Safety of Natural Enemy Importations

Insects may be released as natural enemies of either invasive plants or invasive insects. Both biological weed control and biological insect control show a very high level of safety to vertebrate and human health. There are three safety issues when insects (herbivores, predators, or parasitoids) are imported to a new region: (1) identifi cation of unwanted contaminants in foreign shipments; (2) recognition of organisms that, by virtue of their biol ogy, may be damaging to other biological control agents; and (3) potential damage to nontarget species (e.g., native insects or plants) in the area of release by natural enemies with broad host ranges.

The first two of these safety concerns are addressed by the use of quarantine facilities, which are designed to prevent the accidental release ofnew species into the envir onment following importation. In quarantine, desired natural enemies are separated from all other materials, including miscellaneous insects that might have been accidentally included in the package by the collector, extra neous plant material, or soil. A taxonomist then confirms the species identification of the organism and ensures that all individuals collected are the same species. Voucher speci mens are deposited with an entomological museum for future reference. Natural enemy identification will either indicate the name of the organism or that it is a species not yet described. New species can usually be placed in a known genus, for which some biological information will exist. A sample of the natural enemies are also submitted to a pathologist to determine if they carry any microbial or nematode infections. Ifthey do, they will either be destroyed or, if possible, treated with antibiotics to eliminate the infection. This group of field collected, healthy individuals will then be bred in the laboratory on the target host. This eliminates any undesirable parasitoids (for herbivores attacking weeds) or hyperparasitoids (for insect agents) that might exist in the collected material that, if established, would damage the biological control project by reducing the efficacy of imported natural enemies. For insect para sitoids, rearing for one generation on the target host excludes the possibility that a hyperparasitoid has been obtained by mistake.

The potential for attack on nontarget species after release is minimized by estimating the host range of the natural enemy proposed for release and reviewing that information in light of the fauna or flora in the recipient country. Estimation of an agent's host range is based on

(1) literature records of species known to be attacked by the agent in the region from which it is collected,

(2) negative evidence in the literaure, that is, any species of interest that occur with the agent in its home range but which are not attacked, and (3) data from laboratory tests. Most evidence comes from laboratory host range tests. For herbivorous insects released for weed biological con trol, these laboratory tests include studies of both the adult's oviposition preferences and the feeding prefer ences of the immature stages and, in some cases, adults. For immature stages, tests also include an assessment of a test plant species' suitability to support growth and development. Similar tests can be applied to the study of parasitoids, that is, both oviposition preferences and sur vival of the immature stages on a given host. For predators, feeding preferences of both adults and larvae must be measured.

Historically, estimation of host ranges of herbivorous insects used against weeds began early (in the 1920s), evolving from testing only local crops, to a phylogeneti cally based attempt to define the limits of the host range by testing plants in the same genus as the target weed, then the same tribe, etc. This process has been highly successful in avoiding the introduction of insects whose host ranges are wider than initially thought. Cases of attack of introduced herbivores on nontarget plants have largely been limited to other species in the same genus and have largely been well predicted. Of 117 species introduced into North America, Hawaii, or the Caribbean for biological weed control, only one species (the lacebug Teleonemia scrupulosa, introduced into Hawaii in 1902 against the shrub Lantana camara L.) has attacked nontarget plants that were not either in the same genus as the target weed, or a very closely related genus.

Estimation of host ranges of parasitoids and predators introduced for biological control of insects began in the 1990s, in response to changing views on the ecological and conservation value of native nontarget insects. Techniques for making estimates of arthropod natural enemy safety are less well developed than for herbivorous biological control agents. A few cases of harm from para sitoids or predacious insects to nontarget insects have been reported. Importation of generalist species that have broad host ranges should be avoided because of such potential to harm native insects.

Laws explicit to biological control importations exist principally in New Zealand and Australia. Laws in the

United States regulate importation of herbivorous insects used against weeds, but do not currently regulate impor tation of parasitoids or predators.

0 0

Post a comment