Resource Partitioning Among Parasitoids (Hymenoptera: Braconidae) of Phoracantha Semipunctata in Their Native Range
Total Page:16
File Type:pdf, Size:1020Kb
Biological Control 19, 223–231 (2000) doi:10.1006/bcon.2000.0872, available online at http://www.idealibrary.com on Resource Partitioning among Parasitoids (Hymenoptera: Braconidae) of Phoracantha semipunctata in Their Native Range T. D. Paine, E. O. Paine, L. M. Hanks,1 and J. G. Millar Department of Entomology, University of California, Riverside, California 92521 Received January 11, 2000; accepted August 5, 2000 INTRODUCTION The Eucalyptus longhorned borer, Phoracantha semipunctata (F.), is native to Australia, but it has The Eucalyptus longhorned borer, Phoracantha been introduced without its natural enemies into semipunctata (F.), is distributed throughout the six many parts of the world in which its Eucalyptus spp. biogeographical regions of Australia (Wang et al., host has been planted. The beetle has developed large 1996), where it colonizes Eucalyptus L’Heritier and populations in these novel habitats and has been re- Angophora Cav. spp. (Duffy, 1963). Females lay eggs sponsible for the mortality of large numbers of trees. on the bark surface, frequently under loose bark Although there is a considerable catalogue of the para- (Drinkwater, 1975). Newly eclosed larvae mine sitoids of the beetle in Australia, limited ecological through the outer bark and feed extensively in the information on the assemblage of parasitoids attack- cambium, phloem, and outer xylem tissues (Hanks et ing P. semipunctata is available. We removed bark al., 1993a). In its native range, beetles utilize broken from 40 felled trees, recorded gallery width and bark branches and weakened or downed trees for larval thickness over parasitized larvae, and removed all development (Tooke, 1935; Chararas, 1969). parasitoids. Adult size, sex, and species were recorded Eucalyptus spp. have been widely planted through- when the parasitoid pupae eclosed. Syngaster lepidus out the world as a commercial source of cellulose, fuel, Brulle`,Jarra phoracantha Austin, Quicke, and Marsh, J. maculipennis Austin, Quicke, and Marsh, and J. and timber or as ornamental trees. P. semipunctata painei Austin and Dangerfield were most commonly has been accidentally introduced into many of these collected. The solitary parasitoid S. lepidus preferred regions and has become a significant mortality factor smaller larvae than did the gregarious Jarra spp. The for the trees, particularly in regions with Mediterra- two species with shorter ovipositors, J. maculipennis nean climates (Bytinski-Salz and Neumark, 1952; and J. painei, parasitized larvae under thinner bark Chararas, 1969; Drinkwater, 1975; Ivory, 1977; Caval- than did the other two species with longer ovipositors. caselle, 1980; Cadahia, 1986; Gonzalez-Tirado, 1986; There was a significant positive correlation between Hanks et al., 1993a, 1995a). It has been recognized host larval size and number of parasitoid pupae of the previously that adult beetles are attracted to cut logs gregarious species. Also, there was a significant posi- and to stressed trees in these novel environments tive correlation between host larval size and parasi- (Drinkwater, 1975; Ivory, 1977; Gonzalez-Tirado, toid adult size. The ecological relationships between 1986). Trees that appear to be healthy and growing this assemblage of parasitoids and their beetle host rapidly are colonized by the beetles at densities suffi- may be useful in establishing an effective biological cient to destroy the cambium tissue, resulting in the control program. © 2000 Academic Press death of the host, suggesting that trees are capable of Key Words: solitary parasitoid; gregarious parasi- vigorous growth even under stressful conditions that toid; Cerambycidae; Braconidae; resource partition- favor beetle attack. Consequently, the evaluation of ing; biological control; Phoracantha; Jarra; Syn- tree stress and susceptibility to beetle-caused injury is gaster; Eucalyptus. very difficult. Trees are capable of surviving high levels of moisture stress in the absence of the beetle (T.D.P., pers. obs.), but introduction of the beetle combined with moisture stress has created a significant mortal- 1 Current address: Department of Entomology, University of Illi- ity risk to the trees. However, Hanks et al. (1991b) nois, 505 South Goodwin Avenue, Urbana, IL 61801. demonstrated that high bark moisture content is a 223 1049-9644/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved. 224 PAINE ET AL. significant factor in the resistance of the tree to beetle the end of November 1995 to obtain small wood sam- infestation. ples to test cellulose and pulping characteristics. The As with many other arthropods, many factors may remainder of each trunk was left on the ground at the function to regulate the density of P. semipunctata site. populations. Competition among larvae (Ivory, 1977; The felled trees were sampled between 20 and 31 Mendel et al., 1984; Mendel, 1985; Haddan and Fraval, May 1996. All of the felled trees had been infested by P. 1988), environmental conditions (Hanks et al., 1991a), semipunctata, and if the borer larvae had been para- host tree susceptibility (Hanks et al., 1991b, 1995a), sitized, development of the parasitoid had reached the and host tree suitability (Hanks et al., 1993b) have late larval or pupal stages. Bark was removed from the been examined as mortality factors in geographic loca- entire infested portion of the trunk of each tree to tions where both the trees and the beetles are intro- expose the parasitized beetle larvae. Parasitoid larvae duced species. However, because the insect has not or pupal cocoons were collected and placed into sepa- been a significant problem in Australia (Duffy, 1963), rate vials for each individual host larvae. The width of there has been little work on population dynamics in the host larval gallery, the bark thickness covering the its native range. gallery, the date, and the collection site were recorded Natural enemies have been long recognized as criti- on each vial. cal factors in the regulation of insect populations The braconid wasps that use P. semipunctata larvae within ecological communities (e.g., Hairston et al., as hosts are idiobiont parasitoids. Ovipositing female 1960). Whereas the limitations on beetle population wasps inject a paralyzing venom into the host larvae growth imposed by the relations between the insect prior to oviposition, halting further growth or feeding and their host trees may be similar in the introduced by the larva. Thus, the larval gallery at the time of and native ranges of the insect, the mortality caused by parasitization provides a permanent record of host lar- natural enemies in Australia also may be critical in val size, even though the larva may have been totally regulating the population and limiting its pest status. consumed by the developing wasp. A previous study Several important taxonomic studies have described conducted in Victoria had established that there was a the parasitoid species that utilize P. semipunctata as a strong correlation between the P. semipunctata larval host (Austin et al., 1994; Austin and Dangerfield, length and the width of the larval gallery (Hanks et al., 1997); however, the ecological interactions among the 2000). Consequently, terminal gallery width was used assemblage of parasites exploiting the beetle as a re- as an indication of host larval size to examine the source are not well understood. For the purposes of preference of ovipositing parasitoid females. introducing natural enemies for biological control, it The vials containing the parasitoids were sent by air may be useful to understand the relationships among to the quarantine facility at the Department of Ento- parasitoids and their hosts in the native range so that mology, University of California, Riverside. The vials the efficacy of introductions into new environments can were examined daily for emergence of adult wasps. be optimized for maximum regulation of target host Each emerged adult was briefly anesthetized with CO2, populations. The objectives of the research presented the species and sex were determined, and the length of here were to determine (1) parasitism levels by an one hind tibia was measured. The wasps were subse- assemblage of species in one region of Australia and (2) quently placed into rearing colonies as part of a larger partitioning of resources among parasitoid species and biological control program for P. semipunctata in Cal- sex allocation within species, as a function of larval ifornia. host size and tree bark thickness. These data also The differences among species in host size preference provide a benchmark of information that will be nec- and in bark thickness were determined using analysis essary for evaluation of the success of the introductions of variance (General Linear Models Procedure) and in the new geographic range. Tukey’s Studentized range test (Sokal and Rohlf, 1981) for means separation on SAS (SAS Institute, 1988). MATERIALS AND METHODS Regression analysis (SAS Institute, 1988) was used to determine the relationship between number of parasi- Four E. globulus Labill. and E. nitens (Deane and toid pupae or emerged wasp size and host larval size. Maiden) Maiden plantations (Amcor Plantations Pro- prietary Limited) growing within a 40-km radius of RESULTS Morwell, Victoria, Australia were used as study sites. All trees within a plantation were planted in a single A total of 185 parasitized P. semipunctata larvae year, and all the plantations were established between were discovered when the bark was removed