South African Journal of Botany 72 (2006) 534 – 543 www.elsevier.com/locate/sajb New and interesting records of southern African rust fungi (Uredinales) AR Wood ARC-Plant Protection Research Institute, Private Bag X5017, Stellenbosch 7599, South Africa Received 5 September 2005; accepted 7 February 2006 Abstract A number of rust fungi are recorded from southern Africa for the first time. Aecidium nairobianum is newly recorded from South Africa and Zimbabwe, and transferred to the genus Endophyllum as Endophyllum nairobianum comb. nov. Pucciniosira anthocleistae and Ravenelia ornata are also newly recorded from South Africa. Uredo abri is reduced to a synonym of R. ornata. Also, new details on the life cycle of several species have been elucidated from recent collections. The full life cycle of both Puccinia phyllocladiae and Uromyces kentaniensis are described. Aecidium capense is reduced to a synonym of the former and Aecidium antholyzae of the latter. Uromyces ventosa is reduced to a synonym of the microcyclic Uromyces bolusii. The taxonomic status of several species of rust fungi (Uredinales), recorded from southern Africa, requires changing to accommodate revised generic concepts. The following new combinations are made: Diorchidium gerstneri (Doidge) A.R. Wood comb. nov., Phakopsora nyasalandica (Cummins) A.R. Wood comb. nov., and Uredo doidgeae (Syd. and P. Syd.) A.R. Wood comb. nov. Schroeteriaster stratosus is confirmed as a synonym of Phakopsora stratosa. Melampsora junodii Doidge is reduced to a synonym of Phakopsora vernoniae Jørstad, and Uromyces paradoxus Syd. and P. Syd. is reduced to a synonym of Uredo balsamodendri Cooke. Details of the holotype of Puccinia estcourtensis Gjærum were omitted from the original description. These are supplied here, validating this species as Puccinia estcourtensis Gjærum ex A.R. Wood and Gjærum, sp. nov. © 2006 SAAB. Published by Elsevier B.V. All rights reserved. Keywords: Biodiversity; Mycoflora; South Africa; Rust fungi; Taxonomy 1. Introduction The taxonomy of the rust fungi indigenous to South Africa is relatively well known largely due to a series of publications by Ethel Mary Doidge (1927, 1928, 1939, 1941, 1948a,b). She produced the last comprehensive checklist of the rust fungi (Basidiomycota, Uredinales) occurring in southern Africa (Doidge, 1950) (including Angola, Botswana, Lesotho, Mozambique, Namibia, South Africa, Swaziland and Zimbabwe). Since then there has been little published on indigenous rust fungi from southern Africa, and there have been many changes in family and genus concepts of the rust fungi (Cummins and Hiratsuka, 1983, 2003). A checklist of rust fungi indigenous to southern Africa is in the process of being compiled, incorporating both published new records and taxonomic changes. There are, however, a number of unpublished new records for South Africa, as well as new details on the life cycle of several species, and a number are incorrectly assigned according to currently accepted generic concepts. This paper is one of several describing these new findings, in preparation for the checklist. E-mail address: WoodA@arc.agric.za. Melampsora junodii Doidge, Schroeteriaster stratosus (Cooke) P. Syd and Syd., S. doidgeae Syd. and P. Syd., Uromyces paradoxus Syd. and P. Syd., and Uropyxis gerstneri Doidge are all currently incorrectly placed. These species are reassigned to their correct generic placement. In addition, the identity of the host species of U. gerstneri is given here for the first time. Physopella nyasalandica Cummins, a later addition to the flora of southern Africa (Cummins, 1960) also needs to be assigned to the correct genus. Unfortunately, the details of the holotype of the proposed species Puccinia estcourtensis Gjærum were omitted from the original description (Gjærum, 1988). These details are therefore supplied here, validating this species. 2. Materials and methods Spores from herbarium specimens were mounted in aqueous lactic acid (50%) solution, and dimensions measured at 1000× magnification using a Zeiss Axioskop light microscope. Specimens were photographed using a Zeiss MC63 camera. Measurements of 25 spores from each specimen were taken. 0254-6299/$ - see front matter © 2006 SAAB. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.sajb.2006.02.002 A.R. Wood / South African Journal of Botany 72 (2006) 534–543 Measurements give the 95% confidence limits of the mean, with the minimum and maximum ranges given in parentheses. In all cases, the descriptions refer to specimens collected in South Africa that are listed below. Host plants were identified by comparison with named specimens in the Compton Herbarium, Kirstenbosch (NBG). Several of the species are well described elsewhere, the descriptions based on the same specimens that were examined for this work. No descriptions are given of these species but reference is made to the published descriptions and illustrations. 3. Results and discussion Diorchidium gerstneri (Doidge) A.R. Wood, comb. nov. (Figs. 1 and 2). Basionym—Uropyxis gerstneri Doidge, Bothalia 4:916 (1948). See Doidge (1948a) for a description and illustration. Specimen examined: SOUTH AFRICA: KwaZulu-Natal Province: False Bay, Zululand, on Uvaria lucida Benth. ssp. virens (N.E. Br.) Verdc., 27 June 1944, Gerstner 4817 (PREM 34564, HOLOTYPE). In the description, the host plant was simply listed as Annonaceae indet., the host identity and date of the collection were obtained from the plant specimen (PRE 421564) from which the type was originally removed (see specimens examined for details). Baxter (1959) excluded this species from Uropyxis and suggested it be transferred to Diorchidium, but did not make the transfer. Examination of the type confirms this diagnosis and the species is therefore transferred. The host plants of most species of Diorchidium are members of the Fabaceae. Apart from D. gerstneri, the only other species known on another host family is Diorchidium amapaensis J.F. Hennen and Sotão on Geophila trichogyne K.M. Schuman, Rubiaceae (Hennen et al., 1998). There are three other species of Diorchidium recorded from Africa, all on fabaceous hosts, namely, Diorchidium quadrifidum Cummins (from Ghana and Nigeria) (Cummins, 1960; Eboh, 1984), Diorchidium tetrasporum Cummins (from Zambia) (Cummins, 1960), and Diorchidium woodii Kalchbr. and Cooke (from South Africa) (Doidge, 1927). 535 Endophyllum nairobianum (Cummins) A.R. Wood comb. nov. Basionym—Aecidium nairobianum Cummins. Bull. Torrey Bot. Club 68:471 (1941). Infections systemic, causing witches' brooms. Pycnia absent. Aecidioid telia hypophyllous, scattered unevenly, not in lesions; aecidioid, orange, cylindrical, 175–300 μm diam; peridial margin white, reflexed, deeply incised. Peridial cells firmly joined together, irregular globose to ellipsoid, 25–33 × 17– 25 μm; outer wall striate, 2.5–3 μm thick; inner wall verrucose, 3–4 μm thick. Aecidioid teliospores orange, irregular angularglobose to ovate, (20–)22–24(–29) × (15–)19–20.5(–26) μm, length/width ratio 1:1–1.93 (mean 1:1.19); spore wall hyaline, evenly verruculose, 2–3 μm thick; upon germination producing 4 basidiospores. Specimens examined: SOUTH AFRICA. Gauteng Province: Horn's Nek, Pretoria, on Lippia javanicum (Burm. f.) Spreng., 6 Mar. 1939, AOD Mogg s.n. (PREM 30240); Witwatersrand National Botanic Garden, Johannesburg, on L. javanicum w (Burm. f.) Spreng., 3 Jan. 2004, AR Wood 536 (PREM 58347, K (M) 122452). Limpopo Province: Thabapaswa farm, 10 km WNW of Mokopane, on L. javanicum (Burm. f.) Spreng., 31 Aug. 2003, S Neser s.n. (PREM 58346); near Makapan Caves, on L. javanicum (Burm. f.) Spreng., 11 Nov. 1938, KM Putterill s.n. (PREM 32758). ZIMBABWE. Gazaland, on L. javanicum (Burm. f.) Spreng., 11 Sep. 1917, CFM Swynnerton s.n. (PREM 10714). Two species of Aecidium have been recorded on Lippia (Verbenaceae) in Africa, Aecidium evansii Henn. from Kenya (Nattrass, 1961), Malawi (Bisby and Wiehe, 1953), South Africa (Doidge, 1950), Sudan (Tarr, 1963), Tanzania (Henderson, 1972), Uganda (Wakefield and Hansford, 1949), and Zimbabwe (Doidge, 1950), and A. nairobianum from Kenya (Cummins, 1941; Nattrass, 1961). Cummins (1941) differentiated A. nairobianum from A. evansii because of the former's systemic habit. Otherwise, the two species are morphologically similar (Cummins, 1941). Witches' brooms caused by a rust fungus were recently collected on Lippia javanicum (Burm. f.) Spreng. in South Africa and identified as A. nairobianum due to Fig. 1. Teliospores of D. gerstneri. Scale bar = 20 μm. Fig. 2. Teliospore of D. gerstneri showing detail of ornamentation. Scale bar = 10 μm. 536 A.R. Wood / South African Journal of Botany 72 (2006) 534–543 their systemic habit. Fresh spores, morphologically resembling aeciospores, of A. nairobianum were germinated using the method of Wood (2004). After 24 h each germinated spore had produced four basidiospores from a metabasidium (Figs. 3–5), proving these spores are aecidioid teliospores. For this reason, E. nairobianum is transferred here to the endocyclic teleomorph genus Endophyllum Lév. To determine whether this species had been previously collected in southern Africa, 8 specimens accessioned under A. evansii in PREM were examined, and 3 were found to be E. nairobianum (see under specimens examined). This species is therefore newly recorded from South Africa and Zimbabwe, despite having been collected as early as 1917 in the region. The single specimen from Zimbabwe listed by Doidge (1950) as A. evansii was on examination actually E. nairobianum; therefore, at present, A. evansii is not confirmed as occurring in Zimbabwe though it is likely to occur there. It has been suggested that A. evansii may be the aecial stage of the pantropical, heteroecious, Puccinia versicolor Dietel and Holw. which is parasitic on many members of the Andropogoneae, Poaceae (Cummins, 1953). This is due to its morphological similarity to A. plectroniae Cooke. This latter species is known as the aecial stage of P. versicolor in India, due to crossinoculation experiments using Plectronia parviflora Bedd., Rubiaceae (Thirumalachar and Narasimhan, 1949), and Lantana indica Roxb., Verbenaceae (Patil and Thirumalachar, 1963). Phakopsora stratosa (Cooke) Arthur, Bull. Torrey Bot. Club 44:508 (1917). (Figs. 6 and 7). Basionym—Melampsora stratosa Cooke, Grevillea 10:128 (1882). S. stratosus (Cooke) P. Syd. and Syd., Monogr. Ured. 3:402 (1914). Bubakia stratosa (Cooke) Dietel, Die Naturl. Pflanzenf. 2 (6):48 (1928). Fig. 3. Germinated aecidioid teliospore of E. nairobianum showing 5-celled metabasidium with 4 immature basidiospores. Scale bar = 20 μm. Fig. 4. Germinated aecidioid teliospore of E. nairobianum showing germinated mature basidiospores. Scale same as in Fig. 3. Fig. 5. Germinated aecidioid teliospore of E. nairobianum showing germinated mature basidiospores. Scale same as in Fig. 3. Fig. 6. Vertical section through part of a telium of P. stratosa. Scale bar = 30 μm. Fig. 7. Vertical section through part of a uredinium of P. stratosa. Scale bar = 30 μm. This species is described and illustrated by Sydow and Sydow (1915) and Doidge (1927). Specimens examined: SOUTH AFRICA: KwaZulu–Natal Province: Inanda, on Croton sylvaticus Hochst., 12 Jan. 1881, Medley Wood 466 (PREM 10363, ISOTYPE). Limpopo Province: Entabeni, Louis Trichardt district, on Croton sylvaticus Hochst, 24 Aug. 1938, H.G. Bower s.n. (PREM 35148). The type species of Schroeteriaster, S. alpinus (Schroet.) P. Magnus, is a species of Uromyces, and Schroeteriaster is therefore a synonym of Uromyces (Mains, 1934; Cummins and Hiratsuka, 1983). Examination of the type specimen of S. stratosus revealed it to be a species of Phakopsora (Figs. 6 and 7), confirming Arthur's (1917) diagnosis. It is unknown why Doidge (1950) maintained its position in Schroeteriaster following the publication of Arthur (1917) and Mains (1934). This species is distinct from various species of Phakopsora which parasitise Croton species in the Americas (Sydow and Sydow, 1915). Phakopsora nyasalandica (Cummins) A.R. Wood, comb. nov. Basionym—Physopella nyasalandica Cummins, Bull. Torrey Bot. Club 87:37 (1960). See Cummins (1960) for a description, and Gjærum (1986) describes the urediniospores. Physopella has been established as an anamorphic genus in the Phakopsoraceae (Ono et al., 1992) and its teleomorph species have been reassigned to Phakopsora (Cummins and Hiratsuka, 1983, 2003). This rust species is recorded from Malawi (Cummins, 1960), Uganda (Gjærum, 1986) and Zimbabwe (Gjærum, 1995) on Helichrysum species (Asteraceae). Phakopsora vernoniae Jørstad, Arkiv för Botanik 3(17):567 (1956). A.R. Wood / South African Journal of Botany 72 (2006) 534–543 Melampsora junodii Doidge, Bothalia 2:158 (1927). This species is described by Jørstad (1956) and Gjærum (1986), and the uredinia by Doidge (1927). Specimens examined: MOZAMBIQUE: Rikatle, on Vernonia senegalensis Less., Sept. 1918, H.A. Junod s.n. (PREM 11723, HOLOTYPE of M. junodii). SOUTH AFRICA: Mpumalanga Province: God's Window, N. of Graskop, on Vernonia wollastonii S. Moore, 1 Jan. 2005, A.R. Wood 608 (PREM 58579). Western Cape Province: Wilderness National Park, Wilderness, 33°59'S 22°36'E, on Vernonia mespilifolia Less., 20 Feb. 2002, A.R. Wood 368 (PREM 58356, K (M) 122455). Two recent collections of uredinia on Vernonia spp. (Asteraceae) in South Africa had a morphology identical to that of the uredinial state of both P. vernoniae (Jørstad, 1956) (Fig. 8) and also of M. junodii (Doidge, 1927). The similarity between these species has already been noted by Jørstad (1956). The uredinia of Melamposora (Melampsoraceae) species are morphologically distinct from those of members of the Phakopsoraceae, including Phakopsora (Cummins and Hiratsuka, 2003; Ono et al., 1992; Buriticá and Hennen, 1994), and therefore, this brings into question the status of M. junodii. The type specimen of M. junodii was examined and only uredinia identical to those of the above specimens were found (Fig. 9). No telia were found; however, it was noted that the leaves had extensive insect feeding damage and that frequently the plant pallisade cells were melanised where damaged (Figs. 10 and 11). The ‘telia’ described by Doidge (1927) possibly refers to these melanised pallisade cells. The uredinial state of this rust species belongs to the phakopsoroid anamorph genus Malupa (Ono et al., 1992, Buriticá and Hennen, 1994; Cummins and Hiratsuka, 2003) and is distinct morphologically from that for Melampsora. 537 Vernonia mespilifolia and V. wollastonii are new host records for P. vernoniae. This species has also been recorded from elsewhere in Africa, including Madagascar (Jørstad, 1956), Kenya (Gjærum, 1986) and Uganda (Gjærum, 1986; Wakefield and Hansford, 1949). Telia have only been observed in the type specimen, from Madagascar, and in one other from Kenya (Gjærum, 1986). Wakefield and Hansford (1949) reported two specimens as M. junodii from Uganda. However, one of these, on Vernonia amygdalina Delile, later became the type of Uredo perscita Cummins (Cummins, 1945), the anamorph of Crossopsora perscita Cummins (Cummins, 1960). Urediniospores of C. perscita are larger than those of P. vernoniae (Cummins, 1945). Puccinia estcourtensis Gjærum ex A.R. Wood and Gjærum, sp. nov. The description, in Latin, (Gjærum, 1988; p. 149) and an illustration (Gjærum, 1988; p. 150) have already been published. Holotype here designated: SOUTH AFRICA: KwaZulu– Natal Province, Estcourt District, on Hyparrhenia quarrei Robyns, Feb. 1937, O. West 248 (K (M) 129498). Puccinia phyllocladiae Cooke, Grevillea 10:125 (1882). Anamorph—Aecidium capense Berk. and M.A. Curtis. Proc. Am. Acad. Arts Sci. 4:127 (1858). Pycnia in small groups on cladodes, no lesions though the immediate area may be thickened, amphigenous, usually only 1 but may have 2 or 3, honey coloured, globose, subepidermal, 130–200 × 160–200 μm, with ostiolar trichomes. Aecia associated with the pycnia, on cladodes usually 1 to 3 immediately surrounding a pycnium, but up to 6 when more than 1 pycnium present, aecia on fruit Fig. 8. Vertical section through a uredinium of P. vernoniae, from PREM 58356. Scale bar = 30 μm. Fig. 9. Vertical section through a uredinium of P. vernoniae, from PREM 11723. Scale bar = 30 μm. Fig. 10. P. vernoniae. Vertical section through leaf blade showing melanised host cells, from PREM 11723. Scale bar = 30 μm. Fig. 11. P. vernoniae. Vertical section through leaf blade showing melanised host cells and a uredinium on the abaxial leaf surface (arrow), from PREM 11723. Scale bar = 60 μm. 538 A.R. Wood / South African Journal of Botany 72 (2006) 534–543 numerous, aecidioid, bright orange, 140–300 μm wide, up to 1 mm long. Peridial cells hyaline, irregularly shaped, concave in cross-section, verrucose, 30–47 × 18–29 μm, inner wall 3– 5 μm thick, outer wall 2–5 μm thick. Aeciospores irregular to globose, orange-yellow, (32–)39–41(–48) × (26–)34–35 (–40) μm, length/width ratio 1:1.03–1.46 (mean 1:1.17); wall verruculose, irregularly thickened, 1–2 μm thick at thinnest and 2–3(–4) μm thick at thickest. Uredinia amphigenous on cladodes, supepidermal, erumpent, pulverulent, brown, in small spots, 1–1.5 mm diam. Urediniospores ovate, ellipsoid to obovate, light yellow, echinulate, (32–) 36–38.5(–43) × (24–)27–29(–33) μm, length/width ratio 1:1.07–1.64 (mean 1:1.34); wall 1–2 μm thick; 5–6 obscure equatorial germ pores. Telia amphigenous on cladodes and on stems, morphology as for uredinia, the teliospores replacing the urediniospores in the same sorus. Teliospores subglobose, rounded at both ends, not constricted at septum, (39–)46.5–48(–53) × (31–)37–38(–42) μm, length/width ratio 1:1.05–1.56 (mean 1:1.26); wall smooth, 2–4 μm thick, apex 5–9 μm thick; germ pore of upper cell to one side or apical, germ pore of lower cell below septum; pedicel persistent, up to 140 μm long, usually inserted obliquely. Specimens examined: SOUTH AFRICA. Western Cape Province: Simon's Bay [Simon's Town], on Asparagus sp., no date, C Wright 141 (K (M) 98611, TYPE) O, I (on fruit); Sedgefield, 34°01'S 22°47'E, on Asparagus aethiopicus L., 25 Sept. 1994, AR Wood 2 (PREM 58679) O, I (on cladodes), III; sin. loc. Oct. 1994, AR Wood 97 (PREM 58671, K (M) 103659), O, I (on fruit); Hermanus, 34°26'S 19°18'E, on A. aethiopicus L., 9 Dec. 1997, AR Wood 45 (PREM 58681) O, I (on cladodes), II, III; Cango Mountain Resort, near Cango Caves, Oudtshoorn, 33°24'S 22°16'E, on A. aethiopicus L., 19 Mar. 1998, AR Wood 47 (PREM 58682) III; Plettenberg Bay, 34°03'S 23°22'E, on Asparagus sp., 8 Oct. 1998, AR Wood 128 (PREM 58683) II, III; De Hoop Nature Reserve, E of Bredasdorp, 34°27'S 20°24'E, on A. aethiopicus L., 11 Dec. 1997, AR Wood 44 (PREM 58680) O, I (on cladodes), III; sin. loc. 18 Aug. 2001, AR Wood 350 (PREM 58684) O, I (on fruit and cladodes), III. Sydow and Sydow (1924) considered A. capense as a doubtful species, due to the poor state of the type specimen and the uncertain identity of the host plant. However, this aecial stage is fairly common on A. aethiopicus in the Western Cape province. The morphology of the host of the type specimen is consistent with this plant species, although it is fragmentary in Fig. 12. P. anthocleistae. Vertical section through a pycnium. Scale bar = 30 μm. Fig. 13. P. anthocleistae. Vertical section through an aecidioid telium. Scale bar = 50 μm. Fig. 14. P. anthocleistae. Catenulate two-celled teliospore chains with intercalary cells between each spore, on the right is the peridium. Scale bar = 25 μm. Fig. 15. P. anthocleistae. Peridial cells. Scale bar = 25 μm. A.R. Wood / South African Journal of Botany 72 (2006) 534–543 nature and thus difficult to corroborate. It was, on several occasions, found in close association with P. phyllocladiae and no other rust species that could be associated with this aecial stage was found in the vicinity. Aecia were found on both cladodes and fruit, the latter being the habit of A. capense. Therefore, A. capense is reduced to a synonym of P. phyllocladiae, with A. aethiopicus as a new host record for this rust species. Doidge (1927) only records teliospores. Thirumalachar (1947) first recorded urediniospores of P. phyllocladiae from India, these reported as being produced in abundance. Of the specimens examined, few urediniospores were produced on A. aethiopicus and only in young sori, they were more numerous on Asparagus sp. though still only in new sori and not in abundance. Pucciniosira anthocleistae (Henn.) Henn., Engl. Bot. Jahrb. 38:104 (1907) (Figs. 12–15). Basionym—Schizospora anthocleistae Henn., Engl. Bot. Jahrb. 34:41 (1905). Pycnia amphigenous on lesions that develop up to 12 mm diam. on the leaf blade, slightly caulicolous if they develop along veins, then up to 30 mm long; sub-epidermal, globose, 225–300 μm long, 140–210 μm wide. Telia amphigenous, surrounding the pycnia; orange, aecidioid, deeply immersed, peridium hardly emerging beyond leaf epidermis, opening 539 constricted. Peridial cells oblong, 40–62 × 15–30 μm; wall verrucose, 1–2 μm thick. Teliospores 2-celled, the cells easily breaking apart; catenulate, separated by intercalary cells; oblong, tapering at both ends, (45–)52–56(–66) × (17–)20.5– 22(–26) μm, length/width ratio 1:1.85–3.88 (mean 1:2.55); spore wall hyaline, verrucose, 2–3(–4) μm thick, the apices often slightly thickened, 3–4 μm thick. Specimen examined: SOUTH AFRICA. Limpopo Province: Tzaneen, on Anthocleista grandiflora Gilg., 16 Sep. 2003, S Neser s.n. (PREM 58374); Lowveld National Botanic Garden, Nelspruit, on Anthocleista grandiflora Gilg., 14 Sept. 2004, AR Wood 585 (PREM 58687). This is a new record for South Africa, but has been recorded elsewhere in Africa; from Gabon (Yen and Gilles, 1970), Ivory Coast (Viennot-Bourgin, 1953), Nigeria (Eboh, 1981), and Tanzania (Hennings, 1905). Ebbels and Allen (1979) listed this species as present in Tanzania, but incorrectly as P. mitragynes Diet., despite basing their reference on the record in Hennings (1905). Ravenelia ornata Syd. and P. Syd., Annal. Mycol. 4:437 (1906). (Figs. 16–19). Anamorph—Uredo abri Henn., Hedwigia 47:252 (1908). Pycnia on round pale yellow spots up to 6 mm diam.; subcuticular, hymenium flat, surrounded by paraphyses; 50– 75 μm diam. Uredinia subepidermal, erumpent, paraphysate; Fig. 16. R. ornata. Vertical section through a pycnium showing flat hymenium. Scale bar = 20 μm. Fig. 17. R. ornata. Vertical section through a uredinium showing surrounding paraphyses. Scale bar = 20 μm. Fig. 18. R. ornata. Teliospore showing marginal cells with capitate processes, and urediniospores. Scale bar = 20 μm. Fig. 19. R. ornata. Teliospores showing surface covered with large brown verrucae. Scale bar = 20 μm. 540 A.R. Wood / South African Journal of Botany 72 (2006) 534–543 primary uredinia surrounding pycnia; secondary uredinia solitary, not on leaf spots. Paraphyses marginal, free, usually clavate, 40–50 μm long, 7–8 μm wide below, 10–18 μm wide at apex; side walls hyaline, 1–2 μm thick; apex hyaline to light brown, wall 3–4 μm thick. Urediniospores brown on one side and hyaline to light brown on other, echinulate, spines small, 4– 5 scattered germpores; globose, (18–)19–20(–21) × (18–)18.5– 19(–20) μm, length/width ratio 1:1–1.11 (mean 1:1.04); wall 1 μm thick. Telia as for uredinia. Teliospores hemispherical, brown, 65–95 μm wide, 4–6 cells per diameter; the whole surface covered with large brown verrucae, 4–5 μm long, marginal cells with 1(–2) capitate processes, 13–22 μm long, apex 4–8 μm diam.; peripheral cells one-celled, central cells two-celled; cysts same number as peripheral cells, persistent. Specimens examined: INDIA. Goralehpur, on Abrus precatorius L., 1979, B Rai 289 (K (M) 121999). PHILIPPINES. Panay, Iloilo, on Abrus precatorius L., EB Copeland 87 (K (M) 122003) (TYPE of U. abri Henn.); Luzon, Tarlac Province, Gerona, on Abrus precatorius L., Jan. 1925, M Strong Clemens 6326 (K (M) 122000). SOUTH AFRICA. Mpumalanga Province: near Blydepoort resort, Blyde River canyon, on Abrus precatorius L., 25 Jan. 2001, AR Wood 284 (PREM 58379), O, II; Lowveld National Botanical Garden, Nelspruit, on Abrus precatorius L., 14 Sep. 2004, AR Wood 583 (PREM 58694), II, III. SRI LANKA. Peradeniya, on Abrus precatorius L., Feb. 1917, 5032 (K (M) 121998). This is a new record for South Africa, and the only other place in Africa where it has been reported is Nigeria (Eboh, 1984). No teliospores were found on the first specimen collected from South Africa, the urediniospores and paraphyses were identical to several specimens with teliospores housed at K (see specimens examined). Although the ornamentation of the urediniospores has been described as verruculose (e.g., Sydow and Sydow, 1915; Patil and Thirumalachar, 1970), they were observed to be spinescent. The pycnia observed are the same as those described by Patil and Thirumalachar (1970), which are type 7 (Hiratsuka and Hiratsuka, 1980) and typical of Ravenelia (Cummins and Hiratsuka, 2003). The later collected specimen with teliospores confirmed the identification. The type of U. abri Henn. was also examined and could not be distinguished from R. ornata. It very closely matched the South African specimens and another specimen from Sri Lanka (K (M) 121998). It is therefore reduced to a synonym of R. ornata. Uredo balsamodendri Cooke, Grevillea 19:6 (1890). (Figs. 20–22). Uromyces paradoxus Syd. and P. Syd., Ann. Mycol. 7:543 (1909). Doidge (1927) provides descriptions of both U. paradoxus and U. balsamodendri, the ‘teliospores’ in the former are in fact marginal paraphyses, and the paraphyses described in the latter occur intermixed between the urediniospores. These last mentioned structures may be urediniospore initials rather than paraphyses. Specimens examined: MOZAMBIQUE: Maputo, on Commiphora sp., 13 April 1909, Howard (PREM 623, ISOTYPE of U. paradoxus); Quelimane, on Commiphora sp., 7 Sept. 1913, I.B. Pole Evans (PREM 7386). SOUTH AFRICA: KwaZuluNatal Province: Durban, on Commiphora woodii Engl., June 1882, Medley Wood 689 (PREM 796, ISOTYPE of U. balsamodendri). In the description of U. paradoxus, doubt was expressed as to whether the ‘teliospores’ were in fact true teliospores or paraphyses (Sydow and Sydow, 1909). Cummins (1952) stated they were “undoubtedly paraphyses”, differing only from those of Nothoravenelia commiphorae Cummins in being shorter. Cross-sections through the uredinia of specimens of U. paradoxus confirmed Cummins' opinion that the ‘teliospores’ are paraphyses (Figs. 20 and 22). With only uredinia present, the type specimen of this species is therefore an anamorph, and can not be applied to a teleomorph name. Cross-sections through uredinia of the type specimen of U. balsamodendri show marginal paraphyses and urediniospores identical to the above, although few in number. As this is the first name applied to the anamorph of this species, U. paradoxus is reduced to a synonym of U. balsamodendri. U. balsamodendri is closely related to N. commiphorae, and may be conspecific. Because of the difference in dimensions of Fig. 20. Vertical section through a uredinium of U. balsamodendri. Scale bar = 40 μm. Fig. 21. Showing ornamentation of pedicellate urediniospores of U. balsamodendri. Scale same as Fig. 22. Fig. 22. Vertical section through part of a uredinium of U. balsamodendri showing marginal paraphyses originally described as ‘teliospores’. Scale bar = 20 μm. A.R. Wood / South African Journal of Botany 72 (2006) 534–543 541 Fig. 23. Vertical section through part of a uredinium of U. doidgeae. Scale bar = 20 μm. Fig. 24. Two “sessile” urediniospores within a uredinium of U. doidgeae. Scale bar = 20 μm. the paraphyses, U. balsamodendri is maintained as a distinct species until such time as additional specimens are acquired to determine the relationship between these two species. Uredo doidgeae (Syd. and P. Syd.) A.R. Wood, comb. nov. (Figs. 23 and 24). Basionym—Schroeteriaster doidgeae Syd. and P. Syd., Ann. Mycol. 24:26 (1926). Doidge (1927) provides a description of this species. Specimens examined: SOUTH AFRICA: Gauteng Province: Pyramid Koppies, Kaalplaats, Onderstepoort, Pretoria, on Croton gratissimus Burch., Jan. 1996, M. Mansel s.n. (PREM 54238). Limpopo Province: Waterberg, on Croton gratissimus Burch., 4 Aug. 1919, I.B. Pole Evans (PREM 14173, ISOTYPE). North West Province: Dikhololo resort, E. of Brits, on Croton gratissimus Burch., 22 Jan. 2002. AR Wood 362 (PREM 58580). Only uredinia are present on material examined, and therefore, the species is transferred to an anamorphic genus. This species is affiliated to P. stratosa, but awaits the discovery of its telial state. Urediniospores of this species differ from those of P. stratosa in that the echinulations are less robust and the wall is not thickened at the apex. This species appears to have an undifferentiated peridial layer placing it in the anamorphic genus Milesia (Ono et al., 1992; Buriticá and Hennen, 1994). As there is a lack of consensus in the use of characters used to define Milesia (Ono et al., 1992; Cummins and Hiratsuka, 2003), this species is placed in Uredo until such time as consensus is reached regarding the definition of anamorphic genera of the Uredinales. Uraecium magnisporum Cummins is closely related, differing only in the presence of pycnia (Cummins, 1952). Uromyces bolusii Massee, Kew Bull. 1901:168 (1901). (Fig. 25). Uromyces ventosa Syd., Ann. Mycol. 22:235 (1924). Pycnia, aecia and uredinia absent. Telia on leaves, subepidermal, erumpent, pulvinate, compact, light brown, up to 1 mm in diameter. Teliospores light orange-brown, variable in shape, elliptic to oblong and pyriform, apex rounded to attenuate, base rounded or sub-attenuate, (23)34–37(58) × (16) 21–23(38) μm, length/width ratio 1:1–3.41 (mean 1:1.63); wall smooth, 1.5–3(–4) μm thick; apex thickened, 3–18 μm, with distinct apical germ pore; pedicel persistent, up to 190 μm long, 5–10 μm wide, wall 1–2 μm thick; spores may germinate without a period of dormancy (lepto-form). Mycelium systemic, inducing witches' broom formation. Specimens examined: SOUTH AFRICA. Western Cape Province: Darling, 33°22'S 18°22'E, on Aspalathus acuminata Lam., 1 Jan. 1999, AR Wood 132 (PREM 57111); Stellenbosch, on Aspalathus crenata (L.) R. Dahlgren (=Borbonia parviflora Lam.), Aug. 1923, VA Duthie (P. A. van der Bijl Herb. 1246, held in PREM, TYPE of Uromyces ventosa); Stellenbosch, on Aspalathus crenata (L.) R. Dahlgren (=Borbonia parviflora Lam.), Oct. 1928, L Verwoerd 364 (PREM 46325); Helderberg Nature Reserve, Somerset West, 34°03'S 18°52'E, on Aspalathus cephalotes Thunb. ssp violaceae R. Dahlgren, 10 Dec. 2000, AR Wood 259 (PREM 57841); 2 km S of Simon's Town, Cape Peninsula, 34°13'S 18°23'E, on Asplathus ericifolia L., 12 Aug. 1999, AR Wood 152 (PREM 57112); Waylands Private Nature Reserve, SE of Darling, 33°24'S 18°25'E, on Aspalathus hispida Thunb., 2 Oct. 2000, AR Wood 241 (PREM 57838); Tygerberg Nature Reserve, Bellville, 33°52'S 18°35'E, on Aspalathus hispida Thunb., 12 Aug. 2001, AR Wood 340 (PREM 57842); Jonkershoek, Stellenbosch, 33°59'S 18°59'E, on Aspalathus hispida Thunb., 23 May 2002, AR Wood 373 Fig. 25. Teliospores of U. bolusii with long pedicels. Scale bar = 35 μm. 542 A.R. Wood / South African Journal of Botany 72 (2006) 534–543 (PREM 57844); Dikkop trail, Koeberg Nature Reserve, N of Melkbosstrand, 33°40'S 18°26'E, on Aspalathus ternata (Thunb.) Druce, 23 Nov. 2000, AR Wood 251 (PREM 57839); Muiskraal ridge, between Garcia Pass and Muiskraal, N of Riversdale, 33°55'S 21°12'E, on Aspalathus tuberculata Walp., 17 Feb. 2000, AR Wood 173 (PREM 57837); Helderberg Nature Reserve, Somerset West, 34°03'S 18°52'E, on Aspalathus uniflora L. ssp. wildenowiana (Benth.) R. Dahlgren, 10 Dec. 2000, AR Wood 258 (PREM 57840); Potberg Education Centre, De Hoop Nature Reserve, E of Bredasdorp, 34°23'S 20°32'E, on Aspalathus sp., 24 June 1998, AR Wood 51 (PREM 57110); Muiskraal ridge, between Garcia Pass and Muiskraal, N of Riversdale, 33°55'S 21°12'E, on Aspalathus sp., 17 Feb. 2000, AR Wood 172 (PREM 57836); farm Mooihoek, Robinson Pass between Mossel Bay and Oudtshoorn, 33°48'S 22°04'E, on Aspalathus sp., May 2002, AR Wood 369 (PREM 57843). Eastern Cape Province: Nuwekloof Pass, in the Baviaanskloof Wilderness Area, E of Uniondale, 33°29'S 23°38'E, on Aspalathus sp., Jan. 2000, AR Wood 578 (PREM 58383). Uromyces ventosa was originally described from Borbonia sp. (Fabaceae) and distinguished from U. bolusii (on Aspalathus, Fabaceae) because it occurs on a different host genus, and because of minor differences in spore dimensions between the two species (Doidge, 1927). Borbonia has, however, since been subsumed into Aspalathus (Dahlgren, 1963a), and examination of material has shown that these two fungal species can not be distinguished morphologically. Uromyces ventosa is therefore relegated to synonymy with U. bolusii. The host plant of the type specimen of U. ventosa is morphologically identical to A. crenata (L.) R. Dahlgren. Apart from A. crenata, all the host records above are new, namely, A. acuminata, A. cephalotes ssp violaceae, A. ericifolia, A. hispida, A. ternata, A. tuberculata, and A. uniflora ssp. wildenowiana. This rust fungus is only known from South Africa, as the host genus is almost completely restricted to the Fynbos biome (Dahlgren, 1963b). Uromyces kentaniensis Doidge. in Bothalia 2:30 (1927). Anamorph—Aecidium antholyzae Bubak. in Annal. Mycol. 9:144 (1911). Pycnia amphigenous, in small groups, flask shaped, 90– 130 μm wide. Aecia surrounding pycnia, amphigenous, on minute round spots up to 2 mm diam., orange, aecidioid, 200– 250 μm diam., margin of peridium slightly lacerate, not extending much beyond epidermis. Peridial cells rhomboid, irregular, 21– 33 × 13–26 μm; outer wall striate, 6–10 μm thick; inner wall verruculose, 3–5 μm thick. Aeciospores angular-globose, ovate or ellipsoid, orange, (20–)22–23(–25) × (19–)20–21(–23) μm, length/width ratio 1:1–1.3 (mean 1:1.12); wall hyaline, 1 ìm thick, closely verruculose, no refractile granules. Uredinia amphigenous, solitary or in small groups, on small spots bound by leaf veins, up to 2 mm long, sub-epidermal, erumpent, surrounded by torn epidermis. Urediniospores globose to ovate, orange, (22–)26–27(–35) × (18–)20–21(–25) μm, length/width ratio 1:1–1.7 (mean 1:1.27); wall hyaline, 1–1.5 μm thick; germ pores obscure, several, scattered. Telia amphigenous, minute, 60– 190 μm diam., dark brown, remaining covered by epidermis, in groups bound by leaf veins, up to 2 mm long, paraphyses filling space between telia in groups. Paraphyses light brown, palisadelike, laterally adherent, up to 75 μm long. Teliospores variable, ovate to broadly cuneate; rounded, truncate or conical at apex; attenuate or rarely rounded at base; light brown, darker at apex; (24–)30–32(–38) × (17–)20–22(–26) μm, length/width ratio 1:1–2.1 (mean 1:1.49); wall smooth, (1.5–)2–3 μm thick, apex thickened, 3–9 μm thick; pedicel persistent, up to 30 μm long. Specimens examined: SOUTH AFRICA. Eastern Cape Province: Oesterbaai, 34°09'S 24°39'E, 10 Oct. 1996, AR Wood 30 (PREM 57880) II, III; Western Cape Province: Koppie Alleen, De Hoop Nature reserve, 34°28'S 20°31'E, 1 Aug. 1995, AR Wood 17 (PREM 57878) II; Stellenbosch, 33°56'S 18°50'E, 1 Jul. 1996, AR Wood 28 (PREM 57879) O, I, II; sin. loc., 1 Jul 1997, AR Wood 34 (PREM 57881) O, I; sin. loc., 12 Sep. 1997, AR Wood 37 (PREM 57883) II, III; 4 km NW of Scarborough, 34°12'S 18°23'E, 7 July 1997, AR Wood 35 (PREM 57882) O, I; between Bellvidere and Brenton-on-Sea, Knysna, 34°03'S 23°00'E, 7 Jul. 2000, AR Wood 197 (PREM 57884) II; Victoria Bay, 34°00'S 22°32'E, 30 Aug. 2000, AR Wood 228 (PREM 57885); Hermanus, 34°25'S 19°14'E, 15 Nov. 2002, AR Wood 395 (PREM 58004) II, III. Doidge (1927) described the telia of U. kentaniensis, and later the uredinia (Doidge, 1948b) from Chasmanthe aethiopica. Although the earlier described Aecidium antholyzae was listed (Doidge, 1927, 1950) as occurring on the same host plant and from the same collection area (Kentani, Eastern Cape Province), no connection was suggested between these two names. Chasmanthe aethiopica is common in the Stellenbosch area, and is often infected by a rust fungus. Early in the wet season (winter rainfall, May to June) pycnia and aecia of A. antholyzae occur. From June onwards uredinia of U. kentaniensis appear on the same leaves, with both aecia and uredinia occurring together at least during June and July. Later (until September/ October) telia of U. kentaniensis develop as the leaves begin to senesce. The stages associated with both fungal epithets are consistently found together, and have been observed together on a number of occasions and places. No other telial stage has ever been found in the immediate vicinity to which the aecial stage could be connected. For these reasons, it is believed that A. antholyzae represents an anamorph stage of U. kentaniensis and is here reduced to a synonym. This rust occurs on C. aethiopica from the Cape Penninsula to the Eastern Cape (personal observations), the distribution of the host (De Vos, 1985). The related C. floribunda (Salisb.) N. E. Br., though also common in the Stellenbosch area and which sometimes grows within metres of infected C. aethiopica, has not been observed to become infected (personal observation). Acknowledgements The Compton Herbarium allowed access to their collection for identification of host plants. The South African National Collection of Fungi (PREM), especially Sarie Velthuysen, and the Royal Botanic Gardens (K), particularly Begoña AguirreHudson, provided valuable assistance. Pieter Winter of the A.R. Wood / South African Journal of Botany 72 (2006) 534–543 National Herbarium (PRE) is thanked for answering enquiries about the Annonaceae. References Arthur, J.C., 1917. Relationship of the genus Kuehneola. Bulletin of the Torrey Botanical Club 44, 501–511. Baxter, J.W., 1959. A monograph of the genus Uropyxis. Mycologia 51, 210–226. Bisby, G.R., Wiehe, P.O., 1953. The rusts of Nyasaland. Mycological Papers 54, 1–12. Buriticá, P., Hennen, J.F., 1994. Familia Phakopsoraceae (Uredinales) I. Géneros anamórficos y teliomórficos. Revista de la Academia Colombiana de Ciencas Exactas. Físicas y Naturales 19, 47–62. Cummins, G.B., 1941. Descriptions of tropical rusts—IV. Bulletin of the Torrey Botanical Club 68, 467–472. Cummins, G.B., 1945. Descriptions of tropical rusts—VII. Bulletin of the Torrey Botanical Club 72, 205–222. Cummins, G.B., 1952. Uredinales from various regions. Bulletin of the Torrey Botanical Club 79, 212–234. Cummins, G.B., 1953. The species of Puccinia parasitic on the Andropogoneae. Uredineana 4, 5–89. Cummins, G.B., 1960. Descriptions of tropical rusts—IX. Bulletin of the Torrey Botanical Club 87, 31–45. Cummins, G.B., Hiratsuka, Y., 1983. Illustrated Genera of Rust Fungi (Revised edn.). The American Phytopathological Society, St. Paul. ISBN: 0-89054058-6, p. 152. Cummins, G.B., Hiratsuka, Y., 2003. Illustrated Genera of Rust Fungi. (3rd edn.). The American Phytopathological Society, St. Paul. ISBN: 0-89054304-6, p. 225. Dahlgren, R., 1963a. The genus Borbonia L. incorporated in Aspalathus L. Botaniska Notiser 116, 185–192. Dahlgren, R., 1963b. Studies on Aspalathus: phytogeographical aspects. Botaniska Notiser 116, 431–472. De Vos, M.P., 1985. Revision of the South African genus Chasmanthe (Iridaceae). South African Journal of Botany 51, 253–261. Doidge, E.M., 1927. A preliminary study of the South African rust fungi. Bothalia 2, 1–228. Doidge, E.M., 1928. South African rust fungi II. Bothalia 2, 473–474. Doidge, E.M., 1939. South African rust fungi III. Bothalia 3, 487–512. Doidge, E.M., 1941. South African rust fungi IV. Bothalia 4, 229–236. Doidge, E.M., 1948a. South African rust fungi V. Bothalia 4, 895–918. Doidge, E.M., 1948b. South African rust fungi VI, the species of Uromyces on Iridaceae. Bothalia 4, 919–937. Doidge, E.M., 1950. The South African fungi and lichens to the end of 1945. Bothalia 5, 1–1094. Ebbels, D.L., Allen, D.J., 1979. A supplementary and annotated list of plant diseases, pathogens and associated fungi in Tanzania. Phytopathological Papers 22, 1–89. Eboh, D.O., 1981. A taxonomic survey of Nigerian rust fungi: Uredinales Nigerianensis—II. Mycologia 73, 445–453. Edited by SD Sym 543 Eboh, D.O., 1984. A taxonomic survey of Nigerian rust fungi: Uredinales Nigerianensis—III. Mycologia 76, 179–189. Gjærum, H.B., 1986. East African rusts (Uredinales), mainly from Uganda 5. On families belonging to Gamopetalae. Mycotaxon 27, 507–550. Gjærum, H.B., 1988. Rust fungi (Uredinales) on the genus Hyparrhenia (Poaceae). Mycotaxon 32, 143–160. Gjærum, H.B., 1995. Rust fungi from various countries. Lidia 3, 145–170. Hennen, J.F., Sotão, H.M.P., Hennen, M.M.W., 1998. The genus Diorchidium in the neotropics. Mycologia 90, 1079–1086. Hennings, P., 1905. Fungi Africae orientalis III. Botanische Jahrbücher für Systematik, Pflanzengeschichte und Pflanzengeographie 34, 39–57. Henderson, D.M., 1972. Rust fungi from East Africa: II. Notes from the Royal Botanic Garden Edinburgh 31, 441–446. Hiratsuka, Y., Hiratsuka, N., 1980. Morphology of spermagonia and taxonomy of rust fungi. Report of the Tottori Mycological Institute, Japan 18, 257–268. Jørstad, I., 1956. Reliquiae Lagerheimianae: African Uredinales. Arkiv för Botanik Ser. 2 B. 3, 563–598. Mains, E.B., 1934. The status of the genus Schroeteriaster. Annales Mycologici 32, 256–259. Nattrass, R.M., 1961. Host lists of Kenya fungi and bacteria. Mycological Papers 81, 1–46. Ono, Y., Buriticá, P., Hennen, J.F., 1992. Delimitation of Phakopsora, Physopella and Cerotelium and their species on Leguminosae. Mycological Research 96, 825–850. Patil, B.V., Thirumalachar, M.J., 1963. A new aecial host for Puccinia vesicolor. Current Science 8, 253. Patil, B.V., Thirumalachar, M.J., 1970. Cultural studies on some autoecious rust fungi in Maharashtra. Indian Phytopathology 23, 610–614. Sydow, H., Sydow, P., 1909. Einige neue resp. bemerkenswerte Pilze aus Südafrika. Annales Mycologici 7, 543–547. Sydow, P., Sydow, H., 1915. Monographia Uredinearum. Vol. 3 Pucciniaceae (excl. Puccinia et Uromyces)–Melampsoraceae–Zaghouaniaceae–Coleosporiaceae. Borntraeger, Leipzig, p. 726. Sydow, P., Sydow, H., 1924. Monographia Uredinearum. Vol. 4 Uredineae Imperfectae. Borntraeger, Leipzig, p. 671. Tarr, S.A.J., 1963. A supplementary list of Sudan fungi and plant diseases. Mycological Papers 85, 1–31. Thirumalachar, M.J., 1947. Some noteworthy rusts—II. Mycologia 39, 231–248. Thirumalachar, M.J., Narasimhan, M.J., 1949. Preliminary note on the heteroecism of Puccinia versicolor. Current Science 7, 252–253. Viennot-Bourgin, G., 1953. Urédinales d'Afrique (3e note). Urédinales de Côte d'Ivore (2e note). Uredineana 4, 125–228. Wakefield, E.M., Hansford, C.G., 1949. Contributions towards the fungus flora of Uganda—IX. The Uredinales of Uganda. Proceedings of the Linnean Society of London 161, 162–198. Wood, A.R., 2004. Endophyllum macowanianum, a new combination for Aecidium macowanianum (Uredinales–Pucciniaceae), and a note on E. macowanii. South African Journal of Botany 70, 667–670. Yen, J.-M., Gilles, G., 1970. Les Urédinales du Gabon–IV. Bulletin de la Societe de Mycologie de France 86, 851–863.