A. S. McClay. Alberta Environmental Centre, Canada. R. E. McFadyen. Alan Fletcher Research Station, Box 36, Sherwood, Queensland, Australia 4075

Transcription

1 Bulletin of Entomological Research (1990) 80, Biology of Bucculatrix parthenica Bradley sp. n. (Lepidoptera: Bucculatricidae) and its establishment in Australia as a biological control agent for Parthenium hysterophorus (Asteraceae) A. S. McClay Alberta Environmental Centre, Canada R. E. McFadyen Alan Fletcher Research Station, Box 36, Sherwood, Queensland, Australia 4075 J. D. Bradley c/o The Natural History Museum, Cromwell Road, London, SW7 5BD, UK Abstract Bucculatrix parthenica Bradley sp. n., a moth native to Mexico, is described. It has been released and established in Queensland, Australia, as a biological control agent for its host plant, Parthenium hysterophorus. The moth oviposits on leaves of its host. First and second instar larvae are leaf miners, and later instars feed externally on the leaves. The life cycle occupies about 25 days under field conditions. B. parthenica was narrowly oligophagous in hostspecificity tests. In Mexico the insect is scarce but in Queensland it has become abundant enough to cause extensive defoliation of its host plant at some sites. Its rapid increase in Queensland is attributed to the absence of parasitism. Introduction Parthenium hysterophorus (Asteraceae), an annual herb native to Mexico, has been widely introduced into other tropical and subtropical areas (Towers et al., 1977). It has become a serious weed of grazing land in Queensland, Australia (Auld et al, 1983; Haseler, 1976). From 1978 to 1983, the Queensland Department of Lands sponsored studies on the natural enemies of P. hysterophorus in Mexico. This project, undertaken by the Commonwealth Institute of Biological Control (now CAB International Institute of Biological Control), was aimed at finding biological control agents for release in Queensland. One insect discovered was an undescribed Bucculatrix species (Lepidoptera: Bucculatricidae). After host Correspondence: Dr A.S. McClay, Alberta Environmental Centre, Bag 4000, Vegreville, AB, Canada, TOB 4LO. specificity testing, this insect was released (McFadyen, 1985) and established in Queensland under the name Bucculatrix species "D". It is here described as Bucculatrix parthenica Bradley, and its biology and establishment in Australia are recorded. Taxonomy (Bradley) Bucculatrix parthenica sp. n. (figs 1-4) The species should be cited: Bucculatrix parthenica Bradley Description Adults (fig. 1). Wing expanse, male and female, 7 mm. Head white, frons (front) often faintly tinged with ochreous, mostly on lower part, an admixture of brown or

2 428 A.S. McClay, R.E. McFadyen and J.D. Bradley blackish in fine hair-like cranial tuft. Eye-cap white, usually irrorate with a few sparse dark brown or blackish-tipped scales. Antennal scape white, flagellum greyish fuscous with narrow paler or whitish annulation, pedicel shallowly notched in male. Thorax, tegula and patagium white, tegula and thorax sparsely irrorate with blackish-tipped scales. Forewing ground colouration white, often with a trace of pale greyish suffusion and some scattered dark brown or blackish scales between markings; fasciate markings poorly developed, diffuse, strongest on costal margin as oblique wedge-like patches or streaks of pale ochreous suffusion coarsely irrorate with loosely appressed or roughened blackish-tipped scales, the extent and intensity of colouration variable; antemedian fascia arising from costa at about 1/3, narrowly produced along costa to base, outward-oblique from costa to near middle of wing, thence obsolescent, discernible basad as an obscure patch above dorsum; postmedian fascia comparatively well developed, darkest near costa, outwardoblique from costa then angled sharply inward before reaching middle of wing, becoming weaker but discernible to dorsum, containing a group of raised blackishtipped scales in fold a little above dorsum, outer (median) angle of fascia usually prolonged into a narrow diffuse streak extending to tornus; subterminal fascia represented by a poorly defined broadly triangulate patch on costa, followed by a distinct black apical spot; cilia whitish, suffused with pale ochreous-grey around apex and in tornal area along dorsum, a distinct line of blackish-tipped scales extending medially from apex to near middle of termen. Hindwing pale grey; cilia pale ochreous basally, becoming greyish apically from before middle. Legs greyish white, tarsal segments conspicuously blackish-tipped, tibiae of foreand middle legs suffused with fuscous exteriorly, hind tibia clothed with long pale ochreous hairs. Abdomen pale grey, lighter or whitish ventrally; anal tuft of male whitish. Abdomen of male with an eversible sac of scent-scales in mid-dorsal invagination between segments 2-3. Male genitalia (figs 2, 3). Valva elongate, moderately broad basally, narrowing distally with the ventral margin gently rounded, apex angulate at costa, costal margin strongly setose in apical portion, an admixture of stout cone-like setae towards apex, basal projection of costa strongly sclerotized, curving ventrally; vinculum broadly rounded, with narrow median vertical band; socii erect, slightly dilated apically, sparsely setose proximad of apex; gnathal subscaphium an elongate spinulose strip; anellus broad microscopically spinulose, weaker or sutured medially; aedeagus long, cylindrical, very slightly curved, apex deeply cleft, with five to six minute thorn-like barbs at base of cleft. Female genitalia (fig. 4). Papillae anales membranous, each with an extrusible patch of dentate setae on inner margin; ostium retracted beneath caudal margin of 7th sternite, ventral margin convex, antrum or proximal portion of ductus bursae sclerotized, spiculate internally, the spiculation extending exteriorly onto dorsal wall, distal portion of ductus bursae membranous, narrow, entering bursa copulatrix at juncture of ductus seminalis, bursa elongate-ovate, encircled before middle by a series of spined rib-like structures in longitudinal rows, forming a signum, the spines of each row finer and multi-branched distally. Material examined. Holotype d", AUSTRALIA (ex Mexico): Queensland, Brisbane, ii iii.1988, laboratory reared on leaves of Parthenium hysterophorus (R.E. McFadyen); abdomen in situ. Paratypes, AUSTRALIA (ex Mexico: same data as holotype, 29 cf, 30 9; AUSTRALIA (ex Mexico): Queensland, Brisbane, xi.1983, laboratory reared on leaves of Parthenium hysterophorus, Bucculatrix sp. "D" (R.E. McFadyen), 8 d\ 3 9; MEXICO: Nuevo Leon, colony culture Parthenium hysterophorus, 22.vii.1983, W.A. Palmer leg. (A.S. McClay), lcf. Holotype and paratypes deposited in The Natural History Museum, London, excepting 4 paratypes in the Australian National Insect Collection (ANIC), Canberra, ACT, 4 paratypes in the Queensland Museum, Brisbane, and 4 paratypes in the National Museum of Natural History, Washington DC, USA. Comments. B. parthenica is alien to the several bucculatricids indigenous to Australia, and is closely related to its Nearctic congener B. agnella Clemens, the larva of which feeds on Ambrosia artemisiifolia (Asteraceae) which is also an introduced weed in Australia but is not attacked by B. parthenica. The adults of the two species are similar in size and wing pattern, and accurate identification requires examination of the genitalia. Both sexes of B. agnella are figured in Braun (1963). In the male of B. parthenica the valva narrows and has the ventral margin rounded distally and the apex is angulate at the costa; in B. agnella the valva is nearly parallel-sided throughout its length and is weakly emarginate distally and the apex is broadly rounded at the costa. In the female of B. parthenica the ventral margin of the ostium is convex, whereas in B. agnella it is rounded. Biology (McClay & McFadyen) Life history and host specificity Methods. B. parthenica was collected from P. hysterophorus in and around Monterrey, Nuevo Leon, Mexico. Fieldcollected adults or adults reared from field-collected larvae and pupae were placed in cages with plants of P. hysterophorus. Studies in Mexico were conducted in an open insectary under ambient conditions of temperature, photoperiod, and humidity. They included observations on the life history and initial host specificity tests against a short list of species of Asteraceae. Further host specificity tests were conducted in a quarantine insectary in Queensland against additional species of Asteraceae and cultivated and native Australian plants in a wide range of families (table 1). All tests were conducted by releasing adult B. parthenica into multiple-choice cages containing P. hysterophorus and several other test plant species, and observing subsequent larval damage. Results. Eggs of B. parthenica are laid singly on the upper or lower leaf surfaces of P. hysterophorus. They are translucent, flattened, and oval in shape, longitudi-

3 Biology of Bucculatrix parthenica 429 Figs 1-4. Bucculatrix parthenica sp. n. Fig. 1. Adult, male (wing expanse 7 mm). Fig. 2. Male genitalia, ventral view. Fig. 3. Aedeagus, lateral view. Fig. 4. Female genitalia, ventral view.

4 430 A.S. McClay, R.E. McFadyen and J.D. Bradley Table 1. Plant species used in host specificity tests with Bucculatrix parthenica. 'Plants attacked by B. parthenica. ANACARDIACEAE: Mangifera indica. ANNONACEAE: Annona reticulata. APIACEAE: Apium graveolens, Daucus carota. ASTERACEAE: Ambrosia artemisiifolia, Ambrosia confertiflora, Baccharis halimifolia, Bidens pilosa, Carthamus tinctorius, Cassinia laevis, Chrysanthemum morifolium, Cichorium intybus, Coreopsis lanceolata, Cosmos sulphureus, Dahlia pinnata, Eclipta prostrata, Helianthus annuus, Lactuca sativa, *Parthenium argentatum, *Parthenium confertum, *Parthenium hysterophorus, Rudbeckia sp., Viguiera dentata, Xanthium strumarium, Zinnia elegans. BRASSICACEAE: Brassica oleracea, Brassica rapa. BROMELIACEAE: Ananas comosus. CARICACEAE: Carica papaya. CHENOPODICEAE: Beta vulgaris. CONIFERAE: Pinus elliotti. CUCURBITACEAE: Cucumis sativus. FABACEAE: Arachis hypogaea, Glycine max, Medicago sativa, Phaseolus vulgaris, Stylosanthes humilis. LAURACEAE: Persea americana. LILIACEAE: Allium cepa. LINACEAE: Linum usitatissimum. MALVACEAE: Gossypium arboreum. MUSACEAE: Musa paradisiaca. MYRTACEAE: Eucalyptus grandis. PASSIFLORACEAE: Passiflora edulis. POACEAE: Astrebla elymoides, Panicum maximum, Triticum aestivum, Saccharum officinarum, Sorghum vulgare, Zea mays. PROTEACEAE: Macadamia integrifolia. ROSACEAE: Fragaria x ananassa, Malus sylvesris, Prunus persica, Rosa centifolia. RUBIACEAE: Coffea arabica. RUTACEAE: Citrus sinensis. SOLANACEAE: Capsicum annuum, Lycopersicon esculentum, Nicotiana tabacum, Solanum tuberosum. VITACEAE: Vitis vinifera. ZINGIBERACEAE: Zingiber officinale. nally wrinkled, and measure approximately 0.22x0.14 mm. On hatching, larvae bore directly into the leaf through the attached surface of the egg, and begin mining the mesophyll. The mine usually follows a sinuous path initially, until it reaches a main vein or the edge of the leaf, which it then often follows for the rest of its course. The first two moults occur in the mine, and the third instar larva emerges onto the leaf surface, where it soon moults again. Fourth and fifth instar larvae feed externally on the leaves, either skeletonizing or perforating them. When disturbed, the larvae readily drop from the leaves on silk threads. The third and fourth moults occur in a rounded, loosetextured cocoon, usually adjacent to a midvein on the lower leaf surface. Pupation occurs in a white fusiform cocoon about 4 mm in length, bearing eight to ten sharp longitudinal ridges. The cocoon is surrounded by a palisade of upright silk threads enclosing an oval area about 6x4 mm. The cocoon is usually constructed on a leaf undersurface, but petioles and stems are also used. Larvae of one generation which pupated in a breeding cage in December 1982 made almost all their cocoons on fallen leaves, the sides of the pot, or the floor and walls of the cage. This suggests that overwintering may occur in cocoons constructed in similar sites, such as in litter or on twigs and stones. After adult emergence, the exuviae are left protruding from the end of the cocoon. In cages, adults normally live four to five days but a few may survive for up to 14 days. Mating and oviposition occur within 24 hours of emergence. At C, the life cycle occupies about 25 days. No attack by B. parthenica was observed in the field or laboratory on any plant outside the genus Parthenium. Parthenium confertum Gray, a herbaceous perennial occurring in northern Mexico, was attacked in cage tests. Attack on this species in the field was not observed but seems probable. In one cage test, considerable feeding occurred on Parthenium argentatum Gray (guayule), a shrub occasionally cultivated as a source of rubber. Plants of P. hysterophorus in this test were heavily defoliated, and larvae may have wandered from them onto P. argentatum. In later tests in Mexico and Australia, no feeding was observed on P. argentatum. Population density, phenology, and distribution Methods. Plots were sown with P. hysterophorus at Apodaca, near Monterrey, and samples of 25 plants collected at two-week intervals from April 1979 to April The numbers of naturally occurring larvae and cocoons of B. parthenica in each sample were recorded, along with those of other insects. Surveys of insects on P. hysterophorus were conducted throughout most of the plant's range in Mexico, and the presence of B. parthenica was noted where found. Results. B. parthenica was found in the states of Nuevo Leon, Coahuila, and Tamaulipas. Field populations were always low, and significant host plant defoliation was never observed. Mean populations in the experimental plots over the study period were fourth and fifth instar larvae and pupal cocoons per plant. Most were found in March-April and October; however, total numbers were too small to draw reliable conclusions on seasonal occurrence. In the field, B. parthenica was almost always collected as occasional isolated larvae or cocoons. In cages under ambient conditions, however, it rapidly increased to damaging population levels. Eight specimens of Hymenoptera were reared from larvae and pupae of B. parthenica in Mexico. They belonged to three species: Apanteles sp. (Braconidae), an unidentified braconid (subfamily Hormiinae), and an unidentified ceraphronid (probably hyperparasitic). The rate of parasitism was 20% (n = 40). No parasitoids were reared from any stage of the moth in Australia. Releases and establishment (McFadyen) Methods. The Queensland colony was derived from a small number of individuals field-collected near Monterrey in October Four hundred and forty-three laboratory-reared pupae from these parents were sent from Mexico in July Releases started when permission was granted by the Australian federal plant quarantine authorities in January To mass rear B. parthenica for field releases in Queensland, cages were set up with eight young potted

5 Biology of Bucculatrix parthenica 431 plants of P. hysterophorus and 60 to 100 adults were released into each cage. Three to four weeks later, shortly before emergence of the next generation adults, all plant material in the cage was cut down and released directly onto suitable areas of P. hysterophorus in the field. This method was labour-efficient but only allowed approximate recording of the numbers released (about 600 to 1000 individuals per cage). From January 1984 to February 1985, 118 cages were used in 26 separate releases, three to five cages per release. Releases were made throughout the year but at a reduced level in winter. From March to September 1985, five releases totalling eight cages were made, after which releases ceased and the laboratory colony was terminated. Releases were made in most of the P. hysterophorus infested areas of central Queensland, and in the small isolated area of Toogoolawah in the south-eastern part of the state. Results. Few signs of establishment were seen immediately after releases. Release sites were inspected in April, September, and October 1984, and in January, March, and May At two sites, leaf mines and feeding damage were seen one to three months after release, but not subsequently. At other sites, no damage or signs of recent feeding were seen. As a result, in May 1985 it was decided that establishment had failed and releases should cease. Several sites were inspected again in January and March 1986, and regular samples for the stem-galling moth Epiblema strenuana (Walker) (Lepidoptera: Tortricidae) were taken from July 1986 to January B. parthenica was not seen in these surveys. In January 1987 another inspection was made of insect damage to P. hysterophorus in central Queensland, and B. parthenica was found at every site visited, over a distance of about 500 km. Subsequent visits established that the moth was increasingly abundant from early summer (November-December) to autumn and winter (May-June), and then scarce or absent during late winter and spring. Within this seasonal pattern, local abundance depends on host plant availability. This is in turn dependent on the rainfall, which is highly variable and erratic in central Queensland. In June 1988 B. parthenica was found in an isolated area of P. hysterophorus north of Roma, Queensland, where no releases had been made. The moth must have dispersed over the approximately 120 km from the nearest area with P. hysterophorus. B. parthenica has, however, still failed to establish in the small area of P. hysterophorus at Toogoolawah in south-eastern Queensland. Discussion A detailed evaluation of the impact of B. parthenica on P. hysterophorus in Queensland has not yet been conducted; however, when they are abundant the larvae destroy up to 50% of the total leaf area, which must reduce the plant's competitiveness and seed production. In most areas and seasons, B. parthenica is present but scarce, and has correspondingly little impact. Nevertheless, its short life cycle allows the moth to increase rapidly in favourable conditions. Its overwintering pattern differs from that of the stem galling moth E. strenuana, also introduced from Mexico (McFadyen, 1985), so that the two species are to some extent complementary, with at least one being abundant most of the time. The development of B. parthenica populations in Queensland is in striking contrast to their behaviour in Mexico. It is quite possible that only five years after introduction, more B. parthenica exist in Queensland than in its entire native range. The most likely explanation for the outbreak in Queensland is the absence of parasitism. Although parasitism in Mexico was relatively low (20%), this occurs at a very low density of B. parthenica, suggesting that the parasitoids are efficient at finding their hosts. The rapid increases observed in cages under ambient conditions in Mexico, where B. parthenica was protected from parasitism, are consistent with this explanation. B. parthenica is one of the most host-specific agents available for biological control of P. hysterophorus; while other agents introduced into Queensland are capable of completing development on Ambrosia spp. (McClay, 1985), B. parthenica is narrowly oligophagous within the genus Parthenium. In host-specificity tests, B. parthenica fed only on three species of Parthenium. One of these, P. argentatum, is not thought to be a normal host, as feeding on it was observed only once and was probably due to migration of larvae from heavily defoliated plants of P. hysterophorus. Such a limited host range is typical of Bucculatrix spp.; of the 65 species for which host plant information is given by Braun (1963), 53 are recorded only from a single host genus and 43 from a single species. B. parthenica can therefore be safely introduced in any area where P. hysterophorus is a problem. Acknowledgements We thank Dr Dieter Enkerlin and Dr Juan D. Vega for providing facilities in Mexico at the Instituto Technologico y de Estudios Superiores de Monterrey; Ing. Hugo Miranda Sanchez and Ing. Manuel Valencia for technical assistance in Mexico; Mr C. Whitfort, Mrs C. Lockett, Ms B. Radcliffe, and field staff of the Queensland Department of Lands for technical and field assistance in Queensland; Dr I.D. Gauld (Department of Entomology, The Natural History Museum, London), and the late Dr G.E.J. Nixon (CAB International Institute of Entomology) for identifications of parasitoids; and Dr L.A. Mound, Keeper of Entomology, The Natural History Museum, London, for providing facilities for the taxonomic part of this study. References Auld, B.A., Hosking, J. & McFadyen, R.E. (1983) Analysis of the spread of tiger pear and parthenium weed in Australia. Australian Weeds 2, Braun, A.F. (1963) The genus Bucculatrix in America north of Mexico (Microlepidoptera). Memoirs of the American Entomological Society 18, Haseler, W.H. (1976) Parthenium hysterophorus L. in Australia. Pest Articles and News Summaries 22, McClay, A.S. (1985) Biocontrol agents for Parthenium hysterophorus from Mexico, pp in Delfosse, E.S. (Erf.). Pro-

6 432 A.S. McClay, R.E. McFadyen and J.D. Bradley ceedings of the Sixth International Symposium on Biological Towers, G.H.N., Mitchell, T.C., Rodriguez, E., Bennett, F.D. Control of Weeds, August 1984, Vancouver, Canada. and Subba Rao, P.V. (1977) Biology and chemistry of Par- McFadyen, R.E. (1985) The biological control programme thenium hystewphorus, a problem weed in India. Journal of against Parthenium hystewphorus in Queensland, pp Scientific and Industrial Research 36, in Delfosse, E.S. (Ed.). Proceedings of the Sixth International Symposium on Biological Control of Weeds, August 1984, C-A-B International 1990 Vancouver, Canada.