Mycologia, 105(5), 2013, pp. 1287–1305. DOI: 10.3852/12-395 2013 by The Mycological Society of America, Lawrence, KS 66044-8897 # Pseudocosmospora, a new genus to accommodate Cosmospora vilior and related species Cesar S. Herrera1 INTRODUCTION University of Maryland, Department of Plant Science and Landscape Architecture, 2112 Plant Sciences Building, College Park, Maryland 20742 Cosmospora sensu Rossman (Nectriaceae, Hypocreales, Ascomycota; Gräfenhan et al. 2011) was erected to accommodate nectroid fungi with small, reddish, KOH+, smooth, thin-walled, laterally collapsing when dry, non- or weakly stromatic perithecia (Samuels et al. 1991, Rossman et al. 1999). These fungi have been reported worldwide, but they are assumed to have greater diversity in warm temperate and tropical regions. In addition, they tend to have a higher diversity in recently disturbed stands (1–2 y old) compared to early successional stands (25–27 y old) and old-growth stands in tropical forests (Chaverri and Vı́lchez 2006). In that study, frequently collected species in recently disturbed stands, where newly killed woody substrates and herbaceous debris are prevalent, were members of Chaetopsinectria Lou & Zhuang and Volutellonectria Lou & Zhang, two genera segregated from Cosmospora sensu Rossman (Luo and Zhuang 2010, 2012). Many species of Cosmospora sensu Rossman are parasites of their fungal hosts (see Tsuneda 1982). Among genera segregated from Cosmospora sensu Rossman, some members of Cosmospora sensu stricto grow on basidiomycetes or xylariaceous hosts; species of Dialonectria (Sacc.) Cooke occur on Diatrype Fr. (Diatrypaceae); and Microcera Desm. parasitize scale insects (Gräfenhan et al. 2011). The generic name Cosmospora has been a source of much taxonomic confusion. Rabenhorst (1862) described this genus that was reduced to a subgenus of Nectria (Fr.) Fr. by Saccardo (1883). Much later, it was synonymized with Dialonectria (Moravec 1954), which had been elevated from a subgenus of Nectria to generic rank by Cooke (1884). Rossman et al. (1999) resurrected the generic name Cosmospora based on priority. The group also has been referred to as Nectria subgenus Dialonectria or the ‘‘Nectria episphaeria-group’’ (Booth 1959, Samuels et al. 1991, Rossman et al. 1999). Early on, the group was presumed to be polyphyletic given its range of anamorphs and ecological niches (Samuels et al. 1991), and at one time as many as 70 species were classified under Cosmospora (www.indexfungorum. org). The polyphyly of Cosmospora was confirmed by Zhuang and Zhuang (2006), Luo and Zhuang (2008), Samuels et al. (2009) and Gräfenhan et al. (2011). Following the genus-for-genus concept, that is the delimitation of a genus based on the correlation of Amy Y. Rossman Gary J. Samuels United States Department of Agriculture, Agriculture Research Service, Systematic Mycology and Microbiology Laboratory, B-010A, 10300 Beltsville, Maryland 20705 Priscila Chaverri University of Maryland, Department of Plant Science and Landscape Architecture, 2112 Plant Sciences Building, College Park, Maryland 20742 Abstract: Cosmospora sensu Rossman accommodated nectroid fungi with small, reddish, smooth, thinwalled perithecia but recently was found to be polyphyletic and has been segregated into multiple genera. Not all cosmospora-like fungi have been treated systematically. Some of these species include C. vilior and many specimens often labeled ‘‘Cosmospora sp.’’ The objectives of this research were to establish the identity of C. vilior through epitypication using a recent collection that agrees with the type specimen in morphology, host and geography and to determine its phylogenetic position within Cosmospora sensu lato and the Nectriaceae. A multilocus phylogeny was constructed based on six loci (ITS, LSU, MCM7, rpb1, tef1, tub) to estimate a phylogeny. Results from the phylogenetic analyses indicated that C. vilior forms a monophyletic group with other cosmospora-like fungi that have an acremonium-like anamorph and that parasitize Eutypa and Eutypella (Ascomycota, Sordariomycetes, Xylariales, Diatrypaceae). The group is phylogenetically distinct from other previously segregated genera. A new genus, Pseudocosmospora, is described to accommodate the type species, P. eutypellae, and nine additional species in this clade. Key words: fungal systematics, GCPSR, Mycoparasite, Nectria, one-to-one genus concept Submitted 12 Dec 2012; accepted for publication 4 Apr 2013. 1 Corresponding author. E-mail: csherrer@umd.edu 1287 1288 MYCOLOGIA the teleomorph to its corresponding anamorph (Rossman 1993), Cosmospora was segregated into new or revived genera that correlate roughly with these anamorphs: Chaetopsinectria, Cyanonectria Samuels & P. Chaverri, Nectricladiella Crous & Schoch, Fusicolla Bonord., Macroconia (Wollenw.) Gräfenhan et al., Microcera, Stylonectria Höhn. and Volutellonectria (see Schoch et al. 2000; Samuels et al. 2009; Luo and Zhuang 2010, 2012; Gräfenhan et al. 2011). With the change to one scientific name for each species as directed in the International Code of Nomenclature for algae, fungi and plants (ICN) (McNeill et al. 2012), Chaetopsinectria and Volutellonectria are considered synonyms of the older genera Chaetopsina Rambelli and Volutella Fr. Cosmospora vilior was described as Nectria vilior by Starbäck (1899) with the diagnosis ‘‘Peritheciis discretis, superficialibus, ovoideis, coccineis…Hab. in fungillo valsaceo.’’ Traditionally the name has been applied to collections of cosmospora-like fungi having short, coarsely warted ascospores occurring on black stromata, particularly those of the Xylariales (Weese 1916, Samuels et al. 1990, Samuels et al. 1991). Nectria vilior has been reported to have a wide tropical and temperate distribution (Samuels et al. 1990). Re-examination of the type specimen of C. vilior revealed that its associated host is a species of Eutypella (Nitschke) Sacc. (Diatrypaceae). Our recent molecular analyses suggest that true C. vilior is unrelated to species of Cosmospora that occur on xylariaceous fungi, hereafter referred to as the C. viliuscula species complex. Species of the C. vilior complex occur only on species of Eutypella (Diatrypaceae) while C. viliuscula and related species are restricted to xylariaceous fungi. The current paper deals with the phylogenetic and taxonomic reassessment of the Cosmospora vilior and similar taxa. The objectives of this research are: (i) to establish the identity of C. vilior and stabilize the name using epitypification, (ii) to elucidate the phylogenetic placement of C. vilior and related species within Cosmospora sensu Rossman and in the Nectriaceae, (iii) to describe a new genus, Pseudocosmospora, to accommodate C. vilior and related species and (iv) to describe new species within Pseudocosmospora including the type P. eutypellae. MATERIALS AND METHODS Teleomorph and anamorph morphological characterization.— Herbarium specimens were borrowed from the U.S. National Fungus Collections (BPI), the William and Lynda Steere Herbarium, New York Botanical Garden (NY) and the Linnean Herbarium, Swedish Museum of Natural History (S). Fresh specimens were collected on trips to Argentina, Brazil, Costa Rica, France and USA. For the characterization of the teleomorph, these observations were made for perithecia: shape, size (length and width), color, ornamentation, and habit (e.g. perithecia being solitary or gregarious, immersed in substrata or superficial, stromatic or non-stromatic and collapsing laterally or not when dry). Reaction to 3% w/v potassium hydroxide (KOH) and 100% lactic acid was observed for the perithecial wall. Sections of perithecia (ca. 11 mm thick) were made with a freezing microtome. Measurements of continuous characters (e.g. length and width) were made with Scion Image software beta 4.0.2 (Scion Corp., Frederick, Maryland) and summarized by descriptive statistics (e.g. minimum, maximum, mean and standard deviation). Cultures were obtained from the culture collection at USDA, ARS, Systematic Mycology and Microbiology Laboratory (SMML). Additional cultures were obtained by isolating single ascospores from freshly collected samples with the aid of a micromanipulator and grown in cornmeal dextrose agar (CMD; DifcoTM cornmeal agar + 2% w/v dextrose + antibiotics). Morphological observations of the colony were made by growing three pseudoreplicates of each isolate on CMD and DifcoTM potato dextrose agar (PDA) in an incubator that alternated 12 h/12 h between fluorescent light and darkness at 25 C. Cultural morphology is described based on strains grown on PDA; cultures on CMD exhibit little variability. Colony color is described with the terms in Rayner (1970). Culture growth was measured weekly for 2 wk. The anamorph was observed by cutting an agar block of a culture grown in synthetic nutrient-poor agar (SNA; Nirenberg 1976) under the same conditions mentioned above, covering it with a cover slip and examining it by light microscopy (Olympus BX50; Olympus, Tokyo, Japan). Measurements of continuous characters were made and analyzed as described above. DNA extraction, PCR and sequencing.—The DNA extraction protocol is described in detail by Hirooka et al. (2010). Briefly, the isolates were grown in DifcoTM potato dextrose broth (PDB) and the mycelial mat was harvested after a week. DNA was extracted with PowerPlantH DNA Isolation Kit (MO BIO Laboratories Inc., Solana Beach, California). Six partial loci were amplified. These loci are internal transcribed spacer (ITS; primers ITS5, ITS4; White et al. 1990), large subunit nuclear ribosomal DNA (LSU; primers: LROR and LR5; Vilgalys and Hester 1990), MCM7 (a DNA replication licensing factor; primers: Mcm7-709 for & Mcm7-1348 rev; Schmitt el al. 2009), RNA polymerase II subunit one (rpb1; primers: Crpb1a & rpb1c; Castlebury et al. 2004), translation elongation factor 1-a (tef1; primers: Tef1-728, Tef1-986; Carbone and Kohn 1999) and b-tubulin (tub; O’Donnell and Cigelnik 1997). The PCR reaction mixture (25 mL total volume) consisted of 12.5 mL GoTaqHGreen Master Mix 23 (Promega Corp., Madison, Wisconsin), 1.25 mL for the forward and reverse primers each (10 mM), 1.0 mL dimethyl sulfoxide (DMSO; SigmaAldrich, St Louis, Missouri), up to 5.0 mL genomic DNA template, and RNAse-free water to complete the total volume. PCR reactions were carried out in an Eppendorf Mastercycler thermo-cycler (Eppendorf, Westbury, New HERRERA ET AL.: PSEUDOCOSMOSPORA York) under conditions listed (TABLE I). PCR products were cleaned with ExoSAP-ITH (USB Corp., Cleveland, Ohio). Clean PCR products were sequenced at the DNA Sequencing Facility (Center for Agricultural Biotechnology, University of Maryland, College Park, Maryland) and McLAB DNA sequencing services (San Francisco, California). Sequences were assembled and edited with Sequencher 4.9 (Gene Codes Corp., Madison, Wisconsin). Sequences were deposited in GenBank (SUPPLEMENTARY TABLE I). Phylogenetic analyses.—Two separate phylogenetic analyses were performed on two separate datasets as described below. The first dataset contained a reduced number of isolates of Cosmospora vilior and related taxa as well as species of other cosmospora-like fungi to elucidate their phylogenetic placement in the Nectriaceae. The second dataset contained all isolates of Cosmospora vilior and related taxa to determine their relationships. ITS-LSU, MCM7, rpb1, tef1 and tub sequences were aligned with MAFFT 6 (Katoh 2008) and manually edited if necessary in Mesquite 2.75 (Maddison and Maddison 2011). Gaps (insertions/deletions) were treated as missing data. Alignments were deposited in TreeBASE (http://www. treebase.org; accession number S14038). Maximum likelihood (ML) and Bayesian (BI) analyses were performed on each of the datasets of individual loci first and then on the concatenated dataset. CONCATEPILLAR 1.4 (Leigh et al. 2008) was used to determine whether loci could be analyzed by concatenating the datasets or whether loci should be analyzed separately. Loci were concatenated if the P value was greater than the default a-level of 0.05, which indicated that the null hypothesis (i.e. congruence of loci) could not be rejected. For both ML and BI analyses, jModeltest (Guindon and Gascuel 2003, Posada 2008) was used to infer the models of nucleotide substitution for each locus. Default settings in jModeltest were used: 11 substitution schemes with equal or unequal base frequencies (+F) and invariable sites (+I) and/or rate variation among sites (+G). The base tree for likelihood calculations was ML optimized. Once likelihood scores were calculated, the models were selected according to the Akaike information criterion (AIC). Maximum likelihood (ML) analyses were performed with GARLI 2.0 (Genetic algorithm for rapid likelihood inference; Zwickl 2006) by submitting the job via the GARLI web service at http://www.molecularevolution.org (Bazinet and Cummings 2011), which uses a grid computing system (Cummings and Huskamp 2005) associated with the Lattice Project (Bazinet and Cummings 2008). Fifty independent search replicates were performed to find the best tree. The starting tree was generated with a fast ML stepwise-addition algorithm. Two thousand bootstrap replicates were used for bootstrap analysis. Bayesian analyses were performed in MrBayes 3.2.1 (Ronquist et al. 2012). A majority-rule consensus tree was generated by running four chains for 10 000 000 Markov chain Monte Carlo generations, sampling trees every 100th generation and discarding the first 25% of the sampled trees as burn-in. Tracer 1.5 (Rambaut and Drummond 2007) was used to confirm whether the negative log likelihoods had reached convergence. 1289 RESULTS Phylogenetic analyses: phylogenetic placement of C. vilior and related species within Cosmospora sensu Rossman.—The analysis performed in CONCATEPILLAR failed to reject the null hypothesis of congruence among loci (P 5 0.08). Therefore all loci were concatenated. The concatenated matrix included 22 ingroup isolates that formed five major groups plus two outgroup taxa (Corallomycetella repens and Pseudonectria pachysandricola). It consisted of 3591 base pairs of which 970 were parsimony informative, 318 were parsimony uninformative and 1858 were invariable sites. The topologies of the generated phylogenetic trees in both ML and BI were congruent. The negative log likelihoods for the phylogenetic trees were 216460.154 and 216513.115 respectively. The best tree (ML) is illustrated (FIG. 1). Cosmospora vilior, C. joca and related species formed a highly supported clade (94% BP, 100% PP). This clade is related to Dialonectria, Cosmospora sensu stricto and an orphan group that includes C. flavoviridis, C. obscura and C. stegonsporii. These clades of cosmospora-like fungi were highly supported as well (.70% BP, .95% PP), but the inner nodes connecting these clades were poorly supported. Basal to all of these groups is Microcera, another segregate genus of cosmospora-like fungi. Phylogenetic analyses: relationship among C. vilior and related species.—The null hypothesis of congruence among loci (P 5 0.11) was not rejected in CONCATERPILLAR, and therefore the five loci were concatenated to estimate a phylogeny. The concatenated matrix included 25 isolates belonging to the ingroup and two outgroup taxa (C. repens and M. larvarum). The concatenated matrix consisted of 3353 bp of which 651 were parsimony informative, 456 were parsimony uninformative and 1785 invariable sites. The tree topologies generated with ML and BI were congruent. The log likelihoods for these two analyses were 223 024.3191 and 223 055.9286 respectively. The best tree generated with ML is illustrated (FIG. 2). The combined analyses of Cosmospora vilior and related species revealed that there were as many as 16 independently evolving lineages (5 putative species). Clade I is a complex of species whose hosts are Eutypella species. Three species are recognized within this clade, which include C. vilior and two species described below (Pseudocosmospora eutypellae, P. rogersonii). Sister to clade I is P. eutypae (described below), whose host is a species of Eutypa Tul. & C. Tul. Sister to clade II (P. eutypae + clade I) is clade III, which comprises two monotypic species, C. joca and P. metajoca (described below). Cosmospora joca is 1290 TABLE I. Genes/loci used in the phylogenetic analyses Locus ITS LSU Mcm7 Rpb1 Tef1 Tub 3591 970 318 1858 3353 651 456 1785 MYCOLOGIA Nucleotide substitution TIM2+I+2 TrN+I+G TIM2+I+G TrN+I+G TPM3uf+I+G models Included sites 1384 570 692 365 580 Phylogenetically 94 201 235 236 204 informative sites Uninformative 77 29 80 42 90 polymorphic sites Invariable sites 1028 324 282 25 199 Pseudocosmospora Nucleotide substitution TIMef+I+G TrN+I+G TIM2+I+G HKY+G TIM3+I+G dataset models Included sites 1323 561 642 280 547 Phylogenetically 65 44 211 172 159 informative sites Uninformative 64 166 70 64 92 polymorphic sites Invariable sites 992 327 253 0 213 Primers used (reference) ITS5, ITS4 LR5, LROR mcm7-709for, crpb1a, rpb1c tef1-728, tef1-986 Btub-TI, Btub-T2 (White et al. (Vilgalys and mcm7-1348rev (Castlebury et al. (Carbone and (O’Donnell and 1990) Hester 1990) (Schmitt et al. 2009) 2004) Kohn 1999) Cigelnik 1997) PCR protocol: annealing temperature 53 C, 1 min, 403 56 C, 50 s, 383 50 C, 2 min, 403 66 C, 55 s, 93 55 C, 30 s, 353 and cycles 56 C, 55 s, 353 Cosmospora sensu lato dataset Combined HERRERA ET AL.: PSEUDOCOSMOSPORA 1291 FIG. 1. Phylogenetic placement of C. vilior and related species within Cosmospora sensu Rossman based on a combined fiveloci (ITS-LSU, MCM7, rpb1, tef1, tub) dataset. Best tree generated with ML analysis (216 460.154). Values at branches indicate maximum likelihood bootstrap (ML BP)/Bayesian posterior probabilities (BI PP). associated with a species of Biscogniauxia Kuntze (Xylariaceae), while P. metajoca is associated with a species of Eutypa. All clades corresponding to recognized species received maximum BP and PP support (with one exception). TAXONOMY Pseudocosmospora C. Herrera & P. Chaverri, gen. nov. MycoBank MB802432 Type species: Pseudocosmospora eutypellae C. Herrera & P. Chaverri Etymology: ‘‘Pseudo’’ from Greek referring to the morphological similarity to both the teleomorphic and anamorphic states of Cosmospora sensu stricto. Teleomorph: Stroma absent. Perithecia superficial or slightly immersed in fungal host stroma, scattered to gregarious, subglobose to obpyriform with a blunt papilla, generally less than 250 mm high, soft-textured, smooth-walled, scarlet, KOH+ blood red, LA+ yellow, collapsing laterally when dry, uniloculate. Perithecial surface cells forming textura angularis. Perithecial wall generally 20–30 mm thick, of two regions, outer region of cells forming textura globulsa to t. angularis; inner region of cells forming textura prismatica. Asci unitunicate, cylindrical to narrowly clavate, increasing in size as ascospores mature, without a conspicuous apical ring, with eight spores arranged uniseriately. Ascospores ellipsoidal, oneseptate, slightly constricted at septum, yellow-brown, verrucose, sometimes appearing smooth at maturity. Anamorph in culture: After 21 d at room temperature on PDA, colony surface crustose with no aerial mycelium or cottony with aerial mycelium, rosy-buff, pale-luteous, or salmon-pink. Sporulation on SNA 1292 MYCOLOGIA FIG. 2. Phylogenetic relationship of C. vilior and related species based on a combined five-loci (ITS-LSU, MCM7, rpb1, tef1, tub) dataset. Best tree generated with ML analysis (223 024.3191). Values at branches indicate maximum likelihood bootstrap (ML BP)/Bayesian posterior probabilities (BI PP). usually abundant, arising directly from agar surface. Anamorphic state acremonium-like to verticilliumlike; conidiophores generally simple, unbranched, sometimes verticillately branched, rarely densely aggregated. Phialides monophialidic, cylindrical, hyaline. Conidia ellipsoidal, ovoid, or reniform, smooth, sometimes guttulated, non-septate, hyaline. Habitat: On stromata of diatrypaceous fungi, particularly species of Eutypa and Eutypella, rarely on species of Biscogniauxia. Distribution: Asia, Africa, Europe, North America, Oceania, South America, possibly cosmopolitan. Notes: Pseudocosmospora is similar to Cosmospora sensu stricto in its cosmospora-like teleomorph and acremonium-like anamorph, although they differ in cultural characteristics and host preference. Pseudocosmospora is most common on diatrypaceous fungi except for P. joca, which occurs on a Biscogniauxia sp. The latter does not belong among the known hosts of Cosmospora sensu stricto, although these attack xylariaceous fungi as well as polypores. In general, species of Pseudocosmospora have pinkish colonies, while species of Cosmospora sensu stricto have olivaceous green colonies on PDA. Phylogenetically Pseudocosmospora appears to be closely related to Dialonectria. Both occur on diatrypaceous fungi, although they attack different genera. The genera also differ in their anamorphic state; Dialonectria has a fusarium-like anamorph. KEY TO SPECIES OF PSEUDOCOSMOSPORA 1. 1. On Biscogniauxia (Xylariaceae) . . . . . . . . . . P. joca On Eutypa or Eutypella (Diatrypaceae) . . . . . . . . 2 HERRERA ET AL.: PSEUDOCOSMOSPORA 2. 2. 3. 3. 4. 4. 5. 5. 6. 6. 7. 7. 8. 8. 9. 9. On Eutypa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 On Eutypella . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Ascospores smooth, 6.3–8.7 mm long; colony rosy buff, 6–15 mm diam after 14 d at 25 C on PDA . . . . . . . . . . . . . . . . . . . . . . . . . . . . P. eutypae Ascospores verrucose, 7.7–11.9 mm long; colony salmon-pink, 24–25 mm diam after 14 d at 25 C on PDA . . . . . . . . . . . . . . . . . . . . . . . . . . P. metajoca Perithecia with a discoidal apex . . . . . . . . . . . . . . 5 Perithecia with a blunt apex . . . . . . . . . . . . . . . . 6 Fungal host on Alnus sp.; ascospores smooth, 9– 10.4 mm . . . . . . . . . . . . . . . . . . . . . . . P. pithoides Fungal host on Espeletia sp.; ascospores verrucose, 11–14 mm long . . . . . . . . . . . . P. pseudepisphaeria Conidiophores branching, becoming densely ramulose (fasciculate) on SNA . . . . . . . . . . P. triqua Conidiophores not branching or sparingly branched on SNA . . . . . . . . . . . . . . . . . . . . . . . . 7 Colonies pale-luteous on PDA; conidia reniform, with two guttules at opposite ends, 3.4–7.4 mm long . . . . . . . . . . . . . . . . . . . . . . . . . . . . . P. vilior Colonies white to salmon-pink on PDA . . . . . . . . 8 Colonies white on PDA; ascospores smooth, 10– 15 mm long . . . . . . . . . . . . . . . . . P. metepisphaeria Colonies salmon-pink on PDA . . . . . . . . . . . . . . . 9 Ascospores verrucose, 7.1–12.5 mm long; colonies 7.5–20 mm diam after 14 d at 25 C on PDA; conidia oblong to ellipsoidal, with two guttules at opposite ends, 3.1–6.2 mm long . . . . . . . . . . . . . P. eutypellae Ascospores smooth, 7.9–12.2 mm long; colonies 18– 27.5 mm diam after 14 d at 25 C on PDA; conidia oblong to ellipsoidal, without guttules, 2.9–5.5 mm long . . . . . . . . . . . . . . . . . . . . . . . . . . P. rogersonii Pseudocosmospora eutypae C. Herrera & P. Chaverri, sp. nov. FIG. 3. MycoBank MB802433 Holotype: FRANCE, Poitou-Charentes, Saint George de Rex (Marais Poitevin), on Eutypa sp., 26 Apr 2011, C. Herrera (C.H. 11-01), BPI 884164, ex-holotype culture CBS 133961. Etymology: In reference to its fungal host, Eutypa. Teleomorph: Perithecia solitary, superficial, nonstromatic, subglobose with a discoidal apex, collapsing laterally when dry, scarlet, smooth, 171–200 3 150– 183 mm (mean 5 182 3 165; SD 16.1, 16.5; n 5 3). Asci cylindrical to slightly clavate, with eight spores arranged uniseriately, 54–66 3 5.5–7 mm (mean 5 59 3 6; SD 5.4, 0.7; n 5 4). Ascospores ellipsoid, equally two-celled, one-septate, slightly constricted at septum, smooth, hyaline, 6.3–8.7 3 3.1–4.1 mm (mean 5 7.8 3 3.6; SD 0.7, 0.3; n 5 30). Anamorph: Colonies 6–15 mm diam (mean 5 11.4; SD 4.1; n 5 5) after 14 d. at 25 C on PDA, cottony with rosy-buff aerial mycelium, reverse concolorous. Sporulation on SNA usually abundant, arising directly 1293 from agar surface. Anamorphic state acremoniumlike; conidiophores generally simple, unbranched. Phialides monophialidic, cylindrical, collarette not flared, hyaline, length 36–53 mm (mean 5 41.3; SD 4.1; n 5 13), width at base 1.5–2.0 mm (mean 5 1.8; SD 0.2; n 5 13), width at tip 1–1.3 mm (mean 5 1.1; SD 0.1; n 5 13). Conidia oblong, unicellular, smooth, hyaline, 4.6–6.7 3 1.2–2.1 mm (mean 5 5.7 3 1.7; SD 0.6, 0.2; n 5 30). Habitat: On Eutypa cf. lata (Diatrypaceae) bark. Distribution: France, United Kingdom. Additional isolates examined: UNITED KINGDOM, on Crataegus sp., 1958, S. Francis, culture IMI 73016. Notes: Pseudocosmospora eutypae occurs on Eutypa cf. lata and has small and smooth ascospores. Pseudocosmospora metajoca is the only other species of Pseudocosmospora on Eutypa, but it has longer and verrucose ascospores. Pseudocosmospora eutypellae C. Herrera & P. Chaverri, sp. nov. FIG. 4. MycoBank MB802434 Holotype: USA, Maryland, Beltsville, on Eutypella sp., on dead twigs of unidentified tree, 7 Oct 2008, Y. Hirooka (Y.H. 08-17), BPI 884165, ex-holotype culture CBS 133966 5 A.R. 4562. Etymology: In reference to its fungal host, Eutypella. Teleomorph: Perithecia gregarious, slightly immersed in host stromata, subglobose with a blunt apex to obpyriform, collapsing laterally, scarlet, smooth, 143–303 3 108–205 mm (mean 5 202 3 150; SD 39, 22.7; n 5 22). Asci cylindrical to slightly clavate, eight-spored, uniseriately arranged, 63–78.6 3 5.9–7.9 mm (mean 5 71.6 3 6.7; SD 4.5, 0.6; n 5 27). Ascospores ellipsoid to fusiform, equally twocelled, slightly verrucose, yellow-brown, 7.1–12.5 3 3.6–5.6 mm (mean 5 9.9 3 4.5; SD 0.9, 0.4; n 5 136). Anamorph: Colonies 7.5–20 mm diam (mean 5 12.6, SD 3.4, n 5 18) after 14 d. at 25 C on PDA, sometimes crustose, with or without aerial mycelium, buff, rosy-buff or salmon-pink, reverse concolorous. Sporulation on SNA usually abundant, arising directly from agar surface, sometimes from lateral pegs. Anamorphic state acremonium-like to verticilliumlike; conidiophores simple, unbranched, or branched, becoming densely branched. Phialides monophialidic, cylindrical, collarette not flared, hyaline, length (3.8–)7.8–15.1(–18.5) mm (mean 5 10.2; SD 3.0; n 5 29), width at base 0.9–2.3 mm (mean 5 1.3; SD 0.3; n 5 29), width at tip 0.7–1.2 mm (mean 5 0.9; SD 0.14; n 5 29). Conidia oblong to ellipsoidal, unicellular, with two guttules at opposite ends, smooth, hyaline, 3.1–6.2 3 1.0–2.4 mm (mean 5 4.2 3 1.5; SD 0.6, 0.2; n 5 1 80). Habitat: On Eutypella sp. (Diatrypaceae) bark. 1294 MYCOLOGIA FIG. 3. Pseudocosmospora eutypae. A. Perithecia on natural substrata. Bar 5 200 mm. B. Perithecium in 3% KOH. Bar 5 100 mm. C. Median section of perithecium. Bar 5 100 mm. D. Perithecial surface cells. Bar 5 100 mm. E. Asci. Bar 5 10 mm. F. Ascospore. Bar 5 10 mm. G. Cultures after 3 wk at 25 C on PDA. Bar 5 10 mm. H. Phialide. Bar 5 10 mm. I. Conidia. Bar 5 10 mm. HERRERA ET AL.: PSEUDOCOSMOSPORA 1295 FIG. 4. Pseudocosmospora eutypellae. A, B. Perithecia on natural substrata. A. Bar 5 600 mm; B. bar 5 200 mm. C. Perithecium in 3% KOH. Bar 5 100 mm. D. Median section of perithecium. Bar 5 100 mm. E. Perithecial surface cells. Bar 5 100 mm. F. Asci. Bar 5 10 mm. G. Ascopore. Bar 5 10 mm. H, I. Cultures after 3 wk at 25 C on PDA. Bars 5 10 mm. J, K. Phialides. Bars 5 10 mm. L. Lateral phialidic pegs. Bar 5 10 mm. M. Conidia. Bar 5 10 mm. 1296 MYCOLOGIA Distribution: France and U.S.A. Additional specimens and isolates examined: FRANCE, Oloron, Forêt de Bugangue, on Eutypella sp., on bark of Robinia pseudoacacia (?), 17 May 1993, F. Candoussau & J.D. Rogers (F. 262), BPI 802567, culture CBS 128986 5 G.J.S. 93-15; USA, Kentucky, Clermont, Bernheim Arboretum and Research Forest, on Eutypella sp., on dead branch of unidentified tree, 27 Jun 2010, Y. Hirooka, BPI 884169, culture CBS 133977 5 G.J.S. 10-248; Maryland, Frederick County, Cunningham Falls State Park, on Eutypella sp., on Rhus typhina, 26 Aug 2007, L. Vasilyeva, BPI 878454, culture CBS 129430 5 A.R. 4453; Pennsylvania, Greensburg, on Eutypella sp., Aug 2008, J. Plitschke, culture CBS 133965 5 A.R. 4527; West Virginia, Grafton, on Eutypella sp., on bark of unidentified tree, 26 Jun 2010, Y. Hirooka, BPI 884168, culture CBS 133960 5 C.H. 10-02. Notes: Pseudocosmospora eutypellae is most closely related and similar to P. rogersonii but can be distinguished from the latter by the ornamentation of its ascospores. Pseudocosmospora eutypellae has verrucose ascospores, while P. rogersonii has smooth ascospores. Pseudocosmospora joca (Samuels) C. Herrera & P. Chaverri, comb. nov. FIG. 5. Mycobank MB802435 Basionym: Nectria joca Samuels, Mycol. Pap. 164:21. 1991. ; Cosmospora joca (Samuels) Rossman & Samuels, Stud. Mycol. 42:122. 1999. Teleomorph: Perithecia gregarious, superficial, nonstromatic, subglobose with a minute papilla, collapsing laterally, scarlet at first, becoming blood red, darker at apex, smooth, 375–384 3 317–349 mm (mean 5 380.2 3 336.3; SD 4.9, 16.9; n 5 3). Asci cylindrical to clavate, eight-spored, uniseriately arranged, 90.7–112.6 3 8.8–11.1 mm (mean 5 102.2 3 9.8; SD 8.5, 0.7; n 5 9). Ascospores ellipsoid, equally two-celled, one-septate, constricted at septum, verrucose, yellow-brown, 10.9–14 3 6.4–7.7 mm (mean 5 12.8 3 7; SD 0.8, 0.3; n 5 30). Anamorph: Colonies 4 mm diam (n 5 3) after 14 d. at 25 C on PDA, crustose, salmon-pink to orange colony, reverse concolorous. Rarely sporulating on SNA. Anamorphic state acremonium-like; conidiophores generally simple, unbranched. Phialides monophialidic, cylindrical, hyaline, length 14.3– 24.2 mm (mean 5 18.8; SD 5.0; n 5 3), width at base 1.6–2.4 mm (mean 5 2.0; SD 0.4; n 5 3), width at tip 1.1 (n 5 3). Conidia oblong, unicellular, smooth, hyaline, 3.0–5.5 3 1.3–2.1 mm (mean 5 4.1 3 1.6; SD 0.6, 0.2; n 5 30). Habitat: On Biscogniauxia sp. bark. Distribution: Argentina and Brazil. Holotype: BRAZIL, Amazonas, Pico Rondon, Km 211 on Perimetral Norte, ca. 3 h walk from FUNAI post toward summit, 01u329N, 02u489W, on Biscogniauxia sp., 25 Mar. 1984, G.J. Samuels (1094), Pipoly & Guedes, INPA (not seen), ISOTYPES BPI 802606, NY 00671973 (not seen). Epitype designated herein: ARGENTINA, Rı́o Negro Province, San Carlos de Bariloche, Luma forest, on Biscogniauxia sp., on rotted wood, 15 Apr 2011, A. Romero, BPI 884175, ex-epitype culture CBS 133967 5 A.R. 4779. Notes: The application of the name is restricted here to species of Pseudocosmospora on Biscogniauxia. The isotype and the designated epitype both occur on species of Biscogniauxia. The colony and the anamorph are similar to the morphology described in the original description, although the perithecia and ascospores of the epitype are larger than those reported in the literature. Pseudocosmospora metajoca C. Herrera & P. Chaverri, sp. nov. FIG. 6. MycoBank MB802436 Holotype: NEW ZEALAND, North Island, Mount Williams, on Eutypa sp., on dead woody branch of Beilschmiedia tawa, 7 Mar 2009, A.Y. Rossman & P. Chaverri (P.C. 952), BPI 879088, ex-holotype culture CBS 133968 5 A.R. 4576. Etymology: ‘‘Meta’’ from Greek meaning adjacent and ‘‘joca’ in reference to the fact that it originally was classified as C. joca, and later found to be phylogenetically close to C. joca. Teleomorph: Solitary or gregarious, superficial, nonstromatic, subglobose with a discoidal apex, some collapsing laterally, scarlet, smooth, 222–251 3 204– 213 mm (mean 5 236 3 208; n 5 2). Asci clavate, eight-spored, uniseriately arranged, 62.2–69.2 3 5.9– 7.2 mm (mean 5 65.5 3 6.4; SD 2.5, 0.5; n 5 5). Ascospores ellipsoid, equally two-celled, one-septate, slightly constricted at septum, slightly verrucose, yellow-brown, 7.7–11.9 3 3.3–5.3 mm (mean 5 8.9 3 4.3; SD 0.9, 0.5; n 5 29). Anamorph: Colonies 24–25 mm diam (mean 5 24.5; SD 0.5; n 5 3) after 14 d. at 25 C on PDA, slightly cottony, pale salmon-pink, reverse concolorous. Sporulation on SNA usually abundant, arising directly from agar surface. Anamorphic state acremonium-like; conidiophores generally simple, unbranched. Phialides monophialidic, cylindrical, collarette not flared, hyaline, length 26–49 mm (mean 5 37.3; SD 5.7; n 5 9), width at base 2.1–2.8 mm (mean 5 2.4; SD 0.2; n 5 9), width at tip 1–1.3 mm (mean 5 1.2; SD 0.1; n 5 9). Conidia oblong to ellipsoidal, unicellular, guttulated, smooth, hyaline, 3.8–6.1 3 1.6–3.1 mm (mean 5 4.8 3 2.1; SD 0.6, 0.3; n 5 30). Habitat: On Eutypa sp. (Diatrypaceae) on dead branch of Beilschmiedia tawa. Distribution: New Zealand. HERRERA ET AL.: PSEUDOCOSMOSPORA 1297 FIG. 5. Pseudocosmospora joca. A, B. Perithecia on natural substrata. A. Bar 5 2 mm; B bar 5 200 mm. C. Asci. Bar 5 10 mm. D. Ascospores. Bar 5 10 mm. E. Cultures after 3 wk at 25 C on PDA. Bar 5 10 mm. F. Phialide. Bar 5 10 mm. G. Conidia. Bar 5 10 mm. Notes: Pseudocosmospora metajoca originally was identified as C. joca based on its occurrence on what was thought to be a stroma of a Biscogniauxia species and its salmon-pink culture on PDA. On close examination of the specimen, the host was found to be a species of Eutypa. The colony of P. metajoca grows faster than of P. joca. In addition, P. metajoca has much smaller perithecia and ascospores com- 1298 MYCOLOGIA FIG. 6. Pseudocosmospora metajoca. A, B. Perithecia on natural substrata. A. Bar 5 2 mm; B. bar 5 200 mm. C. Perithecium in 3% KOH. Bar 5 100 mm. D. Median section of perithecium. Bar 5 100 mm. E. Asci. Bar 5 10 mm. F. Ascospores. Bar 5 10 mm. G. Cultures after 3 wk at 25 C on PDA. Bar 5 10 mm. H, I. Phialides. Bars 5 10 mm. J. Conidia. Bar 5 10 mm. pared to P. joca. Pseudocosmospora metajoca differs from P. eutypae, the other Pseudocosmospora on Eutypa, by having verrucose ascospores. Basionym: Nectria metepisphaeria Samuels, Mycol. Pap. 164:29. 1991. ; Cosmospora metepisphaeria (Samuels) Rossman & Samuels, Stud. Mycol. 42:123. 1999. Pseudocosmospora metepisphaeria (Samuels) C. Herrera & P. Chaverri, comb. nov. MycoBank MB802437 Anamorph: Acremonium-like Habitat: On Eutypella sp. (Diatrypaceae) on unidentified bark. HERRERA ET AL.: PSEUDOCOSMOSPORA Distribution: Venezuela (known only from the type collection) Holotype: VENEZUELA, Dist. Federale, vic. Macarao, on Eutypella sp., on unidentified bark, 21 Jun 1971, K.P. Dumont (VE 335), J.H. Haines, G. Morillo & E. Moreno, VEN (not seen), ISOTYPE NY. Notes: The isotype specimen was studied and determined to occur on a Eutypella sp. Based on this host, it can be predicted that C. metepisphaeria would fall within the Pseudocosmospora clade. In addition to the host, the reported acremonium-like anamorphic state supports the placement of this species in Pseudocosmospora. Unique to this species is its smooth ascospores, (10–)11–14(–15) mm long and the white, crustose colony on PDA, reverse brown (Samuels et al. 1991). A culture no longer exists. Pseudocosmospora pithoides (Ellis & Everh.) C. Herrera & P. Chaverri, comb. nov. MycoBank MB802438 Basionym: Nectria pithoides Ellis & Everh., Proc. Acad. Nat. Sci. Philad. 43:247 (1891). Anamorph: Unknown Habitat: On an Eutypella sp. (Diatrypaceae) bark of dead alder. Distribution: British Columbia (known only from the type collection). Holotype: CANADA, British Columbia, bark of dead alder, May 1889, J. Macoun (122), NY 00927939. Notes: The holotype specimen of Nectria pithoides was examined and determined to agree with the concept of Pseudocosmospora in regard to the host, which appears to be a Eutypella species. The perithecia have a prominent discoidal apex, which according to the description gives an impression of being barrel-shaped (pithos from Greek 5 barrel). No asci were observed. The ascospores are ellipsoidal, one-septate, slightly constricted at the septum, smooth, 9–10.4 3 4.1–4.5 mm (mean 5 9.7 3 4.4; SD 0.5, 0.1; n 5 8). Pseudocosmospora pseudepisphaeria (Samuels) C. Herrera & P. Chaverri, comb. nov. MycoBank MB802439 Basionym: Nectria pseudepisphaeria Samuels, Mycol. Pap. 164:34. 1991. ; Cosmospora pseudepisphaeria (Samuels) Rossman & Samuels, Stud. Mycol. 42:124. 1999. Anamorph: Acremonium-like. Habitat: On Eutypella sp. (Diatrypaceae) on branch of Espeletia sp. Distribution: Venezuela (known only from the type collection). Holotype: VENEZUELA, Merida, Parque Nacional Sierra Nevada, near Apartaderos, E. of Laguna 1299 Mucubaji, Laguna Negra, on Eutypella sp., on Espeletia sp., 18 Jul 1971, K.P. Dumont (VE 2277), J.H. Haines, G.J. Samuels & A. Revas, NY 01013169. Notes: Based on our examination of the holotype specimen, the fungal host of C. pseudepisphaeria is a Eutypella sp. The fungal host and the reported acremonium-like anamorphic state support the placement of C. pseudepisphaeria in the genus Pseudocosmospora. Unique to this species are the discoidal perithecial apices, its verrucose, (11–)11.2–13(–14) mm long ascospores, and its white to pale salmoncolored colony (Samuels et al. 1991). A culture no longer exists. Pseudocosmospora rogersonii C. Herrera & P. Chaverri, sp. nov. FIG. 7. MycoBank MB802440 Holotype: USA, New York, Dutchess County, Pawling, Pawling Nature Reserve, on Eutypella sp., 6–8 Oct. 1990, G.J. Samuels & C.T. Rogerson, BPI 1107121, exholotype culture CBS 133981 5 G.J.S. 90-56. Etymology: In honor of Clark T. Rogerson for his work on the Hypocreales that has guided all of us. Teleomorph: Perithecia gregarious, slightly immersed in host stromata, subglobose with a blunt papilla, collapsing laterally, scarlet, smooth, 163–245 3 131–180 mm (mean 5 193 3 152; SD 37, 21; n 5 7). Asci broadly cylindrical to narrowly clavate, eightspored, uniseriately arranged, 54–69 3 5.7–8.4 mm (mean 5 63 3 6.7; SD 4.9, 0.7; n 5 12). Ascospores ellipsoid, equally two-celled, one-septate, slightly constricted at septum, smooth, yellow-brown, 7.9– 12.2 3 3.3–4.9 mm (mean 5 9.6 3 4.1; SD 0.9, 0.3; n 5 86). Anamorph: Colonies 18–27.5 mm diam (mean 5 22.4; SD 3.4; n 5 8) after 14 d at 25 C on PDA, crustose, rosy-buff to salmon-pink, reverse concolorous. Sporulation on SNA usually abundant, arising directly from agar surface. Anamorphic state acremonium-like; conidiophores generally simple, unbranched. Phialide cylindrical, smooth, straight, collarette not flared, hyaline, length 6.8–29.4 mm (mean 5 12.3; SD 5; n 5 30), width at base 1.0–2.3 mm (mean 5 1.5; SD 0.3; n 5 30), width at tip 0.7–1.2 mm (mean 5 0.9; SD 0.13; n 5 30). Conidia oblong to ellipsoidal, unicellular, smooth, hyaline, 2.9–5.5 3 1.1–2.6 mm (mean 5 3.8 3 1.6; SD 0.6, 0.3; n 5 89). Habitat: On Eutypella sp. (Diatrypaceae) bark. Distribution: USA. Additional specimens and isolates examined: USA, New York, Dutchess County, Pawling, Pawling Nature Reserve, on Eutypella sp., 6–8 Oct 1990, G.J. Samuels & C.T. Rogerson, BPI 1107120; New York, Huguenot, YMCA Greenkill Retreat Center, on Eutypella sp., 26 Sep 2009, C. Herrera (C.H. 0902), BPI 884167, culture 5 G.J.S. 09-1384; New York, 1300 MYCOLOGIA FIG. 7. Pseudocosmospora rogersonii. A, B. Perithecia on natural substrata. A. Bar 5 2 mm; bar 5 200 mm. C. Perithecium in 3% KOH. Bar 5 100 mm. D. Perithecial surface cells. Bar 5 100 mm. E. Ascus. Bar 5 10 mm. F. Ascospores. Bar 5 10 mm. G, H. Cultures after 3 wk at 25 C on PDA. Bar 5 10 mm. I. Phialide. Bar 5 10 mm. J. Conidia. Bar 5 10 mm. Bars: A 5 2 mm; B 5 200 mm; C, D 5 100 mm; E,F, I, J 5 10 mm; G, H 5 10 mm. Painted Post, Watson Homestead Conference and Retreat Center, on Eutypella sp., on dead branch of Fagus grandifolia, 17 Sep 2010, C. Herrera (C.H. 10–11), BPI 884166, culture CBS 133978 5 G.J.S. 10-296; New York, Painted Post, Watson Homestead Conference and Retreat Center, on Eutypella sp., on dead branch of Fagus grandifolia, 17 Sep 2010, C. Herrera (C.H. 10-12), BPI 884170, culture CBS 133979 5 G.J.S. 10-297. ornamentation of its ascospores. Pseudoscosmospora rogersonii has smooth ascospores in contrast to P. eutypellae, which has verrucose ascospores. Notes: Pseudocosmospora rogersonii is closely related to P. eutypellae but differs conspicuously in the Basionym: Nectria triqua Samuels, Mycol. Pap. 164:40. 1991. Pseudocosmospora triqua (Samuels) C. Herrera & P. Chaverri, comb. nov. MycoBank MB802441 HERRERA ET AL.: PSEUDOCOSMOSPORA ; Cosmospora triqua (Samuels) Rossman & Samuels, Stud. Mycol. 42:125. 1999. Anamorph: Acremonium-like. Habitat: On Eutypella sp. (Diatrypaceae) on unidentified bark. Distribution: French Guiana (known only from the type collection). Holotype of Nectria triqua: FRENCH GUIANA, Upper Marouini River, vic. roche Koutou, 02u559N, 54u049W, 400 m., on Eutypella sp., on unidentified bark, 17 Aug 1987, G.J. Samuels (5818), J.-J. de Granville, L. Allorge, W. Hahn, M. Hoff, NY 01013269. Notes: Examination of the holotype revealed that the host is a Eutypella sp., which suggests that C. triqua should be placed in the genus Pseudocosmospora. In addition, the reported anamorphic state is similar to that of P. vilior and P. eutypellae in having branching conidiophores branch that terminate with multiple phialides. Cultural morphology in PDA was not reported in the description of Nectria triqua (Samuels et al. 1991). The culture no longer exists. The ascospores are verrucose and (6.8–)7.8–9.7(– 10.5) mm long. Pseudocosmospora vilior (Starbäck) C. Herrera & P. Chaverri, comb. nov. FIG. 8. MycoBank MB802442 Basionym: Nectria vilior Starbäck, Bih. Kongl. Svenska Vet.Acad. Handl. 25:28. 1899. ; Cosmospora vilior (Starbäck) Rossman & Samuels, Stud. Mycol. 42:126. 1999. Teleomorph: Perithecia gregarious, slightly immersed in host stromata, subglobose with blunt apex, collapsing laterally, scarlet, smooth, 195–224 3 136– 183 mm (mean 5 213 3 164; SD 9.1, 14.7; n 5 10). Asci cylindrical to clavate, eight-spored, uniseriately arranged, 59–81 3 5.3–11.0 mm (mean 5 69 3 7.6; SD 6.2, 1.7; n 5 18). Ascospores ellipsoid, equally twocelled, one-septate, slightly constricted at septum, slightly verrucose, yellow-brown, 8.3–13.0 3 4.1– 6.4 mm (mean 5 10.2 3 5.2; SD 1.0, 0.5; n 5 90). Anamorph: Colonies 23–65 mm diam (mean 5 49; SD 16.9; n 5 8) after 21 d at 25 C on PDA, cottony with pale luteous aerial mycelium, reverse concolorous. Sporulation on SNA usually abundant, arising directly from agar surface; acremonium-like to verticillium-like; conidiophores simple and unbranched at first, becoming densely branched. Phialides cylindrical, smooth, straight, collarette not flared, hyaline, length 5.9–19.5 mm (mean 5 12.9; SD 3.4; n 5 9), width at base 1.1–1.8 mm (mean 5 1.5; SD 0.2; n 5 29), width at tip 0.7–1.3 mm (mean 5 1.0; SD 0.2; n 5 29). Conidia reniform, unicellular, smooth, with two guttules at opposite ends, hyaline, 3.4–7.4 3 1.1– 2.3 mm (mean 5 4.8 3 1.7; SD 0.7, 0.3; n 5 90). 1301 Habitat: On Eutypella sp. (Diatrypaceae) bark. Distribution: Argentina and Brazil. Holotype: BRAZIL, Rio Grande do Sul, Santo Angelo pr. Cachoaira, on Eutypella sp., 12 Jan 1893, Gustav Malme (114), S F46424. Epitype designated herein: ARGENTINA, Misiones Province, Iguazú Biological Station, on Eutypella sp., 25 Apr 2011, A.Y. Rossman, C. Salgado, A. Romero, R. Sanchez, BPI 884176, ex-epitype culture CBS 133971 5 A.R. 4810. Additional specimens and isolates examined: ARGENTINA, Tucuman Province, Tucuman, on Eutypella sp., on standing dead branch of Piper tucumanum, 19 Apr 2011, A. Romero, BPI 884174, culture CBS 133970 5 A.R. 4771; BRAZIL, Bahia, Igrapiúna, on Eutypella sp., 12 Aug 2010, P. Chaverri (P.C. 1246), O. Liparini Pereira, D. Pinho, A. Luiz Firmino, BPI 884172, culture CBS 133963. Notes: An epitype was needed to establish an anamorph for P. vilior and to determine its phylogenetic placement. The epitype was selected based on the relatively proximity to the collecting site of the holotype. The application of the name is restricted to species of Pseudocosmospora on Eutypella from South America that have pale-luteous colonies on PDA. However, it is recognized here that P. vilior consists of a species complex. DISCUSSION Genus concept.—The generic concept Cosmospora sensu stricto is based on its type Cosmospora coccinea Rabenh., which has Verticillium olivaceum W. Gams as its anamorph. Although the anamorph bears the name Verticillium Nees, the anamorphic state is acremonium-like (single phialide, unbranched) to verticillium-like (branching into multiple phialides). Accepted species in Cosmospora sensu stricto have an acremonium-like anamorph, and it is the character that circumscribes the genus (Gräfenhan et al. 2011). Conidiophore branching is not unique to the anamorph of C. coccinea because this also is observed in some anamorphs in the Cosmospora viliuscula species complex. Pseudocosmospora (described above) is recognized as a new genus based on the one-to-one genus concept suggested by Rossman (1993) to accommodate C. vilior and related species. The one-to-one genus concept has been used extensively in the Ascomycota to delimit genera (e.g. Gräfenhan et al. 2011; Luo and Zhang 2010, 2012). Briefly, this genus concept suggests that a genus should be circumscribed based on the correlation of its teleomorph to its unique anamorph state and vice versa. The groups circumscribed based on this concept are monophyletic and often supported by ecological traits (e.g. 1302 MYCOLOGIA FIG. 8. Pseudocosmospora vilior. A, B. Perithecia on natural substrata. A. Bar 5 2 mm; B. bar 5 200 mm. C. Perithecium in 3% KOH. Bar 5 100 mm. D. Asci. Bar 5 10 mm. E. Ascospore. Bar 5 10 mm. (F) Cultures after 3 wk at 25 C on PDA. Bar 5 10 mm. G. Phialides. Bar 5 10 mm. H. Conidia. Bar 5 10 mm. Gräfenhan et al. 2011; Luo and Zhang 2010, 2012). In Gräfenhan et al. (2011), the reported hosts for members of Cosmospora sensu stricto were basidiomycetes (e.g. Fomitopsis P. Karst., Inonotus P. Karst. and Stereum Hill ex Pers.) and xylariaceous fungi (e.g. Hypoxylon Bull.). Microcera and Dialonectria species have fusarium-like anamorphs, Microcera species are parasites of scale insects and the lectotype species of Dialonectria, D. episphaeria, is reported on Diatrype stigma (Hoffm.) Fr. (Diatrypaceae; Booth 1959). The host of Dialonectria ullevolea Seifert & Gräfenhan has not been identified, but it is predicted here that the host will be a diatrypaceous fungus. The orphan clade consisting of C. flavoviridis (Fuckel) Rossman & Samuels, C. stegonsporii Rossman, Farr & Akulov and C. obscura Lowen has not been taxonomically revised HERRERA ET AL.: PSEUDOCOSMOSPORA and might require generic recognition. Species in this clade have a fusarium-like anamorphs, but little is known about their fungal hosts. Only the host of C. stegonsporii, Stegonsporium pyriforme (Hoffm. : Fr.) Corda (Diaporthales, Sordariomycetes), has been identified to species. It is possible that all fungal hosts of species in this clade are members of the Diaporthales. The one-to-one genus concept is ideal for the circumscription of genera in the Ascomycota because it forces the study of the holomorph and not only the teleomorph or anamorph. Such view is crucial in shifting to one name (Norvell 2011). Discarding information of either the teleomorph or anamorph to favor one generic hypothesis over the other may result in para- or polyphyletic groups. For example, a weak case could be made to group Cosmospora, Dialonectria, Pseudocosmospora and the orphan clade that consists of C. flavoviridis, C. stegonsporii and C. obscura into one genus because they have a cosmospora-like teleomorph and occur generally on Sordariomycetes. However, when the anamorphs are superimposed on the phylogeny, a paraphyletic group is formed with two groups having acremonium-like anamorphs and the remaining two groups fusariumlike anamorphs. It suggests that the teleomorph state is probably a symplesiomorphic character (or ancestral), while the anamorph represents a synapomorphic character (derived). Moreover, segregation of the discussed genera is supported by specialization to different host taxa. The cosmospora-like teleomorphic state of Pseudocosmospora was correlated here to an acremonium-like anamorph. Our phylogeny (FIG. 2) demonstrates that Pseudocosmospora (BP 100%, PP 100%) is not congeneric with Cosmospora s.str., the only other group of cosmospora-like fungi with an acremonium-like anamorph (Hirooka et al. 2010, Gräfenhan et al. 2011). The two groups differ primarily in their cultural characteristics. In general, Pseudocosmospora produces pinkish colonies while Cosmospora sensu stricto produces olivaceous-green colonies on PDA. Members of each genus considered in this study occur only on a particular group of host fungi. Pseudocosmospora is reported here to occur primarily on Eutypa and Eutypella species (Diatrypaceae) with the exeption of Cosmospora joca (Samuels) Rossman & Samuels, whose host is a species of Biscogniauxia. The genus Dialonectria also occurs on diatrypaceous fungi but has a fusarium-like anamorphic state as do species in the genus Microcera that occur primarily on insects. Species concept.—Genealogical Concordance Phylogenetic Species Recognition was used to delimit species 1303 boundaries (GCPSR; Taylor et al. 2000). According to this operational species concept, putative species are clades that are concordant across all single gene trees. The morphological species recognition also was used to support the species inferences made when applying GCPSR. Hence, inferred species may be associated with unique morphological features that set them apart from other closely related species. Another species concept that could be useful in determining additional characters to delimit species is the ecological species concept. Ecological niches or adaptive zones can be used to delimit species (reviewed in de Queiroz 2007), according to this species concept. Host, an ecological niche, could be a character specific to a particular Pseudocosmospora species. However, host identification of Eutypa and Eutypella to species was not possible based on morphology alone. Identification of the fungal host based on DNA sequences would resolve this problem. Moreover, analyzing DNA sequences of cosmosporalike fungi and their associated fungal hosts would allow testing the hypothesis of cospeciation. Evidence for cospeciation would provide independent evidence for the delimitation of species in this genus. A problem of GCPSR is that it requires multiple individuals per species. By definition a clade is formed by a minimum of two individuals per species (reviewed in Vinuesa 2010). In the case of this paper, only a single collection was made for many of the lineages, and it left us with a dilemma on how to deal with the many singletons present in the phylogeny (FIG. 2). It was decided to use the rule of rarity (reviewed in Lim et al. 2012) to recognize a singleton, Pseudocosmospora metajoca (described below), as a species. This species is morphologically and ecologically distinct from species recognized with GCPSR and other singletons. Pseudocosmospora metajoca occurs on an Eutypa sp. on Beilschmiedia tawa (A.Cunn.) Kirk (Lauraceae), which is a broadleaf tree native to New Zealand and has verrucose ascospores. Also supporting the view that P. metajoca is a distinct species is the relatively long branch length, which indicates that there have been multiple substitutions per site since its segregation. Cosmospora vilior represents a case where morphology is insufficient to distinguish closely related species. This species is characterized by its relatively fast growing, pale luteous colony on PDA. However, the clade probably represents a species complex given the highly supported subclade that consists of the strains AR 4771 and PC 1246. The species complex may consist of up to three species, but the selected epitype strain, AR 4810, is considered closer to the true C. vilior based on geographical proximity to the original collecting site of the type specimen and its host. 1304 MYCOLOGIA Phylogenetic placement of Cosmospora joca.—The phylogenetic placement of Cosmospora joca, the host of which is a Biscogniauxia sp., is puzzling considering that other members of Pseudocosmospora have an Eutypa or Eutypella species (Diatrypaceae) as their host. Two potential explanations for this observation are (i) that the Biscogniauxia sp. represents the ancestral host for Psedocosmospora species or (ii) that the Biscogniauxia host of C. joca represents an independent host shift. Given that Pseudocosmospora species have diatrypaceous and xylariaceous hosts and assuming the first view, Pseudocosmospora could represent a link in the divergence from Cosmospora sensu stricto to Dialonectria (or vice versa). Phylogenies of the fungal hosts have placed the Diatrypaceae as a sister clade to Xylariaceae (Moster et al. 2004, Tang et al. 2009), and these cosmospora-like fungi could have tracked their hosts faithfully as they diverged. Host specificity is not uncommon in the Hypocreales (e.g. species of Cordyceps sensu lato are known to be host specific to insect species; Sung et al. 2007). Co-evolution/cospeciation analyses are needed to test these hypotheses. ACKNOWLEDGMENTS We gratefully acknowledge the assistance of the curators and their staff of the herbaria from which specimens were generously loaned. These herbaria include U.S. National Fungus Collection (BPI), William and Lynda Steere Herbarium, New York Botanical Garden (NY) and Herbarium of the Botany Department, Swedish Museum of National History (S). We also are grateful for the help of Yuuri Hirooka (Forestry and Forest Products Research Institute, Japan), Peter Johnston (Landcare Research, New Zealand), Carlos Mendez (University of Costa Rica), John Plitschke (Pennsylvania), Andrea Romero and Romina Sanchez (Departamento de Biodiversidad y Biologia Experimental, Universidad de Buenos Aires, Argentina) and Catalina Salgado (PSLA, UMD, USA) for contributing in the organization and collection of various specimens on collecting trips. The first author especially thanks ChunJuan Wang (State University of New York, College of Environmental Science and Forestry) for mentoring the author as an undergraduate and guiding the author into the field of mycology. The first author also thanks current and former colleagues at USDA-ARS, SMML (USA) and PSLA, University of Maryland (USA) laboratories, for their moral support. This study was financially supported by the United States National Science Foundation (NSF) PEET grant DEB-0731510 Monographic Studies in the Nectriaceae, Hypocreales: Nectria, Cosmospora, and Neonectria to P. Chaverri, A.Y. Rossman and G.J. Samuels. LITERATURE CITED Bazinet AL, Cummings MP. 2008. The Lattice Project: a grid research and production environment combining multiple grid computing models. In: Weber MHW, ed. Distributed and grid computing—science made transparent for everyone. Principles, applications and supporting communities. Marburg, Germany: Rechenkraft.net. p 2–13. ———, ———. 2011. Computing the tree of life—leveraging the power of desktop and service grids. IPDPS workshops 2011:1896–1902. Booth C. 1959. Studies of Pyrenomycetes IV. Nectria I. Mycol Pap 73:1–115. Carbone I, Kohn LM. 1999. A method for designing primer sets for speciation studies in filamentous ascomycetes. Mycologia 91:553–556, doi:10.2307/3761358 Castlebury LA, Rossman AY, Sung GH, Hyten AS, Spatafora JW. 2004. Multigene phylogeny reveals new lineage for Stachybotrys chartarum, the indoor air fungus. Mycol Res 108:864–872, doi:10.1017/S0953756204000607 Chaverri P, Vilchez B. 2006. Hypocrealean (Hypocreales, Ascomycota) fungal diversity in different stages of tropical forest succession in Costa Rica. Biotropica 38: 531–543, doi:10.1111/j.1744-7429.2006.00176.x Cooke MC. 1884. Notes on Hypocreaceae. Grevillea 12:77– 83. Cummings MP, Huskamp JC. 2005. Grid computing. Educause Rev 40:116–117. de Queiroz K. 2007. Species concepts and species delimit at i on. S ys t B i ol 56 : 87 9– 88 6, do i : 1 0. 10 80 / 10635150701701083 Gräfenhan T, Schroers HJ, Nirenberg HI, Seifert KA. 2011. An overview of the taxonomy, phylogeny and typification of nectriaceous fungi in Cosmospora, Acremonium, Fusarium, Stilbella and Volutella. Stud Mycol 68:79–113, doi:10.3114/sim.2011.68.04 Guindon S, Gascuel O. 2003. A simple, fast and accurate algorithm to estimate large phylogenies by maximum likelihood. Syst Biol 52: 696– 704, doi:10.1080/ 10635150390235520 Hirooka Y, Kobayashi T, Ono T, Rossman AY, Chaverri P. 2010. Verrucostoma, a new genus in the Bionectriaceae from the Bonin Islands, Japan. Mycologia 102:418–429, doi:10.3852/09-137 Katoh K, Toh H. 2008. Recent developments in the MAFFT multiple sequence alignment program. Brief Bioinform 9:286–298, doi:10.1093/bib/bbn013 Leigh JW, Susko E, Baumgartner M, Roger AJ. 2008. Testing congruence in phylogenomic analysis. Syst Biol 57:104– 115, doi:10.1080/10635150801910436 Lim GS, Balke M, Meier R. 2011. Determining species boundaries in a world full of rarity: singletons, species delimitation methods. Syst Biol 61:165–169, doi:10.1093/ sysbio/syr030 Luo J, Zhuang WY. 2008. Two new species of Cosmospora (Nectriaceae, Hypocreales) from China. Fungal Divers 31:83–93. ———, ———. 2010. Chaetopsinectria (Nectriaceae, Hypocreales), a new genus with Chaetopsina anamorphs. Mycologia 102:976–984, doi:10.3852/09-263 ———, ———. 2012. Volutellonectria (Ascomycota, Fungi), a new genus with Volutella anamorphs. Phytotaxa 44:1– 10. HERRERA ET AL.: PSEUDOCOSMOSPORA Maddison WP, Maddison DR. 2011. Mesquite 2.75: a modular system for evolutionary analysis. http:// mesquiteproject.org McNeill J, Barrie FF, Buck WR, Demoulin V, Greuter W, Hawksworth DL, Herendeen DS, Knapp S, Marhold K, Prado J, Prud’homme WF, van Reine GF, Smith J, Wiersema NJ, Turland, eds. 2012. International code of nomenclature for algae, fungi and plants (Melbourne Code). Adopted by the 18th International Botanical Congress, Melbourne, Australia. Jul 2011. Regnum Veg (in press). Moravec Z. 1954. Dialonectria cosmariospora v Československu. Ceska Mykol 8:92–95. Mostert L, Halleen F, Creaser M, Crous P. 2004. Cryptovalsa ampelina, a forgotten shoot and cane pathogen of grapevines. Australas Plant Pathol 33:295– 299. doi:10.1071/AP03095., doi:10.1071/AP03095 Nirenberg HI. 1976. Untersuchungen über die morphologische und biologische Differenzierung in der Fusarium-Section Liseola. Mitt Biol Bund Land-Forstwirt Berlin-Dahlem 169:1–117. Norvell L. 2011. Fungal nomenclature. 1. Melbourne approves a new code. Mycotaxon 116:481–490, doi:10.5248/116.481 O’Donnell K, Cigelnik E. 1997. Two divergent intragenomic rDNA ITS2 types within a monophyletic lineage of the fungus Fusarium are nonorthologous. Mol Phylogenet Evol 7:103–116, doi:10.1006/mpev.1996.0376 Posada D. 2008. jModeltest: Phylogenetic model averaging. Mol Biol Evol 25:1253–1256, doi:10.1093/molbev/ msn083 Rambaut A, Drummond AJ. 2007. Tracer 1.5. http://beast. bio.ed.ac.uk/Tracer Rayner RW. 1970. A mycological color chart. Kew, Surrey, UK: Commonwealth Mycological Institute. Ronquist F, Teslenko M, van der Mark P, Ayres DL, Darling A, Höhna S, Larget B, Liu L, Suchard MA, Huelsenbeck JP. 2012. MrBayes 3.2: efficient Bayesian phylogenetic inference and model choice across a large model space. Syst Biol 61:539–542, doi:10.1093/sysbio/sys029 Rossman AY. 1993. Holomorphic hypocrealean fungi: Nectria sensu stricto and teleomorphs of Fusarium. In: Reynolds DR, Taylor JW, eds. The fungal holomorph: mitotic, meiotic and pleomorphic speciation in fungal systematic. Wallingford, UK: CAB International. p 149– 160. ———, Samuels GJ, Rogerson CT, Lowen R. 1999. Genera of Bionectriaceae, Hypocreaceae and Nectriaceae (Hypocreales, Ascomycetes). Stud Mycol 49:1–248. Samuels GJ, Doi Y, Rogerson CT. 1990. Hypocreales. Mem New York Bot Garden 59:6–108. ———, Lu B-S, Chaverri P, Candoussau F, Fournier J, Rossman A. 2009. Cyanonectria, a new genus for ectria 1305 cyanostoma and its Fusarium namorph. Mycol Prog 8: 49–58, doi:10.1007/s11557-008-0577-x ———, Rossman AY, Lowen R, Rogerson CT. 1991. A synopsis of Nectria subgen. Dialonectria. Mycol Pap 164: 1–47. Schmitt I, Crespo A, Divakar PK, Fankhauser JD, HermanSackett E, Kalb K, Nelsen MP, Nelson NA, Rivas-Plata E, Shimp AD, Widhelm T, Lumbsch HT. 2009. New primers for promising single-copy genes in fungal phylogenetics and systematics. Persoonia 23:35–40, doi:10.3767/003158509X470602 Schoch CL, Crous PW, Wingfield MJ, Wingfield BD. 2000. Phylogeny of Calonectria and selected hypocrealean genera with cylindrical macroconidia. Stud Mycol 45: 45–62. Sung G-H, Hywel-Jones NL, Sung J-M, Luangsa-ard JJ, Shrestha B, Spatafora JW. 2007. Phylogenetic classification of Cordyceps and the clavicipitaceous fungi. Stud Mycol 57:5–59, doi:10.3114/sim.2007.57.01 Tang AMC, Jeewon R, Hyde KD. 2009. A re-evaluation of the evolutionary relationships within the Xylariaceae based on ribosomal and protein-coding gene sequences. Fungal Divers 34:127–155. Taylor JW, Jacobson DJ, Kroken S, Kasuga T, Geiser DM, Hibbett DS, Fisher MC. 2000. Phylogenetic species recognition and species concepts in fungi. Fungal Genet Biol 31:21–32, doi:10.1006/fgbi.2000.1228 Tsuneda A. 1982. Nectria episphaeria, a mycoparasite of Hypoxylon truncatum. Rep Tottori Mycol Inst 20:42–46. Vilgalys R, Hester M. 1990. Rapid genetic identification and mapping of enzymatically amplified ribosomal DNA from several Cryptococcus species. J Bacteriol 172:4238–4246. Vinuesa P. 2010. Multilocus sequence analysis and bacterial species phylogeny estimation. In: Oren A, Papke RT, eds. Molecular phylogeny of microorganisms. Norfolk, UK: Caister Academic Press. p 41–64. Weese J. 1916. Beiträge zur Kenntnis der Hypocreaceen. Sitzungsberichte der Kaiserlichen Akademie der Wissenschaften. Math-Naturwiss Kl 125:465–575. White TJ, Bruns T, Lee S, Taylor J. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis MA, Gelfand DH, Shinsky JJ, White TJ, eds. PCR protocols: a guide to methods and applications. San Diego, California: Academic Press. p 315–322. Zhang XM, Zhuang WY. 2006. Phylogeny of some genera in the Nectriaceae (Hypocreales,Ascomycetes) inferred from 28S nrDNA partial sequences. Mycosystema 25:15–22. Zwickl DJ. 2006. Genetic algorithm approaches for the phylogenetic analysis of large biological sequence datasets under the maximum likelihood criterion [doctoral dissertation]. Univ. Texas Austin Press. 125 p.