Mycological Society of America The Biology and Taxonomy of Rust Fungi Associated with the Neotropical Vine Mikania micrantha, a Major Invasive Weed in Asia Author(s): Harry C. Evans and Carol A. Ellison Source: Mycologia, Vol. 97, No. 4 (Jul. - Aug., 2005), pp. 935-947 Published by: Mycological Society of America Stable URL: http://www.jstor.org/stable/3762243 Accessed: 14/12/2009 07:50 Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=mysa. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Mycological Society of America is collaborating with JSTOR to digitize, preserve and extend access to Mycologia. http://www.jstor.org Mycologia, 97(4), 2005, pp. 935-947. ? 2005 by The Mycological Society of America, Lawrence, KS 66044-8897 The biology and taxonomy of rust fungi associated with the neotropical vine Mikania micrantha, a major invasive weed in Asia Asia during the early part of the 20th century (Choudhury 1972, Holmes 1982, Waterhouse 1994) and by the 1930s it was much used in Malaysia as ground cover in plantation crops such as oil palm and rubber (Thompson 1939, Parker 1972). Nevertheless by the 1960s M. micrantha had assumed the status of a problematic weed in plantation crops, particularly of tea in northeastern India and rubber and oil palm in southeastern Asia (Parker 1972, Holm et al 1977, Waterhouse 1994, Gogoi 2001). The weed has spread more recently to southwestern India, where it is invading both natural and planted forests, as well as subsistence and cash crops at the forest interface (Nair 1988, Evans 1998, Sankaran and Sreenivasan 2001), and also to China (Zhang et al 2004). Biological control, specifically the classical approach involving the introduction of co-evolved natural enemies from the region of origin of the exotic weed target, was suggested initially as a management strategy for M. micrantha by Parker (1972). Later Cock (1982) undertook surveys for insect natural enemies in the Neotropics. A thrips species (Liothrips mikaniae) from Trinidad was introduced subsequently into the Solomon Islands and Malaysia but this failed to become established (Cock et al 2001), and attention has now focused on evaluating the biocontrol potential of fungal natural enemies (Ellison 2004). Evans (1987), using literature and herbarium records, had compiled an inventory of the pathogenic fungi associated with weedy species of Mikania. This was followed by field surveys in southern Brazil (Barreto and Evans 1995), from which it was concluded that several rust species warrant further investigation as potential classical biocontrol agents for introduction into the Paleotropics where these coevolved rusts are absent. The present paper reports on the biology and taxonomy of the rust fungi encountered on M. micrantha during exploratory surveys in the Neotropics. Harry C. Evans1 Carol A. Ellison CABI Bioscience, Silwood Park, Ascot, BerkshireSL5 7TA, UK Abstract: Three microcyclic rust species were collected during surveys of the perennial asteraceous vine Mikania micrantha (Eupatorieae: Asteraceae) throughout its native range in the Neotropics but were absent in its invasive range in Asia. The commonest species, Puccinia spegazzinii with brown te- lioid telia, occurred wherever M. micrantha was found in South and Central America including the Caribbean island of Trinidad. Dietelia portoricensis, with occasional vestigial spermogonia and grayishwhite to pale yellow columnar aecioid telia, was collected only in Costa Rica; while D. mesoamericanasp. nov., apparently restricted to Mesoamerica, can be distinguished by its abundant, yellowish-orange, fertile spermogonia, yellow to pale brown telial columns, larger teliospores, and 4-spored rather than 2spored metabasidia. The fact that all three species share a fundamentally similar symptomatology suggests a common origin. Key words: applied mycology, classical biological control, Dietelia mesoamericana sp. nov., Dietelia portoricensis, mile-a-minute weed, Puccinia spegazzinii, systematics INTRODUCTION Mikania micrantha H.B.K. (Eupatorieae. Asteraceae) is a perennial, sprawling vine with a wide distribution in the Neotropics, which extends from Mexico to Argentina (Holmes 1982). Within this native range it is restricted mostly to riparian habitats, typically occurring around the margins of rivers, lakes and marshy terrain and is rarely weedy or invasive (Cock 1982, Barreto and Evans 1995). In sharp contrast, throughout its exotic palaeotropical range, M. micranthajustifiably has earned the common name of mile-a-minute weed (Holm et al 1977). The plant probably was introduced into a number of botanical gardens in MATERIALSAND METHODS Field collection.-Surveys were conducted from 1996-1999 throughout the native range of M. micrantha in tropical and subtropicalAmerica. Samples of rust-infectedplants were collected and initiallystored in a plant pressfor transport to the UK. Part of each samplewas kept as a voucher Accepted for publication 22 June 2005. 1 Corresponding author. E-mail: h.evans@cabi.org 935 936 MYCOLOGIA specimen for deposit in Herb. IMI and part for infection studies. However it was discovered that the rust species infecting M. micrantha lost viability during drying, which necessitated transporting living, rust-infected plants directly to the UK for the pathogenicity and specificity studies. The method described below gave the greatest success. Mikania micrantha readily roots at a node when in contact with soil and produces upright shoots or "plantlets." Newly infected plantlets (ca. 10 cm tall), typically with only chlorotic spots and pre-erumpent pustules, were selected within a rust-infected patch. The "runner" was cut ca. 2 cm from the plantlet, on both sides, and all extraneous leaves and soil were removed. The roots were wrapped in a damp tissue before being placed in a small plastic bag, which was tied loosely around the stem, and then enclosed within a bigger, self-sealing bag, which was inflated and sealed to form a protective vacuum or bubble around the leaves. Plants were able to survive at least 2 wk and the rust pustules matured during this period. When only mature pustules were available, both healthy and infected plantlets were mixed together and, invariably, the former became infected during transit to the UK. Greenhousestudies.-Plants used for rust-inoculation studies were propagated from cuttings and grown in a 1:1 mixture of general purpose, peat-based potting compost and John Innes No. 2 soil-based compost (J. Arthur Bower, Lincoln, UK). Pre- and postinoculated plants were maintained in an air-conditioned, quarantine greenhouse chamber set at 22 + 5 C, with a 12 h light/dark cycle (metal halide, full spectrum, light intensity range of 8000-13 000 lux). Test plants were grown on plant-support frames and pruned to stimulate fresh shoot production. The inoculation procedure was: shoot meristems were sprayed with a fine mist of distilled water, then mature, rust-infected material was suspended ca. 5 cm above the shoot apices using plant ties attached to the support frame. The plants were transferred to a dew chamber (Mercia Scientific, Birmingham, UK) set at 20 C, for 24 h and returned to quarantine for daily observation. TAXONOMY following rust species and Description of rusts.-The their variants were delimited during the study and the descriptions are based on fresh material obtained from greenhouse inoculations, as well as from dried field collections. Dietelia portoricensis (Whetzel & Olive) Buritica &J.F. Hennen, Flora Neotropica 24: 15. 1980. FIGS. 1-7; 8C, D; 9B, C = EndophylloidesportoricensisWhetzel & Olive, AmJ Bot 4:51. 1917. = Cronartium portoricensis (Whetzel & Olive) Sacc. & Trotter, Syll Fung 23: 851. 1925. =Aecidium expansum Arthur, Mycologia 7:317. 1915. Spermogonia rare, epiphyllous, subepidermal, type 4, effete, honey-colored to brown, globose, 110-150 [pm. Telia aecioid, subepidermal, amphigenous, on leaves, petioles and stems; sori 3-6 mm diam on leaves; sori on petioles and stems often coalescing and occurring in grossly swollen areas; comprising few to many, strongly erumpent, yellow aecioid cups producing grayish-white to pale yellow waxy columns or horns, up to 1-1.5 mm X 240-320 JLm, with an outer peridium of interlocking, rhomboidal, strongly x verrucose, thick-walled (4-6 iLm) cells, (25-)30-42 16-25 rm. Teliospores unicellular, hyaline to subhyaline, smooth, thin-walled, guttulate, globose to ovoid X (13-)15-19 (15-)17-23(-25) Vm; in long persistent chains; often, but not always, separated by prominent, hyaline, smooth, guttulate, ovoid to ellipsoidal intercalary cells, 5-9 X 10-16 [Lm. Metabasidia long, <200 Jim, usually producing only two sterigma-bearing cells separated by a densely guttulate, sterile cell. [Lm); basidSterigmata prominent, long (<60-100 to ovoid, iospores hyaline, smooth, subglobose x 8-12 fim. (12-)14-17 Material examined.-COSTA RICA. LIMONPROV.:Rio Madre de Dios, Siquirres, on Mikania micrantha, 17 May 1998, C.A. Ellison & R.H. Reeder(IMI 393068); same locality and host, 6 Sep 1998, H. C. Evans (IMI 393072). TRINIDAD AND TOBAGO. NORTHERNRANGE:Aripo, on Mikania guaco Humb. & Bonpl., 11 Jun 1945, H. Owen (IMI 6316). Commentary. At the Costa Rican locality M. micrantha was heavily attacked by Puccinia spegazzinii, often with telia from both species intermixed on the same leaf, which caused initial confusion over the identity and interrelationships of the rust species involved. This was clarified during infection studies in the UK. Chlorosis occurred ca. 12-13 d after inoculation and the aecioid telia appeared 8-10 d later (FI(;S. 1, 2). Spermogonia were observed only rarely and these always were effete (FIG. 3). As the telial columns matured under high humidity the peridial wall turned outward and upward exposing the teliospores that germinated to produce extremely long basidia with thickened, crozier-like tips with two, heavily pigmented cells typically separated by a smaller sterile cell (FIG. 6). Sterigmata develop only from the former cells and thus only two basidiospores are formed per basidium (FIG. 7). Sterigmata occasionally were observed to branch, and in such cases 3 or 4 basidiospores may be produced. Dietelia portoricensis has been recorded on various Mikania species, including M. micrantha, from Colombia (Kern et al 1933, Buritica and Pardo-Cardona 1996) and Trinidad (Arthur 1922a), as well as from Puerto Rico by Stevenson (1975), who described it as common in Central America, the West Indies and 937 EVANS AND ELLISON: RUST FUNGI FROM MIKANIA weed South America. This has not been our experience in the Neotropics, nor that of others, who also have failed to record it from Costa Rica (Arthur 1918), Cuba (Arthur andJohnston 1918, Urban 1990), Brazil (Barreto and Evans 1995), Barbados and the Windward Islands (Baker and Dale 1948), Jamaica (Dale 1955) and Mexico (Le6n-Gallegos and Cummins 1981). Examination of the specimen from Trinidad (IMI 6316) showed that teliospore morphology corresponds well with that of the Costa Rican material, however, the peridial cells are verruculose, rather than strongly verrucose, and smaller. In the original description by Olive and Whetzel (1917), who erected the monotypic genus Endophylloides Whetzel & Olive to accommodate this species, spermogonia were considered to be "not formed" and the peridial cells were said to be inconspicuous. Thirumalachar (1949) re-examined the material on M. cordifolia (L.f.) Willd. from Puerto Rico and his sections revealed that the peridial cells are persistent on the spore columns, although he described them as thin and evanescent, which is not the case for the Costa Rican material (FIGS.4, 7). He distinguished Endophylloidesfrom Endophyllum, not by the absence of an evident peridium-one of the generic characteristics used by Olive and Whetzel (1917)--"but by the occurrence of columnar telia." In Endophyllum the aecioid telia are pulverulent. Cummins (1959) also recognized the genus Endophylloides but compared it to the genus Dietelia, with which he suggested it even may be synonymous. Later Buritica and Hennen (1980) concluded that Endophylloidesshould be reduced to synonymy with Dietelia because the presence or absence of intercalary cells was neither a critical nor a consistent character. Thus Cummins and Hiratsuka (1983, 2003) subsequently amended the generic description of Dietelia: "with intercalary cells basally but often not seen." Romero and Carri6n (1998) reported D. portoricensis for the first time from Mexico. However, in their description, teliospore dimensions are given as 18-32 X 14-24 ,Im, whereas those of the type are much smaller, 15-20 X 12-15 IJm (Olive and Whetzel 1917). After further collections in Mexico in the same localities listed by Romero and Carri6n (1998), and subsequent comparison with the Costa Rican material of D. portoricensis, as well as with herbarium material from Guatemala deposited in Herb. IMI as D. portoricensis, it was concluded that both the Mexican and Guatemalan rusts are sufficiently distinct to justify separation at the species level. Dietelia mesoamericana H.C. Evans et C.A. Ellison, sp. nov. FIGS. 8A, B; 9A; 10-18 Spermogonia amphigenia,subepidermalia,typus4, insignitus, laxe gregariavel dense sparsa,mellea vel lutea; globosa ad subglobosa, (80-)120-180(-200) X (70-)100-160 (-180) jLm.Teliaaecioid, amphigenia,subepidermalia;in folia insidentibuset in greges (4-6 mm diam) plus minusve regularesdispositisvel sparsisin nervisfoliorum;in caulibus et petiolis greges incrassatasaepe valde elongata,matricum deformantia;cylindrica,cerinea vel fulva,in cornua vel columnae ceracea formentibus,ad 1.5-2 mm longa vel 160300(-450) ,umlata. Cellulaperidialiafirme conjunctae,hyalinis vel subhyalinus verruculosa vel verrucosa, parietibus 2-3.5 jLm,angulatae,ellipsoidiavel rhomboidia,(19-)28catenulatae,unicel38(-42) x (14-)17-25 im. Teliosporae lulatae, subhyalinae,laeve, tenuitunicatae,guttulatae,obovoideae vel ellipsoideae, 23-30 x 14-21 ,im; cellula intercalaria ellipsoideus, 5-9(-11) X 10-13 JLm.Basidiosporae obovoideae-ellipsoideae,9-12 X 6.5-9 jLm. HOLOTYPUS: MEXICO. VERACRUZ:Laguna Es- condida, Los Tuxtlas, ex Mikania micrantha, 14 Sep 1998, H.C. Evans, A. Romero & G. Carri6n (IMI 393070). PARATYPI: MEXICO. VERACRUZ:Laguna Escon- dida, Los Tuxtlas, ex Mikania micrantha, 3 Feb 1997, R. Segura & M. Martinez (IMI 393066); La Mancha, ex Mikania micrantha, 9 Sep 1998, H.C. Evans & A. Romero (IMI 393073). TAMAULIPAS:Nuevo Morelos, ex Mikania micrantha, 12 Sep 1998, H.C. Evans & A. Romero (IMI 393071). Etymology. Mesoamericana, pertaining to its Mesoamerican distribution. Spermogonia amphigenous, subepidermal, type 4, prominent, arranged in groups in swollen, chlorotic areas, or densely scattered on leaves, petioles and stems, yellowish-orange to honey-colored, globose to X (70-) 100-160 subglobose, (80-) 120-180(-200) (-180) txm. Spermatia abundant, clavate to ovoid, 46 X 2-2.5 jLm. Telia aecioid, amphigenous, subepidermal: on leaves in circular sori (4-6 mm) or densely scattered along midribs and veins; on petioles and stems densely grouped and extensive, often accompanied by hypertrophy of the tissues; cylindrical, as waxy, yellow to tawny brown columns or horns, up to 1.5-2.0 mm X 160-300(-450) jim, encased in a persistent peridium of interlocking cells. Peridial cellshyaline to subhyaline, verruculose to verrucose, thickwalled (2.0-3.5 jLm), ovoid to ellipsoidal to rhomX (14-) 17-25 jLm. Telioboidal, (19-)28-38(-42) spores catenulate, unicellular, subhyaline, smooth, thin-walled, guttulate, obovoid to ellipsoidal, 23-30 X 14-21 iLm;separated by persistent intercalary cells, ellipsoid, 5-9 (-11) X 10-13 jLm.Metabasidia hyaline, smooth, 120-150 X 7-8 jim, densely staining in upper third and 4-celled, each cell producing a prominent often long (up to 60 jLm)sterigma. Basidiospores hyaline, smooth, obovoid to ellipsoid, 9-12 X 6.5-9 Pxm. MYCOLOGIA 938 0 U 4"~ ~~~ t' . \ EVANSAND ELLISON:RUST FUNGIFROMMIKANIAweed Other material examined as Dietelia portoricensis, on Mikania sp.: GUATEMALA.ZARAPA DEPT.:Los Amates, 15 Mar 1905, WA. Kellermanex Herb. Arthur49476 (IMI55440). MEXICO. VERACRUZ: Catemaco, Playa Vicente, 22 Oct 1996, M.K. Seier (IMI 393079). Commentary.Chlorosis typically appeared 7-8 d after inoculation when prominent, swollen lesions formed on both the upper and lower leaf surfaces, as well as on the petioles. Clusters of orange spermogonia developed 15-19 d after inoculation, which imparted a shiny or glistening appearance due to the formation of spermatial droplets (FIG. 10). In contrast to D. portoricensis, however, the subsequent appearance of the telial initials was delayed and the yellow spore columns took <40 d to reach maturity (FIGS. 11-13). In the field in Mexico spermogonial pustules, particularly those on the petioles and stems, frequently were effete and heavily parasitized by a species of Tuberculina (FIG. 19). This initially caused severe problems in trying to establish pure cultures of the rust in the greenhouse in the UK. Only by regular and judicious brushing of the newly emerging parasitized pustules with 70% alcohol, to remove the powdery clumps of reddish-brown to purple conidia (FIG.20), followed by repeated inoculations using healthy telia, were clean rust cultures eventually obtained. A further six rust species have been reported on New World Mikania spp., five of which have twocelled telioid telia and belong to Puccinia (Barreto and Evans 1995), while the other, Uromycesmikaniae Viegas, has unicellular telioid teliospores and an ae- cidial stage (Viegas 1945). Of the six Dietelia spp. described from the Neotropics (Buritica and Hennen 1980) only two species, D. aequatoriensis (Syd.) Bur- itica & J.F. Hennen and D. portoricensis, have telial columns. The former can be distinguished by its smaller, verrucose teliospores. D. mesoamericana can be distinguished readily from D. portoricensis by the consistent presence of fertile spermogonia and the significantly longer, obovoid to ellipsoidal teliospores. In addition the telial columns or horns of D. mesoamericanahave a distinctly yellow to pale brown appearance in fresh material contrasting with the grayish-white columns of D. portoricensis. Separation of the two species is made easier, of course, by working with living cultures. For example the prominent 939 spermogonia of D. mesomericana, densely covering the inoculated leaves, are particularly distinctive and diagnostic. Moreover the germination process can be followed and this has allowed for an additional separation character: the metabasidia of D. portoricensis being predominately 2-spored while those of D. mesoamericana are consistently 4-spored. Differences in peridial cell morphology between the Costa Rican material of D. portoricensis and the holotype of D. mesoamericana(IMI 393070) appeared initially to be significant. However the peridial cells of the Guatemalan specimen (IMI 55440), originally assigned to D. portoricensis (Herb. Arthur), are thickwalled and verrucose, resembling those of D. portoricensis (ex Costa Rica) (FIGS.7, 8D), while the peridial cells of the Trinidadian material (IMI 6316) are thinner-walled and verruculose, similar to the D. mesoamericana holotype (FIG. 8B). However one of the Mexican paratypes of D. mesoamericana(IMI 393071) had an intermediate form. It is concluded therefore that the morphology of the peridial cells is too variable to be a reliable character, and this may be influenced by the position of the cells within the spore column or the condition of the specimen. In fresh greenhouse material telial columns are prominent, <2 mm in length (FIGS. 12-16); while in field-collected herbarium specimens, these columns are much reduced or degraded due to weathering and/ or damage in the plant press. Observations of rust cultures suggest that the telial columns of D. portoricensisare more prone to sloughing from the infected plant than those of D. mesoamericana. Indeed microscopic examination (FIGS. 4, 15, 16) reveals that the peridium of the latter species is a much more coherent structure and is less prone to fracture. In addition detached telial columns of D. portoricensis placed directly on the leaf surface have been observed to produce modified sterigmata or infective hyphae rather than basidiospores that lead to normal infection. The basidiospores of D. mesoamericana are released readily from the intact telia, resulting in heavy infection in greenhouse inoculations, whereas those of D. portoricensis frequently are trapped in the peridium, perhaps due to their large size (FIGS.7, 18), and, as a consequence, the resulting infection is less severe. Further differences between the two species relate FIGS.1-7. Macro-and micromorphology of Dietelia portoricensis, IMI 393068 ex Costa Rica. 1, 2. Lower leaf surface and petiole of Mikania micrantha, with grayish-white, aecioid telial columns, 3 wk after inoculation. 3. Cross section of effete spermogonia. 4. Cross section of telial column, with peridial wall. 5. Chain of unicellular teliospores separated by intercalary cells. 6. Metabasidium with two sterigmata separated by prominent sterile cell (arrow). 7. Mature metabasidium producing two basidiospores only, note thick-walled, verrucose, peridial cells (arrows). Bars: FIGS.3, 4=60 iLm;5-7=10 inm. 940 MYCOLO(IA EVANS AND ELLISON: RUST FUNGI FROM MIKANIA weed 941 B A Fig. 9. Teliospores germination of Dietelia spp. A. D. mesoamericana(IMI 393070), shoNwingdevelopment of four sterigmata. B. D. porto7icensis(IMI 393068), shoNwingtwo sterigmata separated by heavily-pigmented sterile cell. C. D. portoricensis,more advanced stage of teliospore germination showing basidiospore formation. Bars: F(;s. A, B = 20 pLm;C = 10 pLm. to their biology, and specifically the symptomatology observed during the greenhouse inoculations. Thus D. portoricensishas a relatively long biotrophic phase (ca. 2 wk) before the first symptoms of chlorosis appear, after which development of the telia is rapid. In D. mesoamericana,however, chlorotic pustules are visible within 1 wk but this is followed by a long maturation period (ca. 5-6 wk in total) during which fertile spermogonia and then the telial columns de- velop. This is interpreted as evidence of the functional role of the spermogonia within the life-cycle of D. mesoamericana, which is clearly vestigial in D. portoricensis. We were intrigued that the 7uberculina mycoparasite was recorded only on D. mesoamericana and proved to be common in the field and troublesome in the greenhouse. It was never found on D. portoricensis, or on P spegazzinii, and failed to become es- Fli(. 8. Comparative morphology of aecioid telial stage of Dietelia spp. A, B. D. mesoameircana (IMI 393070), showing obovoid-ellipsoidal teliospores (A) and small, mainly ovoid-rhomboidal, verruculose, peridial cells (B). C, D. D. portoricensis (IMI 393068), showing globose-ovoid teliospores (C) and rhomboidal, verrucose, peridial cells (D). Bar = 10 Jim. 942 MYCOLOGIA - y--J? j. ,;i I C 1- * -t * 40, .: I ie, ?*j?."'' ?-? !*` aI 9 A ?? r 0 '' 943 EVANS AND ELLISON: RUST FUNGI FROM MIKANIA weed tablished on either of these two rusts in the greenhouse, despite proximity to heavily-parasitized D. mesoamericana. Lutz et al (2004) have shown that Tuberculina spp. cluster as a sister group of Helicobasidium within the Urediniomycetes and have a unique cellular interaction with the haploid stages of rusts. This association was posited to have a sexual basis, having evolved from a modified mating interaction (Bauer et al 2004). This suggests therefore that D. mesoamericanacolonizes its host in the haploid phase, hence the need for functional spermogonia; while the other two species colonize in the diploid phase, and either totally lack spermogonia (P. spegazzinii), or if present these are effete (D. portoricensis). Without this fundamental knowledge of the biology, Buritica and Hennen (1980) expanded the species concept of D. portoricensis,as envisaged by Olive and Whetzel (1917) and probably based on the Guatemalan material, to include the presence of fertile spermogonia and teliospores significantly larger than those originally described. Faced with this amended species description, Romero and Carri6n (1998) had no option but to include their Mexican collections within D. portoricensis. Puccinia spegazzinii De Toni in Saccardo, Syll. Fung. 7:704. 1888. FIGS. 21-23 Teliospores2-celled, occasionally3-celled, linear-oblong to clavate, (35-)40-65(-72) X (10-) 12-18 pxm,obtuse or rounded abovenarrowbelow,slightlyconstrictedat septum; wall pale yellow to brown, smooth, thin, thickened at apex (2-3 [pm);pedicel colorless,as long as or longer than spore, 8-10 mmdiam, frequentlyswollen (<14 !zm) below spore. Metabasidiumhyaline,60-70 X 5-6 Vpm, becoming swollen (10-12 Rm), densely staining and guttulatein upper half; 4-celled, each cell producing a fertile sterigma. Basidiosporeshyaline,smooth, obovoid-ellipsoidal,9-13(-15) X 58(-9.5) pLm. Telial morphology is highly variable among the strains examined, and the golden-yellow to darkbrown teliospores are embedded in the host tissues within distinctly raised cushion-like sori, 4-7 mm diam. These are produced mainly on the lower leaf surface but also occur on the petioles and stems, where they typically coalesce, forming elongated pustules, often accompanied by gross hypertrophy of the underlying tissues (FIG. 21). Such perennial telial cushions may be hyperparasitized in the field, particularly by species of Eudarluca, the ascostromata of which may replace the rust tissues (FIGS. 22, 23). Materialon M. micrantha,host confirmed by molecular DEPT.: analysis (Ellison et al 2004). ARGENTINA. MISIONES Rio Iguassu,13 Jun 2000, H.C. Evans (IMI393078). BRASan Miguel do Anta, 15 Nov 1996, C.A. ZIL. MINASGERAIS: Ellison &J. Mauricio (IMI 393060); Marip6, 30 Nov 1996, C.A. Ellison & H.C. Evans (IMI 393061). COLOMBIA. CALDASDEPT.: Chinchina, 7 Jul 1996, PS. Baker (IMI 393058). COSTA RICA. LIMONPROV.:Rio Madre de Dios, Siquirres, 17 May 1998, C.A. Ellison & R.H. Reeder (IMI 393069). ECUADOR. NAPOPROV.:Rio Pucufio, Loreto, 28 Mar 1999, H.C. Evans (IMI 393075). IMBABURA PROV.:Rio Mira, Collapi, 22 Mar 1999, H.C. Evans (IMI 393074). NICARAGUA. MATAGALPADEPT.: Selva Negra, Jinotega, 29 Jun 2002, H.C. Evans & R.W. Barreto (IMI 393077). PERU. Cuzco DEPT.:Pilcopata, 18 Nov 2001, H.C. Evans & R.H. Reeder (IMI 393076). SAN MARTINDEPT.: Tabolosos, 30 Jan 1997, H.C. Evans, (IMI 393065). TRINIDAD & TOBAGO. Parrylands, La Brea, Nov-Dec 1996 (composite sample), H.C. Evans & M. Morais (IMI 393062-4); St Patrick, Jul 1997, M.J.W. Cock (IMI 393067); Arima, Blanchisseuse, Jul 1996, M.J.W. Cock (IMI 393057). Additional material examined, host identity not confirmed by DNA sequence analysis. The host is given according to the voucher specimen, however, because M. scandens is confined to North America (King and Robinson 1987), it is more than probable that the true host is M. micrantha: BRAZIL. PARA:Santa Isabel, on M. micrantha, 6 Nov 1963, FC. Albuquerque (IMI 135722). Rio DE JANEIRO:Barra de SaoJoao, on M. micrantha, 19 Aug 1989, R. W Barreto (IMI 345385). COSTA RICA, San Jose, on M. scandens, 23 Dec 1924, H. Sydow (IMI 55443); Oratina, 30 Dec 1915, on M. micrantha, collector unknown (IMI 55444). GRENADA (W. Indies). on M. micrantha, 6 Feb 1946, R.E.D. Baker (IMI 55441). PANAMA. Bocas del Toro, on M. micrantha, 2 Feb 1921, M.A. Carleton (IMI 55442). ST. VINCENT (W. Indies). Dorset Hill, on M. micrantha, 26 Dec 1971, C. Critchet (IMI 163360). TRINIDAD & TOBAGO. NORTHERNRANGE: Siparia, on M. scandens, 3 Oct 1947, W.T.Dale (IMI 32155). FIGS.10-18. Macro- and micromorphology of Dietelia mesoamericana,IMI 393070 ex Mexico. 10. Prominent, yellow, fertile spermogonia developing on upper leaf surface of Mikania micrantha, 18 d after inoculation; inset of spermogonial group with spermatial droplets developing within raised, chlorotic pustules. 11. Cross section of erumpent, fertile spermogonium, showing trichogynes and spermatia. 12. Young telial columns developing along swollen petiole 25 d after inoculation. 13, 14. Mature, yellow telial columns formed on chlorotic, swollen pustules on leaves and petioles 35-40 d after inoculation that later become necrotic (FIG. 14). 15, 16. Cross section of telial columns showing early stage with prominent, basal meristem (arrow) and compact peridial wall (FIG. 15); and mature stage (Fig. 16), 40 d after inoculation. Germinating teliospores showing early metabasidial stage. 18. Maturing metabsidium with four sterigmata; inset of basidiospores, compare FIG. 7. Bars: 11 = 30 ,um; 15, 16 = 115 pLm;17, 18 = 10 pm. 944 MYCOLOGIA _4i-- - _-*a_ , t ^~~5 . .f *<?Ll?? r~~~~~~~~~~~~~?*c v "$.~ t.- '- t . < E _, I. ;??? .s z '?"ill' ?I '" f ;j:?? ; ._'i 8;,?. .*'':? ?i -? W' e. EVANS AND ELLISON: RUST FUNGI FROM MIKANIA weed Commentary. This species first was described on Mikania scandens var. periplocifolia ( = M. periplocifolia Hook. & Arn.; King and Robinson 1987) from Argentina, and also on M. cordifolia, and has a number of synonyms (Barreto and Evans 1995). Arthur (1922b) described P. spegazzinii on several Mikania species from North America, including M. cordifolia, although it was never observed on this host during the present surveys, when it was recorded only on M. micrantha in Argentina, Brazil, Costa Rica, Ecuador, Nicaragua, Peru and Trinidad. Puccinia spegazzinii is a microcyclic rust, producing only telioid teliospores and basidiospores. Spermogonia, aecia and uredinia are unknown in the field (Arthur 1922b, Le6n-Gallegos and Cummins 1981, Barreto and Evans 1995), and greenhouse inoculations have confirmed this reduced life-cycle. DISCUSSION Without authoritative data on both the taxonomy and biology of a potential biocontrol agent, permission to introduce and release this agent will not normally be granted by the relevant quarantine body in the importing country (FAO 1996). Thus, taxonomy is an essential component of any classical biological control program, and this is very much the case with the rusts associated with Mikania micrantha that are described here. This paper reports on three, morphologically distinct rust species occurring on M. micrantha in the Neotropics. However, Viegas (1945) recorded a further species associated with this host in the state of Paraiba, north-east Brazil. This species, Puccinia mikania-micranthaeViegas, is apparently also microcyclic with only teliospores, described as predominately unicellular, rarely two-celled, which would seem to fit the generic concept of Uromycesrather than Puccinia. His illustrations clearly depict a unique species, which was not encountered during the present study, despite the intensive surveys mainly in central and southern Brazil. It may be that P. mikania-micranthae has a distribution limited to north-east Brazil. Puccinia spegazzinii was subsequently selected for host range and pathogenicity screening, which took 3 yr to complete, and these results will be reported FIGS. 19-23. Fungal hyperparasites 945 elsewhere. A fully-tested pathotype of this rust from Trinidad is being held in quarantine in India, prior to release in the Western Ghats (south-west India) and Assam (north-east India); and a pathotype from Argentina, now in quarantine in Beijing, is close to being released in Guangdong Province, Southern China. However, recent screening of the new species Dietelia mesoamericana,has shown that this rust fungus may be the preferred agent for some of the biotypes of M. micrantha currently on an invasive front in both Asia and Australasia. Whether the three species delimited here are more closely related than their morphology suggests is currently being investigated using molecular techniques in order to test the theory, based on their biology, that all three share a common origin or ancestry. These microcyclic taxa may have speciated from a full-cycled Puccinia, with a Central American or Caribbean evolutionary center, as they radiated into the different geographic zones: P. spegazzinii dominating in South America, while D. portoricensis and D. mesoamericana evolved in the Central American-Caribbean and Mesoamerican regions, respectively. Cummins and Hiratsuka (2003) used the term endocyclic for life-cycles such as those described here for Dietelia portoricensisand D. mesoamericana,which are derived from aecial or uredinial states. However, since spores of both germinate to produce basidiospores they are teleomorphs and not anamorphs. Logically, therefore, parental generic names should be used but, without molecular evidence and based purely on morphology, this would be speculative. Buritica and Hennen (1980) elected to place Dietelia in the tribe Pucciniosireae of the family Pucciniaceae. Later, however, Buritica and Hennen (1994) considered grouping all endocyclic genera in a separate family, Endophyllaceae. More recently, Cummins and Hiratsuka (2003) have elected to place endocyclic genera close to the families of suspected parental genera or groups of genera. Thus, Dietelia in their system is now classified in the family Pucciniosiraceae and not Pucciniaceae. We believe, however, that molecular data will show that both Dietelia portoricensis and D. mesoamericanaare closely related to P spegazzinii. of rusts associated with Mikania micrantha. 19, 20. Tuberculina sp. hyperparasitic on, and replacingtelial sori of Dieteliamesoamericana;in field situationin Mexico (arrows)(FIG.19), and in greenhousesituation in UK (FIG.20). 21-23. Pucciniaspegazzinii in greenhouse and field situation,showingnon-parasitizedtelial sori developing on lower leaf surface and in swollen pustuleson petiole, 12 daysafter inoculationwith IMI393976 ex Pilcopata,Peru (FIG. 21); and telia in field ex Tabolosos,Peru (IMI393065), heavilyhyperparasitizedby Eudarlucasp. (FIG.22). Crosssection of rusted stem showingascostromataof Eudarlucasp. completelyreplacingtelia. Bar = 60 pm. MYCOLOGIA 946 ACKNOWLEDGMENTS The assistance of Robert Barreto, Matthew Cock, Gloria Carri6n, Angel Romero and Ricardo Segura in obtaining living material is much appreciated. This work was supported by the UK Department for International Development (DFID) within the Crop Pest Research Program (R6735), and under DEFRA licence No. PHL 182/4869. LITERATURECITED Arthur JC. 1918. Uredinales of Costa Rica based on collections by E.W.D. Holway. Mycologia 10:111-154. .1922a. Uredinales collected by Fred J. Seaver in Trinidad. Mycologia 14:12-24. 1922b. North American Flora 7:574. ,Johnston JR. 1918. Uredinales of Cuba. Mem Torrey Bot Club 17:97-175. Baker RED, Dale WT. 1948. Fungi of Barbados and the Windward Islands. Mycol Pap 25:1-26. Barreto RW, Evans HC. 1995. The mycobiota of the weed Mikania micrantha in southern Brazil with particular reference to fungal pathogens for biological control. Mycol Res 99:343-352. Bauer R, Lutz M, Oberwinkler F. 2004. Tuberculinarusts: a unique basidiomycetous interfungal cellular interaction with horizontal nuclear transfer. Mycologia 96: 960-967. Buritica P, Hennen JF. 1980. Pucciniosireae (Uredinales, Pucciniaceae). Flora Neotropica Monogr 24:1-50. . 1994. Familia Phakopsoraceae (Uredinales) , I. Generos anam6rficos y teliom6rficos. Rev Acad Colombiana Cienc Exactas, Fisicas y Nat 19:47-62. , Pardo-Cardona VM. 1996. Flora Uredineana Colombiana. Rev Acad Colombiana Cienc Exactas, Fisicas y Nat 20:183-236. Choudhury AK. 1972. Controversial Mikania (climber)-a threat to the forests and agriculture. Ind Forest 98:176186. Cock MJW.1982. Potential biological control agents for Mikania micrantha HBK from the neotropical region. Tropic Pest Man 28:242-254. , Ellison CA, Evans HC, Ooi PAC. 2001. Can failure be turned into success for biological control of mile-aminute weed (Mikania micrantha)? In: Spencer PM, ed. Proceedings of the 10th International Symposium on Biological Control of Weeds. Bozeman, USA: Montana State University Press. p 155-167. Cummins GB. 1959. Illustrated genera of fungi. Minneapolis, USA: Burgess. 150 p. , Hiratsuka Y. 1983. Illustrated genera of rust fungi. Revised ed. St Paul, Minnesota, USA: APS Press. 153 p. . 2003. Illustrated genera of rust fungi. 3rd ed. St Paul, Minnesota, USA: APS Press. 225 p. Dale WT. 1955. A preliminary list of Jamaican Uredinales. Mycol Pap 60:1-21. Ellison CA. 2004. Biological control of weeds using fungal natural enemies: a new technology for weed management in tea? IntJ Tea Sci 3:4-20. , Evans HC, Ineson J. 2004. The significance of in- traspecies pathogenicity in the selection of a rust pathotype for the classical biological control of Mikania micrantha (mile-a-minute weed) in Southeast Asia. In: CullenJM, Briese DT, Kriticos DJ, Lonsdale WM, Morin L, Scott JK eds. Proceedings of the 11th International Symposium on Biological Control of Weeds. Canberra, Australia: CSIRO Entomology. p 102-107. Evans HC. 1987. Fungal pathogens of some subtropical and tropical weeds and the possibilities for biological control. Biocon News Info 8:7-30. . 1998. Major Indian weeds of neotropical origin and the possibilities for collaborative biocontrol projects. In: Ferrar R, Muniappan P,JayanthJP, eds. Proceedings of the 4th International Workshop on Biological Control and Management of Chromolaenaodorata. Guam, USA: Agricultural Experiment Station, University of Guam. p 55-62. FAO. 1996. International Standards for Phytosanitary Measures. Code of Conduct for the Import and Release of Exotic Biological Control Agents. Rome, Italy:Secretariat of the International Plant Protection Convention. 21 p. Gogoi A. 2001. Status of Mikania infestation in northeastern India: management options and future research thrust. In: Sankaran KV, Murphy ST, Evans HC, eds. Alien weeds in moist tropical zones, banes and benefits. Peechi, Kerala, India: Kerala Forest Research Institute. p 77-79. Holm LG, Plucknett DL, Pancho JV, Herberger JP. 1977. The world's worst weeds: distribution and biology. Honolulu, Hawaii: University Press of Hawaii. 609 p. Holmes WC. 1982. Revision of the Old World Mikania (Compositae). Botan Jahres Beitr Systematik 103:211246. Kern FD, Thurston HW, Whetzel HH. 1933. Annotated index of rusts of Colombia. Mycologia 25:448-480. King RM, Robinson H. 1987. Genera of the Eupatorieae. Monogr Syst Bot Missouri Bot Gard 22:1-581. Le6n-Gallegos H, Cummins GB. 1981. Uredinales de Mexico (Vol.1). Mexico City: Secretariat de Agricultura y Recursos Hidraulicos. 151 p. Lutz M, Bauer R, Begerow D, Oberwinkler F. 2004. Tuberculina: rust relatives attack rusts. Mycologia 96:614-626. Nair KNN. 1988. Mikania micrantha H.B.K.-a noxious weed in the forests of Kerala. Evergreen 20:13-14. Olive EW, Whetzel HH. 1917. Endophyllum-likerusts of Porto Rico. AmJ Bot 1:44-52. Parker C. 1972. The Mikania problem. PANS 18: 312-315. Romero A, Carri6n G. 1998. Primer registro de Dietelia portoricensis (Uredinales) en Mexico. Rev Mex Micol 14: 53-55. Sankaran KV, Sreenivasan MA. 2001. Status of Mikania infestation in the Western Ghats. In: Sankaran KV, Murphy ST, Evans HC, eds. Alien weeds in moist tropical zones, banes and benefits. Kerala, Peechi, India: Kerala Forest Research Institute. p 67-76. Stevenson JA. 1975. The Fungi of Puerto Rico and the American Virgin Islands. Contr Reed Herb 23:1-743. Thirumalachar MJ. 1949. Critical notes on some plant rustsII. Sydowia 5:23-29. EVANSAND ELLISON:RUST FUNGIFROMMIKANIA weed Thompson A. 1939. Notes on plant diseases in 1937-1938. Malayan Agric J 27:86-98. Urban Z. 1990. A contribution to the rust fungi (Uredinales) of Cuba. Rep Tottori Mycol Inst 28:37-56. Viegas AP. 1945. Alguns fungos do Brasil 4-Uredinales. Bragantia 5:1-144. 947 Waterhouse DF. 1994. Biological control of weeds: southeast Asian prospects. Canberra, Australia: ACIAR. 302 p. Zhang LY, Ye WH, Cao HL, Feng HL. 2004. Mikania micrantha H.B.K. in China-an overview. Weed Res 44: 42-49.