|Farrar, Robert - Bob|
Submitted to: Environmental Entomology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/3/1998
Publication Date: N/A
Citation: Interpretive Summary: Nuclear polyhedrosis viruses (NPVs) are naturally occurring viruses that infect only insects and some other arthropods. They are promising alternatives to chemical pesticides for many important insect pests, especially caterpillars. One of the major drawbacks to using NPVs as biological pesticides, however, is the fact that they are relatively slow acting, usually requiring several days to kill insects. As a result, much attention has been focused on the speed of kill of NPVs, and how it may be increased. However, the way in which many scientists report speed of kill can be misleading. When reporting results of a test of NPVs, scientists often report the time required for half of the insects to die (LT50). Some scientists report data on all test insects, whether they are infected or not, while others report data only for those insects that actually die from NPV. If all insects are included, and some are not infected, the LT50 will be longer than if only insects that die are included. Treatments that increase mortality (decrease the number not infected) thus appear to also decrease the LT50, even though those insects that actually die may do so in the same time. In this paper, we review the scope of the problem, its potential significance, and potential solutions. We suggest that scientists should consistently report data only on those insects that actually die. This will result in more realistic evaluation of NPVs, and will increase the likelihood that they will be used in practical insect management programs.
Technical Abstract: The time-mortality relationship for nuclear polyhedrosis viruses (NPVs) is one of the most important factors to be considered in the development of NPVs as biological control agents. This factor has become of particular interest in recent years with the advent of NPVs that have been genetically modified to kill insects faster. However, much of the data published on time-mortality relationships is calculated in such a way that the results may be misleading. In quantifying time-mortality relationships for NPVs, many researchers calculate median lethal times, or LT50s, using probit or similar analyses. However, some researchers include insects that survive the treatment in their calculations, while other researchers do not include survivors. Simulated data are presented to show that when survivors are included, treatment effects on LT50s are inherently influenced by any treatment effects on rates of mortality. Examples of real data from bioassays of the NPV of the celery looper, Anagrapha falcifera (Kirby), against the corn earworm, Helicoverpa zea (Boddie), and the beet armyworm, Spodoptera exigua (Hubner), that illustrate the same pattern are also presented. Rationale is presented that if survivors are simply not infected, data on them represents no useful information on the time-mortality relationship, and so survivors should generally be excluded from calculation of LT50s, as well as from alternative methods of quantifying speed of kill. Mortality data should be presented as well, but separately, to allow full evaluation of the experiments.