with 2 integuments; gynoecial wall 2 cell layers thick, ending in a short style with 2—4 branches. Fruit 1-seeded, dehiscing by decay of gynoecial wall; gynoecial wall extremely delicate, closely enveloping the seed. Seeds without endosperm, areolate, with a basal raphe, fusiform to obovate, occasionally asymmetrical at apex or recurved; testa hard, brittle, 3 or several cell-layered, pitted or smooth; areolae formed by outer two layers of testa, variously eae shaped, irregularly ope or in 15-60 rows, the end walls often raised, giving the testa a papillose appearance; embryo elongate, without lateral enlargement of cotyledons. TYPE SPECIES: N. marina L. (From the Greek, naias, a water-nymph.)—Bushy-pondweed, Naiad, Water-Nymph. KEY TO THE TAXA 1. Plants dioecious; testa pitted, 4 or more ae oS thick; internodes and Rais WN; surfaces of lamina with prickles; seeds 1.2 m e or wider 0.0.2... 1. Plants monoecious; testa smooth or pitted, 3 cal ne thick; internodes and aa surface of lamina without ae seeds 1.2 mm or less wide, #f 1.2 mm wide, then with smooth testa 2 . Seeds asymmetrical at apex or recurved 3. Areolae broader than long, ladder-like; teeth multicellular ............ 7. N. minor 3. Areolae longer than broad, never ladder-like; teeth unicellular ..... 7 . Seeds strongly recurved; areolae in 10-20 rows; leaves in age stiff, recury ee 4. vis asymmetrical at apex but not ae areolae in : 35 ‘5 ws; leaves n age limp and spreading to ascendin a gracillima 2 Sale symmetrical at apex, not recurved oF 5. Lamina teeth unicellular, (18—) 30-100 per side 6. Testa smooth, glossy; seeds obovate, brown to yellow; anthers 1—loculed 2.N eee . flexilis 6. Testa pitted, dull; seeds fusiform, yellowish-white to greenish-brown; anthers 1— or 4-loculed 7. Leaf sheaths deeply auriculate; restricted in North America to California 4, N 7. Leaf sheaths rounded to truncate; widespread graminea 8 8. Seeds 3.3-3.8 mm long; areolae arranged in N DOGO OWS 2:4 anid aeeiueies dy, duets attests d. N. guadalupensis var. muenschert 8. Seeds 2.5 mm or less ong: areolae in fewer than 9. Teeth evident to unaided eye; anthers 1—-locu 5) c. N. evadalapents var. floridana 9. Pe invisible to unaided eye; anthers 4—locu aminae with 50-100 teeth per side; stem r less in di- meter 3a. N. gu anes var. guadalupensts 10. (minde with less than 50 teeth per side; stem mm or more 3b. N. epee. var. olivacea in diameter 5. Lamina teeth multicellular, 5—25 per side 11. Seeds 2 m m long or longer; areolae arranged in ca. 50 rows, elongate; teeth shorter than rest of the lamina is wide . _N. arguta . Seeds less than 2 mm long; areolae arranged in ca. 20 rows, Ssided, all sides about equal; teeth longer than rest of lamina is wide e —_ 8. N. wrightiana NAJAS subgenus NAJAS Plants dioecious. Stem slender, much branched, rooting at the lower nodes, usually armed with prickles on the internodes. Leaves subopposite or appear- ing whorled, sessile; laminae coarsely serrate, the teeth multicellular, a midrib with prickles. Flowers solitary; anthers 4—loculed. Seeds 2.2-4.5 m long; testa several (more than 3) cell layers thick; areolae arranged es larly. Species 1. TYPE SPECIES: N. marina L. 1. NAJAS MARINA L., PL 281019... 7 9: Ittnera major C. C. = , Fl. Bad. 3: a 1808. I. najas C. C. Gmel., Bad. 3:590. 1808. Najas gracilis vee Mea. Torrey Bot Club ae :61. 1893. . 40. 1903 _ in Martius, Fl. Brasiliensis 3 (3) :804. 1894. marina Vat. as Gs. Br.) K. Schum. in Martius, Fl. Brasiliensis 3G) :725. 1894. N. marina var. angustissima K. Schum. in Martius, Fl. Brasiliensis 3(3):725. 1894. N. mariana var. bollei K. Schum. in Martius, Fl. Brasiliensis 3(3) :726. 1894. 7 a var. ree Rendle, Trans. Linn. Soc. London, Bot. 5:396. 1899. N. marina var. californica Rendle, Trans. Linn. Soc. onan Bot. 5:398. na vat. denticulata Rendle, Trans. Linn. Soc. London, Bot. 5:438. marina var. ebrenbergii (A. Br.) K. Schum. in Martius, Fl. eae 3(3) a ees N. marin . genuina K. Schum. in Martius, Fl. Brasiliensis 3(3) :724. 1894. N. marina var. ae ie Dudley, Cornell Univ. Sci. Bull. 2:104. 1886. 38 N. marina var. latifolia (A. Br.) K. Schum. in Martius, Fl. Brasiliensis 3 (3) :725. N marin var. latior F. Muell. ex K. Schum. in Martius, Fl. Brasiliensis 3 (3) :725. ere var. mexicana Rendle, Trans. Linn. Soc. London, Bot. 5:398. 1899. marina a multidentata (A. Br.) K. Schum. in Martius, Fl. Brasiliensis 1894. 3(3) :72 N. marina vat. muricata (Rafteneau-Delile) K. Schum. in Martius, Fl. Brasiliensis 3 . 1894. S, N. pee var. Pig ii (A. Br.) K. Schum. in Martius, Fl. Brasiliensis marina var. recurvata Dudley, So Univ. Sci. Bull. 2:104. 1886. major All., Fl. Pedem. 2:221 major var. angustifolia A. Br. | aa 2:275. 1864. major var. ebrenbergit A. Br., J. Bot. 2:275. 1864. major var. gracilis Morong, Boe Gaz. (Crawfordsville) 10:225. 1885. major var. intermedia A. Br., J. Bot. 2:276. 1864. major var. microcarpa A. Br., J. Bot. 2:276. eee m ilies var. multidentata A. Br. J. Bot. 2:275. 1864. vajor var. paucidentata A. Br., J. Bot. 2:276. 1864. ape ae ee Delile, ee Egypte, Hist. Nat. 281. 1813. ee . Plants dioecious. Stems 6-45 cm long, 0.5-4.0 mm in diameter, branched upward; internodes 0.3-11.0 cm long, with or without prickles. Leaves 0.5-3.9 cm long; laminae spreading to ascending, 0.4-4.5 mm wide, acute with 1 tooth at apex, coarsely serrate with 8-13 teeth per side, the terminal teeth similar in size and structure to the lateral teeth, the teeth multicellular, the midrib with prickles; sheaths 2.0-4.4 mm wide, wider than the laminae, acute, the teeth similar in size and structure to those of the laminae. Flowers solitary, the involucre brown to slightly pu oe soar flowers 1.7—3.9 mm long, the involucre beaks 0.3-0.7 mm long, 2—lobed, the anther 1.7—3.0 mm long, 4—-loculed; carpellate flowers 2.5-5.7 mm iene. the style 1.2-1.7 mm long, the stigma 3-lobed. Seeds 2.24.5 mm long, 1.2—2.2 mm wide, reddish brown, ovoid, the testa pitted, the areolae 3—4-angled, ca. 0.2 mm long, 0.1 mm broad, irregularly arranged, the end walls slightly raised (Fig. 15s 8A). Chromosome number, 27 = 12 (Viinikka : Brackish or highly mere waters of ponds and lakes, North and South Dakota to western New ; California to Utah, south to Panama; Florida and Caribbean islands (Fig. TYPE: Europe maribus (Hol orype: LINN; Microfiche #1156 IDC 1945, 682:II 6, 7 With its prickly internodes and prickles along the undersurface of the leaves, N. marina is the easiest of our Najas to recognize. The species is variable in the size of leaves, teeth, and seeds and, as a result, has been divided into numerous varieties. These differences usually can be found within single populations and, therefore, do not warrant taxonomic recog- nition. Viinikka (1973), however, observed two cytological races of the species, one race with B chromosomes and one without these accessory chro- Fig. of North America showing the documented distribution of Najas marina. Week see inserted. ) 40 mosomes. He examined one population from North America (Erie County, Ohio) and observed B chromosomes from those individuals (personal cor- respondence). He indicated (personal correspondence) that, in Europe, plants from the two races are morphologically different. Specimens with large seeds and leaves (no size given) invariably do not possess B chromosomes, whereas those with smaller seed and leaves do possess B chromosomes. He has not yet determined how to rank these apparent entities. NAJAS subgenus CAULINIA ( Willd.) Aschers., Fl. Prov. Brandenburg 1:670. 1864 Plant monoecious. Stem slender, much branched, rooting at the lower nodes, the internodes without prickles. Leaves subopposite or appearing whorled, sessile; laminae minutely serrulate to serrate, the teeth unicellular or area the midrib without prickles. Flowers solitary or clustered; anthers 1—- or 4—loculed. Seeds 0.7-3.8 mm long; testa 3 cell layers thick; nee arranged in longitudinal rows. Species 2-9. LECTOTYPE SPECIES: Caulinia flexilis Willd. = N. flexilis (Willd.) Rostk. & Schmidt. (Lectotype here designated. ) 2. NAJAS FLEXILIS (Willd.) Rostk. & Schmidt, Fl. Sedin. 382. 1824, Caulinia flexilis Willd., Mem. Acad. Roy. Sci. Hist. (Berlin) 95. 1798. Fluvialis flexilis (Willd.) Pers., Synops. 2:530. 180 Najas canadensis Michx., Fl. Bor. Amer. 2:220. 1803. N. caespitosus (Maguire) Reveal in Welsh, Atwood, & Reveal, uralist 35:357. 1976. N. flextlis subsp. caespitosa Maguire & Jensen, Rhodora 44:7. 1942 N. flexilis var. congesta Farwell, Rep. Michigan Acad. Sci. 21:348. 1920. N. flexilis var. robusta Morong, Bot. Gaz. (Crawford suille) 10:255. 1885, Great Basin Nat- Plants monoecious. Stems 2.5-50 cm long, 0.2-0.6 mm in diameter, often profusely branched above; internodes 0.16-6.8 cm long, without prickles. Leaves 0.8-4.2 cm long; laminae spreading to ascending, 0.2-0.6 mm wide, acute with 1-2 teeth at apex, minutely serrulate with 35-80 teeth per side, the terminal teeth similar in size and structure to the lateral teeth, the teeth unicellular, the midrib without prickles; sheaths 0.7-1.6 mm wide, wider than the laminae, round, minutely serrulate with 8-10 teeth per side, the teeth similar in size and structure to those of the laminae. Flowers 1 (—2) per axil, the staminate in the upper axils, the carpellate throughout, the involucre bronze; staminate flowers 1.1-2.7 mm long, the involucre beaks 0.7-1.2 mm long, 3-lobed, the anthers 1.1-2.7 mm long, 1-loculed; carpellate flowers 2.5-4.7 mm long, the style 1.5-1.7 mm long, the stigma 3-lobed. Seeds (1.2—) 2.5-3.7 mm long, 0.2-1.2 mm wide, ee brown to yellow, narrowly to broadly obovate, the testa smooth, the areolae 5—6-angled, ca. 0.2 mm long, ca. 0.1 mm wide, in ca. 50 rows, without raised end walls (Fig. 8B). Chromosome number unknown. Lakes and rivers, Nebraska, Missouri, and Maryland, north to southwestern 41 Ontario and Newfoundland; Alberta and Saskatchewan south to western Oregon and central Utah (Fig. 2). TYPE: Pennsylvania (Holotype: B; microfiche # 17094 IDC 7440. 1224: III. 2.) In habit, N. flexilis is most similar to N. guadalupensis. However, when seeds are present, N. flexilis can be easily separated from the latter species by the glossy, smooth, yellowish seeds that are widest above the middle. Several different varieties have been proposed for North American indi- viduals of N. flexilis, e.g. var. robusta Morong, var. congesta Farwell, and var. caespitosa Maguire and Jensen. These varieties are based, for the most part, on differences in vegetative structures. Clausen (1936), in discussing some of these different forms, stated “Different forms of the species do occur, but the characters, such as width of leaves, size and shape of seeds, and habit, occur in all sorts of combinations and can not be correlated with geographical areas to give definite geographical races or varieties which are worth naming.” I have reached the same conclusion. Najas flexilis is probably the most common Najas in the northern United States. Wentz and Stuckey (1971), however, indicated that the species is becoming less common in Ohio. This decrease in abundance apparently cor- responds with the increase in turbidity and general decline in water quality of natural lakes and rivers. 3. NAJAS GUADALUPENSIS (Sprengel) Magnus, Beitr. Gatt. Najas 8. 1870. Plants monoecious. Stems 11.0—90.0 cm long, 0.1-2.0 mm in diameter, profusely branched; internodes 0.1-9.0 cm long, without prickles. Leaves 0.3-3.3 cm long; laminae spreading, 0.2-2.1 mm wide, round obtuse to acuminate with 1-3 teeth at apex, minutely serrulate with 18-100 teeth per side, the terminal tooth slightly larger than but similar in structure to the lateral teeth, the teeth unicellular, the midrib without prickles; sheaths 1.0- 3.4 mm wide, wider than the laminae, round to slightly auriculate, serrulate with 4-8 teeth per side, the teeth similar in size and structure to those of the laminae. Flowers 1—3 per axil, the staminate in the upper axils, the carpellate in the lower axils, the involucre purple-tinged; staminate flowers 1.5-3.0 mm long, the involucre beaks 0.2-1.3 mm long, 4—lobed, the anther 1.0-1.7 mm long, 1- or 4—loculed; carpellate flowers 1.5-4.0 mm long, the styles 0.3-1.5 mm long, the stigmas 4—lobed. Seeds 1.2-3.8 mm long, 0.4- 0.8 mm wide, purple-tinged, fusiform, the testa pitted, the areolae 4—G-— angled, 0.08-0.1 mm long, ca. 0.08 mm wide, in 20-60 rows, the end walls not raised. Chromosome number, 27 = 12, 36, 42, 48, 54, 60 (Chase 1947 a,b). Najas guadalupensis has long been considered a highly variable species (Fernald 1923). Clausen (1936), Ooststroom (1939), Wentz and Haynes (1973), and Haynes and Wentz (1974) discussed variability within the species and suggested that the complex should possibly be considered as LAMBERT AZMUTHAL CQUAL-AREA PROJECTION Fig. 2. Map of North America showing the documented distribution of Najas flexilis. 43 several closely related taxa. Rosendahl and Butters (1935) named N. ol/vacea. It was said to differ from N. guadalupensis in its stouter habit, its larger seeds, and its testa. Clausen (1937) separated N. mwenscheri from ecial stay by its more slender habit and its seeds with smaller areolae. y field work and the examination of thousands of specimens of N. ee ee and most of these available of N. mwenscheri and N. olivacea, 1 have concluded that N. olivacea and N. muenscheri represent populations of N. gwadalupensis at the northern limit of its range. They are slightly different from most individuals of N. guadalupensis. However, inter- mediates do exist. Thus, I think they should be recognized but do not believe that specific rank is justified. I consider varieties as morphogeographic sub- divisions of a species that presumably reflect genetic differences (Kapadia 1963). I have therefore chosen to recognize the taxa that have been called N. olivacea and N. muenscheri as varieties of N. guadalupensts. By having unicellular teeth on the margin of the lamina, N. gauadalupensis resembles N. flexilis and N. graminea. The seeds of N. guadalupensis are pitted; those of N. flexilis are smooth. The sheaths of N. gwadalupensis are rounded to slightly auriculate; those of N. graminea are deeply auriculate. 3a. NAJAS GUADALUPENSIS var. GUADALUPENSIS Caulinia guadalupensis Sprengel, Syst. Veep 1220: 1824, Najas flexilis var. curassavica A. Br., J. Bot. 2:277. 1864. N. flextlis var. fusiformis Chapman, FIL. Seuthern U. S. 444. 1883. N. flexilis var. guadalupensis (Sprengel) A. Br., J. Bot. 2:276. 1864. N. microdon vat. curassavica A. Br., Sitzb. Ges. Naturf. Fr. Berlin. 1868. N. microdon var. guadalupensis (Sprengel) A. Br., J. Bot. 2:276. 1864. N. urbaniana O. C. Schmidt, Fedde Rep. Spec. Nov. 22:99. 1925. Stems 11-75 cm long, 0.1-0.8 mm in diameter. Leaves 0.3-2.8 mm long; laminae 0.2-1.8 mm wide, acute to mucronate, with 50-100 teeth per side; sheaths 1.0-1.9 mm wide, round, with 5-8 teeth per side. Flowers 1-3 per axil; staminate 1.5-2.5 mm long, the involucre beak 0.2-0.6 mm long, the anther elliptic, ce carpellate flowers 1.5-2.0 mm long. Seeds 1.2—2.5 mm long, 0.4-0.6 mm wide, the areolae in ca. 20 rows (Fig. 8 Lakes, rivers, a canals, southern Maine to southern Albena: and Wash- ington, south to Guadaloupe, Curacao, and Panama (Fig. 3). TYPE: Guadeloupe (B, destroyed during World War IT). 3b. NAJAS GUADALUPENSIS var. Olivacea (Rosendahl & Butters) Haynes, comb. & stat. nov. Najas olivacea Rosendahl & Butters, Rhodora 37:347. 1935. Stems 15-40 cm long, 1.0—2.0 mm in diameter. Leaves 0.9-1.8 cm long; laminae 1.5-2.0 mm wide, acute, with 20-40 teeth per side; sheaths 2.5-3.4 mm wide, round to slightly auriculate, with 4-8 teeth per side. Flowers 1 per axil; staminate flowers 2.3-2.8 mm long, the anther ovoid, 44loculed; carpellate flowers 2.7-3.1 mm long. Seeds 2.3-2.5 mm long, 0.6-0.8 mm Fig. 3. Map of North America showing the documented distribution of Najas guadalupensis var. guadalupensis. (Caribbean islands inserted.) 45 wide, the areolae in 20-40 rows (Fig. 8D). Lakes and rivers, lowa to Manitoba, east to New York and Quebec (Fig. TYPE: UNITED STATES: MINNESOTA: Kandiyohi Co.: growing in tufts in 1-3 feet of water, muddy bottoms; abundant in NE bay of Norway Lake, 6 Sep 1933, C. O. Rosendahl and F. K. Butters 6446 (Lectotype, MIN}; isolectotype, MIN! ). (Lectotype here designated. ) 3c. NAJAS GUADALUPENSIS var. FLORIDANA Haynes & Wentz, Sida 5:262. 1974 Stems 7-51 cm long, 0.1-1.7 mm in diameter. Leaves 0.9-3.3 cm long; laminae 0.5—2.1 mm wide, round obtuse to acuminate, with 18-42 teeth per side; sheaths 1.2-2.5 mm wide, round, with 5—8 teeth per side. Flowers 1 per axil; staminate flowers 1.5-2.4 mm long, the anther elliptic, 1—-loculed; carpellate flowers 1.0-3.5 mm long. Seeds 1.6—2.2 mm long, 0.3-0.8 mm wide, the areolae in ca. 20 rows (Fig. In lakes, streams, and canals, Florida, ccna Alabama, and central Georgia (Fig. 4 TYPE: UNITED STATES: FLORIDA: Dade Co.: abundant in brackish water of ditch along St. Rt. 41 at edge of Everglades National Park, ca. 40 mi. W of Miami, 5 Apr 1972, W. A. Wentz 670 (Holotype, US!; isotypes, GH!, MICH!, OS! ). 3d. NAJAS GUADALUPENSIS var. muenscheri (Clausen) Haynes, comb. & stat. nOv. Najas muenschert Clausen, Rhodora 39:59. 1937. Stems 30-90 cm long, 0.8-1.0 mm in diameter. Leaves 0.9-1.3 cm long; laminae 0.5-1.6 mm wide, acute, with 50-100 teeth per side; sheaths 1.0-1.5 mm wide, round, with 4-8 teeth per side. Flowers 1 per axil; staminate flowers 2.0-3.0 mm long, the anther elliptic, 1-loculed; carpellate flowers 2.9-4.0 mm long. Seeds 3.3-3.8 mm long, 0.5-0.7 mm wide, the areolae in 50-60 rows (Fig. 8F). Shallow water of the Hudson River, New York (Fig. 4). TYPE: UNITED STATES: NEw YorK: Greene Co.: Tidal mudflats of Hud- son River, Imbocht Bay, 3 Sep 1936, W. C. Muenscher and O. F. Curtis, Jr. 5495 (Holotype, BH!; isotype, GH! ). 4. NAJAS GRAMINEA Raffeneau-Delile, Descr. a Hist. Nat. 281. 1813. Caulina alagensis Pollini, Hort. & Prov. Veron. 26. Najas alagensis (Pollini) Pollini, Fl. Veron. 3:49. ry N. seminuda Griff. in Voigt, Hort. Suburb. Calc. 694. 1845. Plants monoecious. Stems to 35 cm long, 0.2-0.5 mm in diameter, spar- ingly branched; internodes 0.4-1.9 cm long, without prickles. Leaves 0.8—2.0 cm long; laminae spreading to ascending, 0.5-1.0 mm wide, acute with 2—3 10 r , j XS ; : - | ! mY 6) | as SCALE | 4 LAMBERT ATBIUTHAL CQUAL-ARZA PROJECTION = of . Fig. Map of North America aS oie documented distribution of Najas Ps ines var. Fash ee va (solid dots), guadalupensis vat. muenscheri (solid eae ), and N. guadalupensis var. sees. a ots with open star). ra 47 teeth at apex, minutely serrulate with ca. 40 teeth per side, the terminal teeth similar in size and structure to lateral teeth, the teeth unicellular, the midrib without prickles; sheaths 1.0-1.5 mm wide, wider than the laminae, deeply auriculate, minutely serrulate with 8-15 teeth per side, the teeth similar in structure and size to those of the laminae. Flowers 1-2 per axil, the staminate in the upper axils, the carpellate throughout, the involucre clear in color; staminate flowers 2.0—3.0 mm long, the involucre beaks 1.0- 1.5 mm long, 4-lobed, the anther 1.5 mm long, 4—loculed; carpellate flowers to 3.5 mm long, the styles to 1.0 mm long, the stigmas 2—lobed. Seeds 1.7- 2.5 mm long, 0.4-0.6 mm wide, greenish-brown, fusiform, the testa pitted, the areolae 4—angled, less than 0.1 mm long, less than 0.1 mm broad, in ca. 35 rows, the end walls slightly raised (Fig. 8G). Chromosome number, 27 = 48 (Sharma and Chatterjee 1967 ) Rice fields, ayo (Fig. 5). TYPE: Egypt (M By the deeply ae sheaths, N. graminea is one of the easiest of the North and Central American Najas to recognize. A native of southeast Asia, the species apparently was introduced into California prior to 1946 and now is evidently quite rare there. I have seen no recent specimens of the species from North or Central America. Rendle (1901) placed this species in the section Nwdae based upon the lack of any floral involucre. Wilde (1961), however, demonstrated the presence of a leaf-like involucre and used this as evidence to abolish the section. 5. NAJAS GRACILLIMA (A. Br. ex Engelm. in A. Gray) Magnus, Beitr. Gatt. Najas 29 LO7 0; laa indica (Willd.) Cham. var. gracillima A. Br. ex Engelm. in A. Gray, Man. . (ed. 5) 681. 1868. Plants monoecious. Stems 4.5—-48 cm long, 0.2-0.7 mm in diameter, slightly branched; internodes 0.1—3.2 cm long, without prickles. Leaves 0.6-2.8 cm long; laminae spreading to ascending, 0.1-0.5 mm wide, acute with 2-3 teeth at apex, minutely serrulate with 13-17 teeth per side, the terminal teeth similar in size and structure to the lateral teeth, the teeth unicellular, the midrib without prickles; sheaths 0.5-1.5 mm wide, wider than the laminae, truncate, minutely serrulate with 7-8 teeth per side, the teeth simi- lar in size and structure to those of the laminae. Flowers 1-3 per axil, the staminate in the upper axils, the carpellate throughout, the involucre bronze to purple; staminate flowers 1.5-2.0 mm long, the involucre beaks ca. 0.8 mm long, 2—lobed, the anther ca. 1.3 mm long, 1-loculed; carpellate flowers 0.5-2.7 mm long, the styles 0.3-1.5 mm long, the stigmas 2—lobed. Seeds 2.0-3.2 mm long, 0.4-0.7 mm wide, light brown, fusiform with the style arising from one side at the apex, the testa pitted, the areolae 4—angled, less than 0.1 mm long, less than 0.1 mm wide, in ca. 40 longitudinal rows, 0 _ounene une cena Sena meee eee yoo MLS 8 FO 400 00 te t000 00 OD RL OMETERS LAMBERT AZIMUTHAL EQUAL AREA PROJECTION Fi Map of North eae showing the documented distribution of Naj graminea (solid square), N. ancistrocarpa (dots with open star), N. arguta (solid triangles), and N. ee (solid dots). 49 the end walls raised (Fig. 8H). Chromosome number unknown. In soft water lakes, Nova Scotia to Alabama, west to Minnesota and Mis- souri; California (Fig. TyPE: Northeast United States (B, destroyed during World War II). Najas gracillima is most similar to N. minor, especially in vegetative condition (see discussion under N. minor). Najas gracillima, however, can be separated from the latter species by its style arising off-center at the apex of the gynecial wall and by its areolae being longer than broad. The species, once more common than it is now (Wentz and Stuckey 1971), apparently cannot tolerate pollution and has become exceedingly rare due to the gradual degradation of lakes and streams in eastern United tates. 6. NAJAS ANCISTROCARPA Magnus, Beitr. Gatt. Najas 7. 1870. Plants monoecious. Stems 7-22 cm long, 0.2-1.0 mm in diameter, pro- fusely branched; internodes 0.2-4.0 cm long, without prickles. Leaves 0.8— 2.6 cm long; laminae usually recurved with es 0.1-0.7 mm wide, acute with 1 tooth at apex, conspicuously serrulate with 5S—9 teeth per side, the terminal tooth slightly larger than but similar in structure to the lateral teeth, the teeth multicellular, the midrib without prickles; sheaths 1.0-1.3 mm wide, wider than the laminae, round to slightly auriculate, serrulate with 3-5 teeth per side, the teeth similar in structure and size to those of the laminae. Flowers 1 per axil, the staminate in the upper axils, the carpellate in the lower axils, the involucre bronze colored; staminate flowers 2.0-3.0 mm long, the involucre beaks ca. 0.4—0.6 mm long, 4—lobed, the anther 1.5-2.0 mm long, 4—loculed; carpellate flowers to 2.5 mm long, the styles less than 0.4 mm long, the stigmas 4—lobed. Seeds 2.5-3.0 mm long, 0.5-0.7 mm wide, greenish-brown, recurved, the testa pitted, the areolae 4—angled, to 0.2 mm long, ca. 0.05 mm broad, in ca. 20 longitudinal rows, the end walls raised (Fig. 81). Chromosome number unknown. n lakes, southern Georgia and northern Florida (Fig. 5). TYPE: JAPAN: Yokohama (B, destroyed during World War IL). Najas ancistrocarpa, according to Rendle (1901) and Miki (1935), exceedingly rare and has been known only from three Japanese islands, Honshiu, Shikoku, and Yokohama. The species is either not as restricted as it was once thought to be or it is spreading. Yang (1974) reported it new to Taiwan, and Haynes and Wentz (1974) recorded it for North America. The species is known from four localities in North America: Open Pond and Cane Water Pond in Decatur County, Georgia; Milton in Santa Rosa County, Florida; and Lake Jackson in Leon County, Florida. This species is another of the many found in southeastern United States and southeastern Asia (Graham 1972). The vegetative organs of N. ancistrocarpa resemble those of N. minor and N. wrightiana by the presence of large teeth scattered along the margin of 50 wo SCALE . ae} . HO Be aE FDO OME TED LAMBERT AZIMUTHAL EQUAL- AREA PROJECTION Fig. 6. Map of North America showing the documented distribution of Najas gracillima. 51 quite narrow leaves. However, N. ancistrocarpa is easily distinguished from the latter two species by its recurved fruits (sometimes crescent-shaped). I know of no other Najas with such curved fruits. 7. NAJAS MINOR Allioni, Fl. Pedemon. 2:221.1785. Caulinia fragilis Willd., Mem. Acad. Roy. Sci. Hist. Sree ee 1798. Ger Fluvialis minor (Allinoi) Pers., Synops. 2:530. Ittnera minor (Allioni) C. C. Gmel., Fl. Bad. . oon. 1808. Najas australis Bory ex Cham., Linnaea 4:501. lichotoma Roxb., Hort. Beng. 71. N. fragilis (Willd.) Rostk. & Schmidt, . Sein 382. 1824. N. indica (Willd.) Cham., Linnaea 4:501. 1829. N. minor var. setacea A. Br., J. Bot. 2: oe 1864. N. setacea (A. Br.) Rendle, ans: Linn. oF London, Bot. 5:422. 1899. N. subulata Thuill., Fl. Paris (ed. 2) 500. 1800. N. ternata Roxb. ex. Gric., Notul. 3:183. ee Plants monoecious. Stems 11-120 cm long, 0.2-1.0 mm in diameter, pro- fusely branched near apex; internodes 0.5-5.8 cm long, without prickles. Leaves 0.5—3.4 cm long; laminae usually recurved with age, 0.1-1.2 mm wide, acute with 1-2 teeth at apex, conspicuously serrulate with 7-15 teeth per side, the terminal teeth similar in size and structure to the lateral teeth, the teeth multicellular, the midrib without prickles; sheaths 1.0—3.0 mm wide, wider than the laminae, truncate to auriculate, serrulate to lacerate with 5 to 8 teeth per side, the teeth similar in size and structure to those of the laminae. Flowers 1-2 per axil, the staminate in the upper axils, the car- eee throughout, the involucre light green to purple; staminate flowers m long, the involucre beaks 0.4-2.1 mm long, 2-lobed, the anther ca. 03 mm long, 1—-loculed; carpellate flowers ca. 2.2 mm long, the styles 1.0-1.2 mm long, the stigmas 2—lobed. Seeds 1.5-3.0 mm long, 0.5-0.7 mm wide, purplish, slightly recurved, the testa pitted, the areolae 4—angled, ca. 0.01 mm long, ca. 0.03 mm wide in ca. 15 longitudinal rows, the end walls not raised (Fig. 8J). Chromosome number, 27 = 12 (Sharma and Chatterjee 1967 Ponds, lakes, and slow moving streams; New York to Illinois, south to Florida, Mississippi, and Arkansas (Fig. 7). Type: ITALY: Pedemont (Holotype, TO, photo, UNA! ). Najas minor, native of the Eastern Hemisphere, apparently invaded North America nearly 50 years ago (Merilainen 1968, Wentz and Stuckey 1971). With the gradual eutrophication of the waters of eastern United States, the species has become quite widespread and, in fact, has become the most abundant Najas in some areas (Wentz and Stuckey 1971). Najas minor, with its mature leaves recurved and with its areolae broader than long and arranged in longitudinal rows like the rungs of a ladder, is one of the more distinctive species of Najas. However, young sterile indi- | ° . eo wie Ms | “ oT oO LOMENE f | 7, | am | LAMBELAT AZIMUTHAL EQUAL- AREA PROUTCTION hse Fig. 7. Map of North America showing the documented distribution of Najas minor. Fig. 8. Najas sears SEM micrographs taken at 20 KV. A. N. marina (X 94 )- B. N. flexilis (X 18). CN, | ks var. gua dalicpens (X 284). 'f Seca Bilas var. eee (X ih uddalupensts . floridana (X 334). F, bee var. muenscher ae .N, wanen (X 23%). H. N. grace Mis (X 19). TN. anchitrocarha 204) . J. N. minor (X 193). KL. N., wrightiana (X 43). L. N. arguta (X 184). oO 54 viduals resemble N. gracillima. Merilainen (1968) indicated that N. minor can be separated from the latter species by the former having recurved leaves and the latter having leaves not recurved. Wentz and Stuckey (1971) demonstrated that the leaves of N. minor do not become recurved until late in the growing season. Therefore, during the early growing season, individuals of N. minor could be mistaken for N. gracillima when one utilizes vegeta- tive characters only. 8. NAJAS WRIGHTIANA A. Br., Sitzber. Naturf. Fr. Berlin 17. 1868. Najas multidentata Koch, Ber. Schweiz. Bot. Ges. 44:341. 1935. N. wrightiana subsp. multidentata (Koch) See Bull. Torrey Bot. Club 73:365. 16. Plants monoecious. Stems 9-45 cm long, 0.3-1.0 mm in diameter, pro- fusely branched; internodes 0.3-5.2 cm long, without prickles. Leaves 0.5- 2.6 cm long; laminae ascending, 0.2-1.3 mm wide, acute with 1 tooth at apex, conspicuously serrulate with 8-22 teeth per side, the terminal tooth similar in size and structure to the lateral teeth, the teeth mulicellular, the midrib without prickles; sheaths 0.7-2.8 mm wide, wider than the laminae, round, serrulate with (1—) 3-5 teeth per side, the teeth similar in structure and size to those of the laminae. Flowers 1-2 per axil, the staminate in the upper axils, the carpellate in the lower axils, the involucre brown to red purple; staminate flowers 1.2-1.9 mm long, the involucre beaks 0.3-0.5 mm long, 2—lobed, the anther ca. 1.0 mm long, 4—loculed; carpellate flowers ca. 2.0 mm long, the styles to 0.7 mm long, the stigmas 4-lobed. Seeds 0.7-1.5 mm long, 0.3-0.5 mm wide, whitish, fusiform, the testa pitted, the areolae S—angled, ca. 0.1 mm long, less than 0.1 wide, in ca. 20 longitudinal rows, the end walls raised (Fig. 8K). Chromosome number unknown. Slow moving streams and ponds, Veracruz to Florida, south to Guatemala, Honduras, and Cuba (Fig. 5). TYPE: CUBA (B, destroyed during World War II). Najas wrightiana superficially resembles N. arguta. Clausen (1946) dis- cussed this similarity and suggested that the two species could be separated by teeth of N. wrightiana being longer than the rest of the lamina is wide, whereas those of N. argwta are shorter than the rest of the lamina is wide. I believe, however, the best characters are those of the seed. The seeds of N. wrightiana are short (1.0-1.5 mm long) and have 5—angled areolae arranged in ca. 20 rows. Najas arguta seeds, on the other hand, are 2.1-2.6 mm long and have elongate areolae arranged in ca. 50 rows. 9. NAJAS ARGUTA H.B.K., Nov. Gen. et Sp. Pl. 1:371. 1815. Caulinia tenella C. G. Nees in Neuwied, ae 2:345. 1824. Najas tenera Sched. Gort. Gel. Anz. 2:715. 1825. Plants monoecious. Stems 15-70 cm ae 0.3-1.0 mm in diameter, pro- fusely branched; internodes 0.5-13.5 cm long, without prickles, Leaves 1.2—2.8 55 cm long; laminae spreading, 0.6-1.9 mm wide, acute to acuminate with 1 tooth at apex, conspicuously serrulate with 15—25 teeth per side, the terminal tooth slightly larger than but similar in structure to the lateral teeth, the teeth multicellular, the midrib without prickles; sheaths 1.2-3.4 mm wide, wider than the laminae, round to truncate, serrulate with 2—7 teeth per side, the teeth smaller than but similar in structure to those of the laminae. Flowers 1 per axil, the staminate in the upper axils, the carpellate in the lower axils, the involucre purple-tinged; staminate flowers 2.0-2.2 mm long, the involucre beaks ca. 0.5 mm long, 4-lobed, the anther to 1.6 mm long, 1-loculed; carpellate flowers 1.7—2.1 mm long, the styles ca. 0.5 mm long, the stigmas 3—4-lobed. Seeds 2.1-2.6 mm long, 0.6-0.8 mm wide, yellowish-white, elongate-fusiform, the testa pitted, the areolae sometimes inconspicuous, S—G-angled, to 0.2 mm long, ca. 0.05 mm broad, in ca. 50 longitudinal rows, the end walls raised (Fig. 8L). Chromosome number unknown. In lakes, Panama and the Dominican Republic (Fig. 5) TYPE: Provenit in aquis prope momprox- Regno novogranatensi (P?). ACKNOWLEDGMENTS I am grateful to Dr. W. Alan Wentz for his help during the course of this work. Appreciation is also owed to Dr. Wentz and Dr. John W. Thieret for their reading the manuscript and to the curators of the herbaria from which specimens were borrowed. A portion of the field work was completed at the University of Michigan Biological Station, and a portion of the labora- tory investigations were carried out at The Ohio State University and at Louisiana State University at Shreveport. REFERENCES ASCHERSON, P. 1864. ae ae pea Brandenberg. I. Berlin. 1034 pp. BOSTRACK, J. M. and W. LLINGTON. 1962. On the determination of leaf m i an arene ie species of Ranunculus, Bull. Torrey Bot. Club 89:1 BRAUN. ne 1864. Revision of the genus ae e a J. Bot. 2:274—279. ar S. S. 1947a ‘oes studie enus Najas in oe United States. hesis, Gail Uni pp. 1947b. Polyploidy in an immersed aquatic angiosperm. (Abstract.) Amer. aa Bot. 34: ae 582 CLAUSEN, ale 1936, Studies in the genus Najas in the northern United States. Rhodora 38. es 345. 1937. A new species ot Najas from the Hudson River. Rhodora 39: 57-60. eae ae - 1946. Najas ar ae in Central America and its relationship to N. wright- jana. Bull. Torrey Bot. 73 :363-3 FASSETT, N. C. 1951. calivihe in the New World. Rhodora 53:137- 155, 161- 182, 185-194, 209-22 FERNALD, M. L. 1923. ae on the distribution of Najas in northeastern America. SGT 29% 105-1 09. HAM, A. 1972. Floristics aS ee of Asia and eastern North America. SEE: Publ. Co., New York. HAYNES, R. R. 1974. A revision of North American Potamogeton subsection Pasilli Re Oa Rhodora 76:564-649. 56 1977. The Najadaceae in the southeastern United States. J. Arnold Arbor. 58: oi mn nd W. A. Wentz. 1974. Notes on the genus Najas (Najadaceae). Sida 5: 259-264, HOLGREN, nd W. KEUKEN. 1974. Index Herbariorum I. The Herbaria of the i = ag ~ o H >) aie = oO a oe, o pa Ln 9 i O° oes Oe ° fe eo) 5 ig ~p nn ct oc = o wn a oO = e = ray ae mmon, ae ae Solidago nemoralis Ait. Old- field Goldenrod. Common, prairie strip; pastures. *Solidago petiolaris Ait. Goldenrod. Infrequent, field edges; woodland. Solidago rigida L. Stiff Goldenrod. Rare, railroad prairie strip. olidago ulmifolia Muhl. Elm-leaf Can aiea, eee ae WS ***Taraxacum officinale Wiggers. Dandelion. Common, pastures; prairie ee N. **Tragopogon dubius Scop. Goat's Beard. Infrequent, pate track ballast. N. *Verbesina helianthoides "Michx . Yellow Crownbread. Infrequent, field edge. Vernonia arkansana DC. ee Ironweed. Infrequent, woodland thickets. Vernonia baldwinit Torr. Western Ironweed. Frequent, righ t-of-way. **Vernonia missurica Raf. Missouri Ironweed. Common, prairie strip; fields. Nanthium strumarium L. Cocklebur. Infrequent, cultivated field edges. CONVOLVULACEAE ***Calystegia sepium (L.) R. Br. Hedge Bindweed. Infrequent, railroad thickets. ** Convolvulus arvensis L. ame ‘Bindweed. Frequent, waste ground sites. Cuscuta gronovii Willd. Dodder. Rare, moist lake border on Solidago spp. Cuscuta pentagona Engelm. Dodder. Rare, prairie swale er ***Cuscuta polygonorum Engelm. Smartweed Dodder. Precuiene lake margin; wet sedge-grass meadow on Polygonum spp. 79 * Ipomoea — (L.) Jacq. Ivy-leaved Morning Glory. Infrequent, waste ground habitats. Ipomoea a L. Small White yaameage Glory. Rare, pasture fence row. *Ipomoea pandurata (L.) G.F.W. Meyer. Wild Potato. Infrequent, field edges. ** Ipomoea purpurea (L.) Roth. poate Morning Glory. Rare, railway ballast. I. CORNACEAE Cornus florida L. Flowering Dogwood. Frequent, subcanopy of oak woods. CRUCIFERAE Barbarea vulgaris R. Br. var. arcuata (Opiz) Fries. Yellow Rocket. Common, pas- ture fields; roadsides; right-of-wa ** Brassica juncea (L.) Coss. Indian niece d; quent, waste ground. N. ** Brassica Raber (DC.) L.C. Wheeler var. peers (Stokes) L.C. Wheeler. Charlock. Infrequent, railroad ballast. N. **Capsella bursa-pastoris L. Shepherd's Purse. Infrequent, waste ground. N. **Cardamine parviflora L. var. arenicola (Britt.) O.E. Schultz. Bittercress. Frequent, disturbed open ground; cultivated fields. Dentaria laciniata Muhl. Toothwort. Frequent, open cherty oak woodlan **Descurainea pinnata (Walt.) Britt. var. brachycarpa (Richards ) ree Tansy Mustard. Infrequent, railroad prairie strip. *Draba verna L. Vernal Whitlow Grass. Rare, roadside; cane! ballast. N. **T epidium campestre (L.) R. Br. Fieldcress. Rare, waste ground by tracks. N. Lepidium virginicum L. Peppergrass. Infrequent, pastures; en of-way. Sisymbrium officinale (L.) Scop. Hedge Mastard, Infrequent, railroad ballast. N. ***Thlaspi arvense L. Field Pennycress. Frequent, roadsides; railroad ballast. N. *Thlaspi perfoliatum L. Perfoliate Pennycress. Rare, waste ground. N CUPRESSACEAE Juniperus virginiana L. Red Cedar. Common, oak woods; pastures; fence rows. CYPERACEAE **Carex annectans Bickn. Scdge. Frequent, moist ditches; sedge-grass meadow. ** Carex artitecta Mackenz. Emmon’s Sedge. Rare, cherty open woods. ** Carex bicknellii Britt. Bicknell’s Sedge. Infrequent, woodland border. *Carex bushii Mackenz. Bush’s Sedge. Frequent, railroad prairie strip. Carex cephalophora Muhl. Oval-headed Sedge. Rare, prairie swale. **Carex crinita Lam. Fringed Sedge. Infrequent, wet sedge-grass mcadow. Carex digitalis Willd. Sedge. Rare, oak-hickory shaded woods. Carex frankii Kunth. Frank’s Sedge. Common, lake margin; oe oo Carex gravida Bailey. Heavy Sedge. Rare, wetl foe sedge-grass mc Carex birsutella Mackenz. Hirsute Sedge. Infrequent, shaded oak ead *** Carex lupulina Muhl. Hop Sedge. Common, wet sedge-grass meadow. Carex lurida Wahlenb. Sedge. Common, lake margin; sedge-grass meadow. Carex meadii Dewey. Sedge. Common, railroad prairie strip. Carex molesta Mackenz. Sedge. Rare, roadside ditch along right-of-way. Carex muhlenbergii Schk. Muhlenberg’s Sedge. Rare, prairie swale by lake. Carex scoparia Beats Pointed Broomsedge. Common, sedge-grass meadow. *** Carex squarrosa L. Squarrose Sedge. Frequent, scdge-grass meadow. ** Carex tribulotdes Wahlenb. Blunt Broomsedge. Frequent, sed ge- -grass meadow. m d Cyperus esculentus L. Yellow Nutgrass. Infrequent, cultivated field edges. Cyperus ovularis (Michx.) Torr. Hedgehog Clubrush. Frequent, prairie strip. 80 *Cyperus psendovegetus as Umbrella Sedge. Rare, wet sedge-grass meadow Cyperus strigosus L. Flatsedge. Common, lake margin; sink-hole pond border. **Dyalichlum a ela (L.) Britt. Three-way Sedge. Common, lake margins. -ocharis acicularis (L.) R. & 8. Wire Spikerush. Common, muddy lake margin. ikerush. Common, Ce margin. * * Eleocharis quadran gulata (Michx.) R. . Square-stemmed Spikerush. Common, muddy lake margin; sink-hole a *Eleocharis tenuts Se Id.) Schultes. var. verrucosa (Svenson) Svenson. Slender Spikerush. Frequent, lake margin; prairie swales Fimbristylis See (L & S. Autumn Sede: Common, sedge meadow. *Scirpus atrovirens Wil goes Bulrush. Infrequent, wet sedge meadow. *** Scirpus pendulus Muhl. Bulensh, Common, roadside ditches; sedge meadow **Scirpus validus Vahl. var. creber Fern. Great Bulrush. Frequent, tie spillway overflow; sink-hole pond margin. DIOSCOREACEAE Dioscorea villosa L, Wild Yam. Infrequent, shady oak-hickory woodland DIPSACACEAE Dipsacus sylvestris Huds. Common Teasel. Infrequent, waste ground. N EBENACEAE Diospyros virginiana L. Persimmon. Common, subcanopy of oak woods; fields EUPHORBIACEAE Acalypha gracilens Gray. Three-seeded Mercury. Rare, railway ballast. Acalypha virginica L. toe seeded M Croton capitatus Michx. Hogwort. Rar ste ground of right-of-way **Croton glandulosa L. var. leptentrionali Muell.-Arg. Sand Croton. cultivated field edge; railroad bal Common, waste ground habitats. e. Common, railroad prairie strip. Euphorbi : Tragia betonicifolia Nutt. Noseburn. Rare, moist woodland thicket by ioe FAGACEAE Quercus alba L. White Oak. Common, canopy of oak-hickory ear mers Common, oak Quercus marilandica Muenchh. Blackjack Oak. ang. Post Oak. Common, canopy of oak ann fence Quercus stellata Quercus velutina Lam. Black Oak. Common, canpy of oak woods; fence rows rows. GENTIANACEAE Gentiana puberulenta Pringle. Downy Gentian. Rare, railroad prairie strip. Sabatia angularis (L.) Pursh. Rose Pink. Infrequent, roadside. GERANIACEAE *Erodium cicutarium (L.) L’ Her. Storksbill. Infrequent, waste ground N. bill. Common, disturbed waste ground. Geranium carolinianum L. Crane Geranium maculatum L. Wild Chee Frequent, shaded oak woodland edges Infrequent, oods; fence rows. GRAMINEAE * Aegilops cylindrica Host. Jointed Goatgrass. Rare, railroad waste ground. N. Agropyron repens (L.) Beauv. Quackgrass. R ures. I. m. as mon, eee prairie strip. mmon, pastures; fields; rae strip. Ma Michx. Bove rtyarass. Sea ee right-of-way. ** Aristida oe Poir. Three-awn Grass. Rare, railway ballast. * Aristida oligantha Michx. Prairie Three-awn. Frequent, anes waste ground. Rare, cultivated fields: railway balla *Bromus commutatus Schrad. Hairy Chess. Common, waste sae habitats. N. why Chess. Common, wast nd habitats; fields. N Cenchrus longispinus (Hack.) Fern. Sandbur. Rare, cultivated field edges nna arund Wo : Beauv. Poverty Oatgrass. Frequent, open ds. ** Digitaria ischaemum (Schreb.) Muhl. Smooth ers Frequent, waste ground. <2 Dios sanguinalis (L.) Scop. Hairy Crabgrass. Common, open waste ground ***Echinochloa crusgalli (L.) Beauv. Barnyard Grass. Frequent, moist waste ground Bian indica (L.) Gaertn. Goose Grass. Infrequent, area of tracks, N. ‘El ‘Ym Ue. pecan L. Canada Wild i oe railr MUS ViT gZinich cL. Virginia Wild R Glyceria striata (Lam.) Hitchce. deum ** * Hor usillum Nutt. Little Barl ommon, waste ground; roadsides Sonam ae L. Barley. Rare, cultivated field edges; railroad ballast. I **Koeleria pyramidata (Lam.) Bea Junegrass. Infrequent, prairie stri Leersia oryzoides Sw. Rice Cutgrass. Common, wet sedge-grass meado Leptoc filiformis (Lam.) Beauv. Red Sprangletop. Rare, cultivated field *Lolium perenne e L. Perennial Paes Infrequent, roadside waste ground. I. Muhlenbergia schreberi J.F. mel. Nimblewill Muhly. ek roadsides. ommon, sedge meadow; prairie cules: m oligosanthes Schultes var. seribnerianum (Nash) Fern. Panic Grass. Fre- quent, railroad prairie strip. Panicum rigidulum Bosc. Munro Grass. Common, prairie swale; sedge meadow. mon, lake margin; prairie swales. ** Paspalum aaa Poir. var. glabrum Vasey. Beadgrass. Infrequent, moist waste ground; roadsides 82 ***Phalaris arundinacea L. Reed Canary Grass. Common, wet sedge-grass meadow. Phleum pratense L. Timothy. Frequent, pastures; cultivated fields. I Poa annua L. Annual Bluegrass. Rare, waste baa site. N. Com ie strip; pasture oo scoparium (Michx.) Nash. ‘Little Blu estem. Common, eis strip. Secale cereale L. Rye. Infrequent, roadsides; railroad = lat ** * Setaria fabertt Herrm. Giant Foxtail. Co , open waste ground. N. ** *Seraria glauca ( mo (L.) Beauv. Yellow Foxtail. Conca, waste ground. N. *Setaria viridis (L.) Beauv. Green Foxtail. Infrequent, railroad ballast. So rghastrum avenaceaum (Michx.) Nash. Indian Grass. Common, prairie strip. (L.) Moench. Sorghum. Rare, cultivated fields; right-of- “way. I. ** Spartina pectinata Tink. Prairie Cordgrass. Piccie ** Sporobolus eh ead (Torr.) Wood. aA Diopeed. Rare, pastures. Tridens flavus (L.) Hitche. Purpletop. oe *Tripsacum dali L. Eastern Gamagra vate requent, sedge meadow. *** Triticum aestivum L. Wheat. Rare, eaeae fields; railway ballast Vulpia octoflora ai Rydb. Six es Fescue. Frequent, peseaics: woods. **Zea mays L. Corn. Rare, railway bal nm , prairie strip; joallsides. HALORAGIDACEAE *Prosperpinaca palustris L. Mermaid Weed. Rare, sink-hole pond; lake spillway. HYPERICACEAE ** Hypericum drummondii (Grev. & Hook.) T. & G. Nits-and-lice. Rare, prairie strip. Hypericum gentianoides (L.) BSP. Pineweed. Rare, railroad prairie strip. Hypericum mutilum L. Dwarf St. Johns-wort. Frequent, muddy lake margin. Hypericum perforatum L. Common St. Johns-wort. Frequent, prairie trip. N. Hypericum a. Lam. Spotted St. Johns-wort. Frequent, prairie strip. IRIDACEAE Sisyrinchium angustifolium Mill. Blue-eyed Grass. Rare, cherty woodland. ** Sisyrinchinm campestre Bickn. Prairie Blue-eyed Grass. Common, railroad prairie. JUGLANDACEAE Carya ovata (Mill.) K. Koch. Shagbark Hickory. Common, canopy of oak woods. texana Buckl. Black Hickory. Frequent, canopy of oak-hickory woodland. Carya tomentosa Nutt. Mockernut Hickory. Common, canopy of oak woodland. Juglans nigra L. Black Walnut. Infrequent, open pasture lands Q as JUNCACEAE ** Tuncus acuminatus Michx. Sharp-pointed Rush. Frequent, sedge meadow; swales. Juncus brachycarpus Engelm. Short-fruited Rush. Common, Juncus dudleyi Wieg. Dudley’s Rush. Rare, roadside ditches; sedge meadow Luzula bulbosa (Wood) Rydb. Wood Rush. Frequent, cherty open seed and, LABIATAE Cunila origanoides (L.) Britt. Stone Mint. Infrequent, cherty oak woods. edeoma hispida Pursh. Mock Pennyroyal. Rare, railroad prairie strip. Isanthus brachiatus (L.) BSP. False Pennyroyal. Rare, moist lake margin. ***Lamium amplexicaule L. Henbit. Frequent, open ground; roadsides. N. ***lLamium purpureum L. Dead Nettle. Infrequent, roadside embankments. N. "*“*Lycopus rubellus Moench. Water Horehound. Common, lake margin; sink-holes. “**Lycopus virginicus L. Virginia Bugleweed. Infrequent, sedge-grass meadow Monarda fistulosa L. Wild Bergamot. Frequent, woodland edges; prairie ste Perilla frutescens (L.) Britt. Beefsteak Plant. Rare, moist embarkment. N. Physostegia virginia ana (L.) Benth. False Dragonhead. Rare, shaded thicket. *Prunella vulgaris L. var. lanceolata (Bart.) Fern. Self-Heal. Common, field edges; pastures; r ae of-way; lake border. Pycnanthemum tenuifolium Schrad. Slender Mountain Mint. Frequent, prairie strip; border. Scutellaria lateriflora L. Mad-dog Skullcap. Infrequent, muddy lake margin. **Scutellaria ovata Hill. Heart-leaved Skullcap. Rare, ae oak woodland. Scutellaria parvula Michx. Small Skullcap. Infrequent, open prairie strip. Teucrium canadense L. Wood Sage. Common, coe edges; prairie strip. LAURACEAE Sassafras albidum oe Nees. Sassafras. Common, subcanopy of oak woodland; fence row thicket LEGUMINOSAE Amorpha canescens Pursh. Lead Plant. Rare, railroad prairie strip; fence row. Aptos americana a Ground Nut. ee wet Siceae’ Sa Cassia fasciculata Michx. eee Pea. Common, sar anes oo. Cassia nictitans L. Sensitive Pea. Infrequent, cherty open ds Cercis canadensis L. Red Bud. Infrequent, subcanopy of an hickory woods. *Crotalaria sagittalis L. Rattlepod. Common, railroad prairie strip. * Desmanthus illinoensis (Michx.) MacM. ex Robins. & Fern. Illinois Mimosa. Rare, prairie strip Desmodium ciliare UMuhl. ) DC. Hairy Tick Trefoil. Frequent, Prairie strip. *Desmodium illinoense Gray. Illinois Tick Trefoil. Frequent, ee strip. Desmodium marilandicum (L.) DC. Small-leaf Tick Trefoil. Rare, oak woods. Desmodium paniculatum (L.) DC. Panicled Tick Trefoil. Gonmion: prairie strip. Desmodium rotundifolium DC. Round-leaf Tick Trefoil. Infrequent, open woods. ens ae (Torr.) T. & G. Sessile-leaf Tick Trefoil. Frequent, rail- road right-of-way str Gleditsia pie LH Honey Locust. Infrequent, open oak woods; pastures. Lespedeza capitata Michx. Round-headed Bush Clover. Frequent, prairie strip. ***Lespedeza cuneata (Dumont) G. Don. Sericea Lespedeza. Common, waste ground. I. Lespedeza procumbens Michx. Traili ing ae sae ae open woodland. ***T eshedeza stipulacea Maxim. Korea spedeza. Frequent, waste ground. I. Lespedeza striata (Thunb.) H. & i. eave Lespedeza. Rare, roadsides. I. Lespedeza violacea (L.) Pers. Violet Prairie Clover. Frequent, railroad prairie. Lespedeza virginica (L.) Britt. Slender Bush Clover. Common, eae prairie. Medicago lupulina L. Black Medic. Infrequent, roadside waste grou N. *** Medicago sativa L. Alfalfa. Frequent, pastures; roadsides; field ee L Melilotus albus Desr. White Sweet Clover. Common, roadsides; pastures. I. *Melilotus officinalis (L.) Lam. Yellow Sweet Clover. Frequent, waste ground. I. Petalostemum candidum (Willd.) Michx. White Prairie Clover. Rare, prairie strip. Petalostemum purpurcum (Vent.) Rydb. Purple Prairie Clover. Rare, prairie strip. Psoralea psoralioides (Walt.) Cory. var. eglandulosa (Ell.) Freeman. Sampson's nakeroot. Common, railroad prairie stri Schrankia uncinata Willd. Sensitive Brier. Frequent, railroad prairie strip; open woodland. Strophostyles helvola (L.) Ell. Wild Bean. Rare, fence thicket; Sevan Stylosanthes biflora (L.) BSP. Pencil Flower. Common, open wooc ; prairic strip. Tephrosia virginiana (L.) Pers. Goat's Rue, Frequent, open woods; etre strip. *Trifolium campestre Schreb. Large Hop Clover. Common, waste ground, *Trifolium pratense L. Red Clover. Frequent, fields; roadsides. N. *Trifolium reflexum 1. Buffalo Clover. Rare, roadsides Trifolium repens L. White Clover. Common, dicnubed waste plac Vicia dasycarpa Tenore. Hairy-fruited Vetch. Infrequent, railroad ball ast. N. LILIACEAE Allium canadense L. Wild Onion. Frequent, pastures; railroad prairie strip. *Allinm sativum L. Garlic. Rare, railway ballast edge. I. inm vineale L. Field Garlic. Infrequent, roadsides; pasture fields. *** Asparagus officinalis L. Garden eran Infrequent, fielc edges. I. *Camassia scilloides (Raf.) Cory. Wild Hyacinth. Infrequent, prairie strip. Erythronium albidum Nutt. White Trout Lily. Infrequent, cherty oak woods. ** Melanthinm virginicum L. Bunch Flower. Rare, wet sedge-grass meadow edge. acacia oe (L.) Britton. False Garlic. Common, prairie strip; fields; ww A. as ™ ~— a. odland e Poly, ee biflorum Sle Ell. Solomon’s Seal. Infrequent, shady oak woods. Smilacina racemosa (L.) Desf. False Solomon's Seal. Frequent, moist oak woods. Smilax bona-nox L. Catbrier. Infrequent, open woods; fence rows; roadside. Smilax hispida Muhl. Bristly Greenbrier. Infrequent, shaded oak woodland. LINACEAE Linum sulcatum Riddell. Yellow Flax. Rare, railroad prairie strip. LOASACEAE *** Menzelia oligosperma Nutt. Stick-leaf. Frequent, railroad ballast. LYTHRACEAE Cuphea petiolata (L.) Koehne. Clammy Cuphea. Frequent, open waste ee Rotala ramosior (1..) Koehne. Toothcup. Tafeequent: muddy lake margin MALVACEAE aun theophrasti Medic. Velvetleat. eae, railroad ballast; fields. N. Althaea rosea L.. Hollyhock. Rare, railroad ballast. I. anne digitata Nutt. Poppy Mallow. Rare, on cherty wor ee US eae oo Woolly Rose Mallow. Common, ire margin; sedge **Sida ante L. Prickly Mallow. Infrequent, cultivated field edges. N. MORACEAE ** Morus alba L. White Mulberry. Rare, cultivated field edge; oak woods. I. NYCTAGINACEAE *** Mirabilis nyctaginea (Michx.) MacM. Four O'Clock. Common, railroad ballast. NYMPHACEAE **Nelumbo lutea (Willd.) Pers. American Lotus. Common, emergent in lake. 85 NYSSACEAE *Nyssa sylvatica Marsh. Sour Gum. Infrequent, subcanopy of oak-hickory woods OLEACEAE Fraxinus americana L. White Ash. Infrequent, oak-hickory woodland. ONAGRACEAE Gaura biennis L. Biennial Gaura. Infrequent, railroad prairie strip. * *Iudwigia alternifolia L. Seedbox. Common, margin; sink-hole pond border. Ludwigia palustris (L.) EL var. americana (DC.) Fern. Marsh Purslane. Com mon, muddy ae margin; sink-hole *Oenothera biennis L. Common Evening Primrose. Common, moist waste ground. ** * Qenothera iat Hill. Ragged Evening Primrose. Infrequent, waste ground OPHIOGLOSSACEAE Botrychinum virginianum (L.) Sw. — Fern. Infrequent, moist oak wood- land; sink-hole pond embankmen ORCHIDACEAE *Spiranthes cernua (L.) Richard. Nodding Ladies Tresses. Rare, prairie strip *Spiranthes gracilis (Bigel.) Beck. Green-lipped Ladies Tresses. Rare, railroad prairie strip. OXALIDACEAE **Oxalis dillenii Jacq. Yellow Wood Sorrel. Frequent, waste ground. Oxalis violacea L. Violet Wood Sorrel. Frequent, open woods; railroad prairie PASSIFLORACEAE Passiflora lutea L. var. glabriflora Fern. Passion Flower. Rare, shaded woods. PHYTOLACCACEAE Phytolacca americana L. Pokeweed. Common, disturbed open ground. PLANTAGINACEAE Plantago avristata Michx. Bracted Plantain. Frequent, open disturbed areas. Plantago lanceolata L. English Plantain. infrequent! waste ground; ee N. i ds border. Plantago virginica L. Hoary Plantain. “Antreanent, railroad prairie ‘strip. POLEMONIACEAE Phlox pilosa L. Prairie Phlox. Infrequent, railroad prairie strip. POLY GALACEAE *Polygala incarnata L. Pink Milkwort. Rare, railroad prairie strip **Polygala sanguinea L. Field Milkwort. Common, railroad prairie strip. Polygala senega L. Seneca Snakeroot. Rare, moist shaded woodland by lake. POLYGONACEAE **Persicaria coccinea (Mubhl.) eas Water Smartweed. Common, sink-hole pond; Cedar Ga Ree enue (Michx.) Small. Mild Water Pepper. Common, lake margin rie swales; sedge-grass meadow Bocsicaia lapathitolia (L.) Small. Pale Smmareweed Frequent, lake margin. N. emergent in 86 Persicaria pensylvanica (L.) Sm Pinkweed. Common, lake margin; swales. Persicaria punctata (EI.) Small. eee Rare, lake border by spillway. **Polygonum aviculare L. Knotweed. Frequent, disturbed open waste ground. N. Polygonum scandens L. False Buckwheat. Frequent, wet sedge-grass meadow thick- et ets. Polygonum es eel L. Virginia Knotweed. Rare, moist shaded oak woods. Rumex acetosella L. Red Sheep Sorrel. Common, throughout waste ground. N. *Rumex see L: oe Dock. Common, waste ground habitats. N. POLYPODIACEAE Asplenium platyneuron (L.) Oakes. Ebony Spleenwort. Rare, cherty embankment. Pteridium aquilinnm (L.) Kuhn. Bracken Fern. Rare, cherty open woodland next PORTULACACEAE Claytonia virginica L. Spring Beauty. Frequent, open woods; pastures. POTAMOGETONACEAE Potamogeton diversifolius Raf. Pondweed. Common, emergent of Cedar Gap Lake. PRIMULACEAE Dodecatheon meadia L. Shooting Star. Infrequent, railroad aan strip. ** ysimachia bybrida Michx. Loosestrife. Rare, lake margin thic Lysimachia lanceolata Walt. Loosestrife. Frequent, moist shaded ‘ceellal RANUNCULACEAE Anemone virginiana L. Thimbleweed. Infre ae cherty open woods. Anemonella thalictroides (L.) Spach. Rue Anemone. Rare, cherty open woods. *Delphininm ajacis L. Rocket Larkspur. Rare, noadide gas tial eh Mase Walt. var. crispum Perry. Wild Blue Larkspur. Infre- ue railrc ight-of-way. + Delphiniuum tricorne Michx. Dwarf Larkspur. Rare, cherty moist woodland. Ranunculus abortivus L. Small-flowered Crowfoot. Common, waste ground. — fascicularis Muhl. var. apricus (Greene) Fern. Prairie Buttercup. Fre- uent, cherty oak woodland; prairie strip. ee laxicaulis (YT. & G.) Darby. Spearwort. Frequent, sedge meadow. ia asi micranthus Nutt. Small-flowered Buttercup. Rare, cultivated field. *Thalictrum revolutum DC. Waxy Meadow-rue. Infrequent, railroad prairie. RHAMNACEAE Ceanothus americanus L. var. pitcheri T. & G. New Jersey Tea. Frequent, railroad prairie strip; open woods. ROSACEAE Agrimonia parviflora Ait. Swamp Agrimonia. Rare, sedge-grass meadow edge. Amelanchier arborea (Michx. f.) Fern. Shadbush. Infrequent, oak-hickory woods. k Crataegus crusgalli L. Cockspur Hawthorn. Rare, fence row thicket. Fragaria virginiana Duchesne. Wild Strawberry. Common, prairie strip; woods. Geum canadense Jacq. var. camporum (Rydb.) Fern. White Avens. te 71 } woods near lake. moist shade Gillenia stipulata (Muhl.) Trel. Indian Physic. Infrequent, cherty woodland. **Potentilla norvegica L. Rough ee Ae Rare, spillway _N. Potentilla recta L. Sulfur Cinquefoil. Common, disturbed g nine habita . Potentilla simplex Michx. Old-field Cinquefoil. Cute oak oe noe tairie strip. **Prunus hortulana Bailey. Hortulan Plum. Rare, fence row border. I. **Prunus mahaleb L. Mahaleb Cherry. Rare, oak woodland border. i Prunus munsoniana Wight & Hedrick. Wild Goose Plum. Rare, fence rows. *Prunus persica (L.) Batsch. Peach. Rare, remnant from old orchard. i y of Bak woods. > woods. I osa carolina L. Pasture Rose. Fre t, pastures; prairie strip; woods Rosa multiflora Thunb. Multiflora Rose frequent, fence rov ickets. I >, Infrequent, pastures; prairie strip. *Rubus flagellaris Willd. Northern Dewberry. Common, fence rows; prairie strip; woodland thickets. Rubus pa L. Black Raspberry. Rare, mesic shaded woodland thickets. **Rubus pensylvanicus Poir. Highbush Blackberry. Common, pastures; fields; rail- road prairie strip; roadsides; sedge thickets. RUBIACEAE Cephalanthus ee L. Buttonbush. Wa Common, lake margin; sink- aa pond. Diodia teres ugh Buttonweed. Common, waste ground habitat Galium aparine L. Sea Frequent, mesic oak woods; sedge ee **Galium obtusum Bigel. Blunt-leaf Bedstraw. Rare, shaded oak woods. **Galium tinctorium L. Stiff Bedstraw. Frequent, sink-hole pond border. **Galinm virgatum Nutt. Dwarf Bedstraw. Rare, railroad prairie strip. Hedyotis es Raf. Small Bluets. Common, pastures; prairie s tri Hedyotis longifolia (Gaertn.) Fosb. Long-leaved Bluets p. Infrequent, railroad prairie strip. Hedyotis nigricans (Lam.) Fosb. Narrow-leavec 1 Bluets. Rare, cherty woods. SALICACEAE Salix uneasy Michx. Ward’s Willow. Frequent, lake margin. Salix humilis Marsh. Prairie Willow. eee arise strip; field edge. ** Salix nigra Nich. Black Willow. Common, lake margin; sink-hole pond. SANTALACEAE Comandra richardsiana Fern. False Toadflax. Common, railroad prairie strip. SAXIFRAGACEAE *Heuchera richardsonii R. Br. Prairie Alumroot. Rare, *Penthorum sedoides L. Ditch Stonecrop. Common, lake m 9 railroad prairie argin; sedge Ge Ribes missouriense Nutt. Missouri Gooseberry. Frequent, Sik woods; fence rows. SCROPHULARIACEAE *** Agalinus tenuifolia (Vahl.) Raf. Slender False Foxglove. Frequent, lake margin. **Tindernia dubia (L.) Pennell. False Pimpernal. Infrequent, muddy lake margin. Mimulus ringens L. Monkey Flower. Frequent, sedge-grass meadow zone Pedicularis canadensis L. Lousewort. Infrequent, ence cherty woodland. SUS m : Veronica arvensis L. Corn Speedwell. Frequent, disturbed open ground **Veronica peregrina L. Purslane Speedwell. Infrequent, waste ground sites. 88 Veronicastrum virginicum (L.) Farw. Culver’s Root. Rare, railroad prairie. SOLANACEAE Datura stramonium L. var. tatula (L.) Torr. Jimson Weed. Infrequent, railway Oi er t; cultivated fields salis heterophylla Nees. Clammy Ground Cherry. Rare, railroad ballast. ven subglabrata Mack. & Bush. Smooth Physalis. Frequent, waste ground. **Solanum americanum Mill. Black Nightshade, Rare, cultivated fields. Solanum carolinense L. Horse Nettle. Common, pastures; roadsides; fields. TYPHACEAE Typha latifolia L. Cat-tail, Common, lake margin; prairie swales; ditches. ULMACEAE **Celtis occidentalis L. Hackberry. Infrequent, oak woods; sink-hole pond border. Ulmus alata _ chx. Winged Elm. Infrequent, open oak-hickory wood Ulmus americana L. American Elm. Infrequent, oak woods; pasture edges. Ulmus rubra L. ‘Slippery Elm. Common, oak woodland; “fence rows; pastures. UMBELLIFERAE Angelica venenosa (Greenway) Fern. Wood Angelica. Rare, prairie swale. *Chaerophyllum procumbens (L.) Crantz. Wild Chervil. Common, waste ground. Cicuta maculata L. Water Hemlock. Common, lake margin; sedge-grass meadow. *Pastinaca sativa L. Wild Parsnip. Rare, railway ballast. Polytaenia nuttallii DC. Prairie Parsley. Infrequent, Paileoad prai **Sanicula canadensis L. Black Snakeroot. Infrequent, shaded cae a oos Torilis japonica (Houtt.) DC. Hedge Parsley. Common, waste ground sites. N. Zizia aurea (L.) Koch. Golden Alexanders. Rare, wet sedge-grass meadow. URTICACEAE Boemeria handles (L.) Sw. False Nettle. Infrequent, damp sedge meadow. Pilea pumila (L.) Gray. Clearweed. Infrequent, shaded mesic lake border VALERIANACEAE Valerianella radiata (L.) Dufr. Corn Salad. Common, waste ground habitats. VERBENACEAE Phyla sous (Michx.) Greene. Frog-Fruit. Infrequent, lake marg *Verbena bracteata Lag. & Rodr. Prostrate Vervain. Rare, roadside to fe Verbena conden L. Rose Vervain. Infrequent, Pits d prairie strip. Verbena hastata L. Blue Vervain. Frequent, wet sedge-grass meadow Verbena simples ae: Narrow-leaved Vervain. Frequent, waste ground areas. Verbena stricta Vent. Hoary Vervain. Infrequent, pastures; roadsides. Verbena urticifolia L White Vervain. Common, lake margin; sedge meadow. VIOLACEAE **Viola missouriensis Greene. Missouri Violet. Rare, cherty woodland. Viola pedata L. Bird-foot Violet. Frequent, railroad prairie; oak woods. Viola pratincola Greene. Common Blue Violet. Frequent, fields; oak woods. Viola rafinesquit Greene. Field Pansy. Common, prairi i *Viola sagittata Ait. Arrow-leaved Violet. Infrequent, railroad prairi s. ot = uo] a) a .o c = oO wn 89 VITACEAE ***Parthenocissus quinquefolia (L.) Planch. Virginia Creeper. Infrequent, oak- hickor Vitis aestivalis Michx. Summer Grape. Infrequent, oak-hickory woodlan Vitis cinerea Engelm. Grayback Grape. Frequent, oak-hickory woods; chicken Vitis vulpina L. Frost Grape. Infrequent, oak woodland thickets by lake. REFERENCES BRAUN, E. L. 1950. Deciduous forests of eastern North America. Hafner Publishing Sone New York. 596 pp. : 1955. The phytogeography of unglaciated eastern United States and its interpretation. Bot. Rev. 21: ae 375. CORR D. S. and M. C. NSTON. 1970. Manual of the vascular plants of pee pie Research Bee ates Renner, Texas. 187 ae ee N. M. 1938. cae of the eastern United States. McGraw-Hill ook Company, ey York. 714 LAWRENCE, G. M. ee Taxonomy of vascular plants. Macmillan Publishing Company, Inc., New York. McCRACKEN, M. H. et os logical Survey and W Mis MCQUIGG, J. D. ine Ciisopianhy of ae Unieed States, oe 60—23: Missouri. SDC, ae | Data Service, Washington, D.C. STEYERMARK, J. A. 1940. Studies ‘of the veer of MEE I. Natural plant associations and succession in the Ozarks of Missouri. Field Mus. Nat. Hist. Bot. 475. Ser. 9:349— vations 1961. Peg oe = of Missouri. Missouri Geo- : We sarge history of the Ozark Forest. University of Missouri ~~ Press, s, Columbia, Missou pp. ___.. 1963. Flora of Micur Iowa State ai Press, Ames, Iowa. 1728 pp. USDA, SOIL CONSERVA TION SERVICE. 1975. Southwest Missouri resource con- servation and development project plan, Columbia, Missouri. 131 pp. TAXONOMIC NOTES AND NEW SECTIONS OF CLITORIA SUBGENUS BRACTEARIA (LEGUMINOSAE) ’ PAUL R. FANTZ Fairchild Tropical Garden, Miami, Florida 33156 Botanists have traditionally followed Bentham (1858) recognizing three natural groups in the genus Clitoria, treating them as sections, and ignoring Baker (1879) who elevated two of them to the level of subgenus. Species with woody habits (i.e. trees, tall shrubs, and woody vines) are usually in- cluded in section Clitorianthes Bentham. The name Clitorianthes is a super- fluous name for Bractearia Martius ex Bentham (1837). In recent mono- graphic studies of Clitoria, Fantz (1977) concluded that these three natural groups, on the basis of morphological and distributional data, supported b the scant cytological and developmental data available, should be treated at the subgeneric level. The subgenus of woody members contains half of the 58 species in Clitoria, a number of which are endemic to certain refugia recognized by Prance (1973) and as yet undescribed. The 29 species of this subgenus can be segregated into four sections. This paper provides taxonomic notes on subgenus Bractearia and describes the new sections and two new species. In his revisionary treatment of Clitoria, Bentham (1847) recognized three sections. Section Bractearia Mart. was described for the first time by fruticose habit, trifoliolate leaves and bracteoles equal to or longer than the calyx. Bentham included five species in this section: C. amazonum Mart. ex Benth., C. acuminata Benth., C. racemosa Benth., C. poitaei DC. and C. Pe Benth. In article on plants collected by Schomburgk in British Guiana, Bentham (1839) substituted the name Dendrocyamus for Bractearia justify- ing the change by the fact that Bractearia was used as a generic epithet in the Rubiaceae and as a sectional epithet in the genus Chaetogastra (Melastomataceae ). Although Bentham (1858) continued to recognize three sections in Clitoria, he again substituted a new name for the woody section, now calling it Clitorianthes. No comment was made on the changed named, nor was reference made to any pre-existing names for this section. Bentham here changed the sectional diagnosis of section Clitorianthes to erect shrubs or — studies on Clitoria accomplished at the IFAS Herbarium, Dept. of ny, University of Florida, Gaines ville, Florida. Florida Agricultural Experiment Sater Journal Series No. 1344. SIDA 8(1): 90-94. 1979. o1 tall climbers with three leaflets and a flat or slightly convex, coriaceous legume. He further divided the nine species in section Clitorianthes into two unnamed groups distinguished by the bracteoles. The first group in- cluded six species with bracteoles narrow or much shorter than the calyx. Of the six species included, only Clitoria arborescens (synonym: C. poitaei DC.) had figured in his original list of 1837. The second group was char- acterized by ovate coriaceous bracteoles subequal to the calyx. three species now placed in this group had been assigned to this section in 1837. Their names now were reported as C. amazonum Mart.. ex Benth. (synonym: C. acuminata Benth.), C. hoffmannseggii (a superfluous name for C. arborea Benth.) and C. racemosa Benth. Publications of the next century which included Clitoria were primarily of a floristic nature. Authors adopted Bentham’s treatment and used the name Clitorianthes for the section which included the woody species. Baker (1879), in a floristic treatment of legumes in India, elevated two of Ben- tham’s sectional names to the level of subgenus. Section Clitorianthes Benth. was not included in this revisionary change, presumably because none of its species occur in India. No other floristic work on Clitoria has recognized Baker’s treatment of subgenera, the genus always being divided into sections following Bentham’s treatment of 1858 In the first monographic treatment of the genus in the past century, Fantz (1977) concluded that the three natural groups can be distinguished morpho- logically by the fruits and seeds, supported by differences in the leaves, calices, androecia, and gynoecia. Distributional patterns, presence or absence of cleistogamy, and the limited cytological and developmental data available all support the recognition of these three groups as separate entities, and at the level of subgenus in agreement with Baker. Section Clitorianthes Benth. is thus elevated to the level of subgenus and Bractearia is adopte as the oldest and only legitimate name published for this woody group. Bentham’s diagnosis of the woody group is modified to include additional characteristics which distinguishes it from the other two subgenera. CLIToRIA L. subgenus Bractearia (Mart. ex. Benth.) Fantz, stat. nov. Clitoria L. sect. Bractearia Mart. ex Benth., Ann. Wien. Mus. Natur. 2: 115. 1837. Trees, tall erect shrubs or treelets, and lianas. Leaves 3-foliolate, long- stalked (petiole 4-16 cm), large (leaflets commonly 8-28 cm long x 3-23 cm wide); petiolules elongate (4-10 cm long). Inflorescence woody, panic- ulate, subpaniculate or racemose-nodose; usually several- to many-flowered, the flowers all chasmogamous. Calyx subcoriaceous, multi-nerved, persistent in fruit. Ovary elongate (1-2 cm). Staminal tube elongate (2.5-4 cm). Fruits flat or occasionally weakly convex around the seeds and depressed between them, ecostate, long-stipitate (stipe 12-33 mm), coriaceous, 6-25 cm long x cm wide. Seeds smooth, compressed, orbicular or nearly so, 7-16 mm in diameter; germination epigeal. oe LECTOTYPIC SPECIES: C. amazonum Mart. ex Benth. (LECTOTYPE: Brazil, Para, ad fluy. Amazonum prope Ponte de Mattary, Sep, Martins 2740, M). Bentham (1837) placed five species in section Bractearia when he first described the group. Of these, C. amazonum is designated as the lectotype because it was from the type collection that Bentham obtained the name Bractearia Mart. At Munich, six sheets of the type collection (M 12408- 12413) are contained in one folder, sheet M 12408 bearing the identification Bractearta amazonica Mart. and annotated Clitoria amazonica by Bentham. KEY TO THE SECTIONS OF SUBGENUS BRACTEARIA 1. Bracteoles 6-16 mm wide, coriaceous, 10-40 mm long and subequal to the calyx (it often hidden from view), rarely shorter; inflorescence terminal or axillary (ap- pearing with the leaves), paniculate to acacia the primary lateral aaa ie bear the pedicels either short (1-5 mm long) or elongate (to 30 mm long). 2. Calyx strigulose with oe short Ane ant trichomes, to glabrate, or a pilose, its surface easily seen; inflorescence straight to slightly curved, paniculate to subpaniculate, ered Sements ascending in a nearly straight line (non flexuous); trees or occasionally tall erect shrubs, rarely with a climbing apex. 1. Bractearta 2. Calyx velutinous with short, eee a more or fe any echoes its surface hidden; inflorescence flexuous, at least toward the apex, racemose-nodose, inter- nodal segments weakly to strongly bent in the opposite direction; lianas, occa- sionally erect shrubs or treelets 2. Flexuosae 1. Bracteoles 1-4, rarely 6 mm wide, giiceiae cis ena 2 12 mm ioue and short- r than the ie or rarely élongate and subequal to the calyx, but then always narrow (calyx conspicuous); inflorescence axillary, and then often appearing before the leaves and/or cauliflorous, racemose, the primary ase branches bearing the pedicels lacking or represented by a conspicuous knob to 1 mm long (=nodose). 3. Calyx tube 6-13 mm long with minute lobes 1-3 (—5) mm long; legume slightly convexed around the seeds and ee aaa, ——e between them at maturity; trees or tall shrub- ae rarely lianas. 3. Brachycalyx 3. Calyx tube 10-30 mm long with conspicuous lispes 4— 18 mm iba ; legume flat, sometimes raised around the seeds but not uci between them lianas or rarely shrubs-treelets . . ou Bats # ze 4, ‘Cualiiioras 1. CLrroriA L. subgenus BRACTEARIA (Mart. ex Benth.) Fantz sect. BRAC- TEARIA Clitoria L. sect. Bractearta Mart. ex Benth., Ann. Wien. Mus. Natur. 2: 115. 1837. Clitoria L. sect. Dendrocyamus Benth., Ann. Ant. Hist. 3: 44. 1839; nom. on erfl. Clitoria L. sect. Clitorianthes Benth., Journ. Linn. Soc. 2: 41. 1858; nom. superfl. Abores vel interdum frutices alti. Inflorescentiae sansa ia vel subpaniculatae, lignosae; rami laterales primarii pedicellifert (1) 4-30 mm longi; axes centralis non flexuosus. Calyx pubescentia dispersa vel plerumque glabratus, aes brevibus, appressis, infrequenter pe Bracteolae coriaceae, calycem subaequantes et pler- um que ia. 6-1 m latae et (11-) 1 < mm longae. Legumen planum, coriaccum vel sublignosum, ee latum, (1.5-) 2-4 cm latum, suturis incrassatis; pubescentia glabrata plerumque vel ee Microscopicis uncinatis, magni- trichomatibus ribi repertis secus suturas. ee) Members of section Bractearia are commonly collected in forests along the Amazon River and its major tributaries. The section includes seven species (Fantz, 1977): C. arborea Hoffm. ex Benth.; C. amazonum Mart. ex Benth., C. fairchildiana Howard, C. nervosa Herz., and three to be described in a succeeding article. 2. CLITORIA L,. subgenus BRACTEARIA (Mart. ex Benth.) Fantz sect. Flexuosae Fantz, sec. nov Frutices lignosi saepius volubiles raro arbores. Inflorescentia debilis vel valde flexuosa prope apicem, segmentis internodiorum abruptly flexuosis, subpanicula; ramis laterales primarii pedicelliferi inconspicuis, subsessilibus ad 6 mm _ longae. Calyx pubescentia conferta, trichomatibus brevibus, appressis, aliqguanto sericeis. Brae coriaceae, calycem subaequantes vel calyce paulo breviores vel raro calyce multo breviores, semper latae factae, (6—) 9-15 mm latae, 10-40 mm longae. Flores grandes 6-8 (-9.5) cm. Tubes stamineus elongatus, 3.5-5 cm Bae antheris grandibus, 2-3 mm longis. Legumen longistipitatum (stipes 2.5-4 cm lanum, coflaceum, 1.5-2.5 cm latum, pubescentia appressa. Semina brunnea, incrassata, longitudine latitudine longiore OLOTYPIC SPECIES: C. flexuosa Fantz? (HOLOTYPE: Peru, Tarapoto, Feb. 1856, Spruce 4527, K-Hb. Bentham). ee of section Flexwosa have been collected from forest refugia in u and Eucador. The section includes three species (Fantz, 1977): C. oe Fantz, C. eae Macbride, and one to be described. 3. CLITORIA L. subgenus BRACTEARIA (Mart. ex Benth.) Fantz_ sect. Brachycalyx Fantz, sect. nov rbores vel frutices alti, raro frutex lignosus saepius volubilis. Folia decidua, typice tempore florendi, superne ki uberula, rarius glabra. Stipulae deciduae vel caducae, lanceolatae, acutae, angustae, 1-2 mm latae. Inflorescentia nodoso-racemosa, axillaris, plerumque esa pay oe nye denudato visa praecox. Calycis tubus brevis, Le Dee cupulatis | tubularis), 7-15 mm longus, lobis late deltoideis minutis vel fere Lena 1-4 mm longis, 3-4 mm latis, latitudo longitudinem subaequalante vel ok ae ee ngiore. Bracteolae ane. 2-5 mm longae, 1.5-3 mm_ latae. Legumen pubescens, leniter vel valde depr rs inter semina tempore liberationis. ae ‘suborbiculat vel leviter longiora quam latiora, compressa, vulgo 7-13 mm diam —3 mm es data. HoLornc SPECIES: C. brachycalyx Harms (TYPE: Brazil, Rio Branco, Surumu, p de Serra de Maiacy Feb 1909, Ule 8398, B destroyed during war, photo at MO 675333. Isotype at K) Members of section Brachycalyx are collected from dry tropical forests or occasionally from open grassy areas of savannas at altitudes up to 900 m, * CLITORIA flexuosa Fantz, sp. A Clitoria aeecaeieee Macbride affinis a qua ee is et bracteis eset thas: eee longo-st ipitatis, tubo staminale breviore, oe revi-un rade s, et foliis plerumque infra subpilosis distinquitur. HOLo- cas : Peru, Tarapoto. pee volubil robust, flores rosei, Feb 1856, Spruce 4527 (K- Hb ae Payee mae YPES: Hb. Hooker ne ee PARATYPES: Tarapoto, Spruc (K San Martin, aay Jui, Rio eee ca. 400 m, Se 1934, oe eee coer 766344, GH, MO Tse & 1105521). 94 in northwestern South America and adjacent Panama. Isolated collections of one species have been made as far north as Chiapas, Mexico. The section includes seven species (Fantz, 1977): C. brachycalyx Harms, C. brachystegia Benth., C. dendrina Pittier, C. glaberrima Pittier, and three to be described. 4. CLITORIA L. subgenus BRACTEARIA (Mart. ex Benth.) Fantz_ sect. auliflorae Fantz, sect. nov Frutices lignosi saepius volubtilis vel infrequenter frutices erecti. Inflorescentia vulgo cauliflora et axillaris, racemosa, plus minusve nodosa. Bracteolae amplitudine variae, | sed non grandae et calycem occultantes, typice calycis tubo breviores et angustae, 1-3 (raro —6) mm latae. Calycis deed ne CONS picuos ferens, typice 4-13 (—27) mm longos. Legumen planum vel leniter depressum inter semina, pubescentia typice microuncinata cum ee ee dispersis, effusis vel eae interdum confertis cum trichomatibus uncinatis — iene vulgo suborbiculata, raro oblonga, 3-5 mm incrassata, interdum com a. OLOTYPIC SPECIES: C. bees "Fante! (HOLOTYPE: French Guiana, Karouany, 1857, Sagot 120, K-Hb. Bentham) Members of section Cawliflorae are found mostly in northern South America, one extending into Central America. The section includes twelve species (Fantz, 1977): C. arborescens R. Brown in Ait., C. coriacea Schery, C. javitensis (H.B.K.) Benth., C. leptostachya Benth., C. obidensis Huber, C. sagotii Fantz, C. selloi Benth. and five to be descri é Acknowledgement is made to Helen Correll for the Tata descriptions, and to Daniel B. Ward, Dana Griffin III, and Donovan Correll for reviewing the manuscript aa providing constructive criticisms. REFERENCES BAKER, J. G. 1879. es no. 76: Clitoria. In Hooker, Fl. Br. India, 208-209. L. Reeve & ae ndon. BENTHAM, G. 337, i de Leguminosarum Generibus. Ann. Wien. Mus. ee 2: oe 839. Pameration of Pane eee by Mr. Schomburgk, British Gui- ana. re Wien. Mus. Nat : 4 < er Synopsis - oe genus Cite Journ. Linn. Soc. 2: 33-44. FANT 1977. A monograph of the genus Clitoria (Leguminosae: Glycineae). 08. bp Ph.D. Thesis, University of Florida. PRANCE, G. T. 1973. a, ae ae meek for a seg of Pleistocene forest bas refuges. in the Amazon ed on evidence from the distribution pattern in Caryocaraceae, Chrysobalanacea Sener ce Lecythidaceae. ACTA Ama- zonica 3 (3): 5- 3 CLITORIA seein Fantz, sp. nov. A Clitoria javitensts Cee eae _ a qua calycis, leguminis, et ova rii trichomatibus prominentibus uncin pagina supra folit medionervum pipers um ferenti, eee subsessile, vexillo brevis Serene stylo ovario multo longiore et fructibus parvioribus, semina incras distinguitur. HOLOTYPE: Fr. Guiana, Karo ieee 1857, aes 120 (K-Hb. ee photo at S). IsOTYPES: GH,NY,S,U 37632 A,W NEW SPECIES OF CLITORIA SUBGENUS BRACTEARIA SECTION BRACTEARIA (LEGUMINOSAE) FROM PERU AND ECUADOR’ PAUL R. FANTZ Fairchild Tropical Garden, Miami, Florida 33156 Members of Clitoria L. subgenus Bractearia (Mart. ex Benth.) Fantz sec- tion Bractearia are distinguished by the large coriaceous bracteoles which subequal and obscure the sparsely pubescent to glabrate calyx, (sub-) paniculate inflorescences in which the pair of pedicels are borne at the apex of the primary lateral branches, broad thick-sutured fruits, and by the tall, erect, woody habit (trees or shrubs commonly 5-30 m tall). Section Bractearia includes seven species (Fantz, 1977): C. arborea Hoffm. ex Benth., C. fairchildiana Howard, C. amazonum Mart. ex Benth., C. nervosa Herz., and three newly described below. 1. CLITORIA juninensis Fantz, sp. nov. Frutex 2-4 m altus. Folia Fea subtus subvelutina; foliola obtusa, acumine brevi, 1-7 mm longo. Stipulae deciduae, 8-11 mm longae; stipellae 5-10 mm longae. inlorenennee subpaniculato-racemosae, axillares et terminales, multiflorae; rami laterales primarii pedicelliferi subsessiles ad 5 (-8) mm longi. Bracteolae amplae, coriaceae, oblongo- bee calycem subaequantes, 14-18 (—20) mm longae. Flores 4-5.5 cm; vexillum extus fluvum, sericeum. Stylus 16-17 mm _ longae. Legumen sae planum, aces sutura incrassata. Semina fere orbiculata, anthracina, mpressa, laevis. Shrub, erect, 2—4 m tall. Branches elongated (vide Killip & Smith 23380), hollow, to 10 mm thick, juvenile branches subangular with dense, spreading, rufus trichomes, branches becoming terete and seemingly glabrate, with uncinate trichomes becoming more conspicuous (observe at 30X); bark dark brown, splitting in longitudinal strips, lightened beneath; axillary buds 1-2 mm, scales glabrate. Leaves 3-foliolate, coriaceous, leaflets broadly elliptic, ovate-elliptic, or obovate, apex generally obtuse, rapidly narrowed to a short acumen of 1-7 mm, more or less mucronate, base broadly cuneate to rotund, midrib impressed above, glabrous or bearing short erect and/or uncinate trichomes, primary nerves of 10-16 (—18) pairs, upper surface dark green and glabrous, lower surface pale green and subvelutinous, lamina 8-22 cm long, 4-9 (-12.5) cm wide. Petioles subangular-terete, 4-11 cm ribbed, with moderately dense, spreading, short, rufus trichomes, rachis more com- ecuiices studies on Clitoria accomplished at the IFAS Herbarium. Dept. of Botany, University of Florida, Gainesville, Florida. Florida Agricultural F penmient Station Journal Series No. 1345. SIDA 8(1): 95-101. 1979. 96 pressed, (1.5—) 2-3.5 cm; petiolule subquadrangular, 6-10 mm, with dense, spreading, rufus trichomes. Stipules deciduous, lanceolate-deltoid, acute, 8-11 mm long, 2-3 mm wide; stipels more persistent, linear, acute, 5-10 mm long, 0.2-0.7 mm wide. Inflorescence subpaniculate-racemose, axillary and terminal, solitary, multiflowered, pubescence dense, trichomes short, spreading, rufus; central axis 5—23 cm long, lignose and nodose in fruiting stage, the primary Figure 1. Holotype of Clitoria juninensis (Killip & Smith 23380, NY). 97 lateral branches which bear the pedicels subsessile to ‘ (-8) mm _ long; pedicels paired, 4-7 mm. Bracts ovate, acute, reflexed in age, 4-7 mm long, 2—3 mm wide, pubescence dense, trichomes appressed, eee ee large, coriaceous, oblong-elliptic, Oi bequal to calyx, 14-18 (- mm long, 7-10 mm wide, pubescence appressed. Flowers all chasmogamous, 4-5.5 cm, white to pink with a bluish-tinge. Calyx sparsely pubescent, trichomes appressed, tube 13-17 mm long, 3-5 mm wide at the base expanding to 8-12 mm wide at the throat, lobes ovate-deltoid, acuminate, 5-7 mm long, 3-4 mm wide, ventral lobe narrow, 7-8 mm long. Vexillum pubescence dense, trichomes appressed, tawny, blade 3-3.5 cm wide, claw 8-9 mm. Alae white to pinkish, extending 6-8 mm beyond the carina, blade 18-23 mm long, 5-9 mm wide, claw 13-15 mm. Carina bluish, falcate, blade 9-13 mm long, 4 mm wide, claw 24-27 mm. Staminal tube glabrous, 27-32 mm long, free filaments 3-5 mm long; anthers 1.5-1.7 mm long, 0.5-0.8 mm wide. Gynophore 3 mm; ovary ca. 14 mm long, 1.2 mm wide, pubescence dense, trichomes appressed, yellowish-white; style bearded, 16-17 mm long, geniculate 6-7 mm; stigma flattened, 1 mm diam. Legwme stipitate, slightly exerted beyond oe ae ish-brown, flat, coriaceous, glabrous, thick-sutured, 21 cm long, 2.5-3.2 cm wide; stipe 14-17 mm; beak to 4 mm; dehiscence causing valves to twist one-half of a turn. Seeds nearly orbicular, black, smooth, compressed, 1-2 mm thick, ca. 10 mm long, 9 mm wide; 7 seeds per pod. Fig. 1 TYPE COLLECTIONS: Pert. Shrub, 8—12 ft., with ae branches; standard and wings white, blue-tinged, keel blue, white ee La Merced, Dept. Junin, 700 m, 29 May-4 Jun 1929 Killip & Smith 23380 (HOLOTYPE: NY. ISOTYPES: F 616214, US 1358082). PARATYPES: Clearing along Rio ae a neat “Hacienda 3”, Ciloata Perene, 600 m, Killip & Smith 25136 (NY); Pampa Whaley, Cpa Perené, prov. Tarma, 700-750 m, 14 Feb 1959, Ferreyra 13555 (WIS-2 sheets). Clitoria juninensis has close affinities with Clitoria andrei which is dis- tinguished by the elongated acumens of the leaflets, shorter stipules and tipels, somewhat smaller flowers with larger bracteoles, a nonsericeous vexillum, and by the tree habit. All known collections of C. jaminensis occur from the Dept. of Junin, Peru, where the species is an apparent endemic of highland scrub communities at elevations of 600-2000 m. 2. CLITORIA andrei Fantz, sp. nov. ad 20 m alta. Folia — subtus subvelutina, foliolis abrupte acuminatis, l bre acumine eA, 1-2 cm longo. Stipulae caducae, 4 mm longae; stipellae —5 mm longo. Inflorescentiae elongatae, subpaniculato-racemosae, axillares et termi- nales, multiflorae; rami primarii latecales pedicelliferi a ad 2 mm longi. Bracteolae amplae, coriaceae, oblongae, obtusae, calycem subaequantes, 22-25 mm longae. Flores 4-4.5 cm; vextllum extus fuluum, cum PO Ne Secus nervos. Stylus 17-21 mm. Legumen ignotum. Tree to 20 m tall. Branches subquadrangular becoming terete, to 1 mm diam., hollow, juvenile branches with short appressed to spreading trichomes, becoming glabrate; bark light brown; axillary buds 15-2 mm, bud scales 98 ovate, acute, concave, strigose. Leaves 3-foliolate, coriaceous, leaflets ovate- lanceolate to elliptic lanceolate, apex abruptly acuminate, acumen 1-2 cm, mucronate, base broadly cuneate to rotund, midrib impressed to subimpressed above, often bearing short erect trichomes, primary nerves 13-17 pairs, upper surface dark green, glabrous, lower surface pale green, subvelutinous, lamina 7-20 cm long, 3-8 cm wide. Petiole subangular-terete to compressed Figure 2. Holotype of Clitoria andrei (Andre 4675, GH). oD above, 4-7 cm, pubescence of short, spreading trichomes; rachis subquad- rangular, 1-2.5 cm; petiolules quadrangular, 6-7 mm, trichomes. falcate. Stipules caducous, deltoid, acute, 4 mm long, 3 mm wide; stipels deciduous, linear, 2-5 mm long. Inflorescence elongate, subpaniculate-racemose, axillary and terminal, solitary, multiflowered, pubescence of axes dense, trichomes short, erect, rufus on juvenile portions becoming less dense and whitened with age; central axis 7-20 cm long, the primary branches which bear the pedicels subsessile to 2 mm long; pedicels paired, 6-8 mm. Bracts ovate, acute, reflexed in age, 4-G mm long, 1.5-2 mm wide, pubescence dense, trichomes rufus. Bracteoles large, coriaceous, oblong, obtuse, subequal to calyx, 22-25 mm long, 9-12 mm wide, pubescence strigose. Flowers resupinate, all chasmogamous, pink with bluish tinge, 4-4.5 cm. Calyx pubescence sparse, trichomes short, appressed, tube infundibular, 14-16 mm long, 4-5 mm wide at the base expanding to 9-11 mm wide at throat, lobes deltoid, 5-7 mm long, 2-3 mm wide, ventral lobe 6-8 mm long. Vexillum indumentum of short, appressed, tawny trichomes confined primarily along the nerves, blade 2.5-3 cm wide, claw ca. 12 mm. Alae extended 6-8 mm beyond the carina, blade ca. 16-18 mm long, 4-6 mm wide, claw ca. 15-17 mm. Carina falcate, blade ca. 9 mm long, 3-4 mm wide, claw ca. 24-27 mm. Staminal tube cue ae near apex, trichomes microscopic (observe 30X ), uncinate, tube mm long, free filaments 2-3 mm long, uncinate-trichomed; eee ee mm long, ca. 0.4 mm wide, connective apiculate. Gynophore 3 mm, ovary ca. 14 mm long, 1.4 mm wide, pubescence dense, trichomes yellowish-white; style 17-21 mm, bearded, geniculate 6-7 mm; stigma discoid, 0.5 mm diam. Legwme unknown. Fig. 2. — TYPE COLLECTIONS: Ec cuador. Inter Guayaquil and ee ce 60 m, July 1876, Andre 4675 (HOLOTYPE: GH. IsOTYPES: F 537681, Clitoria andrei has close affinities with Clitorta juninensis of Peru which is distinguished by the obtuse leaflets with a minute acumen, larger stipules and stipels which are more persistent, smaller bracteoles on somewhat larger flowers, a sericeous vexillum, and by the shrubby habit. The species is knowa only from its type locality in Ecuador. 3. CLITORIA moyobambensis Fantz, sp. nov. Arbor. ee trifoliolata svbtus glabrescentia, foliolis acuminatis, acumine elongato, 1.5-—3 cm lon Stipulae amplae, concavae, 10-12 mm longae, 4-6 mm latae; stipellae caducae. ee centiae axillares et terminales, subpaniculato-racemosae, multiflorae; rami laterales primariit pedicelliferi ae ae ad 2 mm longi. Bracteae coriaceae, lanceolatae, breviacuminatae, 8—/1 mm longae, 3—5 mm latae. Bracteolae amplae, coriaceae, oblongo-ellipticae, ee subaequantes, obtusae vel abrupte api- culatae, 12-15 mm longae. Flores 4—5 cm; vexillum extus fulvo-sericeum. Stylus brevis, 14 mm longo. Legumen ignotum Tree. Branches hollow, 4-7 mm diam., juvenile branches subquadrangular, longitudinally striated and sulcate, grooves shallow to deep, pubescence 100 prominent, trichomes uncinate (observe 30X) and short, falcate to appressed (conspicuous at 10X), branches becoming subterete and glabrous with age; bark medium gray, splitting into longitudinal strips, whitish-tan beneath. Leaves 3-foliolate, coriaceous, leaflets elliptic-lanceolate, tapering to acuminate apex, acumen 1.5—3 cm long, to 1.2 cm wide at broadly cuneate base, mid- Figure 3. Holotype of Clitoria moyobambensis (Mathews s.n., K-Hb. Hooker). 101 rib impressed above bearing uncinate trichomes, primary nerves 13-16 pairs, upper surface dark green, glabrous or with short trichomes along major nerves, lower surface light green, glabrous, lamina 12-15.5 cm long, 6-8 cm wide. Petioles conspicuously twisted at base, longitudinally striated and canaliculate, 4-5 cm long, pubescence scattered, trichomes minute, falcate to appressed; rachis similar, 4 cm; petiolules quadrangular, rugose, 8-9 mm. Stipules large, conspicuous, concave nearly halfway around stem, lanceolate- ovate, rapidly tapering above middle to acute apex, 10-12 mm long, 4-6 mm wide; stipels caducous, only one terminal stipel observed (2 mm x 0.4 mm). Inflorescence axillary and terminal, solitary, subpaniculate-racemose, multi- owered, trichomes subappressed; central axis twisting, longitudinally striated and caniculate, to 4 cm long (juvenile? ), primary lateral branches which bear the pedicels subsessile, 1 (—2) mm long; pedicels paired, 8-14 mm. Bracts large, conspicuous, lanceolate, short-acuminate, 8-11 mm long, 3-5 mm wide, trichomes appressed. Bracteoles large, coriaceous, multinerved, oblong-elliptic, obtuse or abruptly apiculate, subequal to calyx, 12-15 mm long, 9-10 mm wide, pubescence ciliolate and strigose. Flowers resupinate, all chasmogamous, 4-5 cm. Calyx infundibular, glabrate, trichomes widely scattered, heii tube a ae eae multinerved, nerves eel 13-15 mm long, 4-6 mm wide at the base expanding to 8-10 mm wide at throat, lobes deltoid, acute, 5-6 mm long, 4 mm wide, ventral lobe 7 mm long. 2 mm wide. Vexillum tawny-sericeous, blade 3-4 cm wide, claw 11 mm. Alae and carina hidden by vexillum and/or damage, not clearly observed. Staminal tube glabrous, 27-28 mm long, free filaments 1-3 mm _ long; anthers 0.8 mm long, 0.5 mm wide. Gynophore ca. 5 mm; ovary 13 mm long, 1 mm wide, lateral surfaces densely pubescent, trichomes uncinate, sutural edges with 1 mm long, ascending trichomes; style 14 mm _ long, bearded, geniculate 6 mm; stigma dark, ca. 0.7 mm diam. with a ring of short, reflexed trichomes at its base. Legume unknown. Fig. 3. TYPE COLLECTION: Peru. Prov. Moyobamba, 1838. Matthews s.n. (HOLOTYPE: K-Hb. Hooker). Clitoria moyobambensis is easily distinguished from other species in the subgenus by the unusually large bracts and stipules, and the pubescence of the ovary, a type unique amongst those species with a tree habit. This species appears to have close affinities with Clitoria juninensis and Clitoria andrei which have similar floral structures. It is known only from the type locality. Acknowledgement is made to Helen Correll for the Latin descriptions, and to Daniel B. Ward, Dana Griffin II], and Donovan Correll for reviewing the manuscript and providing constructive criticisms. REFERENCES FANTZ, . R. 1977. A monograph of the genus Clitoria (Leguminosae: Glycineae). 1066 pp. Ph. D. Thesis, University of Florida. FLORISTIC AND GEOLOGIC ASPECTS OF INDIAN MIDDENS IN SALT MARSHES OF HANCOCK COUNTY, MISSISSIPPI LIONEL N. ELEUTERIUS Botany Section and ERVIN G. OTVOS, JR. Geology Section Gulf Coast Research Laboratory Ocean Springs, Mississippi 39504 ABSTRACT Sixty-two species of flowering plants were collected from Indian middens (shell mounds) located in the salt marshes of Hancock Co., Miss. The diversity of the flora depended on the height of the midden above the marsh surface and increased with increased elevation. The largest midden (Cedar Island), with an elevation of mS ' greatest diversity of species (47). Radiocarbon dating indicates that clam shells (Rangia cnneata) found on the surface of Cedar Island are over 2,600 years old. The geological history surrounding the establishment of the middens is discussed. Indicator plants for these Indian occupation sites 0.5 m or more above MLW are the calciphiles ail silicola, Erythrina berbacea, and Aesculus pav - These shell middens provide the only known habitat of Sageretia minutiflora, the first report of its occurrence in Misicibal INTRODUCTION Salt marshes were apparently utilized by prehistoric humans for a con- siderable length of time. Human occupation sites in the coastal areas of Louisiana were established as early as 12,000 B.P. (Gagliano, 1963). Pres- ently known cultural sites on the Mississippi Coast date back only to the last four millenia. The fact that these early people gathered shellfish as a food item from the estuarine waters and salt marshes is evident by the presence of numerous and extensive shell mounds, composed of the clam Rangia cuneata. Such shell deposits are “kitchen middens” (refuse heaps ) at periodic Indian occupation sites, located close to the source of food. These mounds are relatively enduring features in the changing coastal environment and represent the only remaining local sites of floristic disturbance by pre- historic man, not obscured by natural processes or European man. Over the past several years we have located five previously unreported and undisturbed shell middens in the salt marshes of Hancock County, Mississippi. Although Brown (1936) described the vegetation of 16 mounds and middens in Plaquemines and St. Bernard Parishes (delta areas of the Mississippi River ), SIDA 8(1): 102-112. 1979, 103 Louisiana, on the shell middens we have visited, we noted an array of plan species previously unreported in Louisiana or Mississippi. The Enea conditions that relate to the establishment of these middens are essential for the understanding of the flora’s development. The present paper is the result of a combined effort to study these middens floristically and geo- logically. he middens can only be reached by boat or pontooned helicopter. This isolation probably contributed to their preservation. A study of aerial photo- graphs indicated the location of suspected shell middens and was followed by a general survey of the salt marsh area where five middens in the tidal marsh east of the Pearl River and south of Cadet Bayou were selected for study (Fig. 1). Floristic surveys were conducted seasonally in 4, an periodically in 1975, 1976, and 1977. Geological information was taken from an ongoing, detailed geological study of the Mississippi coast includ- ing core drillings in the marshland and radiocarbon dating of the middens. GEOLOGICAL FRAMEWORK The geological evolution of southern Hancock County (Fig. 1) began after a long period during the Wisconsin glacial stage and the early Holo- cene epoch when the subject area was dry land. Transgressing mid-Holocene sea waters reached it at about 6,300-6,000 years ago. By about 4,000—3,500 years ago, all of the present marsh area was inundated and a low dune ridge (Magnolia Ridge) formed along this new shoreline. With the first arrival of sea waters, silty-muddy and marsh deposits characterized by Ammotium- Haplophragmoides-Miliammina agglutinate foraminifers, formed in the low salinity nearshore waters influenced by fresh waters from the Pearl River. Oyster reefs developed in more saline waters further seaward, while the central and southern-southeastern parts of the present marsh area were cov- ered by the highest salinity waters. Open-Gulf nearshore-type, varied and rich foraminifer fauna (Elphidinm- Be ae Nonion-Ammonia) thrived Fi . The on the bottom sediments there (Fig. 2 igh energy conditions near the shore resuited in the development of ay shoals and in the subsequent emergence of low (1.5-4.0 m), narrow, elongated barrier islands: Point Clear and Campbell Islands. The sand for these developing islands was derived from the east along the Mississippi barrier island-shoal chain (via Cat Island—Square Handker- chief Shoal) and the Hancock barrier islands themselves become parts of another 50 km long barrier-shoal trend (Bayou Sauvage—Hancock County trend, Otvos 1973, Fig. 16) which extended deeply into the present New Orleans area. Growth of the large Mississippi River subdeltas (St. Bernard Parish) prevented further seaward progradation of Point Clear and Campbell Islands. This delta development south-southeast of the Hancock area event- ually cut off the sand supply. It severely restricted open marine water influx and also reduced wave energies. The decreasing salinities south of the two ‘}X9] UT passndstp “suoneso] Udpprul UvIpUy O} JaJaI C-] sJoqUINNY ‘puryysseur AWUNOD YoourpYY YyINog "] sNITy miles oO 1 2 ee ie) 1 2 3 kilometers 4 ? yi ee | wa ALIS =f Ss ue git oe 89°30’ x , NG, ¢ HERON BAY s ; y Aft ae. Yh, ae Ja mn - ’ Cedar Island 2 : * ape 89°27'30" | ei . \ ers MISS. > New AA Orleans~) St Joseph Point AREA OF STUDY (ZZ SAND RIDGES [|] MARSH (inland ps) also swam vO! 105 islands are reflected by the less varied Asmmonia-Elphidium-Nonion- Nonionella fauna in the muddy, sandy-muddy nearshore deposits of the period. Immediately south of Campbell and Point Clear Islands, sediments, presumably mostly from the Pearl River, built the bottoms up to intertidal levels and marsh vegetation was established, protected by the Mississippi River subdeltas. At its peak of development, this marshland extended con- siderably beyond the present southern and eastern marsh shores. At this stage, Indian refuse middens started to dot the banks of creeks in the newly established marshland. Shells of a very low salinity bivalve, the edible Rangia cuneata almost exclusively form the mounds. This is especially so, at the largest midden, “Cedar Lsland”. Some of the other middens also contained lesser quantities of oyster shells (Crassostrea vir- ginica) and the snail (Littorina irrorata). “Cedar Island”, with a surface area of 50 by 150 m (Fig. 1, Site 1), extends about 1.0 m above mean sea level and, according to borings, to about 3.5 m below sea level. Most of the mounds were formed during the Tchefuncte and later cultural periods, characterized by pottery production (Table 1). Each mound represents long periods of intermittent occupation and shell accumulation during seasons of food gathering activities. Subsidence, caused mostly by the compaction of the underlying muddy Holocene sediments, clearly affects the mounds. This is shown by occasional, partially drowned oak trees and red cedars in the marsh along mound peripheries. Due to subsidence, some middens may have been completely buried under surface deposits. Marsh accretion ended between 2,000-1,700 years ago. Termination of the active flow through the protective Mississippi-St. Bernard subdeltas started their disintegration. Renewed coastal erosion increased the adjoining western Mississippt Sound water area and the strength of erosive wave energy on the shores. This initiated the present steady reduction of the south Hancock marshland area. Table 1. Radiocarbon dates of Rangia cuneata from Indian shell mounds, Hancock County salt marshes No. of Date Site (Fig. 1) Dating laboratory Analysis (yrs. B.P.) 1 (Cedar University of Georgia, UGa-353 2650+70 Island ) Geochronology Lab. UGa-354 2900+70 2 Krueger Enterprises GX2653 1295+90 3 University of Georgia, UGa-371 2680475 Geochronology La 4 University of Georgia, UGa-373 2035+65 Geochronology Lab. 5 Krueger Enterprises GX25671 1190+110 5m Ikm Imile Magnolia Bayou PointClear Cedar Three Oaks St. Joseph Ridge Caddy Island Island Bayou Point | Sm{ sea level Pleistocene land surface my au 901 107 BOTANICAL ASPECTS COMPOSITION AND DISTRIBUTION. The vegetation of the Indian middens contrasts conspicuously with the low profile, homogeneous, and rather monotonous vegetation of the sur- rounding, extensive salt marshes. A list of plants found on the middens is shown in Table 2. The marshes are composed primarily of Juncus roemerta- nus, Spartina alterniflora, and Distichlis spicata (Eleuterius, 1972). The most abundant trees on the middens are the live oak (Quercus nigra) and the southern cedar (Juniperus silicola). The prevalence of cedar on the largest midden studied is apparently the basis for its name (Cedar Island; USGS Quadrangle, Grand Island Pass, Miss.-La., 1956, photorevised, 1970). Here 16 oaks with diameters at breast height (DBH) ranging from 6-62 cm and about 50 cedars, the largest with a 45 cm DBH. Other common trees found were Celtis laevigata, Diospyros virginiana, Morus rubra and Zanthoxylum clava-herculis, e understory of site 1, “Cedar Island” is composed of a mixture of shrubs, primarily Ywcca doi folia and herbs. This midden, the most diverse floristically, contained a dense vegetational cover forming an almost 1m- penetrable mass. Other shrubs of frequent occurrence on the top of the mid- dens were: Ilex vomuitoria, Sabal palmetto, Sambucus canadensis, Hypericum hypericoides, Rhus copallina, Erythrina herbacea, Aesculus pavia, and Sageretia minutiflora. There is no obvious zonation of plants on the midden and the general impression is one of over-crowding. Numerous trees and shrubs lean outward from the mound, this “overhang” shading much adjacent marsh. Many of the shrubs are quite tall and often reach about 6 m. One large oak blown down by winds of Hurricane Camille in the east-central island area in 1969 left a large opening that subsequently was invaded by Yucca alozfolia. The south- eastern portion of Cedar Island presently is vegetated primarily by small trees, shrubs, and herbs. Stumps indicate that larger trees grew here in the past. The prevailing southeasterly winds apparently have an effect on the vegetation here, sculpturing and damaging the vegetation, as on the eastern ends of the offshore barrier islands (Eleuterius, 1975; Penfound and O'Neil, 1934; Lloyd and Tracey, 1901; and Miller and Jones, 1967). The periphery of the midden contained shrubs that are common on natural levees along the bayous, upland edges of salt marshes and dredged soil. There were Baccharis halimifolia, Iva frutescens, Borrichia frutescens, and Myrica cerifera. Figure 2. Geological cross section through Holocene deposits in south ock County ypacshland Symbols: 1-clay, mud, sandy mud with Hiphidiem. neds fauna (h ini deposits); 2-clay, mu baer) clay ae mud with Ammonia- Elpbidium. Nonion fauna Cae alnty yi uddy Rae vay mud, mud with Ammotium Ga ae -salt marsh deposits; 5—dune and barrier ridge es va ae ei eee (shell mound). 108 Table 2. Plant species found on Indian middens in Hancock County marshes. Indicator species are those found on shell middens and not in the marshes. They are marked with an asterisk. Scientific Names TREES Celtis laevigata Willd. s L, *Juniperus silicola (Small) Bailey Morus rubra L. rcus virginiana Miller bai olan clava-herculis L, SHRUBS AND WOODY VINES * Aesculus pavia L. *Ampelopsis arborea (L.) Koehne Baccharis halimifolia UL. Borrichta frutescens DC. *“Bumelia lanuginosa (Michaux) Campsis radicans (L.) Seemann ] ! Des Moulins DC. (Pers. ) Pypevcun hvpevicoides Ilex vomitoria Aiton 1 frutescens L (L.) Crantz li *Matelea caroliniensis (Chapman) Woodson Myrica cerifera . Rhus copallina Sabal palmetto a ex Schultes * Sageretia Lishee ih (Michaux ) Sambucus canadensi Serenoa repens (Bartram) Small Sane bona-nox I. Wusterta Daca (L.) Poiret * ie dloifolia L. Trel. HERBS Amaranthus spinosus L. Ambrosia artemistifolia U. Chaerophyllum tainturieri Hooker Cuscuta pentagona Engelm philadelphicus L Eryngium yuccifolium Michaux Eupatorinm — Michaux Euphorbia maculat Ipomoea sagittata Caw Lepidium virginicum UL. Manisuris rugosa (Nucall) Kuntze Common Names Hackberry Persimmon Honey Locust Southern Red Cedar Red Mulberry Live Oa Hercules-Club, Prickly Ash Red aoe Pepper-Vin Groundsel- Tree Sea Ox-eye False a Chittumwood Trum e Marine ivy, Weahe vine Coralbeads Coral Bean St. Andrew's Cross sim Pod Milkweed, Climbing r Vining Milkweed Wax Myrt ‘le Dwarf or Winged Sumac Cabbage Palmetto Ww a Spanish Bayonet, Yucca Thorny Amaranth Ragweed Thoroughwort Forb Morning Glory Poor-mans Pepper Joint Grass Opuntia compressa (Salisbury) Macbride Passiflora lutea UL. Phragmites communis as Physalis angu Bee Phytolacca rigida Sm Rhus radicans L Samolus parv Bees Raf. 109 Prickly Pear Prickly P: Wood Sorrel, Sheep Sorrel Bitter Panicum, Beach Grass Common Rec Roseo Cane Ground Che Poke, Pokebery, Pokeweed Poison I Water Pimpernel Scutellaria ovata Hill Skullcap Solidago sempervirens L. Goldenrod Hog Cane Clumped Cord Grass Venus Looking-Glass (44 Stellaria media eee Chickweed Tillandsia usneoide Spanish Moss Verbesina ee (L. ) Walter Crown Beard Verbesina virginica L Coffee Bea Vicia ludoviciana Nute, Vetch, Deer Pea There were also many woody vines on the middens such as: Cissus mecisa, Wisteria frutescens, Ampelopsis arborea, Smilax bona-nox Cocculus carolinus, Campsis radicans, and Rhus radicans, ELEVATION-TIDE RELATIONS. All the other middens are of relatively low profile, and less than one meter in elevation. The diversity of the herbaceous flora is reduced (Fig. 3) with reduced elevation. Those middens only slightly above the surface of the marsh (0.1 m above MLW ) had fewer species than those on more elevated areas. No tree species were found below 0.5 m elevation (above MLW). Locally tidal amplitude is about 0.3-0.5 m, thus tidal inundation is probably responsible for the reduction of species. PLANT-SOIL RELATIONSHIPS. Although the soil on the middens is generally thin, a very black soil layer 4-8 cm thick is often found and is apparently nutritionally rich, judging from the dense vegetation supported. Roots cia ently rework soil and shell; shell and soil were found mixed at 8-16 c depth. Although high phosphorous content of soils is characteristic of up- land Indian habitation sites (Diety, 1957; Zeiner, 1946), available phos- phorous was generally low, probably due to the relatively high pH of the soil (Hole and Heizer, 1965 RARE SPECIES. Sageretia minutiflora. Mohr (1901) indicated the distribu- tion of Sageretia minutiflora as extending from the Carolinas to Florida and west to Mississippi. Recent floristic surveys have failed to locate this species on the coastal plain of Mississippi and Alabama. Radford et al. (1968) 110 state that it occurs in only two coastal counties of South Carolina and in one in Georgia. Long and Lakela (1971) reported it from Lee County, Florida and Clark (1971) from Mobile Co., Alabama. Specimens were taken from a large shell midden on Dauphin Island. Michael Lelong (personal commu- Sag collected the species from clam shell middens on Little Dauphin sland, Alabama. Shell middens apparently provide the only habitat for this oe and this is the first report of its presence in Mississippi. CULTURED RELATIONSHIPS. Several of the plant species apparently had cul- tural relationships with prehistoric man since identifiable plant parts have been found in burial mounds (personal communication, Dale Greenwell). Seeds of plants of economic value may have been purposefully introduced at the semi-permanent occupation sites. Cedars may have been propagated because of the fine tinder produced from the bark and used in fire making. Coral bean seeds were probably a source of pigment for paint and used as “bead” ornaments. The fruits of the buckeye are historically known (at least locally, from pioneer days) to make a powerful fish poison. Manisurus rugosa consistently found on the more elevated sites is cited by Mangledorf (1974) as an ancestor of modern corn Zea mays. There is no known basis for extend- 50 e Ww = Oo ud w 25 " @ o ac Wi va) = > i e th hh ht t+ + 0.! 0.5 1.0 1.5 ELEVATION (Meters above MLW) Figure 3. Species diversity in ce to ae elevation of Indian clam shell grat in hes marshes of Hancock Co., Miss. The number of plant species increased with increase in elevation. (MLW = mean on water 111 ing such speculation to the other species. Chemical work (pharmaceutical and taxonomic) on these species, might be highly rewarding. CALCIPHILES AS INDICATOR SPECIES. Brown (1938) stated that many species found on Indian mounds and middens that he examined in south Louisiana may be common on alluvial soil elsewhere in that state. He thought that the mound vegetation may have represented a relict. The investigated Han- cock marsh mound vegetation, however, can not be relict, since most of the plant species, especially those considered indicators, are calciphiles Their presence is apparently favored and determined by the large amount of calcium contained in the clam shells. These calciphiles, paradoxically, have not been found on recent deposits of oyster shells, left by European man during the past three centuries. Indicator species (Table 2) are those species that represent consistent and reliable indicators of the shell deposits on the studied middens. This does not preclude their occurrence elsewhere in other habitats. Several of the indicator species are more frequently found on Indian occupation sites than others. They are also found on occupational burial sites (mounds) on the adjacent terrestrial mainland and may also serve as indicator species for both marsh and terrestrial occupation sites (Eleuterius, unpublished data). These are Juniperus silicola, Aesculus pavia, Erythrina herbacea, and Morus rubra. The presence of shell midden within the salt marsh ecosystem provides unique habitats for an unusual flora which increases the diversity of our estuaries. The effects of these specialized, isolated plant communities on the surrounding marsh is of considerable interest for further studies. ACKNOWLEDGEMENTS We thank Dr. Sidney McDaniel for identifying many of the plant species and his helpful comments. Dolores Smith and Darlene Jennings provided technical assistance. REFERENCES BROWN, C. 936. The vegetation of the Indian Hane ares and marshes in pdaeoia a St. Bernard Parishes. In Lower Mississippi r Delta, Reports on the Geology of Plein. and St. Bernard See hes. ree Geol. Surv Bull. No. 8: i ear 1938. The flora of ore deposits in the western Florida pees West | Feliciana Padi sh and e aton Rouge Parish, Louisiana. Jn Contribu to ae eee History - ihe Florida Parishes of Louisiana. Louisiana Geo fil Surv. . No. 12: 59- CLARK, a "OTL. Woody plants of Alabama. Ann. of Missouri Bot. Gard. Vol. 58: 9-2 242. DIETY, E. F. 1957. Phosphorus accumulation in soil of an Indian habitation site. mer. Antiquity 22: 405-409. ELEUTERIUS, L. N. 1972. The marshes of Missis ssippi. Castanea 37: 153-168. te O75 he plant life of a co Hele een nland, associated waters and bar- ne island of Mis ssissippi with Saag to cones as a natural resource. ae ide Mar. Resources of anes ae pe ‘Grane pa a 84-87. GAG ANO, S. M. 1963. A s in aoe of south Tadee and south Micdccipei. Fronds edo See 104-13 1B, HOLE, F. and R. F. HEIZER. 1965. An introduction to prehistoric archeology. Holt, tas Le oe N.Y. 497 LOY . TR L nd § RACY. 1901. a insular flora of Mississippi and Louisi- ana. “Bull Toney Boe ‘Club. 28: 68-10 LONG, R. and O. LAKELA. 1972 Manual of the flora of tropical Florida. Univ. of Miami Press. Coral Gables, Florida. 692 MANGLEDORF, P. C. 1974. meee origin, evolution and improvement. Harvard Univ. Press, ace Mass. MILLER, G. J. a JONES, ai 1967. The vascular flora of Ship Island, Missis- sippl. ee a eioe. MOHR, C. 1901. hei in Alabama. Geol. Surv. Alabama. Vol. 6: 921 pp. OTVOS, E. G, Jr., 1973. Geology of the oe eye Alabama coastal area and nearshore zone. Guido New Orleans Geol. Soc. 67 pp. PENFOUND, W. s M. E. O'NEIL. 1934. a vegetation of Cat Island, Missis- sippi. Ecology 13 RADFORD, A. E., H. EE AHLES, and C. R. BELL. 1968. oo of the vascular flora of the oo Univ. of North Carolina Press. 1183 p ZEINER, H. M. 1946. poe survey of the Angel Mock site, Evansville, Indiana. Amer. J. Bot. 33: 83-90. NOTES NOTES ON THREE SPECIES OF OPHIOGLOSSUM FROM NORTH CAROLINA—While making a study of over 4200 sheets of Ophioglossum from 114 different herbaria, the senior author became aware that the only Ophioglossum common to North Carolina was O. vulgatum VL. var. pycnostichum Fern. Specimens of this fern were examined from 32 counties. The only other species from the state was O. crotalophoroides Walter, and it was known only from a collection made on Dec. 9, 1956 on the lawn of Duke University Marine Station in Carteret County. Since other species of Ophioglossum are much more common than the number of sheets in herbaria indicate, it seemed probable that others could also occur in the Coastal Plains Province of North Carolina. A field trip was made in March, 1976 to look for these ferns in North Carolina. On March 4, 1976, Thomas collected specimens of O. nudicaule var. tenerum from Brunswick County. The locality is: Bennett's Cemetery just west of Singletree Creek, 1.5 miles west of U.S. 17 near Hickman Cross- roads (Thomas 48453). It is common at this site with O. crotalophoroides in sandy soil. No other collections of this fern from North Carolina were made Later, the same afternoon, Thomas collected O. petiolatum new to North Carolina. The locality is: Lawn of Letties Grove Pentecostal Free-Will Bap- tist Church, south of U.S. 17 east of Shalotte, Brunswick County (Thomas 48454). Thomas and Marx spent March 6-7, 1978 searching for Ophio- glossum in eastern North Carolina. We found O. petiolatum from Craven (Thomas and Marx 48476), Carteret (48489), Beauford (48502), Wash- ington (48509), and Dare (48524) counties. Although Ophioglossum crotalophoroides had been known only from one county, the authors found it to be abundant in sandy, grassy areas such as cemeteries and school lawns. We made collections from Brunswick (48452), Craven (48475), Carteret (48482), Beaufort (48506), Pamlico (48503), Martin (48506), Washington (48512), Hyde (48516), and Dare (48525) counties. We searched for O. crotalophoroides in southeasternmost Virginia with no success. We were hampered by a cold, rainy, and windy day that made crawling on the ground very miserable and seeing small plants almost a possibility. No doubt O. crotalophoroides and O. petiolatum will eventually be found in the Coastal Plain area of Virginia. R. Dale Thomas, Biology Department, Northeast Louisiana University, Monroe, LA 71209 and Paul S, Marx, Botany Department, University of North Carolina, Chapel Hill, NC, 27514. SIDA 8(1); 113, 1979. 114 NEW STATE RECORDS FOR RANUNCULACEAE IN THE SOUTH- EASTERN UNITED STATES—During continued studies of the Ranun- culaceae for the forthcoming Vascalar Flora of the Southeastern United States, the following localities constitute, so far as I know, new state records for the species indicated below. 1. RANUNCULUS FASCICULARIS Muhlenberg ex Bigelow. GEORGIA: Dade Co., moist shady deciduous forest, SW of Trenton, 12 Apr 1964, Walliams sa. (UNA!). . RANUNCULUS ‘TRILOBUS Desfontaines. TEXAS: Harrison Co., in a field off Interstate 20 at Waskom, 15 Apr 1974, MacRoberts 982 (LSUS! ) THALICTRUM DASYCARPUM Fischer & Avé-Lallemant. ALABAMA: Greene Co., common in rich low woods 1 mi E of Boligee, 9 May 1969, Thomas 2059 (UNA!) . THALICTRUM DEBILE Buckley. MIssissIPPI: Lowndes Co., common in oak-hickory-cedar ridge in the Black Belt region, 9 mi E of State College, 25 Mar 1968, McDaniel 10397 (GA!). I am indebted to the curators of the various herbaria for their loan of the specimens during this study—Carl S. Keener, 202 Buckhout Laboratory, The Pennsylvania State University, University Park, PA, 16802. CUPHEA CARTHAGENENSIS (JACQUIN) MacBRIDE—THE COR- RECT ORTHOGRAPHY .—With the publication of Radford, Ahles, & Bell’s Manual of the Vascular Flora of the Carolinas (1964) a typographical error was introduced in the treatment of Cuphea carthagenensis ('C. carthagensis’) which was repeated in Correll & Johnston’s Manual of the Vascular Plants of Texas (1970) and Correll & Correll’s Aquatic and Wetland Plants of Southeastern United States (1972). The error now appears more and more frequently on herbarium labels and in print (e.g. SID In hopes of preventing perpetuation of a misspelling I began this note only to face the question of what the correct spelling should be. The original spelling suggested the epithet should be carthagenensis, ee the original Lythrum carthagenense Jacquin (Enumeratio Pl. Carib. 22. 1760) based on plants collected at Cartagena, Columbia, $. A. As a ot place name, however, the epithet would be correctly formed by the stem carthagin- and the adjectival suffix -ensis denoting place of origin (Stearn, W., Botanical Latin, p. 218. 1966). To clarify the matter I consulted Dr. Dan Nicolson, expert in botanical orthography at the Smithsonian Institution. His clarifica- tion is one which many may find useful, particularly since today so few taxonomists are well-versed in the Latin necessary for correct formulation of new epithets. i) ee) PN ‘The problem is whether a latinization of the Spanish Cartagena should or can yield a stem cartagen- (directly Spanish), carthagin- (directly from classical Latin SIDA 8(1): 114. 1979. 115 Carthago, genitive carthaginis) or an intermediate carthagen-. Jacquin’s stem carthagen-, represents a latinization of Cartagena as Carthagena, treating it as if i were a First Declension name (genitive carthagenae, stem carthagen-). This is clearly not classical Latin which would be Carthago oa carthaginis, stem carthagin-) of Third Declension. Purists would object to such a neolatinization as Carthagena but it strikes me as a perfectly reasonable option. As such I believe carthagenensis should be maintained under the paragraph of Art. 7 (ICBN ) which states that when changes made in orthography by earlier ee who adopt geographic names are intentional latinizations, they are to be pre d. “In another case, Panicum chartaginense Swartz (Nov. Gen. Sp. Bs 22. 1788), Swartz ae made yet another neolatinization of . rtagena which he cites as Chartagena. A classical scholar would substitute carthaginense, ae under the Code I believe this would not be acceptable and that the original spelling should stand. “Obviously it is hard to draw a firm distinction between orthographic (in- cluding pee aphic) errors and intentional latinizations. One can say that carthagenense and chartaginense are errors to justify correcting both to cartha- c in cases like this. However, the Preamble to the Code tells us that, ‘Other con- siderations, such as absolute grammatical correctness, regularity or cuphony of names, more or less prevailing custom, regard for oo etc., notwithstanding their undeniable importance, are relatively accessory.’ Under this concept I favor accepting original spelling, in most cases” (Pers. comm., 1978) The correct epithet is thus Cuphea carthagenensis, not C. carthaginensis or C. carthagensis. Users of the floras cited above should correct the name in those works to include the omitted syllable—Shirley A. Graham, c/o Alan Graham, Dept. of Biological Sciences, Kent State University, Kent, OH, 44242. ZAMIA (CYCADACEAE) NEW FOR GEORGIA—This is the first re- port for the natural occurrence of Zamia in Georgia. It is reported by Small (1933) for Florida, occurring the most abundantly in the peninsular portion of the state. Lawrence (1951) states that Zama in the United States is “restricted almost exclusively to Florida, with an additional station reported in adjoining Alabama.” There appear to be no specimens to support the natural occurrence of Zamia in Alabama. Dr. Robert Kral of Wanderbilt University, who has been working intensively on the flora of Alabama for several years, says in a recent communication that to his knowledge it does not and has not occurred naturally in Alabama, but that ‘there is every likeli- hood that someone could have reported it as an escape—.” In northeast Florida Zamia is documented by herbarium specimens north into St. John’s County. The Georgia records are from Camden and Glynn Counties, with Duval and Nassau Counties, Fla., intervening. The classifica- tion and nomenclature of Zama species in Florida are in confusion. Until these problems are resolved it seems best to refer to the Georgia material as SIDA 8(1): 115. 1979, 116 Z. umbrosa Small which Small (1933) reports for “Hammocks, sand-dunes, and shell mounds, NE Fla The first collection presumed to be from Georgia was sent in for naming to the University of Georgia Harbarium in 1928 by Gertrude Proctor. Within a year or so after my arrival at the University of Georgia in 1938 I corre- sponded with Ms. Proctor in regard to the Zamia. The specimen had been given to her for naming from near Woodbine, Camden County. Ms. Proctor could not remember or find out who sent the specimen, but thought it was collected in a hammock. She and I searched considerably for the species in Camden County without success. There seemed to be a strong possibility that the specimen had been collected in Florida by someone living near Woodbine. Hence, I did not report Zamia for Georgia at that time. My beliefs are now altered by my collection of Zama from St. Simon's Island, Glynn County, where, according to Albert Fendig, Sr., a local resident, it occurs in a few scattered natural localities, but nowhere abundantly. M collection was made from one of three plants under a Quercus virginiana tree in a woods dominated by this and other evergreen species. Pinus was scattered, Vit7s common. In view of the above, Zamia should be considered native to Camden and Glynn Co., Georgia. The collections reported are: Camden Co.: {with male cone} Apr 1928, GA18054. Glynn Co.: Duncan 26359. Vigorous plant with about 20 leaves and “fruiting.” —Elev. ca. 25 ft. 17 Sep 1971, GA100484. —Wilbur H. Duncan, Department of Botany, University of Georgia, Athens, GA, 30602. REFERENCTS tae ee G. ee M. 1951. Taxonomy of vascular plants. MacMillan Co., N.Y. SMALL, J. K. 1933. Manual of the eee flora. University of North ee Press Chapel Hill 1554p. GALIUM SPECIES NEW TO THE SOUTHEASTERN UNITED STATES—While preparing a treatment of Galivm for the Flora of the Southeastern United States, | examined specimens which proved to be Galiam tricornutum Dandy and Galium palustre L. Both species are previously un- recorded for the southeastern United States. Galium tricornutum is a Eurasian species which occurs sporadically in the eastern United States and is reported from California (Munz, 1959) and western Oregon (Hitchcock, 1959). This species superficially resembles G. aparine L. Specific collection data for specimens from the southeastern United States are as follows: ARKANSAS: Miller Co.: Red River bottom, E of Texarkana, 3 May 1951, Moore 510145 (UARK). GEORGIA: Oglethorpe Co.: banks of artificial pond, just W of Dry Fork Creek between Lexington and Washing- ton, 17 May 1952, Duncan 13541 (GA). SOUTH CAROLINA: Cherokee Co.: SIDA 8(1): 116. 1979. Li? san ditch, 2.4 mi N of jet. SC Rt. #150 and US Re. #29 on SC Rt. 50 (N of Gaffney), 13 Apr 1957, Ables 22713 (NCU). Greenwood Co.: ae SC #34, 2.5 mi SW of Ninety-Six, 30 Mar 1957, Radford 20103 (NCU). Galium palustre L. is also a Eurasian species which previously has been introduced in the United States in New England, New York, Pennsylvania, Michigan, and Wisconsin (Fernald, 1950) and which closely resembles G. tinctorium L. Specific collection data for the one known specimen from the southeastern United States is as follows: WEST VIRGINIA: McDowell Co.: wet ditch, Anawalt, 17 Jun 1969, Music s.n. (WVA). Presently it is unknown whether this species is established in the area of collection or merely a waif—Cheryl A. Lawson, 960 Oakridge, Duncan, Oklahoma, 73533. — REFERENCES a M. L. 1950. Gray's manual of botany. 8th edition. American Book Co., New York. HITCHCOCK Go Ls A: ee & M. OWNBEY. 1959. ae plants of the Pacific Northwest. Part 4. Univ rsity of Washington Pres eat MUNZ, P. A. 1959. A California ae University of California Dee Bete NOTES ON MITRACARPUS HIRTUS (M. VILLOSUS) AND SCIRPUS CALIFORNICUS IN MISSISSIPPI—Mitracar pus birtus (L.) DC. (Nichol- son, 1977) has been previously reported in the continental United States from southern Texas (Correll & Johnston, 1970, as Miétracarpum hirtum (1.) DC.), central Florida (Ward, 1976, as Métracarpus villosus (SW.) DC.), and most recently from southern Louisiana (Wunderlin, 1979, as Mitracarpus hirtus). In the course of identifying a Rubiaceous collection from R. Dale Thomas, Northeast Louisiana University, | discovered an SMU collection of Mitracarpus hirtus misidentified as Spermacoce glabra from southern Missis- sippi. This is apparently the only known record of it from that state. Collec- tion data: Mississippi. Pearl River Co.: cultivated field with Crotalaria spectabilis, 2 mi N of Henleyfield, 15 Nov 1964, Sargent 8542 (SMU). his is about 12.5 mi NW of Picayune on hwy 43, which is around 15 mi NE of the Louisiana collection reported by Wunderlin. Wunderlin’s report from Louisiana is based on a later collection from St. ae Parish: waste place RR, Abita Springs, 3 Oct 1970, Thieret 32568 I suspect the two collections to have their origin from one introduction possibly originating in Mississippi based on collection dates and localities. Scirpus californicus (C.A. Mey.) Steud. is presently known in the United States from Arkansas and Texas west to California. It is believed that this is the first known collection to be reported from Mississippi. SIDA 8(1): 117. 1979. 118 Mississippi. Hancock Co.: low wet areas, Bay Saint Louis, 10 Aug 1952, Demaree 32719 (SMU). This extends the known range of S. californicus about 275 mi eastward.—Barney Lipscomb, Herbarium, Southern Methodist University, Dallas, TX, 75275 REFERENCES arene D. S. & M. ree ica 0. xas. Texas Sa See Ren 5. typification of taxa: Proposals on ac ‘OL ge D. H. ypification ore names vs 48 and oid pettin of Mitracarpus birtus vs. M. villosus (Rubiaceae). " Manual of the vascular plants of Article Taxon 26: 569- SMALL, J. K. 1933. Flora of the southeastern flora. Publ. by author, New York. WARD, D. B. 1976. Maracas rpus ieee genus new to Florida od. eastern North America. ee a 78: was — WUNDERLIN, R. 79, Not Spermacore and Mitracarpus (Rubiaceae) in 313-316. southeastern United Bae Phyoloeta 41(5) SIDA 8(1): 118. 1979. DOCUMENTED PLANT CHROMOSOME NUMBERS 1979:1° JAMES S. PRINGLE Royal Botanical Gardens, Box 399, Hamilton, Ontario, Canada LEN 3H8 Figs. 1-12 are tracings from photomicrographs. All figures represent mitosis in root tips. Figs. 1, 3, 5, and 7-12 show chromosomes at metaphase; Figs. 2, 4, and 6 show chromosomes at anaphase (2 sets per figure ). ALISMATACEAE ALISMA TRIVIALE Pursh.” (Fig. 1.) 2n=28. CANADA, Reg. Mun. Hamilton- Wentworth: Pringle 1286 (HAM). SAGITTARIA CUNEATA Sheldon.” (Fig. 2.) 2n=22. U.S.A., Mich., Presque Isle Co.: Pringle 1225, 1273 (illus.) (HAM). SAGITTARIA LATIFOLIA Willd.” (Figs. 3,4.) 2n=22. CANADA, Ont., Reg. Mun. Hamilton-Wentworth: Pringle 1285 (HAM). ARACEAE ARISAEMA MACROSPATHUM Benth." (Figs. 5,6.) 2n=28. MEXICO, Mexico, Mun. Temascaltepec: Duncan et al. 2486 (MICH). LILIACEAE CLINTONIA ery ae Raf.’ (Fig. 7.) 2n—=28. U.S.A., Mich., Cheboygan Co.: Pringle 1224 (HAD CLINTONIA Sane a Raf.’ (Fig. 8.) 2n=32. U.S.A., N.H., Coés Co.: Pringle 1277 (HAM). (Documentation tor report in Utech, F., & L.B. Thien, Michigan Bort. 12:122—123. 1973.) RANUNCULACEAE CLEMATIS INTEGRIFOLIA L.” (Fig. 9.) 2n=16. Cult.: Pringle 345 (HAM). COMPOSITAE SOLIDAGO CANADENSIS L. var. HARGERI Fern.” (Fig. 10.) 2n=18. CANADA, = : in. Hamilton-Wentworth: Pringle 1372 (HAM). SOLIDAGO OHIOENSIS Riddell.” (Fig. 11.) 2n=18. Cult. ex U.S.A., Mich., Che- boygan Co.: Pringle 1289 (HAM). SOLIDAGO RIDDELLII Frank.” (Fig. 12.) 2n=18. Cult. ex CANADA, Ont., Essex Co.: Pringle 1250 (HAM). ® Contribution No. 32 from the Royal Botanical Gardens, Hamilton, Ontario, Can- a "Taxon for which the same chromosome count has been published previously. © Taxon for which a different chromosome count has been published previously. “Taxon for which no chromosome count has been published previously. SIDA 8(1): 119. 1979. 120 AP L >» eo \ x Z ” ies a Ate PM yen » ee as ( n \ % 7 a NS = S = 7 i Vy = =a? 3 ™~ , 3 A ) ‘Dv ad a ' oo ARS SEP aN hy wy i ab Nowe / xX) U ~~ ‘ ~\ I ri | ty“ Sex NOTICE VASCULAR PLANTS OF NORTH AMERICA NORTH OF MEXICO The Flora North America project (FNA) was recently revitalized by the Man and Biosphere Program (MAB) as a binational effort between the United States and Canada to produce a conventional flora of the vascular plants of North America north of Mexico using traditional methods. Initial funding for the proposed five-volume work is being provided by the Na- tional Park Service of the Department of the Interior. The /FNA Program Council plans to coordinate the research which will be necessary to produce a floristic publication of high scientific quality pertinent to na- tional needs. he Program Council has appointed an Editorial Subcommittee consisting of Dr. Reed C. Rollins, Chairman, Gray Herbarium of Harvard University; Dr. Howard S. Irwin of the New York Botanical Garden; and Dr. Roy L. Taylor of the University of British Columbia Botanical Garden. Dr. James L. Reveal of the University of Maryland has been appointed Editor. Their function will be to stimulate and coordinate the efforts of the botanical community in the writing of the flora. Toward this goal, the Editorial Sub- committee is currently working on a proposed format for the flora. Initial efforts will be toward the production of a volume treating the monocotyledon- ous plants, with a volume on the sympetalous dicotyledonous plants to follow next. The remaining three volumes will be worked on in the future. Long-term funding for the flora project is being explored by the Program Council, chaired by Dr. Peter H. Raven of the Missouri Botanical Garden. It is hoped that the flora project will be completed by 1990. Individuals wishing additional information or interested in contributing to the project, and in particular treatments of the monocots and sympetalous dicots should write to: Dr. James L. Reveal, Editor College Park, eure’ 3072 4? SIDA 8(1); 121, 1979. | SIDA Susie" VOLUME 8 NUMBER 2 NOVEMBER 1979 CONTENTS Sidus Sidarum—III. Sida rzedowskii sp. nov., including a preliminary discussion of the Sida elliottii species group. Paul A. Fryxell. 123 Taxonomy of the Verbesina virginica complex (Asteraceae ). John Olsen. 128 The vascular flora of Savage Gulf, Tennessee. B. Exgene eee Thomas S. Patrick, Loy R. Phillippe, and David H. We 135 A new species of Centrosema (Leguminosae) from nea and a key to the species in Central America. Paul R. Fantz 152 Halogeton (Chenopodiaceae ) in North Se Will H. Blackwell, J. Douglas Haacke, and Christine O. Hopkin. 157 Changes in hesnaes eo of the southeastern United States. Wilbur H. Dunca 170 A new species of Linum from southern Texas and adjacent Mexico. C. M. Rogers 181 Rediscovery of Spiranthes parksii Correll. P. M. Catling and K. L. Melntosh. 188 Clematis pitcheri T. & G. var. dict = caudal: Dennis, comb. nov. (Ranunculaceae). W. Michael Denni: 194 Field notes on Nigerian Striga (Scrophulariaceae). Lytton J. eee Daniel L. Nickrent, Richard A. Mansfield, and James E. A. Ogbor: 196 NOTES. Hyptis mutabilis (Labiatae) in southeastern United States. 202—Trillinm pusillum (Liliaceae) a oni 209—Range extensions of Brazoria pulcherrima dell Corea Rubus trivialis Michx. var. duplaris (Shinners) Mahler, cea rae note on the distribution of Styrax americana fern and flowering plant flora. 213—Neviusia Wee: (Rosaceae) in Mississippi. 215—-The chromosomes of Collinsonia canadensis L. (Labiatae). 216— Nicolletia aia (Compositae) in Baja California. 216. US ISSN 0036-1488 SIDA, CONTRIBUTIONS TO BOTANY Founded by Lloyd H. Shinners, 1962 Editor & Publisher Wm. F. Mahler SMU Herbarium Dallas, Texas, 75275 Associate Editor Assistant Editor John W. Thieret Barney L. Lipscomb Northern Kentucky Universit SMU Herbarium Highland Heights, Kentucky, 41076 Dallas, Texas, 75275 Guidelines for contributors are available upon request beginning with volume 8. Subscription: $6.00 (U.S.) per year; numbers issued twice a year. © Sida, Contributions to Botany, Volume 8, Number 2, pages 123-216. Copyright 1979 by Wm. F. Mahler SIDUS SIDARUM—III. SIDA RZEDOWSKII SP. NOV., INCLUDING A PRELIMINARY DISCUSSION OF THE SIDA ELLIOTTII SPECIES GROUP PAUL A. FRYXELL Research Geneticist, U.S. Department of Agriculture, Texas AGM University, College Station, TX 77843 Eight North American species of Sida show strong affinities with one an- other and merit recognition as a distinct species group, as was first noted by Small (1898). The characters that unite them include (1) distinctive conformation of the fruits (generally oblate) and of the mericarps (see Figure 1), which are usually about 8 in number; (2) leaf shape that varies from broadly elliptic to narrowly linear (never rhombic or cordate-ovate); (3) a tendency in some species to rose-colored flowers (at least on drying), reaching an extreme in S. rzedowskii with purplish flowers; and (4) a tendency to extreme shortening of the apical internodes in several species, giving rise to congested apical inflorescences The abbreviated apical internodes are well expressed in the newly de- scribed species, Sida rzedowskii, and are suggestive of a similar condition in Sida sect. Malachroideae (Clement, 1957; Fryxell, 1975). In the latter group of species, typified by S. anomala St-Hilaire, the condition is described by Schumann (1891) as “Flores petiolis bractearum foliacearum adnati, saepissime apice ramulorum subumbellato-congesti” and by Kearney (1954) as ““Peduncles adnate to the petiole of the subtending leaf or bract.” Careful examination indicates that the adnation of parts characteristic of sect. Malachroideae is simply an extension and intensification of the shortening of the apical internodes found in S. rzedowskii, S. neomexicana, and S. inflexa and occasionally expressed weakly in S. ell/ottii. Thus, we find in the present group of species an intermediate condition between the extreme characteristic of sect. Malachroideae and the less specialized condition found in the re- mainder of the genus. It is plausible to suggest that this tie indicates a phylogenetic link connecting sect. Malachroideae with the balance of the genus. It should be noted that sect. Malachroideae also has a tendency to rose-colored corollas and elliptic leaves, similar to the tendency in the S. elliottii group. Iam explicitly not suggesting, however, that the members of the S. elliottii group should be included in sect. Malachroideae. They are distinguished from the latter by their erect (not procumbent) habit and by their more numerous mericarps that lack rugulose or muriculate ornamentation on the SIDA 8(2): 123-127. 1979. 124 dorsal walls, as well as by other characters. The S. e/liottii group 1s not now given formal taxonomic status, pending further studies leading to a better understanding of infrageneric groupings in Sida. It is simply recognized for the present as a coherent group, without rank, and the following key is presented as an aid to distinguishing the species. Species are sometimes difficult to distinguish within this group. For example, some eee of characters occurs between S. elliottii and S. lindheimeri (e.g. in Arkansas). In this instance it may be preferable, following more detailed study, to merge the species and distinguish the taxa as S. elliottii var. elliottit and S. elliottii var. texana Torrey & Gray. Similarly in northern Mexico, it is sometimes difficult to distinguish S. ed/votti and S. nomexicana. These taxa are retained in specific rank in the key, following prevalent usage. A detailed analysis of these problems is merited. KEY TO THE SPECIES OF THE SIDA ELLIOTTIIT GROUP 1. Pedicels up to 15 cm long, usually more than twice the length of the subtending eaves 2. Leaves narrowly lanceolate, 7-10(—15) times as long as wide; calyx 6-7 mm long . S. longipes 2. Leaves broadly elliptic, 1.2—2 times as long as wide; ee 9-11 mm long ; otosina 1. ae oe no more than 6 cm long, usually shorter than the caheending lea . Apical congestion of flowers and fruits conspicuous, as a result of oa ener: re the apical internodes . Corolla rose or purple (with yellow center); leaves elliptic, 2-6 times as long as broac . S. rzedowskti 4. Corolla yellowish, sometimes — rose; leaves Tanceolate to linear, usually 0 (—20) times as lon oad 5. Plants freely branched one i base; leaves narrowly lanceolate to linear, 1 broad S. neomexicana pow on plane sparingly branched; leaves broadly elliptic, 10-15 mm a . S. inflex . Apical congestion of flowers and fruits slight “e Calyx 7-10 mm long; pedicel often 2-5 cm long, shorter than to slightly exceeding ian: leaf . &. lindheimert 6. Calyx 5-7 mm long; pedicel usually 1-3 cm feng, shorter than subtending leaf 7. Leaves eee often 15-20 times as long as wide; mericarps 8-11 . S. elliotti 7. Leaves elliptic, usually 1.5—-3 times as long as wide: ‘mericarps 5- .. . S. turneroides SIDA ELLIOTT Torrey & Gray, Flor. N. Amer. 1:231. 1838 (based on S. gracilis Elliott, 1822, non Richard, 1792). (S. rubromarginata Nash, 1896; S. leptophylla Small, 1898). North Carolina south to Florida and west to southernmost Missouri and Arkansas; eastern Mexico from Veracruz and Querétaro north to Nuevo Leén and the Coastal Bend of Texas. ida rubromarginata and S. leptophylla have been maintained by Kearney (1954) and others as distinct from S. ellottii, but they seem to be rela- tively broad-leaved and relatively glabrous variants (respectively) of it. 125 Although Elliott originally described S. elliottii as ane examination of Elliott's type reveals that this is incorrect (Weatherby, 2 SIDA LINDHEIMERI Engelmann & Gray, Boston J. Nat. Hist. 5: 213, 1845. (Sida texana (Torrey & Gray) Small, 1903). Central Texas and Louisiana to Sinaloa (1 specimen); cited from Guatemala by Standley & Steyermark anaes and from Chiapas by Standley (1923), but these may refer to S. ellio SIDA LONGIPES A. Gray, Pl. Wright. 1:19. 1852. Trans-Pecos Texas and Coahuila. SIDA NEOMEXICANA A. Gray, Proc. Amer. Acad. Arts ee ae Trans- Pecos Texas, New Mexico, Coahuila, Chihuahua, and Duran SIDA POTOSINA Brandegee, Univ. California Publ. Bot. 4:184. 1911 San Luis Potosi and Tamaulipas. SIDA INFLEXA Fernald, Rhodora 42:463. 1940. A restricted endemic in south- eastern Virginia. SIDA TURNEROIDES Standley, Publ. Field Mus. Nat. Hist., Bot. Ser. 22:90. 1940. Hidalgo to Tamaulipas. SIDA rzedowskii Fryxell, sp. nov. Planta herbacea perennis, minute stellato-pubescens, internodiis apicalibus abrupte abbreviatis. Laminae foliorum ellipticae, dentatae, utrinque stellato-pubescentes. Pedi- celli in axillis foliorum summorum, solitarii, plerumque in apicem congesti. Calyces aepe in nervis hirsuti. Petala glabra, gene Fructus oblati, in apicem minute pare nee mericarpia 8-11, laevigata vel leviter reticulata, 1-seminalia, in apicem acuta vel rotundata. Herbaceous perennial 1-4 dm tall, the stems branched and suberect, in- vested with minute stellate pubescence, the internodes abruptly shortened at the apices of the branches. Leaf lamina elliptic or oblong, 1-3 cm long, —)3-6 times as long as wide, subacute or obtuse, dentate, 3(—5)-nerved from the base, the lateral nerves inconspicuous, with dense or sparse stellate pubescence above and beneath. Petiole densely pubescent, 4-10 mm long. Stipules 4-7 mm long, linear-lanceolate, eearaes Maes: the petiole. Pedicels 1-10 mm long, solitary in the axils of the upper leaves, mostly aggregated apically (forming inflorescences eee to ae of Side ciliaris Bs owers, reduced leaves, and stipules crowded together because of the shortened internodes). Calyx 4-6 mm long, with pubescence like that of the leaves except sometimes hirsute on the nerves, 10-nerved, 5-lobed, the lobes ovate-acuminate. Petals glabrous (including the claw), 6-8 mm long, markedly asymmetrical, rose to red-purple but yellowish at the base. Staminal column pallid, 2-3 mm high, slightly pubescent or glabrous; filaments 1-1.5 mm long, arising from the apex of the column; anthers yellow, few (ca. 20); pollen yellow. Styles 8-11; stigmas capitate. Fruits oblate, 5-6 mm diameter, 3-4 mm high, apically with minute stellate pubescence; mericarps 8-11, 3-4 mm high, smooth or slightly reticulate on the lateral wall, apically acute or rounded and dehiscent, l-seeded (Fig. 1,D). Seeds 2 mm long, 126 Figure 1. Comparison of mericarps of the four species of Sida. A, Sida lindheimeri (Byrd 42); B, Sida neomexicana (Board s.n.); C, Sida elliottii (Hill 2841) Sida rzedowskii (Hilerio 72). Scale = 5 mm. glabrous, blackish YPE: MEXICO: Hidalgo: Cerro Ventoso, entre Pachuaca y Real del Monte; orilla del camino, matorral de Hechtia podantha; alt. 2500 m; flores de color morado oscuro; 29. viii. 1965, Rzedowski 20560 (holotype: ENCB). Additional Ae examined : Hidalgo: Mpio. Pachuca, 4 km al NE de Pachuca, sobre la carretera a Real del Monte, alt. 2650 sas Sage ee (ENCB). Mpio Tepeapulco, Cerro Tres Penas, alt. 2500 m, Ventutn 245 (ENCB, pf). Edo. México: Mpio. Phen 3 alrededores de la Presa de la Concepcion, alt. 2350 m, Hilerio 72 (ENCB, pf); ca. 2 km al NW de Tepotzotlan, alt. 2350 m, Huerta 41 (ENCB); alrededores de la Hacienda Lanzarote, alt. 2350 m, Rzedowski 35021 (ENCB, pf). Mpio. tae sed as te W de Cerro Sincoahe alt. Rzedowski 34307 (ENCB); al de Huehuetoca, Cerro Ahumada oe del Rancho Nuevo, alt. 2350 m, Roses ski a 03 ag Mpio. ae 3kma de Atizapan de Zaragoza, alt. 3400 m [2400 m?], Cruz 633 (ENCB); ele a Tigre, al NW de Atizapan, alt. 2500 m, Rzedowski 32004 (ENCB, a Distrito Nees Pedregal de San Angel, cerca de San Angel, Rzedowski 1019 (ENCB); al sur de Cerro Zacatepec, Rzedowski 283 (ENCB). Jalisco: Mpio. Be de Moreno, Paso de La Troje, Cerro La Campana, SW of Ojuelos, alt. 2100-2300 m, McVaugh et al. 16832 (ENCB); ia ae SW “of Lagos de Moreno, Waterfall 15659 (SMU-in part). Chiapas: Cerro San Cristobal, San CriscSbal de las Casas, alt. 7000 ft, Breedlove DS San Luis Pe ytosi: Mpio. de Mezquitiq [Mexquitic], km 65 de la carretera San Luis Potosi-Zacatecas; alt. 2090 m; Garcia 665 (CHAPA, pf). The new species is named in honor of Dr. J. Rzedowski, author of Vegetacién de México and coauthor of Flora Fanerogdmica del Valle de México, whose extensive Mexican collections, especially in the Valley of ry Mexico, have brought this species to light. It is especially noteworthy that this species occurs rather commonly in the Valley of Mexico at elevations of 2300-2700 m, higher than any other North American species of Sida is known to occur. Moreover, it appears to be confined to these high eleva- tions, extending from Chiapas to northern Jalisco and southwestern San Luis Potosi. REFERENCES CLEMENT, I. D. 1957. Studies in Sida tira Contr. Gray Herb. 180: 1-91. FRYXELL, P. A. 1975. are sidarum [lI]. a 6: 1-6. ore T. H. 1954. ae key to North American species of Sida L. eafl. W. Bot. 7: eae SCHUMANN, K. 1891. oe In: K. F. P. von Martius, Flora Brasiliensis 12(3): 279-34 SMALL, : K. 1898. eercice in the botany of the southeastern United States.-XIV. Bull. Torrey Bot. Club 25 84. STANDLEY, P. C. 1923. Sida ae Trees and Shrubs of Mexico. Contr. U.S. Nat'l. Herb. oe Rene _ d J. Steyermark. 1949. Malvaceae. In: Flora of Guatemala. Fieldiana, Boe Date. 324-38 WEATHERBY, C. A. 1942, A list of type specimens in Elliott’s herbarium. Rhodora 44: 249-262 TAXONOMY OF THE VERBESINA VIRGINICA COMPLEX (ASTERACEAE) JOHN OLSEN Department of Biology, Southwestern at Memphis, TN 38112 The Verbesina virginica complex contains three taxa, V. virginica L. var. virginica, V. v. var. laciniata (Poir.) Gray and V. microptera DC., in the widespread tropical section Ochractinia. The complex has been a source of taxonomic confusion for some time. This confusion has arisen over the segregation of the southwestern United States taxon, V. microptera, from the more widespread V. virginica and the segregation of the southeastern popu- lations of V. virginica at either the specific or the varietal levels. Verbesina microptera DC. was described from a south Texas Berlandier collection (no. 1442) in 1836. Gray (1883) and Stanford (1976) con- sidered this name synonymous with V, virginica, while Correll and Johnston 1970) treated the two taxa as distinct species, though suggesting they are conspecific. These two taxa are commonly distinguished by the number of wings on the stem and the extent to which these wings continue into the inflorescence. Examination of herbarium material quickly shows these characters are com- pletely unsatisfactory; the number of wings and their extent varies with the robustness of the plant and ranges from none to six in all taxa of this complex. Coleman (1977) has performed experimental crosses on populations of all three taxa in this group and recognized V. virginica and V. microptera as distinct species while maintaining the southeastern United States popula- tions as a variety of the former. He also points out the differing number ray florets that distinguish between V. microptera (10-13) and V. virginica (1-5). My examinations of herbarium and living material supports the conclu- sions of Coleman (1977). Furthermore, V. sicroptera is geographically isolated from V. virginica, the former largely restricted to the region south of San Antonio, Texas, and extending into Coahuila and Nuevo Leon, Mexico, and the latter occurring from west of Austin, Texas, north to Kansas and east to the Carolinas (Figure 1). Correlated with this geographic break (Figure 2) is a distinct morphological boundary as noted by Coleman. Verbesina virginica usually has three ray florets (always less than seven) with 10-13 disc florets, while V. mcroptera usually has 11 ray florets (always more than nine) with ca. 25 disc florets. No exceptions to this meristic difference have been detected in the specimens examine SIDA 8(2): 128-134. 1979. inica laciniata oa rd eudidnigadinaeee nn —_— _————— ead ES | : : virginica var. virginica li Wt a7, il \ | | ¢ ) ov < micropte V. Fig. 1. Distribution of the taxa in the Verbesina virginica complex. SA : ipauuiinecsaanae nee btm: —" ‘ Vie VLE var. 7 OSMUNDACEAE Osmunda cinnamomea Osmunda regalis ie ) Gray—l1 POLY PODIACEAE Polypodium polypodioides (L.) Watt— Polypodium virginianum L.—5 142 SCHIZAEACEAE Lygodium palmatum (Bernh.) Swartz —2 SELAGINELLACEAE Selaginella apoda (L.) Spring—4 VITTARIACEAE Vittaria sp. (gametophyte )—5 II]. GYMNOSPERMS PINACEAE nus echinata Miller—3 Pinus taeda L.—: Pinus virginiana Miller—2, 3, 4 Tsuga canadensis (L.) Carr.—5 CUPRESSACEAE Juniperus virginiana L.—3 Ill. ANGIOSPERMS ACANTHACEAE Ruellia caroliniensis ( Walt.) Steud —6 ACERACEAE Acer negundo 3 Acer pense L—5 Acer rubrum Acer nee naenil eek —S5 ALISMATACEAE Alisma subcordatum Raf.—2 Sagittaria engelmanniana J. G. Smith ssp. longirostra (Micheli) Bogin—3 AMARYLLIDCEAE Agave virginica L.—6 Hypoxis hirsuta (L.) Coville—3, 6 ANACARDIACEAE Rhus copallina L—3 Rhus glabra L.—6 Rhus radicans L.—4 ANNONACEAE Asimina triloba ( APIACEAE Angelica venenosa (Greenway) Fern.— Chaerophyllum tainturieri Hooker—6 Cicuta maculata Cryptotaenia oe (L.) DC.—6 Daucus carota Erigenia bulbosa (Mich. ) Nutt.—5 Osmorhiza claytoniti (Michx.) Clarke —5, 6 Oxypolis rigidior (L.) Raf—l1, 3 Sanicula canadensis L—3, 5 Sanicula gregaria Bicknell—5 Sanicula smallii Bicknell—4 Sanicula trifoliata Bicknell— Taenidia integerrima (L.) Drude—6 Thaspium barbinode (Michx.) Nutt.— t, Zizia aptera (Gray) Fern.—6 AQUIFOI IACEAE Ilex ambigua Scare Torr. var. mon- tana (T. & G.) _— 3, 6 =i decidua Walt.— x opaca Ait.—3, 4 ARACEAE Arisaema triphyllum (L.) Schott—4, 5 ARALIACEAE Aralia racemosa L.—6 Aralia spinosa L.—3, 6 Panax quinquetolium L—S5 ARISTOLOCHIACEAE Aristolochia macrophylla Lam.—5 Aristolochia serpentaria L.—5 Asarum canadense L.— rece arifolia (Michx.) Small—2, ASCLEPIADACEAE Asclepias amplexicaulis Smith—3 Asclepias quadrifolia Jacq.—4, 5 Asclepias tuberosa a ata L— Matelea_ carolinensis a Woodson —6 ASTERACEAE Ambrosia son ae L—3 Ambrosia trifida L.— Antennaria Sia nraeaitels (L.) Hook. —3 Antennaria ska Rydb.—4 Aster cordifolius L.—6 Aster ere L—3, 5, 6 Aster dumosus L.—2 Aster Soe Michx.—3 Aster laevis L—6 Aster mere (L.) ae 6 Aster linariifolius L.— Aster lowrieanus ae Aster patens Ait. Aster paternus Cronquist—2, 3 Aster shortii Lind].— Bidens polylepis Blake—3 Cacalia atriplicifolia L—6 Chrysanthemum leucanthemum L.— Cirsium ae (L.) Sprengel—2 Cirsium muticum Michx.—6 Coreopsis eae L.—6 Coreopsis major Walt. var. stellata utt. foe album = Heterotheca graminifolia (Michx.) Shin- ners—3 rs Heterotheca mariana (L.) Shinners—2, 3 143 Hieracium gronovii L.—2, 3 Hieracium paniculatum L. Hieracium venosum L.—3, Krigia biflora (Walt.) Blake—3, 6 6 Lactuca canadensis L.— solr floridana (L.) Gaertn.—6 Liatris aspera Michx.—3 Liatris microcephala (Small) K. Schu- mann—2 a spicata (L.) Willd —3 Marshallia trinervia (Walt.) Trelease ex ae & Coville— Parthenium integrifolium L.—3 L—3, Prenanthes serpentaria Pursh— Pyrrhopappus carolinianus (Walt.) DC. Rudbeckia fulgida Ait. var. umbrosa on & Beadle) Cs fidbeclia hirta L—3, 6 Rudbeckia laciniata L.—3, 6 Senecio aureus L.— Senecio obovatus Muhl. ex Willd. —4 3 Solidago arguta Ait—3 Solidago boottit Hook. var. bootti Solidago boottii Hook. var. Se iglae ) Cronquist— Solidago caesia L.—6 canadensis L. var. scabra (Muhl.) T. & G—3, 6 Solidago curtisii T. & G—3 Solidago erecta Pursh—2, 3, 6 Solidago flexicaulis L. = Solidago gigantea Ait Solidago -raminifolia “ys a var. Solidago rugosa Cronquist—2, Solidago liaifolia Muhl.—3, 6 Taraxacum officinale Wiggers Verbesina alternifolia (L.) Brit—6 Vernonia altissima Nutt.— —6 Mill. ssp. aspera ( Ait.) 6 BALSAMINACEAE Impatiens capensis Meerb.—6 Impatiens pallida Nutt.—5 144 BERBERIDACEAE Viburnum nudum L.—2, 3 Caulophyllum thalictroides (L.) Michx. Viburnum rufidulum Raf.—2, 6 —_ - CARYOPHYLLACEAE Podophyllum pelratum L.—5 : . ; _. Cerastium holosteoides Fries var. vulgare BETULACEAE (Hartman) Hylander— Alnus serrulata (Ait.) Wéilld.—2 Silene rotundifolia Nutt—4, 5 Betula lutea Michx. f.—4, 5 Silene stellata (L.) Ait. f—3 Silene virginica L.— Betula nigra L.—2 : Carpinus caroliniana Walt.—6 Stellaria pubera Michx.—5 Corylus americana Walt.—6 CELASTRACEAE Ostrya virginiana ( Mill.) K. Koch—%5 Euonymus americanus ma) BIGNONIACEAE Fuonymus aceopun puree ice, —06 Antsostichus capreolata (L.) Bureau— ae ain: 3 sae CISTACEAE 4,0 Campsis radicans (L.) Seemann—6 es tomentosa (Thunb.) Steud. Sheen —6 Lechea racemulosa Michx.—3, 6 > Tradescantia subaspera Ker—6 BORAGINACEAE Cynoglossum virginian CONVOLULACEAE Lithospermum toa Rugel ex DC. Calystegia spithamaca (L.) R. Brown— —4 6 Cuscuta campestris Yuncker—6 BRASSICACEAE Cuscuta compacta Juss.—3 Arabis canadensis L.—6 Ipomoea purpurea (L.) Roth—6 Arabis leavigata’ (Muhl. ex Willd.) Poir.—4, 6 CORNACEAE Cardamine hirsuta L.—6 Cornus alternifolia L. f.—5 Dentaria laciniata Muhl.—5 Cornus amomum Mill.—6 Cornus florida L.—5 BUXACEAE CRASSULACEAE 4 Pachysandra procumbens Michx.—6 Pa eee Sedum ternatum Michx.— Calycanthus floridus L.—4 CYPERACEAE Carex aestivalis M. A. Curtis—3, 6 CAMPANULACEAE é . Carex albolutescens Schwein.— Campanula americana L.—6 Carex albursina Sheldon—3 Lobelia cardinalis L.—6 Carex amphibola Steud.—2 Lobelia inflata L—2, 3 Carex austro-caroliniana Bailey—4 Lobelia puberula Michx.—6 Carex bail ey! Britt. —" Specularia perfoliata (L.) DC.—3, 6 Carex cephalophora Muhl. ex Schkuhr —3 CAPRIFOLIACEAE Carex communis Bailey—2, 4 Lonicera japonica Thunb.—6 Carex complanata Torr. & Hook.—3 Lonicera sempervirens L.—3 Carex crinita Lam,—3 Sambucus canadensis L.—6 Carex debilis Michx. var. rudgei Bailey Sambucus pubens Michx.—5 —l, 2, ° Triosteum perfoliatum L.—3 Carex emmonsi1 Dewey—5 Viburnum acerifolium L.—2 Carex flaccosperma Dewey—1 Viburnum cassinoides L.—2 Carex gigantea Rudge—l Viburnum dentatum L-3 Carex hitchcockiana Dewey—6 Carex incomperta Bicknell—3 Carex intumescens Rudge—l, 2, 3 Carex jamesii Schwein.—6 Carex joorit Bailey—1l Carex laxiflora Lam.—3, 5 Carex lurida Wahl.—3 Carex pedunculata Muhl.— Carex pensylvanica Lam. var. distans Peck—4 Carex plantaginea Lam Carex purpurifera Mac Aenrie= 2: 4 Carex retroflexa nee ex ere ee Carex rosea Schkuhr— Carex swanii noe ) Macca: 3, 6 Carex virescens Muhl. ex Schkuhr—2, 3, 6 Cyperus praclongatus Steud.—6 Cyperus retrofractus ae Torr.—3 Cyperus strigosus Eleocharis obtusa (Willd. ) Schultes— Rhynchospora capitellata (Mich) Vahl 535 Rigethossoe corniculata (Lam.) Gray —l1 Scirpus atrovirens Willd. var. georgianus (H Fern.—3, 6 Scirpus cyperinus (L.) Kunth—1 Scleria oligantha Michx.—6 Scleria triglomerata Michx.—3 DIOSCOREACEAE Dioscorea batatas Dcne.—6 Dioscorea villosa L.—4, 6 EBENACEAE Diospyros virginiana L.—3 ERICACEAE SEs meee (L.) Pursh—6 Epigaea repens L.—3, 4, 6 Kalmia latifolia L—2, 3, 4 yonia ligustrina (L —3 Monotropa uniflora L.— Mo ene: odorata Schwein. in Ell— ae arboreum (L.) DC.—3 Rhododendron alabamense Rehder—l, Rhododendron arborescens (Pursh ) orr.—6 Rhododendron calendulaceum (Michx. ) orr.— 145 Rhododendron cumberlandense Braun Rhodedend {ron maximum L.—5 Rhododendron nudiflorum (L.) Torr.— Vacaniuin arboreum Marsh.— Vaccinium constablaei ey 4 Vaccinium corymbosum L.—3 Vaccinium stamineum L.—3 Vaccinium vacillans Torr.—3, 4 “aS EUPHORBIACEAE Acalypha gracilens Gra Acalypha rhomboidea Rat. 6 Croton monanthogynus Michx pe es corollata L. var. ainnii Aone nall ) hles—3 ic mercurialina Michx.—4 FABACEAE Albizia julibrissin Dur Amphicarpa bracteata (L.) Fern.—6 pios americana Medicus—3 Baptisia tinctoria (L.) R. Brown—2, 3 Cassia marilandica L.—6 Cassia nictitans L—6 Cercis canadensis L.—6 Cladrastis ae (Michx. f.) K. Koch— 5 Clitoria mariana L.—3 Desmodium eee (Muhl. ex W illc Loudon— Desmodium Ae (Muhl. ex Willd.) Wood—6 Desmodium marilandicum (L.) DC— ) jou jae ille Desmodium nudiflorum (L.) DC—2, 6 ee sae ta (L.) DC. var. niculatum— eres ices Soe DC. var. dillenit (Darl.) Isely—3, Desmodium obtusum (Muhl. ex Willd.) =O). Desmodium pauciflorum (Nutt.) DC. —6 Desmodium rotundifolium DC.—6 Galactia volubilis x Britt.—6 Lespedeza bicolor Turcz.—3 Lespedeza cuneata pen G. Don iecpedens hirta (L.) Hornemann—2 aes, Lespedeza hirta X inte Lespedeza intermedia (S. Wea aie —3,6 146 Lespedeza repens (L.) Bart.—3 aed steuvel as t— Melilotus alba Desr.—3 fe oralioies ee Cory—4 Robinia rid )— Robinia i. acacia L—4 carla ee (Soland. ex Sm.) ride— Se ena ice (L.) BSP.—3 Tephrosia virginiana (L.) Pers.—2, 3 Trifolium campestre Schreb.—6 Trifolium pratense L.—6 Vicia caroliniana Walt.—6 FAGACEAE Castanea dentata (Marsh.) Borkh.—3, 4 Fagus grandifolia Ehrhart—2 Quercus alba L Quercus coccinea Meunchh. —3 Quercus falcata Michx.—3 Quercus marilandica Muenchh.—3 Quercus muehlenbergii Engelm.—6 Quercus prinus L.—3 Quercus rubra L.—3 Quercus stellata Wz Quercus velutina pee 3 FUMARIACEAE Dicentra cucullaria (L.) Bernh.—5 GENTIANACEAE Bartonia virginica (L.) a Gentiana quinquefoli ia L— 1 i Swertia caroliniensis (Wa It.) Kuntze— 6 GERANIACEAE Geranium carolinianum L.—6 Geranium maculatum L.—4 HALORATACEAE Proserpinaca pectinata Lam.—1 HAMAMELIDACEAE Hamamelis virginiana L.—2 Liquidambar styraciflua L—4, 6 HIPPOCASTANACEAE Aesculus octandra Marsh.—5, 6 HYDROPHYLLACEAE Hydrophyllum canadense L.—5 Phacelia bipinnatifida Michx.—5 HY PERICACEAE Hypericum ae Walt. var. recognitum Fern. & —3 Hypericum gentianoides “(L) BSP.—2, 3 Hypericum mutilum L—3 Hypericum nudiflorum Michx. ex Willd. —6 Hypericum prolificum L—6 Hypericum punctatum Lam.—2, 6 Hypericum stans (Michx.) P. cer & Robson—3 Hypericum stragalum P. Adams & Rob- son— 6 IRIDACEAE Iris cristata Ait—4, 5 Sisyrinchium a tbidugn Raf.— Sisyrinchium angustifolium Mill—6 JUGLANDACEAE Carya cordiformis (Wang.) K. Koch— Carya glabra ( Mill.) Sweet—6 g. Carya ovata (Mill.) K. Koch—5 Carya pallida (Ashe) Engler & Graebn. —3 Carya tomentosa (Poir.) Nutt.—6 Juglans cinerea L—6 Juglans nigra L—6 JUNCACEAE Juncus canadensis J. Gay ex La Harpe —3 Juncus coriaceus Mackenzie—3 Juncus debilis Gray—1, ° Juncus effusus L.—1, 3 Juncus marginatus Rostk.—3 Juncus repens ch Juncus tenuis Willd —2, 3 la acuminata —6 Luzula bulbosa (Wood) Rydb.—3 LAMIACEAE eoninrona canadensis m virginianum L.—6 i 6 Mosla dianthera Caamition) Michx.—6 Perilla frutescens (L.) Britt.— Prunella vulgaris L.— Pycnanthemum incanum (L.) Michx.— Pycnanthemum muticum (Michx.) Pers. Salvia lyrata L— Salvia urticifolia L—6 Scutellaria elliptica Muhl.—3, Scutellaria pseudoserrata Epling—6 Trichostema dichotomum LAURACEAE Lindera benzoin (L.) Blume—2, 5 Sassafras albidum (Nutt.) Nees—4 LILIACEAE Aletris farinosa L.—3 Allium canadense Chamaelirum ee CL Disporum lanuginosum Miche) Nic- holson—5 Disporum maculatum (Buckl.) Britt.— Erythronium americanum Ker—6 Lilium michiganense Farw.—1 edeola virginiana L.— Polygonatum biflorum (Walt.) Polygonatum pubescens (Willd.) Patch 5 Smilacina racemosa i Desf.—4 Smilax bona-nox L.—6 Smilax ecirrhata (Engelm.) Watson var. hugeri (Small) Ahles—4, 5, 6 Smilax glauca Walt.—3, 4 Trillium erectum L.—5 Trillium grandiflorum (Michx.) Salisb. 5 Trillium recurvatum Beck—5 Uvularia grandiflora Smith—5, 6 Uvularia perfoliata Uvularia sessilifolia L.—2, 3 Veratrum i Robbins—5 Zigadenus eed. Gray—1 LINACEAE Linum striatum Walt.—2, 3 Linum virginianum L var. chon Plan floridanum LOGANIACEAE Spigelia marilandica L—6 147 LORANTHACEAE Phoradendron serotinum (Raf.) M. C. ohnston—5 MAGNOLIACEAE Lirodendron tulipifera L.—4 Magnolia acuminata L.—4 Magnolia tripetala L—6 MALVACEAE Hibiscus syriacus L—6 MELASTOMATACEAE hexia mariana L.—3 Rhexia virginica L—3 MORACEAE Morus rubra L.—G NYSSACEAE Nyssa sylvatica Marsh.—4, 5 OLEACEAE Chionanthus virginicus L.—3 Se (Beadle) J. Wright e ex rein Fraxinus quadrangulata Michx. 6 ONAGRACEAE Circaea lutetiana L. ssp. canadensis (L.) 6 nis L. Oenothera ona Roth—3 ORCHIDACEAE Aplectrum hyemale (Muhl. ex Willd.) Orr. Corallorhiza odontorhiza (Willd.) Nutt. 2G Cypripedium acaule Ait eae ca mee, To var. Seas Willd.) Correll—4 Goose pubescens (Willd.) R. Brown eae spicata (Walt.) Barnhart— Isotria verticillata (Muhl. ex Willd.) Raf.—1 Malaxis unifolia Michx.—2, 3 —4 Orchis spectabilis L Platanthera ciliaris cn Lindely—1, 3 148 Platanthera clavellata (Michx.) Luer— Platanthera cristata (Michx.) Lindley— 1 Platanthera integrilabia (Correll) Luer —1,3 Tipularia discolor (Pursh) Nutt.—6 Triphora trianthophora (Swartz) Rydb. 5 OROBANCHACEAE Conopholis americana (L.) Wallroth— 4 Epifagus virginiana (L.) Barton—%S, 6 OXALIDACEAE Oxalis di ee Jacq.—6 Oxalis stricta L—6 Oxalis aes L.—6 PAPAVERACEAE Sanguinaria canadensis L.—5 ba diphyllum (Michx.) Nutt. PASSIFLORACEAE Passiflora lutea L.—6 PHRYMACEAE Phryma leptostachya L—4 PHYTOLACCACEAE Phytolacca americana L.—6 PLANTAGINACEAE Plantago aristata Michx.—3 Plantago lanceolata L.—3 Plantago rugelit Dcne.—6 Plantago virginica L.—6 PLATANACEAE Platanus occidentalis L—4, 6 POACEAE Agrostis ia (Walt.) Tuckerman 25D. eee gerardit, Vitman—3 Andropogon scoparius Mi Andropogon virginicus L.—6 Aristida dichotoma Michx.—3 oo gigantea (Walt.) Muhl.— ene erectum (Schreb.) Beauv. —6 Bromus purgans L.—6 Bromus secalinus L—2 Calamogrostis cinnoides (Muhl.) Bart. 1,3 Chasmanthium © latifolium ( Michx. ) ates—6 Chasmanthium laxum (L.) Yates—3 Cinna arundinacea L.—6 Dactylis glomerata L.—: anthonia compressa Austin—3 anthonia serivea Nutt.—2 Danthonia spicata (L.) Beauv. ex R. & —)D Decchamibets flexuosa (L.) Trin.—2, 3 Diarrhena americana Beauvy.—6 Digitaria ischaemum (Schreb.) Schreb. ex Muhl. var. itschaemum—3 Digitaria ischaemum (Schreb.) Schreb. uhl. var. violascens (Link) Rad- elesere ord—- Echinochloa muricata (Beauv.) Fern.— 3 jon Eleusine indica (L.) Gaertner—3 Elymus virginicus L.— Eragrostis spectabilis (Pursh ) Steuc jon el|— Erianthus alopecuroides (L.) Ell—6 Festuca elatior L.—3 Festuca obtusa Bichler—6 Holcus lanatus L.—3 Hystrix patula Moench—6 Leersia virginica Willd.—3, 6 Lolium multiflorum Lam.—3 Melica mutica Walt.—2, 4 Microstegium “icine (Trin.) A. -amus—3 Muhlenbergia schreberi J. F. Gmelin— 3 Muhlenbergia sylvatica (Torr.) Torr. ex Gray—6 Panicum an ceps Michx.—3 nicum dichotomum L.— Panicum laxiflorum oon —3, a are longifolum Torr.—1 nm Panicum longiligulatum Nash—1 Panicum philadelphicum Bernh.—3 Panicum polyanthes Schult.—2 Panicum sphaerocarpon Ell.—3 Paspalum laeve Michx.—3 Poa compre 5 Poa oe ‘Hitche —6 Poa pratensis L.—6 Setaria faberi W. Herrmann— Setaria geniculata (Lam. ) Deaig: —3 Sorghastrum nutans 3 et nitida (Bichler) Scribner Sporobs vaginiflorus (Torr.) Wood sce avenacea L.—2, 6 Tridens flavus (L.) te —6 POLEMONIACEAE Phlox amoena eS Phlox carolina L.— Phlox divaricata L_4. 5,6 Phlox glaberrima L—6 Phlox maculata L. ssp. pyramidalis (Smith) Wherry—2, 6 Phlox pilosa L.—6 POLYGALACEAE Polygala curtissii Gray Polygala incarnata L.—3 Polygala senega L.—4 Polygala verticillata L—2, 3 POLYGONACEAE Polygonum cespitosum Blume ~ var. longisetum (DeBruyn) Stewart—6 Polygonum persicaria L.—6 Polygonum punctatum Ell.—6 Polygonum scandens L.—6 Rumex acetosella L.—6 Rumex crispus L.—6 Tovara virginiana (L.) Rat—G PORTULACACEAE Claytonia caroliniana Michx.—5 PRIMULACEAE Lysimachia lanceolata Walt—6 Lysimachia quadrifolia L.—3, 6 Lysimachia tonsa (Wood) Kunth—6 RANUNCULACEAE yeas pachypoda EI].— Anemone quinquefolia L.—6 Ane irginiana L.—6 pe canadensis L.—4 Hepatica acutiloba Hydrastis canadensis L—6 Ranunculus abortivus L—6 149 Ranunculus recurvatus Poir.—5 Thalictrum eine DC.—4, ‘ bescens Purs iG valictrum Fane (L.) Boivin— , 6 Xanthorhiza simplicissima Marsh.—2, 5 RHAMNACEAE Ceanothus americanus L.—6 Rhamnus caroliniana Walt—6 ROSACEAE Agrimonia parviflora Ait—6 Agrimonia pubescens Wallroth—6 Agrimonia rostellata Wallroth—6 Amelanchier canadensis (L.) Medicus— 3, 6 Aruncus dioica (Walt.) Fern.—6 Crataegus crus-galli L—6 Crataegus flabell lata Raa K. Koch—3 Crataegus flava Ait. C rataegus viridis L.— i = Gillenia stipulata (Muhl.) Baillon—o6 oli (L.) Moench—3 Physocarpus opulifolius (L.) Maxim.— 6 Potenulla canadensis L.—2 Pyrus angustifolia Ait.—2, 3, i Pyrus arbutifolia (L.) L. f— Pyrus meclanocarpa (Michx.) ee ) Rosa carolina L.—95 Rosa multiflora Thunb.—6 Rubus argutus Link 5 Rubus betulifolius Small Rubus hispidus L.—2, 3 Rubus occidentalis Rubus phoenicolasius | Maxim.—6 Spiraca japonica L.— RUBIACEAE Cephalanthus occidentalis L—3 Diodia teres Walt.— Diodia ee L—6 Galium aparin Galium circaezans ie —3, 6 Galium latifolium Michx.—5 Galium pilosum Ait.—3, 6 150 Galium triflorum Michx.—3, 6 THYMELAEACEAE Houstonia caerulea L—2, 3 Dirca palustris L—6 oustonia purpurea L.—6 Mitchella repens L.—S, 6 TILIACEAE Tilia heterophylla Vent.—5 SALICACEAE Salix humilis Marsh.—3 TYPHACEAE Salix nigra Marsh.—3, 6 Typha latifolia L—3 SANTALACEAE ULMACEAE Pyrularia pubera Michx.—6 Celtis occidentalis L.—6 Ulmus alata Michx.—6 SAPOTACEAE Ulmus americana L—5 Bumelia lycioides (L.) Pers.—6 Ulmus rubra Muhl.—4 URTICACEAE SAXIFRAGACEAE Boehmeria cylindrica (L.) Swartz—2, 6 Laportea canadensis (L.) Weddell—5, 6 Heuchera americana L.—5, Pilea pumila (L.) Gray—6 Heuchera praviflora Baltes 5 ~~ Heuchera villosa Michx.—5 VERBENACEAE Hydrangea arborescens L. ssp. discolor (Seringe) McClintock—5 Itea virginica L.—1, 2 Mitella diphylla L.—s VIOLACEAE Parnassia asarifolia Vent.—1, 3 Penthorum sedoides L.—6 Astilbe biternata (Vent.) Britt.—5 Verbena simplex Lehmann—3 Verbena urticifolia L— Hybanthus concolor (Forster) Sprengel 5 Tiarella cordifolia L.—5 Viola affinis LeConte—4, 5, 6 Viola blanda Willd.—5 SCROPHULARIACEAE Viola canadensis L.—5 Agalinis tenuifolia (Wahl) Raf.—2, 3 Viola conspersa Reichenb.—4, 5, 6 Aureolaria laevigata (Raf.) Raf.—6 Viola hastata Michx. Aureolaria virginica (L.) Pennell—3 Viola hirsutula Brainerd Chelone glabra L.—3 Viola pedata L—2, 3 Chelone lyonti ae Viola pensylvanica Michx.—5 Gratiola virginiana L.—3 Viola primulifolia L—2, 3 Lindernia dubia (L.) Pennell—6 Viola rostrata Pursh.— Mimulus alatus Ait—6 Viola triloba Schwein.—5 Pedicularis canadensis L—4 Penestemon canescens (Britt.) Britt. ssp. VITACEAE brevisepalus (Pennell) Crosswhite—6 Parthenocissus quinquefolia (L.) Planch. Verbascum blattaria L—6 —4, Verbascum thapsus L.—3 Vitis aestivalis Michx.—3, 4, 6 Vitis cinerea Engelm. ex es SIMAROUBACEAE Vitis rotundifolia Michx.—4 Ailanthus altissima (Miull.) Swingle—5 XYRIDACHAE SOLANACEAE Xyris torta Smith—1 Physalis heterophylla Nees—6 Solanum carolinese L.— THEACEAE Stewartia ovata (Cav.) Weatherby—4, 6 151 REFERENCES BRAUN, E. L. 1950. Deciduous of Eastern North America. The Blakiston Co., Philadelphia, Pennsylvania. 59 CAPLENOR, D. 1955. An snorted list of the vascular plants of the gorges of the Fall Creek Falls State Park. Jou enn. Acad. Sci. 39: 93-108. fest ell 1965. The owe 5 ne gorges of the Fall Creek Falls State Park in ee Jour. Tenn. Acad. Sci. 40: 27-39. CLARK, R. C. 1966. The vascular loa of the Fiery Gizzard gorges in south central Tennessee. eae M.A. Thesis, The University of North Carolina, Chapel Hill. 89 p COMMITTEE FOR TENNESSEE RARE SS 1978. The rare vascular plants of Tennessee. Jour. Tenn. Acad. Sci. 53: 128-133. FENNEMAN, N. M. 1938. Physiography of Eastern United States. McGraw-Hill Book Co., New York. 691 FERNALD, M. L. 1950. rages Manual of Botany. Eighth Edition. American Book JOHNSON, R. G. 1969. A taxonomic and floristic study of the Liliaceae and allied families in the southeastern a States. Unpublished Fh.D. Diss., West Virginia University, ia are 337 p QUARTER MAN, E., B. TURNER and T. E. HEMMERLY. 1972. Analysis of virgin mixed aa Honests in Savage Gulf, Tennessee. Bull. Torr. Bot. Club 99: 228-232. RADFORD, A. E., H. E. AHLES and C. R. BELL. 1968. Manual of the Vascular Flora of the Carolinas. Univ. of North Carolina Press, Chapel Hill. 1183 p SHERMAN, H. L. 1958. The vegetation and floristics of five gorges of the Cumberland Plateau. Unpublished M.S. Thesis, The University of Tennessee, Knoxville. 103 PP. UNDERWOOD, J. K. 1962. Cyperaceae of Tennessee. Unpublished Ms. deposited in the University of Tennessee Herbarium, Knoxville. 80 WOFFORD, B. E., D. H. WEBB and W. M. DENNIS. 1977. State records and other recent noteworthy collections of Tennessee plants II. Castanea 42: 190-193. A NEW SPECIES OF CENTROSEMA (LEGUMINOSAE) FROM NICARAGUA AND A KEY TO THE SPECIES IN CENTRAL AMERICA’ PAUL R. FANTZ Fairchild Tropical Garden, Miami, Florida 33156 Present address: Dept. of Horticultural Science, Box 5216, North Carolina State University, Ralergh, NC 27650 Seymour (1978a) reports four new species of Centrosema from Nicaragua, viz. C. angustifolium (H.B.K.) Benth. C. plwmiert (Turp.) Benth., C. pubescens Benth, and C. virginianum (L.) Benth. However, a specimen of this genus deposited at the herbarium in Paris (P) which was examined in the course of a monographic study of the genus Clitoria (Fantz, 1977), does not accord with these or any other previously described species reported from Central America as listed in revisions and New World floras (e.g. Bentham, 1837, 1839; Standley, 1928; Standley and Record, 1936; Standley and Steyermark, 1946). The new species has larger flowers and bracteoles similar to C. plamieri, but is easily distinguished by the leaflet shape, longer calyx lobes, and the failure of the plant to turn black when dried. Nicaragua, the largest country in Central America, has but one published flora (Ramirez Goyena, 1909-1911), which is reported to be of little value (Blake and Atwood, 1942). A retired minister and botanist, Frank C. Sey- mour, at the age of seventy-three, began a personal cataloguing of the Flora of Nicaragua in 1968. He made seven expeditions to Nicaragua to study the flora firsthand and to collect representative specimens, plus purchasing a number of Nicaraguan collections made by other scientists doing research in that country. This project was financed from Seymour's own personal funds and through the sale of duplicate herbarium specimens. Seymour 8b) summarized his several publications on his expeditions and on the flora of Nicaragua. Now an octogenerian, Seymour's research is culminating in a manuscript which provides a checklist of the flora of Nicaragua. This new species of Centrosema is named in honor of Frank C. Seymour in recog- nition of his contribution to the botanical knowledge of Nicaragua. ou — CENTROSEMA seymourianum Fantz, sp. nov. Folia trifoliata, glabrata; foliola elliptico-oblonga, breviacuminata, infra sparsim Pverouic studies accomplished at the IFAS Herbarium, Dept. of Botany, Uni- versity of pen Gainesville, Florida. Florida Agricultural ee Station Journal Series No. SIDA 8(2): 152-156. 1979. 1399 puberula trichomatibus appressis secus nervos majories. Pedunculi axillares, floribus apicem pedunculorum congestis. Flores grandes, resupinati, 5.5-6 cm ae vexillum extus sericeo-villosum. Bracteolae ae ae calycem subaequantes, 10-20 mm longae, 4-5 mm latae. Calyx late campanule atus, tubo 4—5 mm longa; dentes supert et laterales tubo vix longiores, 5-7 mm longi; dens eae elongatus, 13-15 mm longus; calycis pubescentia inconspicua, in dente ventrali densior, trichomatibus uncinatis. Legumen incognitum. Scandent herb; stems terete, faintly angular, longitudinally striated, hollow, 2-3 mm thick, weakly twining, pubescence scattered, pilose and uncinate (observed latter at 30X). Leaves 3-foliolate, petiolate, glabrate; leaflets pin- nate, broadly elliptic-oblong, rapidly short-acuminate, apex mucronate, base rotund, primary nerves of 10-12 pairs, upper surface green, glabrous, lower surface concolorous, very sparsely pubescent with appressed trichomes along major nerves, lamina 12-13.5 cm long, 6-8 cm wide. Petioles subquangular- terete, densely pilose on pulvinus, sparsely so above, 5 cm long; rachis similar, ca 2.5 cm; petiolules quadrangular, pilose, 6-8 mm. Stipules ovate, striate, persistent, 4-6 mm long, 0.6-0.8 mm wide; stipels to 6 mm long. Inflo- rescence axillary, shorter than the petioles, several-flowered crowded at the apex of the peduncle; peduncles one or two per node, 2-2.5 cm. long, moderately densely pubescent, trichomes short, subappressed; pedicels paired, 8-9 mm. Bracts deciduous, ovate, striate, 6-7 mm long, 2-3 mm _ wide. Bracteoles enclosing calyx and flower in bud, large, coriaceous, striated, ovate- oblong, ciliolate, 10-12 mm long, 4-5 mm wide. Flowers large, papilionace- ous, resupinate, 5.5-6 cm long. Calyx broadly campanulate, upper teeth widely separated from lateral and lower teeth, sparsely pubescent on tube but more densely on teeth, particularly the retell tooth, trichomes uncinate, inconspicuous (observe at 30X), to 0.5 mm long; calyx tube 4-5 mm long, 4 mm wide at base expanding rapidly to 11 mm wide at throat; calyx teeth longer than the tube, upper teeth 5 mm long, 2 mm wide, lateral teeth 7 mm long, 1.7 mm wide, ventral tooth elongated, complicate, 13-15 long, 2.5 mm wide at the base. Vexillum broad, ca. 4.5-4.8 cm wide, compli- cate, densely sericeous-villosus on the outer surface, bearing a basal pouch ca 6 mm in diam. above the 4 mm long claw, and with a broad 1.8 mm sinus above the pouch and between it and a 3 mm long spur. Alae, carina, androecium, and gynoecium hidden within the complicate vexillum glued to the herbarium sheet, thus not observed. Legwme unknown TYPE COLLECTION: NICARAGUA. Bois, environs de Grenade, 40 m, Janvier 1870, Levy 419 (HOLOTYPE: P—plant “A” mounted on left side of ene: Fig The type specimen is mounted on a herbarium sheet along with another specimen labeled as “Herb. Mus. Paris, 1870, no. 20”, which is mounted on the right side and labeled "B” by this author. It is to be excluded as part of the type specimen of Centrosema seymourianum as this specimen belongs to the genus Phaseolus. The type specimen includes only the specimen mounted on the left-hand side of the herbarium sheet, and which has been Mayyy - WEARS, NES. Pads Kanter wcsrag unanes ede FPS : EX 7e MW 20 re 1. ce of Sa Giese seymonrianum (Levy 419, P—plant “A” mounted on ae left side of the herbarium sheet; excluding plant “B”’, Levy [?] 20, ee on the right side of i. ca sheet. ) 155 labeled “A”. A key is Bene below to the species of Centrosema reported to occur in Central America 1. Leaflets 1; petiole winged (Mexico to Colombia) Te sae 3; petiole not winged . Flowers small, 2—4 cm Tone: leaflets commonly small, 3—7.5 cm long; bracteoles —10 mm long. 3. Upper calyx teeth much shorter than the tube; leaflets with main lateral nerves divergent at right angles from the midrib (Honduras to Brazil) Me eh che ta A he Me Msi cic ta a P a alln ins aCsa fce ene caa bese eee let eee 3, Upper calyx teeth subequal to longer than the calyx tube; eee with ma lateral nerves ascending from the midrib. Lea ets anaes oblong or linear; legume short, weakly falcate (Costa Rica Cc a ee ete C. sagittatum to Brazil) oo cccccceeeee . paschorum pies ovate to oblong; legume long, straight. a Calyx teeth very unequal, upper and lateral teeth short, ca 2-3 mm long ay subequal to the length of the tube, lowermost tooth much longer than the tube length, 5-8 mm long; legume 10-20 cm long, 5-7 mm wide; leaflets ovate, pubescent on both surfaces at least in the juvenile state, veinlets somewhat inconspicuous (Mexico to Brazil, W. Indies) Be ct te laid s adh sida etttin ntidlntttns allen at sf a ee ae Tee . pubescens y: Calyx teeth acai all much longer than the length of the abe. teeth ca 6-9 mm long; legume 6-12 ong, 4—5 mm wide; ee ovate to oblong or linear- ae ae veinlets eonspiciously % sed, promi- nent eee United States to Argentina, W. Indies, topical Africa) ere hs C. virginianum 2. Flowers large, 4-6 cm long; leaflets commonly large, (5) 7-14 cm long; bracteoles 10-20 mm long. 6. Leaflets broadly ovate to rhombic-ovate, often drying black; flowers 4—5 cm long; bracteoles exceeding calyx by two to three times calyx length, ca. 15-20 mm long; calyx tube 5-7 mm with obsolete to short teeth to 2 mm long; legume 9-10 mm wide (Mexico to Brazil, W. I Indies, a, Africa ) Aon plumiert 6. Leaflets oblong to elliptic- oblong ¢ or ovate-lanceolate, not deine Giaee flowers 5S-G cm long; bracteoles subequaling length of cal tube and lowermost tooth, 10-13 mm long; calyx tube 4-5 mm long with conspicuous teeth sub- equal to longer than the tube Icngth; legume 7-8 mm wide. 7. Leaves ovate-lanceolate, subpilose below with trichomes spreading on the nerves; calyx sericeo-pilose, trichomes 0.5-1 mm long, conspicuous when viewed at 10X; vexillum gibbous (Mexico) .. C. galeottii . Leaves elliptic-oblong, glabrate with very sparsely appressed trichomes on the nerves; calyx Soeee with scattered uncinate trichomes (to 0.4 mm long), sparse on the tube and more densely compacted on hens al lobes, conspicuous when viewed at 30X; vexillum spurred (Nicar ~~ ie seymourianum Acknowledgement is made to Helen Correll for the Latin description > CENTROSEMA galeotii Fantz, ‘ig nov, Bastonym: Clitoria grandiflora Mart. & Gal., Bull. Acad. Brux. 10: 189. se ee Galeotti 3284 (BR!). Non Geaerens ee Benth. of Brazi 156 and to Daniel B. Ward, Dana Griffin III, and Donovan Correll for reviewing the manuscript and providing constructive criticisms. REFERENCES BENTHUM, o 1837. ae entationes de Leguminosarum Generibus. Ann. Wein. Mus. Natur. 117-120. ie De (ae Generibus Commentations. Ann. Wein. Mus. Natur. 2: 117-120. BLAKE, S. F. and A. C. ATWOOD. aoe: 1: ae ce States Government Printing Office, Washington FANTZ, P. R. 1977. A monograph of the genus Clitorta Cesuminose: 1066 pp. aaa Thesis, University of Florida. RAMIREZ GOYENA, M. 1909-1911. Flora Nicaraguense. 2 vol. 1064 pp. Managua. SEYMOUR, F. C. 1978a. A checklist of the vascular plants of Nicaragua. (Unpub- lished manuscript and personal communication ). 978b. we in Nicaragua. Phytologia 39(6): 500. STANDLEY, C. 28. Flora of the Panama Canal Zone. Lae U.S. Nat. Herb. 27: 215- 1942. Geographical guide to the floras of the Glycineae ). ae Flora of Costa Rica. Pub. Field. Mus. Nat. Hist. Bot. Ser. 18(2): and S. J. REC — Bie ce = and flora of British Honduras. Pub. Field. Mus. Nat. Hist. Ser. 12 and J. Se ee a Ce of Guatemala. Fieldiana Bot. 24(5): 178-181. HALOGETON (CHENOPODIACEAE) IN NORTH AMERICA’ WILL H. BLACKWELL, J. DOUGLAS HAACKE, AND CHRISTINE O. HOPKINS Department of Botany, Miami University, Oxford, Ohio 45056 The history, nomenclature, and geography of Halogeton glomeratus (Stephan ex Bieberstein) C. A. eyer in Ledebour in North America have been investigated with attention given to both literature and herbarium specimens. Halogeton received intensive study in the United States between 1942 and 1961 because of its role in the mass mortality of sheep (oxalate poisoning) on western rangelands. Although it is apparently no less com- mon today in the Intermountain Region, sheep mortality has been reduced through educational efforts, and interest in this plant has declined somewhat in recent years. Confusion has existed in the literature, but a herbarium study indicates only one species of halogeton present in the United States now or historically and its identity is confirmed as Halogeton glomeratus. Halogeton was introduced into the United States (probably from south- central Russia) earlier than previously reported (ie. before 1930), live animals (Karakul) being the probable source of introduction. A number of the unnoticed halogeton infestations in the United States were likely initiated by migrating sheep. This species usually does not self-spread over long distances a should only gradually, at most, extend its present range. t will probably not become as widespread, nor represent as great a threat to western sheep flocks, as once was thought. HISTORICAL PERSPECTIVE Halogeton, an annual Eurasian chenopodiaceous genus, might well have remained an obscure weed but for the alarm it brought when large numbers of sheep were poisoned in the western United States. Although present in the United States at an earlier date, substantial awareness of halogeton did not come until 1942 when 160 head of sheep were fatally poisoned near Wells, Nevada (Erickson et al., 1952; Flemming et al., 1944). Stomach contents of these dead animals were composed almost entirely of this plant "This work was supported in part by a grant (W. H. Blackwell, Principal Investi- gator) from the American Philosophical Society (No. 7991—Penrose Fund) and from the Willard Sherman Turrell Herbarium Fund of Miami University. We seed fully acknowledge curators of the herbaria listed for loaning eee and/or pro- viding information. We also wish to acknowledge Dr. Charles Perino for his assistance in gathering some of the specimen and literature information alluded to in this paper. SIDA 8(2); 157-169, 1979. 158 material. Subsequent feeding tests performed at the University of Nevada confirmed the strong circumstantial evidence of the poisonous nature of this plant (Flemming et al., 1944). The mass poisoning near Wells led to first attempts (in 1943) to control the weed, a mixture of furnace oil and detergent being employed (Erickson et al., 1952). This and subsequent con- trol measures were largely unsuccessfu Whereas halogeton was known (at least by a few people) to be present in the United States before 1942, it was not thought to be harmful (Kings- bury, 1964). In fact, it was even considered by some (Erickson et al., 1952) as fair forage for livestock. The casual regard in which halogeton poisoning was initially held in this country was doubtless due to the fact that there were apparently no reports of its toxicity in the Old World (Dayton, 1951), and thus no special reason to suspect it. However, suspicious small losses of livestock in the western United States had actually preceded the Wells incident (Kingsbury, 1964). After the major poisoning at Wells, numerous minor losses of both sheep and cattle were soon reported in which halogeton was implicated (Erickson et al., 1952). However, a wider recognition of this taxon as a poisonous range plant was to wait until 1945 when 1620 head of sheep perished in a single day near Bridge, in south central Idaho (Morton et al., 1959a). Other severe losses in this Raft River area (Cassia County) included 750 head of sheep belonging to a Mr. John Ward of Almo, Idaho (ZoBell and Silcock, 1950). In fact, the problem became so serious in the mid 1940's that a dozen major sheep ranchers in the Raft River Valley were forced out of business (Stoddart et al., 1951b), finding it necessary in some cases to abandon their ranches. These and other events brought the noxious weed to the attention of the public by 1950, and generated an expansion of re- search by personnel of the U. S. Department of Agriculture. Popularized reports often took the form of alarming accounts of the spread of this “invader,” sometimes with not so subtle innuendoes as to the country of native origin. The following are typical of the frightening titles or inflama- tory headlines of the early 50's: “Sheep- Killing Weed,” Life Magazine, Jan. 15, 1951; “Poison Rides the Range, ’ Reader's Digest (condensed from the Denver Post), December 1953; “Another Russian Invades US.,” the Na- tional Wool Grower, 1950; and “Soviet Weed Kills Sheep,” Oakland Tribune, June 28, 1950. The Life Magazine article contained the statement that more than one-third of the flocks of sheep in the United States were in danger. It was understandably at this time that several agricultural colleges, univer- sities, and range extension agencies in the western states took an active interest in halogeton, initiating various studies into its ecology and possible control. Professional interest was stimulated when the United States Congress appropriated over four-and-a-half million dollars for study in legislation (Halogeton Act, Velie, 1953). Co-operative projects and committees were established in Idaho (Erickson et al., 1952) and Utah (Stoddard et al., — 159 1951b) to investigate its biology and possible eradication. Warnings of the danger of halogeton appeared in agricultural extension service bulletins of several western states in the early 50's, including Arizona (Armer, 1952) where its occurrence has yet to be documented. Besides the usual admonitions of its effect on sheep, of possible habitats in which it might occur, that it is not controlled selectively by 2,4-D or similar herbicide sprays, and recom- mendations for trailing sheep so as to avoid it, these bulletins usually de- scribed the plant. Particular emphasis was given to means of distinguishing halogeton from the less dangerous Russian thistle (Salsola kali L.), another introduced chenopodiaceous plant with which it was sometimes confused. For example, Durrell (1951) stated: “[halogeton] resembles Russian Thistle, which is common throughout the country. However, it has a softer appear- ance, its leaves are sausage-shaped, succulent and fleshy. The bristle at the tip is softer than the spine on Russian Thistle leaves.” The toxicity and precise effects of halogeton on animals received addi- tional attention during the late 1940's and early 1950's. Flemming et al. (1947), who had earlier (1943) reported halogeton as poisonous (contain- ing oxalic acid), indicated the toxic substances to be “salts of oxalic acid which are elaborated and stored during the growth of the plant to maturity,” present as two oxalate salts, water soluble and insoluble. Vawter (1950) subsequently described the mechanism, symptomology, and pathology of halogeton poisoning. Ingestion of only nine ounces will kill a large, mature sheep (weighing 135 to 165 pounds) within six to 10 hours due to “rapidly overwhelming loss of inorganic serum calcium.” Vawter (1950, 1951) listed the symptoms as dullness, prostration and coma. Kidney sections examined under polarized light show large crystal aggregates of calcium oxalate in the tubules. Anderson and Huffman (1957) indicated these changes resulted in acute renal failure. According to Vawter (1950) there “seems to be no practical way of saving sheep after eating a lethal dose of the plant.” The toxic form of oxalate was determined by Dye (1956) to be the soluble form, insoluble calcium oxalate being relatively harmless. Binns and James (1961) discussed the mechanism of sodium oxalate in the blood in pre- cipitating calcium (with resultant alkalosis) and stated that the feeding of alfalfa pellets, supplemented with dicalcium phosphate, was helpful in preventing (not curing) poisoning in sheep. More detailed study of the physiology of halogeton, particularly oxalate content and metabolism, was the concluding phase of an intensive two- decade study. Morton et al. (1959b) determined the oxalate content of plants to range from 18 to 21 percent of dry weight (about the usual range reported in the literature), declining somewhat during the growing season. Most of the oxalate was contained in the leaves, with relatively small amounts in the stems, and high oxalate concentration was associated with a high sodium content. Williams (1960) observed that soluble oxalate content of the plant, as well as growth and succulence, increased when sodium (as 160 NaCl) was added to nutrient solutions. In contrast to Morton et al., Williams found that the concentration of oxalate in the leaves continued to rise during the summer, attaining a peak in September. According to Walliams, the leaves of vigorous plants often contain in excess of 30 percent soluble oxalate on a dry weight basis, with an additional three or four percent of the dr weight of the plant accounted for by insoluble oxalate. A dry-weight soluble oxalate content of 34.5 percent was reported by Cook and Gates (1960), a high percentage compared to most other oxalate-containing plants. Kings- bury (1965) indicated a 10 percent soluble oxalate content to be high for beet tops (considered potentially dangerous forage), well below the usual estimates for halogeton. Both popular and scientific liceracure on halogeton decreased after 1961, although Kingsbury (1964) devoted attention to it in his comprehensive work on poisonous plants of the United States and Canada. Sheep losses from oxalate poisoning also declined, apparently from an awareness among sheep ranchers. Nonetheless, occasional reports of severe loss, particularly in Utah, continued to surface in various newspapers. As recently as 1971 (Associated Press, Salt Lake City, January 23), 1,180 head from a flock of 2,400 died near Garrison, Utah, close to the Nevada border. This report observed that “there’s halogeton all over that area.” Thus, a “slow down” in the literature probably did nor reflect a decrease in populations, but rather indicated better education concerning the plant, a lessening of the earlier somewhat faddish interest in the weed, and the realization that halogeton would not actually kill one out of three sheep in the United States (as suggested by Life Magazine). Questions left unanswered or inadequately answered include: When did halogeton enter the North American flora and by what means? What has been the pattern and the mechanism of its spread? Is 1t more common now than it was say 30 years ago? What is its probable future in the flora? Is more than one species of Halogeton present in the United States? And, per- haps surprisingly, what is actually the correct scientific name(s)? By taking a close look at the history of the literature on halogeton, and by examination of actual specimen records, this paper attempts to deal with these questions. NOMENCLATURE Since its appearance in the flora, three different names for halogeton in North America have found their way into the literature: Halogeton sativus, H. souda, and H. glomeratus. These three names are to be found among specimens in North American herbaria (although H. glomeratus is the common determination ). In 1936, the 77th Intermountain Forest and Range ee Station collection was forwarded to the Washington Office of the United States Forest Service (Dayton, 1951). These plants included two specimens (S-43 and §-44, collected in 1934 by Ben Stahmann, a Forest Service officer, one 161 mile northeast of Wells, Elko County, Nevada) which were the first records of halogeton for the U.S. These two specimens were not immediately identi- fied. However, similar material (Field Museum, F 780344 and F 815074) collected by Mr. Stahmann from Wells in 1935 was subsequently deter- mined by Dr. Paul Aellen, an authority on the Chenopodiaceae, as “Halogeton sativus’ (based on Salsola sativa L., 1762), native to Spain and Algeria. Standley (1937) first published on halogeton in North America, accept- ing Aellen’s determination of Stahmann’s 1935 material as Halogeton sativus (L.) C. A. Meyer. Standley, seemingly unaware of Stahmann’s 1934 collec- tion, commented that the genus was first collected in “1935” at Wells, and speculated that “it will be interesting to learn whether this new addition to the United States flora thrives as well as Bassia hyssopifolia ( Pall.) Kuntze, a related plant of somewhat similar appearance, which, if I remember cor- rectly, also was found in the United States first in Nevada.” Standley further stated that “it [halogeton} is reported to be spreading rapidly.” Confusion persisted in the literature as to whether the initial collection of halogeton was made in 1934 or 1935 (e.g. ZoBell and Silcock, 1950); however, her- barium records clearly show that 1944 is correct. The question of the initial year of collection has not been the only con- fusion enshrouding halogeton in North America. Morton (1941) correctly observed that the publishing author of the combination Halogeton sativus was actually Moquin-Tandon (1840), rather than C. A. Meyer (as suggested by Standley and presumably also by Aellen). Although Meyer (1829) mentioned Salsola sativa, he never employed the combination Halogeton sativus (as did Moquin-Tandon ). Therefore, the correct citation is Halogeton sativus (L.) Moqguin-Tandon, not (L.) C. A. Meyer. However, Morton (1941) went on to discuss the existence of an earlier name for this species, viz. Halogeton souda (Loefling) Macbride (based on Salsola souda Loefling, 1758). Our examination of photographs of the type specimens of both Salsola sativa and S. souda in the Linnaean herbarium (LINN) confirms that they are probably the same species. Thus, if the adventive American taxon were considered to be the Spanish-Algerian species, Halogeton souda would be the correct name (on the basis of priority), with H. sativus relegated to synonymy. Fosberg (1940) used the name H. sowda for halogeton in Nevada. The question of Halogeton sativus vs. H. souda is relatively moot to our considerations, however, because Morton (1941) further indicated that North American specimens are not Halogeton souda (H. sativus) at all but belong instead to a Russian species, Halogeton glomeratus. It 1s apparent that Morton examined Old World specimens of both H. sowda and H. glomeratus, although probably not the type specimen of either. Dayton (1951) subsequently offered a brief review of these taxonomic and nomen- clatural problems, concurring with Morton. However, Dayton stated that the type of H. glomeratus was behind the “Iron Curtain,” presumably in Moscow or Leningrad, and not available oe examination. Thus, ultimate 162 certainty of the identity of North American specimens must have eluded Morton and Dayton. Through the auspices of Dr. V. I. Grubov, we obtained a loan of speci- mens from the Leningrad herbarium (LE), along with photographs (nega- tives) of the type specimen of H. glomeratus. Prints made from these nega- tives matched well with the specimens collected in the United States. Thus, North American specimens can now be classified with reasonable assurance as H. glomeratus, not H. souda (H. sativus). Dayton (1951) and others indicated that the combination Halogeton glomeratus is traced to C. A. Meyer (1829) when he described the genus in Ledebour’s Flora Altaica. Dayton, however, noted that C. F. Stephan (author of a Flora of Moscow) was actually the first to use the epithet (in connection with herbarium material) but that Stephan himself did not pub- lish it. Stephan’s specimen, however, is the type of Halogeton glomeratus, and is the specimen in the Leningrad herbarium. This specimen (lacking a date) is labeled “Salsola glomerata’ in Stephan’s handwriting. The name written by Stephan is a herbarium or “manuscript” name. However, Stephan’s epithet “glomerata’ was later validly published by Marshall von Bieberstein (a colleague of the well-known Russian botanist, Pallas) as Anabasis glomeratus (1806). Meyer (1829) subsequently transferred Bieberstein’s glomeratus to Halogeton. The most complete and accurate author citation of the name (which we have understandably not seen published ) would be Halogeton glomeratus (Stephan ex Bieberstein) C. A. Meyer in Ledebour. Tutin (1964) considered Halogeton glomeratus and H. sativus (= H. souda) to be closely related species. In addition to the geographical separa- tion, the Spanish species is to be distinguished by longer leaves, a more robust habit, longer perianth, and a greater number of stamens per flower. Our examination of Specimens and phototypes lends support to those who eae these two species. However, if a future monographer were to com- ne H. glomeratus and H. souda, the name would then become H. souda on the basis of priority. —_ Halogeton, as is the case with many other chenopodiaceous genera, is in need of revision on a worldwide basis. This is evident from the variation in estimates as to the number of species of which the genus is composed: Ulbrich (1934) indicated that there are three species; Ijin (1936) called for four; and, somewhat inexplicably, Dayton (1951) suggested that there are 12. Regarless of whom one believes, there is agreement that more than one species exists. However, we consider that only one species (H. glomera- tus) has been and is in the flora of the United States. We find no evidence to support Dayton’s (1951) speculation that “it is perhaps only a question of time when one or more of these other species will emigrate to this country.” 163 INTRODUCTION, SPREAD, AND DISTRIBUTION Neither the manner nor the date of introduction of halogeton into the United States has been definitely established (Tisdale and Zappetini, 1953). Several references (Bellue, 1949; Standley, 1937; ZoBell and Silcock, 1950) indicated halogeton first entered the North American flora in 1935, although it was already common in the Wells area by 1934 (Dayton, 1951). Ic is apparent in the case of several “first records” for particular regions that infestations were extensive at the time of discovery (Bellue, 49; Velie, 1953). Even the nature of various of the livestock losses reported, i.e. severe and sporadic, is an indication that it became abundant in certain areas prior to recognition. Stoddard et al. (1951b) set the time of entry back somewhat in suggesting the probable date of introduction from southern Russia “in about 1930.” Binns and James (1961) likewise favored 1930 as a likely date of introduction. Regardless of the exact time of introduction, it appears that halogeton became more common in western North America in the early 1930's. During this period an extremely dry climatic circumstance prevailed throughout most of the West. It is probable that the drought tolerant halogeton extende its populations (at least locally) while many native species suffered decline (Tisdale and Zappetini, 1953). This species does not compete well with healthy native perennial forbs (Fenley, 1952), yet it enters in force when soils are denuded (Erickson et al., 1952). As many as 50,000 winged seeds per plant (Morton et al., 1959a) are Ae locally by the wind (Kings- bury, 1964). This and a wide ecological tolerance range permits se to survive in habitats unfavorable to competitors (Erickson et al., 1952) and to “take over” local disturbed or barren areas quickly. The “Dust Bowl” of the western United States in the 1930's assuredly contributed to its spread. Although undocumented, halogeton must have been present in North America before 1930 (Erickson et al., 2). Erickson and his coworkers suggested three possible sources of Pees 1) impurity in crop seed, 2) imported wool, or 3) importation with breeding stocks of sheep. Durrell (1951) and an anonymous writer in Life Magazine (1951) supported the introduction into this country as a contaminant in imported seed. Erickson et al. (1952) favored wool pelts or live animals as the vector, pointing out that new infestations coincide with sheep movements. Bellue (1949) simply commented that “Halogeton seed seems to have an affinity for and readily sticks to the wool of sheep,” and that in Nevada, Utah, Idaho and Wyoming, the weed has “followed sheep trails and bedding grounds.” Karakul ed among other possibilities, has been mentioned as a potential source of i troduction and the most likely vector from the Soviet Union. In contrast to other breeds (including Angora), the natural range of the Karakul corre- sponds closely to that of halogeton. Karakul are native to desert-like areas e Bokhara Province of southern Russia, and sheep with Karakul blood extend as far westward as the Caspian and Black Seas (Vaughan, 1931). 164 Halogeton Ses is well known from this general area (Ijin, 1936). Erickson et al. (1952) mentioned that both Karakul and ae. breeding stocks were oe into the United States between 1908 and 1925. Went- worth (1948) also alluded to the importation of Karakul early in the twentieth century. Vaughan (1931) identified “three lots” of Karakul that were imported (by Dr. C. C. Young), in 1908, 1912, and 1914 respectively. If Karakul were the vector, then 1908-1914 would be more likely the time of introduction than 1930 to 1934. Circumstantial evidence for the ovine origin of halogeton infestation 1s found in the implicating comparison (Figure 1) between sizeable halogeton populations and major sheep trails of the early part of the century (these trails were in heavy use until about 1915). Several primary routes were used in sheep drives out of California (Wentworth, 1948), including south- ern (leaving Bakersfield), middle (leaving Fresno), and northern (leaving Red Bluff) passages. The southern and middle routes passed along the eastern flank of the Sierras, then across Nevada, proceeding through Wells (a major crail junction), and up into the Raft River Valley of southern Idaho. Tisdale and Zappetini (1953) pointed out that the Raft River Valley area was one of the principal winter ranges for sheep in the early 1900's. The northern route passed directly through the well-documented area of infestation in Lassen County, California, on the way to Wells and southern Idaho. Thus, the correlation between major sheep trails and a number of significant halogeton infestations is, in our Opinion, not coincidental (Figure 1). Localities plotted are based on specimens examined from the following herbaria (Holmgren and Keuken, 1974): A, ARIZ, DAO, F, GH, ID, MU, NY, OSC, PENN, PH, RENO, RM, SMU, UC, US, USFS, UT, and WTU. Additional information on specimens was provided by: CAN, COLO, TEX, and UNM. We consider, by virtue of the native home of the Karakul, its long fur (Persian lamb’s wool) in which seeds might easily become entangled, the time of introduction of Karakul into the United States, and the apparent relationship between old sheep trails and the distribution of the plant, that halogeton was both introduced and subsequently dispersed by live sheep. Introductions by means of fur pelts or grain shipments are conceivable, but not as probable. Karakul skins were mailed in bundles of 500 each from Asia directly to New York City where they were processed (Vaughan, 1931), and therefore had no connection with halogeton growing in the western states. Grain shipments contaminated with halogeton seed could have ac- counted for a few populations of halogeton. Seed contamination has played a major role in the introduction of certain other chenopodiaceous weeds (Blackwell, 1978; Collins and Blackwell, 1979). But if the majority of infestations had actually been caused by contaminated grain, there should be no correlation between its occurrence and well-known sheep trails. Also, railroads for shipping grain were not established early in the 1900's through Figure 1. Distribution of Halogeton glomeratus in North America. Stars with Halogeton populations based on herbarium specimen records. Sheep trails (double broken lines) are modified from E. N. Wentworth, 1948, America’s Sheep Trails. 166 some of the areas of heaviest infestation. Thus, live animals, in particular the Karakul, would seem to have been the vector, and we believe halogeton to have been present and growing in the United States by 1915. It is some- what ironic that the type of animal (sheep) which has been most adversely affected by its toxicity also appears to have been the source of introduction and dispersal. The assumption (e.g. Stoddard et al., 1951b; ZoBell and Silcock, 1950) that halogeton was first introduced into Nevada, from there spreading to the other states (Utah, Idaho, Wyoming, etc.), may be invalid. Even though it spreads locally in great numbers, there is no evidence that it is dispersed across large distances without the aid of an external vector (such as sheep ). Kingsbury (1964) indicated that wind is primarily effective only in local dispersal. Thus, a majority of present extensive areas of infestation probably correspond to the original sites of introduction by migrating sheep. While Wells, Nevada, was one of those early sites, it is mot possible to be sure that it was the earliest. It is reasonable to say only that it was first collected there. It is conceivable, for example, that one or more of the Raft River Valley localities in Idaho might predate the Wells, Nevada, occurrence. There is no proof one way or the other. Since halogeton was present for a number of years, probably in a number of locations, prior to its discovery, it is not likely that the first site of occurrence in this country will ever be conclusively identified. But even if Wells were accepted as the initial locality (as it is in much of the literature), an autogenic spread of halogeton from there to other widely separated localities is less than plausible. Bellue (1949) inferred that halogeton migrated from northwestern Nevada (located across the state from Wells) over the border into Lassen County, California. Yet, this rather isolated occurrence in northeastern California may have had nothing to do with its presence in Nevada. The California infestation could just as well have been initiated in situ in the early 1900's by a sheep drive. Bellue stated that “a survey made subsequent to our first California record in Lassen County in 1946, indicated that Halo- geton had become scattered over several thousand acres.” This extensive infestation did not develop overnight and it seems as feasible to us to specu- late that halogeton had been there for some time (the introduction having taken place at or near the site) as to postulate a recent invasion from adjacent Nevada. We believe that, whereas it is possible to determine a pathway of infesta- tion (and a probable cause), it is not possible to establish meaningful chronology. Halogeton was, we think, independently and separately intro- duced by sheep into scattered localities, sooner or later becoming locally common in some of them; and since these introductions probably took place well before discovery of the plant, the order in which they transpired can- not be determined. If it were possible to establish the chronology of in- festation, however, it is unlikely that it would follow an orderly geographic 167 sequence. Based on herbarium specimen records, and without any necessary implication of the sequence of spread, we can state that the first collections known in the following states were in the following order: Nevada, 193 Utah, 1943; Wyoming, 1943; Idaho, 1946; Montana, 1951; California, 1951; Colorado, 1951; Oregon, 1954; and New Mexico, 1973. Halogeton was ob- served five years prior to collection in California (Bellue, 1949) and at least one year before collection in Idaho (Morton et al., 1959a). Thus, specimen records offer only a rough gauge of the time of awareness of its presence in the various states, and little or nothing about the actual time of its appear- ance in the state's flora. Opinion has varied as to the prognosis for halogeton in North America. Holmgren (1950) stated that “it’s migration into new areas is continuin at an alarming rate,’ and such a sentiment is echoed in popular articles in Life Magazine (anonymous, 1951) and Reader’s Digest (Velie, 1953). More conservatively, Erickson (1952) stated that “the future distribution of halo- geton cannot be forecast with accuracy at this time.” Erickson did mention the possibility that it could eventually spread over the entire Intermountain Shrub Zone (Cold Desert region) and even into the western margins of the grasslands lying to the east. Perhaps it was simply too early, and with too little information at hand, to forecast accurately the future of halogeton in the United States, except to postulate that it would certainly persist (Stod- dart et al. 1951la) and probably would not become any less common with time. It appears that the present major areas of halogeton infestation are essentially those aca in publications of the 1950's. Simple maps pre- sented by Erickson et al. (1952) and the U. S. Poe of Agriculture (anonymous, 1958), outline these areas rather well. To the USDA map, new localities can be added primarily in southern Oregon ia northwestern New Mexico. According to herbarium records, no new “halogeton states,” except for New Mexico, have been added to the list since 1954. Thus, it would seem that whereas halogeton can and does spread locally at a rapid rate, it does not readily establish new widely disjunct areas of infestation by its own devices. Bellue (1949), for example, noted that “the range o Halogeton is still confined to the original general location in the southeastern corner of Lassen County.’ And in the summer of 1977, two of us (Blackwell and Hopkins) carried out an extensive search throughout adjacent Modoc County, California, without turning up a single specimen. Suitable habitats were apparently abundant in Modoc County, yet halogeton had not sprea there in more than 30 years of opportunity from either Lassen County or adjacent Nevada. Morton et al. (1959a) indicated that the total area of halogeton infestation was 10.5 million acres (though it is not clear how this figure was derived). It does not seem that the area of infestation today is much larger than this, although an accurate estimate is lacking. en com- pared with the USDA map of 1958, major areas of infestation do not 168 appear to have increased strikingly. This is not to imply that there have not been numerous instances of local increase, or numerous rather closely spaced small new infestations in various parts of the Intermountain Region. Undoubtedly, the fact that major sheep drives have drastically declined as a means of transporting sheep great distances is related to the present minor spreading of the plant. New areas of halogeton infestation now perhaps most commonly originate in disturbed habitats along roadways (personal observa- tion, Blackwell, southern Oregon, 1977). Infestations in eastern Utah and western Colorado, New Mexico, and Oregon apparently developed in this manner. Thus, spread (not necessarily rapid) along roadsides allows move- ment into poorly managed rangeland. Control of halogeton remains a concern. Of the various alleged methods, biological control has come to be among the more common suggestions. In particular, the use of crested wheatgrass (Agropyron cristatum Gaertn.) as a competitor was recommended (Fenley, 1952; Stoddart et al., 195la) However, this, as various chemical control measures, produced dubious re- sults (Morton et al., 1959a). Halogeton and crested wheatgrass were observed growing successful ly together in Lake County, Oregon (Blackwell, personal observation, summer 1977). Perhaps Morton and his collaborators (1959a ) summarized it best in stating that “the successful control of halogeton does not depend upon any single control method . . . Ultimate control of halogeton depends upon the restoration of vigor and productivity to good range forage plants REFERENCES ANDERSON, W. A. and W. T. HUFFMAN. 1957. Halogeton poisoning in a ewe. J. Am. Vet. Med. Assoc. 130: 330-331. ANONYMOUS. 1951. Sheep- killing weed. Life Magazine, Jan. 15: 55-56. ANONYMOUS. 1958. Reducing livestock losses from halogeton poisoning in the western states. U.S. Dept. Agri. Bull. PA-321 ARMER, W. . tac a stock- poisoning weed, Ariz. Agri. Exp. St. Bull., Ext. Circ. 197. BELL i M. K. ioe Halogeton invades California. Calif. Dep. Agri. Bull. 38(1): 32-306. BIEBERSTEIN, F. A. M. VON. 1806. ieee Societe Imperiale des Naturalistes de Moscov, Memoires, vol. 1 (ed. 2): 108-1 BINNS, W. and L. F. JAMES. 1961. a ee and other oxalic acid poisonings. Proc. Am. Col. Vet. Toxicol. 1960: 5-8. BLACKWELL, W. H. 1978. The history of Russian pigweed, Axyris amaranthoides (Chenopodiaceae, Atripliceae), in North America. Weed Sci. 26: 82-84. COLLINS, S. L. and W. H. BLACKWELL. 1979. Bassia (Chenopodiaceae) in North America. Cg 8: 57-64. COOK, C. W. and D. H. GATES. 1960. a of site and season on oxalate content ie ie. J. Range Manage. 13: 97— YTO W.A. 1951. cae nar o barilla (Halogeton glomeratus). J. eee Manage. 4(6): 375-38 DURRELL, L. W. 1951. Ee ee a new stock-poisoning weed. Colo. Agri. Exp. St. am 170-A. 5 pp 169 DYE, W. B. 1956. Chemical studies on Halogeton eahitrs Weeds 4: 55-60. ERICKSON, L. C., H. L. MORTON, E. W. TISDALE, and G. ZAPPETINI. 1952. The present status of the halogeton problem. Weeds 1(4): aoe 328. FENLEY, J. M. 1952. How to live with halogeton while limiting its spread and_re- ducing loss. Nev. Agri. Ext. Serv. Bull. 106. 20 pp. FLEMMING, C. E.,, M. R. MILLER, L. R. VAWTER, and W. NEILSON. 1943- 1949. Poisoning range plants. Nev. Agr. Exp. St. Ann. Rpts. FOSBERG, F. R. 1940. Chenopodiaceae of Nevada (in part). Contrib. Toward a Flora of Nevada, No. 8: 1-18. Mimeographed. oe A. 1950. Halogeton ee eastern Utah winter ranges. Utah Agri. , St., Farm Home Sci., 11(4 foe P.K. and W. carat er Index herbariorum (6th ed.). Regnum Veg., Vol. es Oosthoek, Scheltema and Holkema, Emmalaan 27. Utrecht, Nether- lands, 397 i M. 1936, Sree In V. L. Komarov and B. K. Shishkin. Flora of the R, 6: 4-272. (English Translation, 1970, by N. Landau. ) ane J. M. 1964, Poisonous plants of the United States and Canada. Prentice Hall, Inc., Englewood Cliffs, New Jersey. 626 pp. er ctale J. M. 1965. Deadly harvest. Holt, Rinehart and Winston, New York. eon e A. 1829. Halogeton In C. F. Ledebour. Flora Altaica, 1: 378-380. MOQUIN-TANDON, C. 1840. Cheaopodearn monographica enumeratio. 182 pp. MORTON, C. V. 1941. A note on u paloget Leafl. West. Bot. 3(2): 45-46. MORTON, H. L., R. H. HAAS, and L. C. TFRICKSON. 1959a. Halogeton and _ its control. Idaho Agri. Exp. St. Res Bull. 307. TON, H. L., R. H. HAAS, and L. C. aa om 1959b. Oxalate and mineral contents of Halogeton glomeratus. Weeds 7(3): 255-264. STANDLEY, P. C. 1937. Studies of Arete plants—VIII. Field Mus. Nat. Hist. Publ., Bot. air hee 225-284. (Publ. 395 ia STODDART, G. T. BAIRD, G. STEW RT, R. S. MARKHAM, and ses oan ‘the halogeton problem in ce Utah Agri. Ext. cee Bull. 050 cnn L. A., H. CLEGG, B. 5S. ee se G. rears 1951b. The halogeton problem on Utah’s ranges. J. Range Manage. 4 ( 223-2 TISDALE, E. W. and G. eee 1953, ae a on 4 aho ranges. J. Range Manage. 6(4): 225-236. TUTIN, T. G. 1964. ee. In Tutin et al. Flora Europaea, 1: 90-108. ULBRICH, E. 1934. Che pee In Engler and Prantl. Die Natiirlichen Pflanzenfamilien (ed. 2), 16c: 379-584. VAUGHA N, H. ue ee ee of livestock in America. R. G. Adams and Co., cattle. “Calif. Vet. 316). VAWTER, L. R. 1951. Teor poisoning in sheep. Natl. Wool Grower 41(2): 24-25. VELIE, L. 1953. Poison rides the range. me Reader’s Digest. December: 79-82. Soles from the Denver Post, Nov. 8, 1953) WENTWORTH, E. N. 1948. America’s sheep Gaile: Iowa State Univ. Press, Ames. 667 pp. WILLIAMS, M. C. 1960. Effect of sodium and Fea salts on growth and oxalate content of halogeton. Plant Physiol. 35: 500-505. ZOBELL, R. S. and ae W. SILCOCK. 1950. nes Russian invades U.S. Natl. Wool Grower 40(4): CHANGES IN GALACTIA (FABACEAE) OF THE SOUTHEASTERN UNITED STATES WILBUR H. DUNCAN Department of Botany, University of Georgia Athens, Georgia 30602 ABSTRACT ifferent names and realignments are presented for several species of Galactia. It is shown that G. volwbilis (L.) Britt. and a macreei M, urtis sensu most recent authors should be called G. regularis (L.) BSP and G. Were (L respectively. G. glabella Michx. is supported as the sean name for what has B . reg G. glabella is shown to of three partly sympatric forms but studies are insufficient to decide what if ¢ uld be assigne : taxonomic status, if any, sho be igned to diagnostic key for eight species and distribution maps for six species are presented. More detailed diagnoses y are presented for better understanding of G. regularis glabella, volubilis, and mollis, specimens of which have been abundantly misidentifie Galactia in the Southeastern United States has interested me for at least 30 ears. Intermittently I have attempted to solve some of the many complex problems inherent in this genus, but only recently have sufficient data and con- aoe been accumulated to justify their presentation. First I described one new species (Duncan, 1977). I am now presenting my remaining con- clusions, fully aware that some may rot hold up under future studies but also realizing the unlikelihood of my making further studies of this subject. Taxa confined to peninsular Florida are omitted due to insufficient study of materials from there. A key to all other species of the Southeastern United States as I interpret them follows. KEY TO SPECIES 1. Leaves with 5—9 leaflets Coe ee eee eG. ebliottii 1. Leaves with 3 leaflets. Se we © ay eee ee ER Se ee 2. Plants erect, pps sulseeseile G. erecta 2. Plants twining or trailing, inflorescences pedunculace 3 . Calyx lobes ana to reddish brown on inner surface when dry; ‘corollas dark colored when dry, persisting after ha sometimes partly present wl fruit is mature G. mollis Calyx lobes preenich: cele to tan on inner bern ace when ays gonollas an pare when dry, falling as they wither or soon thereafte 4, Stems with fine close mostly ieee hairs, the ee. 0. 7 ge or enge: S>) 7. floridana 4, Stems with fine antrorse, retrorse, or spreading haits under mm rarely a few longer, but hairs not dense 5. Hairs on stems antrorse . 6. Internodes only a little longer to usua lll suis than thet Renede leaflet of adjacent nodes, hairs on stems 0.05—0.25 mm long G. > minor SIDA 8(2): 170-180. 1979. 171 6. Several to most internodes (especially those toward the base) much longer than the oe t leaflet of uae nodes, hairs on the stem lon G .1-0.8 mm long : : . glabella 5. Hairs on stems retrorse to i. gereadine ee es ae a al OF Mature flowers 7-9 mm long . . woe ee. GG. regularis 7. Mature flowers 10-18 mm long 8. Longest inflorescences 3-15 cm long, nodes. near the tip ‘close and the 6-9 G. flowers congested; ovules 6— glabella 8. Longest inflorescences 5-55 cm Tong A dower in separated ns and not congested near tip; ovules 10-13 : G. volubilis DISCUSSION OF SPECIES G. ELLIOTTI Nutt. and G. ERECTA ( Walt.) Vail These species have distinct features and are rarely confused with others in contrast with all other species, which are frequently misnamed. G. MOLLIS Nutt. This taxon js often confused with G. floridana. In addition to the characters in the key the legume of G. mollis has dense divergent hairs prominently visible macroscopically while in the former species the hairs are scarcely or not visible macroscopically and are retrorse-appressed to retrorse-spreading to a few scattered hairs being divergent. G. FLORIDANA T. & G. r the present this species is best treated as two varieties, floridana and microphylla Chapm. The latter has shorter internodes, smaller leaflets, shorter petioles, and the inflorescences are shorter than the leaves. Inflorescences are mostly longer than the leaves in var. floridana. Additional studies are needed to test this separation more thoroughly. Growing plants under the same con- ditions is likely to be helpful G. MINOR Duncan Additional aids to the identification of this taxon are: stems geniculate (un- commonly not), largest leaflets 14-28 mm long, inflorescences with 1-3 (4) flowers, the longest inflorescence little if any longer to shorter than the sub- tending leaf, and flowers 11-17 mm long. G. REGULARIS (L.) BSP. r many years there seems to have been little disagreement among tax- onomists concerning the application of the name G. regwlaris. Descriptions, although differing in some details, indicate that each author had the same taxon in mind. These authors include Small (1933), Fernald (1950), Gleason and Cronquist (1963), Wilbur (1963), and Radford et al. (1964). Under this name until now, as indicated by the descriptions, have been included individuals that were prostrate or rarely twining; had inflorescences shorter or longer than the leaves and with one to many flowers, nodes close in the upper part of most inflorescences, and flowers congested or few; flowers 10-18 mm long and with a calyx 6-10 mm long; and stem hairs 0.05—-0.8 mm long and usually retrorse, but frequently antrorse and sometimes spreading [descrip- tion mine}. 172 G. MOLLIS pode tase é oo FLORIDANA MICROPHYLLA ~ Tee DN INK o G. MINOR Wor 1-3. ‘ ag Ls i i we gy a A Os BPpeayt Pps =} L | ' Z am. ii) Although at first glance the type specimen (Virginia — Clayton — ex. Herb. Gronovii in BM) in some respects looks like a specimen of G. regularis sensu recent authors, it differs significantly from the above description in several details. The calyx is only about 5 mm long, the flower only 8 mm long, the upper flower clusters are evenly spaced and not congested, and the hairs on the stem are soft and spreading which is very unusual for G. regalaris as it has been interpreted. These data fit in all respects what has been treated variously as G. volubilis (L.) Britt. (Small, 1933, in part; Fernald, 1950; oe 1963; Gleason and Cronquist, 1963, in part; Radford et al., 1964), G. volubilis var. mississippiensis Vahl, and G. mississippiensis (Vahl ) Rydb. Also the description by Linnaeus (Se Pl. I: 726. — no. 8, DOLICHOS regularis) can as easily be applied to this taxon as to what has been passing as G. regularis. Therefore, the name G. regularis should be used for G. volu- bilis sensu Fernald (1950) et al. During the checking of hundreds of herbarium specimens, I noted that G. regularis and G. mollis often had been confused. The hairs on the stem do seem to intergrade in size and abundance, with G. mollis generally hav- ing the longer and more abundant hairs. A series of measurements of hairs on specimens of known identity probably will show that almost all speci- mens of these two species can be identified by hair characters. Until this is done, except for sterile specimens, there are other means of identification as follows: Calyx greenish-yellow to tan on inner surface when dry; corolla light colored when dry, falling as it withers or soon afterwards; legume Note scattered diver - nt to antrorse-appressed hairs scarcely or not visible macroscopica . G. regularis Calyx brown to reddish brown on inner surface w as dry; ar a anu co ied when dry, persisting after withering, sometimes partly present when fruit is gs with dense divergent hairs prominently visible oe : i 7 ollis A partial synonomy of G. regwlaris should be useful in adjusting to the new application of this name and follows: Galactia regularis (L.) BSP. Prel. Cat. N-Y. 14. 1888. Dolichos regularis L. Sp. Pl. 726. ?Hedysarum volubile L. Sp. PI. a en ?Ervum volubile Walt. Flr. Car. 187. 1788. Galactia mollis Nutt. Gen. 2: 117. 1818. Not Michx., 1803. Galactia pilosa Ell. Bot. S.C. and Ga. 2: 238. 1824 _ Not Nuttall, 1818. Galactia villosa Eaton & Wright. N. Am. Bot. 248. 1840. Not Wight and Arnott, Prodromus Fl. Ind. Orient I. 1834. Galactia volubilis (L.) Britt. Mem. Torr. Club 5: 208. 1894. Sensu:—Small, 933, in part; Gleason & Cronquist, 1963, in part; Fernald, 1950; Wilbur, 1963; Radford et al., 1964. Galactia ia (L.) Britt. var. misséss7ppiensis Vail. Bull. Torr. Bot. Club 22: 500. 189 Galactia ee Vail. Bull. Torr. Bot. Club 22: 500. 1895. G. GLABELLA Michx. In order best to understand the application of this name it is useful to 174 form of GLABELLA GLABELLA , o @ w : | < aN oe , . Ro oo : 4 o U8 : ° = nM * oc q un8 Fal @ Pog =e 7 a Pe ee ; esr oe “54 + eae — ee ed Ge ee | ee aaa ; ert. f : v\e T jt is { eS, } ‘a 4 4 is 1 ek ss STALL st gN \ ope : ee OR ae ; __¥ fa N ~ aS a 5 N NSN 1 N nS “+ PaL-t—2] Ee 7 ; entlizike ¢ ki, EAL wu y, # AR pee “psd ” - i b a a 175 discuss the variability of the plants that have long been designated as G. regularis. Actually there are two and maybe four taxa involved. One is G. minor which has been discussed. The remaining plants of the former G. regularis consist of at least three types of plants. One has retrorse or very rarely retrorse-spreading hairs. The other two types have appressed antrorse or very rarely partly antrorse-spreading hairs. No plants have predominately divergent hairs. The three types have definite distributional patterns which coincide only partly. They have some characteristics in common which are different from those of G. minor, including having the stems straight, not geniculate; several to most internodes much longer than the longest leaflet of the adjacent nodes; the longest inflorescences longer than the subtending leaves, uncommonly shorter, and with several to many flowers, rarely as few as 4, the upper nodes close and the flowers congested in the larger inflo- rescences. It is interesting that Small (1933) reserved the name G, regularis for those individuals having “minutely retrorse-pubescent” stems. However, none of the other species he includes can be the antrorse-haired G. minor. The “retrorse” population is much more abundant than the “antrorse” ones. Its characteristics essentially are those recognized in publications for those plants until now designated as G. regularis. The next oldest name that possibly could apply to this “retrorse” taxon is apparently ERVUM? volubule (Walter Flr. Car. 187. 1788). However, Linneaus (Sp. Pl. II: 750. 1753) had used the epithet previously describing the plants as having “caule volu- bile,” which is unusual for this taxon. Furthermore, I shall show later that voluble applies to another taxon. I believe, however, that Michaux’s Galactia glabella (Fl. Bor. Am. 2: 62. 1803) can be used for the “retrorse” taxon with reasonable confidence. Michaux’s description of the plants as being (translation) “almost glabrous: leaflets oblong-oval, both ends shallowly notched, obtuse: calyx glabrous” is unlikely to apply to any other Galactia, and especially to those taxa known to occur where the type specimen was collected, namely G. volubilis and the “retrorse” plants. At flowering the calyx of G. volubilis is nearly always appressed hairy and only rarely nearly glabrous. On the other hand, the calyx of specimens of the “retrorse” plants is often glabrous or nearly so. The type specimen of G. glabella (Michx. Herb—Col. Co.—S. Carolina —in P) neither supports nor rejects application of the epithet to either of the above taxa. The specimen is apparently from Colleton County, which included at that time all of Charleston County southwest of the Ashley River. The specimen consists of three separate partly twining pieces, all sterile. one of its characters strongly favors either taxon other than the twining which is much more common in G. volwbilis. However, rather than creating a new name because of this possible discrepancy, I favor using G. glabella on the basis of the strong implications involved in Michaux’s describing the calyx as glabrous. His description most likely is based on a specimen from the “retrorse” population. 176 A detailed description of G. glabella (G. regularis, in the sense of previ- ous authors, minus G. szévor and the other “antrorse” populations) should be helpful and follows: Perennial herb. Stems prostrate, straight or sometimes twining, bearing appressed to rarely spreading—retrorse hairs 0.02-0.45 mm, or rarely a few scattered hairs to 0.7 mm long; some internodes, if not all, longer than the longest leaflet of the subtending nodes. Leaves compound, 28-92 mm long; leaflets 3, elliptic to narrowly elliptic to narrowly ovate to ovate, usually retuse, entire, largest per plant 7-20 mm wide and 20-45 mm long, thinly antrorse-appressed hairy beneath, glabrous or sometimes antrorsely scabrous above. Inflorescences axillary, longest 2-15 cm long, longer than the leaves, rarely shorter. Longest pedicels at flowering 2-5 mm long. Flowers 5-13 per inflorescence, 12-18 mm long. Calyx 5.5-9 mm long, glabrous or with antrorse appressed hairs. Longest dehiscing anthers usually 0.88-1.20 mm long (1:17 is less than 0.88 mm long). Legume 35-52 mm long and 4.5- 5.7 mm wide, bearing appressed antrorse hairs. Maximum number of ovules or seeds (including aborted ones) per specimen 6-9(10) per legume. Representative specimens include: VIRGINIA: Southampton Co.: Fernald and Long 8744 (DUKE); NORTH CAROLINA: Pender Co.: Wilbur 4136 (FSU, GA); Scotland Fox and Godfrey 2412 (DUK es SOUTH CAROLINA: Orangeburg Co.: Ables 31751 (NCU). GEORGIA: Chatham Co.: Duncan 21128 (NCU), Duncan 21229 (GA). FLORIDA: Brevard Co.: Kral 5066 (FSU) \ a OQ I have no strong feelings as to how the “antrorse” population should be treated, as a separate species, as a variety of G. minor or G. glabella, or other- wise. It seems useful to others to point out that the Florida material and that from the coastal and adjacent counties into North Carolina have leaflets averaging broader than those of the more inland populations of NC, SC, and G he narrow aspect of the leaflets of this latter group is quite evident in comparison, width-length ratios being mostly around 1:3(4) as compared to mostly 1:2. I have other commitments which will keep me from attacking these and other problems involved for at least several years and so am leaving them for others who I know are interested in the problem. Until further studies can be made, it seems best to include these as forms of G. glabella, providing them with no names. Representative specimens of the antrorse forms include: NORTH CAROLINA: Bladen Co.: Rogers and Blomquist 3120 (DUKE). Moore Co.: Fox and Godfrey 2424 (GA). Richmond Co.: Correll 1079 (DUKE). Scotland Co.: Ahbles 28568 (NCU). SOUTH CAROLINA: ee Co.: Radford 27337 (NCU). Kershaw Co.: hee 23549 (NCU). Marion Co.: Bell 13677 (NCU). GEORGIA: Clinch Co.: Norris 1132 Charlton Co.: Duncan A (GA). McIntosh Co.: Bozeman 2824 (NCU). Set DA: Lee C oldenke 951 (DUKE). Nassau Co.: Godfrey 64135a (FSU), Godfrey and Lindsey 56855 (FSU Useful synonomy for G. glabella: Galactia glabella Michaux. Fl. Bor. Am. 2 oe 1803. Galactia purshii Desv. Am. Sci. Nat. 9: 413. 1826. Ee 177 BOGE KB LQ A ~pae we ONS s, LA n a ad ea 2 Oo WwW ox rsa} 178 — regularis (L.) Britt. Sensu Small, 1933. In part sensu: Fernald, 1950; Gleason & Cronquist, 1963; Wilbur, . Radford et al., 1964. G. VOLUBILIS (L.) Britt. As indicated earlier, G. regularis has been treated by recent authors as G. volubilis. These decisions were apparently based largely on the description by Linnaeus and an eies he pe (Dill. Elth. 173. t. 143. f. 170). Fernald (1950), ai Ibur (1963), and Radford et al. (1964). It is also quite doubtful which of these species the illustration represents. The leaf shape and relatively long pedicels are probably of the latter. The retrorse- spreading hairs on the stem and the antrorse-spreading hairs on the calyx are likely of G. regwlaris. However, I have seen each of these characters on specimens’ of G. macreei sensu above. Although the apex of the standard of the latter is rounded or nearly so and that of G. regularis is retuse, the illus- trations (Dill. Elth., above) of the flower are inconclusive as characteristics of both taxa are displayed. On these bases application of the epithet volabilis is clearly in doubt. Fortunately help is provided by a specimen in the Dillenius Herbarium labeled “Hedysarum trifoliatum scandens.” From 1:1 photo- copies of the specimen and other data provided by F. White, Curator of the Herbaria at Oxford University, Great Britain, the following diagnostic characteristics were derived: Hairs on stems retrorse-appressed, longest in- florescence 29 cm long with well separated nodes, longest pedicels at flowering ca. 4 mm long, longest flower 13 mm long, longest calyx ca. 6 mm long, the corolla light colored when dry. The specimen is clearly of G. macreei sensu the authors indicated earlier. On the basis of priority, however, G. volubilis (L.) Britt. should be used for the taxon. G. volubilis and G. glabella are similar in several ways and have been confused. Both twine, the former occasionally and the latter usually, but when the latter is growing in bare areas it cannot climb and is prostrate like the former. They may be separated as follows: Longest inflorescences 3-15 cm long, upper flower clusters congested; longest flowers 12-18 mm long; longest dehiscing anthers usually 0.88—1.2 less than 0.88 mm long); maximum number of — or seeds (including shared ones) per specimen 6-9, rarely 10(1:26) per legum G. glabella Longest inflorescences 5-55 cm long, all floriferous ane well separated: longest Sl 10-14 mm long ; longest diadas anthers 0.65—0.88 mm long; maximum mber of ovules or seeds (including aborted ones) per specimen usually 10-13, ee 9(1:43) per legume. ee owe . G. volubilis Numerous plants of G. volubilis have also been identified as G. regularis sensu Duncan and also the reverse. This can be prevented by checking the following differences: ee Co., SC; H. E. Ahles 15677; NCU 103886 + 103876.—Long Co., Ga.; J. R. Roseman, 2114; GA 94661. 179 Hairs on the stems usually spreading, to occasionally retrorse-spreading to uncommonly retrorse appressed; leaflets mostly elliptic to ovate; longest inflorescences usually 3 : 5—7, occasionally 4 or 8, or rarely 3 or 9? (1:133) per legume. regularis Hairs on the stems usually retrorse appressed to uncommonly retrorse- redid leaflets narrowly pe to narrowly ovate or approaching the above; longest in- florescences 5-45 cm long, longest pedicels at flowering 3.0-4.0 mm long; longest calyx (5.5) 6-10 mm long; longest flowers 10-14 mm long; longest dehiscing anthers 0.65—-0.88 mm ue hairs on legume antrorse-appressed or rarely antrorse- spreading; maximum number of ovules or seeds (including aborted soar per speci- men usually 10-12, teens 13, or a 9? (1:43) per legume . volubilts Partial synonomy for G. vol#bilis follow oe volubilis (L.) Britt., Mem. Tor rr. Club 5: 208. 1894. Sensu: Small 33) in part, Gleason & Cronauist eae in part. Cina pilosa Nutt. var. macreei Carey T. & G. Fl. N. Am. 1: ae 1838. Galactia pilosa Nutt. var. oa sg T. & G. Fl. N. Am. 1: 287. 1838. Galactia macreet M. A. Curtis, Bost. Jour. Nat. Hist. 1: 120. 1837. Sensu Fernald (1950), Wilbur (1963), Radiord et al. (1964). Galactia volubilis (L.) Britt. var. intermedia Vail, Bull. Torr. Bot. Club 22: 08. 1895. DISTRIBU TION During this study distribution maps were prepared for Galactia minor, the retrorse form of glabella, the antrorse forms of glabella, regularis, volubilis, mollis, and floridana. The specimens were mostly from DUKE, FSU, GA, NCU, VDB, and VSC. A few were from FLAS, GH, NY, and US. These distribution data were essential during my study, and are useful in understanding my interpretations of these taxa. They are presented here for the Southeastern United States. Appreciation is expressed to my colleague, Dr. Samuel B. Jones, for sug- gestions involving the study; to Dr. G. Taylor, Keeper of Botany, of the British Museum, London, who many years ago provided details about the type specimen of Galactia regularis, and to F, White, Curator of the Herbaria at Oxford University, Great Britain. The loan from Duke University Library of the Ph.D. dissertation, “The genus Galactia in the United States,” by Hollis J. Rogers and the loans of specimens through the courtesy of curators of the several herbaria are appreciated. REFERENCES DUNCAN, W. H. 1977. A new species of Galactia (Fabaceae) in the southeastern United States. Phyilosia 37: 59-61. ? The only specimen with 9 seeds is from Yell Co., Ark. 3 The only specimen with 9 seeds is from Iberia Parsich. La. 180 FERNALD, M. L. 1960. Gray’s manual of botany. 8th ed. American Book Co. N. Y. 1632 GLEASON, H. A. and A. CRONQUEST. 1963. Manual of vascular plants of north- oa eae United States and adjacent Canada. D. Van Nostrand Co. Princeton, ote: A. E., H. E. AHLES, and C. R. BELL. 1964. Manual of the vascular flora of the Carolinas. Univ. of North Carolina Press. Chapel Hill. 1182 p. LL, J. K. 1933. Manual of the southeastern Flora. Univ. of North Carolina Press. Chapel Hill. 1554 p WILBUR, R. L. 1963. The leguminous plants of North Carolina. Tech. Bull. No. 151. North Carolina Agr. Exp. Sta. Raleigh. 294 p. A NEW SPECIES OF LINUM FROM SOUTHERN TEXAS AND ADJACENT MEXICO C. M. ROGERS Wayne State University, Detroit, MI 48202 In an earlier publication (Rogers, 1968) one of the yellow-flowered flaxes, Linum rigidum Pursh, was described as comprised of four varieties. Of these the typical variety and var. compactum are plants of northern Texas and northward. The var. berlandieri, with some reason sometimes considere a separate species, is distributed nearly throughout Texas. It is rather readily distinguished from the other varieties by its comparatively thick-walled fruit and coarse sepals. These three varieties, although displaying some variation which may require further study, are reasonably well defined. The fourth variety, var. filifolimm Shinners, however, was described as a “variable population .. . more study is necessary to determine the relationship of the western Texas plants to those of southern Texas, as well as some anomalous collections included here from northern Mexico.” In order to help clarify these relationships, the author has collected additional material of var. flifoliwm and attempted a number of crosses between plants from different parts of its range. Herbarium material from the New York Botanical Garden, Southern Methodist University and the University of Texas, including the Lundell Herbarium, has also been reexamined and thanks are expressed to the respective curators. Within this “variable population,” in addition to Linum rigidum var filifolium, which, as now interpreted, is a plant of western Texas and nearby Coahuila, two taxa, mostly of southern Texas and adjacent Tamaulipas and Nuevo Leon, can be distinguished. These are L. elongatwm (Small) Winkler, a species proposed many years ago (Small, 1907), and previously regarded as a synonym (Rogers, 1968), and L. lwndellii, a previously undescribed species. The latter is named for C. L. Lundell, long a student of the Texas flora and co-collector of the type specimen. The three taxa may be distinguished from one another in the following way: 1. Styles 7-8 mm long; petals mostly ca. 15 mm an lanceolate, acute; petals diffusely a - at the base; annual... L. ea Cie var. Alo 2. Sepals lanceatenaat — ee reddish to wine-colored band below the aidale mostly perennial . L. elongatum. : . Styles 3-4 mm ie pe aie ca. 10 mm long... . L. landellii In the following descriptions, petal and anther coloration is taken from SIDA 8(2): 181-187. 1979. 182 a limited number of living plants, supplemented by scattered information from herbarium labels. LINUM RIGIDUM Pursh var. FILIFOLIUM Shinners in McVaugh, Field & Lab. 17: 136. 1949. Type: McVaugh 7798, 4 Apr 1947, near abandoned wax factory on Rio Grande at mouth of Big Canyon, Brewster Co., Tex. (Holo- type: SMU; isotype: TEX). More or less stiffly branched, essentially glabrous annual herb, 15-35 cm tall; leaves alternate, linear, 10-30 mm long, 0.5-2.0 mm wide; stipular glands mostly present (80% of the specimens examined have stipular glands, but see the discussion which follows the species descriptions); sepals lanceolate, sharply acute, 6-9 mm long, grayish; petals narrowly to broadly obovate, 13-17 mm long, yellow to orange or salmon, brick-red near the base, not banded (Fig. 4a); stamens 5-6 mm long; anthers 1.5—2.5 mm long, yellow- ish; styles 6.5-8.0 mm long; stigmas dark-wine to black; fruit ovate, ca. 3.5-4.0 mm high, 3.0 mm diameter; seeds reddish-brown, ca. 3.0 mm long, 1.25 mm wide; chromosome no. 2 = 15 (chromosome voucher: Rogers 13494, 29 Aug 1976, 4 mi SE of Del Rio, Val Verde Co., Tex. WUD About 70 collections have been examined. The following citations include the Mexican collections and a specimen from each of the counties in Texas. MEXICO. Coahuila: ca. 100 mi NW of Muzquiz, 12 May 1968, Latorre s.n. (TEX Santa Rosa Mts., a Jul 1938, Marsh 1386 (TEX); S of Sabinas, 29 Mar i Rinehart 289 (WU exas. a ee Co.: 23 mi on Eunice Highway, 8 es 1960, Scuddy s.n. (nea rest cs australe but apparently with some features of var. filifolinm) (LL). Brewster Co.: 26 mi E of Marathon, 23 Aug 1947, Warnock “6719 (TEX). Crane Co.: ca. 10 mi NW of Crane, 9 May 1966, Correll 32773 (appears to combine traits of var. filifolinm and var. rigidum) (LL). Culberson Co.: 51 mi W. Bal- morrhea, 29 May 1967, Mears 15654 (TEX). Jeff Davis Co.: SW éf Mt. Livermore, Davis Mts., 15 Jun 1926, Demaree & Palmer 212 (SMU). Kinney Co.: of Del Rio, 29 Aug 1976, Rogers 13493 (WUD). Maverick Co.: 6 mi N of Que- mado, 4 Apr 1959, Correll 20727 (LL). oa Co.: 10 og S of Ft. Stockton, | a 1948, Warnock 7691 (LL). Presidio Co.: ca. 11 mi of Valentine, 25 Jun 1948, York 48185 (TEX). Terrell Co.: 3 mi W of sre 30 Aug 1976, Rogers 13498 D). Val Verde Co.: near Del Rio, 30 Mar 1963, Correll 27114 (LL). distribution is shown in Fig. 1 LINUM ELONGATUM (Small) Winkler, Nat. Pflanzenfam, ed. 2, 19a: 116. 1931. Type: Reverchon 3776, 21 Mar 1903, Laredo (Webb Co.), Tex. (Holotype: NY Cathartolinum elongatum Small, North Amer. Flora 25: 82. 1907. More or less diffusely branched, annual (or merely flowering first year? ) or perennial, essentially glabrous herb, 15—30 cm tall; leaves alternate, linear, 5-25 mm long, 0.5-1.0 mm wide; stipular glands present throughout or at bases of upper leaves only; sepals mostly lance-attenuate, 6-11 mm long; petals broadly obovate, 14-18 mm long, yellow-orange to salmon or brown- 183 ° r 1 L. rigidum var. filifolium . a @ > @ @ e > @ ®e e* Wy ° @ op doe ee elongatum : 3 L. lundellii Figs. 1-3. Distributions of L. rigidum var. filifolium, L. elongatum, and L. lundellii. 184 ish-red, with a prominent wine-colored band below the middle (Fig. 4b); stamens 5-6 mm long; anthers 1.5-2.5 mm long, brick-red; styles 7.0-9.5 mm long; stigmas mostly grayish to wine-colored; fruit ovate, ca. 4 mm high, 3 mm in diameter; seeds reddish-brown, ca 3 mm long, 1.2-1.3 mm wide; chromosome no, 7 = 15 (chromosome vouchers: Rogers 13475, 27 Aug 1976, farm road 2895, 5 mi N of jct. with Tex. 359, Webb Co., Tex.; Rogers 13482, 28 Aug 1976, 1 mi SE of San Ygnacio, Zapata Co., Tex., both WUD). About 65 collections have been examined. The following citations include the Mexican collections and a specimen from each of the counties in Texas. on EXICO. Tamaulipas: 7 mi S of Nuevo Laredo, 12 oa 1964, Fields 45 (TEX); 16 mi S of Nuevo Laredo, 7 Mar 1962, Garza 35 (TEX); 14 mi S of Nuevo Laredo, 24 Mar 1944, Heard & Barkley 14604 (TEX); 10 mi SE of Nuevo Laredo, 8 2 1964, Ibarra 103 (TEX); 12 mi S of Nuevo Laredo, 26 Jun 1963, Rogers 1284. (WUD); 2 mi S of Nuevo Laredo, 26 Jun 1963, Hagens 12846 (WUD); 7 mi d of Nuevo Laredo, 28 Aug 1976, Rogers 13476 (WUD); Highway 2, 5 mi E of jet. with a 85, S of eee Laredo, 28 Aug 1976, Rozen: 13477 (WUD); 4 mi f Nuevo Laredo, 14 Nov 1958, Rollins & Tryon 5802 (LL); Arroyo Coyote, 7 km. Carretra Nacional, 1 Apr 1964, Vazguiz 30 (TEX). U Texas. Cameron Co.: US. 83, 12 mi N of Prowasyls 18 Apr 1965, Cavazos 262 (LL). Dimmit Co.: 5 mi N of Carrizo Springs, 16 Mar 1963, fue SO (TEX). Duval Co.: Realitos, 17 Mar 1934, oe 781 (NY). Jim Hogg Co.: 8 mi S of Hebbronville, 20 Mar 1969. ules 36801 (LL). Jim Wells Co.: 6 mi S of Alice, 20 Mar 1952, Jones 699 (SMU). LaSalle Co.: near Encinal, 27 Aug 1976, Rogers 13470 (WUD). Webb Co.: oe 11 Mar 1944, Crockett 5417 (LL). Zapata Co.: 13 mi N of oa Ygnacio, 31 Jan 1954, Shinners 17656 (SMU). The distribution is shown in Fig. Bes LINUM lundellii Rogers, sp. nov. L. elongato affine, sed semper annuum, glandulis stipularibus foliorum oe nullis, petalis circa LO mm gen et stylis 3-4 mm longis. Type: C. L. & A. A. Lundell 9894, 2 Apr 1941, « gravelly hill, off U.S. 83, W of Sullivan City, Starr Co., Tex. (Holotype: tL . nnual, essentially glabrous herb, 10-40 cm tall; leaves alternate, linear, 5-30 mm long, 0.5-1.5 mm wide; stipular glands absent above, moderately developed below; sepals linear-lanceolate to lanceolate, acute to acuminate, 4-12 mm long; petals obcordate, 7-12 mm long, yellow to orange-salmon, faintly banded near the base (Fig. 4c); stamens 4-5 mm long; anthers 1.0- 1.5 mm long; styles 3.0-4.0 mm long; stigmas mostly dark-wine; fruit ovate, 3.3-4.0 mm high, 2.6-3.1 mm in diameter; seeds reddish brown, 2.5-2.7 mm long, ca. 1.1 mm wide; chromosome no. 7 = 15 (chromosome vouchers: Rogers 13137, 2 Jul 1964, 2 mi N of Roma, Starr Co., Tex.; Rogers 13471, 27 Aug 1976, 5 mi E of Laredo, Webb Co., Tex., both WUD). In addition to the type and the specimens cited for cnomescme number, the following collections have been examined. EXICO, Nuevo Leon: ca. 15 mi SW of Galeana, 19 May 1934, Mueller and M Mueller 470 (TEX): 4 mi S of China, 6 Aug 1964, Rogers 13189 (WUD). Tamaulipas: 5 mi SW of Reynosa, 29 Feb 1944, Painter & Barkley 14412 (LL); ca 185 mi SW of Reynosa, 6 Aug 1964, Rogers 13190 (WUD); 2 mi W of Nuevo eat 28 Aug 1976, Rogers 13478 (WUD). Texas. Dimmit Co.: 2 mi N of Carrizo a 3 Jul 1964, Rogers 13142 (WUD); 3-4 mi NW of Carrizo ia 29 Aug 1976, Rogers 13486 (WUD); = mi NW of Carrizo Springs, 29 Aug 1976, Roger 13488 (WUD). Hidalgo Co.: mi W of LaJoya, 28 Aug 1976, Rowers 134 (WUD); La Joya, 9 Feb 1942, o alker 48 (TEX). Jim Hog Co.: Arroyo ee eae 15 Mar 1964, Correll 29000 (LL). Starr Co.: ca. 1.5 mi N of Roma, 17 Mar 1966, Correll 32270 Glee): pees Falcon Dam, 17 Mar os Ramos et al 7864 Sas X); near Falcon Dam, 2 July 1964, Rogers 13138 (WUD); U.S. Highway 83, 3 mi SE of pape with aes road 2 098, 28 Aug 1976, sles 13480 (WUD); 3 mi N of Roma, 31 Jan 1954, Shinners 17702 (SMU); Falcon State Park, 14 Mar 1968, Wood 83 8 (TEX). The distribution is shown in Fig DISCUSSION Linum lundelli, except for vegetative features such as stature, amount of branching, and leaf size, is a very uniform species and may be readily dis- tinguished from L. elongatunz, the range of which overlaps that of L. landellii, as well as from L. rigidum var. filifolium by style length, which is well correlated with other differences such as petal size and shape, anther size, and the absence of stipular glands at the bases of upper leaves. Its distinctive- ness is supported by the results of attempted crosses with L. elongatum (128 crosses) and with L. rigidum var. filifolinm: (164 crosses), none of which produced functional seeds. In L. landellii the anthers are placed at about the same level as the stigmas (Fig. 4c). In garden-grown plants, by the time the flowers were well opened, pollen had already been deposited upon the stigmas. In these plants. self- pollination was certainly the rule. In L. elongatum and L. rigidum var. fulifolium the styles extend well beyond the anthers (Fig. 4a,b), and there are ordinarily a few hours (depending upon cloudiness and wind condi- tions) during which the anthers have dehisced, but pollen has not been shed. In the Michigan garden no insect pollinators were observed; these may be present in the natural range of the species. There is at least the oppor- tunity for outcrossing. If cross-pollination fails, however, self-pollination ts assured, since the intact corolla, before falling, ordinarily slips upward past the open anthers like an ascending collar, depositing pollen on the stigmas above. Linum elongatum and L. rigidum var. filifolium appear to be closely re- lated. It may be that some hybridization takes place between the two in the field. The results of experimental crosses between the two reflect this possi- bility; one or more rather weakly developed but mature seeds were pro- duced in 14 of 136 crosses attempted. Of a series of features that differentiate the two, possibly none holds for every comparison of individual specimens. For example, most collections of L. elongatum are perennial. This is a very unusual character in the “L rigidum complex” of about twelve species. Linum subteres (Trel.) Winkler, of Utah and Nevada, is the only other which ts tal, stamens and a of L. rigidum var. filifolium (a), L. elongatum Pas a : lundellii (c). All x 2.5. consistently perennial. However, L. elongatum flowers the first year, and a number of herbarium specimens appear to be or are annual. In addition, two collections of L. rigidum var. fulifolium are quite clearly perennial. The petal coloration of L. elongatum makes this one of the most attractive of the flax species. The author's collections and the several dozen plants grown from their seed were uniformly colored in the manner described. Dried specimens, because of the fading of the pigments and because petals have frequently fallen if collections are made late in the day, do not always show petal color. In garden-grown plants, one individual (of about 40 plants), which otherwise had the characters of (and was identified as) L. rigidum var. filifolium, had the petal coloration of L. elongatum. Other- wise the two taxa were easily distinguished on this character alone. The attenuate sepals, which were the basis for the selection of the specific epithet for L. elongatum, can be used for the identification of most specimens, but there is enough variation in both taxa that some specimens would be difficult to place using that character alone. Linum elongatum uniformly possesses stipular glands at the bases of the upper leaves, L. /wndellii uniformly lacks them there, although they may be moderately developed at the bases of the lower leaves. Most collections of L. rigidum var. filifolium have stipular glands also, but with the exception of a single collection from near Del Rio, none of the specimens from Val Verde, Kinney, or Maverick Counties have stipular glands. This may provide another distinguishing feature between L. rigidum var. filifolium and L. elongatum, if one is needed, in the region of closest geographical proximity of the two taxa. Baesd upon the specimens thus far examined, the range of L. rigidum var. filifolium is distinct from those of either L. elongatum or L. lundellii, being restricted in Texas to northwestern Maverick Co. and northwestward, while L. elongatum and L. lundellii are found only from Dimmit County southeastward. 187 REFERENCES ROGERS, C. M., 1968. Yellow-flowered ae of Linum in Central America and western North America. Brittonia 20: 107— SMALL, J. K., 1907. Linaceae. In: North ae Flora 25: 67-87. REDISCOVERY OF SPIRANTHES PARKSIT CORRELL P. M. CATLING AND K. L. McINTOSH Department of Botany, University of Toronto Toronto, Ontario, MSS 1Al1 Spiranthes oS Correll was described in 1947 on the basis of specimens collected by H. B. Parks along the Navasota River (Democrat Bridge) in Brazos County, Texas (Correll, 1947). It appears that the late Dr. Parks was the only person to have seen a living plant. Correll (1950) and Correll and Johnston (1970) reported that this species was endemic to Brazos County but recent attempts to find it there have been unsuccessful (Luer, 1975; . Correll and M. C. Johnston, pers. comm.). Spiranthes parksii is the only North American orchid not illustrated with a photograph in Luer's (1975) recent work on the North American orchids. It is one of the 20 species listed by Ayensu (1975) in an attempt to identify those North American orchids in serious danger of extinction. Spiranthes parksii is not only one of the rarest North American orchids, it is also one of the least well known. While collecting data from the AMES orchid herbarium (Harvard Unt- versity) in 1975, I noticed a sheet labelled “S. cernwa, Hy College—Navasota, 10 miles west of Navasota R. bridge on hwy no. 6, H. B. Parks, 27 Oct 1945” (AMES 63043). This sheet had 12 plants mounted on it, some referable to S. cernua, others to S. parksii. The type of S. parksii (AMES 63039) was available for comparison. On 25 Oct 1978, we had an opportunity to explore the Post Oak Savanna northwest of Navasota. Local residents reported that it had been a dry sum- mer and perhaps for this reason Spiranthes spp. were not easily found. S. cernua (L.) L. C. Rich. occurred sparingly along margins of ponds and streams. Both the normal white open-flowered plants and more or less peloric, yellow closed-flowered plants were seen. A few S. lacera (Raf.) Raf. var. gracilis (Bigel.) Luer were found on dry sandy banks, all past anthesis. Exploring the open banks of a temporary stream surrounded by scattered oaks (Quercus stellata, Quercus marilandica) and Beauty-berry (Callicarpa americana), several Spiranthes orchids were found among the open cover 0 grasses and forbs (Fig. 1). Some of these were closed-flowered S. cernua but others scattered among them had open greenish-white flowers with relatively short rounded petals and an ovate, distally truncate lip. The floral bracts of these plants also differed in having distinctly whitened tips. close examination and dissection of a few flowers indicated that these plants SIDA 8(2): 188-193. 1979. 189 Fig. 1. Open grassy meadow habitat of S. parksii along a temporary stream. Dominant plants include Schizachyrinm scoparinm var. freguens, Sporobolus juncens, and Eupatorium compositifolium,. Photographed 25 Oct 1978, ca. 19 km NW of Navasota, Brazos Co., Texas. 190 were beyond any doubt referable to S. parksii, the description and original illustrations of which we had with he dominant plant associates here included Schizachyrium scoparinm var. frequens, Panicum brachyanthum, Aristida longispica, Sporobolus junceus, Expatorium compositifolium, and Linum medium var. texanum. Chrysopsis pilosa and Ascyrum hypericoides var. hypericoides were also present. A few hundred yards away we found more plants of S. parksii on the banks of another temporary stream (Fig. 2). The aa here was again an open oak woodland (Q. marilandica, Q. stellata, Q. nigra) but with scattered thickets of Forestiera ligustrina, Callicarpa americana and Ilex vomitoria. Spiranthes parksii occurred both on the tops of the banks in open sand with a sparse cover of grasses, and on the sides of banks sometimes in the shade of thickets. Dominant associates here included Andropogon ternarins, Andropogon virginicus, Aristida longispica, Mublenbergia capillaris, Linum medium var. texanum, Eupatorinm compositifolium, and Smilax bona-nox var. hastata. Paspalum setaceum var. stramineum, Ascyrum byperi- coides var. hypericoides and Drosera annua were also recorded as associates at this location. In all, seven plants were found at the first locality and 13 at the second. Soil collected from about the roots of several plants was found to have a pH range of 4.0-5.8, with most readings between 4.5 and 4.9. Since the flower color of S parksii has not been described, various floral parts were compared with the Royal Horticultural Society Colour Chart (1966). The general flower color (7.e. the conspicuous perianth parts) varies from white (155A) to light yellow-green (154D). The dark green color of the ovary (144A-B) extends onto the basal perianth parts for 0.5—2.0 mm. Lateral petals vary from whitish (155A) to yellow-green, (144B-C, 154C). The midvein region is slightly greenish (145C) in white lateral petals and areal green (144A) in greenish lateral petals. The central por- tion of the lip varies from yellow (2D) to yellow-white (158B—C) and yellow-green ea 151C 154C). The calli are white (155A-D). The stig- matic surface is green (144B, C) but the proximal stalk of the column is white (15SA-D) and the pollen masses are yellow (9B) to light yellow- orange (14C). The darkest green color in the flowers occurs in the basal perianth parts, in the lateral petals and on the stigmatic surface. The remain- der of the plant is green (144A) except for the floral bracts which are white 1.5—3.0 mm from the tip. Although no insects were observed pollinating the flowers, the pollinia are easily detached, and the stigmatic surface is viscid, the pollen readily adhering to it. A drop of glucose-rich liquid is secreted near the base of the lip. These characteristics suggest insect pollination, but the ovaries can enlarge and develop seed without pollination. Brown and withered flowers on several plants had the pollen masses intact and the stigmatic surface had not received any pollen. The ovaries of these withered flowers contained Fig. 2. Habitat of S. parksit (lower center). Several plants were growing on the high banks of a temporary stream in the open with Andropogon ternarius, Andro pou VIN SINICHS, Mud aber capillaris, and others (see text). The surrounding trees Ouercus marilandica, and eee stellata. Photographed 25 Oct 1978, ca. 19 km W of Navasota, Brazos Co., Texa Fig. 3. Flowers of S. parksii showing relatively short lateral petals and white-tipped red 25 Oct 1978, ca. 19 km NW of Navasota, Brazos Co., floral bracts. Photograp Texas Oe) immature seeds of which 80-90% were polyembryonic. The occurrence of polyembryony in Spiranthes is associated with adventitious embryony (Swamy, 1948; Catling, unpublished data). With its rounded or oval lateral petals (4.5-6.7 mm long) much shorter than the sepals (5.0-8.0 mm long), and a distally truncate, erose-margined lip (5.0-7.0 mm long), S. parksii appears to be a very distinctive species. The loosely flowered spike and absence of leaves at flowering time also hel to separate S. parksii from other sympatric Spiranthes spp. These features are well illustrated in the drawings provided by Correll (1947, 1950) and in the accompanying photograph (Fig. 3). The green marking along the center of the lateral petals and the white-tipped floral bracts are helpful in field identification. S. parksii keys out readily in keys provided by Correll (1950) and Correll and Johnston (1970). Correll (1947) thought that S. parksii had no close allies in our flora, its affinity being with several Mexican and Central American species. Its taxonomic status and evolutionary relationships are indeed a matter of great interest. In view of the general rarity of the plant, ic seems desirable to document any future discoveries with photographs, measurements and detailed fie!'d notes rather than a large series of collected specimens. We thank D. S. Correll and M. C. Johnston for commenting on the manuscript. REFERENCES AYENSU, E. S. 1975. ee and threatened orchids of the United States. Amer. Orchid Soc. Bull. 44(5): 94, CORRELL, D. S. 1947. A new cS ranibes from Texas. Amer. Orchid Soc. Bull. 16: 400 1950. Native orchids of — America north of Mexico. Chronica Botsaies Co., Waltham, Mass. 399 -and M. C. Johnston. 1970. von of the vascular plants of Texas. Texas Research Foundation Renner, Texas. 1881 pp. LUER, C. A. 1975. The native orchids of the rae States and Canada excluding Florida. New York Botanical Garden, New Y pp. ROYAL HORTICULTURAL SOCIETY. 1966. Colour chart (202 colours each with 4 tints, printed in solid colour). Royal Horticultural Society, Vincent Square. West- minister, S.W.1, England. SWAMY, B. G. L. 1948. Agamospermy in Spiranthes cernua. Lloydia 11(3): 149- 162 CLEMATIS PITCHERI T. & G. VAR. DICTYOT'A (GREENE) DENNIS, COMB. NOV. (RANUNCULACEAE).' W. MICHAEL DENNIS’ Department of Botany, University of Tennessee Knoxville, Tennessee 37916 Clematis pitcheri 'T. & G. is a highly variable, wide-ranging species dis- tributed on a northeast to southwest clinal axis extending from western Indiana, Illinois, and eastern Iowa south to Hidalgo and Queretaro, Mexico (Dennis, 1976). It is distinguished from other closely related taxa by its non-plumose achene tails. In Gray's Synoptical Flora of North America (1895), Robinson recognized four varieties within the C pitcheri complex. More recently Erickson (1943) recognized only two varieties: C. pitcheri . & G. var. pitcheri, which includes those plants with pinnate leaves having leaflets 4-10 cm long and ovate sepals, and C. pitcheri var. filifera (Benth. ) Robinson, which is distinguished from the type variety by having pinnate- ternate leaves having leaflets less than 4 cm long and lanceolate sepals. Clematis pitcheri var. pitcheri was ascribed a distribution from Indiana to eastern Nebraska and south to Texas and var. filifera was reported to occur along banks of streams and in canyons of western Texas, New Mexico and Mexico. In a recent biosystematic study of Clematis subsection Viornae (Dennis, 1976) it was concluded that Erickson’s (1943) recognition of two varieties of C. pitcheri on the basis described above is taxonomically unsound. These characters (leaf size, degree of leaf division and flower shape) vary con- siderably within all members of the subsection suggesting polymorphism and ecotypic variation rather than variation indicative of speciation. How- ever, within C. pitcheri var. filifera sensu Erickson there is a morphologically recognizable element that was treated by Erickson (1943) as a synonym of var. filifera. This element is C. dictyota Greene, which is herein recognized as a variety of C. pitchert. CLEMATIS PITCHERI var. dietyota (Greene) ge stat. et comb. nov. Basionym: Clematis dictyota Greene, Pittonia 5:133. 1903. E: US, Texas, caiee Canyon, at Apr 1902, Tracy ey Earle 256 (Holotype, oo as photo MO!; Isotypes, TEX! US!). [Viorna dictyota (Greene) Heller.] Leaflets generally as pe and more divided than var. pitchert. Sepals light to brownish purple without, tips recurved ‘Contribution from the Botanical Laboratory, The University of Tennessee N.S. 12 2 Present address: Tennessee Valley Authority, EDB, Muscle Shoals, AL 35660. SIDA 8(2): 194-195, 1979. 195 and slightly ae eee and connectives (including extended apex) essen- tially glabrous or occasionally w a few erect trichomes just below the anthers or along the connective or its oe apex. Clematis pitcheri var. dictyota is restricted to soil accumulations among boulders, crevices of rock formations, and stream banks of the Trans Pecos region of western Texas and adjacent New Mexico and Mexico. It is dis- tinguished from var. pitcheri by its glabrous filament and anther connectives. Erickson (1943) did not note this character and referred all small leaved plants of Texas and Mexico with both glabrous and pubescent filaments and connectives to C. pitcheri var. filifera. Correll and Johnston (1970) used pubescence on filaments as a key character to distinguish west Texas popula- tions as C. filifera Benth. However, examination of the type specimen of C. filifera Benth. [Mexico. Prope Leon, 1839. Hartweg 1590 (Holotype, K!; Isotype, LD! )} revealed that its filaments were pubescent. Since the most consistent taxonomic character distinguishing the Trans Pecos populations is the absence of trichomes on the filaments, the name C. filifera cannot be applied to these plants. Review of previously published names and_ study of type specimens indicate that C. dictyota Greene is the appropriate element upon which to base a taxon representative of the Trans Pecos populations. It was described from the region and its type specimen has glabrous fila- ments. Varietal status for the plants of the Trans Pecos region based on the name C. dictyota is therefore proposed. All small-leaved plants in this com- plex occurring in New Mexico, Mexico and Texas that have non-plumose achene tails and pubescent filaments, (ie. C. filifera Benth. sensu stricto ) are referred to C. pitcheri var. pitcheri. REFERENCES CORRELL, D. S. and M. C. JOHNSTON. 1970. Manual of the vascular plants of Texas. Texas Research Foundation, Renner, Texas. 1881 DENNIS, W. M. 1976. A a las study of Copa: section Viorna subsection Viornae. Ph.D. dissertation, Unive of Tennessee. 176 p. ERICKSON, R. O. 1943. eee Clematis section Viorna. Ann. Missouri Bot. GRAY wn "1895. Ranunculaceae. In B. L. Robinson 1895-1897. Synoptical flora of North America, Vol. 1, pt. 1. American Book , New York. FIELD NOTES ON NIGERIAN STRIGA (SCROPHULARIACEAE) LYTTON J. MUSSELMAN, DANIEL L. NICKRENT Department of Biological Sciences, Old Dominion University Norfolk, Virginia 23508 RICHARD A. MANSFIELD, JAMES E. A. OGBORN Weed Science Section, Institute for Agricultural Research Ahmadu Bello University, Zaria, Nigeria Striga is a palaeotropical genus of perhaps as many as 50 species, although the genus has not been monographed. Like many Scrophulariaceae, Striga is root parasitic. In fact, several species are of great economic importance, particularly in the semi-arid tropics. In the rain fed areas of subsaharan Africa S. hermonthica (Del.) Benth. is an ubiquitous weed in guinea corn (Sorghum bicolor) fields. In some areas it is the single most serious para- site of this main crop of subsistence farmers. Millet (Pennisetum typhoides ) may also be attacked by this species. Striga gesnerioides causes — to a variety of crops, including legumes and tobacco. Witchweed (8. aszatica) is the most widespread species in the genus, extending across Africa, the Indian subcontinent, and into Indonesia. It parasitizes corn, rice, sugar cane, and other members of the Poaceae. Striga angustifolia (Don.) Saldanha (= Striga euphasioides Benth.) is an important parasite on sorghum in India. These and a few other species are briefly discussed in Hosmani (1978). Two species have been accidentally introduced to the New World. In the 1950s S. asiatica (L.) Kuntze (= S. lutea Lour.) was discovered in the Carolinas. It is presently known from several counties in southeastern North Carolina and adjacent South Carolina. It is now the object of a federal quarantine and iota program. Very recently S. gesnerioides ( Willd.) Vatke was discovered in Polk Co., Florida (Wunderlin, Musselman and Shuey, 1979). The extent of this infestation is being investigated. The most useful aid to identification of African material is found in Hepper (1963). Characters of diagnostic value in separating species include number of ribs in the calyx, indumentum, relative size of calyx and of corolla tube, size of the bract subtending each flower, and flower color. Like several other genera of parasitic angiosperms that exhibit reduction of vegetative parts, floral characters are the main basis of taxonomic differences in Striga. Plants are often confused in the field, however, particularly S. aspera and S. hermonthica. Despite its agronomic importance, very little information of a descriptive SIDA 8(2): 196-201. 1979. OF nature is available for the genus, especially for those species that are not serious pathogens. The purpose of this paper is to record field observations on several species of Striga observed in Nigera in October and November O78: Specimens were collected from Kazaure in northern Nigeria near the border with Niger and southeast along the main road south to Mokwa. This transect crosses several rainfall isohyets from 750 mm/year in the north to 1500 mm/year near Mokwa and corresponds approximately to the vege- tational zones known as Sudan Savanna, Northern Guinea Savanna, and Southern Guinea Savanna. All species of Striga are considered noxious weeds in the United States and are under a federal plant quarantine. Therefore, all dried specimens from Nigeria were devitalized by treatment with ethylene oxide at 25 Ibs/1000 cu. ft for 2 hours at 70°F or above by the Animal and Plant Health Inspection Service, Plant Protection and Quarantine Programs. These specimens are deposited in the Old Dominion University Herbarium (ODU). Drawings were prepared from FAA preserved material and from Koda- chromes. Silhouettes are from actual specimens. ~ STRIGA HERMONTHICA (Del. ) Benth. This is a strikingly beautiful species with large pink flowers. Of the six species examined, only 8. bermonthica has more than two corollas open per inflorescence branch and flowers that last more than one day (Fig. 1). Widespread throughout the drier regions of Africa, it is quite variable. A cursory examination of populations parasitizing sorghum at Musasa and Zaria (Both Kaduna State, Nigeria) revealed the following corolla variations: upper lobe entire to almost bifid, orifice of throat white or pink, and lower lobes deeply toothed to entire. The extent of this variation among popula- tions remains to be determined as does its source. Of greater agronomic importance is the physiological variation within the species. Preliminary work by C. Parker (personal communication ) of the Weed Research Organi- zation indicates that S. hermonthica may exist as two strains—one attacking millet, the other sorghum. In the present study “millet” S. bermonthica was examined at Mokwa and appeared morphologically identical to that growing on sorghum. Like other species of Srriga, little is known regarding floral biology despite the potential value of such work on a parasite that repro- duces entirely from seed. However, preliminary work by Parker (personal communication) suggests that at least some populations are distinctly out- crossing. — STRIGA ASPERA ( Willd.) Benth. Striga aspera and S. hermonthica are superficially quite similar and may be found growing together. They may be readily separated, however, on the basis of bract size and corolla pubescence. The bract subtending the 198 =p” Pe ee = Mam ~— Z Lint is - cpio wh ZEESEES cases Sw, bre ne § ee ' =a Ue. (Scale for silhouettes equals 10 cm, for drawings 1 cm) ae : Striga hermonthica, portion of inflorescence. 2. Calyces of S. hermonthica. Figure on right shows lateral bract. Smaller structures are bracteoles ; Fig. 3. S. aspera indowescenice: Musselman and Mansfield 5524 Fig. 4. S. aspera, diminutive plant from Kuffena Rock; i te and Mansfield 5530. Fig. 5. Same as Fig. 4, showing as Fig. 6. S. astatica, Tegina; Masel wa Mansfield 553 Fig. 7. American strain of S. astatica trom U.S.D.A ee Laboratory. Fig. 8. Flowers of S. astatica; ee and Mansfield 5531. 199 flower of S. aspera is almost as long as the calyx (Fig. 4) while the bract of S. hermonthica is much shorter than the calyx (Fig. 2). The entire external portion of the corolla of S. aspera is covered with glandular hairs (Fig. 2) while S. hermonthica lacks glands (Fig. 2). Additional differences include the more abrupt angle of the corolla tube of S. hermonthica (Fig. 2, left) and the larger lower leaves. The two species are also somewhat different in their ecology, S. aspera being more commonly found in natural grasslands and S. hermonthica as a weed in sorghum. However, S. aspera has also been found to parasitize sorghum but has been invariably assumed to be S. hermonthica in such cases. Plants of S. aspera from Kuffena Rock, a massive granite Inselberg near Zaria, were diminutive but possessed the characteristic bracts and pubescence (Fig. 5). STRIGA ASIATICA (L.) Kuntz The most widely distributed of all species in the genus and a serious pathogen throughout most of its range, S. asatica is not a problem in Nigeria despite the planting of corn (Zea mays) on a large scale in some places. Plants were growing on an ironstone outcropping near Tegina, Kaduna State, in a stand of native grasses. The overall form of these plants was quite different from that of the strain introduced to the United States (cf. Fig. 6 and 7). The Nigerian plants were much more pubescent than the American strain (Fig. 8). The American strain has recently been shown to be autogam- ous (Nickrent and Musselman 1979). While more careful studies are neces- sary, it appears that plants in the Tegina population were not autogamous. STRIGA GESNERIOIDES ( Willd.) Vatke Unlike the other species in this study, S. gesnerioides does not parasitize grasses. It is a serious parasite of cowpeas (Vigna unguiculata). In other parts of its range it is known to parasitize tobacco (Hosmani 1978). Two dis- tinct strains were observed in this study. The first is the widespread cowpea strain (Figs. 9-12), the second was seen only at Mokwa, Niger State and was parasitizing a weedy legume Tephrosia pedicellata (Figs. 13, 14). To our knowledge, there is no formal taxonomic difference between these two strains. They are, however, quite different in some aspects of their morphology and growth habit. The flowers of the Mokwa strain were more pink than the cowpea strain, the size of the bract was also different (cf. Fig. 9 with 13). Most, striking, however, is the general habit of the plant. The Mokwa strain only rarely branched (Fig. 14), while the cowpea strain branched repeatedly at the ground level (Fig. 9). The strain of S. gesnerioides recently introduced into Florida most closely resembles the Mokwa strain (Wunderlin, Musselman and Shuey 1979). STRIGA MACRANTHA Benth. Unlike the four species discussed above, S. macrantha and S. klingii are 200 Fig. 9. Cowpea strain of Striga gesnerioides, La. GRA (Scale for silhouettes equals 10 cm, Inflorescence of cowpea strain a nd of beate relative to calyx. Fi ig. 13 Fig. 14 Fig. 15. Fig. 106. Fig. 17. Fig. 18. . Mokwa strain of S. gesnerioides, , cL. strain of S. aleneieee showing a: Inflorescence of S. macrantha, of S. klingii, Inflorescence Tegi Mokwa; Flower and bract of S. &lin Face view of corolla, S. Mac aan na; Musselman and Mansfield, Kazure; Musselman and Mansfield, spect. Musselman and Mansfield, Musse for drawing 1 cm) + Has A Nh) [Lad Woy \ Vy \ iA. ad hy oh ae LV Iman and Mansfield, 2. Individual flowers of cowpea strain of S. gesnerioides show size 5534. D225 5533 201 not known to be of any economic importance. Striga macrantha was collected near Tegina, Kaduna State, on an ironstone outcrop where it was growing with S. astatica in rocky soil amongst grasses. It was up to 1.5 m tall and had large white flowers, apparently the only Nigerian species characterized by that corolla color. The flowers are crowded on a dense spike (Figs. 15, 18) and the bracts and calyces are densely pubescent with both glandular and non-glandular hairs. STRIGA KLINGU (Engl.) Skan Great variation in the size of plants were observed in this species. It is apparently a characteristic plant of the dense Hyparrhenia (Poaceae) grass- lands of the Southern Guinea Savanna. When growing amongst tall grass, the plants are quite short while those on the margin of the grasslands are considerably more robust. This species is characterized by small tg flowers in crowded spikes and hispid hairs on the large bracts (Fig. 16, 1 A workshop was held in Khartoum, Sudan in November eae on Striga and Orobanche. Of major concern were S. asiatica, S. angustifolia, and S. hermonthica although some time was devoted to the S. gesnerioides problem. he emphasis of the meeting was on control in its broadest sense including breeding for resistance/tolerance in host species. At present one of the more promising means of control is the use of germination stimulants that cause seeds of the parasite to germinate without a host—thus committing “suicide”. A group of compounds developed at Sussex University are now being tested for their efficacy in this regard. A summary of this meeting and suggestions for future work will be published in PANS. One critical area includes research on taxonomy and host specificity. Thus, much opportunity exist for taxonomic research in the genus Striga. Unfortunately for American botanists interested in conducting a biosystematic study, the species are under federal quarantine and cannot be grown in the United States without special permission. REFERENCES HEPPER, F. N. 1963. Scrophulariaceae. In: J. Hutchinson and J. M. Dalziel. Flora of West Tropical Africa. Second Edition. Crown Agents for Overseas Governments, London. HOSMANI, fe - 1978. Striga (a noxious root parasitic weed). Published by the author, Dhar NICKRENT, D. L ad L. J. MUSSELMAN. 1979. Autogamy in the American strain of witchweed. Hee 31: 253-256. WUNDERLIN, P., L. J. MUSSELMAN and A. G. SHUEY. 1979. Striga ges- Deregiees cei 2 Vatke first record of the species in the New World. Plant Dis Rep. 61: 251-25 NOTES HYPTIS MUTABILIS (LABIATAE) IN SOUTHEASTERN UNITED STATES.—In September 1977 I collected, in far southeastern Louisiana, a member of the Labiatae that was not at the site when I was there in summer from 1963 to 1972. After some struggle through a scant and not-too-helpful literature, I finally identified the species as Hyptis mutabilis (L. Rich.) Briq., a weedy plant widespread in warm America and new to Louisiana. This paper is the result of my curiosity concerning H. muwtabilis. It sum- marizes literature on the species, provides illustrations of the species (there apparently are no good, easily accessible illustrations in the literature ), and gives data on its occurrence in southeastern United States. By the most recent monographer of Hyptis (Epling 1949), H. mutabilis was recognized as a single, highly variable species not divisible into infra- specific taxa (Epling 1949) although in older works it was treated as a species with several varieties or even as several species (Epling 1933, 1936). Its representatives in the United States were treated as two species or as two varieties of one species. Such distinctions fail when many specimens are examined. In Epling’s firs’ work on Hyptis (Epling 1933), he used “var. spicata Brig.” for those representatives of H. mutabilis in the West Indies and the Caribbean coastal regions. (Presumably var. spicata is what today would be called var. matabilis, although this point is not resolvable from Epling’s prose; he never was able to locate the holotype of H. muwtabilis.) In his synopsis of South American Labiatae, Epling (1936) continued to use “var. spicata Brig.” in the same way. However, he commented: “I have sought to indicate here the principal modes of variation within this puzzling com- plex [H. mutabilis],; however, it is impossible to refer numerous specimens with certainty to any one group [variety], and I hesitate to cite specimens within the divisions [varieties] I have indicated.” In Epling’s last work on Hyptis (Epling 1949), he reduced spicata and other varieties recognized in 1933 and 1936 to synonymy under H. mutabilis. Abandoning efforts to distinguish infraspecific taxa, he wrote: “. . . parece que no posse razas bien definidas” (he did, however, refer to his 1936 work where “La indiqué.. . las variantes principales”). In all his papers, he recorded H. mutabilis only from Florida in southeastern United States. The author of the epithet mwtabilis is frequently and incorrectly given as “A. Rich.” (e.g., in Epling 1933, 1936; Small 1933). ia actual author, Louis Claude Marie Richard (usually cited as “L. Rich.” or “L. C. Rich.” ), published Nepetha [sic] matabilis in 1792. Louis eee pea his atten- tion from Nepetha and fathered Achille Richard (“A. Rich.”), who was not born until 1794. SIDA 8(2): 202, 1979. 203 Hyptis mutabilis was recorded from southeastern United States as early 130 years ago. Bentham (1848), under H. spicata, included the note “Florida (h. Torrey! ),” evidently based on a collection (or possibly two) he had received from the herbarium of John Torrey. Two undated sheets— one at NY, one at GH—appear to be duplicates of the specimen(s) Bentham saw; they bear, among equivocal collection data and in handwriting, the annotation H. spicata and the notes “fide Bentham” (NY) and “Benth. in lit.” (GH Gray (1878) ascribed H. spicata to “S. Florida.” Chapman (1889) re- ported it from “Tampa and Jacksonville, Florida’; he gave the same dis- tribution later (Chapman 1897), but the word “introduced” was added with- out explanation (and possibly without justification). Small (1903) recog- nized, under the name Mesosphaerum, two species in the mutabilis complex: Mesosphaerum spicatum and M. mutabile. The former was ascribed to Florida and Alabama, the latter to Florida. (I cannot verify, through herbarium speci- mens, the presence of H. mutabilis in Alabama that early.) Small later (1933) withdrew the Alabama record: using the name Hyptis, he cited H. mutabilis only from Florida and said, of H. spicata, “not now definitely known rom our range.” More recently, a good account of H. mutabilis is that by Standley and Williams (1973), who gave a detailed description of the species but did not ascribe it to conterminous United States at all. Long and Lakela (1971) gave the range as “Fla. to Va. The specimens of H. mutabilis | have seen suggest that, in southeastern United States, this species was originally only from Florida. From that state it has, in the last 3.5 decades, spread northward and westward, reaching Louisiana very recently. The earliest specimens of H. mutabilis available to me from southeastern United States (herbarium of John Torrey, NY, ) were, as indicated above, collected in Florida before 1848. All other pre-1900 specimens are from Florida. Indeed, the earliest non-Florida specimen I saw was collected in southeastern Virginia in 1939 by Fernald and Long. That this specimen represents a non-persistent introduction is suggested by the facts that (1) it is the only Virginia collection I saw and (2) the collection locale is sepa- rated by 500 miles from the main U.S. range of the species. The northward and westward spread of H. mutabilis can be seen in exsiccata: the earliest western Florida (te. the Panhandle) collection I saw is from 1897 (the species was only rarely collected there until the 1940s and later); the earliest Georgia collection, 1940; Alabama, 1952; Mississippt, 1968; and Louisiana, 1977 (my collection was made just a few hundred feet west of the Louisiana- Mississippi border, which the species had obviously just crossed in its west- ward push. The documented distribution of H. mutabilis in southeastern SIDA 8(2); 203. 1979. 204 Fig. Hyptis mutabilis. Documented distribution in southeastern United States ee locale omitted ) United States is shown in Fig. 1. Available specimens of US. H. mutabilis suggest that the species has become much commoner in recent decades. The decade-distribution of speci- mens seen by me is as follows: pre-1848, 1 collection (possibly 2); 1850s, none; 1860s, none; 1870s, 2; 1880s, 1; 1890s, 9; 1900s, 5; 1910s, 1; 1920s, 1; 1930s, 2; 1940s, 7; 1950s, 16; 1960s, 43; and 1970s, 34 (through 1976 except my 1977 Louisiana specimen ). It is, of course, possible that specimen-based data on spread and abundance of H. mutabilis represent vagaries of collection and collectors rather than biological verities. But the data nevertheless strongly support my speculations. Of the 127 conterminous U.S. collections of H. mutabilis | have seen, 93 are from Florida, 21 from Georgia, 6 from Alabama, 5 from Mississippi, 1 from Louisiana, and 1 from Virginia. SIDA 8(2): 204. 1979. 205 Fig. lyptis mutabilis. Flower (left) and fruiting calyx (right) (based on as eee Curtiss 5112, Jacksonville, Florida, NY) The vertical line represents 0). Hyptis mutabilis occurs mostly in disturbed places (e.g., roadsides, rail- road yards, orchards, gardens, lake shores, river banks, fallow fields, ballast ground, thickets, and grazed areas). Some label data, though, suggest that it grows also in less disturbed habitats (e.g., stabilized dunes, hardwood or softwood forests, and hammocks). It is at home in moist or dry soil and in sun or shade. The method of its dissemination is unknown. No label data or literature reports suggest that H. mwtabilis is a troublesome weed. My Louisiana collection (St. Tammany Parish: 8 mi SE of “Slidell along hwy. 190, 8 Sep 1977, Thieret 50223, KNK) was made from a colony of about 25 plants in a roadside weedy area bordering a channel of the Pearl iver. Associates were mostly grasses, including Paspalum spp., Cenchrus incertus, and Eragrostis oxylepis. | especially noted two characteristics of the specimen I collected. First, its corollas, though small, were relatively showy, being oe purple with white blotches; and second, its foliage, crushed, ha oma so delightful that H. ables is certainly one of Louisiana’s eat mints, perhaps second only to Satureja georgiana. The best identifying characteristic for H. mutabilis is the calyx (Figs. 2, 3). About 1.5-2.0 mm long, and campanulate in flower, it enlarges consider- SIDA 8(2): 205. 1979. 206 Fig. 3. Hyptis mutabilis. Section of fruiting inflorescence (Mitchell 350, Jackson Co., Florida, FSU). The vertical line represents 1 cm ably and quickly in fruit, becoming as much as 7.5 mm long and tubular. The 10 longitudinal veins increase in prominence, and strong cross veins develop to connect them. The accrescent calyx was first noted by Richard SIDA 8(2): 206. 1979. 207 (1792), the describer of the species, who wrote, “calycibus defloratis elon- gatus’; the characteristic was probably the basis for the epithet »utabilis. The following description of H. mutabilis is based primarily on specimens, secondarily on literature. Fig. 4. Hyptis mutabilis. Section of plant with dense inflorescences (Duncan 17187, Wayne Co., Georgia, GA). The vertical line represents 4 cm SIDA 8(2): 207. 1979, 208 Plants herbaceous, erect, to 2.2 m high, often much branched, the branches glabrous to villous. Leaves membranaceous, 2.0-18.0 cm long, narrowly to broadly ovate to rhombic (rarely subrotund or even reniform), acute to acuminate at the apex, attenuate, cuneate, truncate, or subcordate at the base, Hyptis mutabilis. Section of plant with sachs ae inflorescences (Shuey eae oa Co., Florida, USF). The vertical line represents 4 cm SIDA 8(2): 208. 1979. 209 finely to coarsely crenate to serrate, sometimes doubly so, variously pubescent to glabrate, passing abruptly or gradually into bracts; petioles 0.5-8.0 cm long. Flowers sessile to very short pedicelled, in bracteate verticels, these arranged in dense or interrupted (Figs. 4, 5), leafy or naked, terminal, spiciform inflorescences 3.0—20.0 cm long; bracts ovate to elliptic, acute to acuminate, prominently nerved. Calyx tubular to narrowly campanulate and 5-2.0 mm long at anthesis, tubular, reticulate-veined, and to 7.5 mm long in fruit, the teeth subulate to setaceous, erect, 0.7-1.2 mm long. Corolla 3.0-5.0 mm long. I am grateful to the curators of the following herbaria for loan of speci- mens: A, FSU, GA, GH, MISS, NCU, SMU, TEX, US, and USF. Dr. Carroll E. Wood, Jr., gave help with the literature—John W. Thieret, Faculty of Biological Sciences, Northern Kentucky University, Highland Heights, 41076. REFERENCES BENTHAM, G. 1848. Labiatae. DC. Prodromus 12:27-603. [H. spicata, 121.} CHAPMAN, A. W. 1889. Flora of the southern United ao 2nd ed. Ivison, Blake- man, & Company, New York. 698 pp. [H. spicata, 637. = . 1897. Flora of the southern United States. 3rd ed. American Book Com- ny, New York, 655 pp. [H. spicata, ee EPLING, C. 1933. Synopsis of the genus Hyptis in North America. Repert. Spec. Nov. Regni Veg. 34:37-130. [H. nn 103-106. 1936. peas of the South American Labiatae, 3. Teil. Repert. Spec. Nov. Resa Veg. Beih. 85:193-288. [H. mutabilis, 259-262.} 1949. See del género Hyptis. Revista Museo La Plata, Secc. Bot., N.S. 7:153- 497. [H. mutabilis, 317-318.} Bee A. 1878. Synoptical flora of North America, Vol. 2, Part 1. Gamopetalae after Compositae. Ivison, Blakeman, Taylor, & Company, New York. 402 pp. [H. spicata, 350; account repeated in ed. 2, 1886.} , R. W., an LAKELA. 1971. A flora of tropical Florida. University of Miami Press, Coral Gables, Florida. 962 pp. [Hyptis, 748-749. RICHARD, L. 1792. Catalogus plantarum, ad societatem, ineunte anno 1792, Cayenna missarum a Domino le Blond. Actes Soc. Hist. Nat. Paris 1(1) :105— 114, [Nepetha mutabilis, sp. nov., 110. SMALL, J. K. 1903. Flora of the southeastern United States. Published by the author, New os 1370 pp. [Mesosphaerum, i.e., Hyptis, 1052-1053; account re- ee in - 25 023): Tae of the southeastern Flora. University of North Carolina Press, Chapel Hill. 1554 pp. Deda 1180-1181.] STANDLEY, P. C., and L. O. AMS. 1973. Labiatae. Mint Family, 317. In a of Coens ee Bot. 24, Part IX, Number 3. [H. mutabilis, 254.} TRILLIUM PUSILLUM (LILIACEAE) IN MISSISSIPPI—Trillivm pusil- lum Michx. is apparently one of the rarer species of the eastern United States. Roe (1978) considers the species to consist of four more or less disjunct populations probably of varietal status. His map shows a wide dis- tributional gap between South Carolina and Texas, which has in part been SIDA 8(2): 209, 1979, 210 filled by recent collections from Sumner Co., Tennessee and Madison Co., Alabama (R. Kral, pers. comm.). We here report T. pusillum from Missis- sippi, thus further bridging this disjunction. 1978 we first collected plants of an unknown Trillivm with pedicellate fruit from Jones Co., Mississippi. It was initially noted as being somewhat different from all previously known species of the genus in Mississippi, but was not definitely identified until comparison with material at US. It is apparently most similar to material from South Carolina and North Carolina. The purpose of this paper is not to clarify the status of the varieties, but the material would therefore be referable to var. pusillum. The later collection of flowering material further confirmed the identity of this plant with T. pasillum. The habitat in which Trillium pusillum was collected by us is a rich alluvial floodplain dissected by several small creeks which empty into Bouge Homa Lake. Inundation occurs probably at least once a year. The first time we collected this species, it was necessary to wade through floodwater to get to the area. The second collection (in flower) was made only one day after the entire bottomland had been flooded. The bottomland forest where Trilliam pusillum was collected has been disturbed by selective logging, but is still reasonably well-shaded by uncut hardwoods. Dominent species are Pinus glabra Walt., Fagus grandifolia Ehrh. and Magnolia grandiflora L. associated with many other hardwood species such as Ulmus americana L., Fraxinus caroliniana Miller, Caprinus caroliniana alt., Quercus lyrata Walt, QO. michauxii Nutt. and Q. nigra L. Associated herbaceous species include ‘Sons parviflorus Raf., Sisyrinchium angusti- foliwm Miller and various species of Viola, Carex, Panicum and Polygala. There is a combination of factors which may have been responsible for Trillium pusillum having been overlooked in Mississippi and possibly other areas before now. First, it flowers very early in the spring, in fact is one of the earliest plants of the area. Also in some years the plant may be nearly inaccessible due to flooding. Certainly the habitat is not unique and sites similar to that where T. pwsillum was collected by us are scattered across the southern portions of Mississippi, Alabama, and Georgia. Specific data for the collections mentioned above are given below. MISSISSIPPI. Jones Co.: 7 m E Laurel, disturbed beech-magnolia-spruce pine woods, wet ground near ne road, 9 May 1978, Morgan & Me- Daniel 468 (IBE, MISSA, duplicates to be distributed), 7 Mar 1979, Morgan 1589 (IBE, MISSA, duplicates to be distributed )—David Morgan and Sid- ney McDaniel, Dept. Biological Sciences, Mississippi State University and Institute for Botanical Exploration, Box EN, Mississippi State, MS 39762 REFERENCE ROE, G. F. 1978. Additions to the Range of Trillium pusillum. Castanea 4: 191-195. SIDA 8(2): 210. 1979. 211 RANGE EXTENSIONS OF BRAZORIA PULCHERRIMA LUNDELL (LAMIACEAE ). —Brazoria pulcherrima Lundell is now known from four counties growing in disturbed loose white sand as a pioneer species. In addition to collections from Leon County, collections from the other three counties are (SMU): Freestone County—Mahler 8552, Anderson County— Mahler 8553; Houston County—Mabler 8545. This taxon was one of the proposed taxa for the Federal List of Endangered Species (Federal Register 41 (117): 24549. 1976) since it was known only from a local area. Funds for this field study were provided by the office of Endangered Species, U.S. Fish and Wildlife Service, Albuquerque, New Mexico. The range of distribution of the species has not been greatly increased but its role as an invader species of recently distributed loose white sands enables the taxon to apparently co-exist with man and his activities even though it is a rare endemic to Texas—IWm. F. Mahler, SMU Herbarium, Dallas, TX 75275. RUBUS TRIVIALIS MICHX. VAR. DUPLARIS (SHINNERS) MAH- LER, COMB. NOV. (ROSACEAE ).—Based on Rubus duplaris Shinners, Field & Lab. 22: 27. 1954. Holotype: Freestone County, 13.6 mi S of Fair- field, 2 May 1953, Shinners 14465 (SMU!). Rubus duplaris Shinners was one of the proposed taxa for the Federal List of Endangered Species (Federal Register 41(117): 24562. 1976) since it was known from only two locations (counties) in Texas at that time. Subsequent field work has resulted in expanding the known distribution of this taxon and observations of the variability of the amount and type of pubescence warrant the recognition of this taxon at the varietal level. This variety commonly grows in the deeper sands (Carrizo) of the oak- red, glandular hairs on the upper and lower leaflet surfaces of both primo- canes and floricanes. In adjacent areas of shallower sands and sandy loam soils, the amount and type of pubescence varies on the primocane and flori- cane leaflets. The specimens from Angelina County are intermediate and possess few red, glandular hairs on the leaflet surfaces. In some locations, Rubus trivialis var, trivialis and var. duplaris grow together with varying amounts of abundance and type of pubescence on the leaflets of the primo- canes and floricanes on the same plant. The Rare Plant Study Center of the University of Texas reported that TEX did not have any collections of Rubus duplaris (A preliminary survey of the a of proposed endangered Texas plants (FR 6/16/76). 19772 U.S: PWS): The range of the distribution of this variety is given in Figure 1 and documentation includes more than thirty specimens (SMU) collected by SIDA 8(2): 211. 1979. ya Wes Trans—Pecos Gale cree oo Fayette Pra Gulf Cone ere 13 VOak-pine he 14 Big Thicke VEGETATION ZONES OF TEXAS FOUWUMNDAUFWNE w= aQ ro 3 > Oe Dd i. n y = rt ny ry) | oon co: = — io = ° * o n rs 3 Hee 1. Documented distribution of Rvbuws trivialis Michx. var. duwplaris (Shinners ) Fig Mahler. Shinners, Kral, Cory, McVaugh, Lipscomb, and Mahler. The vegetation zones in Figure 1 are a compilation of past researchers. I am indebted to Barney Lipscomb and Larry Lodwick for their assistance and to ee se of Endangered Species, U. S. Fish and Wildlife Service, er ee , New Mexico for providing travel funds—Wm. F. Mabler, MU a, Dallas, TX 75275. SIDA &(2)2 212. 1979, 215 A NOTE ON THE DISTRIBUTION OF STYRAX AMERICANA (STYRACACEAE) IN TENNESSEE.—The range of Styrax americana Lam. var. americana, the American Snowbell, is primarily confined to the Atlantic and Gulf Coastal Plain, Piedmont, and Mississippian Embayment as indi- cated by the distribution map of Gonsoulin (Sida 5: 191-258, 1974). The rarity of this taxon in the Interior Low Plateau in general and the apparent absence of it from this Province in Tennessee or from other provinces east- ward in Tennessee is striking. This absence was also noted by Shanks in his lists of Tennessee woody plants (Journ. Tenn. Acad. Sci. 27: 27-50, 1952, 28: 1958-1959, 1953; Castanea 17: 90-96, 1952). In fact, Shanks included it as an indicator species of the lowland forests of the Mississippian Embay- ment in western Tennessee (Journ. Tenn. Acad. Sci. 33: 195-210, 1958). However, recent collections from Stewart County (Schibig, 514, APSU, VSCC, Chester 3288, APSU) definitely adds this species to the ILP Province in Tennessee. Another fact of significance is that the Hamilton County collection cited in the Gonsoulin text is not indicated on the distribution map; hence the addition of this species to the eastern Tennessee flora may be overlooked if only the map is consulted—Joe Schibig, Volunteer State Community College, Gallatin, TN 37066 and Edward W’. Chester, Austin Peay State University, Clarksville, TN 37040. ——t NOTES ON PLANTS OF MISSISSIPPI II. ADDITIONS TO THE FERN AND FLOWERING PLANT FLORA—In the course of field work on the flora of Mississippi, in 1978-79, I have found some ferns and flowering plants which appear not to have been previously collected in Mississippi. None of the species of ferns reported here were included in the most recent treatment of pteridophytes of Mississippi, Evans (SIDA 7(3): 282-297. 1978). Herbarium specimens are in the University of Tennessee, Vanderbilt Uni- versity, and Museum of Natural Science (Jackson). PTERIDOPHYTA Asplenium resiliens Kunze Tishomingo County: Pickwick Lake: Wooded bluffs between Eastport Beach and entrance to Yellow Creek. Crevices of shale. Rare. 4-5 Oct 1978, Rogers 46125, 46195, Athyrium thelypteroides Desv. Hinds County: About 4.0 mi N of Edwards from I-20: Mesic woods in ravines in Loess Hills region. Rare. 20 Oct 1978, Rogers 46264, Tishomingo County: Yellow Creek peninsula, cove N of Cooks Landing. Rare in mesic woods, 19 May 1979, Rogers 46710. SIDA 8(2): 213. 1979. 214 Cheilanthes alabamensis Kunze ilkinson County: Clark Creek natural area: dry border of mixed woods on a road cut. Rare. 7 Aug 1978, Rogers 45417. Cyrtomium falcatum Presl Adams County: About 10 mi NW of Natchez just N of Hwy 555, T8N, R2W, Sect 27; wooded ravines in Loess Hills region. Frequent locally. 10 May 1978, Rogers 45178. Adams County: Loess Hills just N of Quitman Road at the delta, T8N, R3W, Sect. Several plants growing on steep ravine slopes. 16 Aug 1978, Rogers 45540, Adams County: Loess Hills just N of Hwy 552, about 3 mi W of Mt Zion Church, T12N, R13E, Sect 32. Several plants on ravine slopes. 16 Aug 1978, Rogers 45544. Pellaea astropurpurea Link Pohoniees County: Yellow Creek peninsula, T1S, R16W, Sects 24 and 30; in second-growth hardwood-pine on calcareous shale bluffs. Rare 5 Oct 1978, Rogers 45544. Selaginella uncinate Spring Jefferson County: About 1/4 mi W of Alcorn State University farm and campus: ravine slopes and bottom, in full sun. A single large population seen, numbering several hundred individuals. Rare and local. 5 Apr 1978, Rogers 45062. SPERMATOPHYTA Carex grayi Carey Tishomingo County: Yellow Creek Quad, T1S, RI1E, Sect 30, first cove N of Whetstone Branch. Rare and local in rich mesic woods, 15 May 1979, Rogers 46640. Dicentra cucullaria (L.) Bernh. Tishomingo County: Yellow Creek Quad, T1S, RIOE and RIIE, Sects 24, 25, and 30. Rare and local in rich woods and on shale bluffs. 10 Apr. 1979, Rogers 40443. Hydrophyllum appendiculatum Michx. Tishomingo County: Yellow Creek peninsula, T1S, R1I1E, Sect 30; first cove north of Whetstone Branch. Rare in moist cherty deciduous woods. 15 May 1979, Rogers 46638. Hydrophyllum macrophyllum Nutt. Tishomingo County: Yellow Creek peninsula, T1S, R11E, Sect 30; first cove N of Whetstone Branch. Rare in moist rocky deciduous woods. 15 May 1979. Rogers 46637. Kochia scoparia (L.) Roth SIDA 8(2).: 214. 1979. 219 Hinds County: Jackson: along railroad tracks W of Travelodge Motel at Amite and Adams streets. 9 May 1979, Rogers 46621. Phacelia bipinnatifida Michx. Tishomingo County: Yellow Creek Quad, T1S, RIOE and R1I1E, Sects 24, - 25, 30. Rare in rich mesic woods and wooded shale bluffs. 10 Apr 1979, Rogers 46442. Philadelphus hirsutus Nutt. Tishomingo County: Yellow Creek Quad, T1S, R10E, Sect 14. Very rare on calcareous shale bluffs, in pine-oak woods. 15 May 1979, Rogers 46628. Scutellaria ovata var. versicolor (Nutt.) Fern. Tishomingo County: Yellow Creek peninsula, T1S, R11E, Sect 30. First cove N of Whetstone Branch. Rare in rocky, mesic deciduous woods. 15 May 1979, Rogers 46643. Taenidia integerrima (L.) Drude Tishomingo County: Yellow Creek Quad, T2S, R11E, Sect 31. Whetstone Branch; dry rocky south-facing slopes; several plants occurring locally. 11 Apr 1979, Rogers 46503. Tradescantia ernestiana Anders. and Woodson Tishomingo County: Yellow Creek Quad, T1S, R11E, Sect 30; Boo Hollow. Rare in mesic woods. 15 May 1979, Rogers 46650. Trautvetteria caroliniensis (Walt.) Vail Tishomingo County: Yellow Creek Quad, T2S, R11E, Sect 6. Very rare and localized along Short Creek, on calcareous shale, in seepage areas E of road leading to Cooks Landing. 9 Apr 1979, Rogers 46502. Viola pensylvanica Michx. Tishomingo County: Yellow Creek Quad, T1S, R10E, Sect 24; Meat Hol- low. Several plants locally in cherty moist deciduous woods. 14 May 1979, Rogers 46629—Ken Rogers, Mississippi Museum of Natural Science, 111 North Jefferson Street, Jackson, MS 39202. NEVIUSIA ALABAMENSIS (ROSACEAE) IN MISSISSIPPI—Neviusia alabamensis A. Gray, commonly known as Snow Wreath, is a low shrub in the Rose Family. It has been known previously from Tuscaloosa County in Alabama, Conway County, Arkansas, and Butler County, Missouri (Clark, 1971; Small, 1933; Steyermark, 1963). In October 1978 I found a large colony of Snow Wreath, in the hilly woodland of northeast Mississippi, in Tishomingo County, growing on calcareous shale bluffs overlooking the Tennessee River and Pickwick Lake. A second colony was found in November in the same stretch of bluffs. Sun Wreath was growing in full sun on the bluffs which lie in a narrow belt from Eastport northeast to Yellow Creek. Several species of plants found SIDA 8(2); 215, 1979. 216 here are either unique to the bluffs or occur rarely elsewhere in Mississipp1. A few plants associated with Snow Wreath were Dodecatheon meadia, Pellaea atropurpurea, Woodsia obtusa, Solidago sphacelata, Heuchera villosa var. macrorhiza, Delphininm tricorne, Cladrastis lutea, Euonymus atropur- pureus, Fraxinus quadrangulata, Quercus mublenbergii, Quercus prinus, Aesculus glabra. Herbarium specimens (Rogers 46149, 46652, 46670) are in The Univer- sity of Tennessee, Vanderbilt University, and Museum of Natural Science (Jackson). —Ken Rogers, Mississippi Museum of Natural Science, 111 North Jefferson Street, Jackson, MS 39202. THE CHROMOSOMES OF COLLINSONIA CANADENSIS | L. (LABIATAE).—The following preparations represent the first published illustrations of the chromosomes of Collinsonia, a genus of mints confined to eastern North America. The mitotic chromosomes (27 = 50) are shown in Figure 1 and the meiotic chromosomes (7 = 25) in Figure 2—all derived from Collinsonia canadensis L. (PA, Centre Co.: C. S. Keener 1978, PAC). — Carl §. Keener, Department of Biology, The Pennsylvania State University, University Park, PA 16802. * ° na * mes on? 5 80° ~ % 4 . a ga * ft a's oe 84% + nm” Fe, 5 ~ a x ee - caf 0.01 mm ’ NICOLLETIA OCCIDENTALIS (COMPOSITAE) IN BAJA CALI- FORNIA!—A specimen at SD (Valle de San Felipe, southeast of Santa Clara, ca. 31°03'N, 115°13'W, with Larrea, Fouquiera, Simmondsia, and Cercidium at ca. 400 m., 3 Apr 1977, Fred Sproul s.n.), brought to my attention by Dr. R. Moran, documents the occurrence of Nicolletia occidentalis in Baja California. This collection extends the known range of the species some 300 km. south to within ca. 100 km of the nearest known population of N. trifida (cf. my review of Nicolletia, Sida 7:369-374. 1978). The Parishes are not known to have collected in Lower California (cf. ibid.)—John L. Strother, Department of Botany, University of California, Berkeley, CA, 94720. SIDA 8(2): 216. 1979. SIDA susie" VOLUME 8 NUMBER 3 JUNE 1980 CONTENTS The systematics of Hedeoma (Labiatae). Robert S. Irving. 218 Historical account 219 Generic limits and relationships 220 Chromosomal studies ae Breeding habits 223 Natural interspecific hybridization 224 Specific concepts 725 Ecology, distribution, and possible origin 225 Systematic treatment 229 Excluded names 292 Acknowledgements 294 References 294 Taxonomy of Tetragonotheca ( Asteraceae-Heliantheae ). B. L. Turner and Deborah Dawson. 296 New species of Clitoria subgenus Bractearia section Brachycalyx (Leguminosae) from Colombia, Venezuela, and Brazil. Pawl R. Fantz. 304 NOTES. A chromosome count for Juniperus ashei (Cupressaceae) and additional chromosome numbers for Hedeoma (Labiatae). 312—-New and. te- aes combinations in Guwtierrezia (Compositae: Astereae). 313—He sper aloe funifer. (Agavaceac) in Texas. 314—Cwueurbita digitata (C ucarbitnceae ) in Phyllanthus polygonoides (Euphorbiaceae) new to Arkansas. 315. US ISSN 0036-1488 SIDA, CONTRIBUTIONS TO BOTANY Founded by Lloyd H. Shinners, 1962 Editor & Publisher Wm. F. Mahler SMU Herbarium Dallas, Texas, 75275 Associate Editor Assistant Editor John W. Thieret Barney L. Lipscomb Northern Kentucky University SMU Herbarium Highland Heights, Kentucky, 41076 Dallas, Texas, 75275 Guidelines for contributors are available upon request beginning with volume 8. Subscription: $6.00 (U.S.) per year; numbers issued twice a year. Sida, Contributions to Botany, Volume 8, Number 3, pages 218-315. Copyright 1980 by Wm. F. Mahler Pare ss ae ly | = a is , ‘ > 7 7 a ss tf tne ee ee ee - i 7 7 ‘<4 ' | ae < — re qeed - cs = 7 a 7 4 = Frontispiece. Hedeoma microyhyllum Irving. THE SYSTEMATICS OF HEDEOMA (LABIATAE) ROBERT S. IRVING 1422 Summit, Little Rock, AR 72202 The genus Hedeoma (Labiatae) is a group of annual and perennial herbs found throughout southwestern United States and Mexico and in scattered localities in South America. In this treatment it is composed of 38 species and 5 varieties. Hedeoma is a member of the subtribe Melissinae of the tribe Stachyoideae of Briquet (1897). Its generic boundaries have undergone widely divergent modifications during the last 170 years, as indicated by the publication of Persoon (1807), Bentham (1832-1836, and 1848), Gray, (1878), Briquet (1897), and Epling and Stewart (1939). The intent of this work, although by no means definitive, is to present a coherent delimitation of Hedeoma based on extensive field and experimental data. In addition to the problems of generic limits and phyletic position of Hedeoma, there are numerous problems at the specific and intraspecific levels. In previous works, including the most recent inclusive treatment of the group by Epling and Stewart (1939), there was little or no attempt to explore questions of intra- and interspecific relationships or phylogeny. Such an exploration is one of the primary goals of this monograph, and it is believed that the present treatment to a great extent, places the taxa of Hedeoma in natural groupings not only from the standpoint of morphology, ecology, and geography but phylogeny as well. A list of representative specimens is available on request. HISTORICAL ACCOUNT The genus Hedeoma was proposed by Persoon in 1807 to accommodate three species: H. thymoides, H. glabrum, and H. pulegioides. In Bentham’s (1836) treatment of Hedeoma, H. thymoides and H. glabrum were excluded from the eae H. pulegioides was retained as the type species and seven other newly described or previously unrecognized species were added. He also divided the genus into two sections, Mosla and Hedeoma. Between 1848 and the publication in 1876 of the Genera Plantarum, Bentham elevated the section Mosla to generic status and added four additional species to Hedeoma proper bringing the total to 12. Gray’s treatment in 1878 resulted in a circumscription very similar to the present one. Gray recognized eight additional taxa as belonging to Hedeoma, and divided the genus into the sections Exhedeoma, which includes most of the SIDA 8(3); 218-295, 1980. 220 currently accepted species, and Stachydeoma. The latter section was elevated to generic rank by Small (1903); later its two species were disaggregated into the monotypic genera Rhododon and Stachydeoma (Epling and Stewart, 199?) In Briquet’s (1897) treatment of Labiatae, Hedeoma was expanded to include Keithia Benth., Poliomintha Gray, and other related genera. How- ever, Epling and Stewart (1939) returned to the earlier generic delineation of Bentham and of Gray. They believed that Bentham’s account was a more natural, logical, and consistent interpretation of the relationships involved than that of Briquet, who, for lack of a single clearcut difference between these genera lumped them together. The present treatment is essentially that of Epling and Stewart and | am in full agreement with their arguments against Briquet’s expanded generic concept. My treatment differs in that three of the four North Ameri- can representatives of Epling’s genus Hesperozygis have been incorporated into Hedeoma and several new taxa have been added. GENERIC LIMITS AND RELATIONSHIPS There are 11 other genera which, on morphological grounds, are con- sidered closely allied to Hedeoma. Conceivably, all of these genera, including Hedeoma, could be considered to constitute a single very large genus, in fact seven genera, Mosla, Rhododon, Stachydeoma, Poliomintha, Eriothymus, Hoebnea, and Rhabdocaulon have at one time or another been merged with Hedeoma. The genus Z7z/phora, of the Mediterranean region, can be considered the Old-World counterpart of Hedeoma as its overall habit and floral morphology are similar. In Bentham’s treatment of the Labiatae (1832-1836, and 1848), Ziziphora was placed in the tribe Monardeae, thus suggesting considerable morphological divergence of Hedeoma from Ziziphora. Briquet (1897), on the other hand placed Ziziphora next to Hedeoma in his subtribe Melissinae. Ziziphora and Hedeoma can be distinguished by several morphological fea- tures. Z/z/phora possesses a tubular calyx with connivent, subequal, narrowly triangular teeth—a feature very reminiscent of the calyces of Poliomintha. Further, Z7z/phora is separated by a technical character of the anthers; unlike Hedeoma, the lower anther-sacs of Ziziphora are completely or partially aborted The diverse genus Cumila, despite an obvious morphological affinity with Hedeoma, has always been placed distantly from Hedeoma by both Bentham and Briquet. The features which distinguish the genera are subtle. The stamens of Cwnila are seated above the middle of the corolla tube, ascend from the orifice, surpassing the upper lip. The tubular or turbinate calyx has 10-13 nerves, and is rarely more than 5-6 mm long; the teeth are subequal. It should be noted that within Hedeoma, H. floribundum has characteristics of the calyx similar to those found in this genus, perhaps suggesting an even 221 closer relationship than previously supposed. Pogogyne is a small California genus which both Bentham and Briquet reckoned it to be a close relative of Hedeoma. However, careful comparison of the two genera reveals a number of striking differences. The principal features of Pogogyne which distinguish it from Hedeoma are as follows: habit, annual; inflorescence a dense terminal “spike” of cymes; calyx deeply cleft and 15-nerved; stamens 2 or 4; anther sacs parallel; style pubescent; and, chromosome number, 72 = 19 (Ornduff, 1967) The genus Glechon was recognized and treated as a close relative of Hedeoma by Bentham (1832-36 and 1848). Briquet (1897), on the other hand, placed Glechon into a different subtribe. Epling and Stewart (1939) again aligned the genus with Hedeoma. Since the principal distinguishing feature resides in the corolla, it is not surprising that Briquet, who strongly emphasized corolla morphology in his treatment, should fail to indicate close relationship between the two genera. Briefly, Glechon differs from Hedeoma in being composed of South American fruticose herbs with sessile cymules; calyx tube 13-15-nerved, turbinate, with subequal deltoid or deltoid-subulate teeth; corolla ample, abruptly expanded, the upper lip manifestly longer than the lower; stamens genticulate at the base and hairy. The Asiatic genus Mosla was originally placed in Hedeoma, later, as more species became known, it was separated from Hedeoma and given generic rank, Mosla differs from Hedeoma primarily in having a 10 rather than 13-ribbed calyx and in having a racemose inflorescence. Rhododon ciliatus was included in Hedeoma sect. Stachydeoma and later Stachydeoma. In 1939 Epling and Stewart elevated it to generic rank. Rhododon is readily distinguished from Hedeoma. Rhododon is a coarse annual with sessile, long pilose-ciliate leaves and has a dense terminal ‘spike’; a 15-nerved, deeply cleft calyx; no staminodes; a pubescent style; and a chromosome number of 2 = The genera, Hoehna, Eriothymus, and Rhabdocaulon are all believed to be offshoots of the diverse genus Kerthia. Bentham treated Keithia as very close to Hedeoma, but Briquet (1897) included the whole of Kesthia into Hedeoma. Epling not only recognized Keithia as distinct but further divided the group into three genera, a treatment with which, after careful examina- tion of many representatives of these groups, I am in complete accord. Stachydeoma graveolens was included in the genus Hedeoma by both Gray (1878) and Briquet (1897). In 1903, Small created Stachydeoma as a dis- tinct genus based on H. graveolens and H. ciliatum. Unlike Rhododon ciliatus, with which it was previously grouped, Stachydeoma cannot be kept easily as a distinct genus. It is very Hedeoma-like in the details of the calyx and corolla and in the important features of the nutlets. The separating features reside in the pilose, sessile leaves and in its chromosome number, 7 = 9 Poliomintha and Hesperozygis are also allied very closely to Hedeoma. As the differences are more quantitative than qualitative, the genera are con- 222 trasted with Hedeoma in the tabular form below. Hedeoma Poliomintha soccer semishrubs, usually her- Shrubs. eous perennials or annuals. Calyx tube gibbous or saccate. Calyx tubular and symmetrical. Calyx teeth usually acuminate and well Calyx teeth subsequal, convergent to differentiated into upper and lower sets. close the orifice. Calyx annulus in a well-defined ring. Calyx annulus absent or in an irregular ring. Nutlets oblong or orbicular in outline, Nutlets oblong, o 1.5 mm long, not ca. 1.5 mm long or less, usually Reem becoming een when moistened. ing mucilaginous when moistened. 2n — 34, 36, 44, 72, 144. 2n = 36. Hedeoma Hesperozygts ses N. American Primarily S. American Semi-shrubs, of more co ommonly peren- Shrubs with leathery leaves. re or eas herbs, leaves various but not leathery. Calyx 13-nerved. Calyx 13-16 nerved. Calyx tube various but not funnelform. Calyx tube funnelform. Calyx teeth various. Calyx teeth ae ae deltoid lanceolate, erect a par Calyx annulus seated at or near the Calyx annulus usually seated near the juncture of the upper and lower teeth middle of the tube at the tube’s summit. 2n = 34, 36, 44, 72, 144. 2n = 44, The genus Hesperozygis in both its original circumscription (Epling, 1936; Epling and Stewart, 1939) and in subsequent additions (Epling, 1940 ) houses a bewildering array of seemingly unrelated taxa. It is rendered some- what more homogeneous by removing to Hedeoma three of its former species, but the remaining species are problematical still and are likely to remain so until a more comprehensive study is undertaken. All of the above-cited genera are closely interrelated. Though they all have two stamens, they are probably related to the four-stamen groups through the genera Satureja and Melissa. Indeed, Bentham (1832-36) stated that, “Hedeoma scarcely differs from Miécromeria except in the abortion of the upper stamina.” CHROMOSOMAL STUDIES Chromosome numbers for Hedeoma and related genera are reported else- where (Irving, 1976). In summary, 27 = 36 was the most frequently en- 225 countered number occuring in 67% of the taxa reported. Section Alpine was consistently 27 = 44. Aneuploidy (2 = 34) occurred in several taxa distributed along the eastern edge of the generic range: a cluster of popula- tions of H. drummondii in east-central Texas, H. hispidum, H. acinoides and both varieties of H. reverchonii. Polyploidy was found in the southernmost populations of H. costatum (2n = 72) and in H. multiflorum (2n = 72) and H. apiculatum (2n = 144). Poliomintha longiflora was 2n — = 36; Hesperozygis marifolia, 2n = 44; Rhododon ciliatus, 2n = 26 and Stachy- deoma graveolens, 2n =18. Based on an analysis of the distribution of chromosome numbers and the ae of S. graveolens, the genus Hedeoma is judged to be dibasic x = 9, 1 BREEDING HABITS Inbreeding, in a variety of forms, is found in all species of ee The lower fertile pair of anthers, typically unseparated, dehisce and th style, which lies behind the anthers, recurves sharply at its tip into the ex- posed pollen mass. The time between the dehiscence of the anthers and the recurvation of the style, however, provides reproductive variability among the species. A few of the annual species (H. hispidum, H. medium and H. pulegioides ) are primarily cleistogamous and only occasionally are chasmogamous plants encountered. In the small to medium flowered species, which are the majority of Hedeoma species, the flowers open and the dehiscence of the anthers and the recurvation of the style occur almost simultaneously. A few of these species (H. acinoides, H. nanum, and H. costatum) can be — The remainder are chasmogamous inbreeders and their populations are often subtly and consistantly distinct from one another. Hedeoma Dei H. costatum, H. nanum are good examples of this phenomenon Among the larger flowered species (H. hyssopifolium, H. pulcherrimum, and H. ciliolatum) the opening of the flower and the maturation of the anthers and style are separated by periods of time. The anthers open approxt- mately 12-24 hours after anthesis and the style requires 24-48 hours, after anthes’s, to recurve into the exposed pollen. Insects (bees, moths and butter- flies) are commonly observed visiting the large flowered species and pre- sumably outbreeding is the predominant mode of reproduction. Occasionally through a special mechanism some plants of the largest flowered species, such as H. ciliolatum, are obligate outbreeders. Following anthesis in these plants, the anther pair separates and is no longer in the path of the recurving style. In general the mode of reproduction has determined in large part my thinking as to the evolutionary development of the genus, and is well cor- related with other plyletic characteristics. Large flowers and outbreeding are considered primitive traits and are typically associated with the long-lived perennial or semi-shrub habits and very restricted and non-weedy distribu- > 224 tion patterns. Small flowers and inbreeding are believed to be derived char- acteristics and most commonly occur in weedy, widely distributed annuals and short-lived perennials. NATURAL INTERSPECIFIC HYBRIDIZATION From both herbarium and field studies there is evidence of interspecific hybridization involving at least seven species in the genus Hedeoma. The putative hybrids have been documented both morphologically and through the use of secondary plant constituents (Irving, unpublished ms. ) H. DRUMMONDII var. DRUMMONDIT X H. NANUM var. NANUM Both H. drummondii var. drummondii and H. nanum var. nanum occur in pure form in the mountains of southern Nevada and adjacent California, and for each this area represents the extreme western range of their distribu- tion. Usually, these two taxa are quite distinct, but numerous specimen (cited below ), although determined as H. nanum, are actually intermediate between H.d, var. drummondii an m. var, nanum. Acceptance of their hybrid status follows directly from a chromatographic survey of the specimens cited which revealed both chromotographically distinct parents and a summation of phenolic patterns for the intermediates (Irving, unpublished ms.). Representative putative hybrids: ALIFORNIA: San aes Co.: Keystone Canyon, 14 May 1940, Alexander & Kellogg 1483 (DS, GH). NEVADA: Clark Co.: Lee Canyon, 22 Jul 1939, Clokey 8476 (UC). Clark ae Harris Spgs. Rd. 10 Jun = Clokey 8633 (NY). Clark Co.: Kyle Canyon, 14 Jun 1939, Clokey 8478 (DS, GH, IA, MICH, NY, ae POM, US). Clark Co.: Deadman Canyon, 30 May oe Alexander & Kellogg 1614 (DS, Se TEXAS: Val Varde Co.: junction of rts 90 and 277, 4 Jun 1967, Irving fo) CTEXs H. DRUMMONDII var. DRUMMONDII & H. PULCHERRIMUM These two closely related species occur in sympatric populations in the mountains of southern New Mexico around Cloudcroft. A visit to one such population 20 miles northeast of Tularosa revealed the presence of numerous intermediate plants. The latter were assumed to be hybrids since paper chromatography displayed an additive pattern for these intermediates and counts of pollen stainability revealed ca 20% stainable pollen. Both parental species consistently displayed 100% stainable pollen (Irving, unpublished ms. ) aid MEXICO: Otero Co.: 20 mi ne of Tularosa, 5 Aug 1967, Irving 835 a & b (TEX H. DENTATUM X H. HYSSOPIFOLIUM The occurrence of these hybrids were detected solely from morphological analysis of herbarium specimens deposited at LA. The hybrids are similar to H. hyssopifolium but possess reduced gibbous calyces and dentate-entire leaves bo i) i) (the calyces of H. hyssopifolium are usually large and nongibbous and the leaves entire). a eee hybrid specimens: NA: Cochise Co., Huachuca Mts, 5 Sep 1936, C. Epling and W. Stewart s.n. (LA) “0 sheets ) . H. DRUMMONDII &* H. REVERCHONII var. REVERCHONII * H. R. var. SERPYLLIFOLIUM Each of these taxa come together on the arid limestone outcroppings of the Edwards Plateau of Central Texas. Here all three form large sympatric populations on the same chromosome level (27 = 34). Although inbreeding and microhabitat differences work to maintain the integrity of each taxon, hybridization in all possible combinations occurs and much of the inter- and intra-populational variation appears to be the result of past and present gene exchange. Gas chromatography, morphology, and pollen fertility were em- ployed in the detection of hybrids (Irving and Adams, 1973). SPECIFIC CONCEPTS The taxa of Hedeoma are delimited by combined information taken from morphology, cytology, ecology, and geography. I have tried to be consistent in my recognition of species, subspecies, and varieties; that is, I have at- tempted to accord formal names to comparable categories. In several cases, as in H. palmeri and H. acinoides, various intraspecific races have been acknowledged but were not accorded formal taxonomic recognition. In regard to the species concept itself, I shall refer to discussions presented by Raven (1962) and Lewis and Lewis (1955). The species concept presented in these papers most accurately reflects my own beliefs With the categories subspecies and variety, the term “variety” is reserved for divergent subspecific populational units which display genetic and geo- raphic integrity. The term sages: is used as a grouping category for closely allied varieties, much as the subgenus is used to group closely allied sections. This usage is as interpreted by Turner (1965), and Kapadia (1964) ECOLOGY, DISTRIBUTION, AND POSSIBLE ORIGIN As a prelude to a discussion on the ecology, distribution, and possible origin of the species in Hedeoma, 1 would like to cite what I consider to be primitive and derived characters for the genus (Table 1). In an effort to present a phyletic picture, I have adopted the “Wagner Divergence Index Fig. 1) as developed by W. H. Wagner, Jr., and illustrated by Hardin (1957). To derive a divergence index for a given species the characters were ranked 0, 1, or 2, depending on their assumed primitive, intermediate or derived condition, respectively. A numerical total, which might be termed a phyletic index, is obtained by summing the value for each of these char- CALIATUM & = as TOD 5 1) Ss) MS SATUREJOIDES ACIN Sect. Alpine wy, POLIOMINTHOIDES PATR c MOL AN 5 .@) 15 1. Divergence in the genus Hedeoma. Concentric arcs represent levels of ig specialization. Open circ taxa (see text). | es suggest hypothetical ancestors; and solid circles are extant 20 9C¢ Pa acter states for any given taxon. The phyletic indices are then plotted upon concentric circles and arranged so as to form a phyletic “tree” or dendrogram ie these calculations there have been no weighting of characters; for Seaeaple, the characters of the leaf margin carry as much weight as characters relating to corolla size. In many cases “intermediate” states were not found; ee they are not listed in the tabulation. m Table 1, the primitive Hedeoma would be an entire-leaved shrub an ae pubescence and solitary flowers; the calyx tube would be radially symmetrical with erect, united teeth and no annulus; the corolla would be large with a flat upper lip and possessing a definite annulus; nutlets ae be brown, oblong, not becoming mucilaginous when moistened; and there would be 36 large, symmetrical chromosomes. No single taxon has af of the criteria for one state or the other, although several approach the purely primitive or derived condition. In particular, three species emerge as primi- tive for the genus: H. céliolatum, H todsenti and H. palmeri (phyletic indexes 7, 8, and 8, respectively; Fig. t the other end of the spectrum the annual, H. Deda. is the most advanced, possessing 13 of the criteria for the derived condition (phyletic index 24); it is closely followed by acinoides (phyletic index 23) PRIMITIVE AND DERIVED C HARACTERS _ DERIVED (2) PRIMITIVE (0) INTERMEDIATES (1 - — bo . Shrubby or ‘suffrutescent . Leaves entire 3. Hairs simple, plants Awa) relatively glabrous . Flowers solitary . Calyx tube radially symmetrical or nearly so 6. Calyx teeth erect and o CO slightly eeu into upper and lower s . Upper calyx teeth scarcely . Calyx annulus absent . Corolla large bn 12 mm) and wit expanded ee Herbaceous perennials Intermediate Hairs simple; plts pubescent Cymules 3—5 flowered Tube gibbous or sigmoid Intermediate Teeth united for less than length 4 their A broken or undulating ring Less than 12 mm or long tubular Anau Variously toothed Hairs branched Cymules 5 or more flowered Tube saccate Teeth variously reflexed and well differentiated into upper and lower sets Teeth united over 4 their length A definite ring Minute and bar exserted from e iene 228 10. Delayed autogamous/ Intermediate Autogamous and/or allogamous in breeding cleistogamous abit 11. Nutlets not becoming Mucilaginous when mucilaginous when moistened moistened 12. Nutlets oblong in outline Orbicular in outline 13. Chromosome number 2” 2m — 34, 72, or 144 = 36, 44 14. Chromosome large and Small and symmetrical asymmetrical 15. Plants of mesic, montane Intermediate Open-arid habitats habitats The genus Hedeoma is divided into 4 subgenera and it is my impression that each developed early in the evolution of the genus and subsequently each followed its own evolutionary course. Ciliatum and Poliominthoides are believed to be the oldest of the sub-generic lines, with Satwrejoides and Hedeoma occupying the intermediate and youngest ae respectively. No firm evidence is available with which to determine the time of divergence of these phylads, but it is suggested that prototypes for ee. ee de- veloped out of the Madro-Tertiary Geoflora during the Oligocene, or perhaps earlier. As the extant species of Hedeoma are associated with both the Madro-tertiary and Arcto-tertiary floras, the ancestral stocks may have de- veloped in the ecotonal region between these two floras (Axelrod, 1950). he subgenus Ciliatwm is composed of five narrowly endemic species which are spatially and morphologically well-separated. This fact, along with their large number of primitive traits, speaks for a rather lengthy evolutionary history for the group. The species are confined to relict communities of the Sierra Madrean woodland of the dissected Madro-Tertiary flora (Axelrod, 958). Hedeoma ciliolatum and H. todsenti are the most primitive, occuring in mesic pine forests near Galeana, Nuevo Leon, Mexico and the forest communities of the San Andres Mountains of New Mexico respectively. Hi. apiculatum and H. pilosum, are more advanced species occurring in the open, more xeric rock crevices in the mountains of Trans-Pecos Texas. The subgenus — is believed to have had an evolutionary pattern similar to that of Czliatum. Its four taxa, while also very distinct morphologically, are a wider ranging. To the north, H. patrinum, H. montanum, and H, molle are associated with the drier elements of the Sierra- Madrean woodland and to the south H. palmeri, occurs principally in the pine-oak woodlands of the Sierra Madre Oriental. The subgenus Saturejoides can be divided into two geographical and morphological groups: (1) section Alpine in which the members are all 229 montane species of the Sierra Madre Occidental; and (2) section Saturejoides, the species of which occupy a wide variety of habitats. he ecology and distribution of section Alpine suggest a strong affinity with the Arcto-Tertiary Geoflora and the section may well have had its origin within it. The pine forests in which the species of Alpine now occur are be- lieved to be extensions of the Arcto-Tertiary flora which extended southward with the uplifting of the Rocky Mountain chain, The lack of transitions with other species and the disjunctive distribution pattern of its species suggest that the former ranges of species belonging to this section cael have been contracted with the spread of aridity during late Tertiary times. Section Saturejoides is composed of taxa of varying Kae and habitats and can be considered the core of the genus. Xerophytic annuals and peren- nials dominate the section and, for the most part, the North American taxa can be aggregated into four species complexes: 1) H. costatum-plicatum- tenuipes-microphyllum, 2) H. ices aa inltiled op me a6 aUOn a aut: dentatum-hyssopifolium, and 4) H. d dii-reverchonit-4 These complexes typically have wide distribution Paes and display combi- nations of relatively specialized morphological attributes. Each complex in- cludes a less specialized and distributionally restricted montane species, H. costatum, H. diffusum, H. hyssopifolium, and H. pulcherrimum respectively. The xerophytes presumably evolved with the expansion of arid areas and dissection of the Madro-Tertiary Geoflora. The subgenus Hedeoma has been the most difficult with which to contend in terms of origin and phylogeny. It seems to be a recent line, on morpho- logical grounds, but its connection with the other subgenera of Hedeoma is obscure. Further, the species are well separated and do not merge one into the other as might be expected in a youthful phylad. SYSTEMATIC TREATMENT HEDEOMA Pers., Syn. Pl. 131, 1807 Annuals or caespitose, suffrutescent, or herbaceous perennials, forming semi-shrubs, tufts, clumps, or mats. Shoots solitary to numerous, strict, ascending, or decumbent and often producing adventitious roots from the lower nodes; shoots variously branched; arising from a slender or thickened taproot, woody caudex, or trailing rootstock; variously pubescent with hairs simple and spreading or retrorsely curving (tomentose, in H. molle with hairs profusely branched). Leaf pairs crowded or well-spaced. Leaves mem- branous or coriaceous; ascending, spreading, or cernuous; oval, ovate, ellip- tical, oblong, lanceolate, linear or spatulate; entire, crenate, serrate, or dentate (often obscurely so); apex obtuse or acute; base sessile, subsessile, or petio- late; blades tomentose (HL molle), canescent (subgenus Poliominthoides), or variously pubescent with ciliate margins; nervation indistinct to con- spicuous with the nerves forming raised costae on the undersurfaces (H. plicatum, H. pusillum, and H. ciliolatum), the leaf surfaces smooth, glandular- 250 dotted, or glandular-punctate to varying degrees. Flowers solitary or dis- posed in axillary cymes (cymules) 1-31 flowered, well-spaced along the axis or aggregated into a terminal “spike” of cymes; peduncles inconspicuous to several millimeters long; pedicels conspicuous, usually several millimeters long and variously pubescent with spreading or retrorsely curving hairs, bracteoles typically subulate or lanceolate, either shorter or longer than the pedicels they subtend. Calyx 3.0-15.0 mm long, highly zygomorphic to nearly actinomorphic (subgenus Poliominthoides); tubular, tubular-sigmoid, or tubular-funnelform; usually gibbous or saccate below for 14 to over 34 of the tube’s length, variously pubescent and usually with the pubescence diminishing in size and density toward the teeth apices (evenly tomentose in H. palmeri and H. molle); the calyx teeth markedly or scarcely (sub- genus Poliominthoides) differentiated into upper and lower sets; the three upper teeth variously connate forming an upper lip; the connate portion of the lip either short or expanded, and either glabrous or pubescent on its inner face; the lobes semi-ovate, deltoid, narrowly triangular or subulate, glabrous or hirtellous to hirsute-ciliate, laterally spreading or erect and convergent (H. multiflorum, H. drummondi:); the two lower teeth generally not coalesced, narrowly triangular or subulate, and typically recurved and surpassing the length of the upper lip (short, equal to the upper lip in the subgenus Hedeoma), ciliate, or, occasionally glabrous on the margins; calyx with a well-defined ring of simple hairs (the annulus) within which tend to be projected outwardly and partially seal the calyx orifice, occasionally upwardly extended onto the inner face of the upper lip (H. piperitum, H. medium, and H. multiflorum), 0.5-1.5 mm. wide and seated at the juncture of the upper and lower teeth or below it, included or exserted from the calyx tube. Corolla white, blue, lavender, pink, or orange-red, extremely variable, 3.0-45.0 mm long, always zygomorphic with the limbs forming upper and lower lips, glabrous or pubescent within in two decurrent lines along an abaxial groove or channel running along the throat floor or simply pubescent in the throat, also commonly pubescent within the tube, occasionally the hairs forming a well-defined annulus; the tube gradually or abruptly expanded upwardly; the upper lip obcordate to ligulate, concave and subgaleate or flat, straight or reflexed away from the axis, scarcely to deeply emarginate; the lower lip spreading to erect (subgenus Hedeoma), equally three lobed or more commonly with the central lobe exserted beyond the laterals and bilobed. The abaxial stamen pair arising from the middle of the corolla (upper %4 in. H. martirense) and ascending upwardly under the upper corolla lip or surpassing it, the anther sacs nearly parallel to widely divaricate on well-developed connectives; the adaxial pair reduced to stami- nodia, 1.0-2.0 mm long, arising opposite the fertile pair on the adaxial surface, occasionally the staminodia bear abortive anther sacs at their apices (H. pulegiodes). Style long and bifid at the apex with the abaxial segment markedly longer than the adaxial one, glabrous. Nutlets orbicular, ovate, or 251 oblong in outline, ovate or ovate-triangular in cross section, brown or black at maturity, the surface areolate, foveate, or ruminate, weakly to strongly glaucous, usually mucilaginous when moistened. Chromosome numbers, 27 = 34, 36, 44, 72, or 144 (x =9 Type species: Hedeoma pulegioides (L.) Pers. Distribution: North and South America. Centers of distribution appear to be in the mountains of far northwestern Mexico, and the mountains of southern Arizona and New Mexico; also in the Sierra Madre Oriental of east central Mexico. 1. — — KEY TO THE SUBGENERA OF Hedeoma pane caespitose, often forming dense mats, 4.0-12.0 cm high; leaves coriaceous, tire and sessile; flowers solitary or in three-flowered axillary cymules; corolla large - oe 10.0—-32.0 mm long; western Texas, ee Mexico and northern J. Subgenus CILIATUM , Plant. ieee eta annuals or perennials or semi-shrubs; leaves various; flowers in three or usuall ore flowered axillary cymules; corolla 3.0-45.0 mm long, plants variously Babar: 2. Suffrutescent, canescent perennials, usually semi-shrubs, 15.0-100.0 cm. tall; leaves canescent; axillary cymules commonly but not always aggregated into terminal “spikes”; calyces tubular, nearly radically symmetrical with the teeth only slightly differentiated into upper and lower Oe I. Subgenus POLIMINTHOIDES 2. Plants usually herbaceous annuals or perennials; ce variously pubescent but never canescent; calyces bilaterally symmetrical with the teeth markedly differ- entiated into upper and lower sets. 3. Leaves and calyces various; lower calyx teeth usually longer than upper; nutlets brown to dark brown at maturity, commonly glaucous, oblong or occasionally obicular in outline, mucilaginous when moistened, plants vari- Ill ously distributed. . Subgenus SATUREJOIDES 3. Leaves ovate or elliptic, toothed; calyx saccate, lower teeth short; nutlets black at maturity, never glaucous, orbicular in outline or nearly so, at all mucilaginous when moistened. Plants of South America and eastern North America. V. Subgenus HEDEOMA KEY TO THE TAXA OF THE SUBGENUS CILIATUM Leaves with regularly spaced costae on the undersurface and finely ciliate margins; calyces densely pubescent on the inner faces of the teeth; »lants of Mexico 2. Plants 5.0-10.0 cm tall; leaves elliptical-lanceolate, 5.0-10.0 mm long: corolla bright orange-red, large and showy, over 25.0 mm long. 1. H. ciliolatum 2. ae 3.0-5.0 cm tall; leaves ovate, 3.5-5.0 mm long, corolla lavender, less n 15.0 mm lon H. pusillum 2. . Leaves smooth or tuberculate on the undersurface but without regularly spaced costae, margins pilose or tuberculate-ciliate; calyx teeth glabrous on their inner faces; plants of U.S. 3. Plants ca 4.0 cm high; leaves ovate, ca 5.0 mm long, calyces pilose. 3. H. pilosum 3. Plants 10.0-15.0 cm high, leaves, lanceolate-elliptic, ca 10.0 mm long, calyces hispidulous 252 . Leaves obtuse or acute at apex; corolla ca 30.0 mm long, — re as 4. Leaves apiculate at apex; corolla ca 20.0 mm long, pink. 5. H. apiculatum KEY TO THE TAXA OF THE SUBGENUS POLIOMINTHOIDES 1. Plants with sim seca hairs; Mexico il a" ple c 2. Tall shrubs, with axillary 15-31 flowered cymules, commonly aggregated into dense terminal spikes. ay meri 2. Spindly shrubs or suffruticose perennials with 1-5 flowered enue not usually ageregatec 3, Spindle shrubs with the upper calyx teeth narrowly triangular and reflexed away from the axis; calyx annulus more or less included. ...... 7. H. patrinum 3. Suffruticose perennials with the upper calyx teeth subulate and, at least par- tially, convergent with the lower at maturity; calyx annulus well-exserted. montanum . Plants covered by a tomentum of highly branched hairs; Texas. .......... 9. H. molle KEY TO THE TAXA OF THE SUBGENUS SATUREJOIDES . Plants perennial herbs with the shoots decumbent and rhizomatous in some species; leaves ovate, elliptical, oblong, or lanceolate, margins entire or obscurely toothed, base tapered and subsessile; upper calyx lip broad; corolla showy, the upper lip wide; plants of the Sierra Madre Occidental of western Mex XICO. ....., A. sect. Alsine 2. Plants with numerous ascending and decumbent se ee adventitious roots from the lower nodes; nutlets spherical or ne so. 3. Leaves oblong or oblanceolate; are eet toward the branch apices; calyx large, tubular-funnelform, ca 8.0 mm long, 2.0-2.5 mm wide at the summit, the tube not at all gi 4. Leaves crowded with the ea aks uniformly 1.0 mm long, leaves 6.5— 7.0 mn pi cote 0. H. woliedlnetolum 4, Leaves well-spaced, 10.0-17.0 mm long, the bases not decurrent; calyx hirtellous; upper calyx lobes deltoid or semi-ovate; plants . - Am. 3. Leaves elliptical, ovate or subrotund; cymules well-spaced ae the axis; calyx zygomorphic, 5.5-7.0 mm long, 1.5 mm wide at the summit, the tube pibbo ous or saccate. 5. Leaves ovate to narrowly elliptical; the calyx ia Behcon gibbous, the upper calyx lobes triangular; corolla 11.0-12.0 mm long. ......00000.00........ . . jucundum 5. Leaves ovate to sub-orbicular; the calyx tube ee bia accate, forming a =u pouch, the upper calyx lobes semi-ovate or deltoid; oa la 7.0- 8 13. £ ee 2. Plants with ane ascending; nutlets ovoid; leaves with odor of oregan 6. Cymules well-spaced along the upper 3 of the stems; calyx 4.0-6.5 mm 5 eee: corolla 7.0-9.0 mm long. 14. H. patens 6. Cymules aggregated toward the branch apices forming a sn terminal “spike” of cymules; calyx 3.0-3.5 mm long; corolla 5. ae . mm long . aH. Fae ei 259 1. Annuals or aoa the shoots and leaves various; upper calyx lip not usually broad, or if so, t plants annual; corolla various, the up Ee “Pp os or if redered markedly a plants auaual: nutlets oblong, ca tw s long as wide; plants variously distributed, but not in the Sierra Madre Orci dental of. western exico. t. Saturejoides 7. Plants perennial, or annual with the calyx teeth oe convergent and closing the calyx orifice, or with the corolla tbe markedly 8. Upper and lower calyx teeth convergent, closing a o orifice at maturity, or if only slightly so, then plants with entire, long, elliptical-oblong leaves and large corolla, 10.0-14.0 mm long, the lower lip 7.0 mm wide 9. Calyx tube gibbous but not saccate, not tapered upwardly; upper calyx the mountains of south central New Mexico. ............ . . pulcherrimum ee tube conspicuously saccate and tapered into a narrow neck; the upper calyx teeth, clearly convergent with the lower; ae of various crap and distribution. 0. Upper calyx teeth ununited, subulate; nutlets areolate: plants of N. Am xe 11. Plaats annuals or herbaceous perennials, occasionally somewhat t the a0) hirsute, the upper and lower teeth usually Sones convergent; corolla 7.0-11.0 mm long; widespread; odor of peppermint. i drum mondii 11. Plants suffruticose typically averaging over 25.0 cm in height; leaves r dark green, usually less than three times longer than ; calyx coarsely hirsute or hoary villous, the upper and lower teeth incompletely closing the orifice; corolla 8.0-15.0 mm long; west and central Texas to south-central Oklahoma; odor of ] camphor or lemon. 18. H. reverchonti 10. Upper calyx teeth united, triangular; nutlets foveate; as of S. Am 19. H ; ee 8. Calyx teeth not as above, the Bees mae generally spreading and reflexed a maturity leaving the calyx tube Leaves ovate and entire, a ee cere short and gibbous-saccate, the distended region 4 of the length of the tube. 13. Leaves ovate, acute; corolla often dilated upwardly; nutlets areolate; plants of N. Am. 14. Plants with numerous decumbent stems, forming circular mat calyx sania but not saccate, the upper lip pubescent on its inner face; corolla ca 9.5 mm long, restricted to the ee forests of north-central eee (Blapstath) soe Dare, 20. diffusum 14. Plants of various habits but not as above; calyx saccate; oe upper lip ee rous within; corolla 8.0-9.0 mm long, southwestern U.S. and Mexico. 15. ee entire and smooth; widespread. .............. 21. H. nanum 15. Leaves denticulate, costate; endemic to central Mexico. ............ 2 : ie 13. mas linear-elliptical; corolla not dilated upwardly; nutlets foveate; plants of S. Am 23. H. medium 12. Leaves pen eatned (entire and ovate in the two s species from Baja, Caltiorna) calyx tubular or sigmoid, the distended region less than 1 the length of the tube. 234 16. Tufted perennials with the shoots freely branching; axillary leaf buds not usually occurring; leaves oval, ovate, linear or elliptic, plants of central and Mexico. 17. Leaves serrate or dentate; ovat, elliptic, rhombic or oblanceolate aves ovate, oval or elliptic; leaf nerves distinct but not dlievared: curved and branched; calyx sanulus seated at the juncture of Upp er and lower lips. 24. H. costatum 18. Leaves ovate, oblanceolate, elliptic or rhombic; leaf nerves elevated and straight; calyx annulus seated below the jmenire of upper and lower li Ips. 19. Leaves ovate-oblanceolate, dentate; calyx well over 10 mm ; corolla showy, over 30 mm long; known only from the Sierra jardin H, Risiee oO n Mexic 19. Leaves elliptic or hone crac serrate; calyx - 8.0 mm long; corolla 7.0-10.0 mm long; widespread. .................... oo ee 17. Leaves entire or obs ae crenate; ovate, dlise or ie 20. Pouce aes or ovate; oe of Baja California. . Leaves 9.0-13.5 long; internodes elongated; cymules 1 flowere 27. H. tenuiflorum 21. Leaves 5.0-6.0 mm long; internodes short; flowers solitary in the upper leaf axils. 28. H. martirense 20. Leaves lanceolate (at least the upper); plants of eastern Mexico. 22. Axillary cymules 1-3 flowered; calyx 7.0-8.0 mm long; corolla ca mm long, well exserted from the calyx. ........ 29. H. quercetorum 22. Axillary cymules many-flowered; calyx 5.0-6.0 mm long; corolla ca 5.5 mm long, barely exserted from the calyx. ...... 30. H. tenutpes 16. Tall robust plants with long ascending, usually anes hed shoots; axillary eav ieldi a plants of western New Mexico and eastern Arizona, occasionally in northwestern Mexico. 23. Leaves linear or narrowly elliptical, entire, and subsessile; cen i yy flowered; bal slightly gibbous; tien large and showy, long, ca 7.0 wide across the lower lip. ................ 31. H. hysopiolm 23. Leaves elliptical, dentate or cones shortly petiolate or bsessile; ymules 1-15 flowered; calyx m markedly gibbous; corolla 8.0-11 0: mm long, 24. Basal leaves ovate, glabrous, entire. .....0..0.0000000000.. 32. H. ee 24. Basal leaves ovate-rhomboid, yee pubescent, dentate. 200.000.0000... -anbabain 7. Plants annual, the calyx teeth not convergent or the corolla markedly Pie 25. Leaves ovate, sUipecal or rhombic-elliptical, dentate, or entire, petiolate; calyx saccate for ca $+ of the tube’s length; corolla only occasionally cleistogam mous. ... 34. H. acinoides 25. eaves linear or linear-elliptic, at least five times longer than wide, entire sessile; calyx saccate for over 3 of the tube’s length; the upper lip ee the corolla straight, corolla often cleistogamous. 35. H. hispidum KEY TO THE TAXA OF THE SUBGENUS HEDEOMA 1. Plants annual; leaves elliptical and dentate; plants of N. Am. .... 36. H. pulegioides 1. Plants perennial; leaves ovate, obovate, spatulate, crenate or entire; plants of S. Am. 2. Leaves sary crenate; calyx markedly saccate, ca 4.0 mm long; corolla ca 3.0 m long, Braz 37. H. crenatum 23 2. Leaves obscurely crenate, mostly entire; calyx gibbous or ed slightly saccate, 4.0-5.0 mm long, corolla 5.0-8.5 mm long, Bolivia and P 8. H. er es I. SUBGENUS CILIATUM, Irving, subg. nov. Herba casespitosa vel tegeta, 4-12 cm alta; caules herbacei numerosi simplices; folia conferta, coriacea, ovato vel elliptico—lanceolata, integra, sessilia, 3.5-15.0 mm longis, 1.5—3.5 mm lata, margine conspicue ciliata; cymae triflorae vel flores singulares; calyx infundibuliformis vel oe: tubularis; corolla grandis speciosa; mericarpium oblongum madefactum mucilaginum; chromosomatum numerus 2” = 36, 144. Plants casespitose, often forming dense mats, 4.0-12.0 cm in_ height. Stems herbaceous, numerous, unbranched, arising from a woody caudex or trailing rootstock. Leaf pairs crowded, the internodes shorter than 2.0 mm; leaves coriaceous ascending, usually folded with drying, ovate to elliptical- lanceolate, entire, sessile, 3.5-15.0 mm long, 1.5-3.5 mm wide; margins ciliate, the upper and lower leaf surfaces glabrate to puberulent, nervation prominent or inconspicuous; undersurface to varying degrees glandular- punctate between the costae, the upper surface smooth. Flowers typically solitary in the upper leaf axils or in three-flowered cymules (H. apiculatum). Calyx tubular-funnelform, not at all gibbous below, distinctly bilabiate in both the tube and disposition of the teeth, 5.0-10.0 mm long; the upper teeth united for 4-2 their length, ee an upper lip, the lobes tri- angular or deltoid; the lower teeth narrow, triangular; the annulus dense and included. Corolla lavender, pink or red, usually large and showy, 10.0—32.0 mm long, pubescent within. Nutlets oblong in outline, at least twice as long as wide, ovate-triangular in cross section, brown at maturity, the sur- face areolate, glaucous, becoming mucilaginous when moistened. Chromo- some number, 27 = Type species: HEDEOMA CILIOLATUM Epl. 1. HEDEOMA CILIOLATUM (Epl.) Irving, Brittonia 22: 345. 1970. Hesperozygis ciliolata Epl., Rep. Spec. Nov., Beihefte 115: 13-14. 1939. Somewhat robust perennials forming dense clumps 7.0-12.0 cm (10.0 cm) high. Stems numerous, unbranched or only sparsely so, canescent with short retrorse hairs. Leaves coriaceous, folded with drying, elliptical-lanceolate, entire, sessile, 1.7-2.5 mm (2.0 mm) wide, 5.0-10.0 mm (84 mm) long; tapered to a narrowly obtuse or acutish apex, margins distinctly ciliate; upper and lower surface glabrous; nerves of 4-5 prominent pairs, leaves smooth above, punctate below. Flowers in the axis of the upper leaves, soli- tary or rarely in two-flowered cymules, congested somewhat towards the apex; peduncles inconspicuous, ca 1.0 mm in length; pedicels 2.0-4.0 mm long, canescent; bracteoles short, oblong. Calyx coriaceous, 8.0-10.5 mm (9.9 mm) long, the tube tubular-funnelform, 6.5-8.0 mm (7.3 mm) long, 2.0-2.2 mm wide above, densely puberulent; the upper teeth scarcely connate, 236 forming an upper lip, 1.5-2.1 mm long, the lobes triangular ca 2.6 mm wide at the base, 1.0-2.0 mm long, ciliate; the lower teeth lanceolate, ascending, 1.5-3.0 mm long, ciliate, the annulus dense, included, 0.7-1.0 m wide but extended upwardly into the calyx lobes, seated 1.5-2.0 mm below the junc- ture of the upper and lower teeth. Corolla brilliant, orange-red, 30.0-32.0 mm (31.0 mm) long, pubescent within on the palate, densely puberulent within the tube, forming annulus ca 4.5 mm wide; the tube, 25.0-27.0 mm long dilated upwardly, 5.0 mm wide above; the upper lip broadly ligulate 5.0 mm wide, ca 6.0 mm long, distinctly emarginate; the lower lip ca 12.0 mm wide, ca 7.5 mm long. Nutlets ca 0.6 mm wide, 1.2 mm long, the sur- face weakly glaucous. Chromosome number, 27 = 36. Distribution (Fig. 2). Endemic but abundant on gypsum outcrops in Open, mature, pine forests in the vicinity of Galeana, Nuevo Leon. 4,000-— 6,000 ft. Jul-Sep. TYPE: Nuevo LEON. 15 mi southwest of Galeana, 18 Jul 1934, C. H. & M. T. Mueller 1062 (Holotype, LA!; isotypes, F! GH! MICH! TEX!). There are several features of H. ciliolatum which suggest an alliance to Poliomintha. Its large, brightly colored corolla is scarcely distinguishable from the corolla of Poliomintha longiflora, and its inner calyx pubescence is suggestive of the P. longiflora—P. glabrescens complex. However, its habit and clear alliance to H. pusillum, strongly support its inclusion in the sub- genus Ciliatum. In no way can I see any alliance to the South American representatives of the genus Hesperozygis nor to its sole North American representative, H. marifolia, as was suggested earlier by Epling (1939). 2. HEDEOMA PUSILLUM (Irving) Irving, Brittonia 22: 345. 1970. Hesperozygis pusilla Irving, Brittonia 19: 245-247. 1967. Small, perennial, caespitose herbs, 3.0-5.0 cm high, forming mats. Stems numerous, unbranched, canescent with short retrorse hairs. Leaves coriaceous, folded with drying, ovate, entire, sessile, 1.5-2.1 mm (1.8 mm) wide, 3.5-5.0 mm (4.7 mm) long, tapered to a narrowly obtuse or acutish apex, margins finely ciliate, both surfaces evenly puberulent with short appressed hairs; nerves of 4—5 prominent pairs; leaves smooth above, punctuate below. Flowers solitary in the upper leaf axils, congested toward the apex; peduncles short, less than 1.0 mm in length; pedicels 1.0-2.0 mm long, canescent; bracteoles short, ovate-oblong. Calyx 5.2-7.0 mm (6.2 mm) long, coriaceous; the tube tubular-funnelform, 4.5-5.5 mm (5.0 mm) long, densely puberulent; the upper teeth connate for 1/4-1/3 of their length, forming an upper lip, the lobes 1.0-1.5 mm wide at the base, ca 0.7 mm long, the central lobe commonly reduced to about 1/2 the width and height of the two adjacent lobes, the lobes slightly laterally spreading and reflexed; the lower teeth lanceolate, short, ascending, ca 1.0 mm long, densely puberulent on the inner face; the annulus included, dense, ca 0.5 mm wide, 257 extending into the calyx lobes, seated ca 0.5 mm below the juncture of the upper and lower teeth. Corolla lavender, 10.0-11.0 mm (10.5 mm) long, pubescent within but not annulate; the tube, 8.5-9.5 mm long, only slightly dilated upwardly, ca 2.5 mm wide above; upper lip ligulate, subgaleate, straight, 1.55 mm wide, ca 1.5 mm long, distinctly emarginate, the lower lip ca 3.5 mm wide, ca 2.0 mm long. Nutlets ca 0.5 mm wide, 1.0 mm long, the surface, weakly or not at all glaucous. Chromosome number 27 = 36. Distribution (Fig. 2). Known only from the pine forests south of Galeana, Mexico; often sympathic with H. crliolatum. TYPE: Nuevo LEON. 2.5 mi south of Galeana near junction of road to Linares, 5-7 Aug 1965, Irving 471 (Holotype, TEX! isotypes, LA! SMU! TEX! MONTU! ). In general, H. pusillum is a diminutive counterpart of H. ciliolatum. The quantitative differences being reflected to some degree in all of the mor- phological features. The most conspicuous departure of H. pusillum is in its small (10.0-11.0 mm.) lavender corolla which is in marked contrast to the long (30.0-32.0 mm), orange-red corolla of H. ciliolatum. 3. HEDEOMA PILOSUM Irving, Brittonia 22: 340-342. 1970. A small perennial, caespitose herb ca 4.0 cm high, probably forming mats. Stems extremely short, unbranched above, profusely branched below, pube- scent with short spreading hairs. Leaves coriaceous and sharply ascending, appressed to the stems, ovate, entire, sessile, 5.0 mm long, 2.5-3.0 mm (2 mm) wide; tapered to a narrowly obtuse or acutish apex, margins short pilose-ciliate, both upper and lower surfaces glabrous; nerves of 1-2 slightly raised subopposite pairs. Flowers solitary in the uppermost leaf axils; ped- uncles short; pedicels 1.0—2.0 mm long, canescent; bracteoles short, oblone: ciliate. Calyx 7.5 mm long, somewhat coriaceous, the tube tubular-funnelform, dilated upwardly, ca 5.0 mm long, pilose, the hairs long and spreading; the upper teeth connate for ca 1/2 of their length, forming a conspicuous, broad upper lip, ca 2.4 mm long, the lobes triangular, ca 0.5 mm wide at the base, ca 1.2 mm long, laterally slightly spreading and reflexed, pilose ciliate; the lower teeth narrowly triangular-lanceolate, ascending, ca 2.5 mm long, pilose-ciliate; annulus dense, included, ca 1.1 mm wide but seated deep within the tube at ca 2.0 mm below the juncture of the upper and lower lips. Corolla, nutlets unknown. Chromosome number, unknown. Distribution (Fig. 2). Known only from the type locality. Open, exposed limestone, 5,000-6,000 ft. Jul. TYPE: TEXAS. Brewster Co., near top of Baldy Peak, Glass Mountains, 13 Jul 1940, Warnock 352 (Holotype GH! ) Hedeoma pilosum seems to be intermediate between H. pasillum and H. apiculatum but due to the incompleteness of the specimen this position is tentative. Unfortunately a revisit to the type locality did not reveal any trace of H. pilosum and it may well be extinct. 238 Fig. 2. Distribution of Hedeoma ciliolatum (closed circles), H. pusillum (open circles), H. pilosam (open square) and H. apiculatum (closed square), H. todsenti (star 239 4. HEDEOMA TODSENII Irving, Madrofio 26: 184-187. 1979. Suffrutescent caespitose perennials 10.0 to 18.0 cm high. Stems numerous, simple, canescent or puberulent with hairs retrorsely curling. Leaves corta- ceous, entire, lanceolate-elliptic, subsessile, 8.0-14.0 mm long, 2.0-4.0 mm wide; apex obtuse to slightly mucronate; midrib and margins slightly his- pidulose-tuberculate; lower surface deeply glandular punctuate, the upper less so. Flowers solitary and well-spaced in the upper portions of the stems; peduncles ca 1.5 mm long; pedicels ca 3.5 mm long, canescent; bracteoles lanceolate longer than pedicels. Calyx 11.0-12.0 mm long (base to tip of teeth), tubular-funnelform, not gibbous, minutely scabrous on the nerves and glandular punctate between; upper teeth partially connate forming a conspicuously reflexed upper lip, the lobes lanceolate, ca 1.0 mm wide, 1.5 mm long, sparsely ciliate; lower teeth ascending, deltoid-aristate, 3.0—3.5 mm long, sparsely hirsute-ciliate; annulus ca 1.0 mm wide, included and seated ca 2.0 mm below the juncture of the upper and lower teeth. Corolla orange- red, ca 30.0 mm long, 6.0 mm wide at the throat, sparsely pubescent in the tube but not annulate; the upper lip deeply obcordate, erect, up to 5.0 mm long, red in color on the interior margins; lower lip streaked with red, reflexed, the similar rounded lobes ca 3.5 mm long, 3.0 mm wide; anthers exserted. Nutlets ca 2.0 mm long, 1.0 mm wide, dark brown in color and weakly glaucous. Chromosome number, unknown. Distribution (Fig. 2). Known only from the type locality. Open, exposed limestone slopes. Aug—Sept. TYPE: New MEXxIco, Sierra Co., San Andres Mts., On steep north-facing gravelly gypseous limestone hillside in open pinyon, 2 km s of Hardin Ranch, 18 Aug 1978, T.K. Todsen SA 1-78 (aiouee NY; isotypes, NMSU US! TEX!). Hedeoma todsenii is most closely allied to H. apiculatum. The suffruticose habit; the overall patterns of pubescence; the coriaceous, entire, lanceolate leaves and the large tubular calyx with its deep seated annulus are some of the morphological attributes which are nearly identical in both taxa 5. HEDEOMA APICULATUM Stewart, Rep. Spec. Nov., Beihefte 115: 25. 1939. Small, suffrutescent perennials, 10.0-15.0 cm (12.0 cm) high, forming mats. Stems numerous, unbranched, densely clothed in minute retrorsely curling hairs. Leaves coriaceous, somewhat folded upon drying, lanceolate- elliptic (occasionally ovate below), entire, sessile, 1.5-3.5 mm (2.5 mm) wide, 7.0-15.0 mm (10.0 mm) long, tapering to a conspicious but short apiculate apex, sparsely hispiduluse-tuberculate on veins and margins; nerva- tion indistinct; upper surface smooth, lower rugose and punctuate. Flowers sol- itary or in 2—3 flowered axillary cymules, crowded toward the apex; peduncles short, 1.0-1.5 mm long, pedice's 3.0-4.0 mm long, canescent, bracteoles narrowly lanceolate. Calyx 8.5-9.0 mm (8.8 mm) long, chartaceous; the 240 tube tubular-funnelform, flaring above, 5.5-6.5 mm (6.0 mm) long, 1.5—-1.8 mm wide below to 2.0-3.0 mm wide above, minutely hispidulose-tuberculate below, glabrous above; the upper teeth connate for ca 1/4 or less of their length, forming an upper lip 2.0-2.2 mm long, the lobes broadly triangular or deltoid, apiculate at their apices, ca 1.0 mm wide at the base, 1.0-1.5 mm long, laterally spreading and reflexed, hirtellous and tuberculate-ciliate; the lower teeth ascending, broadly to narrowly triangular-aristate, ca 1.2 mm wide at the base, ca 3.0 mm long, conspicuously hirsute-ciliate; annulus dense, ca 1.2 mm wide, with long pilose hairs, seated ca 1.5 mm below the juncture of the upper and lower teeth. Corolla pink, showy, 19.0-20.0 mm long, pubescent in the throat, sparsely pubescent within the tube; the tube ca 10.0 mm long conspicuously dilated upwardly, ca 4.0 mm wide above; the upper uP broadly obcordate in outline, ca 5.0 mm wide, ca 5.0 mm long, the lower lip broad, 9.0-10.0 mm wide, ca 8.5 mm long. Nutlets somewhat pointed at their apex, ca 2.0 mm long, 0.7-0.9 mm wide, the surface weakly glaucous. Chromosome number, 27 = 1 Distribution (Fig. 2). Restricted to the Guadalupe Mountains of Texas and New Mexico, frequent among rocks of the streamway and canyon sides in the south fork of McKittrick Canyon and in the limestone crevices atop adjacent Pine Top Mountain. 5,000-8,000 ft. Jul-Aug. TYPE: TrExAs. Culberson Co., Guadalupe Mts., sunny crevices along stream, McKittrick Canyon, 22 Aug 1931, Moore & Steyer mark 3563 (Holo- type, LA!; isotypes, CAS! DS! GH! MICH! NY! PH!). Il. SUBGENUS POLIOMINTHOIDES Irving, subg. nov. Suffruticosum 15.0-100.0 cm altae incani ad basin valde lignosi; folia crassa membranacea, ae ovata vel elliptica, integra vel dentata, petiolata, 5.0-30.0 mm longa, 2.0-15.0 mm lata; cymulae axillares 3-31 flori, terminales densi interruptique, ageregatae; calyx tubularis, calycis dentes, subulati; series superae et interae minime distincti; corolla magna speciosa; mericarpum oblongum madefactum mucilaginum, chromosomatum numerous, 2” = 36. Suffrutescent perennials, or semi-shrubs, 15.0-100.0 cm tall. Shoots woody below, herbaceous above, ascending or shortly decumbent, arising from a woody caudex or short woody taproot, canescent. Leaves canescent and glandular punctuate, thick-membranous, rotund, ovate, or elliptical, 2.0-15.0 mm wide, 5.0-30.0 mm long, margins entire or dentate; apex acute or obtuse; base narrowly attenuated and petiolate. Axillary cymules, 3—31- flowered, well-space along the upper 1/3-1/2 of the stems, or commonly aggregated into dense, interrupted terminal spikes. Calyx more or less radi- ally symmetrical and tubular, 5.0-8.0 mm long, pubescent evenly dis- tributed; calyx teeth only slightly differentiated into upper and lower sets (to be greater extent in H. patrinum), the upper teeth commonly subulate and scarcely united to form the upper lip; the lower teeth subulate; annulus dense and well-formed. Corolla white to lavender, large and showy, over 241 13.0 mm long, variously pubescent within, the tube dilating upwardly; upper lip ligulate to obcordate, emarginate; lower lip spreading. Nutlets oblong in outline, over twice as long as wide, brown at maturity with the surface areolate, glaucous and becoming mucilanginous when moistened. Chromo- some number 27 = 36. Type species: HEDEOMA PALMERI Hemsl. 6. HEDEOMA PALMERI Hemsl., Biol. Cent. Am. 2: 548. 1881. ele rotundifolium Hemsl., Biol. Cent. Am. 2: 549. 1881. Type: HIDALGO: “Zimapan,” Coulter 1081 (Holotype. K!; photolypes: GH! LA! 7 PH! -US!). hie suffrutescent perennials 35.0-100.0 cm (58.8 cm) high. Shoots long, ascending or erect, arising from a woody caudex, branching sparse but occurring freely along the axis with the branches diverging at wide angles, densely pubescent or puberulent (in many representatives nearly cinerous) the hairs retrorsely curving or curling. Leaves variable, ovate or subrotund, 6.0- 15.0 mm (11.5 mm) wide, 10.0-30.0 (20.0 mm) long; margins crenate or dentate, rarely entire; apex obtuse; base broadly or narrowly attenuated, petiolate with the petiole 2.5-6.0 mm long; pubescence highly variable, the under-surface varying from tomentose to simply pubescent, the margins and upper surface glabrate to pubescent; nerves not prominent. Axillary cymules 15-31-flowered, occurring along the upper 1/3-1/2 of the stems and com- monly terminating in a dense interrupted spike; primary peduncles ca 1.5 mm; pedicels ca 1.0 mm long, inconspicuous; bracteoles narrowly lanceolate usually longer than the pedicels they subtend and frequently as long or longer than the calyx tube (short in University of Illinois collection 994; Mueller 2707, 29956). Calyx 5.0-8.0 mm (6.3 mm) long, the tube radically symmetrical, 3.0-4.0 mm (3.5 mm) long, ca 1.1 mm wide in diameter, somewhat dilated upwardly, puberulent or densely long pilose (Irving 669, Shreve & Tinkham 97794, Cummins 23-63), calyx teeth only slightly differ- entiated into upper and lower lips; the upper teeth scarcely connate, the lobes long subulate (occasionally triangular ), commonly exceeding the lower in length, 1.9-3.7 mm (2.6 mm) long, erect and outwardly spreading; the lower teeth subulate, 2.0-3.3 mm (2.5 mm) long, erect and outwardly spreading; the annulus well-exserted, undulating, ca 0.7 mm wide and seated at the juncture of the upper and lower teeth. Corolla blue-lavender, 14.0— 16.0 mm (15.0 mm) long, pubescent within but not annulate; the tube and throat together 9.0-10.0 mm (9.5 mm) long, conspicuously dilated upwardly, about 3.0 mm wide above; the upper lip broadly to narrowly ligulate, emarginate, 2.0-3.5 mm wide, 5.0-6.0 mm long, the margins and apex revolute; lower lip 5.0 mm wide, 5.0-7.0 mm long. Nutlets ca 0.5 mm wide, 1.0 mm long, the surface minutely areolate and weakly gluacous. Chromosome number, 27 = 36. Distribution (Fig. 3). Abundant in a variety of habitats from Galeana, Nuevo Leon, south to Zimapan, Hidalgo. It is especially prominent near the — 242 city of San Luis Potosi. The habitats range from upper montane pine forests where the plants occur in open, rocky, exposed areas to the relatively xerophy- tic habitats at lower elevations, where the plants abound on open, rocky ex- posed areas (e.g., stabilized talus slopes), 5,000-10,000 ft. Jul-Sep a = = 5) 3 ty a im il r0eeee CT aul Fig. 3. Distribution of Hedeoma palmeri (closed circles), H. molle (open circles), H. patrinum (closed squares) and H. montanum (open squares). 243 TYPE: SAN LuIs POTOSI: chiefly in the region of San Luis Potosi, 22° N. Lat., Alt. 6,000-8,000 ft., 1878, Parry & Palmer 728 (Holotype, K! iso- types, F! LA! MICH! NY! PH! US!; phototype, LA! ). Hedeoma palmeri is quite variable and its populations form a mosaic of morphological patterns. Because of this, no attempt has been made to sub- divide the species into subspecific categories. The densely tomentose populations growing in the Galeana area (the most northern part of H. palmeri’s range) are especially difficult to treat. In the words of Epling (unpublished notes), “these specimens [from Galeana] are such an extreme as to provoke a review of this species.” I.M. Johnston, according to information on herbarium labels (Shreve & Tinkham 9779), thought the plants represented an undescribed species. However, these workers examined only a few extreme specimens which gave a somewhat distorted picture. When many representatives are brought together, the extremes grade one into another. Hedeoma palmeri is quite variable and its populations form a mosaic of morphological patterns. Because of this, no attempt has been made to sub- divide the species into subspecific categories. 7. HEDEOMA PATRINUM Stewart, Rep. Spec. Nov., Beihefte 115: 45. 1939. Spindly, suffruticose perennials or semi-shrubs, 15.0-62.0 cm (31.4 cm) high. Shoots ascending, or commonly shortly decumbent, rhizomatous pro- ducing adventitious roots at the lower nodes, arising from a slender taproot, branching freely along the axes with the branches divergent at wide angles, cinerous above with short retrorsely curling hairs (occasionally spreading), glabrous below with exfoliating bark. Leaves deciduous from the lower por- tion of the stems, spreading, ovate or elliptic, 3.0-5.5 mm (4.2 mm) wide, 5.2-10.0 mm (7.6 mm) long; margins entire or occasionally obscurely den- tate in the upper 1/2 of the blade (Purpus 4588); apex broadly acute or obtuse; based broadly cuneate or rounded, petiole 1.0-1.7 mm long; upper surface densely puberulent, margins and undersurface cinerous and tomen- tulose; nervation inconspicuous. Axillary cymules 1-3 flowered, well-spaced along the upper 1/4 of the branches; primary peduncles (and also secon- daries) conspicuous, 1.0-2.5 mm long; pedicels 1.7—3.0 mm long, cinerous or simply pubescent; bracteoles subulate, 1.0-2.0 mm long, shorter than the pedicels they subtend. Calyx 5.5-8.0 mm (7.0 mm) long; the tube 4.0-5.0 mm (4.9 mm) long, tubular or slightly gibbous below, sparsely or densely pubescent; the upper teeth variously united from scarcely at all (Mueller 3302) to forming an upper lip 1.0-1.2 mm long (McVaugh 23685), the lobes triangular (deltoid in McVaugh 23685) ca 0.4 mm wide at the base, 0.4-0.9 mm long, spreading and outwardly curling at their apices, pubescent on their outer faces but not ciliate or only scarcely so; the lower teeth ascending, recurved, more or less subulate, narrowly triangular, ca 0.5 mm wide at the base 1.5-3.0 mm long, hirsute ciliate; annulus dense, more 244 or less included, ca 0.7 mm wide and seated at the juncture of the upper and lower teeth. Corolla lavender or blue, 15.0-16.0 mm (15.1 mm) long, glabrous in the throat, hirtellous pubescent in the tube forming a loosely defined annulus ca 5.0 mm wide; the tube and throat together 12.0-13.0 mm (12.5 mm) long gradually dilated upwardly, ca 3.0 mm wide above; the upper lip ligulate flat, straight, 1.8 mm wide, ca 3.0 mm long, very shallowly emarginate; the lower lip considerably larger than the upper, ca 6.0 mm wide, ca 5.5 mm long. Nutlets ca 0.7 mm wide, 1.4 mm long, the surface minutely areolate, and weakly glaucous. Chromosome number, 2” = ; Distribution (Fig. 3). Common in the isolated mountains of the Chi- huahuan Desert of central Mexico, mostly in oak woodlands, 6,500-8,100 ft. Jun-—Sep. TYPE: ZACATECAS. Concepcion del Oro, Sierra Madre Oriental 2600- 2700 m, 18-19 Jul 1934, Pennell 17432 (Holotype, LA!; isotypes, F! MICH! NY! PH! US!) Since the original description of H. patrinum in 1939, additional collec- tions have considerably expanded the known distribution of this relatively rare species. Notable collections have been made in the Sierra Mojada by Mueller and in the state of Aquascalientes by McVaugh. These have revealed considerable intra-specific variation. The southernmost collection of this species (McVaugh 23685) is quite different from the more northern “typical” material. Plants of the former are not cinerous with curling hairs; rather they are pubescent with straight, usually spreading hairs; the upper calyx teeth are well united; and their deltoid lobes approach the foliaceous upper lip found in H. jucundum and H. piperitum. The taxonomic position of H. patrinum is not certain. Its shrubby habit and entire cinerous leaves indicate an affinity with H. molle, H. montanum, and H. palmeri. | have therefore, like Epling and Stewart who placed H. patrinum between H, montanum and H. molle, included it in the subgenus Poliominthoides. 8. HEDEOMA MONTANUM Brandegee, Univ. Calif. Publ. Bot. 4: 188. 1913. Robust canescent, suffruticose perennials, 20.0—30.0 cm (28.0 cm) high. Shoots prolific, ascending and arising from a woody caudex or thickened tap- root, branching primarily in the lowermost regions with the branches acsend- ing at close angles, canescent with dense, short, spreading or curling hairs. Leaves deciduous from the lower stem regions at maturity, narrowly elliptical, 6.0-10.0 mm (8.0 mm) long, 2.0-3.1 mm (2.6 mm) wide; margins den- ticulate to dentate. occasionally entire; apex acute, base narrowly attenuated and petiolate, the petiole ca 1.0 mm long, both upper-and lower-leaf sur- faces densely cinerous; nerves somewhat elevated. Axillary cymules well- spaced along the upper 1/3-1/2 of the stems, occasionally somewhat con- gested, typically 1-5-flowered; primary peduncles inconspicuous less than 0.5 mm long; pedicels ca 2.0 mm long, cinereous, bracteoles subulate, ca 245 3.0 mm long, longer than the pedicels they subtend. Calyx 6.0-8.0 mm (7.1 mm.) long; the tube radially symmetrical or nearly so, ca 1.2 mm in diameter, cinereous with tightly curled puberulent hairs evenly distributed from the calyx base to the teeth apices; the calyx teeth only slightly differentiated into upper and lower sets and convergent at maturity; the upper teeth scarcely united for a 1/10 of their length, the upper lip 1.0-2.5 mm long, the lobes slightly shorter than the lower, subulate, ca 0.3 mm wide at the base 0.8—2.5 mm long; lower teeth subulate, recurved but convergent with the upper at maturity, ca 0.5 mm wide at the base, 1.0-2.5 mm long, hirtellous-ciliate; annulus extremely dense, well-exserted, ca 1.0 mm wide, seated approximately at the juncture of the upper and lower teeth. Corolla blue-lavender, ca 15.0 mm long, densely annulate at the juncture of the tube and throat; the tube and throat together ca 12.0 mm long, widening upwardly, ca 3.0 mm wide above; the upper lip ligulate, ca 2.5 mm wide, ca 3.0 mm long, straight, slightly emarginate and concave; the lower lip ca 4.0 mm long, 4.0 mm wide. Nulets ca 0.5 mm wide, ca 1.5 mm long, the surface minutely aerolate nearly smooth and glaucous. Chromosome number, 27” = 36. Distribution (Fig. 3). Arid mountains of southern Coahuila in the Chi- hauhuan Desert. Frequent on open, exposed, limestone ridges and hillsides, commonly associate with Agave lecheguilla. 7,000-8,000 ft. Jun; Sept-Oct. TYPE: COAHUILA. Sierra de la Paila, 7,000-8,000 ft. Oct 1910, Purpus 4964 (Holotype, UC!; isotypes, BM! F! GH! LA! US! phototype, LA! ). Hedeoma montanum is a rather rare species being known by only a few collections from a relatively restricted area of southern Coahuila. It is most closely related to H. molle. 9. HEDEOMA MOLLE Torr., Bot. U.S. Mex. Bound. Surv. 129. 1859. Poliomintha mollis (Torr.) A. Gray, Proc. Am. Acad. 8: 365. 1872. Robust, suffrutescent perennials or semi-shrubs, 16.0-55.0 cm (35.7 cm) high. Shoots numerous and long, ascending, tomentose-canescent or tomen- tulose-cinerous with short profusely branching hairs, arising from a woody caudex; branching almost completely confined to the basal region, the upper branches, when present, short and arising at close angles. Leaves narrowly to broadly elliptical to oval, highly variable, 3.0-10.0 mm (6.1 mm) wide, 8.5-22.0 mm (14.1 mm) long; margins entire or obscurely crenate; apex obtuse to somewhat acute; the base broadly or narrowly attenuated, petiolate, the petiole 0.5-4.0 mm (1.8 mm) long; surfaces evenly tomentose or tomentulose with intertangled, profusely branching hairs; nerves elevated. Axillary cymules 3-15-flowered (typically 10), occurring along the upper 1/3 of the stems, well-spaced or more commonly congested forming a dense but interrupted spike of cymes up to 7.0 cm long; primary peduncles promi- ent, 2.0-3.5 mm long, equal to or exceeding the length of the primary pedicel; pedicels 1.5-3.0 mm long, tomentose; bracteoles subulate as long as or longer than the pedicels they subtend. Calyx 5.2-7.0 mm (6.0 mm) long; 246 the tube nearly iene: symmetrical, or slightly upwardly dilating, ca 1.1 mm wide in diameter, 4.2-6.0 mm long, occasionally slightly gibbous or curved (Correll 30389 and Hinckly & Warnock 46876, respectively ), tomentose with highly branched evenly distributed hairs; the calyx teeth only slightly differentiated into upper and lower sets; the upper calyx teeth connate for ca 1/3 of their length, the lobes subulate or very nar- rowly triangular, ca 0.7 mm wide at the base, 0.3-0.8 mm (0.5 mm) long, slightly laterally nadine or convergent, erect to sharply outwardly curling, the outer face and margins tomentose, the inner face covered with short pilose hairs as an extension of the prominent annulus; the lower teeth subulate, 0.3-0.5 mm wide, at the base, 0.6-1.3 mm (0.9 mm) long, clothed in a branched hair tomentum; annulus exserted, extremely dense, with unbranched, ilose hairs, ca 0.7 mm wide, seated approximately at the juncture of the poorly defined upper and lower lips. Corolla lavender 13.2-14.0 mm (13.6 mm) long, sparsely pubescent in the throat and in the lower tube but not forming a well-defined annulus; the tube and throat together 11.0-12.0 mm (11.5 mm) long conspicuously dilated upwardly, 3.0-3.5 mm wide above; the upper lip broadly ligulate, conspicuously emarginate, ca 1.9 wide, ca 2.2 mm long; the lower lip 4.0-5.0 mm wide, 3.0-4.0 long. Nutlets narrow, ca 0.3 mm wide, ca 1.2 mm long, the surface minutely areolate, and weakly glaucous. Chromosome number, 27 = 36 Distribution (Fig. 3). Restricted to rock slopes and ledges of the Davis, Glass, Chisos, Chinati, and Sierra Vieja mountains of Trans-Pecos, Texas. 1,000-7,000 ft. Jul-Nov. TYPE: TEXAS. Presidio Co., w of Alpine, cliffs near Puerta de Paysano, Sep. Bigelow sn. (Holotype, NY! isotype, NY! ). Hedeoma molle occurs in widely separated populations, however, Judging from its representation in herbaria, it is abundant where it occurs. Specimens from the northern portion of its range (Davis Mountains: Madera Canyon, Cory 14042; Ajuga Canyon, Correll 33341) are unusually large-leaved and show only a slight differentiation between the upper and lower calyx teeth. ne specimen (Correll 33341) has gray-green leaves which measure 24.0 mm long and 16.0 mm wide. To the west, in the more arid Chinati Mountains, the leaves are quite diminutive, measuring ca 7.5 mm long and 2.5 mm wide, and there is a marked differentiation of the upper and lower calyx teeth. (McVaugh 7453 H, molle is most t closely related to H. montanum. III. SUBGENUS SATUREJOIDES, Irving, subg. nov. rbae annuae vel perennes raro suffrutescentes, 5.0-G60.0 cm altae; folia ovata, elliptica vel late elliptica, oblonga, lanceolata vel linearia, 3.0-28.0 mm longa, 1.0— 16.0 mm lata, margine integra vel dentata, apice obtusa vel acuta, ad basin petiolata vel sessilia; cymulae 1—20-florae, remotae; calyx 4.0-10.0 mm longus, valde zygo- morphus; calycis tubus leviter gibbosus vel saccatus; corolla varia (minuta vel longi tubularis vel magna conspicuaque), 53.5-20.0 mm longa; mericarpum oblongum, 247 uplo longiorum quam latiorum, brunneum (ad maturitatem), areolatum glaucum mucilaginum (in stat humectato); chromosomatum numerus, 2” = 34, 36, 44, 72. Herbaceous annuals or perennials, occasionally suffrutescent, 5.0-60.0 cm high. Shoots solitary to numerous, strict, ascending or decumbent, sometimes rhizomatous sending out adventitious roots from the lower nodes, arising from a taproot, trailing rootstock, or woody caudex. Leaves membranous, oval, linear, 3.0-28.0 mm long, 1.0-16.0 mm wide, margins entire to coarsely dentate, apex obtuse or acute, base subsessile to petiolate. Cymules 1—20- flowered, usually well-spaced along the stems, occasionally ne toward a apices forming a short “spike.” Calyx zygomorphic; the tube slightly ous to saccate, 4.0-10.0 mm long, variable in pubescence; the calyx an well-differentiated into upper and lower sets; the upper teeth variously connate forming as upper lip; the lobes subulate, deltoid, variously ciliate, commonly spreading and reflexed; the lower teeth subulate and recurved, ciliate; annulus well developed. Corolla usually lavender-blue, variable from minute (H. hispidum) to long tubular (H. costatum) to large and showy (H. pulcherrimum. H. diffusum, and H. byssopifolium), 5.5-45.0 mm long. Nutlets oblong or ovate in outline usually twice as long as wide, occasion- ally orbicular, brown at maturity, the surface areolate or foveate, glaucous and mucilaginous when moistened. Chromosome numbers, 27 = 34. 36, 44, i Type species: HEDEOMA COSTATUM Gray. The subgenus Satwrjoides is subdivided into two sections, Alpine and Saturejoides. \t is the largest subgenus in Hedeoma and its species are reticulately related one to another e most homogeneous unit within the subgenus is the section Alpine. It is unified by a number of features. Section Saturejoides, on the other hand, is a more diversified group which presents greater difficulty in characteriza- tion and circumspection. A. SECTION ALPINE Irving, sect. nov. e perennes 1.0-5.0 dm altae; surculi multi ascendentes vel decumbentes; asi ee fere sphaeroideum vel ovoideum, diametro suo duplo longium vel brev Perennial herbs 10.0-50.0 cm tall. Shoots few to numerous, ascending or decumbent and rhizomatous, sending out adventitious roots from the lower nodes, arising f slender taproot. Leaves glandular, punctate on or obscurely serrate, apex obtuse, base typically tapered and_ subsessile. Cymules 3-7-flowered. Calyces variable between species, conspicuously zygomorphic, 4.0-8.0 mm long; the tube tubular or saccate; the upper teeth united to form a conspicuous upper lip, the lobes narrowly triangular or 248 deltoid; the lower teeth narrowly seo or subulate, recurved. Corolla showy, 5.0-17.0 mm long, the upper lip obcordate, commonly reflexed, the lower lip various, spreading. Nutlets ae or ovate in outline, nearly spherical, ovoid, ovate in cross section, ca as long as wide, brown at maturity, the surface areolate. Chromosome number, 27 = 44. ype species: HEDEOMA PIPERITUM Benth. A number of similarities, in addition to the morphic ones, unite the species of section Alpine: (1) they are found at the higher elevations, (2) they are the only representatives of the genus occurring in the Sierra ee Occidental of Mexico and, except for the South American species, H. ; galaefolium, they are all found in this western range; (3) they all oe a chromosome number of 27 = 44. From these diverse lines of evidence, it is difficult to deny the naturalness of section Alpine. The weakest evidence for the groups naturalness comes from morphology, since the combination of characters which unite them are subtle. 10. HEDEOMA POLYGALAEFOLIUM Benth., Lab. Gen. et Sp. 367. 1834. Small wiry perennial herbs, ca 10.0 cm high. Shoots decumbent and rhizo- matous forming adventitious roots from the lower, buried nodes, branching freely along the axis; glabrate or with a few minute retrorsely curved hairs. Leaf pairs crowded with internodes uniformly ca 1.0 mm long. Leaves ascending, readily deciduous from the lower portions of the older stems, chartaceous becoming folded with drying, oblong to oblanceolate, 2.0—2.5 mm (2.3 mm) wide, 6.5-7.0 mm (6.8 mm) long, margins entire, apex obtuse, base narrowly attenuate and subsessile to a decurrent base; leaves glabrous throughout; nervation consisting solely of a prominent midrib. Cymules 1-3-flowered, congested toward the branch apices; peduncles 2.0- 2.5 mm long; pedicels 4.5-5.0 mm long, puberulent with retrorsely curved hairs; bracteoles lanceolate shorter than the pedicels they subtend. conspicuously zygomorphic, ca 8.0 mm long; the tube tubular—funnelform 4.5 mm long not at all gibbous below, dilating and flaring upwardly, ca 2. mm in diameter above, glabrous; the upper teeth united for ca 2/3 of their length forming an upper lip ca 2.5 mm long, pubescent within, the lobes narrowly triangular, acuminate at the apex, ca 0.5 mm wide at the base, ca 1.5 mm long, slightly spreading and reflexed, the margins glabrous; the lower teeth triangular—subulate, recurved, ca 3.0 mm long, hirtellous—ciliate; annulus dense ca 1.2 mm wide, seated ca 1.0 mm below the juncture of the upper and lower teeth but extended into the faces of the upper teeth, more or less included. Corolla ca 10.0 mm long. Nutlets 0.8 mm wide, ca 0.8 mm long, brown at maturity, the surface smooth and aerolate. Chromo- some number, unknown. aa ecaalner (Fig. 12). Brazil, province Rio Grande do Sul. ZIL. 1841, Sellow s.n. (Holotype, K! isotype LA!: phototypes, US! ). = “ LA! NY! a 11. HEDEOMA BELLUM (Epl.) Irving, Brittonia 22: 345. 1970. Hesperozygis bella Epl., Bull. Torr. Bot. Club 67: 512. 1940. Small wiry perennial herbs, 15.0-30.0 cm in height. Shoots few to numer- ous, typically shortly decumbent (up to 7.0 cm) and rhizomatous sending out adventitious roots from the lower, buried nodes; sparsely branching; stems densely puberulent with the hairs spreading above and _ antrorsely curving below. Leaves spreading or cernuous, readily deciduous from the lower portions of the older stems, chartaceous, the margins often becoming revolute upon drying, oblanceolate-elliptical, 3.4-4.0 mm (3.6 mm) wide, 10.0-17.0 mm (15.0 mm) long, margins entire, apex obtuse, base gradually attenuate and subsessile; glabrous above, strigulose on the margins and midrib; nerves consisting of a prominent midrib and inconspicuous second- daries. Cymules 2-3-flowered, congested in the upper 1/4 of the stem forming a short terminal “spike”, peduncles distinct but short, ca 1.5 mm long; pedicels 3.0-4.0 mm long, hirtellous with hairs spreading. Calyx conspicuously zygomorphic, ca 8.0 mm long; the tube ca 5.0 mm long not gibbous below, dilating upwardly, ca 2.5 mm wide above, sparsely hirtellous; upper teeth connate for approximately 2/3 of their length forming a con- spicuous upper lip, ca 2.5 mm long, pubescent within with short pilose hairs, the lobes more or less deltoid or semi-ovate, conspicuously apiculate at the apex, ca 1.0 mm wide at the base, 1.0 mm long, sharply reflexed, strig- ulose-ciliate; the lower teeth sharply recurved, broadly triangular and taper- ing into a long aristate apex, ca 14 mm wide at the base, 3.4-3.8 mm long, hirsute-ciliate; annulus dense ca 1.0 mm wide, seated ca 1.0 mm below the upper lip, included. Corolla ca 12.0 mm long, pink and red dotted in the throat, minutely hirtellous within; the tube ca 10.0 mm long, grad- ually dilating upwardly; the upper lip obcordate, emarginate, straight or slightly reflexed, ca 2.0 mm long; the lower lip ca 5.0 mm broad, ca 3.0 mm long. Nutlets ca 0.8 mm wide, ca 0.8 mm long. Chromosome number, un- known. Distribution (Fig. 4). Hedeoma bellum is known only from the type sae where, according to label data, it occurs on damp stream banks. i T : JALISCO. Sierra Madre Occidental, San Sebastian, east of Arro Santa ee 1500 m, 18 Jan 1927, Nee 1514 (Holotype, UC! oe F! NY!) Vegetatively, especially in the leaves and in the inflorescence, H. bellum is similar to H. patens. It might also be noted that the common name of this species “oreganum’, undoubtedly derived from its characteristic odor, is also the vernacular name given to H. patens. In the finer details of the calyx, one of the most elaborate in the genus Hedeoma, it is unique among the Nozth American representatives, but is similar to the equally restricted South American species, H. polygalaefolium. 250 12. HEDEOMA JUCUNDUM Greene, Pitt., 1: 156. 1888. Dense, caespitose perennial herbs, 10.0-15.0 cm in height. Shoots typi- cally numerous, widely ascending or shortly decumbent; branched profusely at the base, and frequently producing adventitious roots from the lower eee Sse fi ee l eenera Tl ~ : UF ‘ig. 4. Distribution of Hedeoma patens (open circles), H. floribundum (open square), H. jucundum (closed circles), H. bellum (open triangle) and H. piperi (closed squares). 251 nodes, unbranched above or sparsely so; stems moderately to densely hirtel- lous above with retrorsely curving or occasionally spreading hairs, becom- ing puberulent below. Leaves spreading or ascending, chartaceous with the margins frequently becoming revolute Ee drying, ovate to narrowly elliptical, 4.0-6.0 mm (5.3 mm) wide, 8.0-12.0 mm (11.0 mm) long, margins entire to ee serrate, apex “obtuse, base shortly attenuate and subsessile; glabrous above, strigulose on the margins and the midrib below; nerves not conspicuously raised. Cymules 3-5-flowered, well-spaced along the upper 1/2 of the stems; peduncles inconspicuous less than 1.0 mm in length; pedicels 3.0-5.0 mm long, moderately hirtellous to puberu- lent. Calyx 6.0-7.0 mm (6.5 mm) long; the tube 3.5-4.5 mm (4.0 mm) oe slightly gibbous below for ca 1/2 of its length, the distended region 1.3 mm wide, dilating upwardly to ca 1.5 mm wide, sparsely to moder- ere hirsute-hirtellous; the upper calyx teeth connate for ca 1/2 of their length forming a pronounced, laterally rounded upper lip ca 2.3 mm long, the margins outwardly extended, the lobes triangular, ca 0.8 mm wide at the base, ca 1.2 mm long, sharply reflexed and laterally spreading, hirsute— ciliate; the lower teeth recurved, narrowly triangular—subulate, ca 8.0 mm wide at the base, ca 3.0 mm long, ciliate; the annulus dense ca 1.0 mm wide and arched upwardly into the connate portion of the upper lip, included. Corolla 11.0-12.0 mm (11.5 mm) long, white or pink with violet mark- ings on the lower lip, glabrous within; the tube 9.0-10.0 mm long weakly dilating upwardly, ca 2.0 mm wide above; the upper lip obcordate in outline, ca 2.5 mm wide, 3.0 mm long, reflexed; the lower lip broad, ca 7.0 mm wide, ca 5.0 mm long. Nutlets ca 0.8 mm wide, 1.0 mm long, the surface minutely areolate, not glaucous. Chromosome number, 27 = 44 Distribution (Fig. 4). Sierra Madre Occidental, common in the pine forests west of the village of El Salto, Durango. 7000-9000 ft. Aug—Oct. TYPE: DuRANGO. From the Sierra Madre, west of Durango, 8100 ft. Sep— Oct. 1881, Forrer sn. (Holotype, ND! isotypes, F! GH! LA! NMC! NY! PH! POM! UC! US!). Hedeoma jucundum is morphologically uniform and is most closely re- lated to H. piperitum. 13. HEDEOMA PIPERITUM Benth., Lab. Gen. et Sp. 730. 1835. Cunila piperita Moc. et Sesse ex Benth., Lab. Gen. et Sp. 730. 1835. Loose, wiry, perennial herbs, somewhat caespitose, 10.0—20.0 cm tall. Shoots few to numerous, widely ascending or shortly decumbent; branching from the base, and frequently producing adventitious roots from the lower nodes, upper shoots typically unbranched or bearing a few short branches; stems densely hirsute with spreading hairs in the upper regions, hirtellous to puberulent below. Leaves spreading or drooping, membranous, broadly ovate to sub-orbicular, 5.0-13.0 mm (8.0 mm) wide, 7.0-18.0 mm (12.5 mm) long, margins entire or obscurely serrate, apex obtuse, base shortly attenuate 222 and subsessile; glabrous above, strigulose below on the nerves and margins; nerves not eae raised. Cymules 3—7-flowered well-spaced along the upper 1/2 of the stems; peduncles inconspicuous less than 1.0 mm in length; pedicles 2.0-5.0 mm_ long, hirsute. ee strongly ee reaga 5.5-6.5 mm (6.0 mm) long; the tube 3.0-4.0 mm (3.5 mm) long, tinctly saccate below for ca 2/3 of the tube’s ne the distended region 1.5-2.0 mm wide, sparsely to moderately hirsute; the upper calyx teeth united well over 1/2 of their length to form a reflexed, laterally rounded, upper lip ca 2.0 mm long, the margins pba flaring, hirtellous—ciliate, glabrous or pubescent on the inner face, the lobes semi-ovate to deltoid, ca 0.7 mm wide at the base, 0.7 mm ion. ciliate; the lower teeth recurved, narrowly triangular, ca 0.8 mm wide at the base, ca 2.3 mm long, hirsute— ciliate; annulus moderately dense, ca 0.5 mm wide and arched somewhat into the upper lip, included. Corolla, 7.0-8.5 mm (8.0 mm) long, white with maroon markings on the lower lip, puberulent to glabrous within; the tube 5.5-7.0 mm (6.5 mm) long gradually dilated upwardly, the upper lip obcordate in outline, ca 2.3 wide, 1.5-2.0 mm long, straight or very slightly reflexed; the lower lip broad, 3.5-4.5 mm wide, ca 3.0 mm long. Nutlets 0.8 mm wide, ca 1.0 mm long, the surface minutely areolate, not glaucous. Chromosome number 22” = 44, Distribution (Fig. 4). Pine forests of central Mexico, 6,000-10,000 ft. Particularly abundant in sunny, open slopes associated with rocky lava or volcanic ash. Aug—Oct. YPE: Mexico. Mocino et Sesse (lectotype, here designated BM!; iso- types, F! MA, photoisotypes, LA! TEX! ). The specimen bearing the her- barium name C. piperita Moc. et Sesse was observed by Bentham in the Dunant herbarium which obtained the type from the herbarium of Pavon, Hedeoma piperitum is quite distinct and relatively homogeneous through- out its narrow range. The salient morphological features which distinguish this species are its caespitose, broadly ovate-suborbicular leaves, and saccate calyx with its well-developed, expanded, upper lip. It is closely related to H. jucundum. Examination of the westernmost populations of H. piperitum (Cheron, Michoacan) reveal several subtle morphological departures from the eastern populations around Mexico City. The western populations vary in the direction of H. juwcundum; that is, they have a more robust habit; thick, oes leaves ( Beal 4); and retrorsely curling stem pubescence 13) (Beal 14. HEDEOMA PATENS Jones, Contrib. West. Bot. 12: 70. 1908. Perennial herbs 20.0-55.0 cm high, averaging 30.0 cm. Shoots numerous, herbaceous above, decumbent (often one-sidedly so), woody below, branch- ing freely, densely puberulent or hirtellous with hairs retrorsely curling. Leaves variable, laxly spreading to drooping, membranous, broadly to narrowly elliptical to oblanceolate, 2.0-8.0 mm (4.0 mm) wide, 5.0-22.0 mm (12.0 — 253 mm) long, margins entire or obscurely crenate, apex obtuse, base narrowly attenuate, short petiolate (seldom exceeding 1.5 mm) or subsessile; glabrous above, strigulose below and on the margins; nerves inconspicuous. Cymules well-spaced along the upper 2/3 of the stem with 3-7 flowers per cymule (typically 3-5), peduncles obscure to ca 2.0 mm long; pedicels 3.0-5.0 mm long, typically exceeding or equaling the length of the calyx tube, densely puberulent. Calyx 4.0-655 mm (6.0 mm) long, moderately zygomorphic; the tube 3.5-5.0 mm (4.0 mm) long, slightly gibbous below for 1/3-1/2 its length, the distended region 0.7-1.6 mm wide, puberulent to hirtellous with hairs strongly antrorsely curling; the upper calyx teeth connate for 1/3-1/4 of their length, forming a short upper lip 1.2 mm long, the lobes deltoid, ca 0.5 mm wide at the base, ca 0.5 mm long, laterally spread- ing, weakly reflexed, strigulose-ciliate; the lower calyx teeth recurved, sub- ulate highly variable as to length, ca 0.5 mm wide at the base, 1.2—2.3 mm (1.8 mm) long, typically hirsute—ciliate, annulus dense, ca 0.5 mm wide, typically seated ca 0.5 mm below the juncture of the upper and lower teeth and included, occasionally exserted (Knobloch 5407). Corolla 7.0-9.0 mm (8.0 mm) long, pink, glabrous within; the tube 5.0-7.0 mm (6.0 mm) long, the throat scarcely dilated upwardly (ca 2.0 mm wide above); the upper lip obcordate in outline, ca 2.0 mm broad at the widest point, ca 2.6 mm long, reflexed; the lower lip, ca 5.0 mm wide, ca 4.5 mm long. Nutlets not quite twice as long as broad, 0.5 mm wide, ca 0.9 mm long, the surface minutely areolate, and weakly glaucous. Chromosome number 2” = 44. Distribution (Fig. 4). Pine forests of the Sierra Madre Occidental. 5000- 7000 ft. Aug—Oct. TYPE: CHIHUAHUA. Sierra Madre Mts., “Guayanopa Canyon”, 5000 ft., 23 Sep 1903, Jones s.n. (Holotype, POM! phototype, LA! ). Hedeoma patens is an exceedingly variable species. The specimens which I have examined seem to form a transition between H. hyssoptfolium and Hl. pulcherrimum (SECTION Saturejoides), and H. jucundum and H. piperi- tum (SECTION Alpine). Epling and Stewart (1939) placed H. patens close to H. pulcherrimum. Admittedly, they resemble each other in many respects, particularly in habit and vegetative morphology. They can be readily separated from each other, however, by the important details of the calyx and corolla. 15. HEDEOMA FLORIBUNDUM Standley, Rep. Spec. Novy., Beihefte 115: 31. 192). Perennial herbs 20.0—32.0 cm high. Shoots woody below, ascending or de- cumbent, branching freely along the axis with the branches ascending at close angles, clothed below in long spreading pubescence, above puberulent with hairs tightly retrorsely curling. Leaves laxly spreading to cernuous, membranous, narrowly elliptical or oblanceolate, 4.0-5.0 mm (4.7 mm) wide, 15.0-18.5 mm (15.9 mm) long, margins entire or obscurely crenate, 254 apex obtuse, base narrowly petiolate with the petiole ca 1.0 mm _ long, glabrate above, puberulent below; nerves inconspicuous. Cymules aggregated toward the branch apices forming a short terminal spike, ca 4.0 cm long, 3-7-flowered (typically 3-5), peduncles and pedicels contorted; peduncles inconspicuous to ca 1.0 mm long; pedicels 2.0-2.5 mm long, shorter than the calyx tube, puberulent. Calyx 3.0-3.5 mm oe Uses Bel aaa puberlulent with antrorsely curling hairs, the tube 2.5-3.0 mm long. The tube weakly distended ca 1.0 mm wide; the upper and loca: teeth only slightly differentiated; the upper teeth united for ca 1/3 of their length forming an upper lip ca 0.5 mm long, the lobes erect, deltoid ca 0.4 mm wide at the base, ca 0.4 m long; the lower teeth ca 0.5 mm wide at the base, 0.5 mm long; annulus included, ca 0.5 mm wide and seated just below the juncture of the upper and lower lips. Corolla 5.0-6.0 mm long, glabrous within; the tube 4.0-5.0 mm long gradually widening upwardly, upper lip obcordate, 1.0-1.2 mm wide, 1.0-1.5 mm long; the lower lip ca 2.5 mm wide, 2.5 mm long. Nutlets 0.5 mm wide, 0.9 mm long, the surface minutely areolate, glaucous. Chromosome number, unknown. Distribution (Fig. 4). Pine forests along the Rio Mayo in the state of Chihuahua, ana . Oct YPE: CHIHUAHUA. Cans Quicorichi, Rio Mayo, 6 Oct 1935, Gentry 1938 (Hol eae F!; isotype, he vegetative features of H. floribundum are similar to those of H. patens, and they have similar distributions. The salient differences between these two species reside in the inflorescence and calyx. Unlike H. patens, the inflorescence of H. floribundum is a dense terminal spike and the calyx shows a greater trend toward radical symmetry, both in the tube and in the disposition of its teeth. B. SECTION SATUREJOIDES, Irving, sect. nov. Herbae annuae vel cee: raro suffrutescentes, 5.0-60.0 cm altae; folia varia, margine integra vel aepe serrata vel dentata, ie basin petiolata vel subsessilia; 4 a ened : *s; coroll: vel magna aera ue, 53.5-20.0 mm ieee mericarpium oblongum, circa duplo longiorum quam latiorum brunneum (ad maturitatem), areolatum vel foveolatum. Herbaceous annuals or perennials, occasionally suffrutescent and woody below, 5.0-60.0 cm high. Shoots solitary to numerous ascending or decum- bent, occasionally rhizomatous and forming eee roots. Leaves glan- dular punctate, variously shaped. Margins entire, serrate, or dentate, apex acute or narrowly obtuse, base petiolate or subsessile. Ce 1—20-flowered. Calyx 4.0-10.0 mm long, zygomorphic; the tube slightly gibbous to distinctly saccate; the upper teeth variously connate, the lobes subulate or narrowly triangular; the lower teeth subulate and recurved; both upper and lower teeth variously ciliate. Corollas inconspicuous to large and showy, 5.5—45.0 mm long; the upper lip ligulate, straight or reflexed, flat or sub-galeate; 255 the lower lip spreading. Nutlets oblong, at least twice as long as wide, ovate- triangular in cross section, brown at maturity, the surface areolate or foveate, glaucous. Chromosome numbers, 2” = 34, 36, 7 TYPE: HEDEOMA COSTATUM Gray. 16. HEDEOMA PULCHERRIMUM Woot. & Standl., Contr. U.S. Natl. Herb. 16: 168. 1913. Robust, perennial herbs, 15.0-40.0 cm in height (typically between 20.0 and 30.0 cm). Shoots few to numerous, ascending or very shortly decumbent, woody below, herbaceous above; branching frequently in the lower half, the branches long and widely ascending, stems below clothed in villous pubescence, above sparsely to densely puberulent with the hairs tightly retrorsely curling. Leaves thick membranous, elliptical-oblong, 8.0-18.0 mm (13.4 mm) long, 2.0-6.5 mm (4.2 mm) wide, margins entire to obscurely serrate, base attenuated, short petiolate or subsessile, apex obtuse; leaves glabrous above, strigulose below and on margins; nerves conspicuous. Cymules well-spaced along the upper 2/3 of the stems with 3-7 flowers per cymule (typically 3-5); peduncles short; 1.0-2.0 mm long; pedicels 2.0—3.5 mm long, minutely puberulent; bracteoles short. Calyx 6.0-7.0 mm (6.6 mm) long; the tube 3.0-5.0 mm (4.0 mm) long, slightly gibbous below for ca 1/2 of the tube’s length, the distended region ca 1.2 mm wide, typically moderately hirsute; the upper calyx teeth connate for 1/3-1/4 of their length forming an upper lip, 1.5-2.0 mm long, the lobes very narrowly triangulate-lanceolate (nearly subulate), ca 0.3 mm wide at the base, ca mm long, laterally spreading and reflexed (with maturity becoming erect and convergent with the lower teeth), hirsute or hirtellous—ciliate, the lower calyx teeth recurved, subulate, ca 0.5 mm wide at the base, ca 2.7 mm long; annulus dense, ca 1.0 mm wide, seated at the juncture of the upper and lower teeth, exserted. Corolla 10.0-14.0 mm (12.4 mm) long, blue, sparsely villous within the throat and densely short pubescent in the tube forming a well-defined annulus, 2.0-3.0 mm in width; the tube 8.0- 10.0 mm (9.0 mm) long, the throat strongly dilated, ca 5.0 mm wide; the upper lip broadly ligulate, slightly emarginate, 3.0-3.3 mm broad, 3.0-4.0 mm long, straight (not upwardly reflexed); the lower lip broad ca 7.0 mm wide, ca 5.0 mm long. Nutlets ca 0.6 mm wide, 1.2 mm long, the surface minutely areolate, weakly glaucous. Chromosome number 27 = 36 Distribution (Fig. 5). Restricted to the Sacramento and White Mountains of south central New Mexico where it is common in Pine or Spruce-Fir forests, Poca) in disturbed areas; 5000-9000 ft. Jul-Sept. TYPE: New Mexico. Lincoln Co., collected in the White Mountains, 6500 fet., 30 July 1897, Wooton 241 (Holotype, US! isotypes, DS!, LA! ND! NMC! NY! POM! UC! US!). H. pulcherrimum has been difficult taxonomically to place. In the original description, Wooton and Standley state: “This species [H. pulcherrimum] seems to be on the dividing line between the two principal groups of species: 256 its upper calyx lobes are not foliaceous expanded as in one group [Sect. Alpine} nor narrowly subulate as in the other but are triangulate-lanceolate and hispid-ciliate like those of the second group [Sect. Satwrejoides}". Al- though recognizing its transitional nature, they thought it closest to H. jucundum. Epling and Stewart however, aligned H. pulcherrimum with H. I have treated H. pulcherrimum as being most closely related to H. drummondii. Except for characters of the calyx, the western elements of the H. drummondiu complex are identical, or nearly so, to H. pulcherrimum. Moreover, natural hybridization occasionally occurs between these species with the F, hybrids vigorous and partially fertile (ca 20%). Tes, we poeoee eee Te meme im eae ~~"... ° fe} ween lo 7 wet ee °. _- - -< fone) -— — = -<. -. 4 Wot, pay ea! Fig. 5. United States distribution of Hedeoma pulcherrimum (squares), H. drummondti (open circles), H. reverchonit var. serpyllifolium (triangles), and H.+r. var. reverchonii (closed circles) 27 17. HEDEOMA DRUMMONDJ Benth. Lab. Gen. et Sp. 1836. Hedeoma ciliatum Nutt., Journ. Acad. Nat. Sci. Phil., N.S. 1: 47. TYPE: NEW MEXICO: se Fe Co., Santa Fe, Gane 5M, eee GH! isotype, K!) Hedeoma ovatum A. Nels., Bull. Torr. Bot. Club 31: 245. TYPE: a Pole Creek, 30 Jun 1895, Nelson ae sees RM!; isotype, GH ie ee Rydb., Bull. Torr. Bot. Club 36: 695. 1909. TYPE: NEBRASKA. Banner Co., Cliffs of Canyons, Aug 1890, Rydberg x297 (Holotype, NY!) Hedeoma Senee aie Flora Rocky Mts. 750. 1917. Based on H. longifloru Hedeoma drummondi Bal var. seta Irving, Britt. 22: 342. 1970 TYPE: MEXICO, San Luis Pot 7.8 mi w of Guadalcazar, 10 ‘Ate 1965, Irving 632 (Holotype, TEX!; ae MONTU! ) Annuals or robust perennials, 15.0-45.0 cm high. Shoots few to numer- ous, shortly decumbent and/or frequently spreading upwardly at wide angles, branching freely along the lower 1/2 to 2/3 of the stems with the branches typically long, straight, and spreading at angles of ca 45°; stems densel pubescent or puberulent above with the hairs tightly retrorsely curling, glabrate below. Leaves bright green, cernuous, thick membranous, linear to elliptic-oblong, 1.2-4.0 mm (2.2 mm) wide, 5.0-11.0 mm (7.7 mm) long, entire, obtuse at apex, narrowly attenuated and short petiolate or sub- sessile, glabrous or sparsely pubescent above, densely pubescent below; nerves inconspicuous. Cymules somewhat crowded along the upper 1/2-3/4 of the stems with 3-7 flowers/cymule,; peduncles inconspicuous to ca 1. mm long; pedicels 2.0-3.5 mm long, pubescent; bracteoles subulate ca 1/2 as long as pedicels. Calyx finely hirsute, 5.0-6.0 mm (5.6 mm) long, con- spicuously saccate for ca 2/3 of its length, the distended region 1.2-1.3 mm wide; the tube tapering to a slender neck ca 0.5 mm wide closing the orifice after anthesis; the upper calyx teeth slightly connate, aaa or very narrowly triangular, ca 0.2 mm wide at the base, 1.0-1. m long, erect and dorsiventrally convergent with the lower teeth to oe the orifice, hirsute-ciliate; the lower teeth subulate, recurved and arching over the upper teeth, 0.3 wide at the base, 1.5-2.0 mm long, ciliate; annulus dense ca 1.5 mm wide and seated ca 1.5 mm below the juncture of the upper and lower teeth. Corolla blue, 7.0-9.0 mm (8.0 mm) long sparsely pubescent in the throat, densely to below forming an annulus 0.7-1.5 mm wide; the tube 1.0 mm wide; the upper lip ligulate 1.0-1.5 mm long; the lower lip 2.5—3.0 mm wide and 2.5-3.0 mm long. Nutlets ca 0.4 mm wide, ca 1.2. mm long. Chromosome numbers, 27 = 34, 36. Distribution (Fig. 5). Various habitats from Montana to San Luis Potosi, Mexico; and western California to western Nebraska. Jun—Sep. TYPE: Nuevo Leon. Near Monterrey, 1828, Berlandier s.n. (Holotype, isotypes, BM! GH! ). . drummondiw is a morphologically variable species and a number of 258 races can be discerned within it. In central Texas along the escarpment of the Edwards Plateau the plants take the form of annuals or biennials with linear leaves, sparsely pubescent calyx, and the teeth strongly converging. Many of these populations are aneuploid (27 = 34) and hybridize with both varieties of H. reverchonit. Towards the west and north, the plants take the form of tufted caespitose perennials with leaves more oblong- elliptic and the calyx to varying degrees sealed at the summit. In southern Nevada hybridization with H. nanum produces still another morphological grouping. To the west, mesic montane forms possess long cernuous elliptical- oblong leaves and calyx teeth only very loosely closed at the summit. This last form is common in the sha Mountains of west Texas and in the Sacramento Mountains of central New Mexico where it hybridizes with pulcherrimum. The eastern Mexico populations, occurring along the eastern edge of the Sierra Madre Oriental, vary in the direction of H. reverchonii var. serpyllifolium. 18. HEDEOMA REVERCHONII Gray Key to the varieties of H. reverchonit 1. Plants 25.0-60.0 cm tall branching usually restricted to the base and the upper 1/2 = on shoots; leaves gray green, ca 3.7 mm wide, 11.7 mm long; calyx 6.0-7.0 mm , very coarsely hirsute to hoary pilose; the upper ne widely oe ein ae 10.0-15.0 mm long, the tube conspicuously dilated _ plants restricted to the eastern edge of the Edwards Plateau and Lampa ys ut Plain of central Texas, and south central Oklahoma; lemon-scented . “pic = eee iauiav ee paras . Aor. var. aN 1. Plants 15. 0-40. 0 cm tall branching freely one the 2h; of the shoots; leaves < green, ca 3.0 mm wide, 8.3 mm long; calyx 5.0—6. m long, coarsely hirsute, upper ee erect or only slightly spre Aue la a “oral small, 8.0-10 mm long, the tube only very slightly dilated Uae plants from eastern — of Edwards eee south to the Central Coastal Plain Texas and west uth- eastern New Mexico; camphor-scented ... 8b. Hur. var. eh Sse 18a, HEDEOMA REVERCHONII Gray var. REVERCHONII, Syn. Fl. ed 2, 2: 460. 1886 Hedeoma drummondii - reverchonit A. Gray, Syn. Fl. 2: 363, 8. Hedeoma latum Small, Fl. S.E. U.S. 1040, 1337, 1903. TYPE: a an Dallas Co. Rocky prairies, near Ses Texas, Jun—Sept. Reverchon (Curtis 2020) (Holo- type, NY! isotypes, BM! F! GH! ND! NY! SMU! US!) Robust, suffruticose perennials, 25.0-60.0 cm high, with the shoots of the previous season usually persistent. Shoots ascending or shortly decum- bent, arising from a stout woody taproot, branching more or less restricted to the very base, branches long and well-developed; stems densely pubescence, nearly canescent above with tightly retrorsely curling hairs, glabrous below. Leaves gray-green, eat erect, coriaceous, elliptic-oblong and highly vari- able in size, 3.0-5.0 mm (3.7 mm) wide, 9.0-14.0 mm (11.7 mm) long, margins entire, apex hae we mars attenuated and subsessile, sparsely pubescent above, densely so below and on the margins; nerves inconspicuous. 299 Cymules 2—4-flowered (occasionally more) well-spaced along the upper 1/2 of the stems; peduncles 1.0-1.5 mm long; pedicels 3.0-4.0 mm long, clothed in tightly retrorsely curling hairs, bracteoles short. Calyx 6.0-8.0 mm (6.6 mm) long; the tube 4.0-5.5 mm (4.2) long, saccate below tapering to a narrow neck partially sealing the calyx orifice, hoary villous with the hairs antrorsely curling at their tips and up to 1.5 mm long; the upper calyx teeth not connate or slightly so, the lobes subulate, ca 0.2 mm wide at the base and ca 1.0 mm long, laterally spreading, slightly reflexed but with the base convergent to the lower teeth and closing the orifice, hirsute— ciliate; the lower teeth subulate, recurved 2.0-2.5 mm long, ciliate; annulus extremely dense ca 1.5 mm wide and seated ca 1.5 mm below the juncture of the upper and lower teeth. Corolla large and showy 10.0-15.0 mm (12.0 mm) long, white or lavender, variously marked on the throat, sparsely pubescent in the throat, densely so in the tube forming a well-defined annulus; the tube dilating distally to ca 5.0 mm wide above; the upper lip ligulate and sub-galeate, ca 5.0 wide at the base, 4.0-5.0 mm long, the lower lip ca 4.0-5.0 mm wide, ca 5.0 mm long. Nutlets ca 0.6 mm wide, ca 1.2 mm long, areolate and glaucous. Chromosome number, 27 = 34. Distribution (Fig. 5). Open, exposed, calcareous outcrops in southern Oklahoma and central Texas along the fault escarpements of the Edwards Plateau and Lampasas Cut Plain. Also collected infrequently on calcareous outcrops in east Texas, Arkansas and Alabama, as yet however, it is un- reported in Louisiana and Mississippt. TYPE: Texas. Brown Co., Rocks, Blanket, 9 Aug 1877, Reverchon 755 (Holotype, GH!; isotypes, F! NY! US!). 18b. HEDEOMA REVERCHONI Gray var. SERPYLLIFOLIUM (Small) Irving, Britt. 22: 333-345. 1970. Hedeoma serpyllifolium Small, Bull. N.Y. Bot. Gard. 1: 287. 1896. deoma sanctum Small, Bull. N.Y. Bot. Gard. 1: 287. 1896. TYPE: TEXAS. Bexar Co., San Antonio, Apr 1853, Thurber s.n. (Holotype, NY!; isotype, GH!). Robust, suffruticose perennials, 15.0-40.0 cm high, the shoots of the previous season persistent. Shoots numerous, ascending or decumbent, branching freely along the lower 2/3 of the stems, the branches long and ascending, densely pubescent above with retrorsely curling hairs, puberulent below. Leaves dark green typically falling from the lower 1/2 of the shoots with maturity, ascending, thick membranous, elliptical-oblong, 2.24.0 mm (3.0 mm) wide, 6.0-10.0 mm (8.3 mm) long, margins entire, apex obtuse, base narrowly attenuated, petiolate or subsessile, glabrate above, densely pubescent below; nerves inconspicuous. Cymules usually 2—4-flowered, some- what crowded along the upper 2/3 of the stems, developing to one-side; peduncles 1.0-1.5 mm long; pedicels 2.0-3.0 mm long clothed in retrorsely curling puberulent hairs, bracteoles short. Calyx 5.0-6.0 mm (5.5 mm) long; the tube saccate below tapering distally and partially closing the tube, 260 3.0-4.0 mm (3.7 mm) long, densely hirsute pubescent with the hairs spread- ing or antrorsely curved; the upper calyx teeth slightly connate, subulate, ca 0.2 mm wide at the base, 1.0-1.2 mm long, erect or slightly laterally spreading, somewhat convergent with the lower and closing the orifice, hirsute-ciliate; the lower teeth subulate ca 0.2 mm wide at the base, 1.5—2.0 mm long; annulus very dense, ca 1.5 mm wide and seated ca 1.5 mm below the juncture of the upper and lower teeth. Corolla chiefly white, occasionally lavender, 8.0-10.0 mm (9.0 mm) long, sparsely pubescent in the throat, densely so below forming an annulus; the tube dilating upwardly to ca 2.0 mm wide; the upper lip ligulate ca 1.0 wide at the base, ca 1.5 mm long; the lower lip ca 3.2 mm wide, 3.2 mm long. Nutlets ca 0.5 mm wide, ca 1.2 mm long, areolate and glaucous. Chromosome number, 27 = 34. Distribution (Fig. 5). Abundant in open calcareous outcrops and road- cuts from the Balcones fault escarpment of central Texas to southeastern ou bees and southward into the south Texas Coastal Plain. Jun—Sep. TEXAS. Kerr Co., Kerrville, 1600-2000 ft, 25-30 Apr ee Helle We ( Holoty pes, NY! isotypes, GH! IA! MICH! MCS! ND NY! PH! SMU! UC! US!) 19. HEDEOMA MULTIFLORUM Benth., Lab., Gen. et Sp. 367. 1834. Hedeoma gilliesti Benth., Lab. Gen. et Sp., 367. 1834. TYPE: ARGENTINA. Cerro de Archiras ace _Mendosa, 1820, Gillies s.n. (Holotype, K!; isotype, LA!; phot types LA! NY! PH!). Micromeria ee, Fisch. et Meyer, Ind. Sem. Hort. Petrop. X: 56. 1845. not located. Satureia ites (Benth.) Brig., in Engler & Prantl. Nat. Pflanzanf. ed. 1, IV, 3a, 300. 1897. Based on Hedeoma gilliesii Benth. — bonariensis (Fisch. et Meyer) Briq., in Engler & Prantl. Nat. svn ed. IV, 3a, 300. 1897. ae on eee bonariensis Fisch. et Mey ie caespitose perennials forming tufts, 5.0—25.0 cm in height. Shoots numerous densely pubescent or puberulent with retrorsely curling hairs on the upper herbaceous portions of the stems, becoming glabrate below on the lower woody portions, branching mostly confined to the lower half, the branches widely ascending to nearly 45° angles. Leaves deciduous from the lower stem regions with maturity, spreading, menage: linear-lanceolate, usually five times longer than wide, 1.0-1.5 n (1.2 mm) wide, 5.5-8.5 mm (7.1 mm) long, margins entire, apex acutish, base narrowly attenuated and subsessile, upper and lower leaf surfaces and margins strigose; nerves indistinct. Axillary leaf buds develop on the immature stems yielding 2-3 eaf pairs. Cymules 1-5-flowered (commonly 1-2-flowered ), well-spaced along almost the entire length of the shoots; peduncles inconspicuous, less than 0.5 mm long, pedicels 3.0-4.0 mm long, retrosely puberulent; bracteoles lanceloate, typically shorter than the pedicels they subtend. Calyx 5.0-6.0 n (5.8 mm) long; the tube 3.9-4.5 mm (4.1 mm) long conspicuously saccate for ca 2/3 of its length, the distended region 1.2-1.5 mm wide, 261 the tube above the pouch constricted and tapered into a narrow upper neck, 0.8-1.0 mm wide, partially closing the calyx orifice; calyx sparsely and minutely hirtellous with antrorsely curling hairs; the upper calyx teeth connate for 1/3-1/2 of their length forming an upper lip, 0.8-1.2 mm long, pubescent on its inner face, the lobes narrowly triangular, spreading and slightly reflexed, ca 0.3 mm wide at the base, 0.4-0.8 mm _ long, hirtellous-ciliate; the lower teeth subulate, 1.5-2.0 mm long; annulus dense, ca 0.7 mm wide and seated ca 1.0 mm below the juncture of the upper and lower teeth, extended upwardly into the inner face of the upper lip, included. Corolla ca 8.2 mm long, glabrous within; the tube ca 7.0 mm long gradually dilated upwardly, ca 2.0 mm wide; the upper lip ligulate, emarginate, 1.0-1.2 mm wide, ca 2.0 mm long; the lower lip ca 3.5 mm wide, ca 3.0 mm long. Nutlets ca 0.7 mm wide, 1.4 mm long, the surface foveate, brown at maturity, strongly glaucous. Chromosome number, 27 = Distribution (Fig. 12). Rocky calcareous outcrops of rivers and roadcuts in southern Brazil, Uruguay, and northern Argentina. Nov—Feb. TYPE: BRAZIL. 1830, Sello 1062 (Holotype, B destroyed; lectotypes here designated K! isotypes LA!; photolectotypes, LA! NY! PH! ). Hedeoma multiflorum is most closely allied to H. drummondii of the Northern Hemisphere and differs from it primarily in the details of the calyx. 20. HEDEOMA DIFFUSUM Greene, Pittonia 3: 338. 1898. Hedeoma CE sal Greene, Pitt. 3: 339. 1898. TYPE: ARIZONA. Cocon Co. near Flagstaff, 29 Aug 1884, Jones 404 or 4104 (Holotype, US! ee BM! CAS! DS! GH! LA! MCS! NY! POM! UC! US!) Perennial herbs forming dense, circular, dwarfed mats with numerous decumbent shoots, 7.0—20.0 cm long; branching prolifically at the base and generally unbranched distally, stems puberulent above with retrorsely curl- ing or spreading hairs, pubescent with spreading hairs below. Leaves spread- ing, membranous, ovate, 2.0-455 mm (3.0 mm) wide, 3.5-0.0 mm (5.2 mm) long, margins entire, apex narrowly obtuse or acute, base rounded or attenuated and petiolate, petioles 5.0-1.0 mm long; the blade glabrate or puberulent above, densely hirtellous-puberulent below and on the margins; nervation inconspicuous. Cymules 1—3-flowered, occurring well-spaced along nearly the full length of the numerous shoots; peduncles 0.5—-1.0 mm long; pedicels 2.0-3.0 mm long puberulent with retrorsely curving hairs, bracteoles subulate, shorter than the pedicels they subtend. Calyx 5.0-6.0 mm (5.3 mm) long; the tube 3.0-4.0 mm (3.6 mm) long more or less tubular, slightly gibbous below, the distended portion ca 1.0 mm wide dilated up- wardly, hirtellous-hirsute with spreading hairs; the upper calyx pes united for ca 1/3 of their length or less, forming an upper lip 1.0-1.2 mm long, pubescent on its inner face, the lobes narrowly triangular, 06 1.3 mm long, laterally spreading and reflexed or erect, hirsute-ciliate; the lower teeth 262 subulate, recurved, 1.5—2.0 mm long, hirsute-ciliate; annulus dense, included or slightly exserted, 0.5-1.0 mm wide seated just below the juncture of the upper and lower teeth. Corolla ca. 9.5 mm long, blue, pubescent within on the throat, annulate below; the tube ca 7.5 mm long upwardly inflated, ca 2.5 mm wide above; the upper lip ligulate, subgaleate, and straight but with the margins and apex revolute, ca 2.3 mm wide, ca 2.0 mm long, emarginate, the lower lip ca 5.0 mm wide, ca 5.0 mm long. Nutlets ca 0.5 mm wide, 1.1 mm long, the surface areolate and strongly glaucous. Chromo- some number, 27 = 36. Distribution (Fig. 6). Pine forests in the vicinity of Flagstaff, Arizona; common on rocky outcrops and exposed areas, ca 7000 ft. Jul-Sep. TYPE: ARIZONA. Coconino Co. near Flagstaff, 1883. Rusby 786. (Holo- type, GH! isotypes, LA! MICH! NY! PH! UC! US!). Although I accept the name and type of H. diffusum designated by Greene, I cannot accept his morphological concept of the species. From his discussion Greene apparently had a broad concept of this taxon. he indicated, the range of the species was “from north to south through- out the state [Arizona}”. Thus, the distribution would extend into the range of H. nanum. H. diffusum, as it is realized here, is restricted to the Pinus ponderosa forest of the Flagstaff area. This distribution also concurs with that accorded H, diffusum by Epling and Stewart (1939). Hedeoma diffusum is most closely related to H. nanum. The ovate, entire, acute leaves, gibbous calyx, and the ad- and abaxially inflated corolla are all common morphological features. n 21. HEDEOMA NANUM (Torr.) Briq. Key to the varieties of Hedeoma nanum. 1. Coarse, sane Sipe with shoots few to numerous; leaves 5.0-16.0 mm wide, 8. ong; cymules 1—7-flowered i well-spaced along the axis; calyces ieee, 6.0-8.0 mm long, slightly gibbous or saccate below; plants of central and western Arizona. 21c. . Plants not a H.n. var. macrocalyx — abov 2. Caespitose ois als forming dense tufts with shoots extremely numerous, retrorsely puberulent-pubescent; leaves 2.1-4.5 mm _ wide S- n long; cymules 3—5-flowered, well-space along the axis; calyces 4.5-5.6 mm _ long, weakly saccate below; plants eee to southern-most Nevada and contiguous areas in California and Arizon lb. H.n. var. californicum 2. Annual or perennial herbs of v various habits, shoots solitary to numerous, pube- scence spreading or seeps ae pilose or hirsute; leaves 2.6—-6.0 wide, 5.5-13.0 mm long, cymules 3—15-flowered (tpically 8-10) well- Been or congested, forming a dense ‘ aie ; calyces 4.0-5.2 m oo distinctly saccate; plants widespread—Arizona, New Mexico, Texas and Mexi ve Hin. var. nanum 21a. HEDEOMA NANUM (Torr.) Brig. var. NANUM, in Engler & Prantl, Die Naturlichen Pfl. EX, 3a, 294. 1897. Hedeoma dentatum var. nanum Torr., Bot. Mex. Bound. 130. 1859. 263 6. Distribution of Hedeoma diffusum (triangle), H. nanum va Fig. eae | ake H. (squ nanun. n. var. macrocalyx (open circles), and H. mn. vat. ie nin 264 Hedeoma sche ee Gray, Syn. Fl. N. Am. 2: 368. 1878. Based on H. dentatum var. nanum Hedeoma nanum ae Pittonia 3: 339. 1898. Based on H. dentatum var. nanum sacl nanum (Torr.) Brig. var. typicum pee ae Spec. Nov. Beihefte 115: . 1939. Based on H. dentatum var. nanum Tor cea var. mexicanum aan Rep. Spec. nov., Beihefte O LEON. Calcareous hills near Monterrey, 7 Apr CAS! F! GH! K! LA! MICH! Hon nanum (Torr.) 15: 29. 1939. TYPE: Nu i508: Pringle 10203 (Holotype, US! isotypes, MSC! NY! PH! SI! SMU! UC! phototype, LA!) Plants annual or perennial herbs, 5.0-35.0 cm (14.1 cm) high. Shoots solitary to numerous, erect, ascending, or shortly decumbent up to 5.0 cm; branching primary at or near the base, typically unbranched above; shortly pilose or hirsute above, puberulent below, the hairs spreading or retrorsely curling. Leaves ascending to cernuous, ovate to oval, variable in dimension, 2.5-6.0 mm (2.9 mm) wide, 5.5— n (7.4 mm) long, margins entire occasionally obscurely crenate, apex a acute to narrowly obtuse, base broadly rounded to narrowly attenuated, petiolate, the petiole 0.5-2.5 mm long; the leaf surface glabrous to sparsely pubescent above, margin and undersurface hirtellous pubescent; nerves inconspicuous. Cymules 3-15- flowered (usually between 8 and 10), well-spaced along shoots, or com- monly very congested forming a dense but interruped spike of cymes; pe- duncles inconspicuous to 3.0 mm long; pedicels long, typically exceeding the length of the calyx tube, 3.0-6.0 mm long, clothed in hirsute-hirtellous spreading hairs; bracteoles a shorter than the nae they subtend. Calyx 4.0-5.2 mm (4.9 m long; the tube 2.5—3.7 1 (3.3 mm) long, saccate or occasionaly eres below for 1/2 to 2/3 bE — tube's length, the distended region typically more than 1/2 the tube’s length, 1.0-1.5 mm wide; the calyx tube densely hirsute to hirtellous with the hairs spread- ing; the upper calyx teeth connate for slightly less than 1/2 their length forming a conspicuous upper lip, 1.0-1.3 mm long, the lobes narrowly triangular-lanceolate, occasionally approaching subulate, ca 0.3 mm wide at the base, 0.4—0.7 mm long, laterally spreading, and reflexed (commonly at nearly right angles to the axis), hirtellous-ciliate, the inner face of the lip pubescent; the lower teeth subulate, recurved and not conspicuously exceed- ing the upper lip in length, 1.2-2.0 mm long, hirsute or hirtellous-ciliate; annulus dense, included, ca 0.7 mm wide, seated ca 0.8 mm below the junc- ture of the upper and lower calyx teeth, and continued upwardly into the inner face of the lower lip. Corolla 8.0-9.0 mm (8.9 mm) long, pubescent within the throat in two decurrent lines, the tube 6.0—7.0 mm long, strongly inflated, ca 3.0 mm wide above; the upper lip subgaleate, the margins and apex revolute, ca 2.0 mm wide, ca 2.0 mm long; the lower lip ca 4.0 mm wide, ca 3.5 mm long. Nutlets oblong, ca 0.5 mm wide, ca 1.1 mm long, areolate and weakly glaucous. Chromosome number, 2” = 36. Distribution (Fig. 6). Common on rocky, usually calcareous bluffs, ridges, — 265 roadcuts and streamways. Southern Nevada and adjacent California, southeast to southeastern Arizona, east to Trans-Pecos Texas and south to San Luis Potosi, Mexico. Apr—Jul. TYPE: TEXAS. El Paso Co., Rocky hills of the Rio Grande, near El Paso, 4 May 1852, Bigelow, sm. (Holotype, NY!; isotypes, DS! K! US!). Hedeoma n. var. nanum is a variable taxon which harbours a number of more or less distinct races. One such race received formal recognition as var. mexicanum by Stewart and Epling. It is an annual with bristly, spread- ing stem pubescence and a somewhat shortened calyx and corolla. Repre- sentative collections include, Whiting 950 (LA); and Lundell 5030 (LA, MICH). Another well-marked race occurs in northern Arizona. It is char- acterized by long, trailing stems (up to 40 cm long), and exceptionally long internodes, peduncles and pedicels (the peduncles up to 3.0 mm long, and the pedicels up to 6.0 mm long). The following are representative: Howell 26454 and 26409; Bailey s.n., Clover 5134, Wooton 308, and Whit- ing 1089. 21b. HEDEOMA NANUM (Torr.) Brig. var. CALIFORNICUM Stewart, Rep. Spec. Nov., Beihefte 115: 29. 1939. Dense, caespitose perennials forming tufts, 11.0-25.0 cm (17.1 cm) high. Shoots extremely numerous, ascending or shortly decumbent up to 4.0 cm long; branching almost exclusively at or near the base, densely pubescent- puberulent above with hairs retrorsely curling, puberulent below. Leaves small, spreading to cernuous, ovate, 2.0-4.5 mm (3.4 mm) wide, 3.4-8.5 mm (5.0 mm) long, margins entire or obscurely cremate, apex acute, base broadly to narrowly rounded, shortly petiolate, the petiole 0.5-1.5 mm long; leaf surfaces puberulent above, margins and undersurface hirtellous- pubescent; nerves inconspicuous. Cymules 3—5-flowered well-spaced along the upper 2/3-3/4 of the shoots; peduncles 1.0-2.7 mm_ long; pedicels 3.0-4.3 mm long; bracteoles subulate, shorter than the pedicels which they subtend. Calyx 4.5-5.6 mm (5.1 mm) long; the tube 3.0-4.0 mm (3.3 mm) long, weakly saccate below for ca 2/3 of the tube’s length, the dis- tended region 1.2-1.5 mm wide, hirsute-hirtellous to puberulent with the hairs spreading, the upper calyx teeth connate for slightly less than 2 of their length forming a conspicuous upper lip, 0.6-1.5 mm long, the lobes triangular, ca 0.4 mm wide at the base, 0.4-0.8 mm long, laterally spreading and slightly to sharply reflexed, hirtellous-ciliate, the inner face giabrate or pubescent; the lower teeth subulate, recurved, not conspicuously exceeding the lower in length, 1.5-2.0 mm (1.8 mm) long, hirsute-ciliate; annulus dense, included ca 0.7 mm wide and seated ca 0.7 mm below the juncture of the upper and lower teeth. Corolla as in H. nanum var. nanum ca 88 mm long. Nutlets 0.6 mm wide, 1.2 mm long, strongly glaucous. Chromo- some number, 27 = Distribution (Fig. 6). Frequent on rocky limestone outcrops in the 266 mountains of the Mojave Desert of California and contiguous areas of southern Nevada and northwestern Arizona. Apr—June. TYPE: NEVADA, Clark Co., Good Springs, 1 May 1905, Jones s.n. (Holo- type, LA!; isotypes, CAS! Ds! F! GH! POM! US!). Hedeoma #. vat. californicum is a relatively uniform taxon separated from the varieties nanwm and macrocalyx by its multi-stemmed, tufted habit, finely puberulent stems and microphyllous leaves. 2lc. HEDEOMA NANUM (Torr.) Brig. var. MACROCALYX Stewart, Rep. Spec. Nov., Beihefte 115: 29. 1939. Somewhat coarse, robust perennial herbs, 15.0-55.0 cm (27.0 cm) high. Shoots numerous, branches ascending at close angles, cansescent above, pub- erulent below, hairs retrosely curling. Leaves spreading or cernuous, ovate, oval, or elliptical, 5.0-16.0 mm (7.3 mm) wide, 8.0-25.0 mm (12.4 mm) long, margins obscurely crenate, apex broadly acute or obtuse, base broadly rounded to attenuated, petiole 1.5-5.0 mm in length; finely puberulent above, margins and undersurface hirtellous-puberulent, nerves inconspicuous. Cym- ules 5—7-flowered, well-spaced along the upper 2/3 of the shoots; peduncles 0.5-1.5 mm long; pedicels 3.0-5.0 mm long and typically exceeding the length of the calyx tube; bracteoles subulate usually shorter than the pedicels they subtend. Calyx 6.1-8.0 (6.8 mm) long; the tube 4.0-5.0 mm (4.4 mm) long, slightly gibbous or saccate below for ca 1/2 of the tube’s length, the distended region 1.3-1.8 mm wide, sparsely to densely hirsute-hirtellous with hairs spreading; the upper calyx teeth connate for slightly less than 1/2 of their length to form a long and conspicuous upper lip 1.5—2.7 mm (2.0 mm) long, lobes narrowly lanceolate, ca 0.3 mm wide, ca 0.8-1.2 mm long, hirsute-hirtellous-ciliate; the lower teeth subulate, recurved and not conspicuously exceeding the upper lip in length, 2.0-3.0 mm (2.5 mm) long, hirsute-ciliate; annulus dense, included, ca 0.7 mm wide and seated ca 0.8 mm below the junction of the upper and lower calyx teeth. Corolla as in H. n. var. nanum, 8.0-9.0 mm long. Nutlets ca 0.6 mm wide and 1.4 mm long, strongly glaucous. Chromosome number, 27 = 36. Distribution (Fig. 6). Rocky limestone outcrops and hills of arid central and southwestern Arizona. May—Aug. TYPE: ARIZONA. Pinal Co., Higley, Queens Creek, 19 May 1926, Peebles and Harrison 1916 (Holotype, LA! ). 22. HEDEOMA MICROPHYLLUM Irving, Brittonia 22: 338-340. 1970. Tufted perennials, 10.0-17.0 cm high. Shoots numerous, decumbent or ascending with branching more or less restricted to the base, puberulent or hirtellous with hairs retrorsely curving. Leaves numerous and crowded, ovate, 3.0-5.0 mm (4.0 mm) wide, 4.0-6.5 mm (5.0 mm) long, margins obscurely serrate, apex acute, base rounded or weakly cuneate, petiolate with a petiole ca 1.0 mm long; surfaces glabrous above, minutely puberulent 267 below and on the margins; nerves conspicuously elevated. Cymules 1-3- flowered (occasionally up to 5-flowered), crowded along the upper 1/2 of the stems; peduncles inconspicuous, less than 1.0 mm long; pedicels ca 3.0 mm long, clothed in hirtellous spreading hairs; bracteoles shorter than the pedicels they subtend. Calyx 5.5-6.5 mm (6.0 mm) long; the tube 4.0-4.5 mm (4.1 mm) long, saccate below but not markedly so for ca 1/2 of the tube’s length, the distended region ca 1.2 mm wide, moderately hirsute with spreading or slightly recurving hairs; the upper teeth connate for ca 1/3 of their length forming an upper lip, ca 1.5 mm long, the lobes reflexed and spreading, narrowly triangular, ca 0.4 mm wide at the base, ca 1.0 mm long, hirsute or hirtellous-ciliate; the lower teeth recurved, subulate, ca 0.5 mm wide at the base, 2.0 mm long, hirsute or hirtellous, ciliate; annulus moderately dense, included, ca 0.5 mm wide and seated at the juncture of the upper and lower teeth. Corolla blue, ca 8.5 mm long, glabrate within; the tube ca 7.0 mm long, slender, only slightly dilated upwardly. ca 0.5 mm wide; the upper lip ligulate, slightly concave, reflexed and emarginate, ca 1.0 mm wide, ca 1.5 mm long; the lower lip ca 2.5 mm wide, ca 3.0 mm long. Nutlets ca 0.5 mm wide, ca 1.0 mm long. Chromosome number, 2” = 36. Distribution (Fig. 7). Canyons near San Luis Potosi, Mexico, Jul-Sep. TYPE: SAN Luis Potosi. 10 km east of Guadalcazar, Aug 1966, Irving 693 (Holotype, TEX! ). 23. HEDEOMA MEDIUM Epl., Rep. Spec. Nov., Beihefte 115: 30. 1939. Caespitose perennials, 10.0-25.0 cm high. Shoots numerous, densely puberulent with retrorsely curling hairs on the upper herbaceous regions of the stems, glabrate below the woody portions; branching confined to the lower half of the plant with the branches ascending at close angles. Leaves deciduous from the lower stem regions with maturity, linear-elliptical, 1.5-5.0 mm (2.4 mm) wide, 6.5-17.0 mm (8.0 mm) long, margins entire, apex acutish, base tapered and subsessile, upper surface glabrate to puber- ulent, lower surface and margins puberulent; nervation indistinct with only the midrib visible. Axillary buds developed on the immature stems yielding 2-3 leaf pairs. Cymules 1—5-flowered, well-spaced along the upper 2/3 of the shoots, peduncles inconspicuous, less than 0.5 mm long, pedicels 4.0-5.0 mm long, retrorsely puberulent, bracteoles lanceolate, typically shorter than the pedicels they subtend. Calyx 4.2-5.0 mm (4.5 mm) long; the tube 2.5-3.0 mm (2.8 mm) long, conspicuously saccate for ca 2/3 of its length, the distended pouch ca 1.5 mm wide, constricted above; calyx tube glabrate- puberulent; the upper teeth connate for over 2/3 of their length to form a broad, upper lip, 1.0-1.2 mm long, pubescent on its inner face, the lobes deltoid, spreading and reflected, ca 0.3 mm wide at the base, ca 1.3 mm long, the margins glabrate or hirtellous-ciliate, the lower teeth subulate, recurved, 1.4—2.0 mm long, hirtellous or hirsute-ciliate; annulus dense, in- 268 cluded, ca 0.5 mm wide, seated approximately at the juncture of the upper and lower teeth, extended upwardly onto the inner face of the upper lip. Corolla ca 5.0 mm long glabrous within; the tube ca 4.5 mm long scarcely dilated upwardly, ca 1.2 mm wide; the upper lip ligulate, emarginate, straight, ca 1.3 mm wide, ca 1.5 mm long; the lower lip ca 3.0 mm wide, ca 2.0 mm long. Nutlets ca 0.5 mm wide, 1.0 mm long, the surface foveate, and glaucous. Chromosome number, unknown. Distribution (Fig. 12). Prairies and pasture lands of Uruguay and Argen- tina. Nov—F TYPE: UruGuay. Sorino, 1 Dec 1935, Gallinal sn. (Holotypes, LA!). Epling and Stewart comment that H. medium is ‘a species of uncertain quantity and disposition.” I, too, am in a quandry over its natural relation- ships and have placed it only tentatively with H. nanwm and its allies. The leaf morphology and foveate nutlet surface are very similar to those features of H. multiflorum and in habitat and general habit it resembles H. hispidum. 24. HEDEOMA COSTATUM Gray. A number of species have been created from the variable Hedeoma costatum complex. Most of the species stem from an attempt, by many workers, to formalize the marked differences observed in morphology between mesophytic and xerophytic representatives of this complex. costatum, H. tenellum, H. quinquenervatum, H. pringeli, and H. ee are all names accorded by various authors to the mesophytes. H. pulchellum, H. albescentifolium and H. convisae are names which were applied to the xerophytes. The taxonomically significant variation between these two basic groups resides primarily in length of bracteoles, degree of leaf denta- tion, and overall pubescence pattern. However, after careful analysis of herbarium material, I have been unable to recognize these basic morpho- logical groups at the specific level. Instead, I have maintained the complex as a single species separated into two varieties: var. costatum, representing the mesophytic populations; var. pulchellum, the xerophytic populations. Representatives of the southern most populations of H. costatum (Chia- pas, Mexico) were tetraploid in chromosome number (22 = ; ese plants, however, displayed no discernable morphological differences accom- panying the increase in the ploidy level. Key to the varieties of Hedeoma costatum. . Plants ane pubescent but not canescent with hirsute-villous hairs; leaves with margins dentate, often obscurely so; bracteoles narrowly tria poe usually ap ter than the pie corolla long exserted; plants of central southern Mexic are 24a. He var. pera ; Plants canescent throughout with stiff hirsute- villous hairs; leaves with margins coarsely dentate; bracteoles subulate, usually conspicuously longer than the eer sa corolla short Lacely exserted from the a t long exserted; plants of norther Mexico and southwestern United States... 24b. H.c. var. ealeh allan — —y 269 ig. 7. Distribution of Hedeoma costatum var. costatum (closed circles ) , H. cos- tatum vat. ee oe circles), H. microphyllum (closcd triangles), H. pilica- tum (closed squares), H. tenuipes (open squares), H. guercetorum (open Meaneles). and H. ee (asterisk). 270 24a. HEDEOMA COSTATUM Gray var. COSTATUM, Syn. Fl. 2: 363. 1878. Hedeoma costatum Hemsl., Biol. Centr. Am. Bot. 2: 547. 1881. TYPE: CHIAPAS. 1864-1870. Ghieshrecht 815. (Holotype, K!; isotypes, BM! GH! NY!; Cenaon GH! LA! 1). Hedeoma tenellum Hemsl., Biol. Centr. Am. Bot. 2: 549. 18 NUEVO LEON. Guajuco, Feb-Oct 1890, Palmer ae (Holotype, K!; isotypes, GH! PH! US!; phototype, LA!). Hedeoma pringlei Briq., Ann. Conser. et Jardin Bot. Geneve 2: 182. 1898. EVO LEON. Wooded slopes of Sierra Madre near Monterrey, 13 Jul 1888, Pringle 1911 (Holotype G!; isotypes, BM! GH! K! LA! MSC! ND! NY! PH! POM! UC! US!; phototype, LA!). Hedeoma permixtum Briq., Ann. Conserv. et Jardin Bot. Geneve 2: 183. 1898. Based on H. pringlei. Hedeoma quinquenervatum Bartlett, Proc. Am. Acad. 44: 634. 1908. TYPE: NUEVO adre above Monterrey, 3000 ft, 9 Apr 1906, Pringle 10241 (Holotype GH!; isotypes, CAS! F! IA! LA! MICH! MSC! NMC! SMU! TEX! C! US!). Extremely variable, loose or tufted perennials, 4.0-36.0 cm (15.0 cm) high. Shoots few to numerous, ascending or shortly decumbent, branching copiously in the lower 1/4 of the plant, the upper branches sparse and short; moderately to densely hirsute in the upper stem regions, the hairs spreading or antrorsely curving, puberulent below. Leaves spreading, mem- branous, ovate, oval, or elliptical, 3.0-13.0 mm (6.0 mm) wide, 5.0—28.0 mm (10.0 mm) long, margins strongly to obscurely serrate-dentate, apex acute, base broadly rounded or attenuated, the petiole 1.0-4.0 mm long; glabrate or pubescent below with the margin ciliate; nerves not conspicu- ously elevated. Cymules 3-7-flowered (typically 3-5), well-spaced along the upper 1/2 of the stems or becoming more or less congested toward the apex; peduncles 1.0—3.0 mm long; pedicels 4.0-5.0 mm long, hirtellous with the hairs spreading or retrorsely curving; bracteoles triangular, shorter than the pedicels they subtend. Calyx 5.0-10.0 mm (6.4 mm) long; the tube 4.0-7.0 mm (5.0 mm) long, slightly gibbous for 1/3-1/2 of its length, the distended region, ca 1.1 mm wide, moderately or densely hirsute or hirtellous with the hairs spreading and/or antrorsely curling; the upper calyx teeth connate for 1/4-1/3 of their length forming an upper lip, the lobes narrowly triangular to subulate, 0.2-0.5 mm wide at the base, 1.5- 2.5 mm long, laterally spreading and reflexed, hirsute or hirtellous-ciliate; the lower teeth recurved, subulate, ca 0.5 mm wide at the base, 2.0—3.0 mm long, hirsute or hirtellous-ciliate; annulus moderately dense, included, ca 0.5 mm wide and seated at the juncture of the upper and lower calyx teeth. Corolla, well-exserted from the calyx, 9.0-20.0 mm (12.0 mm) _ long, pubescent in the throat and the tube but not forming a well-defined annulus; the tube 8.5-18.5 mm (11.5 mm) long, slightly dilated upwardly, ca 2.0 mm wide; the upper lip flat, ligulate, slightly reflexed upwardly and emar- ginate, ca 1.5 mm wide, 1.5 mm long; the lower lip 3.0-4.0 mm wide, 2.0-3.0 mm long. Nutlets 0.6 mm wide, ca 1.2 mm long, the surface eal minutely areolate, weakly glaucous. Chromosome numbers, 27 = 36, 72. Distribution (Fi oy Pine Oak Forest of the Sierra Madre Oriental, Nuevo Leon to Hidalgo and Chiapas. 8000-10,000 ft. Jun—Sept. TYPE: CHIAPAS. 1864-1870, Ghiesbreght 815 (Holotype, GH!; isotypes, BM! K! LA! NY!; phototypes, GH! LA! US!). 24b, HEDEOMA COSTATUM Gray var. PULCHELLUM (Greene) Irving, Brit- tonia 22: 345. 1970 Hedeoma pulchellum Greene, Leaflets Bot. Obs. 1: 213. 1906. Basionym. ee ma albescentifoliunm Bartlett, Proc. Am. Acad. 44: 633. 1908. TYPE: CHI- HUA. Santa Eulalia Mountains, Apr 1885, Pringle 133 (Holotype, GH!; aoe F! NY! POM! PH! UC! US!). Hedeoma convisae A. Nels., Am. Bot. 23: 270. 1936. TYPE: NEW MEXICO. Eddy Co., Carlsbad Caverns, chaparral, Apr 1930, Convise 6 (Holotype, RM!; isotype, LA!). Densely tufted perennials up to 15.0 cm high. Shoots numerous, ascend- ing, or shortly decumbent; branching in the lower half; canescent with retrorsely curling hirsute hairs. Leaves spreading, membranous, ovate or elliptical, 3.0-6.5 mm (48 mm) wide, 5.0-10.0 mm (7.8 mm) long, margins coarsely serrate, apex sharply acute, base rounded or shortly cuneate with a petiole ca 1.0 mm long; puberulent to nearly villous above, strigose to villous below and on the margins, nerves not conspicuously raised. Cymules 3—5-flowered, typically congested in the upper 1/2 of the stems; peduncles 1.0-2.0 mm long; pedicels ca 4.0 mm long, hoary-villous with the hairs spreading or pubescent with the hairs retrorsely curling; bracteoles subulate usually conspicuously longer than the pedicels they subtend. Calyx 6.5-8.0 mm (7.2 mm) long; the tube 4.5-5.5 mm (5.0 mm) long, saccate- gibbous for ca 1/2 of its length, the distended region ca 1.3 mm wide, typically hoary-villous, occasionally pubescent, hairs spreading or antrorsely curving; the upper teeth connate for ca 1/4 of their length forming an upper lip 2.0-3.0 mm long, the lobes slightly reflexed and laterally spreading, subulate, ca 0.3 mm wide at the base, 2.0 mm long, hirsute-ciliate; the lower teeth recurved, subulate, ca 0.4 mm wide at the base, 2.0-3.0 mm long, hirsute-ciliate; annulus dense, included, ca 0.5 mm wide and seated at the juncture of the upper and lower teeth. Corolla dimorphic: in the first form white, slender and about equal to the length of the calyx, the lobes minute; in the second form blue, long and well-exserted from the calyx tube, 10.0-20.0 mm (14.0 mm) long, pubescent in the throat and tube but not annulate; the tube 8.5-18.5 mm (10.5 mm) long, slightly dilated upwardly ca 1.5 mm wide; the upper lip ligulate, slightly reflexed and emarginate ca 1.5 mm wide, ca 1.5 mm long; the lower lip 3.0-4.0 mm wide, 2.0-3.0 mm long. Nutlets ca 0.6 mm wide and 1.2 mm long, the surface minutely areolate, weakly glaucous. Chromosome number 2” = 36. Distribution (Fig. 7). Arid calcareous mountain regions of northern Mexico, western Texas and southern New Mexico. 3000-7000 ft. May—Jul. zie TYPE: New Mexico. Sierra Co., Kingston, 6600 ft, 19 May 1905, Metcalf 1599 (Holotype, ND!; isotype, NMC!; phototype, LA! ). 25. HEDEOMA JOHNSTONII Irving, SIDA 7(2): 103-106. 1977. Wiry perennial herbs up to 30.0 cm tall; shoots ascending or decumbent, numerous, branching primarily from the base and rooting at the lower nodes, glabrous except for a minute tuft or retrorsely curling hairs at each node. Leaves spreading or ascending, stiff membranous in texture, glabrous, conspicuously glandular punctate on the lower surface, ovate-oblanceolate, 0.0-20.0 mm long, 4.0-7.0 mm wide, dentate with 4-5 pairs of teeth terminating the costae, apex acute, base attenuate to a short petiole ca 1.5 mm long, midrib and secondaries conspicuously elevated. Cymules 1—3- flowered, well-spaced along the upper one-third of the stem; primary peduncles ca. 1.0 mm long; primary pedicels ca. 3.0 mm long; bracteoles lanceolate, equal to or slightly longer than the pedicels. Calyx 10.0-15.0 mm long (base to tip of upper teeth), tubular-funnelform, not gibbous, sparsely hirsute; upper teeth partially united forming a small but conspicu- ously reflected lip, the lobes triangular ca 1.0 mm long, sparsely ciliate; lower teeth deltoid below ca 1.0 mm wide, abruptly tapering to an aristate apex, ca 1.0-2.0 mm long, hirsute-ciliate, annulus located just below the teeth, included. Corolla showy, long and slender, up to 45 mm long from its base to the tip of the upper lip, pink-violet, broadly annulate within where seated in the calyx; upper lip short, ca 4.0 mm long, 3.0 mm wide, emarginate; the lower lip spreading, ca 5.0 mm long, ca 3.0 mm wide, Nutlets 2.0 mm long, 1.0 mm wide. Chromosome number unknown. Distribution (Fig. 7). Canyons and igneous slopes of the Sierra Del Jardin, Coahuila, Mexico. 1500-2250 m. Jul-Sep. TYPE: COAHUILA: Canyon Hundido on north side of Pico de Centinela, Sierra del Jardin, 8 km e. of Rancho El Jardin. “Steep canyon through 1 wee sierra, gravelly and sandy loam derived from extrusive igneous rocks, 15 2250 m.” 27 July 1973: M. C. Johnston, F. Chiang, ‘i L, Wendt and D. Riskind 11803. Hedeoma johnstonii is a distinctive species apparently endemic to the steep igneous slopes of the Sierra del Jardin, Coahuila, Mexico. Its habit 1s very reminiscent of Hedeoma bellum, an endemic of the mountains of the western coast of Mexico, but, in the details of its calyx and corolla is more closely allied to Hedeoma costatum and its allies. It is readily separated from all species of Hedeoma by its exceptionally long and showy corolla. 26. HEDEOMA PLICATUM Torr., Bot. U.S. Mex. Bound. Surv., part 2, 130. 1859 Perennial herbs, 15.0-37.0 cm high. Shoots typically numerous, ascend- ing or very shortly decumbent, copiously branched towards the base, sparsely so above with the branches short; hirtellous above with the hairs retrorsely 273 curving, puberulent below. Leaves spreading or ascending, membranous, elliptical or rhombic, 3.0-6.0 mm (4.4 mm) wide, 5.4—10.0 mm (7.9 mm) long, margins coarsely serrate, with the serrations typically confined to the upper 1/2 of the blade, apex acute, base broadly attenuated to a short petiole, ca 1.0 mm long; glabrous or puberulent above, puberulent or hirtellous below on the nerves and margins; nerves conspicuously elevated. Cymules 3-7- flowered (typically 3-5), well-spaced along the upper 2/3 of the stems; peduncles short ca 1.0 mm long; pedicels 3.0-4.0 mm_ long, hirtellous pubescent with retrorsely curling hairs; bracteoles short. Calyx 4.5-6.0 mm (5.9 mm) long the tube 3.4-4.2 mm (4.0 mm) long, pronouncedly gibbous for ca 1/2 of the tube’s length, the distended region ca 1.3 mm wide, the tube puberulent with short spreading hairs or with hairs antrorsely curling; the upper calyx teeth connate for 1/2 of their length forming an upper lip ca 3.0 mm long, the lobes narrowly triangular-lanceolate, ca 0.3 mm wide at the base, ca 1.4 mm long, laterally spreading and sharply reflexed, hirsute or hirtellous-ciliate; the lower calyx teeth subulate, recurved, ca 0.4 mm wide at the base, ca 2.0 mm long, hirsute-ciliate; the annulus moderately dense, included, ca 0.5 mm wide and seated at least 1.0 mm below the juncture of the upper and lower teeth. Corolla blue, 7.0-10.0 mm (9.0 mm) long, very sparsely pubescent in the throat in two decurrent lines, sparsely short pubescent in the tube; the tube 5.5-8.5 mm (7.5 mm) long, dilated upwardly, ca 2.5 mm wide above; the upper lip straight, concave and sub- galeate, slightly emarginate, ca 2.0 mm wide, ca 1.5 mm long, the lower lip 2.5-3.0 mm wide, 2.5-3.0 mm long. Nutlets ca 0.5 mm wide, 1.0 mm long, the surface minutely areolate, weakly glaucous. Chromosome number 2n = Distribution (Fig. 7). Pine-oak woodlands of the mountainous areas of southern New Mexico, western “trans-Pecos” Texas, and the northern Mexico, 5000-8000 . aes TYPE: AS. Jeff Davis Co., near Fort Davis in the Limpia creek area, dry ravines near the Limpia Mts, July 1849 Bigelow s.n. (Holotype, NY!; isotypes, K! US!) Hedeoma plicatum is a variable species which intergrades into H. costatum var. pulchellum. Hedeoma c. var. pulchellum is a xerophytic, spring flower- ing plant, whereas H. plicatwm is a mesophytic, late summer or fall flower- ing plant. Morphologically. two characters have emerged to distinguish the two species. The leaves o . plicatum are more or less rhombic with raised straight costae on the undersurface; the leaves of H.c. var. pulchellum, on the other hand, are ovate with the nerves unraised and usually curved and branched. The annulus of H. plicatuwm is seated at least 1.0 mm below the juncture of the upper and lower calyx while that of H.c. var. pulchellum is positioned directly at the juncture. 27. HEDEOMA TENUIFLORUM Brandegee, Zoé 5: 254. 1908. Robust perennials, 30.0-35.0 cm high with numerous shoots long and 274 ascending, bristly-hirsute pubescent with spreading hairs above, puberulent below, branching basally from a woody caudex. Internodes elongated, up to 5 cm long. Leaves spreading or ascending, membranous, elliptical-ovate to somewhat rhomboid, 3.0-4.0 mm (3.9 mm) wide, 9.0-13.5 mm (11.0 mm) long, margins entire, or obscurely crenate, apex acute, base narrowly attenuate petiolate, the petioles 2.0-4.0 mm long; margins and both sur- faces puberulent with short, stiff, spreading hairs; nervation indistinct, secondaries barely visible. Cymules 1-3-flowered, well-spaced along the upper 1/2 of the stems; peduncles ca 1.5 mm long; pedicels 3.54.0 mm long, densely hirsute with spreading hairs; bracteoles subulate shorter than the pedicels they subtend. Calyx ca 9.0 mm long; the tube ca 6.0 mm long, distinctly gibbous below, the distended region ca 1.0 mm wide, the tube dilated above, pubescent with short spreading hairs; the upper teeth con- nate for slightly less than 1/2 of their length forming a somewhat broad- ened, rounded upper lip ca 2.0 mm long, the margins more or less outwardly extended, the lobes reflexed, narrowly triangular-subulate, ca 0.3 mm wide at the base, ca 1.0 mm long, hirtellous-ciliate; the lower teeth triangular- subulate, slightly recurved, ca 0.4 mm wide at the base, ca 3.0 mm long, hirsute-ciliate; the annulus ca 0.7 mm wide but seated ca 1.0 mm below the juncture of the upper and lower teeth. Corolla 17.0-18.0 mm long, exserted, pubescent within but not annulate; the tube ca 15.0 mm long not dilated above. Nutlets ca 0.5 mm wide, ca 1.3 mm long, no mature nutlets seen. Chromosome number, unknown. sila Known only from the type locality. TYPE: BAJA CALIFORNIA. El Rancho Viejo, 30 Apr 1889, Brandegee SM. ease UC!; isotypes, PH! US! ). Several features of this rare species suggest an affinity with H. oblongi- folimm. However, until there is more material available for study H. tenwz- florum will be retained with H. costatum and its allies. 28. HEDEOMA MARTIRENSE Moran, Trans. San Diego Soc. Nat. History Te 2127s 1909. Loose, wiry perennial herbs, 8.0-10.0 cm high forming mats or clumps. Shoots numerous, strongly decumbent and forming adventitious roots from the lower nodes; branching copiously throughout, the branches ascending at close angles; densely hispid-hirtellous above with spreading or retrorsely curving hairs up to 0.5 mm long, glabrate below with the lower portions becoming somewhat woody. Shoots of the previous season persistent. Leaf pairs crowded with the internodes 2.0-5.0 mm long, usually confined to the upper 1/4 of the shoots. Leaves membranous, spreading, ovate 2.0-3.0 mm wide (2.7 mm), 5.0-6.0 mm long (5.4 mm), margins entire, apex obtuse, base petiolate, the petiole 1.0-1.5 mm long; leaf surfaces glabrous above and on the margins, strigulose below. Nerves raised. Flowers solitary in the upper leaf axils and restricted to the distil portions of the branches, pa) peduncles 2.0 mm long, pedicels 2.0-3.0 mm long, hirtellous with spread- ing hairs; bracteoles linear 1.0-2.0 mm long, shorter than the pedicels they subtend. Calyx 8.0-8.5 mm long, tubular, slightly sigmoid, saccate below for ca 1/3 of its length, ca 1.2 mm wide, flaring distally; the tube 5.5—6.5 mm long, weakly strigulose below the hairs spreading; the upper lip con- nate for ca 1/3 of its length, the lobes triangular ca 0.5 mm wide at the base, ca 1.5 mm long, slightly reflexed, ciliate; lower teeth 0.7 mm wide at the base, 2.5 mm long, triangular-lanceolate, ciliate; annulus ca 0.5 mm wide, weakly developed, seated at the juncture of the upper and lower lips, included. Corolla white or pinkish, long and slender, 20.0—22.0 mm long, glabrous within; the tube ca 19.0 mm long dilated above, ca 2.5-3.0 mm wide; the upper lip obcordate, slightly galeate, ca 2.0 mm wide, 2.5-3.0 mm long, shallowly notched; lower lip ca 4.0 mm wide, ca 3.0 mm long. Filaments quite short ca 3.0 mm long and arising just below the juncture of the upper and lower corolla lips (unique for Hedeoma). Nutlets brown, oblong, 0.5-0.7 mm wide, 1.4-1.6 mm long. Chromosome number, un- known. Distribution. North facing pine-fir covered slopes and on east facing open escarpments of N. Central Sierra San Pedro Martir, 7-9,000 ft. May—Aug. TYPE: BAJA CALIFORNIA. East rim of Sierra San Pedro Mare N. of Perba Buena, 1 June 1968, Moran 15069 (Holotype, SD 67322!; isotypes, K, MEXU, MONTU! UC, US). Hedeoma martirense is a rare and somewhat enigmatic taxon. In many of its morphic features, its large showy solitary flowers, its decumbent rhizo- matous habit, and its short leaf internodes, it shows strong phenetic affinities to H. diffuswm. In the remainder of its characters however, especially in its long, slender exserted corolla, and leaf structure it is most closely related to H. tenuiflorum, also endemic to Baja California. 29. HEDEOMA QUERCETORUM Epl., Rep. Spec. Nov., Beihefte 115: 41. 139. Lax perennial herbs, 15.0—30.0 cm high. Shoots few, ascending or shortly decumbent, branching basally with the branches widely ascending; puber- ulent-pubescent above with retrorsely curved hairs, glabrate below. Leaves spreading or ascending, membranous; the lower leaves ovate-elliptical, 4.0-5.0 mm (4.5 mm) wide, 10.0-14.0 mm (12.0 mm) long, the upper leaves lanceolate or narrowly elliptic, at least four times as long as wide, 2.0-3.0 mm (2.4 mm) wide, 12.0-30.0 (14.7 mm) long, margins entire or obscurely crenate, apex acute or narrowly obtuse, base narrowly attenuate and subsessile, the surfaces pubescent or glabrate above, sparsely pubescent below and on the margins; nerves more or less elevated. Cymules 1-—3- flowered, well-spaced along the upper 1/2 of the shoots, peduncles 0.5-1.0 mm long; pedicels 3.0-6.0 mm long, pubescent or puberulent; bracteoles linear-lanceolate. Calyx 7.0-8.2 (7.7 mm) long; the tube 4.0-5.5 mm (4.9 276 mm) long, tubular-sigmoid and slightly gibbous below, the distended region 1.3-1.5 mm wide, densely hirsute pubescent with spreading hairs; the upper calyx teeth united less than 1/5 of their length forming an upper lip 1.5-2.0 mm long, the lobes narrowly triangular, 0.3-0.4 mm wide at the base, 1.5-1.8 mm long, spreading, erect or slightly reflexed, hirsute- ciliate; the lower teeth triangular-subulate, recurved, ca 0.5 mm wide at the base, 2.7—3.0 mm long, hirsute-ciliate; the annulus sparse, included, ca 0.6 mm wide seated at the juncture of the upper and lower teeth. Corolla lavender, long exserted, ca 17.0 mm long, pubescent within the tube form- ing a broad annulus; the tube long and narrow, not at all dilated upwardly, 15.0 mm long, 1.2 wide above; the upper lip obcordate-ligulate, emar- ginate, ca 1.0 mm wide, ca 1.5 mm long; the lower lip ca 5.0 mm wide, ca 3.5 mm long. Nutlets oblong-ovate, ca 0.6 mm wide, ca 1.2 mm long, the surface areolate and glaucous. Chromosome number, unknown. Distribution (Fig. 7). Oak woodland associations of the Sierra Madre Oriental of eastern Mexico in the states of Nuevo Leon and Tamaulipas. 7000-9000 ft. Jul-Sep. TYPE: Nuevo LEON. Sierra Madre Oriental, San Francisco Canyon, ca 16 miles SW of Pueblo Galeana, 7500-8000 ft, 16 May 1934, C.H. & MT. Mueller 411 (Holotype, LA!; isotypes, GH! TEX! ). Hedeoma quercetorum is very closely related to tenwipes. The morpho- logical differences reside principally in the number of flowers per cymule, the length of the calyx, and the length of the corolla. The variation in these characters as displayed by these two species, however, is no greater than that displayed in the single species H. costatum, but, until further field studies are carried out, they will be treated as distinct. 30. HEDEOMAS TENUIPES Epl., Rep. Spec. Nov., Beihefte 115: 39. 1939. Lax perennials, 25.0—30.0 cm in height. Shoots strict or ascending, branch- ing primarily in the lower half with the branches widely ascending from the main axis; stems pubescent-puberulent with retrorsely curling hairs above, glabrate below. Leaves spreading, membranous; the lower leaves broadly ovate-elliptical 4.5-8.0 mm (6.1 mm) wide, 1.0-1.7 mm (1.3 mm) long; the upper leaves linear-lanceolate, at least five times longer than wide, 1.8-3.0 mm (2.2 mm) wide, 12.0-17.0 mm (15.3 mm) long, margins en- tire or obscurely crenate, apex acute, base narrowly attenuated and sub- sessile, surfaces glabrate above, pubescent below with short hairs on the nerves and margins; nerves slightly raised. Cymules 3- over 20-flowered, well-spaced along the upper 1/2 of the stems, occasionally congested, peduncles ca 1.0 mm long; pedicels 4.5-6.7 mm long, sparsely pubescent; bracteoles linear, nearly equal to the pedicels they subtend. Calyx 5.0-6.0 (5.6 mm) long; the tube 3.0-3.7 mm (3.6 mm) long, tubular and not at all gibbous below, or only slightly so, ca 1.0-1.2 mm wide, hirtellous pubescent with the hairs spreading or antrorsely curved; the upper calyx 277 teeth united less than 1/5 of their length forming an upper lip 1.0-1.5 mm long, the lobes narrowly triangular, ca 0.3 mm wide at the base, 0.9- 1.2 mm long, slightly reflexed, hirsute-ciliate; the lower teeth triangular- subulate, erect or slightly recurved, ca 0.4 mm wide at the base, 2.0-2.5 mm long, hirsute-ciliate; the annulus, included, sparse, ca 0.5 mm wide, seated at the juncture of the upper and lower teeth. Corolla small, white, barely exserted from the calyx tube, glabrous within, ca 5.5 mm_ long; narrowly tubular, ca 1.0 mm wide above; the upper lip ligulate, ca 1.0 mm wide, ca 0.8 mm long, emarginate; the lower lip ca 2.0 mm wide, 1.2 mm long. Nutlets ca 0.6 mm wide, ca 1.3 mm long, the surface areolate, glau- cous. Chromosome number 27 = 36. Distribution (Fig. 7). Oak woodlands of the Sierra Madre Oriental, Tamaulipas, Mexico. 2000-5000 ft. Jul-Sep. TYPE: TAMAULIPAS. “Cerro Zamora near El Milagro,” 25 Aug 1930, Bartlett 1117 (Holotype, F!; isotypes, MICH! LA!; phototype LA! ). 31. HEDEOMA HYSSOPIFOLIUM Gray, Proc. Am. Acad. L1: 96. 1876. Plants perennials forming dense, wiry tufts or large clumps, 15.0-47.0 cm (29.1 cm) high. Shoots numerous, ascending, or decumbent, rhizomatous, producing adventitious roots at the nodes of the short subterranean por- tions of the stems; the upper regions glabrate or pubescent, but usually puberulent with retrorsely curved hairs, puberulent or hirtellous below; branching almost exclusively confined to the basal regions. Leaf pairs more or less crowded, equal to or exceeding the internodes. Leaves glabrate membranous, narrowly lanceolate to linear, at least five times longer than wide, and commonly more, 1.0-2.0 mm (1.8 mm) wide, 8.0-18.5 mm (13.4 mm) long, margins entire, the apex narrowly acute, the base tapered and sessile; nerves prominent but not strongly elevated. Axillary buds form a cluster of 2-3 axillary leaf pairs. Cymules 1—7-flowered (typically 3- flowered), well-spaced or somewhat congested along the upper 1/2 of the shoots; peduncles 0.5—-3.0 mm long; pedicels 3.0-5.0 mm long, puberulent to pubescent; bracteoles variable, subulate-lanceolate, typically exceeding the length of the pedicels they subtend. Calyx long and slender, 6.0-8.3 mm (7.5 mm) long, the calyx tube 4.9-5.5 mm (4.2 mm) long only slightly or not at all gibbous below, tubular-funnelform, dilated above, 1.0-1.5 mm wide, typically glabrous or puberulent with short stiff spreading hairs in the proximal region, glabrous distally; the upper calyx teeth connate for ca 1/5 of their length forming an upper lip 1.0-1.6 mm long, the lobes narrowly triangular to subulate, ca 0.2-0.5 mm wide at the base, 0.9-1.6 mm long, erect and parallel or slightly laterally spreading, glabrous or tuberculate on the margins; the lower teeth long subulate, recurved, ca 0.2 mm wide at the base, 1.7—-3.0 mm long, hirsute or hirtellous-ciliate; annulus dense, 0.7-1.0 mm wide, exserted, seated approximately at the juncture of the upper and lower teeth and extended upwardly onto the in- 278 nerfaces of the teeth. Corolla lavender; 11.0-16.0 mm (14.1 mm) long, pubescent within the tube but not annulate; the tube dilating upwardly, but not abruptly so, 2.0-3.0 mm wide above; the upper lip broadly ligulate in Outline, emarginate, with margins outwardly flaring, 1.5-4.0 mm wide, 2.2-5.0 mm long; the lower lip up to 7.0 mm wide, 5.0 mm long. Nutlets ca 0.6 mm wide, ca 1.3 mm long, the surface areolate, weakly glaucous. Chromosome number, 27 = 36 Distribution (Fig. 8). A common member of the herbaceous understory flora of mature, pine stands in the mountains of southeastern Arizona, southwestern New Mexico and northern Mexico, lower montane to sub- alpine associations, 6000 to 10,000 ft. Most commonly a member of the upper montane association and found frequently in recently burned areas. Aug—Sep. TYPE: ARIZONA. Graham Co.: Mt. Graham, 9000 ft. Aug eee pie rock 418 (Holotype, GH!; Fragment, LA!; isotypes, F! K! US!; phototype LA! ). Hedeoma hyssopifolium is probably the most easily recognized taxon in the genus Hedeoma. This distinctiveness is due to its narrow, linear leaves, cylindrical, puberulent calyx, and large showy flowers. Although readily recognized, it nevertheless houses considerable variation. The most notable variation is revealed in the disjunct populations of northern Mexico (Fig. 8). These populations are easily recognized and perhaps deserve varietal status. Their glabrous stems, subulate-filiform calyx teeth, and sparce annulus, all combine to set these plants apart from their more northwestern counter- arts. Hedeoma hyssopifolium intergrades into H. oblongifolium and H. denta- tum, most often through hybridization at the lower altitudes where these species come is contact. 32. HEDEOMA OBLONGIFOLIUM (Gray) Heller Key to the varieties: 1. Corolla glabrous within, leaves broadly a plants of U.S. sapcates 2a. H. oe var. oblongifolium 1. Corolla annulate within, leaves eae elliptical, plants of Mexico. .....00..0......... 32b. H. ea var. mexicanum m 32a. HEDEOMA OBLONGIFOLIUM (Gray) Heller var. OBLONGIFOLIUM, Muhlenbergia 1: 1900. Hedeoma piperitum Benth. var. oblongifolium A. Gray, Proc. Am. Acad. 8: 366. 1872. Basion rc thymoides A. Gray var. oblongifolinm A. Gray, Syn. Fl. N. Am. 2: 362. Stout, herbaceous perennials, 20.0-55.0 cm high, forming robust tufts. Shoots few to many, usually under 10, densely pubescent above with retrorsely curling hairs to hoary-pilose above with long, spreading hairs, puberulent below, stems ascending strictly from a slender rootstock; branching con- fined to the basal regions with secondary branches few and arising well after the maturity of the primaries. Leaves spreading, membranous, nar- rowly elliptic-ovate or linear-elliptic, at least twice as long as wide, 5.5 mm (4.3 mm) wide, 8.5-14.0 mm (11.9 mm) long, margins denticulate, apex narrowly and sharply acute, base tapered and shortly petiolate with the petiole 0.9-1.5 mm long; the surfaces glabrate or pubescent above, undersurface and margins somewhat unevenly hirtellous; nerves more or less raised. Basal foliage distinct, broadly ovate, entire, glabrous above, minutely hirtellous below. Axillary buds arise early in the shoot develop- ment and form a cluster of 2—3 leaf pairs. Cymules 5—15-flowered, dense and congested along the upper 2/3 of the shoots; peduncles inconspicu- ous to 0.5 mm long; pedicels 4.0-7.0 mm long and commonly exceeding the length of the calyx tube, sparsely to densely hirtellous; bracteoles subulate-lanceolate ca equal to the length of the pedicels they subtend. Calyx 5.5-7.0 mm (6.2 mm) long; the tube 4.0-5.0 (4.5 mm) _ long, gibbous below for ca 1/3 of the tube’s length, the distended region 1.0-1.2 mm wide, hirtellous of short spreading hairs to densely hirsute- pilose; the upper calyx teeth connate for ca 1/5 of their length forming an upper lip 1.0-1.3 mm long, the lobes narrowly triangular, 0.1-0.3 mm wide at the base, 0.7-1.0 mm long, laterally spreading and reflexed, ciliate; the lower teeth subulate, recurved, ca 0.2 mm wide at the base, 1.3—2.0 mm long, hirsute-ciliate; the annulus dense somewhat exserted, ca 0.5 mm wide, seated approximately at the juncture of the upper and lower teeth. Corolla lavender, 10.0-11.0 mm (10.5 mm) long, glabrous within; the tube 8.0-9.0 mm (8.5 mm) long and only slightly dilated upwardly, ca 1.5 mm wide above; the upper lip, flat, ligulate, ca 1.2 mm wide, ca 2.0 mm long, emarginate, conspicuously shorter than the lower lip, and reflexed away from the axis; the lower lip ca 4.0 mm wide, ca 4.0 mm long. Nutlets ca 0.5 mm wide, ca 1.0 mm long, the surface areolate and weakly glaucous. Chromosome number, 27 = 36 Distribution (Fig. 8). Common on the slopes of the oak-woodland, chaparral associations in mountainous regions of central and southern Ari- zona, New Mexico and adjacent areas of northern Sonora, Mexico. 5000- 8000 fr. Aug—Sep. PE: New Mexico. 1851-52, Wright 1515 (Holotype, NY!; isotypes, BM! F! GH! K! PH! UC! US!) Hedeoma oblongifolium is more or less intermediate between the mesic montane species H. byssopifoliwm and the xerophytic H. dentatum. 32b. HEDEOMA OBLONGIFOLIUM Ao Heller var. MEXICANUM Irving, Brittonia 22: 342-344. 1970 Stout perennials 28.0-40.0 cm high, forming tufts. Shoots ascending, usually under 10, pubescence of tightly retrorsely curling hairs above, 280 puberulent below, branching mostly confined to the lower regions of the stem with the secondary branches long and ascending at close angles. Leaves spreading, thick membranous, narrowly elliptical or linear, at least three times as long as wide, 1.5-4.0 mm (2.8 mm) wide, 9.0-13.0 mm (10.8 Fig. Distribution of Hedeoma hyssopifolium (open squares), H. oblongifolinm var. ae (closed circles), H. oblongifolium var. mexicanum (open circles), and H. dentatum (open triangles). Zed mm) long, the margins denticulate with the teeth confined to the upper regions of the blade, the apex sharply acute, the base tapered and sub- sessile; surfaces glabrous to glabrate above, glabrate below; nerves elevated. Basal foliage distinct, narrowly elliptic to ovate, entire, and evenly glab- rate. Axillary buds arise early in the shoot development and form a cluster of 2-3 leaf pairs. Cymules 5-10-flowered, dense and congested, well-spaced or crowded along the upper 2/3 of the shoots; peduncles in- conspicuous; pedicels, 4.0-5.0 mm long, puberulent; bracteoles lanceolate, ca equal to the length of the pedicels they subtend. Calyx 5.2-5.8 mm (5.5 mm) long; the tube 3.7-4.2 mm (4.0 mm) long, gibbous below for ca 1/2 of the tube’s length, the distended region 1.0-1.2 mm wide, slightly tapered above, hirtellous with short, stiff, spreading hairs; the upper calyx teeth only slightly united for ca 1/5 or their length forming an upper lip 0.9-1.0 mm long, the lobes narrowly triangular, subulate, 0.1-0.2 mm wide at the base, 0.8-0.9 mm long, laterally spreading and reflexed; the lower teeth subulate, ca 0.3 mm wide at the base, 1.2-1.7 mm long, hirsute-ciliate; annulus dense partially exserted, ca 0.5 mm wide and seated approximately at the juncture of the upper and lower teeth. Corolla lavender, 8.0-9.0 mm long, annulate within of short pilose hairs seated at the juncture of the tube and throat; the tube 6.8-7.8 mm (7.0 mm) long, slightly dilated upwardly, ca 2.0 mm wide above; the upper lip ligulate, emarginate, ca 10 mm wide, 1.2 mm long, conspicuously shorter than the lower lip, and re- flexed away from the axis; the lower lip ca 3.0 mm wide, ca 3.0 mm long. Nutlets ca 0.5 mm wide ca 1.0 mm long, the surface areolate and glaucous. Chromosome number, unknown. Distribution (Fig. 8). Common in the oak-woodland and chaparral asso- ciations in the mountains of northern Sonora and Chihuahua, Mexico. 6000-9000 ft. Aug—Oct. TYPE: SONORA. Puerto del Cumarito, Sierra de la Cabellera, 5400 ft. 6 Oct 1941, White 4638 (Holtoype, MICH!; isotypes, GH! US.). 33. HEDEOMA DENTATUM Torr., Bot. U.S. Mex. Bound, Surv., part 2, 130. Tall herbaceous perennials, 29.0-40.0 cm tall, forming robust clumps. Shoots long, few to many, densely hirtellous pubescent with spreading or curling hairs; branching confined to the lower 1/2, the branches long, ascend- ing and arising well after the maturity of the primary shoots. Leaves spreading, ovate, elliptic, or rhombic, at least twice as long as wide, 3.0-5.0 mm (4.1 mm) wide, 8.0-17.0 mm (10.3 mm) long, margins coarsely serrate-dentate, apex sharply acute, base tapered and subsessile, upper sur- face sparsely strigulose to glabrate, under-surface and margin hirtellous with short, stiff, spreading hairs; nerves raised. Basal foliage broadly ovate or obovate, coarsely serrate with both surfaces densely and evenly hirsute-pilose. Axillary buds develop early in the shoot development and form a cluster of 282 2-3 leaf pairs. Cymules 5-15-flowered (typically between 5-7), dense and congested along the upper 1/2 of the shoots; peduncles 0.2-1.1 mm long; pedicels 3.2-4.5 mm long and commonly exceeding the length of the calyx tube, pubescent; bracteoles lanceolate equal in length to the pedicels they subtend. Calyx 4.8-5.5 mm (5.2 mm) long; the tube 3440 mm (3.8 mm) long, gibbous or slightly saccate for 1/3-1/2 of the tube’s length, 0.9-1.2 mm wide below, slightly tapered above, hirtellous with short, stiff, spreading hairs; the upper teeth slightly connate for ca 1/5 of their length forming an upper lip 1.0-1.6 mm long, the lobes subulate, ca 0.2 mm wide at the base, 0.7-0.9 mm long, laterally spreading and reflexed, hirtellous- ciliate; the lower teeth subulate, recurved ca 1.5-1.7 long, hirsute-ciltate; annulus dense somewhat exserted, ca 0.5 wide, seated approximately at the juncture of the upper and lower teeth. Corolla lavender, 8.0-9.0 mm (8.5 mm) long, annulate within with short pilose hairs; the tube 6.5-7.5 mm (7.4 mm) long, only slightly dilated upwardly, ca 1.5 mm wide above; the upper lip flat, ligulate, ca 1.2 mm wide, 1.5 mm long, emarginate, conspicuously shorter than the lower lip, and reflexed away from the axis; the lower lip ca 3.5 mm wide, ca 3.5 mm long. Nutlets ca 0.5 mm wide, 1.0 mm long, the surface areolate and glaucous. Chromosome number, 27 Distribution (Fig. 8). Common on open, well-drained soils in the oak woodland and chaparral associations in the mountains of central and south- eastern Arizona, and adjacent Mexico, SO00—10,000 ft. Aug—Oct. TYPE: SONORA. Santa Cruz, Sep 1851, Thurber s.n. (Holotype, NY!; isotypes, F! GH! US!). Hedeoma dentatum intergrades with both H. hyssopifolium and H. ob- longifolium. Hybridization probably accounts for the latter as there is ac least one documented case of gene exchange between H. dentatum and H. hyssopifolium in the Huachuca mountains. Epling and Stewart (1939) have postulated that H. dentatum is the result of past hybridization between H. oblongifolium and H. plicatum. However, it seems more likely that H. dentatum was selected out of peripheral populations of H. oblongifolium in response to xeric conditions. 34, HEDEOMA ACINOIDES Scheele, Linnaea 22: 592. 1849. Delicate, ephemeral annuals, 5.0—30.0 cm (15.2 cm) high, arising from a very slender taproot. Shoots solitary to numerous, erect, ascending or very shortly decumbent for 1.0-2.0 cm; branching absent or restricted to the lower 1/2 of the shoots with the branches ascending from the main axis at wide angles; densely pubescent above, nearly lanate with the hairs strongly retrorsely curving or curling, puberulent below. Leaves spreading or cernu- ous, membranous, narrowly or broadly ovate, elliptical, or rhombic-elliptical, 3.5-8.0 mm (5.8 mm) wide, 9.0-25.0 mm (14.2 mm) long, margins en- tire or obscurely serrate with the serrations restricted to the distil 1/2 of 283 the blade, apex obtuse, base rounded to narrowly attenuate, petiolate with the petiole 1.0-4.0 mm (2.6 mm) long; the surfaces glabrate or minutely strigulose on nerves and margins; nerves except for the midrib unelevated. Cymules 3—15-flowered and typically congested toward the apices forming a short dense spike of cymules; peduncles inconspicuous, less than 1.0 mm long; pedicels 2.0-5.0 mm long, shortly pubescent; bracteoles lanceolate- subulate, ciliate, the primary pair equal in length to the pedicels they sub- tend. Calyx 5.0-6.1 mm (5.3 mm) long, thin membranous; the tube rela- tively long and slender, 4.0-4.55 mm (4.1 mm) long, the lower 1/3 distinctly saccate below, the distended region ca 1.2 mm wide, the tube giabrate or sparsely hirtellous on the lower 1/2 and with the hairs confined to the primary nerves; upper calyx teeth united for ca 1/3 of their length forming small but definite upper lip, 0.9-1.2 mm long, the lobes broadly to narrowly triangular, ca 0.2 mm wide at the base, 0.8-1.1 mm long, parallel or only slightly reflexed, densely hirtellous-ciliate; the lower teeth narrowly triangular to subuate, recurved, ca 0.4 mm wide at the base, 1.0-1.2 mm long, hirtellous-ciliate; the upper and lower teeth partially closing the orifice through convergence; annulus very weakly represented, consisting of only a few short hairs, ca 0.4 mm wide, seated ca 0.4 mm below the junc- tion of the upper and lower teeth. Corolla pink, 9.0-13.0 mm (11.0 mm) long, the throat very slightly dilated upwardly ca 0.6 mm wide above; the upper lip ligulate and reflexed at right angles to the corolla tube, very slightly emarginate, 1.0-1.5 mm wide; 2.0-4.0 mm long; the lower lip 4.0-5.0 mm wide, 3.0-4.0 mm long. Nutlets small, 0.4 mm wide, ca 0.8 mm long, the surface areolate, not glaucous or only very slightly so. Chromosome number, 2n = 34 Distribution (Fig. 9). Common, in wet years, in the rocky limestone soil along the eastern periphery of the Edwards Plateau and Lampasas Cut Plain. It also occurs in limestone pockets of the South Coastal Plains of Texas and in the calcareous hills surrounding Monterrey, Mexico. Apr- May. TYPE: TExas. Comal Co., in the vicinity of New Braunfels, April, Roemer sn. (Holotype, not located ). Within this species a form can be recognized that is small, more heavily pigmented, glabrate and with entire leaves. Based on field observations in Travis Co., Texas, this form is apparently adapted to more xerophytic locations and often forms subpopulations several inches to several yards in, Size: In its interspecific relationships, H. acinoides is somewhat of an enigma. Its morphological attributes, in combination, are quite unique. Although Epling and Stewart (1939) have suggested an affinity with H. guercetorum, I can find no morphological or chemical substantiation for such an idea. In its overall habit and vegetative morphology, H. acimoides shows some arity with the eastern annual, H. pulegioides. The calyx, corolla, and =, simi 284 seed morphology, however, readily separate H. acinoides from this latter species. H. acinoides has the same chromosome number (2” = 34) as the par- tially sympatric taxa, H. reverchonii, Hr. var. serpyllifolium, H. drummondiu rT | | PP ty e Piet | r . e 2 82 e 4 ry e e * e e e e e : e ee ‘ ~ e le ry “ye > , seen) ° 8 © . e ae e e he \ e ; \ e \ ) \ ee . e \ 4 ; \ \ \ . eee WF : Fig. 9. Distribution of Hedeoma acinoides (closed circles). 285 (in part), and H. hispidum. As the latter 4 taxa are apparently closely related, it is quite possible that H. acinoides is a highly derived species stemming from the H. drummondii complex. Morphological support of this is seen in the convergence of the upper and lower calyx teeth, a feature also characteristic of H. drummondii and its allies. 35. HEDEOMA HISPIDUM Pursh, Fl. Am. Sept. 2: 414. 1814. Hedeoma hirtum Nutt., Gen. 1: 16. 1818. Nom. superfl. TYPE: ARKANSAS {Terr.?}. On open alluvions of the Missouri, June. Nwétall s.n. (Lectotype. here designated GH!; isotype, K!). Ziziphora hispidum (Pursh) R. & S., Syst. Mant. 1: 179. 1822. Based on H. hispidum. Hedeoma hispidum Pursh forma simplex Lalonde, Nat. Canad. 85: 71. 1958. TYPE: QUEBEC. Pontiac Co., Vinton, 27 Aug 1956, P. Lowis-Marie 5614 (Holotype, CAN; isotype, GH! ). Coarse annuals, 9.0-40.0 cm (21.6 mm) in height, arising from a very slender taproot. Shoots slender, sharply quadrangular, usually solitary, occa- sionally numerous, erect or shortly decumbent for 2.0-3.0 cm; branches absent or consisting of 2-6 pairs, long ascending and usually confined to the lower 1/3 of the plant; shoots densely pubescent above with relatively long, retrorsely curling hairs, becoming puberulent below. Leaves typically spread- ing, coriaceous with the margins frequently becoming revolute, linear or linear-elliptic, usually over 5 times longer than broad, 1.0-3.0 mm (2.2 mm) wide, 11.0-21.0 mm (16.4 mm) long, margins entire, apex obtuse, base narrowly attenuated and subsessile, surfaces glabrate above, strigulose below and on the margins; nerves not conspicuously elevated. Cymules born along the upper 2/3 of the stems, occasionally along its entire length, well-spaced or commonly congested forming a pronounced “spike”, 3-15-flowered but typically 5-8-flowered; peduncles inconspicuous less than 1.0 mm in length; pedicels 2.0-5.0 mm long, hirsute; bracteoles stiff and acute, narrowly lanceolate to subulate, hirsute-ciliate, longer than the pedicels they subtend. Calyx highly zygomorphic, 4.9-6.0 mm (5.2 mm) long; the tube 2.8-3.2 mm (2.9 mm) long, markedly saccate below for ca 3/4 of its length, 1.5-2.0 mm wide below, pubescent with hairs long hirsute and spreading; the upper calyx teeth connate for ca 1/2 of their length forming a broad, expanded upper lip, 1.2-2.0 mm long with margins outwardly flaring and exceeding the width of the tube below, the connate portion silky pubescent within on its inner face, the lobes narrowly triangular, laterally spreading, hirtellous-ciliate, ca 0.5 mm wide at the base, 0.5—0.9 mm long; the lower teeth recurved and subulate, 2.0-2.5 mm _ long, hirsute-hirtellous-ciltate; annulus dense, slightly exserted, ca 0.5 mm wide and seated at the juncture of the upper and lower teeth. Corolla dimorphic; the first form blue or white, small, disproportionately so in reference to the calyx, ca 5.3 mm long and scarcely exserted from the calyx tube, glabrous within. The second form blue, considerably larger, ca 6.0-7.0 mm long, and well exserted from 286 the calyx, glabrous within, the pe lip subgaleate, the lower spreading. utlets ca 0.6 mm wide, ca 1.2 mm long, the surface areolate and strongly glaucous. Chromosome number, on = =: Distribution (Fig. 10). Rich, loamy soil of prairie and pasture lands from Vermont west to Alberta, Canada, south to south-central Texas and eastern Florida, along the Gulf coastal plain. Apr—June PE: On the banks of the Missouri, New¢tall s.n. (Holotype, PH! ). Hedeoma ae is one of the most widespread of the taxa of Hedeoma, rivaled only by H. drummondii and H. puleg‘oides in total area of distri- bution, Because of its breeding behavior, its annual habit, and chromosome num- ber, H. hispidum is believed to be derived. Its morphology can be traced to two possible precursors—H. nanum ae H. drummondii, themselves closely related to one another. In past studies H. bispidum has always been associated with H. pulegioides to which it displays a close similarity in its overall morphology. However, differences in nutlet structure, fine details of the calyx, corolla, and chromo- some number clearly set H. hispidum apart from the latter (see discussion following treatment of H. pulegioides). IV. SUBGENUS HEDEOMA Loose, wiry perennials of robust annuals, 5.0-40.0 cm high. Shoots solitary to numerous, erect, ascending or shortly decumbent, arising from a slender taproot. Leaves well-spaced, membranous and spreading, ovate, or elliptical, 3.5-11.0 mm wide, 5.0-32.0 mm long, margins dentate or crenate, base petiolate, the leaf surfaces strigose or glabrous, glandular dotted. Flowers in 1-15-flowered axillary cymules. Calyces strongly zygomorphic, 0 mm long, the tube weakly to strongly saccate for over 1/2 of its length; the upper teeth united for 1/2 to 2/3 of their length to form a broad, ex- panded, upper lip, the lobes semi-ovate, deltoid, or triangular, erect or re- flexed; the lower teeth short scarcely exceeding the upper lip in length, ciliate. Corolla short, 3.0-8.5 mm long (in H. pulegioides and H. crenatum barely exserted from the calyx tube); the upper lip broadly ligulate, deeply emarginate (for ca 1/2 of its length); the lower lip not spreading, more or less held erect, nearly equally 3-lobed. Nutlets orbicular in outline or nearly so, black in color at maturity, the surface ruminate, not becoming mucilaginous when moistened, not glaucous. Chromosome number, 27 = e subgenus Hedeoma is composed o species which, despite their wide geographical separation, appear to form a natural unit. The salient morphological features which hold the group together are the highly zygomorphic calyx, small corolla, and orbicular, black, nonmucilaginous nutlets. Fig. 10. Distribution of Hedeoma hispidum. p 4s - NS Sill meee oN ecco By ‘ ° meee ' nN ‘ ‘ ‘ ane \ \ \ \ \ “Sox \ a ‘ ’ e ’ | . l e 1 | Bn-- | as | 4 nd “44 ‘ Py ia ee rh : | 7 J mbescost 2. ; a a ef ! ' , a) : ! : Faas : 1 a rr : al P ; ‘ ‘e ' ! : 1 ‘ ' yO i oe ' ‘ ' a4 ! ' : i : et ae “ i : : : ! Poe ahora 5 We : \ ; =: ' : bees Sout L : ee eee i cs H ss : wend ! ! ' 3 ‘ ; H ' t t Pann ' mee ! met <) “a 87 A 288 36. HEDEOMA PULEGIOIDES (L.) Pers., Syn. 113. 1807. Based on Melissa pulegioides. Melissa pulegioides L., Sp. Pl. ed 1,593. 1753. Cuntla pulegioides L., Sp. Pl. ed. 2, 30. 1772. Based on Melissa pulegioides. Ziziphora pulegivides (L.) Desf., Table. ed. I: 55. 1804; ed. II: 66. 1815. Based on Melissa pulegioides. Hedeoma Met gee (L.) Pers. forma simplex Lalonde, LaRevue D’Oka 32: 151. 958. TYPE: QUEBEC. Powe 10 Aug 1937. Desilets s.n. (Holotype, Robust annuals, 10.0-40.0 cm high, arising from a slender taproot. Shoots typically solitary, erect, straight, branching freely in the middle region with the branches long and ascending, densely puberulent with retrorsely curling hairs above, glabrate below. Leaves spreading or cernuous, deciduous from the lower 1/2 of the plant with maturity, membranous, narrowly to broadly elliptical or nearly oblong, 4.0-11.0 mm (5.9 mm) wide, 12.0-32.0 mm (19.0 mm) long, margins obscurely to sharply dentate, apex narrowly obtuse, base attenuated and petiolate, the petiole 2.5-7.0 mm long; leaf surfaces glabrous or minutely strigulose on the nerves; nervation not prominent. Cymules 3-15-flowered, typically well-spaced along the upper branches, occasionally congested; peduncles 0.5-1.0 mm long; pedicels 3.0-3.5 mm long, puberulent; bracteoles shorter than the pedicels. Calyx highly zygomorphic, 3.5-4.5 mm (4.1 mm) long; the tube 2.0-2.5 mm (2.2 mm) long, markedly saccate below for ca 2/3 of its length, the dis- tended region 1.3—1.6 mm wide, sparsely hirtellous or glabrate below with the hairs spreading; the upper calyx teeth connate for ca 2/3 of their length forming a broad, expanded, upper lip, 2.0-2.5 mm long with the margins rounded, and exceeding the width of the tube below, the connate portion of the lip glabrous on a inner face, the lobes narrowly triangular, ca 0.5 mm wide at the base, 0.5-0.7 mm long, laterally spreading and reflexed, glabrous on the eee the lower teeth short, barely exceeding the upper lip in length, subulate and recurved, ca 0.5 mm wide at the base, 1.5-2.5 mm long, hirsute-ciliate; the annulus sparse, included, ca 0.5 mm wide and seated at the juncture of the upper and lower lips. Corolla minute, barely exserted from the calyx tube, ca 4.0 mm long, glabrous within; the tube ca 3.0 mm long, dilated core, ca 1.5 mm wide above; the upper lip ligulate, ca 0.8 mm wide at the base, ca 1.0 mm long, straight, deeply emarginate; the lower lip erect, eee 3-lobed, ca 2.0 mm wide, 1.0 mm long; nutlets 0.9 mm wide, 1.0 mm long, black, the surface ruminate, not glaucous. Chromosome number 27 = 36. Distribution (Fig. 11). Open woodlands and waste ground throughout central and eastern U. S., eastern Kansas and Oklahoma eastward to South Carolina and northward to Nova Scotia, Canada. Aug—Sep. TYPE: Plants of the Gronovian Herbarium of which Clayton 514 has been suggested as the standard (Epling 1939) (Lectotype, here designated BM). bo Cc \o Hedeoma pulegioides is a very widespread and distinct species which has, traditionally, been grouped with H. hispidum. Indeed, these two species form- ed a separate section in the revision by Epling and Stewart (1939). Although there are several morphological features shared by these species, notably the annual habit and the saccate calyx, their differences, as for example the features of the nutlet, outweigh their similarities. Hedeoma pulegioides is most closely related to H. crenatum of Brazil. Only habit and leaf shape distinguish these two taxa from one another. coor —_— rf hie oh 4 D | l | el, % Dae LEE ILLS ' See ——TEeOeOOOOOOOOOF ol, EF EE = \ = ° git e al .* \ ie ®ee f= cet * ad --gi he o/* yoo Pia De tg” off Oo e 1 fleets. A ae e e ° e @iur fe ‘S Wb ® fe \e - -"@ = re ‘ e e e- ‘ ~ -_--- z 2 eo °? oie e e eo) 1-7 KS) e \ e@ , 7 awae Ce .? oe f 3 are re 2 ee ef) < . a . e —— e@ 1e ° ° e ‘eer d oe .- ye e 7 e -- a et” e eo = ae — aie = e \ ° » bs ayeectrr es an 1 ¢ e sNews? ra e eN\ee net LL ee le 386 : orev e/e ce Oe a Ae = 'o3 ee o-- Pe n-- ~@ 7 ee---7N e ‘ - i . yee ee ‘e 220, ~ RRP APL a ELLE hd e rT \ ee & ' e' A ! \ - t x ‘\)__---- fo ‘f U) ==, H ' ( 1 s lt a een? ~---=! lo -- . B kixe VA me a — I gpg i Ree ROAD TNE a a ARPA EAPE CD 2 agreement ate in Pie Mi MO GE PTE REIN RT ~ ARR, fon Ss, oe aL LCL PIRES I PAP AEA AL. = maa —o 12 Pa re Tl OCHQH PLIGG IE IO RGR ROS CLALIT \ ve 3 | PPR, st » SSIS SHS mh { i) , ) a Fig. 11. Distribution of Hedeoma pulegioides. 290 37. HEDEOMA CRENATUM Irving, Britt. 22: 344-345. 1970. Based on Pseudocunila montana Brade. Hedeoma Ee le Benth. var. montannm Dusén, Arch. Mus. Nac. XIII. 35. 903. TYPE: BRAZIL. Serra do Itatiaia, March 1894. Ule 199 (Holotype, R). Pseudocunila montana Brade, Rodriguesia 7: 27. 1944. Cunila montana Brade ex Epl., Britt. 7: 140. 1951. TYPE: BRAZIL. Rio de Janeiro, Serra dos Orgaos, Estado do Rio Castello, 2100 m, 17 Jan 1945. Segadas Vianna 675 (Holotype, LA!). Hedeoma montanum (Brade) Epl. & le Britt. 15: 369. 1963 (not Brandegee 1913). Based on Cunila montana Brade jon Loose, wiry perennials, perhaps biennials, 15.0-30.0 cm high. Shoots numerous, typically decumbent and rhizomatous, producing adventitious roots from the lower nodes, densely pubescent with spreading or retrorsely curving hairs, branching mostly at or near the base. Leaves spreading, occurring all along the stems, membranous, ovate-spatulate, 4.0-5.0 mm (4.6 mm) wide, 9.0-10.0 mm (9.5 mm) long, glabrous, margins crenate especially along the upper 1/2 of the blade, apex obtuse, base rounded and petiolate with the petiole 2.5-3.0 mm _ long; nervation inconspicuous. Cymules 1-3-flowered, well-spaced along the upper 1/2 of the stems; peduncles inconspicuous; pedicels ca 2.5 mm long and puberulent; bracteoles shorter than the pedicels they subtend. Calyx highly zygomorphic, ca 4.0 mm long; the tube ca 2.0 mm long, hirtellous, markedly saccate for over 3/4 of its length, the distended region ca 1.0 mm wide; the upper calyx ag connate for ca 1/2 of their length rn an expanded, upper lip, 0 mm long, with the margins rounded, the lobes semi-ovate, ca 0.7 mm wide at the base, ca 1.0 mm long, more or less erect, hirtellous-ciliate; the lower teeth short barely exceeding the upper lip in length, narrowly triangular, ca 0.5 mm wide at the base, ca 2.0 mm long, hirtellous-ciliate, erect or slightly recurved; annulus dense, included, ca 0.5 mm wide, seated just below the juncture of the upper and lower teeth. Corolla minute, barely exserted from the calyx tube, ca 3.0 mm long, pubescent within; the tube ca 2.0 mm long, slightly dilated upwardly, ca 1.0 mm wide at the summit; the upper lip broadly ligulate ca 1.5 mm wide, 1.0 mm long, straight, flat, and deeply emarginate; the lower lip erect, nearly equally 3-lobed, ca 2.0 mm wide, ca 1.0 mm long. Nutlets 0.8 mm wide, 1.0 mm long, the surface ruminate, and not glaucous. Chromosome number, unknown. Distribution (Fig. 12). Brazil, in the mountains of Minas Geraes and Rio de Janeiro. 6,000-7,000 ft. Nov—Jan. TYPE: BRAZIL. Rio de Janeiro, Serra dos Orgaos, 2200 m, A.C. Brade 11,500 (Holotype, RF! ) 38. HEDEOMA MANDONIANUM Wedd., Chlor. And. 2: 148. 1860. Hedeoma rev iflorum Griseb. in Lechler, Berb. Am. Aust. 58. 1857, nomen. TYPE: U gapata,”’ Jun 1854, Lechler 1848 (Holotype, K! Hedeoma adscendens Rusby, Phytologia 1: 74. 1934. TYPE: "BOL OVIA. Cordillera rap Al = a 6G on = | = e | a e a e e® e Ce of 7 Ad x = : y He 12. Distribution of Hedeoma ee (closed circles), H. medium (open crenatum (open square), mandonianum (closed square) and H. ne ee (asterisk), 292 Real, top of the pass on the Tipuani-Ancoma-Sorata Trail, down to Sorata, 13,000 ft, 20 Apr 1936, Tate 783 (Holotype, NY!; isotypes, GH! LA! PH! US!; phototype, LA!) Loose, densely branched perennials, 5.0-15.0 cm high. Shoots numerous, intertangled, ascending or shortly decumbent, rhizomatous, sending out ad- ventitious roots from the lower nodes, hirtellous above with short, stiff, spreading hairs, puberulent below. Leaves spreading or ascending, mem- branous, ovate to obovate 3.5-5.0 mm (4.6 mm) wide, 5.0-8.0 mm (6.4 mm) long, margins in the upper half of the blade crenate, apex acute, the base rounded or cuneate with a small petiole, less than 0.5 mm long; surfaces glabrate to strigose above, strigose below and on the margins; nervation distinct. Cymules 1—4-flowered, occuring along the upper 1/2 of the stems but somewhat congested near the branch apices, peduncles inconspicuous; pedicels 2.0-3.5 mm long hirtellous; bracteoles subulate shorter than the pedicels they subtend, ca 1.5 mm long. Calyx 4.0-5.0 mm (4.7 mm) long; the tube 2.5-3.2 mm (3.0 mm) long, gibbous or saccate below for ca 1/2 of its length, 1.1-1.5 mm wide, puberulent or hirtellous with fine spreading hairs; the upper calyx teeth connate for ca 1/2 of their length forming an upper lip 1.2-1.6 mm long, the lobes deltoid to Dearne erect or slightly reflexed, 0.3-0.5 mm wide at the base, 0.6-1.0 mm long, hirsute-ciliate; the lower teeth, recurved, subulate, scarcely eo the upper lip in length, ca 0.4 mm wide at the base, 1.3-2.0 mm long, hirsute-ciliate,; annulus in- cluded, very sparse, seated approximately at the juncture of the upper and ie! lips, ca 0.S mm wide. Corolla 6.0-8.5 mm long, glabrous within, the e 5.0-6.5 mm long gradually dilating upwardly, ca 1.5 mm wide at the summit; the upper lip obcordate, deeply emarginate, straight and flat, ca 2.0 mm wide, 1.0-1.5 mm long; the lower lip not spreading, 3.0-4.0 mm wide, 20-25 1 mm long, the central lobe broadened and well-exserted from the laterals. Nutlets 0.8 mm wide, ca 1.0 mm One black, the surface ruminate, not glaucous. Chromosome number, unknow Distribution (Fig. 12). Andes Mts. of ee Peru and northwestern Bolivia. 8000-14,000 ft. Jan—Jul. TYPE: BoLiviA. Province of La Paz, Cordillera de Sorata, 1858, Mandon 5.2. (Holotype, P; phototype, LA! ) Hedeoma mandonianum is most closely related to H. crenatum of southern Brazil. EXCLUDED NAMES ng arkansana Nutt., Trans. Am. Phil. Soc. N.S. 5: 186. 1837. = Satureja glabra ee angulata Tharp. = ee pea ies — ) Epl Hedeoma bicolor (Wats) Briq., Engler & Prantl, Nat. Pflanzent, ed. 1, IV, 3a, 294. 1897. = Poliomintha eeaions Gray. Hedeoma bracteolata Nutt., Gen. Am. 2: Add. 1818 = ? Conradina sp. (fide Epling). Hedeoma caerulescens Benth. D. C. Prodr. 12: 245. Nov. 1848 = Satureja caerulescens Epl. 293 yee a Benth. D. C. Prodr. 12: 245. Nov. 1848. = Rhododon ciliatus (Ben Hedcome coccinea (Benth.) Briq., Engler & Prantl, Nat. Pflanzenf., ed. 1, IV, 3a, 295. = Rhabdocaulon coccineus (Benth.) Epl. Hedeoma ee (Benth.) Briq., Engler & Prantl, Nat. Pflanzenf., ed. 1, IV, 3a, 295. 1897. = Rhabdocaulon cease (Benth.) Epl. Hedeoma glabra Nutt., Gen. G: 16138 ar ee glabra Fern. Bria. Engler & Prantl, Nat. Pflanzenf., ed. 1, IV, 3a ray. Hedeoma glabrum Pers., Syn. 2: 131. 1807. =? Satureja glabrella Brig. (fide Epl- ing). Hedeoma oe Briq., Ann. Conserv. et Jard. Bot. Geneve 2: 185. 1898. = Cunila galioides Benth. Hedeoma wee (Benth.) Brig., Engler & Prantl, Nat 295. 1897. = Rhabdocaulon gracilis ae Epl. Hedeoma gracilis var. eee (Benth. , IV, 3a, 295. . Pflanzenf., ed. 1, IV, 3a, , Engler & Prantl, Nat. Pflanzenf., ed. = Rhabdocaulon eras (Benth.) Epl. Peds: gracillima ean Contr. West. Bot. 15: 150. 1929. = Satureja brownet Briq. Hedeoma graveolens Chapman ex Gray, Proc. Am. Acad. 8: 1873. = Stachydeoma Sess ns Small. oma oe Briq., Ann. Conser. Jard. Bot. ae ‘flor Hedeoma ile Briq., Bull. Herb. Boiss Benth. Geneve 1. 1897. = Poliomintha . ser. 2, 7: 611. 1907. = Cunila spicata Be Hedeoma incana Torr., Bot. U.S. Mex. Bound. Surv. 130. 1859. = Poliomintha incana orr.) A. Gray. Hedeoma italtantae awra, Oestr. Bot. Zeitschr. 31: 70. 1881 myrtoides (St. Hill ex Benth.) Epl. se ie longiflora (Gary) Brig., Engler & oo Nat. Pflanzenf. ed. 1, IV, 3a 294, 7. = Poliomintha longiflora A. a the ee (Schauer) Briq., Engler ie Peat Nat. Pflanzenf., ed. 1, IV, 3a 294. = Hesperozygis marifolia (Schauer) E Hedeoma oe Regel, Gartenfl. 357. . = Hesperozygis 864, = = Mos ocymoides Hamilt. Lab. Gen. et Sp., 367. 1834. = Mosla ocymoides Hamilt. Briq., Engler & Prantl, Nat. Pflanzenf., ed. 1, IV, 3a. 295. = Hesperozygis nitda ae ) Ses pulegioides (L.) Pers. B-mexicanum, Lab. Gen. et Sp., 366. 1834. = As H. pulegioides is nareported from Mexico, Bentham’s §-mexicanum undoubt- edly refers to another species; however the type was not located. Hedeoma purpurea Kellogg. Proc. Calif. Acad. 5: 52. 1873. = Satureja sp. (fide Eplin 2): Hedeoma saris (Benth.) Brig., Engler & Prantl, Nat. Pflanzenf., ed. 1, IV, 3a. 294. 1897 = Eriothymus rubiacens (Benth.) Schmidt. Hedeoma schwackeana Glaziou, Bull. Soc. Bot. Fr. 48, Mem. III, 557. 1911 = Hesperozygis myrtoides (St. Hill ex Benth.) Epl Hedeoma scutellarioides (Benth.) Briq., Engler & a Nat. Pflanzenf., ed. 1, IV 3a. 294. 18 = Hoehnea ere ieee ee ade eee var. pilosa (Schmidt) Bri “Eng - & Prantl, Nat. Pflanzenf., IV, 4-295. 1897. = ee sntelleroides (Benth.) Epl. Hedeome sappuliciaes Torr., Pac. R.R. Rep 23. 1856. = Peeoayne serpylloides Gray. eee stenodonta Briq., Bull. Herb. Boiss., ser. II, 7, 612. 1907. = Rhabdocaulon stenondontus (Briq.) Epl. 294 Hedeoma subequale Epl., Bull. Torr. Bot. Club 74: 513. 1947. = Satureja (Quar- doquia) micromiroides Briq. Hedeoma taxana Cory, Rhodora a8: 405. 1936. = Rhododon ciliatus (Benth.) Epl. oe thymoides Pers., Syn. 2: 131. 1806. = Thymus serpyllum L. (fide Bent- han Hedevm villosa (Benth.) Briq., Engler & Prantl, Nat. Pflanzenf., ed. 1, IV, 3a. 295. = Rhabdocanlon lavanduloides (Benth.) Fpl Pens villosa var. lavanduloides (Benth.) Briq., Bo ler & Prantl, Nat. Pflanzenf., ed. 1, IV, 3a. 295. 1897. = Rhabdocaulon lavanduloides (Benth.) Epl. ACKNOWLEDGEMENTS I should like to give special thanks for Drs. B. L. Turner and Tod F. Stuessy and to Joellen Henthorn for their support and Fear esi throughout this undertaking. Thanks are also owed to Dr. M. C. Johnston for his Latin diagnoses, to the late L. H. Shinners for his bibliographic assistance and to the curators of the various herbaria from which specimens were borrowed. Based on a dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the University of Texas at Austin 78712, much of the initial work was conducted under an NSF Traineeship GB-6914 to the Graduate School of the University of Texas at Austin. Subsequent work was assisted by NSF Grants GB 12910 and 08559. Mrs. Elna Widman is gratefully acknowledged for her assistance in the preparation of the manuscript as is the New York Botanical Garden for granting permission to reprint Figure 1 (in part) from Irving (1970) as the frontispiece. Rita Paulaski did the original drawing. REFERENCES AXELROD, D. I. 1950. Evolution of desert vegetation in western North America. Carnegie Inst. Wash. Publ. 590: 215-360. AXELROD, D. L. 1958. Evolution of the Madro-Tertiary Geoflora. Bot. Rev. 24: 433-509. BENTHAM, G. 1832--1836. Labiatarum Genera et Spec London BENTHAM, G. sae ee In A. P. de Candolle, cea. Scan Nautralis Regni oo 27-603. - D. ate 1876. Genera Plantarum 2: 1188-1189. s 5 BRIQUET,. . 1897. Labiatae. In A. Engler and E. K. Prantl, Die Natirlichen Manzoni 4(3a): 183-375. EPLING, 939. ale on the Linnean types of American Labiatae. J. Bot., Bric. & aha . = 1945-19 ee SAG opsis of the South American Labiatae. I. Repert. Sp. Nov. o L. '85(1): 1- 06. 1935; Il. 85(2): 97-192. 1936. III. 85 (3): 193- 388. 1936; . 85(4): 289-341. 1937. (cf. 85(2): 132-134. . 1940. Supplementary notes on American Tabigae, Bull. Torr. Bot. Club 67: 509-534, ., and fe S. Stewart. 1939. A revision e Hedeoma with a review of allied genera. Report. Spec. Nov. Regi Veg., are 15: 1-49. GRAY, A. sg ec al Flom of Nort ay cee 7 Part I. New rk. HARDIN, J. W. 1957. A revision of the American eee meres eae 9: 145-195. 295 IRVING, R. S. 1970. Novelties in Hedeoma (Labiatae). Brittonia 22: 338-345. 1976. Chromosome and numbers of Hedeoma (Labiatae) and related Genera. Systematic Bot. 1: 46—56. , and R. P. Adams. 1973. Genetic and Biosynthetic Relationships of Monorecsence In Recent advances in Phytochemistry, Vol. 6, Academic Press, N.Y. KAPADIA, Z. J. 1964. Varieties and subspecies. A suggestion towards greater uni- formity. Taxon 12: 257-2 LEWIS, H., and M. E. LEWIS. 1955. The Genus Clarkia. Univ. Calif. Publ. Bot. eee ORNDUEF, R. (ed.). 1967. Index to Plant Chromosome Numbers for 1965, Regnum Vegetabile 50: 1-128. PERSOON, C. H. 1805-1809. Synopsis Plantarum, Pari RAVEN, P. H. hee The Systematics of Oepoibers eee Chylismia, Calif. Pub. Bot. 34(1): 21. anEAEL, J.K. 1805. Flora of the Southeastern United States. Pub. by the author, New Yor TUR RNER , B. 1956. A real study of the genus Hymenopappus (Com- poste). ae 58: 163-308 TAXONOMY OF TETRAGONOTHECA (ASTERACEAE-HELIANTHEAE) B. L. TURNER and DEBORAH DAWSON The University of Texas, Dept. of Botany, Austin, TX 78712 Tetragonotheca was first proposed by Linnaeus in 1753 to accommodate a single epappose species, IT. helianthoides, which had been collected in the Carolinas by Gronovius. The genus receives its name from the four large, valvate, outer involucral bracts which house the florets and their associated receptacular scales or palea. Gray, in 1841, described the second known species, Tetragonotheca ludoviciana. Because it had somewhat different, epappose, achenes he originally placed this in a newly erected genus, Halea. This was subsequently transferred to Tetragonotheca. Engelmann and Gray, in 1848, added a third species, Tetragonotheca texana. Buckley, in 1861, added the fourth species, T. repanda, although Gray subsequently reduced this to varietal status under T. ludoviciana. The pres- ent treatment accepts the specific status of T. nies and notes its relation- ship to be with T. texana instead of T. ludoviciana. In short, we recognize four allopatric species for the genus. So far as known, they do not intergrade or hybridize in nature. CHROMOSOME COUNTS Turner (1959) was the first to report chromosome numbers of Tetrago- notheca, finding m= 17 pairs in three of the species examined. Shortly thereafter he counted the fourth species (T. helianthoides, n = 17). Voucher for the plants concerned are listed in Table 1. All counts were made from me‘otic material at late diakinesis. At this stage bivalents show, character- istically, two chiasmata (Fig. 1). Subsequent onan a 1) have also found these several taxa to be diploid with 2 = 17 pat GENERIC RELATIONSHIPS Most workers have placed Tetragonotheca in the tribe Heliantheae. Sub- tribal disposition has been somewhat more controversial. Both Bentham (1873) and Gray (1886) include it in the subtribe Verbesininae, as di Hoffmann (1893). Stuessy (1977), however, placed the genus in the sub- tribe Helianthinae. Species from genera of yet other tribes have been inadvertently (or ignorantly) placed in Tetragonotheca (cf., excluded species), the most notable being Guwizotia abyssinica (Coreopsideae) and Rumfordia (Melam- SIDA 8(3): 296-303. 1980. 29) podineae). By inference, then, some relationship to these taxa may be postulated. It should be noted, however, that both Gwizotia (Baagde, 1974) and Rumfordia (Sanders, 1978) are but dubiously positioned in these tribes, having strong relationships with the Heliantheae. It is our feeling that Tetragonotheca is relatively remote from any extant genus. Were it to have unspecialized, multiseriated involucral bracts, it would probably be positioned in the subtribe Helianthinae, near Vigwiera. ACKNOWLEDGEMENTS n preparing distribution maps we have compiled records from those herbaria listed below (abbreviations according to Lanjouw and Stafleu, Index Herbariorum, Regnum Vegetabile 31: 1964). We are grateful to the insti- tutes concerned for the loan of this material: the number of sheets borrowed, upon which the distributional maps are based, is shown in parenthesis. L MO (94)* DUKE (15)* NC (G2) PM <90)* x (92) * Pow. 20" SMU (82) GH (100) * TENN (41) LAF (4) oe (100) LL (50) C (29)* MICH (27)* - (122) * This paper was originally begun in the early sixties by Mr. William L. McCart, graduate student at the University of Texas, Austin. Annotation of material from the above institutions marked with an asterisk (*) was undertaken at that time. Subsequent annotations have been made by the present authors Table 1. Chromosome counts in Tetragonotheca. Species Weucher* or sete dinee ” number (pairs ) T. belianthoides FLA. Alachua Co.: Pom a: et + al (1974) 17 FLA. Columbia Co.: Turner 4927. aby FLA. Escambia Co.: Turner 4919 17 GA. Harris Co.: Jones et al. ]: 17 MISS. Lamar Co.: Jones et al. (1966) 17 T. ludoviciana TEX. Montgomery Co.: Turner (1959) 17 T. repanda TEX. Atascosa Co.: Turner 4979. 17 TEX. Jim Hogg Co.: Strother (1972). 17 TEX. Medina Co.: Turner (1959). 17 T texana TEX. Travis Co.: Turner (1959). 17 TEX. Travis Co.: Urbatsch (1974). 17 * On file, Plant Resources Center, Austin. xs 8 oo YT Fig. 1. Meiotic configuration at diakinesis in Tetragonotheca helianthoides (X 2600). TETRAGONOTHECA L. Tetragonotheca L., Sp. Pl. 2: 903. 1753. Bikera Adans., Fam. 2: 130. 176: Gonotheca Raf., im Med. Repos. N. ren 5: 352. 1808. Halea T. & G., Fl. N. Amer. 2: 304. 1842. Tetragonosperma Scheele, Linnaea 22: ee. 1849. Coarse, erect, perennial, caulescent herbs with strong tap roots. Leaves Opposite, simple, varying from nearly entire to coarsely-toothed to some- what pinnatifid, their sessile bases often connate-perfoliate. Heads large, solitary, the disk and ray flowers yellow. Involucre double, the outer of 4 large, ovate phyllaries which are valvate in bud and remain united below forming a 4-angled or winged cup; the inner phyllaries membranous, as many as the ray flowers, and partly enclosing their achenes. Receptacles convex or conical, with narrow, membraneous chaff. Ray flowers pistillate, fertile. Disk flowers perfect, fertile. Style branches flattened, with marginal, stigmatic lines, the apices with an elongate hispid appendage. Achenes 27 quadrangular to sub-terete, truncate or rounded at the summit; pappus wanting or of several to numerous small, short scales. Basic chromosome number, x = 17 pairs. Type species: Tetragonotheca helianthoides L. Key to species of Tetragonotheca 1. Pappus of 16 or more well-developed scales; plants of deep sandy soils in eastern S. 2; me mostly ete ae on the main stem few, remote, and an reduced; T. repanda 2. Leaves not mostly basal, those on main stem well-developed, markedly perfoliate, the ae usually pore peduncles 2-12(15) cm _ long; plants of eastern Tex 2. T. ludovictana . Pappus absent or a aor developed fringe of ciliate scales (rarely comprised o 15 or fewer lacerate or fimbriate scales up to 1.0 mm long); plants of the south- — 3. Mid-stem leaves (3)4-8 cm wide, not perfoliate; involucral Beas mecde 15 mm long or more; plants of the southeastern U.S. (east of the Mississippi River ) 1. T. helianthoides 1. TETRAGONOTHECA HELIANTHOIDES L. Sp. Pl. 2: 903. 1753. Type: VIRGINIA. Gronovins 301. Phototype (GH! ) Gonotheca helianthoides (L.) Raf., Med. Repos. N. York 5: 352. 1808. Perennial herbs from stout tap roots. Stems commonly tufted, 3-10 cm tall, densely pubescent with crisp, spreading hairs (rarely glabrate or villous). Leaves broadly ovate to elliptical, sessile, mid-stem blades 5-20 cm long, 4-10 cm wide, acute or acuminate at the apex, remotely but sharply toothed or merely crenulate to nearly entire. Heads, excluding the rays, 2-4 cm across. Peduncles 5-8 cm long. Involucral ee broadly ovate, 2—3 cm long, 1-2 cm wide. Ray florets 8-13, the ligules yellow, 2-4 cm long, 0.5-1.0 cm wide; tube 6-8 mm long, pubescent with spreading hairs. Disc florets greenish yellow; tube 1.5—2.0 mm long, sparsely to densely, crisp- spreading pubescent; throat 4-5 mm long. Achenes broadly clavate, turgid, subterete to somewhat 5(4)-sided, 4—6 mm long, striate, glabrous, epappose. Chromosome number, #7 = 17 II DISTRIBUTION AND ECOLOGY (Fig. 2). Mostly in sandy soils of cut-over pine and oak woodlands; southeastern United States, not known west of the Louisiana River. Flowering Apr—Aug. Except for its vestiture, the species is remarkably uniform considering its wide distribution. On both morphological and chemical grounds (Urbatsch, pers. comm.), T. elianthoides seems closest to T. Iudoviciana, ancestral populations from which it was presumably derived through regional isola- tion, loss of pappus and numerous other small but, in consort, significant diagnostic features. Pes * as aves ye oy . Sere sf ae > Cok ak Ks ws < % yr yf 5% ie S Ley Wan a yee uae! > 9 va relays reAee ay, w AC RS gave senne 7 SIT Fig. 2. Distribution, by county, of Tetragonotheca helianthoides. 301 2. TETRAGONOTHECA LUDOVICIANA (T. & G.) Gray ex Hall, Pl. Texanae Halea ludoviciana T. & G., Fl. N. Amer. 2: 304. 1841. Type: LOUISIANA. “Wes- tern Louisiana.” Dr. Pen ae. $2. (apparently collected before Jun 6, 1837, according to notation on the holonpe): Holotype (N Perennial herbs from stout, elongate tap roots. Stems mostly single but sometimes tufted, especially following injury, densely white villous to glab- rous. Leaves broadly triangular-ovate to somewhat repand, especially below, the mid-stem leaves nearly always connate or perfoliate, irregularly coarsely dentate to crenulate. Heads, excluding the rays, 1.5—3.5 cm across. Peduncles mostly short, slender, 2-8(12) cm long. Involucral bracts ovate, 1.0-1.8 cm long, 0.8-1.2 cm wide. Ray florets mostly 13, the ligules yellow, 1.0-1.8 cm long, 0.5-0.7 cm wide; tube 4-5 mm long, glabrous to sparsely pubescent. Disc florets yellow; tube ca 1.5 mm long, sparsely pubescent; throat 4—5 mm long. Achenes 4.5-6.0 cm long, 5(6)-sided, sparsely appressed pube- scent; pappus a crown of 20-30 short, a entire or erose, scales, 0.5-1.5 mm long. Chromosome number, 2 = DISTRIBUTION AND ECOLOGY (Fig. 3). Deep oe usually sterile, white soils of southeastern Texas and adjacent Louisiana. Flowering, Apr—Aug. 3, TTETRAGONOTHECA TEXANA Gray & Engelm., Proc. Amer. Acad. 48. 1848. Type: TEXAS. Guadalupe Co.: Hills between the Guadalupe and Cibolo rivers, 12 mi W of New Braunfels, according to label data (MA), Apr 1846, F. Lindheimer 431. (Holotype, GH!; isotypes MO!, NY!, UC!, US!). Halea texana (Gray & Engelm.) Gray, Proc. Amer. Acad. 83. 1849. Te ees lyratifolium Scheele, Linnaca 22: 166. 1849. Type: TEXAS. Guadalupe Co.: Rocky soil, Cibolo River, Apr-May 1846? Roemer s.n. (Holo- type, B Perennial herbs from tough tap roots. Stems mostly slender, solitary or tufted, only moderately branched above, 3-6(7) dm tall, frequently tinged with maroon, finely pubescent to glabrate. Leaves oblong, oval or elliptic in outline, pinnatifid, incised or repand, the lower-most merely connate, the upper connate-auriculate, 3-10 cm long, 2-4 cm wide. Heads, excluding the rays, 1-2 cm across. Peduncles elongate, slender, (8)10-25 cm long. Involucral bracts 1.0-1.5 cm long, 1.0-14 cm wide. Ray florets 8-13, the ligules yellow, 4-5(6) mm wide; tube ca 4 mm long. Disc florets yellow; tube ca 1 mm long, glabrous; throat ca 4 mm long. Ac cr obpyramidal, 3-4 mm long, pubescent, eppapose or sometimes with 4-12 small, ciliate scales. Chromosome number, 2 = 17 I DISTRIBUTION AND ECOLOGY (Fig. 3). Calcareous, rocky soils, or some- times in mixed gravelly stream bottoms of south central Texas, extending into north central Mexico. Flowering Apr—Oct., depending on rains. 302 4, TETRAGONOTHECA REPANDA (Buckl.) Small, Fl. Southeastern US. p 1250, 1340. 1 Helea rep ene Buckl., Proc. Acad. Nat. Sci. Phila. 458. 1861. Type: TEXAS. ve orpus Christi:’’ May w/o year, eae s.n. (Holotype, PH! ). . N. Amer. 2: 256. 1884. Nueces Co.: “ne Halea Indoviciana var. repanda (Buckl.) Gray, Syn. F Perennial herbs from stout, elongate, tap roots. Stems 1-5 dm tall at first single but soon forming a short bushy habit through the production of stout laterals from the root-crown, minu soft pubescent, particularly at the base and below the nodes. Leaves irregularly dentate, 3-12 cm wide, [a 4]r. ludoviciana [*.]7. repanda [|r texana Fig. 3. Distribution of Tetragonotheca species. 303 12-24 cm long; blades of the rosette triangular with slender petioles; stem- leaves ovate, lanceolate to repand, narrowed into petiolar-like bases. Heads, excluding the rays, mostly 1.5-2.5 cm across. Peduncles stout, (5)10—40 cm long. Involucral bracts ovate to triangular-ovate, 1.1-2.0 cm long, 1.0- 1.8 cm wide. Ray florets 16-24, golden-yellow, the ligules 2-3 cm long, 0.5-0.8 cm wide; tubes 5-6 mm long. Disc florets yellowish-green; tube glabrous to sparsely pubescent, 1-2 mm long; throat 4-5 mm long. Achenes 4.7-5.5 mm long, turgid, 4(5) sided, sparsely hispid-pubescent, especially along the angles; pappus a crown of 20-30 entire or erose scales, 0.3-1.5 mm long, those of the ray about 1/3 as long as those of the disc. Chromo- some number, # = 17. IL. DISTRIBUTION AND ECOLOGY. (Fig. 3). Deep sandy soils of southern Texas and possibly adjacent Mexico. Flowering Mar—Nov., depending on rains, Gray treated the species as a variety of Tetragonotheca ludoviciana but, as noted by Urbatsch e¢ al. (1978; unpubl.), on chemical grounds it is closest to T’. texana. It is also similar to the latter morphologically. In fact, occasional roseate or mowed plants of 7. texana may be taken for T. repanda, but these can be readily identified by their epappose or fimbriate scales of the achenes, by their smaller leaves and generally slender stems. Excluded Species Saks a abyssinica Ledeb, Ind. Sem. Hort. Dorpat Suppl. 1824. = Gwizotia 74). eee parviflora Jacq., Enum. Pl. Carib. p. 28, 1760 (from description and Tetragonotheca guatemalensis Coulter, Bot. Gz 16: 99. 1891. = Rumfordia ee malensis (Coult.) Blake, Wash. Acad. Sci. 18: 25. 1928 (Cf. Sanders, 1978 REFERENCES BAAG@E, J. 1974. The genus Gwizotia (Compositae). A taxonomic revision, Bot. 39): BENTHAM, G. (1873). Notes on the cer aeeey see and geographical dis- tribution of Compositae. J. Linn. as (Bot.). 13: 335-577 GRAY, A. 1886. Compositae. In Ed. 2, Fl. N. Aer ie 48— 480. HOFFMANN, O. 1893. Compositae. : Die natiirlichen Pflanzenfamilien 4: 87-387. POWELL, A. M., D. W. KYHOS and P. RAVEN. 1974. Chromosome numbers in Compositae. X. Amer. J. Bot. 61: 909—< SANDERS, R. 1978. Taxonomy of arse Vteraceae Brittonia 2: 302-316. STROTHER, J. 1972. Chromosome numbers in western North American Compositae. Amer. J. Bot. 59: 242-247. STUESSY, T. 1977. Heliantheae-systematic review. In The Biology and Chemistry of See II [Heywood ef al., eds.}] Academic Press, London. TUR , B. L. 1959. Meiotic chromosome counts for 12 species of Texas Com- ote ‘Busse 11: 173-17 URBATSCH, L., N. FISHER and J. BACON (1978). Flavonoids, sesquiterpene, lactones, and polyacetylenes from Tetragonotheca. In Bot. Soc. Amer. Misc. Ser. Pbl, 156: 6. NEW SPECIES OF CLITORIA SUBGENUS BRACTEARIA SECTION BRACHYCALYX (LEGUMINOSAE) FROM COLOMBIA, VENEZUELA, AND BRAZIL’ PAUL R. FANTZ Fairchild Tropical Garden, Miami, Florida 33156 Present address: Dept. of Horticultural Science, Box 5216, North Carolina State University, Raleigh, NC 27650 Members of Clitoria L. subgenus Bractearia (Mart. ex Benth.) Fantz section Brachycalyx Fantz are characterized by short tubular to cup-shaped calices with minute lobes, inflorescences emerging from denuded nodes cere than the sane of the leaves, fruits which become slightly convex around the seeds and are co spicuously depressed between the seeds at maturity, i dae leaves, and esas an erect habit (trees or tall shrubs). Section Brachycalyx includes seven species (Fantz, 1979): Clitoria brachystegia Benth., Clitoria brachycalyx Harms, Clitoria glaberrima Pittier, Clitoria dendrina Pittier, one the three species newly described below. 1. CLITORIA hermannii Fantz, sp. nov. Frutex ad 2 m altus. Folia trifoliata, foliola suborbiculata vel late ovalia, obtusa et abrupte minute acuminata, supra trichomatibus brevibus mollibus appressis, subter velutinis. Inflorescentia 132 cm longa, axillaris a nodis denudatis ante foliis anic 4 mm longis. Flores resupinati, albi, 4.8-5.2 cm. Calyx brevitubularis, dense appresso- pubescens; tabus 11-13 mm _ longis; lobi breves, 3-4 mm longi. Vexillum extus fulvo-sericeum. Tubus staminalis 24-26 mm, apice trichomatibus uncinatis. Legumen incognitum. Erect shrub to 2 m tall. Branches solid, pubescence moderately dense with short, appressed to suberect trichomes; axillary buds 3-3.5 mm. Leaves 3-foliate, subcoriaceous, leaflets suborbicular becoming broadly oval when expanded, apex obtuse, abruptly minutely acuminate, base weakly cor- date to rotund, midrib impressed above, primary nerves of 8-10 pairs, upper surface soft pubescent, trichomes short (ca 0.5 mm), appressed, lower surface velutinous, juvenile leaflets 4-7 cm in diameter, lamina expanding to 10-12 cm long by 7-9 cm wide. Petiole elongate, 6.5-15 cm, pubescence dense, appressed; rachis (1.5) 2—2.5 cm; petiolules quadrangular, tomentose, 5—8 mm. Stipules caducous (?— not observed); stipels caducous, ‘Taxonomic studies on Clitoria accomplished at the IFAS Herbarium, Dept. of Botany, The University of Florida, Gainesville, Florida. Florida Agricultural Boek. 1346. ment Station Journal Series No. SIDA 8(3): 304-311. 1980. 305 linear, acute, 3-nerved, 2-2.5 mm long, 0.5 mm wide, sparsely pubescent. Inflorescence axillary, multiflowered, at denuded nodes appearing earlier than the leaves, subpaniculate with primary lateral branches which bear the pedicels subsessile to 4 mm long; axes crowded toward branch apices, pubescence dense; central axis 1.5-2 cm long; pedicels 5-7 mm. Bracts persistent, 1.5-2 mm long, 1 mm _ wide, pubescence dense, appressed. Bracteoles minute, ovate to suborbicular, obtuse, 2-3 mm long, 2 mm wide, pubescence appressed, dense. Flowers resupinate, white, 4.8-5.2 cm. Calyx short-tubular, pubescence dense, appressed, tube 11-13 mm long, 5-G mr wide at base expanding to 8-9 mm wide at throat, lobes minute, broadly ovate-deltoid, 3-4 mm long, 2-2.5 mm wide, ventral lobe 4-5 mm long. Vexillum with indumentum on dorsal surface ae appressed, tawny, blade ca 3 cm wide, claw broadly cuneate, 8- ong. Alae extended well beyond carina by 6-9 mm, blade 19-23 mm av 6-10 mm wide, 2-15 mm. Carina blade falcate, 10-12 mm long, 4 mm wide, claw 21-24 mm. Staminal tube 24-26 mm long, uncinate- spube escent near apex, vexillary stamen coherent near middle of tube, free above and below, free filaments 2-5 mm long; anthers lanceolate, 2-2.5 mm long, 0.5 mm wide, connective apiculate. Gynophore 5 mm, with a few spreading white tri- chomes near apex; ovary 14-15 mm long, 1.3 mm wide, pubescence white slightly tinged yellowish, dense; style bearded, 16-17 mm, genicula te 7-8 mm; stigma subcapitate, 0.5 mm in diameter. Legwme unknown. Fig TYPE COLLECTIONS: COLOMBIA. Vichada: shrub 18 in., ae leafless, fl. white, grassy llanos, along Rio ea ae José . Ocuné, 100 m, 19 Jan 1944, Hermann 10994 (HOLOTYPE: NY. Iso GH). PARATYPES: an Vichada: 6 km NE of Nee on Ilanos ine Rio a ca 36 km NE of San José de Ocuné, 100 m, 21 Jan 1944, Hermann 11019 (NY, US 1879110). Clitoria hermannii appears similar to C. dendrina and C, brachycalyx in its vegetative aspects, but the larger flowers, smaller stipels, and other morphological comparisons indicate closer affinities with Clitoria brachy- stegia of Ecuador, which is easily distinguished from C. hermannii by the elongated inflorescences (5—22 cm) and acuminate leaflets which are pilose below. All known collections of this species are from localities near San José de Ocuné, Colombia 2. CLITORIA canescens Pittier ex Fantz, sp. nov. Clitoria canescens Pittier, nom. in sched. Frutex. Folia trifoliata, taliola lanceolate elliptica, acuminata, supra glabra, subter u entia 1.5—3.5 cm vel longi c race toris lignosis pee vel ica O super ramis primariis lateralibus ad 6-7 mm longis portati. Flores parva, resupinati, 13.5 cm. Calyx brevitubularis, canescens, tubus 13-15 mm, lobi 3-4 mm. Vexillum ex ts canescens, praesertum in albastro. Twbus staminalis Eee 28 mm once Legumen incognitum. Shrub. Branches subterete, solid, internodes in weak zigzag pattern, ju- 306 venile branches with pubescence uncinate (vidi 30X) plus scattered, short, appessed trichomes (vidi 10X), becoming glabrous, axillary buds 3 mm. Leaves 3-foliolate, coriaceous, leaflets lanceolate-elliptic, apex acuminate, base broadly cuneate, midrib impressed above appearing as if it is in a shallow groove, primary nerves of 9-10 pairs, upper surface faintly glossy, PUTT : Ha AS GENS cae NRE SOE ae Hee SUAS SS GP CURRMRE tis ded Nitec 194425 Sliteria co SG Suter Sh Set, a . aes is ye ‘ . a Sheed Us de et JaefSenrs Mla, white, ¥Ou Gresky Lisnes, me 1 a. < abong the Kio Viehada CANCELLED“S Berenice - San sead ue ocunt zat SORES i TS Sys 7 Figure 1. Holotype of Clitoria hermannit (Hermann 10994, NY). 307 glabrous, lower surface dull, pubescence on nerves, inconspicuous, short, appressed, lamina 15-17 cm long, 6-7 cm wide. Petioles subquadrangular- terete, 9 cm; rachis 3.5 cm; petioles subquadrangular, rugose, glabrate, 8-10 mm. Inflorescence terminal, racemose-nodose to subpaniculate, pedice!s borne on central axis at lignose knobs or occasionally on a primary lateral branch of 6-7 mm; axes compressed laterally, subquadrangular, twisting, pubescence moderately dense, appressed; central axis 1.5-3.5 cm long or possibly longer (apex broken); pedicels strigose, 6 mm. Bracts minute, ovate, acute, strigose, 1-2 mm long, 0.5-1 mm wide. Bracteoles minute, ovate, acute to short-acuminate, 2-3 mm long, 1-1.5 mm wide, strigose. Flowers resupinate, small, 3.5 cm, purple (?-in dry state the appearance is similar to C. dendrina which has purple flowers). Calyx dark-colored, with short-appressed trichomes (buds densely pubescent, canescent), tube short, 13-15 mm long, 4-5 mm wide at base expanding to 7 mm at the throat, lobes minute, broadly deltoid, short-acuminate, 3-4 mm long, 2.5-3 mm wide at base, ventral lobe linear, 6 mm long, 0.5-0.7 mm wide. Vexil- lum with indumentum on dorsal surface dense, canescent, appressed, blade obovate, 1.5 cm wide, claw 6-7 mm long, cuneate. Alae extended beyond the carina by 4-5 mm, blade 17 mm long, 3-5 mm wide, claw 16 mm. Carina blade falcate, acute, 16 mm long, 4-5 mm wide, claw 20 mm. Staminal tube glabrous, 28 mm long, free filaments 3-5 mm; anthers 3 mm long, 0.5 mm wide. Gynophore 3 mm long, brownish-black, moderately densely sericeous, white trichomes; ovary 13 mm long, | mm wide, blac with very dense pubescence, trichomes white, sericeous; style black near base and sericeous, bearded above, 18 mm long, geniculate 3-4 mm; stigma compressed, 0.5 mm in diameter. Legwme unknown. Fig. 2 TYPE COLLECTION: VENEZUELA. Amazonas: sitios arenoso de la boca del Gahama, Cano §. Miquel, ae alto Rio Negro, 26 Mar 1942, Williams 14893 (HOLO- TYPE: VEN 403 The type specimen bears the name “Clitoria canescens Pittier, sp. n.” typed on the label and is placed in a separate type folder bearing this name. No record of the publication of Pittier’s name has been found. Pittier (1944) did not include this species in his key to the species of Clitoria in Venezuela. Dried flowers of this species morphologically suggest Clitoria gies more than any other species in the genus, although C. canescens can cee easily by the canescent vexillum and calyx tube pacer during the bud stage), longer calyx tubes, longer style, and the sparse i ie of the leaves. In addition, the inflorescences are less crowded with flowers and occasionally bear primary lateral branches, the ovary lacks any yellowish pigmentation, and other floral structures differ in their measure- ments. Clitoria canescens is known only from the type locality. 3. CLITORIA froesii Fantz, sp. nov. Frutex caulibus scandentibus. Folia trifoliata, foliola lata ovata vel late ovalia vel 308 HERBARIO NACIONAL DE VENEZUELA > itoria panencans Pittior BP. Ne oe Ameo oe SS RS. Pear cf _ Hy iy cal Figure 2 Ls Holotype of Clitorta canescens (Williams 14893, VEN 4038) = ceostece Figtha PSR A de Agriciturs y Cris 309 late ee acuminata, supra hirsuta, subter pilosa. Inflorescentia 1.5—3 cm, pauci- flora. Flores resupinati, parvi, 3.5-4 cm. alyx peas tubus 12-13 mm, i 3-4 mm. Vexillum extus sericeum. Tubus staminalis glaber, 24-28 mm, antherae 2.4-3 mm, connectivo —. apiculato. Legumen incognitum. es a) n fay Pp Ko) us} 5 igy wn wn oa! Ee og ig) a a o =) 2 is -_ W Liana. Branches 3-9 mm in diameter, solid becoming hollow in larger diameter branches, juvenile branches longitudinally striated, angular-terete, pubescence densely uncinate (vidi 30X) beneath the erect to appressed macrotrichomes (vidi 10X), the latter deciduous with age, mature branches becoming glabrate; axillary buds 5 mm. Leaves 3-foliolate, thick- membranaceous, conspicuously pubescent, leaflets slightly asymmetrical, broadly ovate to broadly oval or broadly oblong, apex acuminate, acumen to 1.5 cm nae base rotund, midrib and some of the basal primary nerves weakly raised above, pcb densely pubescent, primary nerves of 7-11 pairs, upper surface dark green, minutely hirsute, texture similar to fine sandpaper, lower surface light green, pilose, nerves densely pilose-hirsute, lamina 10-25.5 cm long, 7-16.5 cm wide. Petioles elongate, 13-24 cm, subterete, trichomes dense, erect; rachis 4-6.5 cm; petiolules 5-7 mm, quadrangular, trichomes dense, erect, tawny. Stipules caducous, lanceloate, acute, 4-7 mm long, 2 mm wide on juvenile branches to 10 mm x 3 mm on more robust branches; stipels caducous, linear, acute, weakly 3-nerved, 3-6 mm long, 0.5-1 mm wide. Wnflorescenee axillary and terminal, racemose- nodose; axes solitary, occasionally branched near base, few flowered, 1.5—3 cm, trichomes dense, erect; pedicels 6-7 mm. Bracts minute, ovate, acute, 2-3 mm long, 2 mm wide, densely pubescent. Bracteoles ovate, acute, mm long, 2 mm wide, densely pubescent. Flowers resupinate, small, 3.5 cm. Calyx pubescence dense, appressed, bigs short, 12-13 mm long, 4-6 mm. wide at the base expanding to 7-9 mm wide at the throat, lobes minute, deltoid, acute, (3) 4 mm long, 2 mm ade ventral lobe subequal, 1-1.5 mm wide Vexillum with indumentum sericeous on the dorsal surface, ae 2.5 cm wide, claw 4-6 mm. Alae extended beyond the carina by 4-6 m blade 18-21 mm long, 4-8 mm wide, claw short, 7-9 mm. Carina ae 9-13 mm long, 4 mm wide, claw 12-16 mm. Staminal tube glabrous, 24-28 mm long, vexillary stamen coherent below to near the middle, free filaments 2-4 mm long; anthers lanceolate, 2.4—-3 mm long, 0.5-0.8 mm wide, connective long-apiculate. Gynophore 2-3 mm; ovary 10-11 mm long, 1.2-1.3 mm wide, pubescence dense, white; style 14-15 mm, bearded, geniculate 6 mm; stigma capitate, 0.5 mm in diameter. Legume unknown. TYPE COLLECTIONS: BRAZIL. Amazonas: terra fir rma, Ma oe on Rio Maria, basin Rio Negro, a Jan — noe 12441/185 (HOLOTYPE: Y-Hb. Krukoff. SOTYPE: A). PARATYPE: BRAZIL. Amazonas: Pirapuca on Rio oe 1941, Froés 12431/175 (S-Hb. at Wall). Clitoria froesi is the only species of section Brachycalyx with a climbing 310 habit, but the short calyx with minute lobes, the pubescence on the upper surface of the leaflets, and the lack of cauliflorous flowers suggest that this species should be assigned to section Brachycalyx rather than section Cauliflorae which is characteristically composed of species with a climbing Figure 3. Holotype of Clitoria froesti (Froes 12441/185, NY). onal habit. Clitoria froesi has close affinities with Clitoria dendrina which is dis- tinguished by the arboreal habit, smaller calyx, smaller anthers, more crowded inflorescences, and leaflets which are velutinous below and broadly ovate to rhombic-ovate to orbicular. Clitoria froesii is known only from western Amazonas, Brazil. Froes noted that the local Indians call this plant “Darume Vine” and plant it in fair quantities, considering it to be the most toxic plant in the region of the Rio Icana. Frées does not indicate for what purpose the toxin is used by the Indians. Other South American species of Clitoria are known to produce toxic seeds which are used locally as a fish poison. Acknowledgement is made to Helen Correll for the Latin descriptions, and to Daniel B. Ward, Dana Griffin II, and Donovan Correll for reviewing the manuscript and providing constructive criticisms. REFERENCES FANTZ, PAUL R. 1979. Taxonomic notes and new sections of Clitoria subgenus Bractearia (Leguminosae). Sida 8: 90-94. PITTIER, HENRY. 1944. Leguminosas de Venezuela I: Papilionaceae. Bol. Tech. Min. Agric. Caracas 5: 49. NOTES A CHROMOSOME COUNT FOR JUNIPERUS ASHEI (CUPRES- SACEAE) AND ADDITIONAL CHROMOSOME NUMBERS FOR HEDEOMA (LABIATAE)—lIn a_ study of possible hybridization in Juniperus, some years ago, I was able to establish a chromosome record for Juniperus asher Buchholtz. To my knowledge this is the first documented chromosome report for this species. The count was taken from root tip material using excised embryos grown on nutrient enriched agar. With this technique no pretreatment (stratification) of the seed material was necessary. During recent biosystematic studies of Hedeoma and allied genera (Irving et al., 1979) additional chromosome numbers were established for three previously uncounted taxa. These counts, derived from root tips, sup- plement Hedeoma chromosome numbers reported earlier (Irving, 1976). The numbers for H. montanum and H. nanum var. macrocalyx (2n = 36) are consistent with those of related taxa. H. multiflorum of Uruguay and Argentina, however, was tetraploid (22 = 72) but is closely related to H. drummondi (2n = 36) of Mexico and western USS. JUNIPERUS ASHEI oo tz (Fig. 1) 2” = 22. USA, Texas, Hays Co.: R. S. Irving s.n. (MONTU). HEDEOMA MONTANUM Brandegee 27 = 36. MEXICO, Coahuila, Sierra de Parras: R. S. Irving 77-7 (TEX). HEDEOMA NANUM (Torr.) Brig. var. MACROCALYX Stewart, 27 = 36 U.S.A, Ariz., Yavapi Co.: R. B. Oxford 428 (ASU). HEDEOMA MULTIFLORUM Benth. 2” = 72. URUGUAY, Mercedes: R. S. FIGURE 1. Chromosomes of Juniperus ashei. Tracing from photomicrograph. SIDA 8(3): 312. 1980. 313 Irving 7715 (TEX).—Robert S. Irving, 1422 Summit, Little Rock, AR 72202. - REFERENCES IRVING, R. S. 1976. Chromosome numbers in Hedeoma (Labiatae) and related genera. Syst. oe 1: 46-56. oe, RENHOLTS, & D. D. IRVING. 1979. Artificial hybridization in Hedeoma Gablatae): Syst. Bot. 4: 1-15. NEW AND RE-INSTITUTED COMBINATIONS IN GUTIERREZIA (COMPOSITAE: ASTEREAE)—In connection with monographic work on what is commonly known as the Xanthocephalum complex, and treat- ment of these genera for the Flora of the Chihuahuan Desert Region (M. C. Johnston, and collaborators, in preparation), the combinations listed below are necessary. Justification of such dispositions and full synonymy will be provided in a forthcoming doctoral dissertation on the group. There is strong morphological and chromosomal evidence supporting the close alliance of the species of the former genus Greenella with those of Gutierrezia, reflected here by the transfer of Greenella arizonica and G. amulosa into Gutierrezia. The third species of Greenella, G. discoidea Gray (Proc. Amer. Acad. Arts 19: 2. 1883), which is known only by the type material, has been found by this author to be a rayless form of Nanthocepha lum wrightii (Gray) Gray. Since I have here returned this latter species to Gutierrezia also, the genus Greenella as a whole is reduced to synonymy. The only objection that Gray himself had to this placement was ray-floret color, and since there are several South American gutierrezias with white rays, maintenance of a distinct genus on that basis is meaningless. As a result of these transfers and the changes of status of two of the taxa involved, there are now 14 North American species of Gutierrezia. Gymno- sperma glutinosum is closely related to this group and may be transferred to Gutierrezia in the future; this taxonomic decision must await additional study. 1. GUTIERREZIA conoidea (Hemsley) Lane, comb. nov. Based o Xanthocephalum conoideum Hemsley, Biol. Centr. Amer. 2: 109-112. 1882. 2, GUTIERREZIA ALAMANI Gray, Smithsonian Contr. Knowl. 3(5): 91 (Pl. Wright. 1: 91). 1850. This species has been known most recently as Xanthocephalum linearifolium (IDC) Greenman (Publ. Field Mus. Nat. Hist. Bot. series 2: 345. 1912), based on Keerlia linearifolia DC (Prod. 5: 309-310. 1836). However, when placed in Gutierrezia, it must take Gray's epithet because of the pre-existence of G. linearifolia Lag. (Gen. et Sp. Nov. 30. 1816) 3. GUTIERREZIA ALAMANI Gray var. megalocephala (Fernald) Lane, comb. & stat. nov. Based on: Xanthocephalum megalocephalum Fernald, SIDA 8(3); 313. 1980, 314 Proc. Amer. Acad. Arts 36: 505. 1901. 4. GUTIERREZIA sericocarpa (Gray) Lane, comb. nov. Based on: Xanthocephalum sericocarpum Gray, Proc. Amer. Acad. Arts 15: 31-32. 1880. 5. GUTIERREZIA WRIGHTI Gray, Smithsonian Contr. Knowl. 5(6): 78 (Pl. Wright. 2: 78). 1883. Gray himself later transferred this species to Nanthocephalum (Proc. Amer. Acad. Arts 8: 632. 1873), but my work has convinced me that the original placement is more correct. 6. GUTIERREZIA arizonica (Gray) Lane, comb. nov. Based on: Greenella arizonica Gray, Proc. Amer. Acad. Arts 16: 81-82. 1880. 7. GUTIERREZIA ramulosa (Greene) Lane, comb. nov. Based on: Green- ella ramulosa Greene, Pittonia 1: 302-303. 1887. 8. GUTIERREZIA SPHAEROCEPHALA Gray, Mem. Amer. Acad. Arts ns. 4(1): 73-74 (Pl. Fendl. 73-74). 1849. This species was transferred by Shinners to Xanthocephalum and considered by Solbrig to be a synonym of Gutierrezia glutinosa, however, in my treatment it will receive specific status in Gutierrezid. 9, GUTIERREZIA TEXANA (DC) T. & G. var. glutinosa (S. Schauer ) Lane, comb. & stat. nov. Based on: Hemiachyris glutinosa S. Schauer, Linnaea 19: 724. 1847.—Meredith A. Lane, Department of Botany, Univer- sity of Texas, Austin, TX 78712. HESPERALOE FUNIFERA (AGAVACEAE) IN TEXAS.—Hesperaloe funifera (Koch) Trelease was collected in the course of field work for a Texas Natural Area Survey of the Devil’s River—Dolen Falls area in Val Verde Co. (Smith, J. M. and M. Butterwick. 1975. A Vegetational Survey of the Devil's River—Dolan Creek Area. In A Natural Area Survey Part VI of VUI. Division of Natural Resources and Environment, The University of Texas, Austin, Texas. Available at the Texas State Library ). This species is known from northern Mexico from eastern and central Coahuila to Nuevo Leon (one station) and San Luis Potosi (several stations) (R. Engard, Desert Botanical Garden, Phoenix, pers. comm.). A small Texas population was observed in shallow limestone soils of rocky upper slopes, ca. 5 km ENE of the Finnegan Ranchhouse (29° 56’ N, 100° 55° W). An additional individual has been transplanted at the entrance of the Finnegan ranch. Hesperaloe funifera with its large (to 5 cm wide and 2 m long), rigid leaves and purplish-green, glaucous flowers is easily distinguished from H. parvi- flora (Torr.) Coult. which has linear, arcuately spreading leaves (2.5 cm wide and 1.2 m long), and rosey-red to salmon colored flowers. Collection data: In rocky limestone soil at the entrance to the Finnegan Ranch (29° 56’ N, 100° 58’ W), ca 65 km NW of Del Rio, Val Verde Co., 2 Aug SIDA 8(3): 314. 1980. gD 1979, Deal sn. (TEX-LL)—Mary Butterwick, Bureau of Land Manage- ment, Phoenix, AZ 85017, and Jackie M. Poole, Department of Botany, Uni- versity of Texas, Austin, TX 78712, CUCURBITA DIGITATA (CUCURBITACEAE) IN TEXAS.—A speci- men of Cucurbita, collected during field studies for a Texas Natural Area Survey of Fresno Creek, proved to be a new state record. Cucurbita digitata Gray was previously known from southern New Mexico, Arizona, and California with Mexican localities in Sonora and Chihuahua. Cucurbita digitata has palmately 5-parted leaves which are narrowly and deeply lobed, a character that easily distinguishes it from the other Texas species. Collection data: Infrequent in gravelly alluvium along the banks of Fresno Creek, ca 4.0 km N of the Smith House of the Big Bend Ranch (29° 23’ 30” N, 103° 51’ ’0” W/), Presidio Co., 30 Sep 1975, Butterwick 1726 (TEX-LL) —Mary Butterwick, Bureau of Land Management, Phoenix, AZ S017; PHYLLANTHUS POLYGONOIDES (EUPHORBIACEAE) NEW TO ARKANSAS.—On June 21, 1979, Phyllanthus polygonoides Nutt. was dis- covered 414 miles northeast of Eureka Springs, Carroll County, Arkansas. The suffruticose plants are characterized by ascending whiplike branches clustered on a caudex, obovate leaves with acute or mucronulate tips, and fruiting calyx lobes 1.5-2.5 mm long. Hundreds of the plants were scattered in glades surrounded by oak-juniper woodlands and in an old overgrown roadbed. They were growing in exposed shallow, gravelly soil underlain by limestone. Webster (A revision of Phyllanthus (Euphorbiaceae) in the Continental United States. Brittonia 22: 44-76, 1970) listed the range of P. polygonides as “Oklahoma and Texas south to the Mexican plateau, with outliers in New Mexico, Missouri, and Louisiana.” The discovery of this station in northwest Arkansas suggests that the Missouri plants (known only from Stone County in southwest Missouri) are not disjunct but rather are at the northeast periphery of the species range which is likely continuous from Oklahoma to southwest Missouri. Dr. Edwin Smith, University of Arkansas, Fayetteville (pers. comm.) indicated to me that the flora of northwest Arkansas has been less than adequately worked and habitats similar to that described here are common in the region. Specimens (Brooks 14254) are on deposit at KANU, MO, NY and UARK.—Ralph E. Brooks, University of Kansas Herbarium and State Biological Survey of Kansas, 2045 Avenue A, Campus West, Lawrence, KS 66044. SIDA 8(3): 315. 1980. SIDA CONTRIBUTIONS TO BOTANY VOLUMES = ——SNUMBER 4 DECEMBER 1980 CONTENTS A new Bahamian Euphorbia and some new combinations. Donovan S. Correll. aly Cyperus difformis L. (Cyperaceae) in North America. Barney L. Lipscomb. 320 The distribution of Habranthus tubispathus (L’Her.) Traub in South America and North America—Texas and Louisiana. W. C. Holmes and Christopher J. Wells. 328 “> The migration and establishment of Juncus gerardii (Juncaceae ) in the interior of North America. Ronald L. Stuckey. 334 Notes on the flora of the Chinati Mountains, Presidio County, Texas. 348 Emily J. Lott and Mary L. Butterwick. NOTES. Distribution records for Digitaria bicornis in eastern United States. 352.— Najas marina: new to the Indiana flora. 353—Gaura odorata Lag. (Onagraccae ) in Louisiana. 354—A note on flower color of Wahlenbergia linarioides (Lam.) A. DC. in northwest Florida. 355—Brachyeletrum erectum and Talinum rugospermum, new species to Texas and notes on Schoenolirion wrightit. 355—Holosteum um- bellatum (Caryophyllaceae) in Ontario. 356. INDEX TO VOLUME 8. Barney L. Lipscomb and Jessica Procter. 359 US ISSN 0036-1488 SIDA, CONTRIBUTIONS TO BOTANY Founded by Lloyd H. Shinners, 1962 Editor & Publisher Dallas, Texas, 75275 Associate Editor Assistant Editor John W. Thieret Barney L. Lipscomb Northern Kentucky University SMU Herbarium Highland Heights, Kentucky, 41076 Dallas, Texas, 75275 Guidelines for contributors are available upon request beginning with volume 8. Subscription: $6.00 (U.S.) per year; numbers issued twice a year. Sida, Contributions to Botany, Volume 8, Number 4, pages 317-372. Copyright 1980 by Wm. F. Mahler A NEW BAHAMIAN EUPHORBIA AND SOME NEW COMBINATIONS DONOVAN S. CORRELL Fairchild Tropical Garden, Miami, Florida 33156 With the project on the Flora of the Bahamian Archipelago nearing com- pletion we find a few loose ends that still need attention. This is one of them. EUPHORBIA abbreviata Correll, sp. nov. Frutex crassus brevis —— ad 70 cm altus, aliquancum cupreus, dense et grosse osus, ramis glabris, typice cum textura dura et rigida. Nodi annui instructi vaginis aes sdawialive isles Cilia primaria 2-3 mm longa. Folia opposita saepe contiguae, sine appendiculis. Capsula glabra, circa 2 mm _ longa, era crasso circa 1 mm longo; semina lactea, levia in uno extremo acuta, circa 1.3 mm _ longa, ecarunculata angulis rotundatis. Stout short bushy shrub to about 70 cm tall, somewhat copper-colored, densely and coarsely branched with the branches typically hard-tissued and rigid, glabrous. Nodes annual, provided with stipular sheaths that are ciliolate adaxially; primary internodes 2-3 mm long. Leaves opposite and solitary, subsessile, fleshy, entire, glabrous, often clustered at the nodes, globose- orbicular or shortly oblong-rectangular or narrowly cuneate-obovate, 1-3 mm long. Cyathia solitary in the axils, very shortly pedunculate, about 1 mm long; glands rather large and prominent for the cyathium, blood-red or blackish, often contiguous, without appendages. Capsule glabrous, about 2 mm long, supported by a stout pedicel about 1 mm long; seeds white, pointed at one end, smooth, about 1.3 mm long, with rounded angles, ecarunculate. YPE COLLECTION: Bahama Islands, Little Inagua, on sandy thinly coppiced slope about saline lake in southwest sector of island, plants 0.7 m tall, coppery-colored, August 12-16, 1975, D. S. Correll 45998 (holotype, NY; isotypes, F, FTG, GH, JJ K, MO Additional collecti tions: Caicos Islands, Long Cay, northern end, March 7, 1911 F. & C.M. Millspaugh 9237 (NY, US); Long Cay, July 28, 1930, Paul Bartsch s. w. (US). Grisebach, in his Flora of British West Indies, p. 52. 1864, described Euphorbia vaginulata from Turks Islands, and it is to this taxon that the presently described species has previously been referred. In my explorations I have collected the rather widespread E. vaginulata on Grand Turks Island, throughout Great Inagua, in the Caicos Islands (Pine Cay, Providenciales ), SIDA 8(4): 317-319. 1980. 318 Fig. 1, a, branchlet of Exphorbia vaginulata; b, branchlet of Exphorbia abbreviata. Drawn by Priscilla Fawcett. 319 on Crooked Island and at its extreme northern limit on San Salvador Island. It is a rather slender plant with flexible, “rubbery” branches that produce elongate internodes and linear-oblong leaves. The presently described species is local in its occurrence. It has a short, stout trunk from which arise rigid thick branches that have much-abbreviated internodes and suborbicular-quadrate to shortly elliptic leaves. The following transfers are necessary for these plants to be coordinated in our Flora. They mainly represent a difference in taxonomic interpretation. EUPHORBIA exumensis . Isp.) Correll, comb. nov. aa syce exumensis Millsp., Field Mus. Bot. 2: 301. 1909. . lechioides var. poets (Millsp.) Burch, Ann. Missouri Bot. Gard. 53: 92, 9. 1966 EUPHORBIA proctorii (Burch) Correll, comb. nov. Chamaesyce proctorii Burch, Ann. Missouri Bot. Gard. 63: 378. 1976. EUPHORBIA wilsonii (Millsp.) Correll, comb. nov. Chamaesyce wilsonit Millsp., Field Mus. Bot. 2: 301. C. lechioides var. wilsonii (Millsp.) Burch, Ann. iVissourh Bot. Gard. 53: 91, 99. 1966. CYPERUS DIFFORMIS L. (CYPERACEAE) IN NORTH AMERICA BARNEY L. LIPSCOMB Herbarium, Southern Methodist University Dallas, TX 75275 Cyperus difformis is a widespread species of southern Europe and the tropics of the eastern hemisphere (Kiikenthal, 1936; Koyama, 1961; Ohwi, 1965). It is spreading and becoming established in North America like many other oriental introductions, e.g., C. haspan, C. iria, and C. rotundus. In North America, it was first noted in México, then in the United States, and now can be reported in Nicaragua, Central America. Early American collections and literature on the species date back to the 1850's. In his monograph on North American Cyperaceae, Torrey (1836) did not list or mention C. difformis. Later Torrey (1859) reported a collec- tion (1851-1852) as C. lateriflorus from México. Britton (1886) noted the presence of C. J/ateriflorus in North America as being nearly allied to C. difformis of the old world and may be the same. Howell (1934) reported C. difformis in the United States from California; Fernald (1935), from Virginia. Numerous other collections have expanded the known range into Alabama, Arizona, Louisiana, North Carolina, and Oklahoma. The known distribution of C. difformis is given in Figure 1. Cyperus difformis is native to the tropics of the eastern hemisphere spe- cifically tropical and western Asia. It is a common weed in a number of habitats but most commonly found in rice paddies. The annual plant, with culms 15-60 cm tall, is similar in habit to C. haspan but can be distin- guished from it by its subglobose spikes and obovate scales (Fig. 2). Correll & Correll (1972) has a description of the species. C. difformis was placed in section Fusci by Kiikenthal (1936) and there it is most closely related to C. fuscus of Europe and western Asia which also has been introduced into North America. HISTORY AND LITERATURE The earliest known American collection of C. difformis is that of Charles Wright from Sonora, México, in 1851. Wright's specimen 1950 was des- cribed as a new species, C. lateriflorus, by Torrey (1859). Ayers (1946) did not mention the Wright specimen in his treatment of the genus in México citing only those previously cited by McGivney (1938) who cites Wright “1960” from “New Mexico.” Wright actually collected the species in Sonora, and McGivney’s citation of “1960” was apparently a misprint or was miscopied because the label clearly shows 1950. SIDA 8(4): 320-327. 1980, 400 MILES. 600 KILOMETERS 400 8 8 8 2 1 = Distnto Federal MCE x We NA flacotecas” Fig. 1. Distribution of Cyperus difformis in North America. Long curved arrows indicate probable points of introduction and short straight arrows possible migration of the species. Wright’s collecting was done in association with the U.S. Mexican Boun- dary Commission in 1851. He accompanied Colonel Graham of the Topo- graphical Engineers between 2 Sep-4 Oct on a trip from Santa Rita, New Mexico, to Santa Cruz, Sonora. During the latter part of this trip (29 Sep), Wright collected the Cyperus just east of Santa Cruz. Wright's (1949-52) Fig. 2 tis) a " Ie ut Ky a : FAN wi ae ats N NP \ : NM ~~ Mh ae hh | iH Ne Db ae cy \ Ye Vy ig ; : d if rity in ¥ sh iyi ANS Ae Bi WN: a oe 4 i) feats wa a fy. i oy Sn Pad, . Cyperus difformis L. A. habit. B. achene, C & D. scale (from Delahoussaye 523 field notes give the following data: along mountain torrents east of Santa Cruz, 29 Sep 1851, Wright 674. Wright's field number of 674 is the same s “Wright 1950” assigned by Gray for distribution. Wright's collections of 1849-1852 from southwestern United States cre- ate a problem for botanist who try to work with his plants. Gray, who received and processed the specimens, sent them out under a new set of numbers and not under Wright's own field numbers. Wooton (1906) noted that the specimens of the 1851 and 1852 collections were under the same printed label and this made it difficult, if not impossible, to fix localities and exact dates of collections. Most of the printed labels do not show the habitat or exact locality and date as given in Wright's field notes (Wooton, 1906). Since Wright’s collection, additional collections of Cyperus difformis in 1890 and 1909 were reported by McGivney (1938) in southern Mexico. It was 25 years later that the species was first collected in the United States. Howell (1934, 1949, 1958) first reported it from California followed by Fernald (1935) from Virginia. It was then reported from Arizona by Kear- ney and Peebles (1951). Additional reports from the United States are Thieret (1964) from Louisiana, Correll and Correll (1972) from Oklahoma, Tyndall et al. (1977) from North Carolina, and Godfrey and Wooton (1979) from Alabama. INTRODUCTION AND DISTRIBUTION The data on introduction and distribution of Cyperus difformis follows a chronological order based on label data taken from specimens examined. The data of the earliest collections, as well as others, does not necessarily indicate the time of introduction, establishment or the direction of migration. Cyperus difformis was first collected on the North American continent in northern México. However, this first collection does not establish this locality and date as the time or place C. difformis was introduced. The earliest refer- ence to C. difformis in North America was Torrey (1859). A search of earlier literature [Bentham, (1839-1857); Humboldt et al. (1815); and Kunth (1822)} yielded no mention of C. difformis. However, since the plant was probably present (maybe even several years) before collection, it would be reasonable to assume that the approximate time of introduction was sometime between 1815-1851. The first collection was reported by Kiikenthal (1936) who cites Wright 1950, Santa Cruz, New Mexico. Mc- Givney (1938) also cited it from New Mexico and it has since been errone- ously reported from New Mexico in various treatments, e.g. Correll & Cor- rell (1972), Godfrey & Wooton (1979), Kearney & Peebles (1951, 1960), and Tidestrom & Kittell (1941). Additional specimens examined: MEXICO: Guanajuato: 1 mi NW of Salamanca, Waterfall & Wallis 13901 (SMU). Jalisco: near La Barca, Pringle 3261 (ENCB). Michoacan: valley of the Lerma, Pringle 3261 (MO); vicinity of Morelia, Arsene 324 3173 (MO); vicinity of Morelia, Arsene 3092 (MO). Based on herbarium specimens from México and the United States C. difformis appeared in the United States flora some 65 years later than ico. It was first collected in California in 1915: Butte Co.: very abundant in canals, ditches, roads, and fields, Jul 1915, Johnston W132] (NY). Howell (1934) published the first United States report of C. difformis based on a 1921 collection from California: Butte Co.: weed abundant in rice fields, Nelson, 25 Nov 1921, McLearn 15 (CAS) Additional specimens examined: CALIFORIA: Alameda Co.: 16th Street Statin, Oakland, 8 Jul 1935, Jussel sn. (CAS). Butte Co.: rice fields, 9-10 Aug 1921, Kennedy s.n. (UC); Sanford slough, SW corner of Butte Co., Mason & Grant 12636 UC); Gray Hodge Game Refuge NW of Pennington, Mason 12737 (CAS, UC); Sacramento Outing Club, Mason & Grant 13024 (DS, UC); rice plots, Biggs Rice Experiment Station, Tucker 2680 (CAS, SMU, UC); eastern outskirts of Oroville, Howell 34637 (NY). Colusa Co.: Colusa rice fields, 18 Sep 1925, ye SM. (UC); rice fields, Dunshee 14 (UC); rice fields, 3 mi W of Williams, Howell 19838 (CAS, DS, MO, NY, SMU, UC); 8 mi N of Colusa, Mason & Grant 12983 (UC); 2 mi W of Colusa on Williams ee Mason & Grant 13011 (UC); 1.6 mi W of Williams, Mason & Grant 13116 (UC); rice field, 1 i Nobs & Smith 1877 (UC). Fresno Co.: es River E of Burrel, Mason & Smith 8341 (DC, UC). Glen Co.: edge of pond S of Willows Migratory Wildfowl Refuge, Beetle 3061 (UC). Kern Co.: Tupan hwy 1/4 mi S of Adohr Rd., Twisselmann 6580 (CAS, NY); Lake Barak near Delano, Twisselmann 8957 Madera Co.: Crocker-Hoffman Ranch near Merced, Wagnon 4a (UC); San Joaquin River just E of Firebaugh, Si aie 9131 (CAS). Marin Co.: Mt. Tamalpais, camp above Marin Woods, Howell 20771 (CAS). Merced Co.: 0.5 mi S . Lake Yosemite, 27 Jul 1938, Wagnon sv. (UC); Crane Ranch S of jct of M an Joaquin rivers, eee & Smith 8328 (UC); Los Banos Bird ne Jussel 90 Ve io) nae 5 5 * (CAS); San Joaquin River 0.25 mi S of hwy 152 2.9 mi E of Santa Rita Park, Wiggins & Wiggins 20701 (DS, ENCB, NY). Nevada Co.: Wolf oo 1 mi N of Nevada-Placer Co. line, Nobs & Smith 1534 (UC). Plumas Co.: near Virgilia, Howell 35064 (NY). pene Co.: American River at Fair Oaks, Crampton 1713 (CAS). San Diego Co.: Balboa Park, Gander 8893 (DS, Oy: San Francisco Co.: Lake Merced, 7 Aug 1953, Bush s.n. (CAS). Santa Barbara Co.: Las Positas Rd. near Fox Canyon, 12 Oct 1962, Pollard s.n. (CAS, ENCB); Las Positas Rd., near jet of Veronica Springs = 3 Nov 1962, Pollard sn. (MO, NY, SMU); 3 Nov 1963, Pollard s.n. (CAS). Sonoma Co.: Guerneville, Rabtzoff 2919 (CAS). Stanislaus Co.: 4 mi S of Oakdale, Hoover 57 (CAS); Claus, Hoover 1628 (UC). Sutter Co.: 6.3 mi E ot Bue Landing, 6 Jul 1938, Cantelow sm. (CAS); Sutter by-pass on Yuba Knight’s Landing hwy, Mason 12705 (NY, UC); 1.7 mi S of Robbins, 7 Bee (NY). Tehama Co.: Pacific hwy 0.5 mi N of Butte Co. line, Heller 14804 (MO, NY, UC). Tulare Co.: Pixley Nat. Wildlife Refuge, Tip ebaone 11590 (CAS). KNOWN RECORDS FROM OTHER STATES VIRGINIA. The collection of C. dfformis from Virginia in 1934 (Fernald, 1935) is apparently the only one known from the state. This probably repre- sents a separate and direct introduction from Asia than from the west coast. Its most probable source of introduction is in rice-straw used in packing, 325 as suggested by Fernald. ARIZONA. Cyperus difformis was first reported as likely to occur in Arizona by Kearney & Peebles (1942). “Not known definitely to occur in Arizona but has been collected at Santa Cruz, Sonora, about 20 miles south of the international boundary.” The following specimens are known from Arizona. Maricopa Co.: swamp area below dam of Lake Pleasant Regional Park, 8 Oct 1964, Lehto 5461 (NCU). Mohave Co.: Goose Lake of Havasu Lake Refuge near Toprock, 27 Oct 1948, Monson s.n. (CAS). These collec- tions probably represent introductions from México or California. LOUISIANA. Thieret’s report of C. difformis in Louisiana is based on the following collection: Lafayette Parish: rice field 13 mi W of Lafayette, 23 Oct 1963, Delahoussaye 71 (SMU). Through correspondence, Thieret indicated he had later collected the species in Lafayette Parish but at a different locality: Lafayette Parish: rice fields, 2 mi NW of Indian Bayou, 10 Sep 1977, Thieret 50236 (KNK). These collections are probably inde- pendent introductions from Asia. HOMA. The Oklahoma report was made by Correll and Correll (1972): Leflore Co.: gravel bars along spillway of Lake Wister, 17 Sep 1970, Correll & Correll 39765 (TEX). One additional specimen can be reported from Johnston Co.: moist soil along Blue River, 24 Sep 1977, Puckett 36 (SMU). This specimen was a gift from Dr. Mickey Cooper of Cameron Uni- versity and was collected on a class field trip through southeastern Oklahoma. The source and origin of introduction of these are questionable. It is possible that they have originated from the populations in Louisiana. ALABAMA. Limestone Co.: bottoms of Tennessee River of Wheeler Na- tional Wildlife Refuge N of Dekatur, 22 Sep 1970, Kral 41221 (ENCB, NCU, SMU). C. difformis has only recently been reported from the state by Godfrey & Wooten (1979). NORTH CAROLINA. Dare Co.: ca 1.5 km N of Duck, along the shore- line of Currituck Sound, with Scirpus americanus, 2 Oct 1978 Tyndall 791 (ODU, SMU). It was first collected 14 Oct 1974 at the same location, Tyndall sn. (ODU) and has only recently been reported from North Caro- lina by Tnydall et al. (1977). DISTRIBUTION IN CENTRAL AMERICA Early studies—e.g., Clarke (1908), Hemsley (1882-1886), and Standley (1930)—do not list C. difformis from Central America. Hemsely’s report of the species is that of C. Jateriflorus from Sonora. The only collection examined was collected in 1969 in Nicaragua. Mata- galpa: beside pond, Rte 1, S of Dario, Calabazas, 25 Dec 1969, Atwood 2566 (SMU). This has recently been reported from Nicaragua by Seymour (1980) as C. haspan with duplicates at BM, ENAG, F, GH, NY, WDP. The nearest known localities are reported in this paper and are about 1000- 1100 miles to the northwest in Michoacan. It is unknown whether or not 326 C. difformis has become established in Nicaragua. The specimen was mature when collected and had already dropped “seed,” thus probably establishing the original source or increasing the existing population. SUMMARY AND DISCUSSION Cyperus difformis was originally described by Linnaeus in 1756. The habitat noted by Linnaeus was in India. It is suspected and generally accepted that, by means of man, especially with the introduction of rice, C. difformis has spread to the new world tropics and subtropics of Central and North America, Since the introduction of it into the United States, probably around 1910-1915, it has become established through the southeast and southwest from Virginia to California. Since being introduced in California it has spread over much of the state as a weed and has become a potential problem in certain rice-growing regions. ACKNOWLEDGEMENTS Thank you John Thieret for taking time to read and review this paper. Your comments and suggested changes are much appreciated. I thank the herbaria, CAS (24), DS (6), ENCB (4), KNK (1), MO (6), NCU (2), NY (12), SMU (9), and UC (24), who kindly loaned specimens. A total of 88 sheets were borrowed. The numbers in parenthesis are numbers of sheets borrowed. REFERENCES cae B.B. 1946. The genus Cyperus in México. Catholic Univ. Amer., Biol. Stud. BRITTON, N.L. 1886. A preliminary list of North ae er of Cyperus, with descriptions of new forms. Torrey Bot. Club. Bull. 13: —216. CLARKE, C. B. 1908. The Cyperaceae of Costa Rica. a ts "Natl. Herb. 10: 443-471. CORRELL, D.S. & H.B. CORRELL. 1972. Aquatic a wetland plants of pig western United States. Environmental Protection Agency. Washington, D.C. FERNALD, ae 1935. Midsummer vascular plants of oe Virginia. —. dora 37: 378-4 ee REY, R. K & J. W. WOOTEN. 1979. Aquatic and wetland plants of south- astern United States. The University of Georgia Press, Athens. 272. HEMSLEY, W. B. 1882-1886. Biologia Centrali-Americana. Vol. 3: 449. Porter and Dulau & Co., Lon HOWELL, J. T: 1934, Tepes difformis L. in California. Leafl. West. Bot. 1(10) : 104. _ . 1949, ee Flora. Univ. of California Press, Berkeley. 88. , P. H. VEN, & P. RUBTZOFF. 1958. A flora of ‘San Francisco, Cali- focaia: The ae - San Francisco, California. 47. KEARNEY, T. H. & R. H. PEEBLES. 1942. eee plants and ferns of Arizona. S.D.A. _ Publ. 423. Washington, D.C. 15 1951. Arizona flora. Univ. of California Press, Berkeley. 149. . Arizona flora (2nd ed, with suppl. by T. J. Howell & x E =F es ene Tey. ar California Press, Berkeley. 149. aed. KOYAMA, T. 1966. Classification of the family Cyperaceae (3). Quart. J. Taiwan Mus. 14: 159-194. KUKENTHAL, G. 1936. In Engler, Das Pflanzenreich. 4(20): 1-315. MC GIVNEY, SISTER M.V.D.P. 1938. A revision of the eae Eucyperus found in the United States. Catholic Univ. Amer., Biol. Ser. No. 26. OHWI, J. 1965. Flora of Japan (in Faglich): Smithsonian Institution, Washington, D.C. OLE eee F. C. 1980. A check list of the vascular plants of Nicaragua. Phytologia p. 66. Mem Ge P. C. 1930. The Cyperaceae of Central America. Publ. Field Mus. Nat. Hist., Bot. Ser. 8(4) : 237-292. THIERET, J. W. 1964. More additions to the Louisiana flora, Sida 1(5): TIDESTROM, I. & S. T. KITTELL. 1941. A flora of Arizona and — Mexico. Catholic Univ. of America Press, Washington, D.C. 766. TORREY, J. 1836. ae of North American Cyperaceae. Ann. Lyceum Nat. Hist. a Aa oe = 859. In cot on ac United States and Mexican Boundary Survey. 2(1): TYNDALL, R. W., J. W. USHER, & G. F. LEVY. 1977. Cyperus difformis L. a North Carolina state as and additions to the flora of Dare County. J. Elisha Mitchell Sci. Soc. 93: WOOTEN, E. O. oe Souther localities visited by Charles Wright. Bull. Torrey Bot. Club. 33: 561- WRIGHT, C. 1849- ede Wright's field notes and collection numbers. L. H. Shinners’ copy at THE DISTRIBUTION OF HABRANTHUS TUBISPATHUS (L’Her.) TRAUB IN SOUTH AMERICA AND NORTH AMERICA— TEXAS AND LOUISIANA W. C. HOLMES Biology Department, Northwestern State University, Natchitoches, LA 71457 and Institute for Botanical Exploration, Mississippi State, MS 39762 and CHRISTOPHER J. WELLS Biology Department, Mississippi State University, Mississippi, MS 39762 and Institute for Botanical Exploration ABSTRACT The plant known as Habranthus texanus (Herb.) Steud. is considered synonymous with H. tubispathus (L'Her.) Traub. Its distribution includes warm temperate South seventeenth or early eighteenth centuries possibly by Spanish missionaries. The plant known as Habranthus texanus (Herb.) Steud. is cited by Sealy (1937) as the only species of the genus to occur in the wild outside of South America. Despite being considered endemic to Texas by Correll and Johnston (1970), this plant has been reported in Louisiana as early as 1934 by Dormon. It has a long and confusing nomenclatural history. This appears to be caused, in part, by the plants of the United States never having re- ceived clear distinction from certain warm temperate South American ex- pressions of the genus. This paper is intended to clarify the nomenclature and distribution of this interesting plant and provide information on its ecology and occurrence in the United States. The plant, known locally as the “Copper Lily,” is a scapose perennial arising from a brown ovoid bulb of about 1-2 cm in diameter. The apex of the bulb is abruptly narrowed into a tubular-like sheath composed of the remains of the leaves. The bulb bears 2-6 linear, deep green grass-like leaves that are 3-5 mm wide and up to 20-25 cm in length. Leaves normally wither prior to flowering. The scape is 10-20 cm tall and bears one flower. The bulb may flower more than once per year, occasionally possessing two scapes, each with a flower or fruit in various stages of development. Borne SIDA 8(4): 328-333. 1980. a2) slightly beneath the flower is a tubular scarous bract ending with a bifid, long attenuated tip. The pedicel is about twice the length of the bract. The flower is 2-3 cm long, slightly zygomorphic, and inclined. Tepals are obovate with rounded apices ending in an apiculate claw-like tip, this more pro- nounced on the outer whorl. The perianth is bright yellow, with the outer surfaces tinged with burgundy, the color deepening toward the base. The six stamens are slightly declinate and in fascicles of four different lengths and apparently attached to the base of the perianth. At the point of attach- ment are scarous, lacinate enations. The style is about two-thirds the length of the corolla (about the length of the longest stamen) and ends in a trifid stigma. The inferior ovary is somewhat tubular, three-loculate, and possesses very narrow, sub-herbaceous wings that extended onto the pedicel. At ma- turity the fruit is ovoid, three-lobed, and contains numerous black wafer-like seeds about 3—5 mm in diameter and 0.55 mm thick. Field observations and data from herbarium specimens indicate the Copper Lily proliferates in dryer, well-drained, but disturbed areas, of low vegetation. Maximum flowering occurs from July through October following periods of heavy rainfall, At such times it may exhibit complete aspect dominance. Sporadic, but very light flowering may occur at other times during this period. Flory (1939), in reviewing Paces (1913) embryological work on this plant, pointed out that this was apparently the first known incidence of diploid parthenogenesis, while at the same time having a sexually produced endosperm. Crane (1978) refers to this condition as semigamy. This appears to be the common condition in the Zephyrantheae tribe of the Amaryllidaceae (“rain lillies’). Ic gives a special advantage in allowing self pollination without genetic penalty in this group whose erratic and unpredictable flower- ing may lower the chance of insect pollination (Charles Crane, pers. comm. ) The synonymy of this species reflects not only its uncertain specific status but also the close affinities that exist between Habranthus and Zephyranthes. he plant of the United States was first designated Zephyranthes texana by Herbert. Upon proposing the genus Habranthus, Herbert himself gave it varietal status as H. andersonianus var. texanus. Steudel reinstated the plant to specific status as H. texanus. Later, Green created the combination, Atamasco texana. Direct comparison of the plants of the United States (known as Habranthus texanus) with those of South America (part of the H. andersonii complex) showed no differences that warrant retention of specific status for H. texanus. In fact, no differences could be detected and correlated with geography to give a definite geographical race or varlety worth recognizing. Our opinion is that the name H. texanus rests solely on its disjunct distribution rather than an actual morphological distinction. This point of view is shared by Sealy (1937), Alexander (1939), and Ravenna (1970). Ravenna also showed that the type of H. andersonii (of which H. texanus is here considered synonymous) is identical to the type of H. tubispathus (’Her.) Traub, the basionym being Amaryllis tubispatha. — 330 Therefore, the correct name for the only Habranthus outside of cultivation in the United States is H. inca pies synonymy is as follows: Zephyranthes andersonii (Herb.) Steud., Nom. Bot. - - A 70. 1840. m : 9, Zephyranthes andersonii (Herb.) Penis & Hooker a on Pl. Ae 724, 1883. Zephyranthes andersonii (Herb.) Baker, Handbook Amaryll. 37. 1888. Atamasco texana Green, Pittonia 3: 187. Atamasco Wie G. M., Rev. Fac. Let. & Cisne. 19: 227. 1914. The distribution of the species includes southern Brasil, Uruguay, Argen- tina,, central and southern Chile, and the United States (east Texas and northwest Louisiana). The species has, under various of the synonyms cited above, been reported in much of the Antilles and northern South America (see Pulle, 1966; Grisebach, 1864; Adams, 1972; Moscoso, 1943; and Urban, 1903). However, the plant of those areas has a white flower and is referable to Zephyranthes puertoricensis Traub. Herbaria abbreviations follow that of Index Herbariorum (ed. 6). except for Institute for Botanical Exploration (IBE), Mississippi State, Mississippi. Specimens examined: NORTH AMERICA: United States: Texas. Anderson Co.: Wilcox, near Palestine 1 Aug 1943, Barkley s.n. (TE astrop Co.: Texas Hwy. 71 on the first ll E of Garfield, 5 Oct 1966, Mears 1017 (TEX); Bexar Co.: San Antonio, Oct Thurber (F); Brazos Co ollege Station, 7 Sep 1969, n ee 1108 (SMU): Caldwell Co.: 4.65 mi N of Lockhart on Texas Hwy. 183, 1966, Mears 680 (TEX); Calhoun Co.: Port Lavaca, Gentry 46 (F); Colo- oes ee Eagle Lake, Oct 1930, Biology Class sn. (TEX): Comal Co.: New Braunfels, Oct 1850, Lindheimer 1207 (F, SMU); DeWitt Co.: without further location, 18 Jul 1942, Riedel s.n. (TEX); Fayette Co.: Muldoon, 20 Jul 1950, Ripple 51-713A (TEX); Jim Wells Co.: 15.2 mi S of Alice, 13 Sep 1955, John- ston 2781A (SMU); Gonzales Co.: 4.5 mi S of Belmont, 27 Sep 1958, Correll 20466 (LL); Karnes Co.: Karnes City, 29 Aug 1953, Johnson 1302 (SMU); Mc- Lennan Co.: M-K-T RR N of Gapshead, Jul 1946, Smith 56 (TEX); Milam Co.: U.S. Hwy. 190, ca 10 mi NE of Cameron, 26 Oct 1963, epee 63-1833 (SMU, TEX); Nacogdoches Co.: E. Austin St., Nacogdoches, 7 Jul 1955, Lacey 32 (SMU); San Patrico Co.: ca : mi NW of Sinton, 24 Oct 1948, ae 6643 (TEX); Travis 0.: mi S of Victoria, Cory sae eel Waller Co.: Hempstead, 10 Jun Ble Hall 654 (F); “Willia amson Co.: 1 SW of Georgetown, 2 Oct 1944, Wolcott 7 (TEX); Walker Co.: Near eRe Jun 1914, Young s.n. (TEX); Louisiana. oo. Parish: Columbia Park, Shreveport, 31 Jul 1977, MacRoberts 262 (LSUS); (IBE, NATC); Sabine Parish: ca 3/4 mi W of Natchitoches-Sabine Parish line on La. — 6, 18 Sep 1978, Holmes 3344 (IBE, NATC). SOUTH AMERICA. a tina. Buenos Aires. La Casada, Troucoso 1296 (F):; Brasil: Minas Gerais. Serra d Se ee 11 Nov 1938, “Bopreto 8809 (F); Chile: Cordillera de Chile, 1(90, Bene Hanenke 2122 (F); Paraguay: Cordillera de Altor, Mar 1903, Frebrig 1017 (F); Uru Montevideo. Sayago, 3 Jan 1930, Herter 85498 (F) and 85494 (F); Rahal pa: Dec 1925, Herter 79892 (F) and 79891 (F) Since the plant formerly known as Habranthus texanus is considered synonymous with H. twbispathus, it is doubtful it is native to both southern South America and Texas and Louisiana. Several points make it probable that the plant is native to southern South America and its disjunct distribu- tion of well over 10,000 kilometers is better explained in another way. These include the following: . Most other species of Habranthus (ca 23, Ravenna, 1970) are native to Sout Fev which undoubtedly is the center of distribution for the genus. Two ae nae, since 1937, been described fro Although plants pospelgiclly identical to the plants of the United States exist, considerably more variation is present in the South American populations this species. The variation is well documented by Ravenna (1970) who proposed eee infraspecific names based on flower size and minor color differences. If in- troduced into the United States, it would almost certainly have come from one population and not exhibit the total range of variation present in the species (Founders Principle). eta coupled with its a habit would explain and insure a great amount of morphological uniform eee of herbarium specimens, field ae and cultivation support thi is argumen . The plants of the United States inhabit disturbed places of ee usually grassy wevetuion (lawns, roadsides, railroads, etc.). he plant is very abundant in west-central Louisiana only in the areas settled or developed at the same time Texas was being settled by the Spanish. These in- clude the City of Natchitoches, founded in 171 14, and the El Camino Real (now roughly following Louisi ana hway 6) a trail Baa in the early 1700's con- abundant at Los aes, nee easternmost Spanish mission Pande in L/L, which is 15 miles west of Natchitoches on 1 Camino Real, miles from the state line. It is abundant on y 6, westward from Los Adaes for eo five miles. Th t has not been oe om any other place in Louisiana, of this plant in these, the oldest inhabitated areas of Louisiana, and with | es that existed with Spanish Texas, (also being settled at the same time) apparently indicates that it may have been introduced into nai in the same mann same time as it appears it was introduced in Texa In considering the possible ways in which Habranthus tubispathus could be introduced into the United States, natural methods can hls be dis- counted because the pl ant does 1 not appear to possess either | ong or moderate through the digestive tract seems doubtful. 332 The introduction of Habranthus tubispathus appears best explained by human activity. This was first proposed by Sealy (1937) at the suggestion of H. H. Hume of the University of Florida. He proposed that Spanish missionaries transported the plant between 1690 and 1703 during the estab- lishment of the various missions in Texas. It should be noted, as pointed out by Flory (1938), that the only records available concerning the estab- lishment of Spanish missions in Texas cite that they were founded by Spaniards coming out of Mexico and H. texanus (H. tubispathus) has not been reported in that country. This would certainly preclude chance intro- duction from Mexico, but not intentional introduction if the plant were cultivated there and could not become naturalized due to unfavorable con- ditions. The plant's slow rate of spread favors intentional human transport. Documented in Natchitoches some 50 years ago, and probably being present there for a much greater period of time, the plant has failed to spread out- side of the city, as cited above. When its distribution area in the United States of well over 125,000 square miles is considered, it is obvious that either the plant arrived here much earlier than the Europeans (which we feel the evidence does not support), or was a direct, intentional introduc- tion as an ornamental or possibly for some other use not known. today. In no other way could the plant have spread to occupy its present distribu- tion area. We suggest ic was brought into Texas from South America, pos- sibly via Mexico. Once established at one site, it was carried to other settle- ments. Under favorable conditions it escaped, became established, which further hastened its spread. These we feel are the only conditions that explain the plants’ present distribution in the United States. It appears to be only a matter of time before the plant spreads to other favorable habitats in Louisiana and possibly the southeast United States. ACKNOWLEDGEMENTS We wish to thank the curators of the hebaria for their loan of specimens and Sidney McDaniel of the Institute for Botanical Exploration for his review and comments on the manuscript. Appreciation is also extended to W. S. Flory of Wake Forest University and Charles Crane of the University of Missouri, Columbia, for their suggestions and comments. REFERENCES ADAMS, C. D. 1972. Flowering plants of Jamaica. Univ. West Indies Press, p. 8-79 ALEXANDER, E. J. 1939. Habranthus andersonii. Native of Texas, Argentina, Uruguay, and Chile. Addisonia 22: 45-46. CORRELL, D. C. and M. C. JOHNSTON. 1970. Manual of the vascular plants of Texas, nae Research Foundation, Renner, Texas. p. 418-419. CRANE, C. F. Apomixis anc crossing mi eue a in some Zephyrantheae. Un- pubeoned oa Dissertation, University of Texas, Austin, 1978. RMON, C. 1934. Wildflower of eee Doudleday, Doran, & Co. Garden City, New York. p. jos 222 FLORY, W. S. 1938. Cytotaxonomic notes on the genus Habranthus. Herbertia 5: 51-153 GRISEBACH, A. H. R. ies Flora of the British West Indian Islands, Lovell Reeve and Co., London p. 584 MOSCOSO, R. M. 1943. Calon Floraes Domingensis. p. 83. PACE, L. 1913. Apogamy in Atamasco. Bot. ees 56: 376-396. PULLE, A. 1966. Flora of Surinam, Vol. I, Pt. 1. E. S. Brill, Leiden, p. 444-445 REVENNA, P. F. 1970. Contibutions to the ee American oe ia Plant Life 26: 73-103. SEALY, J. R. 1937. se Pyrolirion, Habranthus, and Hippeastrum, Journ. Royal Hort. Soc. 63: 195- URBAN, I. 1903. Symb. fae 4(1): THE MIGRATION AND ESTABLISHMENT OF JUNCUS GERARDII (JUNCACEAE) IN THE INTERIOR OF NORTH AMERICA RONALD L. STUCKEY Department of Botany, College of Biological Sciences, The Ohio State University, Columbus 43210 Juncus gerardii Loisel (Juncaceae), a member of the rush family occurs along the Atlantic and Pacific coasts of North America and locally inland on the continent. On the Atlantic coast, the plant is often a dominant mem- ber of the upper littoral zone where it may form extensive stands in. salt marsh meadows at the upper limit of ordinary tides (Nichols, 1920). Ap- parently once confined to the coastal salt marshes, J. gerardii has evidently migrated inland where it has been reported from ten midwestern and Great Lakes states and the provinces of Manitoba and Ontario in the past 120 years. More recently it has invaded Colorado and Utah in western United States. This paper presents the documentation for the migration and estab- lishment of J. gerardii as isolated occurrences in the interior of North America. The progressive range extension over a period of approximately 120 years and the present known North American distribution of J. gerardii based on herbarium data and published accounts are mapped in figure 1. Although a true rush, J. gerardii is commonly called black grass. Other common names that have been applied to the species are given in Table 1. In earlier North American scientific literature, the species was reported under the name J. bulbosus L. EAST COAST DISTRIBUTION AND EARLY OCCURRENCES IN THE INTERIOR In eastern North America, the apparent native distribution of J. gerardii is in the Atlantic coastal salt marshes. Its range is usually stated as extend- ing from Newfoundland, Quebec, Nova Scotia, New Brunswick, the coastal states of the United States to Virginia and sometimes to Florida. I have verified herbarium specimens of its coastal range from Newfoundland to Virginia. However, the southern limit of its Atlantic coastal range is not clear from the literature. In his classic revision of the North American species of Juncus, Engelmann (1866) reported the southern limit of J. gerardit as North Carolina and Florida and cited as authorities, Curtis and Ware, respectively. This range to Florida has been repeated by Fassett (1940), Fernald (1950), Muenscher (1944), Small (1933), Wiegand (1900), and mapped by Hultén (1958). However, its range is given as far south as Virginia in Gleason (1952), Gleason and Cronquist (1963), SIDA 8(4): 334-347. 1980. ope) TABLE 1. COMMON NAMES OF Juncus gerardii LOISEL OTHER THAN BLACK GRASS REGION OF THE COMMON NAME UNITED STATES WHERE USED REFERENCES Black-grass Rush Florida Ward (1968) Gerard’s Rush New York; Hanmer (1940); Ohio Schaffner (1914, 1932) Hog-rush Southeastern United States Small (1933) Nut grass Maryland McAtee (1933) Salt-marsh Rush New England Seymour (1969a) Saltmeadow Rush British Columbia Taylor and MacBryde (1977) and only as far south as Delaware by Potter (1932). It is not reported for the Carolinas by Radford, Ahles, and Bell (1968) or from North Carolina by Beal (1977), but is listed for Florida by Ward (1968). According to Eleuterius (1976), J. gerardii is replaced by J. roemerianus beginning in New Jersey and extending southward in the east coastal salt marshes to Florida and on the Gulf Coastal Plain to Texas. What is believed to be the earliest record of occurrence of J. gerardit inland in the United States appears in the North American Botany by Eaton and Wright (1840), where they reported the species for Michigan. No specimen has been located to verify this record. A second inland occur- rence in the Great Lakes region was reported as “near Chicago .. . Vasey,’ by Engelmann (1866). This second record was apparently a on specimens obtained in 1862 in vacant lots at Chicago, Illinois, by George Vasey (GH, US). Vasey’s specimens substantiate the statement in the 5th edition of Gray’s Manual that J. gerardii was “rare along the Great Lakes” (Gray, 1867). The previous edition of the Manwal noted that the species was “common along the coast from New Jersey northward” with no mention of locations in the interior on the continent (Gray, 1863). Gray’s statement in the Sth edition of the Manual served as the basis for the record cited in three early catalogues of the flora of Michigan (Wheeler and Smith, 1881, Beal and Wheeler, 1892; Beal, 1905), rather than the citation of the pee in Eaton and Wright (1840). The beginning of the inland spread of J. gerardii is documented in New Jersey and Pennsylvania where the species was first reported as “Not fre- quent” from a specimen obtained in 1866 on ballast ground at Petty’s Isl land Be near Philadelphia in the Delaware River (Smith, 1867). Herbarium 2a mens (MO, PENN, PH, US) from ballast habitats in Camden and Phila- delphia date from 1863 to 1876 and document this early introduction. The species has spread northward into Leigh, Montgomery, and Northampton Counties, Pennsylvania, where it its known from alluvial flats and wet meadows along railroads, as documented by specimens dating from 1924 into the 1960’s (PENN, PH). Schaeffer (1949) reported that the popula- tions in Northampton County “probably represent plants which were intro- duced by salt hay.” ESTABLISHMENT IN THE MIDWESTERN AND GREAT LAKES STATES Expans‘on of the range of J. gerardii in New York evidently came in the mid-to late-nineteenth century. Torrey (1824) gave its distribution in that state as “common in the vicinity of New-York,” and later Torrey (1843) wrote “Borders of creeks and ditches in salt marshes: common on Long Island and Staten Island.” Gray (1863) in the 4th edition of his Manual noted its distribution as “Salt marshes; common along the coast from New Jersey northward.” However, in the 5th edition, Gray (1867) first noted it as “also in saline marshes of W. New York.” This apparent range extension was probably based on the specimens obtained in 186 and 1865 at Salina, Onondago County, by George W. Clinton (MO, PH) which also was the basis for the citation in Engelmann’s revision (1866). In a catalogue of plants of that a Paine (1865) noted J. gerardi from “Rivuletsides and wet banks, east Salina and west of Onondaga Lake, in thick patches. Common there.” ees years later, the species was listed as “sparingly, on the brackish marshes E. of Montezuma” in the Cayuga Lake Basin in western New York (Dudley, 1886). In central New York, additional locations are known from Fall Creek, Myers Point, the Union Springs (Wiegand and Eames, 1926), at Silver Springs near Gaines- ville in Wyoming County, and at Greigsville near York in Livingston County (Zenkert, 1934). At the Silver Springs locality, J. gerardii was noted on 30 August 1926 as common in a number of places along Wolf Creek where salt and brine from the Remington Sale Factory was being dumped into the stream (Muenscher, 1927a, b). Additional localities in central and Figure 1. Distribution of Juncus gerardit in North America, with emphasis on its spread into the interior of the continent. Large symbols show locations of cited speci- men records and periods of ae they were obtained, according to the following key: Triangles in circles (1860-1 884), squares in circles (1885-1919), dots in circles (1920-1949), stars in ioe Ae 1979). Small symbols indicate the general range of the species along the coast. Dots show the apparent native range along the east coast; squares show the ns on the west coast. In addition to herbarium records, the east coast range is derived from maps by ee (1960), Hultén (1958), Potter (1932), Roland (1945), and Rousseau (1974 338 western New York include Hamburg, Ithaca, Lansing, McLean, Piffard, Retsof, Springport, and Syracuse, as derived from the list of cited specimens. The species has apparently spread in the man-created saline habitats of central and western New York, although some writers have doubted its being non-indigenous to the area (Wiegand and Eames, 1926). A century later, J. gerardii continues to persist in Onondago County, the same county where it was initially discovered in the interior of the state of New York (Faust, 1961). In the eastern part of the state, the species has been located on rocks in the bed of Kinderhook Creek at Stuyuesant Falls in Columbia County (McVaugh, 1958) and in moist depressions on dredgings from the Hudson River at Rensselaer (BH, CU) indicating that it had moved inland along the Hudson River in the 1930's. In Ohio, J. gerardii has always been considered native, including the most recent treatment by Braun (1967), although she allowed that the species was “pethaps introduced along railroad tracks.” First recorded by Beardslee (1874) as “rare” from “Lake [Erie] Shore,” this record was repeated by Kellerman and Werner (1893). The earliest known specimens are those obtained in 1891 and 1895 at Cleveland by Edo Claassen (OS), which served as the basis for the record in his Cuyahoga County list of plants (Claassen, 1895). These Cuyahoga County specimens also served as the basis for Schaffner’s only record in his two catalogs of Ohio vascular plants (Schaffner, 1914, 1932). A second record was taken 34 years later at Phalanx in Trumbull ene by Almon N. Rood (OS). The plants were “abundant in low, wet swale,” but were erroneously reported as J. greenei (Schaffner, 1933), and later coueeiel (Schaffner, 1935). Additional specimens obtained by Rood in 1936 were labeled as “Phlanax Station, wet swale by R.R., supporting the hypothesis that the species had probably invaded via thé railway. More recent records have come from along a railroad at Crestline in Crawford County by Floyd Bartley (OS) and from a ditch at the Inter- national Salt Company at the mouth of the Cuyahoga River in Cuyahoga County by Allison Cusick (KE). The only known ane of J. gerardii for Michigan are those obtained by C. K. Dodge in 1911 and 1915 at Port Huron in St. Clair County and y O. A. Farwell in 1931 and 1932 at Wayne in Wayne County (Voss, 1972). The 1911 specimens came from railroad yards. The 1931 plants were from wet ditches, and as reported by Farwell (1941) were then be- lieved to be the first record for Michigan. Voss (1972) has stated that this species is “quite probably not indigenous in Michigan.” In Indiana, Peattie (1930) first reported J. gerardii as very rare from brackish wet sand in Indiana Harbor, Lake County, but attempts to locate a specimen from this locality have been unsuccessful. In 1935, however, the species was first found to be well established on dry open ground along the Nickle Plate Railroad four miles east of Kokomo, Howard County by C. M, Ek (Hermann, 1938, 1940). In 1942, the species was obtained farther 52) north in the next county, Miami County, along the Pennsylvania Railroad, three miles west of Bunker Hill by C. M. Ek (GH, MO) Vassey’s record of J. gerardii from near Chicago has long been disputed. Jones and Fuller (1955) excluded it from the flora of Illinois, because the species was “not attributed to Illinois, or indeed even mentioned, in any of Vasey’s published reports on the flora of Illinois, . . .” Their conclusion was accepted until more recent studies have verified its occurrence in the Chicago area, based on specimens collected in 1900 and 1956 (DeFilipps, 1964, 1966; Mohlenbrock, 1970; Swink, 1969, 1974). KNOWN RECORDS FROM OTHER STATES VERMONT: First reported “on the margin of a marshy meadow very near the railway station at New Haven Junction: where it was believed to have been imported in “saltmarsh hay often used in packing” (Eggleston, 1904). Not reported earlier in the flora of Vermont (Brainerd, Jones and Eggleston, 1900), nor in the most recent flora of Vermont (Seymour, 1969b). WISCONSIN: Reported from near railroad yards in Milwaukee and She- boygan Counties (McIntosh, 1950), based on specimens obtained in the 1930's, and in Racine County as early as 1900 (Swink, 1969, 1974). The most recent specimen, taken in 1977, is from a wet area along railroad tracks in Fox Point, Milwaukee County (OSH). MINNESOTA: Stated by Hultén (1958) as a weed along railways in Minnesota, with a mapped location in the southern part of the state evi- dently based on the record from Martin County obtained in 1950. The species is also cited for Clay and Kittson Counties (McGregor and Barkley, 1977) based on records from the 1960's. The specimen from Clay County is J. alpinus (shoulder of roadside ditch, Muskoda, 8 Jul 1960, O. A. Stevens 2231, DAO, US), and the specimen from Kittson County is J. compressus (roadside at Caribou, 28 Jun 1962, J. W. Moore 26083, TRT). NORTH DAKOTA: Listed for Cass and Richland Counties (Stevens, 1961; McGregor and Barkley, 1977). The record from the latter county is dated C MISSOURI: Observed during the years 1964 through 1966 in the southern part of the Ranken yard of the Terminal Railroad Association, east of Compton Avenue, in St. Louis (Muehlenbach, 1969). KENTUCKY: Obtained in 1978. This specimen is the most recently known record and is from a population of plants in diesel sludge mud of railroad yard near tracks, near junction of West 19th Street and Augusta Street, Cov- ington, Kenton County (OS). Selected Specimens: ILLINOIS: [Cook Co.}: Vacant lots, Chicago, 1862, Vasey 5.n. GH, US); railroad and roadside, Chicago, 1956, Glassman 3788 (CHI, ILLS, NP). INDIANA: Howard Co.: Railroad, Kokomo, 1935, E& s.n. (GH, IND), 1936, Deam 57045 (NY). Miami Ca Railroad, Bunker Hill, 1942, Ek s.n. (GH, MO). KEN- TUCKY: Kenton Co.: Railroad yard, Covington, 1978, Applegarth 465 (OS). MICHI- 340 GAN: [St. Clair Co.]: Railroad freight yards, Tunnel Station, 1911, Dodge s.n., (MICH); damp ground, Port Huron, 1915, Dodge s.n. (US). [Wayne Co.}: Diche: Wayne 1931, Farwell sn. (GH), Farwell 8897 (MICH). MINNESOTA: Martin Co.: Railroad, oe 1950, Moore 20689 (GH, MIN). Mrssour!: St. Louis Co.: Rail- road, St. Louis, 1964, Muehlenbach 2356 (MO, US): 1965, Muehlenbach 2505, 2522, 2476 (MO); 1966, Muehlenbach 2625 (MO). NEW JERSEY: [Camden Co.} : Camden, 1863, Canby s.n. (MO): ballast. Petty’s Island in Delaware River, 1866, Burk s.n. (MO), Diffenbach s.n. (MO, PH), Parker s.n. (GH, MO, PH); ballast, Kaighns Point, 1876, Martindale sn. (PH, US). NEW YorK: Cayuga Co.: Brackish meadow east of Salt Creek, Montezuma, 1885, Dudley sn. (CU); brackish place by railroad sta- tion, Union Springs, Springport, 1919, Eames & Wiegand 11730 (CU, GH); salty meadows, Montezuma, 1919, Wiegand, Randolph, & Eames 11729 (CU). Columbia Co.: Bank of creck, Stuyvesant Falls, 1933, McVaugh 997 (NY, PENN); rocks in Kinder Creek, Stuyvesant Falls, 1945, McVaugh 3126 (CU). Erie Co.: Meadow, Hamburg, 1939, Glowenke 3868 (PENN). Livingston Co.: Ditches along roadside, salt factory, Piffard, 1926, Muenscher 16495 (CU, GH); stream bank, salt factory, Retsof, 1926, Muenscher 16495 (CU). [Onondaga Co.}: Salina, 1864, Clinton s.n. (MO), 1865 (PH); salt marshes, Syracuse, 1880, paar 2973 axe salt marshes, Onondaga Lake, 1883, Habertr 954 (CAN, US), (MO). Rensselaer Co.: Moist depressions in pees Hudson River, Sees 1939, House 27070 (BH, CU). Tompkins Co.: Railroad, Myers, Lansing, 1918, Eames 9559 (CU, GH); railroad ditch, McLean, 1920, Wiegand & Muenscher 13448 (CU, GH). Wyoming Co.: Salt Sek Silver Cee 1926, Muenscher & Burkholder 16498 (CU, GH). NoRTH a KOTA: Cass Co.: Railroad, Fargo, 1954, Stevens 1497 (DAO, NY, US). Richland ane soil, Colfax, ie Ball 694 (US). OHIO: Crawford Co.: Railroad, Crest- me 1960, Bartley s.n. (NY, OS). Cuyahoga Co.: Cleveland, 1891, 1895, Claassen sm. (OS); ditch, Tatecnioaal Salt Company, mouth of Cuyahoga River, Cleveland, oe Cusick 13723 (KS). Trumbull Co.: Wet swale, Phalanx Station, ee Rood . (OS); wet swale, railroad, Phalanx Station, 1936, Rood 1077 (KS, NY, P N). eee LVANIA: Lackawanna Co.: Roadside ditch, Carbondale, 1946, ira 6812 (GH, PENN). Leigh Co.: Raileond. Laury’s Station, 1924, Pretz 12155 (PH); meadow, Allentown, 1960, Schaeffer 6095 (PH). a es Co.: Railroad, Hat- boro, 1943, oe 59871 (PH). Northampton Co.: adow, Butztown, 1946, Schaeffer 22738 (PH); railroad, Northampton, 1947, oe 25947 (PH); shore, Siegried, 1961, haeffer 63226 (PH). {Philadelphia Co.]: — ground below the navy yard, Phil piskg 1866, Parker Ly Aas PH). VERMONT: [Addison Co.}: Railroad, New Haven Junction, 1897, B era sn. (GH). ae NSIN: [Racine Co.}: Lake beach, i habor, 1900, Wadmond 2740 (CU, MIN, WIS). Milwaukee Co.: Railroad Estabrook Park, Shorewood, 1939, Shinners 1069 (PENN, WIS); railroad, Fox Point, 1977, Harriman 14125 (OSH). Sheboygan Co.: Railroad yard, Sheboygan, 1933, Goess! s.n. (WIS). o DISTRIBUTION IN THE INTERIOR OF CANADA In Canada, J. gerardii was early reported by Macoun (1888) from only coastal marshes in Nova Scotia, New Brunswick, Quebec, and Vancouver Island. In the interior of Canada, in Ontario, J. gerardii was not included among the introduced plants of that province (Montgomery, 1956). How- ever, the species has been known from the province as early as 1903, based on specimens obtained by William Scott (CAN, CU, DAO, TRT) from a railroad roundhouse at Niagra Falls. J. gerardii has been located in Ontario at other isolated sites along railways, usually at stations were salt has been 341 sprinkled to melt snow in the winter. The most recent records seen are from 1974, 1975, and 1976, (CAN, DAO, TRT). In the Toronto region, Catling and McKay (1975) have described the plant communities at two saline sites, one a salt storage depot on the west side of the Don River near the Bloor Street viaduct and the other a snow dumping site on land- fill at the foot of Leslie Street, where J. gerardii is plentiful with other halophytes. Boivin (1952) reported the species as a weed along railway embankments at Cochrane et Hope i in Ontario and at Brandon in Manitoba. Selected Specimens: MANITOBA: Weed along railway embankments, Brandon (cited by Boivin, 1952). ONTARIO: Carleton Co.: Slightly saline soil, at Mississippi River bridge east of Arnprior, 27 miles west of Ottawa, 1959, Dore & Cody 17425 (MSC); river, Rockcliff Park, Ottawa, 1947, Calder, Dore, & Cody 1641 (MO). [Cochrane Co.}: Railway yards, ae 1952. Baldwin 3912 (CAN, GH). Durham Co.: Abandoned railway, Hope Township, 1948, Reeve 87 (DAO). Essex Co.: 1913, Malte sn. (CAN); oe open saline ground, Windsor o Factory, Windsor, 1975, Catling & McKay sn. (CAN, TRT). Huron Co.: Saline soil, south side of Maitland River, below Sifto Salt beta operation, Goderich, 1975, Catling & McKay 5.n. (CAN, TRT). Welland Co.: Railroad peondaouse: Niagra Falls, 1903, Scott s.n. (CAN, CU, DAO, TRT); Montrose railway yard, Niagra Falls, 1976, Catling & Mcintosh s.n. (DAO, TRT); ditch between two rail embankments, Bertie Township, Niagra, 1976, Catling & Riley sm. (DAO, TRT). York Co.: Dump heap, Ashbridge’s Bay, Toronto, 1904, Scott s.n. (CU, TRT); wet depression receiving run off from rock salt storage depot, opposite Don Valley Brickyard, Don Walley, Toronto, 197 4, Catling & McKay sn. (CAN, TRT); in snow dumping area, foot of Leslie Street, Toronto, 1974, Catling & McKay s.n. (TRT). DISTRIBUTION IN WESTERN NORTH AMERICA On the west coast, J. gerardii is known from the coastal salt marshes in the Puget Sound area in the state of Washington, on Vancouver Island, and in southern British Columbia. Some question has been raised as to whether or not the species is native there because the morphology of the plants more closely resemble those from Europe, whence they may have been intro- duced at an early date (Hitchcock, Cronquist, and Ownbey, 1969). The species was first reported from salt marshes at Victoria, Vancouver Island by Macoun (1838), based on his collection of 26 Jun 1887 (GH). How- ever, the species was not included in the Flora of Washington by Piper (1906). The earliest known specimen from Washington is dated 1917 and comes from Lopez on San Juan Island where it was obtained by S. M. and Mrs. E. B. Zeller (GH). Additional records were added in the 1920's and 1930’s. Muenscher (1941) wrote the J. gerardii was local in salt marshes and cited two localities, South Bellingham and Marietta in Whatcom County. St. John (1928) considered the plants on the west coast as a native species and proposed the name J. fucensts. Inland in the Rocky Mountains of western United States, J. gerardii has been located in two states: 342 RADO: Known from wetlands at one site in Adams County and two locations in Boulder County, as reported by Hermann (1975). The earliest record is from the latter county, dated 1952. UTAH: Reported as he reg and established in salt marshes near Salt Lake City” (Tidestrom, 1925). This record is apparently based on a speci- men from a salt grass pasture near Salt Lake Say (Jun bree, G. QO. Bateman 2., US). The species was not included in the Intermountain Flora by Cron- ae (1977): SOME GENERAL CONSIDERATIONS AND CORRECTIONS It has become apparent that the spread of J. gerardii inlard into the Great Lakes region and midwestern United States has come about by either migra- tion of the species via railroads or by establishment in inland saline marshes, as documented in the above specimen citations and summarized in Table 2. f the 58 known inland stations east of the Great Plains, 26 of them (44% ) are from along railroads and 13 of them (22%) are from inland saline habi- tats in New York, Ohio, and Ontario. Most, if not all, of the inland saline habitats cited here have been artificially created since the coming of Euro- pean man. Other significant localities are man-made roadside ditches and ballast ground, which account for the habitat at eight localities (14%). The occurrence in isolated sites and especially in man-created habitats which account for 80% of the known localities, strongly suggests that J. gerardii is a non-indigenous member of the wetland flora of the Great Lakes region, mid-western, and certain western areas in the United States. Furthermore, the occurrence of J. gerardii in the Finger Lakes region of New York, for example, conforms to Svenson’s idea that human agencies have been perhaps the most effective means of distributing these halophytic species into that region, rather than surviving as remnants from a_ post-Pleistocene marine submergence (Svenson, 1927). As pointed out by Eggleston (1904) in Vermont and by Schaeffer (1949) in Pennsylvania, J. gerardii has been transported as a salt hay in packing. In Massachusetts, Deane (1915) noted that black grass was cut in the sum- mer and fall and used for packing on celery beds to keep the stocks from freezing during the winter. The plants also can form “balls” or “pebbles” from matted sods which are thrown upon beaches during ditching opera- tions (Ganong, 1905). These sods are rolled about, worn down, and rounded by action of the waves and may even be transported Potter (1932, 1934) cited J. gerardii as an example of a halophytic species whose distribution supported the idea of a post-Pleistocene marine connec- tion between Hudson Bay and the St. Lawrence River basin. Those species which he used as evidence occurred in the southern region of Hudson Bay at James Bay, were disjunct to the Gulf of St. Lawrence, and then ranged southward along the Atlantic coast. J. gerardii, however, showed certain irregularities by having isolated occurrences in the Finger Lakes region of 343 TABLE 2. SUMMARY OF INLAND HABITAT TYPES OF J. gerardii IN RELATION TO THE NUMBER OF LOCALITIES IN THE MIDWESTERN AND GREAT LAKES STATES AND IN MANITOBA AND ONTARIO HABITAT NUMBER PERCENTAGE OF TOTAL OF LOCALITIES NUMBER OF LOCALITIES Railroad 26 _ : 4A Salt marsh; saline or brackish marsh 13 22 Roadside ditch 4 7 Ballast ground and dredgings 4 7 Other habitats 11 20 Total 58 100% New York, at the southern end of Lake Michigan, and on the Pacific coast. This irregularity has now been explained. Earlier, Peattie (1922) had cor- rectly omitted J. gerardii as an example of a natural member of the Atlantic Coastal Plain element in the Great Lakes region. Moreover, as reported and mapped by Rousseau (1974), J. gerardii is not known at James Bay, and he has stated that Potter’s specimen at the Gray Herbarium is J. alpinus, as determined by L. Cingq-Mars. The report from a James Bay by Macoun (1888), which Rousseau could not verify, apparently is from a local James Bay at Vancouver Island, British Columbia. Macoun’s published report is clear, “Salt marsh at the head of James Bay, Victoria, Vancouver Island, 1887. (Macoun.),” and agrees with the date on the label of his specimen (24 June 1887, Macown 27,869, CAN). ACKNOWLEDGEMENTS My thanks are extended to Mr. Mark Davis who assisted with this research during the summer of 1977. My thanks are also extended to the curators of the herbaria who made specimens and records available. Those herbaria in which I have examined specimens are: BH, CAN, : : : ; MO, MSC, NYS, OS, OSH, PENN, PH, US, Those herbaria from which only records were seen are: GH, MIN, NY, TRT, WIS. REFERENCES BEAL, E. O. 1977. A manual of marsh and aquatic vascular plants of North Carolina with Sagan data. North Carolina Agric. Exp. Sta., Raleigh. Tech. Bull. No. 247. 29 BEAL, W. J. 1905. es ae Rep. Michigan Acad. (1904) 5: 1-147. ane ewe . F. WHEELER. 1892. Michigan Flora. Ann. a State Board Agric. Michigan (1891) 30: 471- Oo. (Reprinted and repaged 1-180 344 BEARDSLEE, H. C. 1874. Catalogue of the plants of Ohio, including flowering plants, ferns, mosses and liverworts. Painesville, Ohio. 19 pp. (Reprinted Ann. Rep. Ohio State Board mee (1877) 32 . ooe 363, Tous: Rhodora 54: een JONES, and W. W. EGGLESTON. 1900. Flora of Vermont. A lst of the ere and seed plants growing . cultivation. Vermont Bot. Club, Contrib. Bot. Vermont, VHI. Burlington. BRAUN, E. L. 1967. The Monocotyledoneae tof ol uo}. Cat-tails to Orchids. With the Gramineae by Clara G. Weishaupt. The Ohio State Univ. Press, Columbus. CATLING, P. M. and S. M. McKAY. 1975. oo of halophytic plants in the Toronto region. Ontario Field Biol. 29(1): 50-55. CLAASSEN, E. 1895. List of monocssdonous oe vascular-cryptogamous plants of Cuyahoga ae Ohio. 3rd Ann. Rep. Ohio State Acad. Sci. 26- CRONQUIST, A. 1977. Juncaceae, pp 47-67. Im Arthur Cronquist, Apc H. Holm- gren, James L. ae and Patricia K. Holmgren. Intermountain Flora: plants of the Intermountain 8 U. S. niv. Press, New York. 584 DEANE, W. 1915. Floral ahs in a salt marsh during reclamation. Rhodora 17: 203-22??. DE FILIPPS, R. A. 1964. A taxonomic study of Jwncews in Illinois. Amer. Midl. Naturalist 71: 296-31 . 1966. Dist bation of Juncus in Illinois. Phytologia 13: 44-64. DUDLEY, W. R. 1886. The Cayuga Flora. Part I: A catalogue of a phaenogamia growing without cultivation in the Cayuga Lake Basin. ll ve home Ze I-X XX, 1-32 + Additions and Corrections, Index to Orders and ae i-v. EATON, A. and J. WRIGHT. 1840. North American botany, comprising ae native and os cultivated plants, north of Mexico: Genera arranged eae to the artificial and natural ener 8th ed. Elian Gates, Troy, New York. 62 er W. W. 1904. Addenda to the flora of Vermanae Rhodora 6: 139-144, ELEUTERIUS, L. N. 1976. The distribution of Jancus roemerianus in the salt marshes of North America. Chesapeake Sci. 17: ode ENGELMANN, G. 1866. A revision of the rth American species of the a Juncus, with a description of new or eres known species. Trans. St. Loui Acad. Sci. 2: 424-498. ERSKINE, D. S. 1960. The plants of Ae ae Island. Plant Res. Inst. Res. Branch, Canad. Dept. Agric. Publ. 1988. 27 FARWELL, O. A. 1941. Notes on the ae Pee. VIII. Pap. Michigan Acad. Sci. (1940) 26: 3-20. FASSETT, N. C. 1940. A manual of aquatic See) McGraw Hill Book Co., New York. 382 pp. (Reprinted an revision appen ugene gden. Univ. Wisconsin Press, Madison. 405 pp. 1957, 1960, 1966, 1969, 1972, ‘1975. FAUST, M. E. 1961. Checklist of he vasculac plants of Onondaga County, New York. Bull. ee a Nat. Sci. No. 9. 83 FERNALD, M. L. 1950. Gray’s manual of earn 8th ed. American Book Co., Ixiv + 1632 pp. (Reprinted by D. Van Nostrand Co., New York. 1970). GANONG, W. F. 1905. On balls of vegetable matter fron sandy shores. Rhodora 7: 41-47 GLEASON, A. 1952. The new Britton and Brown illustrated flora of the north- eastern ae States and adjacent Canada. Vol. I. New York Bot. Gard., New York. ae bP Vascular A. Vol. Six The Monocotyledons. Columbia Inc. CRONQUIST. 1963. Manual of vascular plants of ie United nee ane adjacent Canada. D. Van Nostrand Co., Inc., New York. 345 GRAY, A. 1863. Manual of botany of the northern United States, including ace Kentucky, and all a of the Mississippi. 4th ed. Ivision, Phinney, & Co. York. 743 pp. + X Is. 1867. Mea of the botany of the northern United ee including the disteiee « east of the Mississippi and north of North Carolina and essee, arranged according to the natural system. 5th ed. Ivison, Blakeman, Taylor ee ee. New York. 703 pp. + XX pl. HAMMER, C. C. 1540. Plants of Fishers Island. Torreya 40: 65— HERMANN, F. J. 1938. New or otherwise interesting plants ee eae Rhodora 40: 77-83. «1940. Juncus, pp. 290-302. In Charles C. Deam. Flora of Indiana. Dept. ic: nservation, Div. Forestry, i 1236 pp. (Reprinted in the series, Re- 197 prints of i Floras, Vol. 6. Verlag von J. Cramer, Lehre. : oe an the mae (Juncus spp.) of the Rocky Mountains and alse an U. S$. A. Forest Service Gen. Tech. Rep. RM-1 HITCHCOCK, C. L., A. ‘CRONOUIST and M. OWNBEY. 1969. Vascular piaats of the Pacific Nonhwest Part 1: vascular cryptogams, gymnosperms, and monocotyle- dons. Univ. Washington Press, Seattle. 914 pp. HULTEN, E. 1958. The Amphi-Atlantic plants and their eee connec- tions. Kungl. Svenska Vetenskapsakademiens Handlingar Fjarde Serien. Band 7. Nr. 1. 340 pp. fea Otto Koeltz a Sa Taunus/B. R. D .) d KELLERMAN, W. A. and W. C. WERNER. “1893” [1894]. Catalogue of Ohio plants. Rep. Geol. Surv. Ohio 7: 56-406. MACOUN, John. 1888. Catalogue of Canadian plants. Part IV—Endogens. Geol. & Nat. Hist. Soc. Canada. 2(4): 1-248. McATEE, W. L. 1933. Some oe names of plants V. Torreya 33: 81-86. McGREGOR, R. L. and T. M. BARKLEY, eds. ion Atlas of the of the Great Plains. Iowa es Univ. Press, ee 600 pp. McINTOSH, J. A. 1950. ose reports on the flora of Wisconsin. XXXIV. Liliales. Trans. Wisconsin Acad. Sci. 40: 215-242. McVAUGH, R. 1958. Flora of os ee County area, New York. New York State Mus. & Sci. Serv. Bull. No. 360. 400 MOHLENBROCK, R. H. 1970. The iUliserared flora of Illinois: flowering plants, flowering me to rushes. Soudhers Illinois Univ. Press, Carbondale and Edwards- ville. 272 MONTGOMERY, F. H. 1956. The ae plans E ee growing outside of cultivation (Part I). Trans. Roy. .31: 91- MEUHLENBACH, V. 1969. ee es new to ne ae flora (III). Ann. Missouri Bot. Gard. 56: 163-171. MUENCHER, W. C. 1927a. Appendix VII. Salt plants of Wolf Creek, p. 87. In A biological survey of the Genesee River system state of New York Conservation oe Supplemental to Sixteenth Annual Report, 1926. Albany. 100 pp. + 5S maps. __ 1927b. Spartina patens | other saline plants in the Genesee valley of western iene York. Rhodora 29: 138-139. _ 1941. The flora of oo County state of Washington. Published by the he-gathon Ithaca. 134 pp. + Index. 1944. Aquatic Plants of the United ae Comstock Publishing Co., Ithaca, oe York. 374 pp. (Reprinted aa 1964, 7] Jes NICHOLS, G. E. 1920. The vegetation of Sein VII. The associations of de- positing areas along the seacoast. Bull. Torrey Bot. Club 47: 511-548. 346 PAINE, J. A., Jr. 1865. oe i. as found in Oneida County and vicinity. Rep. Recents Univ. New-York. PEATTIE, D. C. 1930. Flora of fee iana dunes: A handbook of the flowering plants and ferns of the Lake Michigan coast of Indiana and of the Calumet Dis- trict. Field Mus. Nat. Hist., Chicago. 432 1922. The Atlantic coastal plain element in the flora of the Great Lakes. Rhodota 2 24: 57-70, 80-88 PIPER, C. 1906. Flora of the state of Washington. Contrib. U. S. Natl. Herb. 11: 657, POTTER, D. 1932. Botanical evidence of post-Pleistocene marine connection between eee Bay and the St. Lawrence basin. Rhodora 34: 69-89, 101-112. (Reprinted Contrib. ie) Herb. Harvard Univ. 99.) 934. Plants collected in the southern portion of James Bay. Rhodora 36: a 132 Qu 27 4-284, RADFORD. A. E., H. E. AHLES and C. R. BELL. 1968. Manual of the vascular flora of tht Carolinas. Univ. North Carolina Press, Chapel Hill. 1183 ROLAND, A. E. “1945” [1947]. The flora of Nova Scotia. Proc. & Tins Scotian Inst. Sci. 21: 95-64 127 figs. 1948. ROUSSEAU, C. 1974. Geoetaphic floristique du Quebec-Labrador: principales especes Lary ies. Les Presses de l'Université Laval, Queb SCHAEFFER, R. 1949. The vascular flora of Northampton Conny, Pennsyl- ani: aie in botany presented to the faculty of the graduate school in partial fulfillment of the oe for the degree of docor of philosophy, Univ. cree ee Iphia., 702 SCHAFFNER 1914. Catalog a Ohio vascular plants. Ohio Biol. Surv. 1; 125— 247. Bull 2. - 1932. Revised catalog of Ohio vascular plants. Ohio Biol. Surv. 5: 87-215. Nova 2. (Reprinted Truro Printing & Publishing Co., Truro. cm ibution des 98 pp + 1933. Additions to the revised catalog of Ohio vascular plants. I. Ohio J. Sci. 33: 288-294. 1935, Additions to the revised catalog of Ohio vascular plants. HI. Ohio J. Sci. 35: 297-303. SEYMOUR, F. C. 1969a. The Flora of New England. A manual for the identification of all vascular plants, including fern and fern allies and flowering plants growing without cultivation in New England. Charles E. Tuttle Co., Rutland, Vermont. xvi + 596 pp. : . 1969b. The Flora of Vermont: A manual for the identification of ferns and flowering plants growing without ee in Vermont. 4th ed. Agric. Exp. Sta. Bull. 660. Univ. Vermont, Burlin 393 pp SMALL, J. K. 1933. eee of sg ees flora. Univ. North Carolina Press, Chapel ae xxi + 15 SMITH, A. On ae of aoe observed near eager Proc. Acad. Nat. Sci. Sea 19: 15-24 (Reprinted in R. L. Stuckey, ed. Essays on North oe — geography from i Bees century. Arno ee Inc., New York. ST. Se 1928. New plants from British Columbia. Rep. Pro . Nat. Hist. British Columbia 1927; E14 (Reprinted Contrib. Washington ae Coll. Bot. Dept. a STEVENS, O. A. 1961. New records for North Dakota. Seer 63: 39-4 SVENSON, H. K. 1927. Effects of post-pleistocene oe subm North America. Rhodora 29: 41-48, 57-72, 87-93, 114. SWINK, F. 1969. Plants of the Chicase region. The ae ae Lisle, Hlinois. Lies In eastern 347 s _. 1974. Plants of the Chicago region. 2nd ed. The Morton Arboretum, Lisle ~ Illinois. 474 pp. TAYLOR, R. L. and B. MacBRYDE. 1977. Vascular plants of British Columbia: A descriptive resource inventory. Bot. Gard. Univ. British Columbia Tech. Bull. No 4. xxiv + 754 pp. TIDESTROM, I. 1925. Flora of Utah and Nevada. Contrib. U. S. Natl. Herb. 25: - 665 + pls. 1-15. (Reprinted in the series, Reprints of US-Floras, Vol. 3. Verlag n J. Cramer, Lehre. TORREY. J. 1824. Flora of the northern and middle sections of the United States; or, A systematic arrangement and description of all the plants hitherto discovered in the United States north of Virginia. T. and S. Swords, New York. _ «1843. A Flora of the state of New-York, comprising full descriptions of all the ‘indigenous and naturalized plants hitherto discovered in the state; with remarks on their economical and medical properties. Vol. II D. Appleton & Co., New York. 572 p VOSS, E. G. 1972. MMichieee Flora: A guide to the ee: fication and occurrence of the native and naturalized seed-plants of the state Part I gymnosperms and monocots. Cranbrook Inst. Sci., ae Hills, Bull. 55, and Univ. Michigan Herb., Ann Arbor. xviii + 488 WARD, D. B. 1968. Checklist of the vase) @y noe of Florida Part 1. Agric. Exp. Sta., Univ. Florida, ne Tech. Bull. 726. 72 pp. WHEELER, C. nd E. F. SMITH. 1881, cues Flora. Ann. Rep. State Hort. Soc. Michigan Eciaan 10: 427-529. (Reprinted with title page and repaged 1-105, as catalogue of the phaenogamous and vascular cryptogamous plants of Michigan, indigenous, a eaees: and adventive. ) WIEGAND, K. M. 1900. aed tenuis Willd. allies. Bull Torrey oe Club 27: 511-527. nd A. J. EAMES. ice The flora has the Cayuga Lake Basin, New York. 1 pp i and some of its North American ~ Cornell Univ. Agric. Exp. Sta. Mem. 92. ZENKERT, C. A. 1934. The flora of the Niagara frontier region. Ferns and flower- ing plants of Buffalo, N.Y., and vicinity. Buffalo Soc. Nat. Sa. Bull. 1 6: 1-x, 1-328. NOTES ON THE FLORA OF THE CHINATI MOUNTAINS, PRESIDIO COUNTY, TEXAS EMILY J. LOTT Herbario Nacional, Instituto de Biologia, UNAM, Apartado Postal 70-233, México 20, D.F., México MARY L. BUTTERWICK Bureau of Land Management, Phoenix District Office, 2929 West Clarendon Avenue, Phoenix, AZ 85017 The first definite records of botanical observations in the Chinati Moun- tains are those of V. Havard (1885). G. C. Nealley (1888) collected grasses and forage plants and made general observations in western Texas, visiting the “Chenate” Mountains in late October, 1887. M. S. Young made general collections in the foothills of the Chinatis in September, 1914, but probably never reached the main mass of the mountains (Tharp 1962). The most extensive botanical explorations of the Chinatis were begun by L. C. Hinck- ley in August, 1936. Collecting trips were made to Chinati Peak in 1945 by C. H. Muller and Rogers McVaugh. B. H. Warnock began working in the area both alone and with Hinckley during the 1940's; specimens from these and subsequent trips are deposited at TEX-LL and SRSC herbaria. The present authors made collecting trips into the area in June and October of 1977. Voucher specimens will be deposited in the University of Texas at Austin Herbaria (TEX-LL). Nomenclature for species previously reported from Texas is in accordance with Correll and Johnston (1970). The rugged summits of the Chinati Mountains rise over 1400 m above the surrounding pediments, dominating the 30 km expanse between Pinto Canyon and Shafter in south central Presidio Co. The Chinati Mountains are one of three major mountain ranges in the Big Bend area of Trans-Pecos Texas that are of igneous origin, the others being the Chisos and Davis Mountains. The considerable erosive force of tributaries to the Rio Grande, such as Pinto and Dead Horse canyons, has uncovered various sedimentary rocks of Permian and Cretaceous age. These localized outcrops of limestone, sandstone, and shale, primarily along the margins of the intrusive and volcanic rocks, are the substrates for plant assemblages that are noticeably different from those found on neighboring igneous slopes. n the limestone hills just N of Chinati Peak, associated with Agave lecheguilla, Dasylirion leiophyllum, Quercus pungens, Mortonia scabrella, Leucophyllum frutescens, Ayenia pilosa, and Bouteloua ramosa, we found two species of particular interest. SIDA 8(4): 348-351. 1980. 349 POLYGALA NUDATA Brand. On limestone hills N of Chinati Peak between Dead Horse and Pinto Canyons; Butterwick 3807B; 9 Jun 1977. This is the fourth known United States locality for this perennial Polygala, which until recently has been placed in synonymy with the Mexican P. minutifolia Rose (Wendt, 1978). Polygala nudata is characterized by its upright broom-like incurved-puberulous stems and short seed pubescence. PETERIA SCOPARIA Gray. Infrequent on limestone hills N of Chinati Peak; Butterwick & Lott 3829; 11 Jun 1977. This species is previously unreported from the Chinatis but its presence there could be expected. It is poorly repre- sented in herbaria and generally occurs in small, scattered populations. Indian Cave Canyon, a side canyon to Dead Horse Canyon, on the N side of the Chinatis, provides a sheltered, more mesic habitat for Prunus serotina, Fraxinus cuspidata, F. arizonica, Quercus gambelii, Q. grisea, Symphoricarpos sp., Heuchera rubescens, Aquilegia sp., and Pericome caudata in its upper reaches, which lead to Chinati Peak. In this canyon we collected a species of fern reported for the first time from the Chinatis. POLYPODIUM THYSSANOLEPIS KI. Uncommon in crevices of cliffs and boulders, lower Indian Cave Canyon, with Garrya ovata, Selaginella rupincola, Bommeria hispida, Eupatorium rothrockii, Ungnadia speciosa, Silene laciniata, Sedum wrightii, and Cheilanthes wrightii; Butterwick & Lott Oct 1977. According to T. L. Wendt,’ who is preparing a treatment of Chihuahuan Desert Polypodium, this collection is of an undescribed variety of the south- western United States and northern Mexico. It was previously known in Texas only from the Davis and Chisos Mountains. The summit of Chinati Peak is surprisingly level, resembling a high mesa top. Stipa tenwissima occurs beneath the numerous gray oaks, and Bouteloua gracilis covers the more open areas. Sedum wrightu, Talinum parviflorum, Heuchera rubescens, Aletes acaulis, and Triodanis biflora are a few herbaceous species collected on or near the peak. OPUNTIA POLYACANTHA Haw. var. TRICHOPHORA (Engelm. & Bigel.) Coult. Locally common among grasses on the summit of Chinati Peak; A. M. Powell et al. 3109; 7 Jun 1977. This collection represents the southernmost known population of this widespread species, which ranges as far north as Peace River, Alberta. The variety occurs in Trans-Pecos Texas and the Pan- handle, to Colorado and Arizona through western Oklahoma. GAURA HEXANDRA Gomez Ortega subsp. GRACILIS (Woot. & Standl.) Raven & Gregory. Rare in rocky igneous soils near summit of Chinati Peak, ca 2313 m elev; Butterwick & Lott 3774; 7 Jun 1977. Scarlet flowers, red- dish stems with white spreading hairs, and truncate fruit bases distinguish this species, a new state record for Texas and an extension of its range 1The undescribed variety of Polypodium mentioned above is now published as P. thyssanolepis Kl. var. riograndense Wendt (Amer. Fern J. 70: 5-11. 1980). 350 from Arizona and New Mexico to Sonora, Chihuahua, and Durango, and now to western Texas. Two large canyons on the S§ side of the Chinati Mountains yielded note- worthy collections. Both are relatively closed drainages whose walls provide some protection from the southwesterly winds. Of particular interest was Tinaja Prieta (Palillos) Canyon, with its numerous tinajas and seep springs. PLUMBAGO SCANDENS L. On low gravelly terrace of lower Tinaja Prieta Canyon, just above falls, and med the upper reaches of the canyon; Butterwick & Lott 3872; 16 Jun 1977. The range in Texas is now extended from Cameron and Hidalgo counties in extreme southern Texas. Known also from Pima Co., southern Arizona, and from northern Mexico and south- ern Florida, this species is widespread in the tropical Americas. SIPHONOGLOSSA LONGIFLORA (Torr.) A. Gray. Rare among boulders in lower part of E branch of San Antonio Canyon, § side of Chinati Mountains; Butterwick & Lott 3864B; 15 Jun 1977. The essentially glabrous S. longiflora has petioled leaves and clusters of white tubular flowers in the upper axils which easily distinguish it from the coarsely pubescent S. pilosella. The Chinati Mountains locality for this species extends its range eastward from southern Arizona and Sonora and is apparently the first documented report of the species from Texas. This species is not included in the Manual of the vascular plants of Texas (Correll & Johnston 1970), nor are any previ- ous Texas collections on deposit in TEX-LL, SRSC or GH. Rickett (1970) cited the species as “reported in western Texas” but stated that he had seen no specimens of it from Texas. ACKNOWLEDGMENTS We thank P. H. Raven for determination of the Gawra specimen and T. L. Wendt for his verification of our pteridophyte and Polygala determi- nations. J. Henrickson helped locate specimens of Siphonoglossa longiflora, and R. A. Hilsenbeck verified our determination and graciously provided useful information on this species. We are grateful to A. M. Powell for reviewing the manuscript, and to D. E. Deal for arranging permission with the landowners. Texas Natural Area Survey is gratefully acknowledged for financial and logistical support of field work in the area. REFERENCES CORRELL, D. S. and M. C. JOHNSTON. 1970. ines of the vascular plants of Texas. Texas Research Foundation, Renner, Texa HAVARD. V. 1885. Report on the flora of western and southern Texas. Proc. U. S. 533. NEALLEY, G. C. 1888. Report of an investigation on the forage plants of western Texas. U. . Dep. Agr. Div. Bor Bull. 6. U. S. Government Printing Office, Wash- ington, RICKETT, - W. 1970. Wild flowers of the United States. Vol. 3: Part 2 (Texas). McGraw-Hill, New York (publication of the New York Botanical Garden) peak THARP, B. C. and C. V. KIELMAN, eds. 1962. Mary S. Young's journal of botanical explorations in Trans-Pecos Texas, August- eis 1914. Southwest. Hist 65: 366-393; 512- a WENDT, T. L. 1978 systematic study of cae. section Rhinotropis (Poly- galaceae). Ph.D. .. University of Texas, Aust NOTES DISTRIBUTION RECORDS FOR DIGITARIA BICORNIS IN EAST- ERN UNITED STATES—Henrard’s Monograph of the genus se (1950) restricted the distribution of Digitaria bicornis (Lam.) R ; tropical crabgrass, to tropical Asia. Swallen (1963) described D. dieters a later synonym of D. bicornis, based on type material from Jamaica and gave the distribution in the United States as Florida and Texas. Gould 5) reported D. bicornis in Texas as occurring in the southeastern prairies and coastal marshes. Correll & Johnston (1970) stated that D. diversiflora (ie. D. bicornis) is probably indigenous to and is common on the Rio Grande Plains. Recent collections of D. bicornis | have made in eastern United States have shown that this taxon is much more widespread than previously be- lived. It was found to be common and abundant on the sandy coastal plain of the southeastern states. The range of the species is now known to include eastern North and South Carolina, and eastern Texas (Fig. 1). The distribu- tion of this species seems to be limited to areas of coarse-textured soils and moderate to high rainfall. Populations of D. bicornis and D. ciliaris (Retz. ) Koeler, southern crabgrass, frequently grow intermingled, which possibly accounts for D, bicornis being frequently mistaken for the latter, better- nown species. Morphological distinctions between D. ciliaris and D. bicornis are given by Gould (1975) Voucher specimens (TAES) of D. bicornis were collected at locations in six states from which the species has not previously been reported: North Carolina, South Carolina, Georgia, Alabama, Mississippi, and Louisiana. This crabgrass was found to be an important invading species of cultivated soils throughout the southeastern coastal plain. Specimens were not collected north of North Carolina; however, there seems to be no environmental reason why this species should not occur in sandy coastal areas of Virginia and Maryland. Texas collections I have made extended the range of D, bicornis into the pineywoods and post oak savanna vegetational regions of the eastern portion of that state. The northernmost collection site in Texas was only 30 miles from the Arkansas border and 60 miles from the Okla- homa border; future collections extend the range of the species north into Avlainsas and Oklahoma. t D. Webster, Department of Range Science, Texas AEM University, Colleve Station, TX 77843 REFERENCES CORRELL, D. S. and M. C. JOHNSTON. 1970. Manual of the vascular plants of Texas. Texas Research me Renner, Texas. 1881 pp. GOULD, F. W. 1975. Grasses of Texas. Texas A&M University Press, Col tion, Texas. 653 pp. ege Sta- SIDA 8(4): 352. 1980. eee) Figure 1. Distribution of Digitaria bicornis in southeastern United States. HENRARD, J. T. 1950. Monograph of the genus D¢gitaria. Leyden, Universitare Per Leiden. 999 pp. SWALLEN, J. R. 1963. New species of Digitaria and Trichachne. Rhodora 65: 22-307. NAJAS MARINA: NEW TO THE INDIANA FLORA—On a recent plant collecting trip through the Midwest, two collections were made of Najas marina L., the prickly naiad, in northeastern Indiana. After consult- ing the herbaria at IND and ND (acronyms follow Holmgren & Keuken, 1974), I have determined that this species is new to the Indiana flora. Col- lecting data for N. marina in Indiana are as follows: Steuben Co., off hwy 120 near I-69 at Green’s Lake, ca 7 mi N of Angola, 20 Jul 1979, Davenport 1424; Noble Co., on hwy 3 at W side of Cree Lake, S of South Milford, 20 Jul 1979, Davenport 1427. Voucher specimens for both collections are deposited at UNA. SIDA 8(4): 353. 1980. 354 In North America, Najas marina is found infrequently in brackish or highly alkaline waters of the western United States, Texas, the Great Lakes states, and Florida (Haynes, 1979). Whether it is a native or a naturalized J is under question. After studying its distributional history, Wentz & Stuckey (1971) concluded that N. marina was introduced into the Ohio region, and fairly recent collections in Michigan (Near & Belcher, 1974), Wisconsin (Ross & Calhoun, 1951), and Illinois (Winterringer, 1966) seem to bear out this conclusion. Its occurrence in Indiana, therefore, is to be expected and indicates the further spread of this species in the Great Lakes region —L, J, Davenport, Ecology and Systematics Section, Department of Biology, The University of Alabama, University, AL 35486. REFERENCES —— vi R. 1979. Revision of North and Central American Najas (Najadaceae) . —56 nae P. K. and W. KEUKEN. 1974. Index herbariorum, part 1: The eae of the world. Utrecht, Netherlands. 397 pp. NEAR, K. A. and R. O. BELCHER. 1974. New _— for Najas minor and marina in eee Michigan. Michigan Bot. 1-185. ROSS, J. G. and B. M. CALHOUN. 1951. ee reports on flora of Wisconsin. XX XIII. Naladaae Trans. Wisconsin Acad. Sci. 40: 93-1 WENTZ, W. A. and R. L. STUCKEY. 1971. Se changing distribution . ae genus Najas CNiadacess) <1 in aoe Ohio J. Sci. 71: 292-302. WINTERRINGER, G. S. 1966. Aquatic note plants new for Illinois. Rhodora 68 221-222. on GAURA ODORATA LAG, (ONAGRACEAE) IN LOUISIANA.—Speci- mens from a small population of Gawra found in Lafayette were identified through the keys of Munz (1965) and Correll and Johnston (1970). Com- parison with specimens borrowed from SMU confirms identification as Gaura odorata Lag., which can now be recorded from Louisiana. Lafayette Parish: rhizomatous perennial growing aggressively with Cynodon dactylon on un- cemented square around telephone pole in parking lot on N. University Ave. near corner of Louise St. Lafayette, Vincent 2310 (LAF), 18 Jun 1979. The plants were observed to flower and produce fruit sporadically from March to October 1979 in spite of several mowings. They were vigorously producing new shoots in March 1980 and thus may persist and disperse in the area. Their presence in Lafayette marks a significant eastward range extension for the species, a native of Texas and adjacent Mexico that is rare in east Texas. Previously unreported from Louisiana, G. odorata was not included in the preliminary report on Louisiana Onagraceae by Ellis and UrBatsch (1979). Duplicates are being sent to AC, GA, NCU, NLU, NO, SMU, and VDB. SIDA 8(4): 354. 1980. 355 I thank Dr. William F. Mahler for the loan of specimens—Karl A. Vincent, University of Southwestern Louisiana, Lafayette, LA 70504 REFERENCES CORRELL, D. S. and M. C. JOHNSTON. 1970. Manual of the vascular plants of Texas. Texas Research see Renner, Texas. ELLIS, R. R. and L. E. URBATSCH. ar Preliminary reports on the flora of Soar IV. Onagraceae. ee 44: 28-38. UNZ, P. A. 1965. Onagraceae. North American Flora. New York Botanical Garden. Series II, part 5: 1-231. A NOTE ON FLOWER COLOR OF WAHLENBERGIA LINARIOIDES (LAM.) A. DC. IN NORTHWEST FLORIDA—A recent article by D. B. Ward (Phytologia 39: 1-12, 1978) includes a key for distinguishing the two known species of Wahblenbergia which occur within the state of Florida: W. marginata (Thunb.) A. DC. and W. linarioides (Lam.) A. DC. The key was adapted and expanded from a previous key and descriptions pre- sented in an article by R. K. Godfrey (SIDA 1: 185, 1963). In both of these papers the flower color of W’. /inarioides is stated as blue During more than ten years of botanizing in the Pensacola area I have encountered W’. linarioides at numerous sites along roadsides and in sandy pinewoods. All of the observed specimens have had white flowers. I have recently discussed this with R. K. Godfrey, and he, too, has seen primarily white-flowered specimens of W’. linarioides in recent years. 1 conclude that current populations of W.. linarioides in northwest Florida are predominantly white-flowered, and that if blue-flowered individuals exist they are rare indeed. The fact that W. linarioides has white flowers makes it easy to distinguish in the field from W’. marginata, which has blue flowers, and botanists collecting in northwest Florida should take note of this. The following white-flowered specimens of W. linarioides may be examined in the herbaria indicated: Burkhalter 3908 (UWFP), 6374 (UWFP, CAS) —James R. Burkhalter, University of West Florida, Pensacola, FL 32504. BRACHYELETRUM ERECTUM AND TALINUM RUGOSPERMUM, NEW SPECIES TO TEXAS AND NOTES ON SCHOENOLIRION WRIGHTIU—tThe grass Brachyeletrum erectum (Schreb.) Beauv. is gener- ally distributed throughout the eastern United States, including Oklahoma and Louisiana (Hitchcock, 1950; Gleason, 1968). It is not surprising there- fore, that it occurs in eastern Texas. Brachyeletrum erectum was initially found in 1971 (McCrary 171 ASTC) and has since been collected from the same location (Kyle 40, Jun 1976 ASTC; Nixon 8723, Aug 1978 ASTC). Generally this species occurs in dry to moist wooded areas. We SIDA 8(4): 355. 1980. 356 found it within Nacogdoches County along a branch of Naconiche Creek about 15 miles northeast of Nacogdoches, Texas. Several small populations were present. It was within a forest dominated by Fagus grandifolia, Carpinus caroliniana and Liquidambar styraciflua (Nixon and Raines, 1976). Talinum rugospermum Holz. was collected from the extreme northern portion of Nacogdoches County along FM 1087 (Nixon 7752, Aug 1977 ASTC; Nixon and Marietta 9492, Jul 1979 ASTC). The species occurred occasionally in open, rather disturbed areas, on deep sandy soils. The reported distribution of T. rvgospermum was the north central United States—lIa., Minn., Wis., Ill, and Ind. (Rickett, 1969; Gleason, 1968). It grows in sandy soils or on sandstone in that area. It is somewhat amazing, therefore, that this species is encountered so far south. Page wrightii Sherman (S. texanum (Scheele) Gray; Sherman, 79) is a rare small white-flowered species that grows on moist prairies in nee Alabama, and East Texas (Hardin, Austin, Brazos, and Walker counties). It has been proposed as an endangered species (U.S. Dept. of the Interior, 1976) and was last collected in Texas in 1963 in Walker County (Sherman 560, 15 Apr 1963 VDB). Schoenolirion wrightii was collected in Apr 1978 (Marietta and Nixon 61 ASCT) in an open prairie like area in Jasper County. This collection extends the known range of S. wrighti somewhat and verifies its present existence in East Texas—Elray S. Nixon and Kay L. Marietta, Biology Department, Stephen F, Austin State Univer- sity, Nacogdoches, TX 75961; Michael McCrary, National Plant Materials Center, Beltsville, MD 20705. REFERENCES GLEASON, H. A. 1968. New Britton and Brown illustrated flora of the northeastern United States and adjacent Canada. Hafner Publishing Company, Inc. New York. HITCHCOCK, A. S. 1950. Manual of the grasses of the United Sie. United States Dept. of Agr. Misc. Pub. No. 200, Washington D.C. NIXON, E. S. and J. A. RAINES. 1976. so creekside vegetation of Nacogdoches County, Texas. Texas J. Sci. 27: RICKETT, H. W. 1969. Wild Rowers - a United States: Volume 3, Texas. Mc- Graw-Hill Book Company, New Y SHERMAN, H. L. 1979. Evidence “of misapplication of the name Schoenolirion texanum (Scheele) Gray (Liliaceae). Southw. Nat. 24: 123-126. U. S. DEPARTMENT OF THE INTERIOR. 1976. Endangered and threatened species —plants. Federal Register 41: 24524-24572. HOLOSTEUM UMBELLATUM (CARYOPHYLLACEAE) IN ON- 10.—The spread of Holosteum umbellatum L., jagged chickweed, in the United States has been discussed by Shinners (1965). This species, evi- dently native to Europe and/or western Asia, is now found throughout the United States, but the colonies tend to be widely separated and in some SIDA 8(4): 356. 1980. 357 cases not long persistent. It is most common in the warmer regions, and records from Michigan and New England remain few. The only recent collections of H. wmbellatum from Canada have been from southwestern British Columbia (Scoggan, 1978). Campbell (1895 ) reported having found H. wmbellatwm at Kettle Point, Lambton County, Ontario, in 1894. This record, if correct, represented a considerable westward expansion of the known range of H. wmbellatum in North America at the time (see quotations in Shinners 1965). Later authors, however, have found no specimens documenting this report and have cited it with a question mark (e.g. Scoggan 1978) or omitted it. Specimens from Campbell's her- barium, including many from Kettle Point, are in M MG, but, according to information kindly provided by Marcia J. Waterway, there is no speci- men among them now or formerly identified as H. #mbellatwm, either under Holosteum or in a genus that might likely have been confused with Holosteum (other Alsinoideae; Androsace). It seems probable, therefore, that this report was erroneous, and that the specimen was reidentified soon thereafter, especially if, as the title of Campbell’s paper impies, the plants in question flowered in summer. (Holostewm wmbellatum is in mature fruit by mid-spring, and the remains of the plants disappear in summer.) There have been no other reports of H. wmbellatum from eastern Canada. Flowering stems of H. wmbellatum were brought to me for identification on 18 April 1980 by J. A. Runnells of Stoney Creek, Ontario. On 2) spit following Mr. Runnells’ directions, I found a population of H, umbellatum comprising thousands of plants along the Canadian National tracks near the Royal Botanical Gardens’ Rock Garden in Hamilton, Hamilton- Wentworth Region (43° 17’ 17” N, 79° 53’ 32” W). Here the railroad runs through a cut in a sandbar of Lake Iroquois age, with H. wmbellatum growing on a sunny south-facing slope of the cut, in dry sandy soil, intermixed with Bromus tectorum (Pringle 1936, DAO, HAM, OAC, MO). On 5 May 1980, I found numerous plants of H. wmbellatum in Burlington, Halton Region, growing as weeds in crushed-stone walks in the R.B.G.’s Hendrie Park (43° 177 30” N, 79° 52’ 30” W, 148 km ENE of the Hamilton popula- tion), on nearly level terrain, in full sun, associated with Erophila verna and Sedum acre (Pringle 1937, HAM). Some plants were also found in nearby lawns where the soil is gravelly, derived from Lake Iroquois beach deposits (Pringle 1938, CAN, HAM, UWO). Holosteum umbellatum can readily be distinguished from related plants of the subfamily Alsinoideae by its umbellate inflorescences and by its petals, which are jaggedly few-toothed at the apex but not regularly bifid. Following Walters’ (1964) treatment of Holostewim, these Ontario popula- tions represent H. wmbellatum ssp. umbellatum—Iames S. Pringle, Royal SIDA 8(4): 357. 1980. 358 Botanical Gardens, Box 399, Hamilton, Ontario, Canada L8N 3H8. (Con- tribution No. 40 from the R.B.G.) REFERENCES CAMPBELL, R. 1895. Some of the rarer summer flowers of Canada. Canad. Rec. Sci. 6: 342-351, SCOGGAN, H. J. 1978. Satente In: The Flora of Canada 3: 690. Natl. Mus. Nat. Sci. Publ. Bot. No. SHINNERS, L. H. 1965. | ven eases go Ra vaee population ex- plosion and fractionated suicide. Sida 2: 119-1 oo S. M. 1964. Holosteum L. In: ae a G., et al., eds. Flora Europaea 1: 136. Cambridge: University Press. SIDA 8(4): 358. 1980. SID CONTRIBUTIONS TO BOTANY INDEX TO VOLUME 8 Names of contributing authors are in capital letters. New scientific names are in boldface. Synonyms are in italics. Abutilon theophrasti 84 Ac alypha gracilens 80, 145; rhomboidea Acer ooiado 142; pensylvanicum 142; rubrum 137, 142; saccharinum 75, 138, 14 Achillea millefolium 77 Acorus calamus Aegilops cylindrica 81 Aesculus glabra ae Rar 138, 146; pavia 102, 107, Agalinus aa a ie Agavaceae 314 Agave lecheguilla 245, 348; virginica 142 Agrimonia parviflora 86, 149; pubescens 149; rostellata 14 Agropyron cristatum 168; repens 81 Agrostis hyemalis 81; perennans 148; 81 Ailanthus altissima 150 Aizoaceae 75 Albizi oer 145 eee acaulis 349; farinosa 147 lisma HRS -aquatica 73, 75; sub- cordatum 142; triviale 119 Alismataceae 75, 142 Allium canadense 84, 147; sativum 84; vineale 84 Alnus surrulata 144 Althaea rosea 84 Amaranthaceae 75 Amaranthus hybridus 75; spinosus 108 42 bidentata 77; trifida 77 Amelanchier arborea 74, 86; canadensis 149 . Amorpha canescens 83 76, 142 Andropogon gerardii 72, 81, 148; sco- parius 148; ternarius 190, 191; vir- , 148, 190, 191 Anemone quinquefolia 149; virginiana Angelica venenosa 88, 142 Anisostichus capreolata 144 Annonaceae Antennaria plantaginifolia 77, 142; soli- taria 142 Anthemis cotula 77 Apiaceae 142 pios americana 74, 83, 145 Aplectrum hyemale 147 Apocynaceae 76 Apocynum cannabinum 76 Aquifoliaceae Aquilegia 349; canadensis 14 Arabis canadensis 144; laevigata 144 5 Aralia racemosa 142; spinosa 142 Arctium minus 77 Arisaema macrospathum 119; triphyllum 42 Aristida dichotoma 81, 148; longespica 90; ee Ri ciclochiae ae 14 Aristolochia an 138, 142; ser- pentaria 142 Arnica wilsonii 344 Aruncus dioica 149 gee gigantea 148 Asarum can Meclepiadacens 6, 142 Ascleptias emplestcau 142; hirtella 76; incarnata 76; purpurascens 76; quad- rifolia 142; syriaca 76; ated 76, 360 142; variegata 142; verticillata 76 Ascyrum hypericoides var. hypericoides 190 Asimina triloba 142 Asparagus officinalis 84 Aspleniaceae Asplenium montanum 141; platyneuron a 141; resiliens 213; ghizophy liom > — 14 Aster 35; cordifolius 142; divaricatus 142; dumosus 143; infirmus 143; laevis 143; lateriflorus linarii- folius 143; SOwaTEE ANS 14 eeu folius 77; pate : > paternus 143; pilosus var. demotus 77; praeal- simplex 143; so lidagineus 143; sur- culosus 143; umbellatus 137, 143 Asteraceae Astilbe biternata 150 Atamasco texana 329, 330; tubispatha 330 Athyrium asplenioides 141; pyenorat pon 138, 141; thelypterioides 141, 213 Aureolaria ievieats 150; wiewinies 150 Austrobassia 57 Avena sativa 81 Axyris amaranthoides 168 Ayenia pilosa 348 Baccharis halimifolia re 198 Balsaminaceae 76, 14 Baptisia leucantha 83. leucophaca 83; tinctoria 14 Barbarea vulgaris var. arcuata 79 Bartonia virginica 146 Bassia 57; hirsuta 57, 59, 60; hyssopi- folia 57, 58, 60, ap 161 — Betula lutea 144; nigra 144 Betulaceae 76, 144 ae bipinnata 77, 143; frondosa 77 ; polylepis 77, 143 a 76, 144 — era 29 BLAC Sih eee H.. JRa D7; Blechnaceae Bochmeria Te 88, 150 Bommeria hispida 349 ~~ ch 7, 10 eke dissectum 14 1; virginianum 85 ea gracilis 349; ramosa 348 ae eae erectum 148, 354 ractearla amazonica 92 Bele, juncea 79; coe var. pinnati- fida 79 Brassicaceae 144 Brazoria pulcherrima 211 Bromus commutatus 81; japonicus 81; ee ns 148; secalinus 148; tectorum tS sD a RALPH E. aoe ves ae 8: anne 150 BURK , JAMES R. 354 Seni Wee 315, 348 Buxaceae 144 Cacalia atriplicifolia 143 Calamogrostis cinnoides148 Callicarpa americana 188, 190 Callirhoe digitata 84 Callitrichaceae 7 Call ses 34, 35, 56; heterophylla 76; terrestris 76 C he ccae 144 Calycanthus floridus 137, Calystegia aa iy, spithamaca 144 Comsanida cee a 144 Campanulaccae 76, Campsis radicans i. en 109, 144 Capparidaceae 76 Caprifoliaceae 76, oe Capsclla bursa-pastori Cardamine hirsuta 144: arenicola 79 Carduus nutans 77 Carex 73; aestivalis 144; albolutescens 144; albursina 144; amphibola 144; ~! \o ; parviflora var. meadii 79; molesta 79; muhlenbergii 79; pedunculata 139, 145; pensyl- vanica var. distans 145; plantaginea 145; purpurifera 145; retroflexa 145; rosea 145; scoparia 73, 79; squarrosa 79: swanii 145; tribuloides 79; vire- scens 145; vulpinoidea 73, 79 Caroliana 35 Carpinus caroliniana 139, 144, 210, 355 Carya cordiformis 146; glabra 146; 146; pallida 146; texana 74, 82; tomentosa 74, 82, 138, 146 Guia 77, 144, 355 Cassia marilandica 145; nictitans 83, 145 Castanea dentata 146 Cathartolinum elongatum 182 CATLING, P. M. 188 Caulinia 35, 36; alagensis 45; flexilis 40; fragilis 51; guadalupensis 43; indica 51; minor 51; tenella Caulophyllum thalictroides 144 Ceanothus americanus 149; var. pitcheri 86 Celastraceae 77, 144 Celastrus scandens 77 Celtis laevigata 107, 108; occidentalis Ceachnae incertus 205; longispinus 81 Centaurea maculosa 77 Centrosema 152; angustifolium 152, 155; galeotii 155: grandiflorum 155; Cephalanthus occidentalis 73, 87, 14 Ceras ae Le ratum 77; holosteoides var. vulgare 144; vulgatum Cercidium 216 Cercis ae, 83, 14 Chaero phyllum procumbens 88; taintu- 142 Ghanaians luteum 138, 147 ee exumensis 319; lechioides ar. exumens is 319; var. wilsonii 319; ems z 319 c eR latifolium 148; laxum Cheilanthes alabamensis 314; lanosa 141; wrighti1 3 Chelone glabra 150; lyonii 138, 150 Chenolea 58 Chenopodiaceae 57, 77, 157 Chenopodium album 77 imaphila maculata 145 Chondrophylla 14 361 Chrysanthemum leucanthemum 77, 143 Chrysopsis pilosa Cichorium Hes vi: Cicuta maculata 142 Cimicifuga mre 140, 149 Ciminalis 16 Cinna arundinacea 73, 148 Circaea lutetiana ssp. ome 147 Cirsium altissimum 3; muticum 143; vulgare 77 Cissus incissa 108, 109 Cistaceae 77, 144 Cladrastis lutea 138, 145, 216 Clarkia 295 Claytonia caroliniana 138, 149; virgin- ica 86 Clematis oe dictyota mace integrifolia 119; pitcheri 194; var. dictyota a 194, 195; var. anes ie 195; reticulata 8; virginia Clintonia osalis Clitoria 90; gee a 90, 91; amazo- 7; arborescens 91, 94; brachycalyx 93, 94, 304, 305; brachystegia 94, 304; subgenus bractearia 90, > sect. brachycalyx 93, 304; sect. ibractearia 5; sect. caulifl > sect flexuosa 93; froesil 311; glaberrima 94, 304; Peione 155; hermannii 304, 305, 306; hoffmannseggii 91; javitensis A juninensis 95, 96, , 99, 1; leptostachya 94; mariana 145; ee bambensis , 100, 101; nervosa 3, 95; obidensis 94; poitaei 90, 91; pozuzoensis 93; rac $a ; sagotii 94; selloi 94 COLLINS, SCOTT L Collinsonia emnadengs io 146, 216 Comandra richardsiana 87 Commelina communis 77; erecta var. angustifolia Commelinaceae 77, 144 Compositae 65, 77, 216, 313 Conopholis americana 14 Convolvulaceae 78, 144 Convolvulus arvensis 78 362 Corallorhiza odontorhiza 147 Coreopsis auriculata 143; major var. stellata 143; palmata 77 Cornaceae 79, 144 Cornus alternifolia 144; amomum 139, 144; florida 74, 79, 144 CORRELL, DONOVAN S. 317 “orylus americana 76, 144 Crassulaceae 144 Crataegus crusgalli 86, 149; flabellata 149; flava 149; viridis 149 Crotalaria sagittalis 83 < Croton capitatus 80; glandulosa var. septentrionalis 80; monanthogynus 80, 145 Cruciferae 79 “ryptotaenia eter 142 Cucurbita digita Cucurbitaceae 135 Cunila 220; galioides 293: montana 290; origanoides 82; piperita 251: pulegioides 288 spicata 293 Cuphea carthagenensis 114, 115; petio- lata 84 Cupressaceae 79, 142, 312 Cuscuta cam pestr ris 144; compacta 144; pentagona 78, ’ Co. be 350, 321, 322, 323, 324, 325, 326, 327: fees 79; 3? praclongatus 145; pseudovegetus 80; retrofractus 145; rotundus 320; strig- osus 80, 145 yee eae, ee a oe calceolus var. pubesc s 138, Cyrtomium otn Dr Cystopteris bulbifera 141; protrusa 141 48 ? Datura stramonium var. tatula 88 Daucus carota 88, 142 DAVENPORT, L. J. 1, 353 DAWSON, DEBORAH 296 Delphinium ajacis 86; 149, 216 Dendrocyamus 90 Dennstaediaceae 141 Dennstaedtia punctilobula 141 Dentaria laciniata 79, 144 Deschampsia flexuosa 148 Descurainea pinnata var. brachycarpa 79 Desmanthus illinoensis 83 Desmodium ciliare 83; cuspidatum 145; glutinosum 145; iinoense 83; mari Diarrhena americana 148 Dicentra cucullaria 146, 214 Digitaria bicornis 351, 352; ciliaris 351; diversiflora 351; ischaemum 81, 148: , > Dioscorea batatas 145; villosa 80, 145 Dioscoreaceae 80, 145 Diospyros 72; virginiana 74, 80, 107, 8, 145 Dipsacaceae 80 Dipsacus sylvestris oi Dirca palustris 139, Disporum ee ve maculatum 38, 147 Distichlis spicata 107 Dodecatheon meadia 85, 216 Dolichos regularis 173 Draba verna 79 Dracocephalum virginianum 146 Drosera annua 190 Dryopteris goldiana 138, 141: inter- media 138, 141; marginalis 141 Dulichium erundmaceuin 73s DUNCAN, Meee H. 116, 170 Ebenaceae 80, Echinacea ne 77, 78; purpurea 143 Echinochloa crusgalli 81; muricata 148 s 57 Eleocharis 73; acicu aa 80; obtusa 80, 45; i ema a 73, 80; smallii 80; tenuis var. verrucosa 80 Hee hoe carolinianus 143; tomen- tosus 14 Eleusine indica 148 ELEUTERIUS, ee N. 10 Elymus canadensis 72, 81; virginicus 73, 81, Epifagus virginiana 148 Epigaea r Eragrostis cilianensis 81; oxylepis 205; spectabilis 72, 81, 148 Erechtites hieracifolia 143 Erianthus alopecuroides 148 Ericaceae 145 Ericala sedifolia 18 Erigenia bulbosa 142 Erigeron 65; annuus 78, 143; bigelovii 70; breweri 70; canadensis 78, 143; chiangii 65, 67, 68, 70; foliosus 70; var. confinis 69; var. hartwegil 69; hessii 70; kuschei 70; p hiladelphicus 108; pulchellus 78, 143; strigosus 78, 143 Eriodictyon californicum 7 Eriothymus 220; rubiaceus 293 Erythronium albidum 84; americanum 147 Euonymus americanus 138, 144; atro- purpureus 144, 21 Eupatorium album 43; aromaticum 43; compositifolium 189, 190; hys- sopifolium ; incarnatum 143; perfoliatum ; purpureum 143; rothrockii 349; rotundifolium 143; rugosum 143; serotinum 78, , 143 Euphorbia 317; abbreviata 317, 318; corollata var. zinniiflora 145, dentata 80; exum mensis 319; maculata 80, 108; mercurialina 145; proctorii 19; supina 80; vaginulata 318; Fagus oe 138, 146, 210, 355 FANTZ, PAUL R. 90, 95, 152, 304 ates elatior 148; obtusa 81, 148; pratensis 81 ee autumnalis 73, 80 Fluvialis 35; flexilis 40; minor 51 Forestiera eee 190 216 Fouqui sone vitginiane a 86 raxinus americana 85; var. biltmoreana 47; arizonica 349; ie 210; 1 49; cuspidata 349; pe eee 147, 216 363 Froelichia gracilis 75 FRYXELL, PAUL A. 123 Fumariaceae 1 Galactia 170; elliottii 170, 171; erecta 70, 171; floridana 170, 179; var. microphylla 172; glabella 170, 171, 73, 174, 175, 176, 179; macreei 170, 178, 179; minor 170, 172, 175, 176, 179; méssissippiensis 173; mollis 170, 171, 172, 173, 179; pilosa 173; 173, 175, 176, 178, 179; villosa 173: volubilis 145, 170, 171, 173, 177, 178, 179; var. txtermedia 179; var. i aes 13 Galium >; aparine oie 149; circaezans florum 150; virgatum 87 Garrya ovata 349 Gaura biennis 85; hexandra ssp. gracilis 349; odorata 353 Gentiana 14; acaulis 16; aquatica 14; beamanii 16, 17, 19, 21, 22, 23, 25, 27, 28; cespitosa 17; chimboracensis 18; edifolia var. casapaltensis 18; fieldiana 16, 17, 19, 23, 24, 26, 27, 28; perpusilla 15, 16, 17, 23, 26, 28, a. 32; podocarpa 14: prostrata 15, . 2420, ey Soe puberulenta 80; milio 17, 23, 26, 27 6295013 see 14; qu rae 146; saponaria 146; sedifolia 15, 17, 19, 20, 21, 23, 27, 28, 32; var. B grandi- Geranium oe 80, 146; macu- latum 80, 14 Geum ue 149; var. camporum 86 Gillenia stipulata 86, 149; trifoliata 14 Glechon Gleditsia tes 8, 74, 83, 108 Glyceria striata 73, 81 Gnaphalium helen! 143; oan 78, 143; purpureum 78, 143 Cancthen 298; helianthoides 299 Goodyera pubescens 14 GRAHAM, SHIRLEY A. 115 Gramineae 81 364 Gratiola virginiana 150 Greenella 313; arizonica 314; ramulosa 314 Guizotia abyssinica 296, 303 Gutierrezia 3; alamani 313; var. ieeaiicenheld 313; arizonica 314, conoidea 313; slutinosa 314; ramu- losa ericocarpa 314; a ee 314; texana var. glutinosa 314, wrightii 314 He byanthus gndeiionienas 330; var. texanus 330; andersonii 332; texanus 328, 330, 331, 332; tubispathus 328, 330, 331, Hackelia virginiana 76 Halea 398: sadoviciatia 301; var. re- eA : eee 219, 312; acinoides 223, 225, 85; 27, 234, 282, 283, adscendens 290; albescentifolinm 268, 292; bellum 230, 249, 250 we costatum 223, 229, os 247, 255, 268, 272, 276; var. costatum 268, 269, 270; costatum var. pulchellum 268, 269, 271; crenatum 234, 286, 289, 290, 291, 292; dentatum 234, 278, 279, 280, 281, 282; var. nanum 262, 264; denudata 293: dif- fusum 229, 233, 247, 261, 262, 263, 275; drummondii 223, 233, 256, greggii — oe ae eat 247, 268, 285, 286, 280: Tena simplex 285; hyssopifolium 223, 224, 229, 229, 255, 247. 255. 271 OT 8, 279, 280, 282; incana 293; italianiae 293; johnstonii 234, 269, 272; jucun dum 232, 244, 250, 251, 252, 253; latum 258; longiflora 293; pia 66, molle 228, 229, 230, 232, 242, 244, 245, 246; montanum 228, 232, 242, 244, 245, 247, 312; ele Herden 223, macrocalyx 262, 263, 266, 312; var mexicanum 2 var. 62, 263, 265, var. typicum 264; napalensts oo nitda 293; oblongi- folium 234, 274 at 279, 282; var. mexicanum 778. longifolium 278, en noatiien 957; palmerit 225, 227, 230, 232, 241, ee 243, 244; patens 232, 249, 252, 254; patrinum 228, 232, 240, 244; permistun 268, 270; pilosum 228, 231 ; 238; piperi- tum 230, 232, re ae 250, 251, 252, 253; var. en 278; plicatum 229, 234, 269, 272, 273; subgenus Poliominthoides 231, 240; polygalaefolium 231, 248, 249, 291; montanum 290; pringeli 268, 270; pulchellum 268, 271; niilehenr mum. 225,229) 233,247,299, 25), 256, 258; pulegioides 219, 223, 230, 231, 234, 283, 286, 288, 289; B- mexicanum 293; forma simplex 288; purpurea 293; pusillum 229, 231, 236, 237, 238; quercetorum 234, 269, 293; serpyllifolium 259; serpylloides 293, stenodonta 293; subequale 294; tenellum 268, 270; tenuiflorum 234, 273, 274, 275; tenuipes 234, 269, 276; texana 294; thymoides 219, 264, 294; var. oblongifolium 278; todsenii 227, 228, 232, 238, 239; villosa 294; var. lavanduloides 294 ; glomeratus 161, 162, i 165, 168, ee sativus 160, 161; souda 160, 161, Haloragidaceae 82, 1 Hamamelidaceae 1 Hamamelis virginiana 146 Hedyotis crassifolia 87; longifolia 87; nigricans Hedysarum volubile 173, 179 hirsutus 78, 143. microcephalus 143. mollis 78 Heliopsis helianthoides 78, 143 Hesperaloe funifera 314; parviflora 314 Hesperozygis 220; bella 249; ciliolata 35; marifolia 223, 293; myrtoides 293; nitda 293; pusilla 236 Heterotheca graminifolia 143; mariana Heuchera americana 150; parviflora 150; richardsonii 87; rubescenes 349; villosa 150; var. macrorhiza 21 Hexalectris spicata 1 Hexastylis arifolia 138, 142 Hibiscus lasiocarpos 74, 84; syriacus 14 Hickoria olivaeformis 8; pecan 8 3 Hieracium gronovii 143; paniculatum 4 Hippocastanaceae 146 (C328 Holosteum umbellatum 355, 356, 357 Holubia 14; pyrenaica 14 Holubia 14; pyrenaica 14 Heleboueniie 14 HOPKINS, CHRISTINE O. 157 Hordeum vulgare Houstonia caerulea 150; purpurea 150 Hyas 35 365 Hybanthus concolor 150 Hypericum denticulatum var. recog- i 46; drummondii 82; gentia- noides 82, 146; hypericoides 107, 108; multilum 82, 146; fardtdocain 146; ee 82; prolificum 146; punctatum 82; stans 146; stragalum 146 Hypoxis hirsuta 75, 142 Hyptis mutabilis 202, 203, 204, 205, 206, 207, 208, 209; spicata 203, 209 A toria 107, Impatiens cece 76, 143; pallida 143 Ipomoea hederacea 79; lacunosa 79; pandurata 79; purpurea 79, 144: sagittata 108 6 VING, ROBERT S&S. 219, 313 a brachiatus 82 Isotria verticillata 137, 147 Itea virginica 137, Ittnera 35; major 37; minor 513 najas 3 Iva annua 78; frutescens 107, 108 Juglandaceae 82 Juglans cinerea 146: nigra 82, 146 Juncaceae 82, 146, 334 Juncus 73; acuminatus 74, 82; brachy- 337, 338, 339, 340, 341, 342, 343; saab 338; interior 74, 82; margina- s 146; nodatus 82; repens 137, 146; roemerianus 107, 335, 344; tenuis 74, 82, 146, 347 Juniperus ashei 312; silicola 102, 107, 108, 111s virginiana 74.79, Kalmia latifolia 137, 145 366 KEENER, ‘aa S. 114, 216 Keithta 22 Kochia Sais 214 Koeleria pyramidata 72 Krigia biflora 78, 143; oppositifolia 78 Kuhnia eupetotioides 7§ Lactuca Rene 78, 143; floridana 143; serriola 78 famincere 146, jum amplexicaule 83; purpureum Laportea canadensis 150 Larrea 216 Lauraceae 83, LAWSON, CHERYL A. 1 Lechea racemulosa 144; tenuiiela 77 1; virginica 148 304 tH oO we “eR a 5 fs) n D = 0 ee campestre 7 79; virginicum 79, Leptochioa filiformis 81 bicolor 145; capitata 83; cuneata 83, 145; hirta 145; & inter- media 145; intermedia 145; proc bens 83; repens 146; striata stipulacea 83 steuvei 146; wiolacea ica 8 3; vir Leucophyllum frutescens 348 Liatris aspera 78, 143; microcephala 143; peal 78; spicata 143 Ligusticum canaden 42 Lilium michiganense 137, 147 Linaceae 84, 147 Lindera benzoin 147 Lindernia dubia 87, 150 Linum elongatum 181, 182, 183, 185, 186; lundellii 183, 184, 185, 186; pactum 181; filifolium 181, 182, 183, 185, 186; striatum 147; subteres 185; sulcatum 84; virginianum var. fori: danum 147 LIPSCOMB, BARNEY L. 118, 320 Liquidambar styraciflua 8, 137, 139, 146, 355 Liriodendron tulipifera 138, 147 Lithospermum arvense 76; canescens 76 tuberosum 144 Loasaceae 84 Lobelia cardinalis 144; inflata 144; puberula 144; spicata 76 ee 147 Lonicera japonica 144; sempervirens -44 81 Lolium multiflorum 148; perenne Loranthaceae 14 LOTT, E J. 348 Ludwigia paige 85, 147; palustris var. americana 8 Luzula acuminata 146; bulbosa 82, 146 Lycopodiaceae 141 Lycopodium lucidulum 141; obscurum 141; porophilum 141; tristachyum 1 Lycopus rubellus 83; oo 83, 146 Lygodium palmatum Lysimachia hybrida 86; lanceolata 86, 149; quadrifolia 149; tonsa 149 Lythraccae 84 Lythrum carthagenense 114 ie cae acuminata 138, 147; grandi- flora 210; tripetala 147 tees 147 , WM. F. 211, 212 Saeed pean 138 Malaxis unifolia 138, 147 Malvaceae 84, 147 Manisuris rugosa 10 MANSFIELD ee A. 196 MARIETTA KAY L. 355 Marshallia cinetwie 139, 143 Matelea carolinensis 108, 142 McDANIEL ee 210 McINTOSH, K. Medeola virginiana oie Medicago lupuli ina — sativa 83 Melanthium virgin 4 Melastomataccac M7 Melica mutica 1 Melilotus alba on officinalis 83 Menzclia oligosperma se ogee 203; spicatum 203 Micromeria 222; bonariensis 260 Nicest vimineum 148 imulus alatus 150; ringens 87 Mirabilis nyctaginea 84 Mitchella repens 139, 150 Mitella diphylla 138, 150 Mitracarpus hirtus 117, 118; véllosus 117, 118 Mollugo verticillata 75 Monarda clinopodia 146; fistulosa 83 Monotropa hypopithys 145; uniflora 145 Monotropsis odorata 138, 145 Moraceae 84, 147 MORGAN, DAVID 210 Mortonia scabrella 248 Morus alba 84; rubra 107, 108, 111, 147 Mosla 220; ocymoides 293; dianthera 146 Muhlenbergia capillaris 190, 191; schre- beri 81, 148; sylvatica MUSSLEMAN, LYTTON J. 196 Myosotis virginica 76 Myrica cerifera 107, 108 Najadaceae 34 Najas 34; alagensis 45; ancistrocarpa 40; dichotoma 51; flexilis 37, 40, 41, 42, 43; ssp. os O, 41; var. congesta 40, var. curassavica 43; var. lees a var. guadalupensis var. busta 4 ly poe 37; gracillima 47, 49, 50, 54: graminea 37, 43, 45, 47, 48, 52; guadalupensis 41, 43; var. floridana 7, 45, 46, 52; var. guadalupensis 44, 52; var. muenscheri 37, 45, 46, 52; var. Olivacea 37, 43, 46 51; var. gracillima 47; latifolia 37; latior 37; major 28; var. angustifolia a Le 38; marina 35, 36, 39, 52 353; var. angustifolia 37; var. as ma 37; var. bollei 37; var. var. genwina 37; var. gracilis 37; var. latifolia 38; var. latior 38; var. mexi- cana 38; var. multidentata 38; var. muricata 38; var. paucidentata 38; var. recurvata 38; microdon var. curassavica 43; as guadalupensis 43; , 54, 56; var. setacea 51; muewschen a 4); piuusesied nuda 495; 51; SH ulate 51; multidentata 54. 307 Nelumbo lutea 73, 84 Nepetha multabilis 202 NELSON, GUY L. 65 Neviusia alabamensis 215 NICKRENT, DANIEL L. 196 Nicolletia occidentalis 216; trifida 216 NIXON, ELRAY 355 Nothoscordum bivalve 84 Nyctaginaceae 84; Nymphaea 34, 56, 84 Nyssaceae 85, 147 Nyssa sylvatica 74, 85, 137, 147 Oenothera 295; biennis 85, 147; laciniata 85; tetragona 147 OGBORN, JAMES E. A. 196 Oleaceae 85, 147 OLSEN, JOHN 128 Onagraceae 85, 147, 353 Ophioglossaceae 85, 141 Ophioglossum 113; crotalophoroides 113; nudicaule var. tenerum oe petiolatum 113; vulgatum var. pyc- nostichum 113 Opuntia compressa 109; polyacantha var. ae ie a stricta 109 Orchidaceae 85, Orchis specail es 147 Orobanche een 148 Osmorhiza claytonii 142 Osmundaceae 141 Osmunda cinnamomea 137, 141; regalis 137, 14 Ostrya virginiana 144 OTVOS, ERVIN C., Jr. 102 Oxalidaceae 85, 148 Oxalis dillenii 85, 109, 148; stricta 85, 148 Oxydendrum ee 137, 145 Oxypolis rigidior Pachysandra eaten te se Panax quinquefolium 140, Panicum amarum 109; 2 an 148; boscii 148; brachyanthum 190; capil- lare 81; clandestinum 81, 148; dichotomiflorum 81; dichotomum 12 Ol: philadelphicum 148; poly- anthes 148; rigidulum 7 81; sphaerocarpon 148; virgatum 73, 81, 109 368 Papaveraceae 148 Parnassia asarifolia 137, 150 Parthenium integrifolium 143; hispid- um 7 Parthenocissus quinquefolia 89, 138, 150 Paspalum 205; laeve 81, 148; pubi- flo var. glabrum 81; setaceum ar. Passifloraceae 85, Passiflora lutea ne. 148; var. glabri- flora 85 Pastinaca sativa PATRICK, [ous 5:13) 7, ee Pellaea atropurpurea 141, 214, Penstemon canescens 150: eee 87; tubaeflorus 87 Pennisetum typhoides 196 85; hydropiper- : lapathifolia 85; pensylvanica 86; punctata Petalostemum candidum 83; purpureum 83 Peteria scoparia 349 Phacelia bipinnatifida 138, 146 Phaethusa oe 13° PHILLIPPE, LOY R. 135 Phleum pratense 82 Phlox amoena 149; carolina 149; eee cata 138, 149; glaberrima 139, maculata eee ssp. pyramidalis i pilosa 85, 139 Phoradendron serotinum 147 Phragmites communis 109 Phryma Cee 148 Phrymaceae 148 Phyla oe 88 Phyllanthus polygonoides 315 Physalis angustifolia er heterophylla 88, 150; subglabrata 8 Physocarpus opulifolius 2 Phytolacca americana 85; rigida 109 Phytolaccaceae 85 Pilea pumila 88, 150 Pinaceae 142 inus Here 137, 139, 142; glabra 8, taeda 142: virgiana 138, 139, Plantaginaceae 85, 148 an fae aristata 85, 148; lanceolata 89; 148; pusilla ae rugelii 85, ee 85, Platanaceae vas Platanthera ciliaris 137, 147; clavellata 48; yee as 48; integ- ale 137, 139, Platanus Paci , 148 Sone andens Poa ua ae 148 Poa compressa 148; languida 139, 1 pratensis 82; 148 Podophyllum peltatum 76, 144 Pogogyne 331; serpylloides 293 Polanisia dodecandra 76 Polemoniaceae 85, 14 Poliomintha 220; glabrescens 236, 293; incana 293; longiflora 223, 236, 292, as - Polygala curtissii 149; incarnata 85, 149; minutifolia 349; nudata 349; sanguinea 85; senega 85, 149; verti- 49 Polygonaceae 85, 149 Polygonatum biotin 84, 147; pubes- cens 138, Polygonum aviculare 86; cespitosum var. longisetum 149; persicaria 149; punctatum 149; scandens 74, 86, 149; virginianum 8 Polymnia canadensis 143; uvedalia 143 Polypodiaceae 86, 141 eeu lone 141; thyssano- lepis; virginianum 141 Polystichum scrontichides 141 POOLE, JA M. 315 Polytaenia ae 88 Portulacaceae 86, 149 Potamogeton 35, 55; diversifolius 73, 86 Potamogetonaceae = Potentilla canaden 149; aa 149; recta 86. ae 86, Prenanthes altissima 143, aspera 78; 149 PRINGLE, JAMES S. 14, 119, 356 Prosperpinaca palustris 82; pectinata > Prunella vulgarus 146; var. lanceolata 83 Prunus americana 149; hortulana 87; maheleb 87; munsoniana 87; persica 74, 87; serotina 74, 87, 149, 349 Psendocunila montana 290 Psoralea psoralioides 146; var. eglandu- losa 84 Preridium aquilinum 86, 141 Pycnanthemum incanum 147; muticum 147; tenuifolium 83; linifolium 7 Pyrolirion 333 Pyrrhopappus carolinianus 78, 143 Pyrularia pubera 1 Pyrus angustifolia 149; arbutifolia 149; ioensis 87; malus 74, 87; melano- sae ae 74, 80, 138, 139, 146; coccinea 146; daraddii 7; falcata 139, 188, 190, 191; velutina 74, 80, 146; virginiana 108 Ranunculaceae 86, 149 Ranunculus 35, 55; abortivus 86, 149; fascicularis 114; var. apricus 86; laxicaulis 86; ete 86; recurva- tus 149; trilobus 114 Ratibida pinnata 78 Rhabdocaulon 22 coccineus 293, denudatus 293; eraalis 293; iavond loides 294; wae 2 Rhamnaceae 86, 149 Rhamnus caroliniana 149 Rhexia mariana 147; virginica 147 Rhododendron alabamense 145; arbore- scens 139, 145; calendulaceum 145; cumberlandense 145; maximum 145; nudiflorum 137, 145 Rhododon 220; ciliatus 221, 223, 292, 293, 294 Rhus 72; aromatica 74, 76; e 74, 76, 107, 108, 142: cotinoides 7; glabra 74, ae 142; radicans 74, 76, 109, 138, Breck oe ee 145; cornicu- lata 137, 145 Ribes missouriense 74, 87 369 Robinia 146; pseudo-acacia 146 ROGERS, C. M. 181 ROGERS, KEN 215, 216 Rosa carolina 87, 149; multiflora 87, ; betulifolius 149; “igs 211; flagellaris 87; hispidus 149; occidental: 87, 149; pennsyl- vanicus 74, 87 phoenicolasius 149; trivialis var. duplaris 211, 212; var. trivialis 211 Rudbeckia fulgida var. umbrosa 143; hirta 78, 143; laciniata 143; triloba 78 Ruellia caroliniensis 142; humilis 75 Rumex acetosella 86, 149; crispus 86, 149 Rumfordia 296, 297; aaa 303 Sabal palmetto 107, Sabatia angularis 80; fide 146 Sageretia minutiflora 102, 107, 108, 109 Sagittaria cuneata 119; “aie rama 142; lat hope yo Fee OF aet ie ola Salicaceae 87, Salix eine HE at humilis 87, 150; nigra 73, Salsola kali 159; ie on ae 161 Salvia lyrata 147; anientoles 14 Sambucus canadensis 74, 76, un 108, 144; pubens 144 Samolus parviflorus 109, 210 Sanguinaria canadensis 148 Sanicula canadensis 88, 142; gregaria 142; smallii 142; trifoliata 142 Santalaceae 87, 150 Saponaria officinalis 77 Sapotaceae 15 Sassafras 72; albidum 74, 83, 147 Satureja 222; bonariensis 260; brownei 293; caerulescens 292; georgiana 205; glabra 292, 293; glabrella 293; gillie sii 260; _micromoroides 29 0 Schizachyrium scoparium 72, 82; var. 0 Schoenolirion wrightii 354, 355 Schrankia microphylla 146; uncinata 84 Scirpus 73; americanus 325; atrovirens a7 80; var. georgianus 145; californicus 117; cyperinus 145; pendulus 80; validus var. creber 80 Scleria oligantha 145; triglomerata 145 Scrophularia marilandica 87 Scrophulariaceae 87, 180, 196 Scutellaria elliptica 147; lateriflora 83; ovata 83, 109; var. versicolor 215; parvula 83; —— rrata 147 Secale cereale 8 Sedum ternatum 1 138, 144; eas 349 Selaginella apoda 142; rupincola 349; ncinata 214 Selaginellaceae 142 Senecio aureus 143; obovatus 143; smallii 143 Serenoa repens 108 Setaria faberii 82, 149; geniculata 149; slauca 82; viridis 82 Sida anomala 123; elliotit 123, 124, 124, be 12 otosina 124 rzedowskii 123, eee 84; texana io turner 125 Siegesbeckia laciniat 132 Silene antirrhi 77; laciniata 349; rotundifolia ivr stellata 77, 144; virginica 77, 144 ia integrifolium 78; trifoliatum 43 eee rad Simmondsia 2 Siphonoglossa longiflora 350; pilosella 0 Sisymbrium officinale Sisyrinchium ees 146; angusti- folium 82, 146, 210; campestre 82 Smilacina racemosa 84, 147 Smilax bona-nox 83, 108, 109, 147; var. hastata 190; ecirrhata var. hugeri 147; glauca 147; hispida 84, 147; rotundifolia 2 Solanaceae 88, Solanum americanum 88; carolinense ree arguta 143; Res var. boottit A; var. eumees 14 caesia a canadensis 78; scabra 143; curtisit 143: erecta "143. flexicaulis 143; gigantea 143; gr minifolia var. nuttallii 143; hispida laris 78 riddellii 9; rigida 78; roanensis 143; r semper- virens 109; sphacelata 216; “ulmifolia 8, 143 Sorter avenaceum 72, 82; nutans 49 ae bicolor 82, 196; halepense 82 Spartina alterniflora cynosuroides 109; patens 345; pectinata 73, 82; spartinae 109 Specularia perfol | ie 144 Spermacoce glabra Sphenopholis ere ue Spigelia semi 147 Spiraea japonica Spiranthes cernua 35, 193; gracilis 85; aoe 188; ae Bory 189, 190, 191, era junceus 189, 190; vagini- florus 82, 149 Stachydeoma 220; graveolens 221, 223, 2¢ dy fy Xo Stellaria media 77, 109; pubera 144 Stewartia ovata 150 Stipa avenacea 149; tenuissima 349 Streptopus amplexifolius 13 a angu ale 196, 201; asiatica 198, 199, 201; aspera 196, 197, 198, 199; euphasioides 196; gesnerioides 196, 199, 200, 201; hermonthica 196, ot 198, 199, 200; klingii 199, 200, 201; /utea 196; macrantha Aas 200, 201 Strophostyles helvola 8 ’ Serlophiorum diphyllum 138 148 Stylosanthes biflora 84, 146 Styracaceae 213 Styrax americana 213 Suaeda hyssopifolta 58 wertia caroliniensis 146 Symphoricarpos 349; ee 76 Tacnidia integerrima 14 15 Talinum parviflorum ve rugosper- mum 354, 35 Taraxacum officinale 78, 14° Tephrosia pedicellata 199; 146 84, virginiana Tetragonosperma 2 5 ie 301 si oa ee abyssinica 303; temalensis 303; Pel nietes 296, oe. 298, 299, 300; ludoviciana 296, 297, 299, 301, 302, 303; parviflora 303; repanda 296, 297, 299, 302, 303; texana 296, 297, 299, 302, 303 149; dasycarpum 114; debile 114; pubescens 149; revo- lutum 86; thalictroides 149 Thaspium barbinode 142 Theaceae 150 helypteris yeaa 141; nove- boracensis 1 THIERET, Sonne ile arvense 79; pestliscm 79 THOMAS, R. DALE 113 TOE On RALPH L. 70 Thymelaecae 150 Thymus serpyllum 294 Tiarella cordifolia 150 Tilia eee 138, 156 Tiliac Tillandsi sia usneoi oy oe Tipularia disc olor 9 Tradescantia ernestiana 215; ohiensis 77; subaspera 144 Tragia ln tn 80 Tragopogon dubiu ‘Trautvetteria i ae 215 Trichachne 352 richomanes petersii 139, 141 Trillium cuneatum ia. erectum 138, 147; grandiflorum 147; oo 209, 210; recurvatum 138, Triodanis biflora 76, 349: ae - riosteum angustifolium 77; perfoli tu Triphora trianthophora 148 Tripsacum dactyloides "53, 82 Triticum aestivum 82 Tsuga eee 137, 138, 142 TURNER 296 Typha latifolia 73, 88, 150 Typhaceae 88, 150 Ulmaceae 88, 150 Ulmus alata 88, 150; americana 88, 371 0, 210; Lee 74, 88, 150 Uvularia grandiflora 147; perfoliata aboreum 145; constablaei 145; corymbosum 145; stamineum Wate nace eae 88 Valerianella radiata 88 , 14 Verbascum blattaria F 150; thapsus Me Wenacese 88, 150 erbena bracteata 88; canadensis 88; simplex 88, 150; stricta 88; urticifolia 50 Verbesina alternifolia 143; helian- thoides 78; laciniata 130; microptera 128... 130, Tol ae loa: Mol Gine 133; occidentalis 109; polycephala 132; rumicifolia 134; 128, 129, 130, 132, laris 133; var. laciniata 128, 132; var. palmeri 134; virginica 128, 131 Se altiss 143; arkansana ; baldwintt 78: missurica 78 ere arvensis 87; peregrina 87 Veronicastrum virginicum 33 Viburnum acerifolium 137, 144; cassi- noides 144; dentatum a nudum 137, 144; rufidulum 74, , 14 Vicia caroliniana 146; eee 84; Viola affinis 150; blanda 138, 150; canadensis 150; conspersa 138, 150; hastata 150; hirsutula 150; missouri- ensis 88; oa 88, 150; pensyl- vanica 150, 15 primulifolia i rafinesquii 88; rostrata 150; sagittata 88; triloba 1 50 Violaceae 88, 150 Vitaceae 89 Vitis aestivalis 89, 150; cinerea 89, 150; aay lia 150; vulpina 8 ttaria 142 oe 142 ae Vulpia octoflora 82 Wahlenbergia linarioides 354; margin- ata WEBB, DAVID H. 135 WEBSTER, ROBERT D. 351 WELLS, CHRISTOPHER J. 328 Wisteria frutescens 108, 109 WOFFORD, B. EUGENE 135 Woodsia obtusa 216 Woodwardia areolata 137, 141; virgin- ica 137, 138, 139, 141 Xanthium strumarium Xanthocephalum 313; conoidenm 313; lineartfolium 313; megalocephalum 313; sericocarpum 314; wrightii 313 Barney Lipscomb and Jessica Procter Herbarium Southern Methodist University Dallas, Texas 75275 Xanthorhiza simplicissima 139, 149 Xyridaceae 150 Xyris torta 137, 150 a i 107, 108 Zam ee clava-herculis 107, 108 Zephyranthes 329; andersonit 330; commersoniana 330; SNE ER 330; texana 229, 330 Zigadenus leimanthoides 137, 139, 147 Zizia aptera 142; aurea Ziziphora 220; hispidum 285; pule- gioides 288