occasional papers of the Farlow Herbarium of cryptogamic botany No. 9 July, 1976 Harvard University, Cambridge, Massachusetts Donald H. Pfister | A Synopsis of the Genus Pulvinula A New Combination in the Genus Gymnomyces Norton G. Miller Studies on North American Quaternary Bryophyte Subfossils |. A New Moss Assemblage from the Two Creeks Forest Bed of Wisconsin Edited by: Reed C. Rollins Kathryn Roby occasional papers of the Farlow. Herbarium of cryptogamic botany No. No. No. No. No. No. No. No. No. 1. Sylvia A. Earle: Hummbrella, a New Red Alga of Uncertain Tax- onomic Position from the Juan Fernandez Islands (June 1969). . |. Mackenzie Lamb: Stereocaulon arenarium (Sav.) M. Lamb, a Hitherto Overlooked Boreal-Arctic Lichen (June 1972). . Sylvia A. Earle and Joyce Redemsky Young: Siphonoclathrus, a New Genus of Chlorophyta (Siphonales: Codiaceae) from Panama (July 1972). . |. Mackenzie Lamb, William A. Weber, H. Martin Jahns, Siegfried Huneck: Calathaspis, a New Genus of the Lichen Family Cladonia- ceae (July 1972). . |. Mackenzie Lamb: Stereocaulon sterile (Sav.) M. Lamb and Stereocaulon groenlandicum (Dahl) M. Lamb, Two More Hitherto Overlooked Lichen Species (March 1973). . |. Mackenzie Lamb: Further Observations on Verrucaria serpu- loides M. Lamb, the Only Known Permanently Submerged Marine Lichen (April 1973). . Bruce H. Tiffney and Elso S. Barghoorn: The Fossil Record of the Fungi (June 1974). . Donald H. Pfister: The Genus Acervus (Ascomycetes, Pezizales). 1. An Emendation. Il. The Apothecial Ontogeny of Acervus flavidus with Comments on A. epispartius (May 1975). . Donald H. Pfister: A Synopsis of the Genus Pulvinula. A New Combination in the Genus Gymnomyces. Norton G. Miller: Studies on North American Quaternary Bryophyte Subfossils. |. A New Moss Assemblage from the Two Creeks Forest Bed of Wisconsin (July 1976). A SYNOPSIS OF THE GENUS PULVINULA DonaLp H. PFIsTER! SUMMARY Seventeen species of the genus Pulvinula are discussed in this treatment. A synoptic key to these species is provided as are descriptions of most of them. A review of the pertinent literature is given and comments on the morphology of members of the genus are provided, One new species, P. neotropica Pfister is described and several new com- binations are made. Listed with the synonyms and excluded species of Pulvinula are a number of spherical-spored species of Pezizales which were examined in the course of this study and which are referred to other genera. INTRODUCTION Although Pulvinula Boudier is a well-defined member of the Pezizales, only Boudier (1907) previously attempted to treat all of the species of the genus together. The many species are scattered through the genera Crouania, Barlaea, Barlaeina, Detonia, and Lamprospora. To ferret them out, type studies have been necessary. In this preliminary report on the genus, I have brought together information gleaned from both the litera- ture and study of specimens. At present, single collections are heavily relied upon to provide details. In recognition of this, I have been very conservative in my treatment. I have preferred to cover collections which are at minor variance with described species by means of comments on those species rather than to describe new taxa prematurely. Therefore, the number of species treated is fewer than will finally be assigned to the genus. During the study of this genus for the Flora Neotropica project, it became apparent, as a practicality, that the temperate species required first attention. Nevertheless, many of the comments on species refer to collections from the Caribbean and will indirectly serve as a guide to the species of Pulvinula in that area. GENERAL COMMENTS The genus Pulvinula was originally mentioned by Boudier (1885) who listed Peziza convexella Karst., Peziza sanguinaria Cooke, and Peziza constellatio Berk. & Br. as species belonging to it. Later Boudier (1907) revised the nomenclature of two of the species and included a full generic description. At that time five species were included: Pulvinula cinnabarina (Fuck.) Boud., P. carbonaria (Fuck.) Boud., P. constellatio (Berk. & Br.) Boud., P. haemastigma (Wedw. ex Fr.) Boud. and P. subaurantia (Bomm. & Rouss.) Boud. The genus was not acknowledged until Le Gal's (1953) treatment of the genus from Madagascar. In the meantime, species of 1Farlow Herbarium and Department of Biology, Harvard University, Cambridge, Massachusetts 02138. 2 DONALD H. PFISTER this genus were referred to Barlaea, Barlaeina, Crouania, Detonia, and Lamprospora. Le Gal (1953) selected Peziza convexella as type of the genus from the three original species mentioned by Boudier. Both she and Boudier considered this species a synonym of Pulvinula haemastigma, a view which I do not hold. For further comment on P. haemastigma see the discussion under P. convexella. In recent treatments, only Rifai (1968), Moravec (1969) and Pfister (1972) have attempted to describe and delimit species of Pulvinula. These treatments are unfortunately limited by the geographical areas which were covered. Taxonomic criteria which I consider of value in this genus are: the size of the ascospores; the size of the ascus; the presence or absence of croziers; the size of the apothecia; apothecial color; the type of substrate on which apothecia are produced. To some extent the number of ascospores per ascus is useful but there is variation among collections in this respect. The number of ascospores per ascus ranges from eight to four. All the asci of a single apothecium rarely contain eight ascospores. Pulvinula tetra- spora and P. lacteoalba are the only species where the asci consistently have four spores. Ascospores of all members of the genus are smooth. The construction of the excipular tissue is uniform among the species. However, though useful and diagnostic at the genetic level, it is of little use in distinguishing among species. The medullary excipulum is com- posed of textura intricata, the hyphae of which are thin, rarely exceeding 4 »m in diam. The ectal excipulum is composed of textura globulosa, the cells of which are sometimes compressed, or of textura angularis. Figures 1, 2A, and 2B show details of cross sections of apothecia. The asci of a majority, though not all, of the species here placed in Pulvinula arise from prominent croziers which form a characteristic horseshoe-shaped, two-pronged base. This particular type of pleurorhy- chous crozier (Chadefaud, 1943) is illustrated diagramatically in figure 3. Crozier formation has been observed in the majority, though not all, of the species of Pulvinula. There is variation from species to species in the attenuation of the ascus base and position of the basal septa; however, this type of ascus is always obvious no matter what variation might accompany it. In one species, P. miltina, there are generally no croziers but in several specimens I have examined there were occasional croziers. This is unusual, since in other species croziers are either always present or always absent. The genus has been placed in the Aleurieae (Dennis, 1968; Rifai, 1968; Korf, 1972), based on the occurrence of guttulate spores, of hyphoid hairs when present, and of carotenoid pigments. I do not feel this position is satisfactory. The species of the genus form a discrete unit which only superficially resemble other members of this tribe. The uniqueness of the genus is also evident in Arpin’s (1969) chemical analysis of pigments of Pulvinula constellatio in which he found a new monocyclic carotenoid 4 DONALD H. PFISTER Fic. 2, A. Pulvinula albida. Portion of ectal excipulum. ru (isotype) & approx. 200. B. Pulvinula neotropica, Portion of ectal excipulum. ru (Pfister 811) X approx. 200. Scale equals 25 um. designated as P472. For the moment, however, rather than recognize yet another tribe in the heterogeneous family Pyronemataceae, I prefer to leave the genus in the Aleurieae. Since Pulvinula appears to be eymno- Pl g) hymenial, whereas at least some of the other Aleurieae are cleistohymenial, é / developmental studies may shed some light on its position. IDENTIFICATION OF SPECIES As an aid for the identification of species, I have constructed a synoptic key to the accepted species of Pulvinula using the method presented by Korf (1972). Using this key, the distinctive species can be easily identified. Certain species, however, cannot be separated without consulting descrip- tions for precise details of size, color, etc., which if included, would have made the synoptic key unwieldy. The numbers given in the key are those under which the name appears. The species are arranged in alphabetical order. If the fungus has more than one of the described characters, its numbers appear in lightface type. Those in boldface appear in only one category. 1. ASCI 1—1. Normal number of ascospores per ascus a. asci 4-spored 7, 16 b. (asci 8-spored or number of spores 8 or fewer within a single apothecium ) A SYNOPSIS OF THE GENUS PULVINULA 5 1—2. Form of ascus base a. (tapering toward the base ) b. ascus base abrupt 3, 9 1—3. Presence of croziers a. (croziers present ) b. croziers absent 9, 13, 15 1—4. Ascus length a. asci less than 150 um long 2, 6, 8, 17 b. asci 150-250 um long, 3, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 16, 17 c. asci greater than 250 um long 1, 4, 5, 13, 15, 16 2. ASCOSPORES 2—1. Ascospore size and shape a. ascospores elliptical 14 b. ascospores globose, 9-11 um in diam 1, 2, 6, 10, 12 c. ascospores globose, 11-16 um in diam 3, 6, 7, 8, 9, 11, 12, 13, 16, 17 d, ascospores globose, 16 wm in diam or larger 3, 4, 5, 15, 16, 17 3, PARAPHYSES 3—1. Branching at the apices a. (apices mostly unbranched ) b. apices branched 2, 4, 5, 8, 9 3—2. Apices of paraphyses a. (apices hooked or circinate ) b. apices straight 1, 6, 15 3—3. Paraphyses width a. (less than 2 um in diam) b. greater than 2 um in diam 4, 5, 9, 11, 15 4. SUBSTRATE 4—1,. Composition and condition of substrate a. (apothecia on soil) b. apothecia on burned material, scorched soil, etc. 2, 3, 8, 11 5. APOTHECIA 5—1. Apothecial color a, apothecia white or light colored 1, 6, 7, 12, 16 b. apothecia brightly colored 2, 3, 4, 5, 8, 9, 10, 11, 13, 14, 15, 17 ACCEPTED SPECIES 1. Pulvinula albida (Rick) Pfister, comb. nov. Fig. 2A = Detonia albida Rick, Brotéria 5: 29. 1906. Apothecia gregarious, 6-8 mm in diam. Disc white, margin somewhat convoluted at least when dry. Ectal excipulum up to 138-161 »m thick at the base of the apothecium, composed of globose to compressed globose cells 11-13 X 11-16 pm in diam, randomly arranged toward the outside and producing loose hyphoid hairs which form a tomentose outer layer. 6 DONALD H. PFISTER Medullary excipulum composed of textura intricata of hyphae 1-2.2 »m in diam. Subhymenium not a distinct layer. Hymenium about 280 um thick. Asci 265-282 16-20 um, 8-spored. Bases of asci with prominent two- pronged croziers. Ascospores uniseriate, globose, hyaline, generally with no prominent oil globule, smooth-walled, 10.2-12 pm diam. Paraphyses gracile, 1-1.5 wm at base, 1.5-2 pm apically, straight or only slightly curved. SPECIMEN EXAMINED. Brazil. On soil, Sad Leopoldo, Rio Grande do Sul, 1932, det. Rick (FH). The specimen in the Farlow Herbarium general mycological collection, which appears to be part of the type collection, is the only specimen of this species I have seen. The species is close to the neotropical examples of Pulvinula globifera but differs in that the asci in the Rick specimen are much longer. It is also similar to P. convexella but differs in the form of the paraphyses which are straight in P. albida rather than curved as in P. convexella and the apothecia are white rather than orange. 2. Pulvinula archeri ( Berk. in Hook.) Rifai = Pexiza archeri Berk. in Hook., The Botany of the Antarctic Voyage. II. Flora Tasmaniae 2:274. 1859. = Barlaea archeri ( Berk.) Sacc., Syll. Fung. 10:5. 1892. = Humaria archeri ( Berk.) Cooke, Handbook of Aus- tralian Fungi. p. 256. 1892. = Barlaeina archeri ( Berk.) Sacc. & Trav. in Sacc., Syll. Fung. 20:138. 1910. = Pulvinula archeri (Berk.) Rifai, Verh. K. ned. Akad. Wet. II 57(3) :213. 1968. = Pexiza gemmea Phil. in Cooke, Mycographia p. 236, fig. 398. 1879. = Barlaea gemmea (Phil. in Cooke) Sacc., Syll. Fung. 8:112. 1889. — Lamprospora pyrophila Snider, Mycologia 28:484. 1936. ? = Barlaeina strasseri Bres. in Strasser, Verh. Zool.—Bot. Ges. Wein 55:613. 1905. Apothecia gregarious to scattered up to 7 mm in diam. Disc orange, convex, margins more or less even. Ectal excipulum relatively small-celled, cells (10-) 13-15 am in diam, globose to subglobose. Medullary excipulum of textura intricata, hyphae 2 pm in diam. Hymenium less than 150 pm thick. Asci 135-145 x 10-11 »m with or without prominent croziers, taper- ing toward the base. Ascospores globose, generally with a single large oil droplet, smooth-walled, 9-10.5(-11.5) »m in diam. Eight or fewer asco- spores per ascus. Paraphyses slender, 1-2 »m in diam, circinate and pig- ment filled, mostly branched apically. SPECIMENS EXAMINED. Austria. Holotype of Barlaeina strasseri, Sonntagberg 26. IX. 1904 (s). U.S.A. Holotype of Peziza gemmea, on fallen leaves of Sequoia sempervirens (Harkness 876) (x); on soil in burn area, Benton Co., Mary's Peak, Philomath, Ore., H. J. Larsen (no. 247), 11. IX. 71 (osu); on burned material, Polk Co., Camp Kilowan and vicinity 5 mi. S. of Falls City, Ore., H. J. Larsen, 27. V. 1972 (osu); on soil in fire spot, Benton Co., Mary’s Peak, Philomath, Ore., H. J. Larsen (no, 252), 17. IX. 1971 (osu). A SYNOPSIS OF THE GENUS PULVINULA 7 | Fic. 3. A-H. Diagrams illustrating the formation and development of asci in most species of Pulvinula. A collection from Oregon (Benton Co., Mary’s Peak Campground, 27. V. 1972) sent for examination by Harold Larsen agrees with this species except in the form of the paraphyses, where they are practically straight and mostly unbranched. Based on collections I have seen from the western United States, I pre- sume this is the most common species found on burned wood and sur- rounding soil in that part of the country. Both Peziza gemmea and 8 DONALD H. PFISTER Lamprospora pyrophila were described from material from the western states. The holotype of L. pyrophila in the mycological herbarium at the University of Washington was not available for study. 3. Pulvinula carbonaria (Fuck. ) Boud. = Crouania carbonaria Fuck., Jahrb. Nassauischen Vereins Naturk. 27-28 :64. 1873. = Peziza sanguinaria Cooke, Mycographia p. 14, fig. 19. 1879 (a name change). = Pulvinula carbonaria (Fuck.) Boud., Hist. Class. Discom. d’Eur. p. 70. 1907. Apothecia gregarious, 1-3 mm in diam. Dise at first subconcave then flat to repand, orange-scarlet. Receptacle smooth, lighter than the disc. Ectal excipulum about 115 »m thick, composed of textura globosa to com- pressed angular cells, 9-11 16-23 um. Ectal excipulum of textura in- tricata, hyphae about 2 pm in diam. Subhymenium not a distinct layer. Hymenium about 215 nm thick. Asci 196-230 »m long, 4-8 spored. Ascus base moderately broad with a prominent crozier. Ascospores uniseriate, globose, hyaline, with a single large guttule and some smaller satellite guttules, smooth-walled, 15-17 ym diam. Paraphyses gracile, up to 1.5 pm in diam throughout; mostly strongly bent or curved apically with abun- dant carotenoids. SPECIMENS EXAMINED. Fuckel, Fungi rhenani no. 2482 (ru). U.S.A.: Burnt ground in woods, Burbank, E. Tenn., R. Thaxter (7944), Aug. 20-Sept. 5, 1887 (rH). Burnt spot in hemlock woods, Purgatory Swamp, Norwood, Mass., D. H. Linder, 27. IX. 1936 (FH). 4. Pulvinula cinnabarina ( Fuck.) Boud. = Crouania cinnabarina Fuck., Jahrb. Nassauischen Vereins Naturk, 27-28:64. 1873. = Peziza laeterubra Cooke, Mycographia p. 14, fig. 20. 1879 (a name change, non Peziza cinnabarina Schw.). = Lamprospora laeterubra (Cooke) Lagarde, Ann. Mycol. 4:217. 1906. = Pulvinula cin- nabarina (Fuck.) Boud., Hist. Class. Discom. d’Eur. p. 70. 1907. [= Lamprospora cinnabarina (Fuck.) Moser in J. Gams, Kleine Kryptogamen- flora 2a:112. 1963. ( basionym not cited) ]. Apothecia usually gregarious, up to 7 mm in diam. Disc vermillion, drying red, margin wavy. Receptacle smooth where exposed. Ectal ex- cipulum up to 115 »m thick at the base of the apothecium; composed of radially arranged elongate cells, 7 < 11 um in diam. Medullary excipulum composed of textura intricata of hyphae up to 4 um in diam. Subhymenium not a distinct layer. Hymenium about 280-320 ym thick. Asci 260-275 « 20-23 wm, 4-8-spored (8 predominating in type collection), The ascus tapers toward the base with a definite crozier. Ascospores uniseriate, hyaline with one or a few guttules, smooth-walled, 16-20 pm in diam. Paraphyses thin, up to 2.2-4 »m in diam throughout, curved or only slightly curved apically, sometimes branched. A SYNOPSIS OF THE GENUS PULVINULA 9 SPECIMENS EXAMINED. Fuckel, Fungi rhenani no. 2481 (FH). U.S.A.: On soil among mosses, Deep Creek, Great Smoky Mountains National Park, Tennessee, 11. XI. 1970. P. E. Powell and D. H. Pfister (349) (FH). 5. Pulvinula convexella (Karst. ) Pfister, comb. nov. = Peziza convexella Karst., Not. Sallsk. Fauna Fl. Fenn. Forh. 10:123. 1869. == Leucoloma convexellum ( Karst.) Rehm, Ascomycetes Lojkani p. 8. 1882. == Barlaeina convexella ( Karst.) Sacc., Syll. Fung. 18:24. 1906. = Peziza constellatio Berk. & Br., Ann. Mag. Nat. Hist. Ser. 4. 17:142. 1876. = Leucoloma constellatio (Berk. & Br.) Rehm, Ber. Naturhist. Vereins Augsburg 26:5. 1881. = Aleuria constellatio (Berk. & Br.) Gill., Champ. France, Discom. p. 207. 1888. = Plicariella constellatio Lindau in Engler & Prantl, Nat. Pflanzenfam. 1(1):180. 1897. = Pulvinula con- stellatio (Berk. & Br.) Boud., Hist. Class. Discom. d’Eur. p. 70. 1907. = Barlaeina constellatio (Berk. & Br.) Rehm in Dodge, Trans. Wisconsin Acad. Sci. 7:1037. 1914. = Crouania livida Rehm in Hazsl., Math. Naturwiss. Ber. Ungarn 21: 262. 1886. = Barlaeina livida (Rehm in Hazsl.) Sacc. & Trott. in Sacc., Syll. Fung. 22:621. 1913. Misapplication: Pulvinula haemastigma (Hedw. ex Fr.) Boud. sensu Boud., Hist. Class. Discom. d’Eur. p. 70. 1907. Apothecia less than 8 mm in diam. Disc yellow-orange. Ectal excipulum of globose, subglobose to angular cells. Medullary excipulum of narrow diameter hyphae forming textura intricata. Asci 230-255(-270) pm X 16-20 pm. Eight spores or sometimes fewer per ascus. The base of the asci with prominent two-pronged croziers. Ascospores smooth-walled, uniseriate, globose, hyaline with one or more oil droplets, (16—) 18-20 pm in diam. Paraphyses up to 2.3 pm in diam, moderately curved and reg- ularly branched apically. This is the type species of the genus Pulvinula selected by Le Gal (1953). Boudier (1907) synonymized this species with P. haemastigma (Hedw. ex Fr.) Boud., a synonymy which has been since followed by most workers, including Le Gal. I do not follow this position and consider P. haemastigma a nomen confusum. My reason is that the illustration upon which the species concept must now be based, in the absence of a speci- men, is not adequately diagnostic in several critical features, such as the form of the paraphyses, the form of the apothecium, and the configuration of the crozier at the ascus base. In addition, there is no information on ascus or ascospore size. Boudier (1907) pointed out that Saccardo (1889) had interpreted Peziza haemastigma differently than he had. Neither Le Gal nor subsequent workers have designated a lectotype from among the eligible specimens in the Karsten Herbarium. There are two specimens bearing practically the same information which might have been referred to by Karsten. Having examined the specimens, I designate number 1273 lectotype of P. convexella. 10 DONALD H. PFISTER There have been five varieties and one subspecies described and at- tributed to Pulvinula convexella or to specific names listed in its syn- onymy. These are listed below. Such a proliferation of infraspecific taxa suggests something of the diversity within this complex species. Peziza constellatio var. fuckelii Cooke, Mycographia p. 45, fig. 82. 1879. Barlaea constellatio var. minuta F. E. Clements, Bot. Surv. Nebraska 4: 10. 1896. Pulvinula convexella subsp. tjibodensis Penz. & Sacc., Malpigia 16: 202. 1901) Detonia constellatio var. aurantiaca F, E. Clements & E. §. Clements, Cryptogamae Formationum Coloradensium no. 114. 1906. Lamprospora haemotostigma [sic] var. gigantea Thind & Batra, Jour. Indian Bot. Soc. 38: 221. 1959. Pulvinula haemastigma var. luteoflava Moravec, Ceska Mykol. 23: 25. 1969. CRITICAL SPECIMENS EXAMINED. Finland, Lectotype: Tavastia australis, Tammela, Mustiala, ad terram nudam ad rivuli marginem in horto, 30. VIII. 1867, leg. et det. P. A. Karsten (Herb. Karsten 1273); Travastia australis, Tammela, Saloris (=Soloinen), in via, 6. VII. 1878, leg. & det. P. A. Karsten (u, Herb. Karsten 1275); Tavastia australis, Tammela, Mustiala, 30. VIII. 1867, leg. et det. P. A. Karsten (u, Herb. Karsten 1272). U.S.A. On moist rotted log in shade in spruce-fir forest, Gothic, Colorado, leg. Mary Alt, 8. VII. 1963 (osu) and eight additional specimens from North America in FH general herbarium. A number of specimens from several exsiccati examined in the course of this study were issued under one of the synonymous names. Those which are unquestionably referrable to Pulvinula convexella are as follows: H. Sydow and P. Sydow, Mycotheca germanica, nos. 496 (sub Barlaea con- vexella), 1620, 1621 (sub Barlaea constellatio); P. Sydow, Mycotheca Marchica, no. 2862 (sub Crouania constellatio); H. Rehm, Ascomyceten, no. 406 (sub Detonia constellatio); E. Bartholomew, Fungi Columbiani, no. 2909 (sub Barlaea subaurantia ); J. B. Ellis and B. M. Everhart, North American Fungi, no. 2306 (sub Barlaea constellatio); F. E. and E. S. Clements, Cryptogamae Formation Coloradensium, no. 114 (sub Detonia constellatio aurantiaca); A. Allescher and J. N. Schnal, Fungi Bavarici Exsiccati, no. 942 (sub Barlaea constellatio ); W. Krieger, Fungi Saxonici, no. 1038 (sub Barlaea constellatio ). 6. Pulvinula globifera ( Berk. & Curt.) Le Gal = Peziza globifera Berk. & Curt., J. Linn. Soc., Bot. 10: 366. 1868. == Barlaea globifera (Berk. & Curt.) Sacc., Syll. Fung. 8: 114. 1889. = Hum- aria globifera (Berk. & Curt.) Cooke, Handbook of Australian Fungi p. 256. 1892. = Barlaeina globifera (Berk. & Curt.) Sacc. & Trav. in Sace., Syll. Fung. 19: 139. 1910. == Pulvinula globifera (Berk. & Curt.) Le Gal, Prodr. Flore Mycol. Madagascar 4:94. 1953. My Caribbean collections of Pulvinula globifera have all been white, as described earlier (Pfister, 1972). Rifai (1968) also gives a full descrip- tion of Pulvinula globifera (Berk. & Curt.) Le Gal. The geographical range and the degree of intraspecific variation remain to be clarified. Rifai pointed out that Le Gal (1953) misinterpreted this species, and A SYNOPSIS OF THE GENUS PULVINULA 11 considers the species description by Le Gal to be P. orichalcea (Cooke ) Rifai. The collection studied by Gamundi (1966) is also probably P. orichalcea although I have not seen her specimen. The morphology of the paraphyses of P. globifera is variable; occasionally they are curved and at times become inflated apically and below the septa. This is especially evi- dent at the margin. W. C. Denison’s collection from Arizona (osu 23,991) might also be referred to Pulvinula globifera. | have no data on the color or size of the apothecia when fresh. Two varieties of Pulvinula globifera have been described. They are listed below, but no attempt has been made to reevaluate their status. Both have recently been treated. Rifai (1968) reported that Peziza globi- fera var. etiolata was not significantly different from the holotype of Peziza globifera to consider it distinct. I have not reexamined Cooke's specimen. Barlaea globifera var. sphaeroplea (Berk. & Curt.) Sace., Syll. Fung. 8:114. 1889. = Sphaerosporella brunnea (Alb. & Schw. ex Fr.) Svréek & Kubicka fide Rifai (1968). Peziza globifera var. etiolata Cooke, Mycographia p. 236, fig. 399. 1879. = Pulvinula etiolata (Cooke) Le Gal, Prodr. Flore Mycol. Madagascar 4:91. 1953, see Rifai (1968). SPECIMENS EXAMINED: eleven specimens from Puerto Rico and Guadeloupe (FH). 7. Pulvinula lacteoalba J. Moravec, Ceska Mykol. 23: 231. 1969. This is probably a 4-spored form of Pulvinula globifera and should not, perhaps, be considered a distinct species. For a description see the original publication by Moravec. SPECIMENS EXAMINED. Holotype: Ad terram nudam in olla cum Sparmania africana in domo-Liban, districtus Jiecin, 22. V. 1966, leg. Jiri Moravec (PR 674713). 8. Pulvinula laeterubra (Rehm) Pfister, comb. nov. = Barlaea laeterubra Rehm, Ann. Mycol. 5: 516. 1905. = Barlaeina laeterubra (Rehm) Sacc. & Trott. in Sacc., Syll. Fung. 22: 622. 1913, == Detonia laeterubra (Rehm) Dodge, Trans. Wisconsin Acad. Sci. 17: 1037. 1914, = Lamprospora wisconsinensis Seaver, North American Cup-Fungi (Operculates ) p. 69. 1928 (a name change). Apothecia gregarious, 14 mm diam. Disc salmon-red (sometimes yellow ). Ectal excipulum 90-115 pm thick composed of textura angularis, the cells of which are about 11 17 pm in diam, cells more or less radially arranged. Medullary excipulum composed of textura intricata of hyphae 2 wm in diam. Subhymenium not a distinct layer. Hymenium about 180 »m thick. Asci (140-) 161-184 (-200) um, 1-8 ascospores per ascus, ascus base has a prominent two-pronged crozier. Ascospores uniseriate, globose, hy- 1 DONALD H. PFISTER aline generally with indiscrete oil droplets, smooth-walled, 11-13.8 »m in diam. Paraphyses thin (about 1 um), curved and commonly branched. SPECIMENS EXAMINED. U.S.A. Holotype of Barlaea laeterubra, [on soil], Madison, Wisconsin, 7. V. 1904, Harper (412) (s, Herb. Rehm), isotype (FH). Also six speci- mens in FH general herbarium. There are two Puerto Rican collections (DHP 1320 and 1329) which differ slightly from Pulvinula laeterubra: they are yellow, and the para- physes are broader. Seaver (1928) proposed the name Lamprospora wisconsinensis, when he treated Barlaea laeterubra in Lamprospora, since he also treated Lamprospora laeterubra (Cooke) Lagarde. See the list of excluded species for comments on this fungus. He gave the spore size as smaller than 10 pm, which is not the case in the type collection where the ascospores are 12- 14 »m in diam. Red-orange North American collections with spores from 12-14 um in diam, have been erroneously referred to Pulvinula constellatio and P. convexella. They should be referred to Pulvinula laeterubra. 9. Pulvinula miltina ( Berk. in Hook.) Rifai = Peziza miltina Berk. in Hook., Fl. Nov. Zealand. 2: 199. 1855, = Barlaea miltina ( Berk.) Sacc., Syll. Fung. 8: 113. 1889. = Humaria miltina (Berk.) Cooke, Handb. Austral. Fungi. p. 256. 1892. = Barlaeina mil- tina ( Berk.) Sacc. & Trav. in Sacc., Syll. Fung. 19: 139, 1910. = Pulvinula miltina ( Berk.) Rifai, Verh. K. Ned. Akad. Wet. II 57(3): 204. 1968. For a description of this species of Pulvinula see Rifai (1968). Several Caribbean collections are similar to Pulvinula miltina but with spores 9-11.5 ym in diam, and asci 140-165 pm long. These measurements are smaller than those given for P. miltina by Rifai. The Caribbean col- lections are all on charred wood. They have the same broad ascus base as is present in P. miltina, but might well represent an undescribed species. These collections are listed below. SPECIMENS EXAMINED. Guadeloupe, F.W.I. On charred wood and surrounding soil, Le Mamelles, Guadeloupe National Park, January 5, 1974 (DHP 813) Fx. Puerto Rico. On burned bamboo, El Verde, Luguillo National Forest, April 8, 1974. (DHP 1331) Fu; on burned branches, as above (DHP 1325) Fu. 10. Pulvinula mussooriensis (Thind, Cash & Singh) Batra & Batra = Lamprospora mussooriensis Thind, Cash & Singh, Mycologia 51: 457. = Pulvinula mussooriensis (Thind, Cash & Singh) L. R. Batra & S. W. T. Batra, Kansas Univ. Sci. Bull. 44: 167. 1963. Pulvinula mussooriensis is very near P. niveoalba except in color and in ascus length. In P. mussooriensis the asci exceed 200 »m and the hymen- ium is yellow, whereas in P. niveoalba the asci are 160-180 wm and the hymenium is white. Both have variable spore numbers with 8 spores per ascus predominating. Whether the two species should be distinguished A SYNOPSIS OF THE GENUS PULVINULA 13 from each other is a question which can be resolved only by additional field collections and by additional distributional data. Both are known only from the type collections. SPECIMEN EXAMINED. India. Isotype of P. mussooriensis, (Brewery Road, Mussoorie), on soil amid mosses, Aug. 12, 1956 (spr). 11. Pulvinula neotropica Pfister, sp. nov. Fig. 2B Apothecium ad 4 mm diam, convexum, flavovirens, in sicco luridum. Asci 165-177 11-14 pm, octospori. Ascospori globosi laevigati (12—) 13— 14(-15) pm. Paraphyses filiformae ad 4 »m diam ad apices. In ligno ustulato, Holotypus: on burned wood, El Yunque, Luquillo National Forest, Puerto Rico, D. H. Pfister (1342) and J. D. Rogers, 9. IV. 1974 in FH. Apothecia gregarious, 24 mm diam. Disc pale yellowish-greenish, con- vex, drying buff, margin more or less even. Receptacle pulvinate, smooth or slightly furfuraceous. Ectal excipulum 30-45 um thick, of globose cells and/or laterally compressed cells, cells 10-15 > 25-32 ym arranged radially, 3-5 cells deep. Medullary excipulum composed of more or less tightly interwoven textura intricata of hyphae, about 2 »m in diam. Sub- hymenium not clearly differentiated from the medullary excipulum. Hy- menium about 200 pm thick. Asci 165-177 « 11-14 um, arising from prominent croziers, not broad at base, 8-spored. Ascospores uniseriate, globose, hyaline, with a single large oil globule, smooth-walled (12) 13- 14(-15) pm. Paraphyses thin, 2-3 pm below, 3-4 »m toward the apex, mostly curved in the upper portion but not strongly. SPECIMENS EXAMINED. Puerto Rico. Holotype: on burned wood, El Yunque, Luquillo National Forest, D. H. Pfister (1342) and J. D. Rogers, 9.1V.1974 (FH); on charcoal in fire spot, El Toro Trail, El Yunque, D. H. Pfister et al. (639), 9.VII.1973 (FH). Guadeloupe, F.W.I. On soil and charred wood, Les Mamelles, Guadeloupe National Forest, 5.1.1974, D. H. Pfister (811), Martha Sherwood, and Steve Carpenter (FH); burn site, Parc Tropicale, D. H. Pfister (1249), S. Carpenter, M. Sherwood, 10.1.1974 (FH). 12. Pulvinula niveoalba J. Moravec = Pulvinula niveoalba J. Moravec, Ceska Mykol. 23:231. 1969. Pulvinula niveoalba can be distinguished by its relatively small apothe- cia (0.6-3 mm in diam), white hymenium, ascus size (160-180 »m) and ascospore dimensions (9.5-12.2 ym ). SPECIMEN EXAMINED. Czechoslovakia. Holotype, ad terram humidam nudam in societe Pulvinulae haemastigmae, viae cavae in piceto prope Drhleny, districtus Mlada Boleslav, 23. VI. 1969, leg. Jiri Moravec (PR 674714). 13. Pulvinula orichalcea (Cooke ) Rifai = Peziza orichalcea Cooke, Mycographia p. 235, fig. 397. 1879. = Barlaea orichalcea (Cooke) Sacc., Syll. Fung. 8:114. 1889. = Barlaeina 14 DONALD H. PFISTER orichalcea (Cooke) Sacc. & Trav. in Sacc., Syll. Fung. 19: 140. 1910. = Pulvinula orichalcea (Cooke) Rifai, Verh. K. Ned. Akad. Wet. II. 57(3): 213. 1968. Le Gal's (1953) description of Pulvinula globifera is, according to Rifai, based on P. orichalcea. I have seen two collections of this species from the Caribbean. SPECIMENS EXAMINED. Jamaica. On soil, trail from Whitfield Hall to Portland Gap, to Blue Mountain Border of St. Thomas and Portland Parishes, R. P. Korf et al., 17. I. 1971 (CUP-MJ-593 ); on soil, Traveller’s Rest, Silver Hill Gap on the border of Port- land and St. Andrew Parish, R. P. Korf et al., 8. I. 1971 (CUP-MJ-103). 14. Pulvinula ovalispora Boud. Fig. 1 = Pulvinula ovalispora Boud., Bull. Soc. Mycol. France 33: 16. 1917. This is the only species of the genus reported to have ellipsoidal spores. When describing Pulvinula ovalispora, Boudier indicated that, although the spores were oval, all other characteristics agree with those of the genus. The ascus bases are forked as is typical of the genus. The type collection consists of a single, small, partially dissected apothecium, thus details of the excipulum were not studied. SPECIMEN EXAMINED. Algeria, Ad terram nudam, Februario, leg. Rene Mairé (pc). 15. Pulvinula salmonicolor (Seav. ) Pfister = Lamprospora salmonicolor Seaver, Mycologia 17: 47. 1925. = Pulvi- nula salmonicolor ( Seay.) Pfister, Phytologia 24: 211. 1972. Apothecia usually solitary, 5-7 wm diam. Disc yellow-orange to orange, convex, drying buff-salmon, margin even. Receptacle pulvinate, smooth where exposed. Ectal excipulum up to 65 »m thick at the base of the apothecium, composed of globose to slightly laterally compressed cells, 30-35 x 40-45 pm, becoming somewhat radially arranged at the margin, random at the base of the apothecium. Medullary excipulum composed of tightly interwoven hyphae (textura intricata), 2-3 wm in diam. Sub- hymenium not clearly differentiated from the medullary excipulum. Hymenium about 375-400 ym thick. Asci 242-286 > 22-23 ym, 8-spored, ascus base lacking croziers. Ascospores uniseriate, globose, hyaline, gen- erally with a single oil globule, smooth-walled though the cytoplasm sometimes gives them a granular appearance, 20-23 um diam. Para- physes stout, 4 wm at the base, 7-10 pm apically, straight. SPECIMENS EXAMINED. Guadeloupe, F.W.L. On soil among mosses, La Soufriére, July 20, 1973, D. H. Pfister (594) and W. Sarriera; on soil, Les Mamelles, Guadeloupe National Forest, 5.1.1974, D. H. Pfister (887), Martha Sherwood, and Steven Carpenter (FH); on soil at base of a banana plant, Camp Jacob, Saint Claude, 500-550 m, 7.1. 1974, D. H. Pfister (1063), Martha Sherwood, and Steven Carpenter (FH); on soil, Pare Tropicale, Basse Terre, 10.1.1974, D. H. Pfister (1197), Martha Sherwood, and Steven Carpenter (FH), Puerto Rico. On soil among mosses, El Yunque, 9.IV.1974, D. H. Pfister (1340 & 1343). A SYNOPSIS OF THE GENUS PULVINULA 15 The species was discussed earlier (Pfister, 1972). Recent collections have permitted this more complete description. 16. Pulvinula tetraspora (Hansf.) Rifai = Lamprospora tetraspora Hansf., Proc. Linn. Soc. New South Wales 79:126. 1954. == Pulvinula tetraspora (Hansf.) Rifai, Verh. K. ned. Akad. Wet. II. 57(3) :207. 1968. Misapplication: Pulvinula etiolata (Cooke) Le Gal, Prodr. Flore Mycol. Madagascar 4:91. 1953. = Peziza globifera var. etiolata Cooke, Myco- graphia p. 236, fig. 399. 1879. Rifai (1968) provides a complete description of this species, which, in addition to Pulvinula lacteoalba, are the only two species thus far known to have exclusively 4-spored asci. 17. Pulvinula sp. There is a distinct species of Pulvinula which resembles P. convexella and P. carbonaria but which differs in that the apothecia are on soil rather than on charcoal and the asci are smaller than in P. convexella. The only specimen of this species I have seen is in Fuckel, Fungi rhenani no. 2290) (FH). This specimen was given the name Crouania humosa (Fr.) Fuckel. However, Fuckel’s description and specimen do not agree with Fries concept of the species, thus that epithet is not available. Future type studies may provide an available name for this species. SYNONYMS, EXCLUDED SPECIES AND COMMENTS ON SPHERICAL-SPORED SPECIES OF PEZIZALES In searching for species of Pulvinula, a number of species were exam- ined which were referrable to other genera of the Pezizales or were synonyms of accepted species of Pulvinula. They are listed alphabetically below under the accepted name if one could be determined. LAMPROSPORA ASPERELLA (Rehm) Boud., Hist. Class. Discom. d’Eur. p. 69. 1907. = Crouania asperella Rehm, Hedwigia 24:226. 1885. = Barlaea asperella (Rehm) Sacc., Syll. Fung. 8:113. The placement of this species in Lamprospora by Boudier (1907) seems to be correct. The structure of the excipulum agrees with that found in Lamprospora sensu Rifai (1968). Seaver (1928) synonymized it with L. crec hqueraulti. LAMPROSPORA ASTROIDEA (Hazsl. in Cooke) Boud. Hist. Class. Discom. d'Eur. p. 68. 1907. = Peziza astroidea Hazsl. in Cooke, Mycographia p. 29, fig. 49. 1879. = Barlaea astroidea (Haszl. in Cook) Sace., Syll. Fung. 8: 111. 1889. Peziza astroidea was treated by Boudier (1907) as Lamprospora. I have not seen a specimen of this species but follow Boudier’s placement 16 DONALD H. PFISTER of it. Cooke’s plate shows the apothecia with fimbriate margins which would certainly substantiate Boudier’s deposition of it. BARLAEINA CENTROSPORA Kirschst., Ann. Mycol. 33:206. 1935. This is a species of Lamprospora sensu stricto and is close or identical to Lamprospora crechqueraulti. The specimen should be reexamined when a critical revision of the genus Lamprospora is done. Holotype in B examined. PLICARIA CHAIGNONI Pat., Bull. Soc. Hist. Nat. Autun. 17:154. 1904. = Pulparia planchonis (Dun. ex Boud.) Korf, Pfister, and Rogers, Phytologia 21:206. 1971. Holotype in rH examined. LAMPROSPORA CHOPRAIANA Batra, Mycologia 52:665. 1960. This is not a Pulvinula but type material is too scanty to make an identi- fication. Rifai (1968) excluded it from Lamprospora. PULVINULA CONSTELLATIO (Berk. & Br.) Boud., Hist. Class. Discom. d’Eur. p- 70. 1907. Pulvinula constellatio has been accepted as a distinct taxon by most workers. I feel it intergrades with P. convexella and tentatively regard it as a synonym as did von Hohnel (1917). Pulvinula constellatio is part of a complex of species, including P. cinnabarina, P. convexella, and P. laeteru- bra, which is yet to be adequately defined. Complete synonymy of this species is given under P. convexella. BARLAEA DISCOIDEA P. Henn. & E. Nym., Monsunia 1:33. 1900. = Barlaeina discoidea (P. Henn. & E. Nym.) Sacc. & Syd., Syll. Fung. 16:710. 1902. = Pulvinula discoidea (P. Henn. & E. Nym.) Batra, Univ. Kansas Sci. Bull. 44:143. 1963. The type material of this taxon was not located. In this group, where microscopic features are critical for proper and accurate identification, no attempt is made to place it. PULVINULA ETIOLATA (Cooke) Le Gal, Prodr. Flore Mycol. Madagascar 4:91. 1953. = Peziza globifera Berk. var. etiolata Cooke, Mycographia p. 236, fig. 399. 1879. Rifai (1965) considered Le Gal’s use of this name a misapplication. According to him, the specimen described is referrable to Pulvinula tetra- spora. Peziza globifera var. etiolata, according to Rifai, is indistinguishable from Pulvinula globifera. I have not studied Cooke's specimens. BARLAEINA FEURICHIANA Kirschst., Ann. Mycol. 33:205. 1935. This is a species of Lamprospora sensu stricto. Holotype in B examined. PEZIZA GEMMEA Phil. in Cooke, Mycographia p. 236, fig. 398. 1879. This is a synonym of Pulvinula archeri (Berk. in Hook.) Rifai. A SYNOPSIS OF THE GENUS PULVINULA 17 BARLAEINA HENNINGSI Kirschst., Notizbl. Bot. Gart. Berlin-Dahlem 15:830. 1943. This is a species of Lamprospora, close or identical to L. crec hqueraulti. Holotype in B examined. CROUANIA KNAJASCHENSIS Karst., Hedwigia 23:37. 1884. = Barlaea knajas- chensis (Karst.) Sacc., Syll. Fung. 8:113. 1889. = Lamprospora knajaschen- sis (Karst. ) Boud., Hist. Class. Discom. d’Eur. p. 68. 1907. No specimen of this species could be located in the Karsten Herbarium at Helsinki. From the original description, it appears to be a species of Pulvinula since it is said to have smooth spores, curved paraphyses which are relatively narrow, and asci within the proper size range. However, it was treated by Boudier (1907) as a Lamprospora. Without a specimen the question is irresolvable. PEZIZA LAETERUBRA Cooke, Mycographia p. 14, f. 20. 1879. = Lamprospora laeterubra (Cooke) Lagarde, Ann. Mycol. 4:213. 1906, This name was proposed as a substitute for the epithet cinnabarina when Cooke wished to transfer Crouania cinnabarina Fuck. to Peziza. Both Cooke (1879) and Lagarde (1906) described and illustrated the globose ascospores as being finely reticulate. I have not examined their specimens, but Fuckel’s specimen in Fungi rhenani no. 2481 is definitely a Pulvinula and has smooth ascospores. The fungus with ornamented spores illustrated and discussed by Cooke and Lagarde is apparently without a legitimate name. CROUANIA Livipa Rehm ex Hazsl., Math. Naturwiss. Ber. Ungarn. 21:262. 1886. This is a synonym of Pulvinula convexella, which see for complete syn- onymy. Apparently Rehm did not intend to publish this as a new species since the packets are annotated as Peziza convexella Karst. and the species was published as Leucoloma convexellum (Rehm 1882). Specimen examined: in excavatione terrae callis “Riu mare” infra alpen Retyezat Transsylvaniae 8/1872, Lojka sub 1870. BARLAEINA PLATENSIS Speg., Anales Mus. Nac. Hist. Buenos Aires ser: 3. 1:70. 1902. The ascospores of this species are not spherical but rather are broad ovoid-ellipsoid, 10-12 % 9-10 »m in diam and smooth. The hairs are flexuous and blunt tipped and arise from = globose cells of the outer layer of the ectal excipulum. The cells of the ectal excipulum are brown colored. The paraphyses are characteristic in that they are swollen apically and many times swollen again below the septa giving the paraphyses a slim moniliform appearance. This may represent a species of Leucoscypha. Holotype in Lps examined. 18 DONALD H. PFISTER LAMPROSPORA PYROPHILA Sydner, Mycologia 28:484. 1936. The holotype is deposited at the University of Washington herbarium but for some reason is not available for study. I wish to thank Dr. Amy Y. Rossman of Oregon State University for locating the specimen in the Uni- versity of Washington herbarium. Since a number of specimens from the northwest were examined and found to agree with both the original description of Lamprospora pyrophila and with Rifai’s (1968) description of Pulvinula archeri, Lamprospora pyrophila is here considered to be a synonym of Pulvinula archeri. BARLAEINA ROSEA Kirschstein in Schieferdecker, Die Schlauchpilze der Flora von Hildescheim p. 97. 1954. This is a species of Lamprospora. It appears to be very similar to L. dictydiola Boud. Holotype in B examined. PEZIZA SANGUINARIA Cooke, Mycographia p. 14, fig. 19. 1879. The name was introduced by Cooke as a nomen novum to replace the epithet carbonaria when he transferred Crouania carbonaria to Peziza. See Pulvinula carbonaria for complete notes on this species. BARLAEINA STRASSERI Bres. in Strasser, Verh. Zool.—Bot. Ges. Wein 55: 613. 1905. This is a Pulvinula close, if not identical, to Pulvinula archeri. The small ascospores, 8-10 wm in diam, and the apically branched paraphyses are very characteristic. Holotype in B examined. PEZIZA SUBAURANTIA Bomm. & Rouss., Bull. Soc. Roy. Bot. Belgique 23: 134. 1884, = Barlaea subaurantia (Bomm. & Rouss.) Sacc., Syll. Fung. 8: 114. 1889. = Pulvinula subaurantia (Bomm. & Rouss.) Boud., Hist. Class. Discom. d’Eur. p. 70. 1907. I was unable to locate the holotype of this species in either BR or in PC, thus its placement within the genus is unresolved. The specimen in Fungi Colombiani no. 2909 issued under this name is Pulvinula convexella. ACKNOWLEDGEMENTS I wish to thank the curators and directors of the following herbaria for searching for and forwarding specimens to me for study: The Herbarium and Library, Royal Botanic Gardens, Kew; Instituto de Botanica C. Spegazzini, La Plata; Section for Botany, Swedish Museum of Natural History (Naturhistoriska Riksmuseet), Stockholm; Botani- cal Museum, University of Helsinki, Helsinki; Muséum National d’Histoire Naturelle, Laboratoire de Cryptogamie, Paris; National Fungus Collections, Agricultural Research Center-West, Beltsville, Maryland; Plant Pathology Herbarium, Cornell University, Ithaca, New York; Botanischer Garten und Botanisches Museum, Berlin-Dahlem; Herbarium, Department of Botany and Plant Pathology, Oregon State University, Corvallis, Oregon. I also wish to thank Dr. Richard P. Korf for reading the manuscript and making valuable suggestions. A SYNOPSIS OF THE GENUS PULVINULA 19 LITERATURE CITED Bouvier, E. 1885. Nouvelle classification naturelle des Discomycétes charnus. Bull. Soc. Mycol. France. 1: 91-120. —————— . 1907. Histoire et Classification des Discomycétes d’Europe. Klinchsieck, Paris. 221 pp. Cooke, M. C. 1879. Mycographia, seu icones fungorum. I. Discomycétes. Williams and Norgate, London. 267 pp. Dennis, R. W. G. 1968. British Ascomycetes. J. Cramer. Lehre. 455 pp. Gamunpi, I. J. 1966. Nota sobre Pezizales bonaerenses con comentarios sobre el “status” de algunos géneros. Revista Mus. La Plata. Secc. Bot. 10: 47-68. Houne., F. X. R. von. 1917. Mycologische Fragmente, CXX—CXC. Ann. Mycol. 15: 293-383. Korr, R. P. 1972. Synoptic key to the genera of the Pezizales. Mycologia 64: 937-994. LacarpE, J. 1906. Contribution a l’études de Discomycétes charnus. Ann. Mycol. 4: 125-256. Lr Gat, M. 1953. Les Discomycétes de Madagascar. Prodr. Flore Mycol. Madagascar 4: 1-465. Moravec, J. 1969. Nékteré operkulatni diskomycety nalezené v okresech Mlada Boleslav a Jitin. Some operculates found in the districts of Mlada Boleslav and Jitin. Ceska Mykol. 23(4) : 222-235. Puiuies, W. 1887. A Manual of the British Discomycetes. Kegan Paul, Trench and Co., London. 461 pp. Prister, D. H. 1972. Notes on Caribbean Discomycetes. II. Two species of Pulvinula from Puerto Rico. Phytologia 24; 211-215. ReuM, H. 1882. Ascomycetes Lojkani lecti in Hungaria, Transsilvania et Galicia. Friedlander & Sohn, Berlin. 70 pp. Rirar, M. A. 1968. The Australasian Pezizales in the Herbarium of the Royal Botanic Gardens, Kew. Verh. K. ned Akad. Wet. II 57(3): 1-295. Saccarpo, P. A. 1889. Sylloge Fungorum hucusque cognitorum. 8. Patavii. 1143 pp. SEAVER, F. J. 1928. North American cup-fungi (Operculates). Published by the author. New York. 284 pp. STUDIES ON NORTH AMERICAN QUATERNARY BRYOPHYTE SUBFOSSILS I. A NEW MOSS ASSEMBLAGE FROM THE TWO CREEKS FOREST BED OF WISCONSIN! Norton G. MILLER? SUMMARY Plant fossils in an exposure of Two Creeks Forest Bed peat about 11,850 years old, near Green Bay, Wisconsin, include 32 species of mosses representative of a variety of forest and nonforest habitats. The assemblage, which contains few aquatics and many calciphiles, establishes that a diverse flora of temperate, boreal, and arctic mosses occurred in northwestern Wisconsin just prior to and during active glaciation. Most of the mosses identified presently grow in Wisconsin or elsewhere in the upper Great Lakes area, but two, Aulacomnium turgidum and Hypnum bambergeri, now rarely occur farther south than arctic and subarctic regions. Present-day occurrences of some of these species in the Great Lakes region may, in part, date from the period when forest bed sediments accumulated. The represented vegetation, an open, more or less dry Picea glauca woodland with rich fens and dry sites, perhaps on dune sand, is inferred from pollen spectra, mosses, cones, seeds and twigs obtained from the peat. The Two Creeks Forest Bed is a buried organic deposit first described in detail from exposures along the shore of Lake Michigan near the base of the Door Peninsula in northeastern Wisconsin. The forest bed contains a record of vegetation and soil that developed during an ice-free period between two glacial advances late in Wisconsinan time. Both glaciations overrode northeastern Wisconsin leaving behind tills which under- and overlie the forest bed. These deposits, the buried forest bed, and associ- ated sediments contain a record of environmental conditions that may extend back 14,000 radiocarbon yrs B.P. The forest bed is generally con- sidered to represent part of the interval between 11,000 and 12,500 yrs B.P. ( Black, 1970). Paleobotanical and paleoenvironmental studies of the buried forest were first carried out by Wilson (1932, 1936) who provided a compre- hensive analysis of the deposits and their contained fossils. Foremost among plant materials recovered were mosses, 19 of which were identified in forest bed peats and clays from two sites (Cheney, 1930, 1931; Wilson, 1936). The orientation and manner of growth of certain mosses in the 1 The first in a series aimed at providing a better understanding of the development of distribution patterns of North American bryophytes during Pleistocene and Holocene times. The usual American usage of Holocene, i.e., postglacial time, which extends from about 10,000 yrs Before Present (B.P.) to now, is being followed, even though some feel that the “postglacial” is best treated as a subdivision of the Pleistocene. Because the geologic setting and general paleobotany of deposits containing fossil Bryophyta help to determine what environmental conditions prevailed when burial occurred, these topics are treated to some extent also. To document the sometimes fragmentary remains of bryophytes encountered in Quaternary sediments, illustrations or citations of published drawings of fossil material are generally provided. The term subfossil indicates that specimens, though found in a fossil context, are not petrified or coalified, but consist of essentially unaltered tissues that almost always lack cellular contents. 2 Harvard University Herbaria, 22 Divinity Avenue, Cambridge, MA 02138 21 22 NORTON G. MILLER deposits, distribution of species within the sediments, and the preponder- ance of aquatic and subaquatic mosses facilitated reconstruction of the plant communities represented and geological events associated with deposition of the upper till. Culberson (1955) also studied mosses in the forest bed and reported species not previously found. In addition, mosses are mentioned by Schweger (1969) as occurring in a deposit of Two Creeks peat about 50 km northwest of exposures along the Lake Michigan shore. The vegetation of northeastern Wisconsin during Two Creeks time, based on pollen analysis, has been treated by West (1961) and Schweger (1969). LOCATION AND GEOLOGIC SETTING The forest bed exposure under consideration occurs in Scott, Wisconsin, mostly in the southwest quarter of Sect. 23, T. 24 N. / R. 21 E., on the Norbert F. Peters farm, about 6.5 km northeast of the city of Green Bay. According to unpublished data assembled by the late F. T. Thwaites when he visited the pit on 30 June and 1 July 1958, the sediments were exposed on a north-south trending wall of a borrow pit, then about 75 m long and 10 m deep (Fig. 1 & 2). The topmost deposit was a pale red glacial till, the base of which was fairly level and not over 2.5 m below the top of the exposure. Sediments below the till, however, were mixed and were not organized into horizontal strata to any great extent. At certain places dark red, obscurely stratified clay occurred immediately beneath the till, but at others sand was present. Thwaites’ manuscript notes record that peat, associated logs, and other woody materials (Fig. 2) occurred intermixed with the clay at an isolated position along the wall. However, in 1957, Pfleger (Rubin & Alexander, 1960, p. 153) noted an organic rich silt and sand layer about 2 m thick with tree trunks and forest litter below the till. Since active digging was going on at this time, these observations indicate the organic bed and associated sediments were variable laterally. Red clay also underlies the peat, but sediments beneath this clay were covered or remained unexcavated, and therefore are not described in information available to me. Stratigraphic relationships at the Peters’ borrow pit are less clear than at comparable exposures along the western shore of Lake Michigan, but Thwaites (ms. notes) interpreted deposits in the two areas similarly. He considered sediments at the pit to have been laid down in a lake im- pounded in front of the ice that produced the upper till. In his opinion, the peat-log mass was rafted in because of the absence of roots in the position of growth. R. F. Black, who studied the site several times in the 1960's, interprets (in litt.) the organic matter as having been incor- porated into lake sand when the rising lake level drowned the forest in front of the advancing ice. He notes that none of the trees were in growth QUATERNARY BRYOPHYTE SUBFOSSILS 23 Fic. 1. Sediments exposed on west-facing north-south wall of borrow pit near Green Bay, Wis- consin, summer 1958. Organic deposit about midway up the wall. (Photograph courtesy of H. H. lltis. ) Fic. 2 Close-up of most of the organic deposit showing logs and other woody material; a deformed peat layer visible at left. Summer 1958. (Photographs courtesy of H. H. Iltis. ) 24 NORTON G. MILLER position and that the organic bed was folded and disturbed greatly by the overriding ice. Although these workers indicate possible redeposition of organic materials, some of the blocks of sediment I studied have an un- mixed vertical succession of species that is consistent with a shift from more or less dry and open forest communities to those that are wetter. At least part of the forest bed, therefore, appears to have been preserved as it had accumulated. The following age determinations have been published for wood as- sociated with peat in the borrow pit: 11,940 + 390 yrs B.P. (Y-147X, Preston et al., 1955, p. 958) and 11,140 + 300 yrs B.P. (W-590, Rubin & Alexander, 1960, p. 153). These determinations agree favorably with the generally accepted age of the Two Creeks Forest Bed, 11,850 + 100 yrs B.P. (Broecker & Farrand, 1963), which has been taken by some to mark the end of Twocreekan time (Black & Rubin, 1968 ). To summarize, the Two Creeks forest developed on lacustrine sediments deposited out of a lake that stood higher than the present level of Lake Michigan. The lake beds accumulated on till whose precise age is not known but is often correlated with an advance of the late Woodfordian Cary ice or a somewhat more recent readvance. This lower till was not recorded at the Peters’ borrow pit but can be seen along the shore of Lake Michigan and elsewhere in the immediate area. Forest became established on the lake sediments following a lowering of the lake level, and soil development and peat accumulation occurred. Glacier ice readvanced into the region about 11,850 yrs B.P. causing the lake level to rise and de- stroy the forest. Lacustrine sediments accumulated on top of the soil for some time following this date. Ice later spread over northeastern Wiscon- sin and then receded, leaving behind a deposit of reddish drift generally considered to correlate with the Valders till, described from somewhat farther south in Wisconsin. Recent data published by Evenson (1973) indicate that till above the forest bed is younger than the Valders, al- though this interpretation has been contested by Black (1974; see also Evenson et al., 1974). Evenson (1973) has proposed the name Two Rivers till for that present above the Two Creeks forest bed. METHODS Ten large, dry blocks of sediment (ca. 20 x 10 & 10 cm) were placed individually in beakers, covered with a solution of 0.5% trisodium phos- phate and heated gently to loosen clays and silts and rewet the plant materials (Benninghoff, 1947). Inorganic materials were washed from the organic fraction through a small mesh sieve with distilled water after about 3 hr of treatment. Residues were soaked for several hours in distilled water to leach out all trisodium phosphate and were bottled in dis- tilled water or 70% ethyl alcohol until studied microscopically. QUATERNARY BRYOPHYTE SUBFOSSILS 25 To determine the variability of residues, subsamples removed from a well-mixed slurry of the wet fossil material were separated into their component species. The area occupied by individual taxa was determined by measuring a tightly packed layer of fragments, one plant deep, using a grid divided into square centimeters. This procedure was repeated twice for two residues to establish the reproducibility of the sampling method, and data from three additional sediment blocks were gathered. Data for individual samples were calculated in percentages. A sample of the organic material cut from the center of a dry sediment block was analyzed for pollen following procedures outlined in Faegri and Iversen (1964). Pollen counts from inorganic sediments were not made because of possible contamination and uncertainty about the orig- inal orientation of sediment blocks in the exposure. RESULTS A compact moss layer, resting directly on an accumulation of forest litter or soil, occurred in each sediment block. The combined thickness of the organic material varied from sample to sample but the maximum noted was 5 cm. Intermixed clays and silts were absent except for a small amount of inorganic sediment found with plant fragments along the upper and lower limits of the organic layer. No plant fossils were found embedded in the silts and clays which under- and overlie the organic horizon. Pollen analysis and vascular plant megafossils. Pollen counts obtained from samples of the thin (ca. 3 cm thick) moss layer in two sediment blocks are given in Table 1. The principal trees represented are spruce (Picea spp.) and tamarack (Larix laricina) which together account for about 80% of the percentage base. Pine pollen is nearly absent. About 15% of the sum consists of various nonarboreal (NAP) taxa, especially mem- bers of the Cyperaceae, which comprise about one-half of the NAP totals. These pollen assemblages agree with results obtained by West (1961) from samples of Two Creeks peat exposed along the shore of Lake Michi- gan, with the exception of somewhat higher spruce (80%) and lower Cyperaceae (usually < 5%) representation in most of his spectra. A pollen spectrum in Schweger (1969) from sediments at the borrow pit contains much less spruce (ca. 40%) and more Cyperaceae and Gramineae (ca. 45%) than either spectra reported here. Though differing quantitatively, the counts are similar in terms of pollen types represented, and the differ- ences probably relate to the mixed nature of the “composite sample” Schweger processed (i.e., peat and organic silt mixed). My analyses were of single, relatively thin, strata. The organic bed contains abundant cones, twigs and needles of spruce and tamarack (Fig. 3-9). Seeds of the two genera were also found, but no other seeds (or fruits) were noted. Although some spruce cones are 26 NORTON G. MILLER TABLE 1. POLLEN CONTENT OF MOSS LAYER AT THE PETERS’ BORROW PIT, BROWN CO., WISCONSIN AP? Sample 1 n % Pinus 2 0.6 Picea 241 68.5 Larix 31 8.8 Betula -- — Carpinus-Ostrya — _ Quercus 4d 1.1 Carya 1 0.3 Ulmus 1 0.3 Populus 9 2.6 AP totals 289 82.1 NAP* Alnus — — Myrica - - Salix 7 2.0 Cyperaceae 23 6.5 Gramineae 8 2,2 Ambrosia vi 2.0 Artemisia 7 2.0 Xanthium il 0.3 High-spine Compositae 10 2.8 NAP totals 63 17.9 MISC” Polypodiaceae 3 0.8 Moss spores _ _ Unidentifiable 24 6.4 Sample 2 % n 2 0.5 271 71.5 35 9.2 1 0.3 2 0.5 1 0.3 1 0.3 1 0.3 10 2.6 324 85.5 1 0.3 1 0.3 2 0.5 24 6.3 11 2.9 ‘i 1.8 5 1.3 4 1.1 55 14.5 7 1.7 1 0.2 17 4.2 “Percentage base = sum arboreal pollen (AP) + nonarboreal pollen (NAP) . Percentage base = sum AP + NAP + MISC QUATERNARY BRYOPHYTE SUBFOSSILS af Fics. 3-9. Cones and twigs from Two Creeks peat at Peters’ borrow pit.—3—5. Cones of Picea glauca (Moench) Voss.—6. Probable cone of Picea mariana (Mill.) B.S.P.—7—8. Cones of Larix laricina (DuRoi) K. Koch.—9. Twig of Larix laricina. small for the species, most were identified as white spruce (Picea glauca) because the cones are at least two times longer than broad and their scales have entire margins. Because cones beneath living white spruce trecs are somewhat variable in size and may be decayed to the point that the scale margins are no longer entire, identifications were made using only well- preserved specimens. About nine of every ten cones in the sediment blocks were of P. glauca, the remainder being assignable to P. mariana. Bryophyte remains. Mosses in some sediment blocks were preserved mostly as unfragmented plants, indicating that water transport was mini- mal and that in situ burial probably occurred. Other sample residues contain mixtures of species from different habitats, and in these cases it seems likely that the assemblage was brought together by moving water, No liverworts were found. In order to determine species composition at various levels in the organic bed, two sediment blocks showing clay—moss/ forest litter—clay layering were cut in two vertically and the cut edges were soaked in water 28 NORTON G. MILLER to facilitate examination. A compact, peaty forest litter, 3 to 4 cm thick was present in both blocks. Materials of Drepanocladus uncinatus and Thuidium abietinum (subordinate) occurred near the top of the litter, together with cones and needles of white spruce and tamarack. While both of these mosses occur in a variety of nonforest habitats, D. uncinatus is a typical species of moist to dry conifer stands, and T. abietinum is sometimes found in open white spruce forests on dry soil. Downward in the litter, the mosses drop out and the amount of degraded plant material increases. Above the forest litter and on top of the mosses was a dense mat of Tomenthypnum nitens, which generally occurs in wet, calcareous habitats that can be forested or open. Relatively few conifer needles and no cones were found with this moss, indicating a decline in the abundance of trees at the site. A shift from more or less dry to wetter conditions indicates a rise in the water table possibly relating to flooding associated with the advance of glacier ice or to periglacial activity. Variability in the composition of moss assemblages obtained from five sediment blocks is given in Table 2. Three of the samples yielded fifteen or more species while two had eight or nine. Some species occur in all samples but others are present in just one or a few. Four samples are dominated by Tomenthypnum nitens and from one to seven subdominant species occur with it. About half of the identified species are rare either in a given sample or in all examined materials. It is tempting to conclude that numerical relationships between species listed in the table also held in nature. However, the percentage values, which though similar to field determined estimates of percent cover, are not strictly comparable to measurements of abundance in nature for at least two reasons: sample residues are in part mixtures of species from distinct communities that may have existed at different times, and it is impossible to reconstruct the orientation of the mosses as they actually grew, therefore making measure- ments of area only an approximation of the original situation. Mosses identified in the forest bed are given below, with an indication of their abundance in the residues. The following scale is used: rare—10 or fewer fragments; sparse—11 to 20; rather abundant—21 to 40; abundant—41 to 60; and very abundant— > 60. A fragment varied from isolated leaves to leafy branches to essentially complete plants. Notes are also given on the present distribution of identified taxa (particularly in the Great Lakes region), diagnostic characteristics and the occurrence of structures associ- ated with the sexual reproductive cycle (e.g., archegonia, antheridia, sporophytes, etc.). Unless otherwise specified, a record documenting a recent collection of the species in Wisconsin has been found (see Cheney & Evans, 1944; Forman, 1967). Species indicated by an asterisk have not been previously reported from the forest bed. Crum’s recent book (1973) on the mosses of northern Michigan contains helpful notes on the habitat preferences of many of the species found in the forest bed. QUATERNARY BRYOPHYTE SUBFOSSILS TABLE 2, AREAS (PERCENT COVER) OCCUPIED BY MOSS SUBFOSSILS IN SAMPLES ONE TO FIVE Sample—Subsample Designation Species Tomenthypnum nitens Bryum pseudotriquetrum Campylium polygamum C. stellatum Brachythecium turgidum Ditrichum flexicaule Drepanocladus uncinatus Thuidium abietinum Hylocomium splendens Scorpidium turgescens Hypnum bambergeri Distichium capillaceum Drepanocladus aduncus var. polycarpus Eurhynchium pulchellum Amblystegium serpens Tortella fragilis Drepanocladus sp. i IN recurvirostrum Dicranella heteromalla Mnium marginatum Tortella sp. T. tortuosa Myurella julacea Encalypta procera Aulacomnium palustre A. turgidum unknown TOTALS *From samples analyzed for pollen 1 2 3 4 a® b? a® b* 67.9 67.2 54.7 52.8 70.3 83.8 7.7 7.3 101 122 08 £03 88 64 91 112 63 £76 1.6 18 8.1 6.1 6.3 trace —- O02 5.1 4.1 _ — 1.1 18 4.1 4.1 9.3 — 44 9.1 a — 3.9 1.3 5.5 3.6 _ — — 6.7 1.1 0.5 — —- 2.3 — — — = 0.8 — _ — 4) 4.6 — _ trace trace 3.0 4.] -_ = = ae 15 m0) — trace 0.7 0.9 - — — _ 0.5 0.4 trace — — 03 — 0.5 trace — — — —_ — 02 — — — 0.2 0.2 — — — _ 0.2 0.2 — — — trace 0.2 — trace trace = — — — trace trace — _ _— — trace trace _ — 99.9 100.1 100.0 100.2 100.0 100.0 29 trace 0.2 trace trace trace trace trace 100.0 30 NORTON G. MILLER 14 Fics. 10-16. Subfossil mosses from Two Creeks peat at Peters’ borrow pit.—l10. Sphagnum papillosum Lindb., cross section of branch leaf.—11. same, hyaline cell and pores, outer face of branch leaf.—12. Dicranella heteromalla (Hedw.) Schimp., plant with apical antheridial cluster.—13. same, two leaves.—14. Encalypta procera Bruch., leaf.—15. same, cells and papillae from upper part of leaf.—16. Tortella fragilis (Drumm.) Limpr., plant with leaf tips mostly broken off. *Sphagnum papillosum Lindb.—Rare, one branch leaf only. Identified on the basis of trapezoidal chlorophyllose cells, which are broader on the inner than outer face, presence of a resorption furrow, and hyaline cells with relatively few pores. No papillae occur on the inner lateral walls of hyaline cells, making the material var. laeve Warnst. A fairly common, widespread species in the Great Lakes region. Fig. 10 & 11. Ditrichum flexicaule (Schwaegr.) Hampe—Abundant, leafy plants matted together or solitary, in one case intermingled with Bryum pseudo- triquetrum, Campylium stellatum and Tortella fragilis. Not known to QUATERNARY BRYOPHYTE SUBFOSSILS 31 occur in Wisconsin but present in the Straits of Mackinac region, Michi- gan. Fossil material from New York State is illustrated in Miller (1973). Distichium capillaceum (Hedw.) B.S.G.—Abundant, gametophytic ma- terials only. Referred here with fair certainty, but the related D. inclina- tum (Hedw.) B.S.G., which differs primarily on the basis of sporophyte characters, may also be present. Forman (1967) cites collections of D. capillaceum from rotting logs and soil at forested sites on the Door Penin- sula. Fossil material illustrated in Miller (1973). *Dicranella heteromalla (Hedw.) Schimp.—Rather abundant. Not re- ported before from North American Quaternary deposits. Some specimens bear antheridia. Fig. 12 & 13. Encalypta procera Bruch—Rather abundant, occurring mainly as iso- lated leaves or as stem fragments with few leaves, only one large leafy plant found. Preservation fair; upper leaf cells with forked papillae; lower hyaline, nonpapillose cells mostly decayed. All leaves lack any indication of an awn; filiform brood bodies are absent. No doubt still a member of the Wisconsin flora, since records of Encalypta streptocarpa Hedw. in Cheney and Evans (1944) probably refer to E. procera. Apparently not recognized previously as a fossil in North America. Fig. 14 & 15. *Trichostomum tenuirostre (Hook. & Tayl.) Lindb.—Rare, eight well- preserved leafy plants. The specimens agree well with most herbarium collections so-named, but North American materials of Trichostomum need revision. Leaves of the fossil specimens clearly show the typical region of basal hyaline cells that do not extend up the margin as in Tortella. Apparently unknown in the present flora of Wisconsin but oc- curring in the Upper Peninsula of Michigan. No records of its fossil occurrence in North America are known to me. Tortella fragilis (Drumm.) Limpr.—Sparse, mostly leafy plants, a few with setae, and several isolated leaves. An infrequent moss of calcareous situations in the Great Lakes region. Fig. 16. °T. inclinata (R. Hedw.) Limpr.—Rare, three plants, somewhat poorly preserved but with certain leaves showing the characteristic concave, + cucullate apex. A rare moss in the Great Lakes area where it has been collected most often on dune sand. Unknown from Wisconsin but reported from several counties bordering the Straits of Mackinac, Michigan. A previous fossil record exists from northwestern New York State in a deposit a few hundred years older than the Two Creeks Forest Bed (Miller, 1973, q-v. for illustrations of fossil material ). T. tortuosa (Hedw.) Limpr.—Rather abundant, leafy plants, preserva- tion generally good. This species and T. fragilis are known from recent collections made in Door County, Wisconsin, at stations near the fossil bed. Illustrations given in Miller (1973). *Bryoerythrophyllum recurvirostrum (Hedw.) Chen—Sparse, somewhat degraded leafy plants. Fossil material illustrated in Miller (1973). 32 NORTON G. MILLER Fics. 17—25. Subfossil mosses (cont.).-17. Mnium marginatum (With.) Brid. ex P. Beauv., plant.—18. same, three leaves from same plant.—19. same, leaf margin with a paired teeth.—20. same, upper leaf cells.—21. Myurella julacea (Schwaegr.) B.S.G., fragment of plant.—22. same, edge of leaf with teeth and papillose cells.—23. Thuidium recognitum (Hedw.) Lindb., nearly com- plete plant.—24. same, stem leaves from same plant.—25. same, paraphyllia. QUATERNARY BRYOPHYTE SUBFOSSILS 33 *Tortula ruralis (Hedw.) Gaertn., Meyer & Scherb.—Rare, two leaves. Identification probable but the characteristic roughened awn is poorly preserved. A common calciphile usually of open, + dry situations through- out the Great Lakes area. Illustrations in Miller (1973). Bryum pseudotriquetrum (Hedw.) Gaertn., Meyer & Scherb.—Abun- dant, leafy plants and a few isolated leaves. Many plants were found matted together, and some individuals in such clumps carried a broken seta. Sporophyte bearing plants were searched for archegonia and/or antheridia, but out of approximately 25 specimens, only two could be shown to be synoicous and two dioicous (only archegonia seen). Plants of the two types were not intermixed in the same clump. One plant with a terminal antheridial cluster was also found. Dioicous collections of this species are sometimes referred to var. pseudotriquetrum and synoicous ones to var. binum (Schreb.) Lilj. Other species of Bryuwm may be present in the fossil materials, but in the absence of capsules and peristomes, identifications could not be made. Leaves of fossil B. pseudotriquetrum from New York State are illustrated in Miller (1973). *Mnium marginatum (With.) Brid. ex P. Beauy.—Rather abundant, leafy plants and isolated leaves, preservation good but the double row of teeth along the leaf margin difficult to demonstrate. A few isolated leaves were broadly ovate (cf. Fig. 18). Upper leaf cells of fossil material are + quadrate with thickened corners and vary in size from 20-28 pm. A wide- spread species in the Great Lakes region where it grows on soil and rock in open or more dense forests. Fig. 17-20. *Aulacomnium palustre (Hedw.) Schwaegr.—Rather abundant, leafy plants and isolated leaves, small marginal teeth sometimes apparent to- ward the leaf apex. A common moss in temperate and boreal America; occurring in a variety of usually moist to wet habitats including forests and open areas. *A. turgidum (Wahlenb.) Schwaegr.—Rare, one leaf-bearing stem tip and eight isolated leaves. Fossil materials identified as this species of Aulacomnium because the leaf apices are obtusely rounded and somewhat hooded. A predominantly arctic and subarctic moss in North America, and apparently not a member of the present flora of Wisconsin, the species occurs disjunctively near the north shore of Lake Superior. Fossil material of the species 12,100 yrs old is known from northwestern New York State (Miller, 1973, q.v. for illustrations); the species has been found in inter- glacial deposits on Banks Island, arctic Canada (Kuc, 1974). *Orthotrichum obtusifolium Brid.—Rare, two plants, preservation fair. Leaves of the fossils have bluntly rounded apices, unipapillose cells and plane margins. No brood bodies were seen. Identification probable but not positive because the fossil leaves in general have a shorter costa than that present in herbarium material. Orthotrichum obtusifolium occurs widely in north temperate and southern boreal North America, where it is 34 NORTON G. MILLER a fairly common epiphyte, particularly on bark of species of Populus ( Vitt, 1973), pollen of which occurs in the forest bed peat. The genus has ap- parently not been reported before from North American Quaternary deposits. *Myurella julacea (Schwaegr.) B.S.G.—Sparse, stem fragments with leaves, preservation fair to good. An uncommon species in the Great Lakes area where it grows on moist soil in forests, especially Thuja swamps, and sometimes on rock (e.g., crevices in cliffs) in + open woods. No other North American Quaternary fossil occurrences are known. Fig. 21 & 22. Thuidium abietinum (Hedw.) B.S.G.—Abundant, large plants and fragments, leaves sometimes decayed from stems and branches. Known from several Quaternary deposits elsewhere in North America. °T. recognitum (Hedw.) Lindb.—Abundant, complete plants and large fragments. Fossil material agrees in all characters with herbarium speci- mens. Not known with certainty from other North American Quaternary deposits. This moss and T. abietinum have been collected on the Door Peninsula and have a wide distribution in the Great Lakes area and else- where in North America. Thuidium recognitum occurs at open sites on soil and rock and less frequently in dry to moist, + open forests. Fig. 23-25. *Campylium polygamum (B.S.G.) C. Jens.—Very abundant, leafy plants. Widely distributed in North America; in the Great Lakes region occurring on soil in bog forests, swamps, and wet open areas. Fig. 26 & 27. C. stellatum (Hedw.) C. Jens.—Abundant, fossil materials sometimes recognizable at a glance by their rich, coppery brown color, but otherwise determinable by the leaves which have a short double costa (vs. costa strong and single in C. polygamum). Extending northward from the temperate zone in North America, C. stellatum often grows in + open, calcareous habitats, especially rich fens, but also sometimes at forested sites. Fig. 28 & 29. *Amblystegium serpens (Hedw.) B.S.G.—Abundant, leafy stems mixed with other mosses or solitary. A common, widespread species of wet places, often on soil or rotting wood, in both forest and nonforest habitats. Not apparently recorded before as a subfossil in North America. Fig. 30. Drepanocladus aduncus var. polycarpus (Bland. ex Voit) Roth—Rather abundant, leafy plants. An aquatic or semiaquatic moss, usually of open or wooded calcareous places, found throughout the Great Lakes region and northward. The variety is not known from other deposits of the forest bed, although vars. aduncus and pseudofluitans Sanio occurred in Wilson’s samples (Cheney, 1930, 1931). Numerous reports of D. aduncus and some of its varieties are reported from Quaternary sediments else- where in glaciated portions of North America. D. revolvens (Sw.) Warnst.—Rare, only two leafy fragments. Leaves of fossil material are nonstriolate and lack inflated alar cells. Recent collec- QUATERNARY BRYOPHYTE SUBFOSSILS 35 tions of this and the following species have been made in Door County, Wisconsin (Cheney & Evans, 1944). D. uncinatus (Hedw.) Warnst.—Very abundant, large fragments or intact plants, sometimes with one or several setae that lack capsules. Dense mats occurred in some sediment blocks. A common moss of open or dense boreal conifer forests; in the Great Lakes area occurring par- ticularly in Thuja swamps. Fig. 31. Scorpidium turgescens (T. Jens.) Loeske—Rare, two leafy plants. Un- known in the present flora of Wisconsin but found at several places in Michigan and adjacent areas, there reaching its present southern limit of distribution in central North America. Throughout its range, this moss occurs most frequently at wet, nonforest, calcareous sites, particularly rich fens. Illustrations of fossil material from New York State are in Miller (1973). 27 o.5mm o.5mm Fics. 26-30. Subfossil mosses (cont.).—26. Campylium polygamum (B.S.G.) C. Jens., plant. —27. same, three leaves from same plant.—28. C. stellatum (Hedw.) C. Jens., plant.—29. same, two leaves from same plant.—30. Amblystegium serpens (Hedw.) B.S.G., leaf. 36 NORTON G. MILLER Tomenthypnum nitens (Hedw.) Loeske—Abundant, entire plants and fragments, one individual with two setae. Another calciphile, occurring in open or forested, wet habitats; found in North America from the Great Lakes region northward into the Arctic. Fossil material illustrated in Miller (1973). *Brachythecium turgidum (C. J]. Hartm.) Kindb.—Abundant, leafy plants; large fragments (3-4 cm) have a few short branches. Stem and branch leaves are entire and in them the costa extends to just beyond the center of the leaf. Apparently unknown in the present Wisconsin flora, although the species has been collected in Michigan, Culberson (1955) Fics. 31-35. Subfossil mosses (cont.).—31. Drepanocladus uncinatus (Hedw.) Warnst., plant with two setae.—32. Brachythecium turgidum (C. J. Hartm.) Kindb., fragment of plant.—33. same, three leaves from same plant.—34, Eurhynchium pulchellum (Hedw.) Jenn., fragment of plant.—35. same, two leaves. QUATERNARY BRYOPHYTE SUBFOSSILS 37 has reported B. salebrosum (Hoffm.) B.S.G. from samples of the forest bed. Fig. 32 & 33. *Eurhynchium pulchellum (Wedw.) Jenn.—Rather abundant, branched leafy plants and isolated branches. A species widespread in North America that usually grows on soil and rotting logs in forests; also extending north- ward to the tundra. A common moss of Thuja swamps and bog forests in the upper Great Lakes region. Fig. 34 & 35. *Hypnum bambergeri Schimp.—Abundant, fragments with and without short branches. A short double costa (occasionally single) and an elevated cluster of reddish-brown, thick-walled alar cells characterize leaves of the fossils, which match recent herbarium materials in all important characters. The present North American range of this moss is limited mostly to arctic and subarctic regions. The species is an apparent calci- phile. Ilustrations of fossil material from the Wisconsin locality are in Miller (1976). *Hypnum pallescens (Hedw.) P. Beauv.—Rare, one plant with branches and several unbranched leafy fragments; identified on the basis of ser- rulate leaves with a double costa and quadrate alar cells. At present found widely in north temperate and boreal North America (and extending southward in the mountains), this moss occurs in forests on the bases of trees and sometimes on humus or other substrata. *Hylocomium splendens (Hedw.) B.S.G.—Abundant, some nearly com- plete plants and fragments. A common moss of moist forest soil throughout boreal North America but also found northward into the tundra and southward in mountain forests. Recent collections from places on the Door Peninsula are cited by Cheney and Evans (1944). DISCUSSION The vegetation cover of northeastern Wisconsin during Two Creeks time has been interpreted as closed canopy spruce forest (West, 1961) or open boreal woodland (Schweger, 1969), two more or less floristically similar vegetation types occurring today at places in the North American boreal forest. Local variation in Twocreekan plant communities existed because fossil assemblages from different deposits vary in composition. Spruce, however, is uniformly the major arboreal component. Pollen or cones of white spruce predominate in some instances (West, 1961, this report ); black spruce is more abundantly represented in others (Schweger, 1969; Wilson, 1936). In present-day boreal America white and black spruce usually occupy dry and wet soil sites respectively, and the same situation no doubt prevailed during Two Crecks time. Nonarboreal pollen (NAP) totals, indicative of the amount of open, nonforest vegetation, also vary. Spectra from some samples have as little as 5% NAP, others contain 20-25% or more. The existence of communities with herbs and shrubs is 38 NORTON G. MILLER compatible with white and black spruce forests because similar mixtures of nonforest and forest vegetation now occur in parts of boreal America. At the Peters’ borrow pit megafossil material of white spruce is most abundant and lesser quantities of tamarack and black spruce occur. Forest communities on drier soil types must therefore have been prominent in the area during accumulation of forest bed sediments. Nonarboreal pollen is also sufficiently abundant to indicate the presence of plant communities containing or perhaps dominated by herbs, especially members of the Cyperaceae. The large assemblage of mosses found in the peat (Table 3) provides a way to define the communities more precisely, since all of the mosses represent extant species whose habitat preferences are known. The species are about evenly divided between those characteristic of forests and those occurring in nonforest habitats, several of which are indicated. Species listed more than once in the table are those that exhibit broad habitat tolerances. The forest mosses support the view that dry to moist but not wet soil conditions prevailed. Species of dense to open forest stands are present. Some of them grow on litter or rotting logs in existing spruce forests (e.g., Drepanocladus uncinatus, Eurhynchium pulchellum, Hylocomium splen- dens); others occur on mineral or humus-rich soil (Dicranella heteromalla, Bryoerythrophyllum recurvirostrum, Mnium marginatum). It can be in- ferred that similar habitats existed in forests of Two Creeks time. Two epiphytic species, Orthotrichum obtusifolium and Hypnum pallescens, were found, although the latter occasionally grows on soil or rocks also. Some of the forest mosses are calciphiles. These perhaps grew on or close to calcareous substrata beneath the forest bed. Other species typical of acid substrata indicate that humus accumulation was fairly advanced. The variety of habitats indicates well established forest and not pioneer communities. Of the nonforest mosses recovered from the peat, species of rich fen communities, which develop in association with shallow, calcium-rich water, are well represented. The distribution and character of rich fens are less fully known in North America than in Europe, although descrip- tions of some to the west of James Bay, Canada, are available (Sjérs, 1961, 1963; and Persson & Sjérs, 1960, for ecological data on mosses in this region). Species characteristic of wetter areas of rich fens are separated in the table from those often occurring in drier parts and emphasize the relative scarcity of aquatic mosses in the peat samples. The predominantly aquatic moss genera Calliergon, Drepanocladus and Scorpidium are better represented in forest bed exposures along the shore of Lake Michigan (Wilson, 1932, 1936) than at the borrow pit, perhaps indicating a greater abundance of poorly drained sites in the former area during Two Creeks time. A few mosses listed in the fen margin category (e.g., Aulacomnium palustre, A. turgidum, Sphagnum napillosum) prefer more acid substrata, QUATERNARY BRYOPHYTE SUBFOSSILS TABLE 3. SUBFOSSIL MOSSES, FROM TWO CREEKS FOREST BED EXPOSED AT PETERS BORROW PIT, BROWN COUNTY, WISCONSIN, BY HABITAT TYPE Forest: mostly humus or mineral soil Dicranella heteromalla Campylium polygamum Bryoerythrophyllum recurvirostrum Amblystegium serpens Bryum pseudotriquetrum Drepanocladus uncinatus Mnium marginatum Eurhynchium pulchellum Orthotrichum obtusifolium (tree bark ) Hypnum pallescens (tree bark ) Thuidium abietinum Tomenthypnum nitens T. recognitum Hylocomium splendens Nonforest: rich fen, aquatic or semiaquatic Campylium stellatum Scorpidium turgescens Drepanocladus aduncus var. polycarpus Tomenthypnum nitens D. revolvens Nonforest: fen margin, + moist, mostly calcareous Sphagnum papillosum Aulacomnium turgidum Dicranella heteromalla Myurella julacea Distichium capillaceum Campylium polygamum Ditrichum flexicaule C. stellatum Tortella fragilis Amblystegium serpens Bryoerythrophyllum recurvirostrum Tomenthypnum nitens Bryum pseudotriquetrum Brachythecium turgidum Mnium marginatum Hypnum bambergeri Aulacomnium palustre Nonforest: moist rocks or mineral soil Distichium capillaceum Trichostomum tenuirostre Encalypta procera Mnium marginatum Tortella fragilis Aulacomnium turgidum Bryoerythrophyllum recurvirostrum Myurella julacea Nonforest: + xeric, well-drained soil, e.g., sand Tortella tortuosa Thuidium abietinum T. inclinata T. recognitum Tortula ruralis 39 40 NORTON G. MILLER and these perhaps grew on + dry peat hummocks or other noncalcareous sites within fens. Mosses limited to rock or mineral soil in open or semi- forested situations were few (e.g., Encalypta procera, Trichostomum tenuirostre ), although certain other species present in the peat sometimes grow in these habitats. The small group of mosses generally characteristic of xeric, open areas may have grown on well-drained dune or beach sand, which probably accumulated during fluctuations in the level of the lake then occupying the Lake Michigan basin. Peat samples from the Peters’ borrow pit have yielded a moss flora con- sisting of 32 extant species. The following seven additional mosses have been reported by previous workers from other exposures of the forest bed (Cheney, 1931, 1932; Culberson, 1955): Brachythecium salebrosum (Web. & Mohr) B.S.G., Bryum tortifolium Funck ex Brid., Calliergon cordifolium (Hedw.) Kindb., C. stramineum (Brid.) Kindb., Drepanocladus sendtneri (Schimp.) Warnst., D. vernicosus (Lindb. ex C. Hartm.) Warnst. and Scorpidium scorpioides (Hedw.) Limpr. No one ecological grouping of species dominates the flora, although aquatic or semiaquatic mosses may be more abundant at some localities. All species except seven are present members of the Wisconsin flora. Of these, however, five are known at stations in northern Michigan (Crum, 1973), and it is probable that they also occur in Wisconsin. Based on records cited in Cheney and Evans (1944) and Forman (1967), 16 of the 39 species (41%) are represented by recent collections from Brown, Door, Kewaunee and Manitowoc counties, which comprise the Door Peninsula. Two mosses that occur in the borrow pit peat, Aulacomnium turgidum and Hypnum bambergeri, are far out of place based on their present ranges. While several disjunct stations for the former are known in the Thunder Bay District of Ontario, about 500 km north of Green Bay, the southern limit of its continuous distribution in North America roughly coincides with the northern edge of the boreal forest. The species also occurs southward in alpine regions both in the East and West. Found as far south as the Gaspé Peninsula, Quebec, in the East and near Banff, Alberta, in the West, Hypnum bambergeri also has a predominantly arctic and subarctic distribution in North America. Based on material in five herbaria (CANM, FH, MICH, NY, US), this moss presently is found no farther south in the midcontinent region than northern Manitoba, some 1200 km north of the fossil occurrence. The presence of arctic and subarctic mosses in Wisconsin during Two Creeks time with others still found in the upper Great Lakes region and to the south is phytogeographically important. Such a mixed assemblage establishes that the Two Creeks flora was composed of species of different current geographical affinities. Existing disjunct stations for certain arctic species in the upper Great Lakes area may thus relate to Two Creeks time when such northern species occurred beyond their present ranges. QUATERNARY BRYOPHYTE SUBFOSSILS 4] Other arctic and subarctic mosses present as fossils in deposits about 12,100 and 13,300 years old in northern Michigan and northwestern New York State are discussed in Miller (1973) and Miller and Benninghoff (1969). While 95% of the Twocreekan moss flora of northeastern Wisconsin is still represented in this area or nearby, some of the calcicolous species currently are rare in the upper Great Lakes region and attain their greatest abundance northward. Certain of them and other species of similar distri- bution also occur to the east in the two deposits mentioned above. The species include Catoscopium nigritum, Distichium capillaceum, Ditrichum flexicaule, Meesia uliginosa, Myurella julacea, Tortella fragilis, Scorpidium turgescens and others. Their presence in the assemblages indicates that much of the Great Lakes region 13,300 to 11,850 years ago was edaphically (and climatically) suited to the occurrence of such calcicolous species. Since the mosses now occur in the same general region as they did during the waning phases of glacial activity in late Wisconsinan time, their present distribution in the Great Lakes area can be viewed as an example of per- sistence at suitable habitats. That the species may be recent immigrants is also possible but perhaps less likely because of evidence provided by the fossil record. Similar assemblages of subfossil mosses from postglacial deposits, i.e., those less than about 10,000 years old, will help establish the more or less continuous presence of the species. This topic is more fully discussed in Miller (1976). The moss flora of northeastern Wisconsin during Two Creeks time was surprisingly diverse considering that it developed at a time of active glaciation. However, based on the richness of the flora as it is now known, study of material from other localities will continue to add species and improve the vegetation interpretation presented here and by other workers. ACKNOWLEDGMENTS Samples used in this study were collected by Hugh H. Iltis (University of Wisconsin, Madison) who wishes to acknowledge with appreciation support from that university’s Research Committee of the Graduate School and Dean of Letters and Sciences. IItis had earlier discovered the organic deposit in the company of Walter Biichmann. Samples of the organic bed were sent to me through the courtesy of W. C. Steere. I am grateful to Prof. R. F. Black of the University of Connecticut and Prof. Iltis for supplying descriptions of the borrow pit and of sediments exposed there. Howard Crum and L. E. Anderson examined some of the specimens on which the moss identi- fications are based, and I thank them for their comments. The habit drawings of mosses are the work of Marion Seiler whose care in their preparation is appreciated. LITERATURE CITED BENNINGHOFF, W. S. 1947. Use of trisodium phosphate with herbarium material and microfossils in peat. Science 106: 325, 326. Biack, R. F. 1970. Glacial geology of Two Creeks Forest Bed, Valderan type locality, and Northern Kettle Moraine State Forest. Wisconsin Geol. Nat. Hist. Sur. Inf. Cire. 13. 40 pp. 42 NORTON G. MILLER ——————— . 1974. Late Pleistocene shorelines and stratigraphic relations in the Lake Michigan basin: Discussion. Geol. Soc. Amer. Bull. 85: 659, 660. ——————— and M. Rusin. 1968. Radiocarbon dates of Wisconsin. Trans. Wisconsin Acad. Sci. Arts Letters 56: 99-115. Broecker, W. S. and W. R. Farranp. 1963. Radiocarbon age of the Two Creeks Forest Bed, Wisconsin. Geol. Soc. Amer. Bull. 74: 795-802. CHENEY, L. S. 1930. Wisconsin fossil mosses. Bryologist 33: 66-68. ——————— . 1931. More fossil mosses from Wisconsin. Bryologist 34: 93, 94. ——————~ and R. Evans. 1944. The mosses of Wisconsin. Trans. Wisconsin Acad. Sci. Arts Letters 36: 171-224. Crum, H. 1973. Mosses of the Great Lakes forest. Contrib. Univ. Michigan Herb. 10. iii + 404 pp. Cu.Berson, W. L. 1955. The fossil mosses of the Two Creeks Forest Bed of Wisconsin. Amer. Midl. Nat. 54: 452-459. Evenson, E. B. 1973. Late Pleistocene shorelines and stratigraphic relations in the Lake Michigan basin. Geol. Soc. Amer. Bull. 84: 2281-2297. ———— , W. R. Farranp and D. F. EscuMan. 1974. Late Pleistocene shorelines and stratigraphic relations in the Lake Michigan basin: Reply. Geol. Soc. Amer. Bull. 85: 661-664. Farcri, K. and J. Iversen. 1964. Textbook of pollen analysis. 2nd Rev. Ed. 237 pp. Munksgaard. Copenhagen. Forman, R. T. T. 1967. New and uncommon Wisconsin mosses. Bryologist 70: 1L15—- 117. Kuc, M. 1974. The interglacial flora of Worth Point, western Banks Island. Geol. Sur. Can. Paper 74-1B: 227-231. Miuier, N. G. 1973. Lateglacial plants and plant communities in northwestern New York State. Jour. Arnold Arb. 54: 123-159. ——————— . 1976. Quaternary fossil bryophytes in North America: A synopsis of the record and some phytogeographic implications. Jour. Hattori Bot. Lab. in press. a and W. S. BEeNNINGHOFF. 1969. Plant fossils from a Cary—Port Huron Interstade deposit and their paleoecological interpretation. Geol. Soc. Amer. Special Paper 123: 225-248 + pl. 3. Persson, H. and H. Syérs. 1960. Some bryophytes from the Hudson Bay Lowland of Ontario. Sv. Bot. Tidskr. 54: 247-268. Preston, R. S., E. Person and E. §. DeEvey. 1955. Yale natural radiocarbon measure- ments II. Science 122: 954-960. Rusin, M. and C, ALExANDER. 1960. U.S. Geological Survey radiocarbon dates V. Amer. Jour. Sci. Radiocarbon Suppl. 2: 129-185. ScuweceEr, C. E. 1969. Pollen analysis of Iola bog and paleoecology of the Two Creeks forest bed, Wisconsin. Ecology 50: 859-868 + 1 pl. SyOrs, H. 1961. Forest and peatland at Hawley Lake, northern Ontario. Natl. Mus. Canada Bull. 171: 1-31. ——————— . 1963. Bogs and fens on Attawapiskat River, northern Ontario. Natl. Mus. Canada Bull. 186: 45-133 + pls. 3-6. Virr, D. H. 1973. A revision of the genus Orthotrichum in North America, north of Mexico. Bryophytorum Bibliotheca 1. 208 pp. + 60 pls. West, R. G. 1961. Late- and postglacial vegetational history in Wisconsin, particu- larly changes associated with the Valders readvance. Amer. Jour. Sci. 259: 766— 783 + 2 pls. Witson, L. R. 1932. The Two Creeks Forest Bed, Manitowoc County, Wisconsin. Trans. Wisconsin Acad. Sci. Arts Letters 27: 31-46. a . 1936. Further fossil studies of the Two Creeks Forest Bed, Manitowoc County, Wisconsin. Bull. Torrey Bot. Club 63; 317-325. A NEW COMBINATION IN THE GENUS GYMNOMYCES Donatp H. PFISTER While studying some members of the Hynangiales I examined speci- mens of Octaviania redolens Cunn. Based on the reexamination of the holotype, I propose that it be referred to the genus Gymnomyces Massee and Rodway. Gymnomyces redolens,! known only from New Zealand, represents a distinct, well-delimited species of the genus. It differs from G. pallidus Massee and Rodway, described from Tasmanian material, in the size and shape of the basidiospores and in the number of them produced on each basidium. The basidia of G. pallidus produce four spores each, as opposed to G. redolens in which the basidia consistently produce but a single spore. In this respect G. redolens resembles G. monospora Stewart and Trappe (1975). Both species are unlike other members of the genus since they produce single-spored basidia. However, they differ from each other in several features. The basidiospores of G. redolens have amyloid material covering many of the ornamentations and have low amyloid connecting lines between the spines. On the other hand, the basidiospores of G. mono- spora were described as having heavy amyloid deposits in the form of dots or collars at the base of the spores. Basidiospore size also distinguishes the two species. The basidiospores of G. redolens are globose, rarely sub- globose, 12-15 wm in diam, excluding ornamentations. Those of G. mono- spora are subglobose to broadly ellipsoid, 15-21 12-18 pm, excluding ornamentations. Additionally, the basidiospore ornamentations in G. monospora are in the form of unconnected spines rather than as those in G. redolens which join toward their bases to form a low, incomplete reticulum. Cunningham’s (1942, 1944) descriptions of Octaviania redolens fail to mention the presence of sphacrocystis in the tramal plates. He mentioned only the presence of interwoven hyphae. There are, however, numerous sphaerocystis in this layer. I wish to thank Joan M. Dingley, Auckland Plant Disease Division, for providing the specimens for study. SPECIMENS EXAMINED. Holotype: on ground, Te Aroha, 250 ft., Auckland, May, 1940, G. H. Cunningham, 10141 (ax). Three specimens collected by J. M. Dingley in Auckland (Ak 6355, 12334, 28585 ). LITERATURE CITED CunnincuaM, G. H. 1942. Two additional New Zealand Gasteromycetes. New Zealand J. Sci. Technol. 23: 172B-173B. ——-—--— . 1944. Gasteromycetes of Australia and New Zealand. Published by the author. 226 pp. Stewart, E. L. and J. M. TRAPPE. 1975. Gymnomyces monospora sp. Nov. Mycotaxon 2: 209-213. 1Gymnomyces redolens (Cunn.) Pfist. comb. nov. = Octaviania redolens Cunn., New Zealand J. Sci. Technol. 23: 172B. 1942. 43