ve PHYTOLOGIA An international journal to expedite botanical and phytoecological publication Vol. 66 June 1989 No.5 CONTENTS ,GUY L. NESOM, Taxonomy of Erigeron sect. Polyactis (Compositae: Astereae) SOSSSSSSSSSSSSSHSSSSSSHSSSSSSSHSSS SSS SSSSSSSSSSSSSHSSHSSSSSSSSHSSSSSSSSSSSSOSSSOSSOSSOO 41 5 B.L. TURNER, New species and combinations in Mexican Viguiera (Asteraceae-Heliantheae) SSSSSSSSSSSHSSSSSSHSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSESSES 456 B.L. TURNER, New combinations in sect. Eremosis of Vernonia (Asteraceae) SSHOSSHSSSSSHSSSSSSSSSSSSHSHSSSSOHSSSSSSSHSSSSOHOSSOSSHHSSSSSSOSSSSSSSOSSSSOSOO 462 LIBRA ARY ~ =m FF) JQQ0 AUG 7 ¢ Wo Ne Published by Michael J. Warnock 185 Westridge Drive Huntsville, Texas 77340 U.S.A. Price of this number $3.00; for this volume $16.00 in advance or $17.00 after close of this volume; $5.00 extra to all foreign addresses; 512 pages constitute a complete volume; claims for numbers lost in the mail must be made immediately after receipt of the next following number for free replacement; back volume prices apply if payment is received after a volume is closed. Phytologia (June 1989) 66(5):415-455 TAXONOMY OF f8 ERON SECT. POLYACTIS (COMPO E: AS IG A TEREAE ) Guy L. Nesom Department of Botany, University of Texas, Austin 78713 U.S.A. ABSTRACT Erigeron sect. Polyactis (Less.) Nesom is prominently char- acterized by arching-pendant buds, white ray corollas that sharply reflex at maturity and a pappus of 10-12 caducous bristles that break off cleanly at their very base. The bristles are completely absent in five species and usually replaced by a short, hyaline corona. The section is typified by FE. delphinifoltus Willd. and comprises 19 species, most of which are endemic to the Sierra Madre of western México. Nine new species are proposed here: E. annuactis, E. basaseachensis, E. caulinifolius, E. cir- culis, E. coroniglandifer, E. eruptens, E. nacoriensis, E. podophyllus and E. rhizomactis. One new combination is proposed: E. dactyloides (Greenm.) Nesom. The taxonomic history of the name of the section is presented. Achaetogeron A. Gray is included among the synonyms of Erigeron sect. Polyactis. KEY WORDS: Erigeron, sect. Polyactis, Asteraceae, México, systematics. The present study treats a monophyletic group of 19 species, which are formally described here as a section of Erigeron. Among these are taxa with a typical pappus of bristles as well as several that have only a low, coroniform pappus without bristles. The latter group includes Achaetogeron wishizent A. Gray, the type of the genus Achaetogeron A. Gray. STATUS OF THE GENUS ACHAETOGERON Numerous Erigeron-like species with pappus bristles absent or reduced in various ways have been placed into Achaetogeron since its original recognition (Gray, 1849). Loss of pappus bristles in Erigeron, however, appears to have occurred independently in at least six lineages, and more than once in at least one of these (Nesom, unpublished). In addition to Erigeron wislizeni 415 416 PHY TOLOGIA volume 66(5):415-455 June 1989 (A. Gray) E. Greene, the following four species have pappus bristles absent or reduced and have been treated as Achaetogeron: E. galeotti (A. Gray ex Hemsl.) E. Greene, E. narcissus Nesom, E. gilensis Woot. & Standl. and E. coronarius E. Greene. Each of these species represents a separate lineage, each of which includes other epappose taxa. Further, EF. strigosus Muhl. ex Willd. has ray achenes without bristles and could be considered to represent another lineage in which loss of bristles has occurred. I could find no features in plants of any of these lineages that would separate them from typical Erigeron (Nesom, 1980). E.L. Greene (1891) was the first to make formal transfers of Achaetogeron species to Erigeron and a number of more contemporary botanists have com- mented on the probable congeneric status of the two genera (e.g., Shinners, 1946; De Jong & Longpre, 1963; Pinkava & Keil, 1977). I published a group of new names and new combinations that completed the transfer of Achaetogeror, to Erigeron (Nesom, 1982). The present paper is the first to treat a monophyletic group of Erigeron species that includes both pappose and epappose taxa. ERIGERON SECT. POLYACTIS - ITS NAME AND CHARACTERISTICS With several seasons of field experience in species-rich southern Chi- huahua, I feel reasonably confident in presenting this treatment of a mono- phyletic group of Erigeron species poorly or not at all represented in most herbaria. Only three of the new species proposed here are described from sin- gle collections of a taxon that I have not personally observed in the field. As noted in the discussions below, however, a number of problems remain. Fur- ther collecting and study in remote areas, particularly in northern Durango, will be necessary to clarify the taxonomy of Erigeron sect. Polyactis. Erigeron sect. Polyactis (Less.) Nesom, comb. et stat. nov. Based on Polyactis Less., Syn. Comp. 188. 1832; non Link, 1809. Type species: Erigeron delphinifolius Willd. Polyactidium DC, nom. nov., Prodr. 5:281. 1836. Stenactis Cass., Dict. Sci. Nat. 37:485. 1825. Type species: Erigeron delphinifolius Willd.; non sensu Less., 1832; non sensu Nees, 1832; non Erigeron sect. Stenactis (Cass.) Torr. & Gray, Fl. N. Amer. 2(1):172. 1841. Achaetogeron A. Gray, Mem. Amer. Acad. Arts, n. ser. 4(Pl. Fendl.):72. 1849. Type species: Achaetogeron wislizent A. Gray. When Cassini first formally described the new genus Stenactis in 1825, he cited three species as members, Erigeron (Aster) annuus (L.) Pers., E. delphinifolius and E. alpinus L. and stated that the description was drawn Nesom: Taxonomy of Erigeron sect. Polyactis 417 from the first two. One of the characters that he emphasized as important in delimiting this group was the double pappus, but his description was also specific about the caducous nature of the pappus. Of the three species cited, only E. delphinifolius has a caducous pappus. The next year (Dict. Sci. Nat. 39:404. 1826), he removed EF. annuus into a segregate genus, Phalacroloma Cass. Still later, in redescribing Stenactis (Dict. Sci. Nat. 50:483. 1827), he specifically stated that the description was drawn from E. delphinifolius, while including in the genus several more species with a double but non-caducous pappus. The species cited in 1827 as composing Stenactis primarily were the ones he had recognized in 1822 (Dict. Sci. Nat. 25:96) as sect. Diplopappus of the genus Diplopappus. However, in one of the first circumscriptions of Diplopappus (Dict. Sci. Nat. 13:308. 1819), Cassini formally cited four species, which are now recognized as Chrysopsis (D. lanatus and D. intermedius), Aster, and Erigeron annuus (D. dubius). Erigeron delphinifolius was mentioned in the 1819 discussion as a fifth species, but was not given a name as Diplopappus until 1822. Since he later removed both erigerons from this group, it is clear that the original reference of the name Diplopappus was to Chrysopsis or Aster and that the first supraspecific name for which EF. delphinifolius is typical is Stenactis Cass. Torrey & Gray’s use of Stenactis at the sectional rank in Erigeron re- ferred to six species that are not closely related to FE. delphinifolius and none that were included among those cited by Cassini as belonging to Stenactis. Erigeron annuus was moved by Torrey & Gray in the same publication to sect. Phalacroloma and E. alpinus to sect. Erigeron, although E. delphini- folius is a Mexican species and would not have been included in their floristic treatment. Later, however, Gray pointed out (Proc. Amer. Acad. Arts 8:648. 1873) that sect. Stenactis Torr. & Gray was a synonym of sect. Phoenactts Nutt. and that his own misapplication was based on the earlier misinter- pretation of Stenactis by Nees & DeCandolle. In Gray’s Synoptical Flora (1(2):219. 1884), he did not recognize Stenactis as a section, but he did cor- rect his earlier concept of the name by noting Cassini’s original association of Stenactis with E. delphinifolius and the close similarity of E. neomezicanus with that species. Because, however, the type species of Stenactis Cass. was excluded (see ICBN, Article 48.1) in the original publication by Torrey & Gray, sect. Stenactis should be attributed solely to those authors and cited as such, not as a new combination based on Cassini’s genus. Lessing’s publication of the genus Polyaciis was superfluous (as well as a later homonym), based on a misunderstanding of Cassini’s work, since Less- ing retained the genus Stenactis with the sole species EF. annuus. DeCandolle corrected Lessing’s creation of a later homonym, but Polyactidium, too, was superfluous, a synonym of Stenactis. Nees, like Lessing, used Stenactis to circumscribe a genus with EF. annuus and several other species but did not 418 PHY TOLOGIA volume 66(5):415-455 June 1989 include E. delphinifolius. Polyactis, however, appears to be the first name typified by E. delphinifolius that is available for use at the rank of section. Plants of sect. Polyactis are remarkably uniform in numerous aspects of their morphology. Almost all of the species produce pinnatisect leaves and the buds have a characteristic arching-pendant aspect that is easily recognized, even at a distance. The ray flowers are narrow and reflex sharply at the tube-ligule junction so that the corollas are held down around the involucre (this is often apparent even on herbarium specimens). The style branch (disc corolla) lengths and collecting appendage shapes are uniform. The pappus bristles, when present, are mostly 10-12 in number and the bristles have lateral constrictions at the very base, clearly perceptible with the SEM and break off there with only slight pressure. The achenes of some species are bristleless and usually have instead a hyaline corona or ring of partially fused scales or setae. The behavior of the phyllaries is linked to the condition of the pappus, whether coroniform or of bristles. In those species with bristles, the phyl- laries reflex or spread after maturation and release of the achenes; with an evolutionary loss of the bristles, the phyllaries remain erect, even after the achenes are dispersed. The reflexing behavior is more pronounced, however, in species of Erigeron with persistent rather than caducous bristles. I know of no other species of Erigeron with the peculiar arching-pendant aspect of the buds nor of any with basally caducous pappus bristles. Reflexing ray corollas, however, are found in various other parts of the genus and | believe this feature is diagnostic in other species groups as well. In particular, in plants of the E. coronarius E. Greene group (including E. pumilus Nutt.), the ray corollas reflex and these species may be closely related to those of sect. Polyactis (Nesom, in prep.). In the morphological descriptions that follow, all measurements were made from pressed specimens. Several types of trichomes, all multicellu- lar, are found among the species treated here as well as the whole genus: (Type A) uniseriate, usually comprising the most conspicuous vestiture, de- scribed as hirsute, pilose, strigose, etc.; (Type B) uniseriate, very small and usually inconspicuous, not found to be of diagnostic value at the species level and not mentioned as such in the descriptions; (Type C) biseriate, glandu- lar, described as “granular-glandular” in small size to “stipitate-glandular” when larger and capitate. [lustrations of these trichomes are found in Nesom (1976). Erigeron sect. Polyactis Leaves commonly pinnately coarsely toothed or lobed, less commonly en- tire, the mid-region of the blade oblong and of even width. Heads shallowly hemispheric to cupulate, markedly impressed below, buds arching-pendant; Nesom: Taxonomy of Erigeron sect. Polyactis 419 phyllaries linear-lanceolate, with whitish but usually not scarious lateral zones, in 3-4(-5) series of nearly equal length, but the inner often shorter than the outer, reflexing to spreading after achene maturation and release in the species with a pappus of bristles, remaining erect if pappus coroniform; re- ceptacles flat to convex, smooth or alveolate. Ray flowers 50-230 in 2-4 series, corollas 5-16 mm long, the ligules 0.6-1.2 mm wide, white and drying white or sometimes cyanic in several species, sharply reflexing at the tube-ligule junction at maturity. Disc corollas narrowly tubular-funnelform, 1.8-4.0 mm long, the tube slightly constricted, barely or not at all inflated or indurated above the constriction; style branches 0.4-0.7 mm long, including the deltate to narrowly triangular collecting appendages 0.1-0.2 mm long. Achenes com- pressed, mostly oblong-oblanceolate, (0.8-)1.0-1.3(-1.8) mm long; pappus of mostly 10-12 (11-17 in E. podophyllus) brittle, basally caducous bristles, or the bristles absent and replaced by an outer corona or series of scales, usually basally fused, 0.1-0.2(-0.5) mm high. Chromosome number, n=9, 18 pairs. All the species of Erigeron sect. Polyactis occur in the western Sierra Madre of México. Thirteen species of sect. Polyactis are known only from Chihuahua, Sonora and Durango. Two of them, E. neomezicanus and E. oreophilus, range into the southwestern United States and the type species, E. delphinifolius, has the major part of its range in the trans-volcanic moun- tains of south-central México. Most occur above 1700 meters and are found primarily in oak to oak-pine or pine woodlands but sometimes range into as- sociated grasslands. Erigeron neomezicanus and FE. oreophilus occur as low as 1100 meters. The most significant differences among the species of sect. Polyactis are in the type of pappus (coroniform or bristles) and the presence or absence of rhizomes, stipitate-glandular trichomes and clasping leaves. Erigeron poly- cephalus and E. annuactis have a distinctive leaf shape different from other species. Comments on possible interspecific relationships follow individual descriptions, but there is not sufficient basis for attempting to outline a hy- pothesis of phylogeny for all the species. I do believe, however, that the species of Erigeron sect. Polyactis with a coroniform pappus (vs. pappus of bristles) comprise a distinct lineage. This switch in pappus morphology apparently has entailed not only loss of the bristles, but the elaboration of a corona, which is a structure probably derived from the “outer” series of setae or scales. See the comments following E. coroniglandifer for a caveat regarding this hypothesis. SPECIES TAXONOMY Artificial key to the species of sect. Polyactis 1. Achenes with a coroniform pappus, without bristles ). Achenes with a pappus of bristles. ..... s .w.nscoows «sonwo ewe aneneant- (6) 420 PHYTOLOGIA volume 66(5):415-455 June 1989 2. Cauline leaves clasping or subclasping ....................25. (3) 2. Cauline,Jeaves not at all clasping» ...i2isss,20. ade (4) 3. Plants producing slender, scale-leaved rhizomes; upper stem pubescence appressed, eglandular® ©... 152202. ck. oot ee rae eee E. rhizomactis 3. Plants without rhizomes; upper stem pubescence spreading, usually prominently stipitate-glandular ................... E. coroniglandifer 4. Basal leaves strongly bipinnatifid, produced at tips of slender rhi- zomes; cauline leaves of erect stem ca 35; heads 6 mm wide; phyl- lariés 2/5-3.0 mm long... hie es Wace. eee E. circulis 4. Basal leaves entire to toothed or lobed, never strongly bipinnatifid; cauline leaves 10-30; heads 7-17 mm wide; phyllaries 3.1-5.5 mm jong?’ ss 22%..8: Sec bes SO, Po ee ee eee (5) . Phyllary margins thick, whitish; rays 115-180; pappus a corona 0.2-0.5 mm high, with an erose or scaly margin; receptacles alveolate FE. wis- hzent or 5. At least the inner phyllaries with relatively broad, hyaline margins; tays 40-230; pappus a blunt, cartilaginous rim less than 0.05 mm high; receptacles punctate to very shallowly alveolate ........... E. griseus 6. Plants annual 6. Plants perennial 7. Stems, leaves and phyllaries stipitate-glandular; leaves clasping to sub- Clasping: fo0/5. 222 be. Lae a ee ee E. inoptatus 7. Plants eglandular or the phyllaries sometimes minutely granular-glandular; leaves not, at all clasping. ...05...02-25.0% 20 vo, ac eee (8) 8. Cauline leaves pinnatifid to bipinnatifid ........ E. delphinifolius 8. Canline leaves entire .... 5s. acne ats ee E. annuactis Ne} . Cauline leaves clasping to subclasping 9. Cauline leaves not at all clasping 10. Stems eglandular, appressed-strigose at least near the heads; cauline leaves obovate to oblong-obovate, the midregions elliptic. SNe tee crc cree ae tect caso wereoe eee E. caulintfolius 10. Stems stipitate-glandular, hirsute-pilose; cauline leaves oblong- oblanceolate to oblong-lanceolate, the midregions narrowly oblong Se eee E. seemannit 11. Leaves serrate or crenate, the blades elliptic to elliptic- oblanceolate; capitulescence-corymbose