Biological Control of Lippia (Phyla Canescens): Surveys for the Plant and Its Natural Enemies in Argentina
Total Page:16
File Type:pdf, Size:1020Kb
Biological control of lippia (Phyla canescens): surveys for the plant and its natural enemies in Argentina A.J. Sosa,1 M.G. Traversa,2 R. Delhey,2 M. Kiehr,2 M.V. Cardo1 and M.H. Julien3 Summary Lippia, Phyla canescens (Kunth) Greene (Verbenaceae) is a fast-growing, mat-forming plant native to South America. It is a weed in Australia, where it was introduced as an ornamental during the nineteenth century. The knowledge about the biology of lippia is currently limited to unconcluded taxonomical studies; there is scarce information about the ecology and natural enemies in the native range. Surveys for the plant and its natural enemies were initiated in Argentina in 2005 to determine its distribution and to search for possible biological control agents, both insects and phytopathogens. We have found Phyla sp. in 54 out of 102 sites sampled, mostly east of 66°W, circumscribing the weed to the Chaco Domain. In places where the plant was present, at least 20 arthropods and 16 fungi were found. Among insects, the most promising candidates are three flea beetles (Chrysomelidae): two species of Longitarsus and Kuschelina bergi Harold. Pathogens include the rust Puccinia cf. lantanae Farl., Cercospora cf. lippiae Ellis and Everh. and three Colletotrichum spp., associated with leaf spots and stem cankers. Additional information on their biology and host specificity is required to propose any of these as biological control candidates. Keywords: plant distribution, arthropods, pathogenic fungi. Introduction South Africa, New Zealand and Australia (Kennedy, 1992). This plant is a weed in Australia, where it was Lippia, Phyla canescens (Kunth) Greene (Verbenaceae) introduced as an ornamental plant during the second is a fast-growing, mat-forming plant. It is widespread half of the nineteenth century. It is a major threat to in and thought to be native to South America (from biodiversity and riparian areas and has a significant southern Ecuador, throughout Peru, Chile, Argentina, impact on conservation and grazing systems due to its Uruguay, Paraguay, Bolivia and Brazil) (Collantes, increasing density and distribution (Julien et al., 2004), et al., 1998; Múlgura de Romero et al., 2003). It was causing annual losses of cattle production of 38 million also recorded from fossil pollen in Santa Fe and Buenos Australian dollars and even greater estimated losses of Aires Provinces, in Argentina (Alzugaray et al., 2003; environmental service. Current short-term and unsus- Fontana, 2005), reinforcing this area as a centre of tainable control methods include the use of herbicides, origin. cultivation and grazing management (Earl, 2003; Ju- P. canescens has been reported naturalized from lien et al, 2004). Biological control is proposed as the France, Spain, Italy, Algeria, Botswana, Senegal, Egypt, sustainable method for this weed and may be the only option for conservation areas, woodlands, forests and along stream banks. However, until this study, there 1 USDA-ARS South American Biological Control Laboratory. Bolivar was little information about the plant and its natural 1559 (B1686EFA), Hurlingham, Buenos Aires, Argentina. 2 Laboratorio de Patología Vegetal, Departamento de Agronomía, enemies in the native range. Universidad Nacional del Sur. (8000), Bahía Blanca, Buenos Aires, Worldwide, Kennedy (1992) recognized nine spe- Argentina. cies of Phyla including P. canescens. In Argentina, an 3 CSIRO Entomology European Laboratory. Campus International de additional species, Phyla reptans (Kunth) Greene was Baillarguet. 34980 Montferrier sur Lez, France. Corresponding author: A.J. Sosa <[email protected]>. recognized together with Phyla betulaefolia Greene © CAB International 2008 and P. canescens (Múlgura de Romero et al., 2003). 211 XII International Symposium on Biological Control of Weeds Australia has P. canescens and Phyla nodiflora Greene P. canescens is a small, prostrate plant, often (Munir, 1993); however, there is confusion between growing as an understory plant and therefore diffi- these two species in the literature, and it remains un- cult to see. Sampling was conducted along roadsides certain whether P. nodiflorais native to Australia or not every 100–180 km. At each site, the presence of the (Leigh and Walton, 2004). plant was checked, and if present, natural enemies There are no records of arthropods associated with were collected. In 27 sites where plants were diffi- Phyla spp. and few for fungal pathogens. In North Ame- cult to identify in the field (because of absence of rica, two biotrophic fungi: Oidium sp. (powdery mil- inflorescences), specimens were collected and culti- dew) and Meliola lippiae Maubl. (black mildew) and vated in the greenhouse for identification. Repeated the following necrotrophs: Cercospora lippiae Ellis sampling was conducted at two (region 4) and ten and Everh. (leaf spot), Sphaceloma lippiae Baines and sites (regions 1–3) in each region at different times Cummins (anthracnose) and Sclerotium rolfsii Sacc. of the year. Soil was sampled at 14 sites with and (southern blight) were identified onP. nodiflora and/or without lippia, and pH and relative humidity were Phyla lanceolata (Michaux) Greene (Farr et al., 1989). recorded. Samples were analysed for phosphates, ni- C. lippiae (syn. Pseudocercospora lippiae) occurred on trates, nitrites, ammonia, calcium, phosphorous, iron Phyla strigulosa (Mart. and Gal.) Moldenke and Phyla and humus (LaMotte Combination Soil Model STH- spp. in the Caribbean, Asia and Africa (Ellis, 1976). 14). This information was examined with Reciprocal On P. nodiflora (=Lippia nodiflora, including also Averaging (PC-ord 4). P. reptans?) in India, C. apii Fresen. emend. Crous and U. Braun was recorded (Crous and Braun, 2003) and Natural enemies in South America, Meliola lantanae Syd. and P. Syd. and Phoma zappaniae Speg. (Viégas, 1961). In Ar- Arthropods: Arthropods were collected directly from gentina, the rust Puccinia lantanae Farl. was found on plants using aspirators or from plant parts attacked by P. canescens in Salta Province, on Phyla sp. in Tucu- endophagous species (e.g. miners or gall formers) and mán Province and in Paraguay (Viégas, 1961; Lind- taken to the laboratory for rearing. Material was placed quist, 1982). There are also records of Prospodium in plastic containers (8 cm diameter, 5 cm high) with spp. on Lippia (including Phyla) in South America moistened tissue paper and leaves of the plant as food, (Viégas, 1961). and kept in growth chambers at 25°C and 12 h light. To gain knowledge of the natural distribution, centre All adults that were collected or reared were sent to of origin and natural enemies of P. canescens in South taxonomists for identification. Relevant biological in- America, surveys were initiated in December 2005. formation was recorded. The aim was to identify suitable arthropod and phyto- Laboratory studies were conducted for the flea pathogen biological control agents for Australia. beetle collected in two places: Tres Arroyos (38.52°S, 60.51°W) and Nueva Atlantis (36.85°S, 56.69°W) in Buenos Aires Province. Eggs laid by field-collected Methods and materials adults were kept in plastic containers (10 cm diameter, 2 cm high), and P. canescens leaves were added to the Surveys resulting larvae. The last instars were reared in a bigger P. canescens is recorded in 17 provinces of Argen- container, and when the larvae decreased their activ- tina (Múlgura de Romero et al., 2003). Considering ity (prepupal stage), they were transferred to another these records and bibliographical information, sur- container (8 cm diameter, 5 cm high) with soil as sub- veys were conducted in four ecological regions of strate. Argentina from December 2005 to February 2007 Pathogens: Plants were inspected for symptoms of (Figure 1): (1) Wetland Chaco, a humid area with a disease, representative samples were collected, and rainy season in summer, characterized as wetlands the presence of fungi was checked in the lab. When with patches of mesquite forests, which includes the necessary, samples were placed into humid chambers north-eastern provinces of Entre Ríos, Santa Fe, Cor- to encourage sporulation. Isolations were made by rientes, Chaco and Formosa; (2) Dry Chaco (similar placing disinfected leaf and stem pieces onto different to the wet Chaco but mostly grassland) and Yungas agar culture media: potato dextrose agar (PDA), wa- (subtropical mountain rain forest), including the north- ter agar (WA) and specific media (Phyla leaf-oat meal western provinces of Jujuy, Salta, Santiago del Estero, agar—a modification of carrot leaf-oat meal agar; Tucumán, Catamarca, La Rioja, Córdoba, Mendoza Dhingra and Sinclair, 1985). The mycelial cultures and San Luis; (3) Pampas, grassland of Buenos Aires were exposed to an UV regime for sporulation. Fungi Province; and (4) Transition zone between Southern were then identified. The isolated fungi are main- Chaco, Pampas and Patagonia, a cold and dry area tained on PDA. Desiccated specimens of infected including Río Negro, Neuquén and Mendoza but plants have been incorporated in the herbarium of the wet in the Río Negro basin where P. canescens was Phytopathology Laboratory of the University of Bahía found. Blanca. 212 Biological control of lippia (Phyla canescens): surveys for the plant and its natural enemies in Argentina P. reptans and some possibly intermediate forms were Results found. High phenotypic variation and poor distinguish- ing characters (presence or absence of conspicuous sec- Surveys ondary leaf venation) made it difficult to discriminate Phyla sp. was recorded in 54 of 102 sites sampled, these species in the field. Specimens that were taken mostly east of 66°W and north of 40°S, from sea level for identification turned out to be both P. reptans and (Buenos Aires Province) to 2100 m (Volcán, Jujuy P. canescens. Elsewhere, populations found were eas- Province) suggesting a natural distribution in the Chaco ily identified asP. canescens. Region and in the Pampas (Chaco Domain) (Cabrera Preliminary analyses of soil samples did not detect and Willink, 1980; Figure 1). In the northern half of any obvious differences between sites with or without region 2, pure and mixed stands of P.