Non-Target Effects of Grass-Specific Herbicides Differ among Species, Chemicals and Host Plants in Euphydryas Butterflies
This study examined the effects of grass-specific herbicides (graminicides) on three species of caterpillars in the genus Euphydryas in the greenhouse. Graminicides are often used in restoration of butterfly habitat to reduce invasive grasses, and their benefits for habitat have been demonstrated in numerous studies. While habitat can benefit, the effects of the herbicides on caterpillar or adult butterflies are unclear. In this research, one experiment compared the effects of three graminicides as their commercial formulations (clethodim as Envoy Plus, sethoxydim as Poast, and fluazifop-p-butyl as Fusilade DX, all applied along with the adjuvant NuFilm IR, as well as controls of NuFilm IR and of water) on E. colon reared on Plantago lanceolata. A second experiment assessed the effects of fluazifop-p-butyl on three species (E. colon, E. editha coloni, and E. phaeton) each raised on two different hostplants (one native and one novel species). In both experiments, they raised collected eggs on P. lanceolata to the second instar, and then applied the herbicides to the plants and larvae at recommended field rates.
In the first experiment, survivorship of E. colon was reduced by sethoxydim but not significantly different for clethodim or fluazifop-p-butyl. Total iridoid glycosides did not vary among treatments, but graminicide-treated larvae had significantly higher aucubin than control larvae. Catalpol levels varied among the groups and were significantly lower than the control in clethodim-sprayed larvae.
In the second experiment, fluazifop-p-butyl herbicide treatment did not influence survival or development time of any of the Euphydryas species. In general, the herbicide consistently reduced growth on P. lanceolata but had no consistent effect on larvae reared on their native hosts. Graminicide only influenced the foraging frequency of one species on one hostplant (E. phaeton on its native hostplant Chelone glabra). Graminicide reduced the size of larvae groups in E. colon and E. phaeton, and group size also differed among Euphydryas and hostplant species. In this experiment, the hostplant effects were noticeably stronger than graminicide effects. Some of this effect for E. colon could have been due to the fact that the larvae experienced food shortages on the native hostplant Castilleja because it was senescing during the experiment.
The experiments showed both lethal and sublethal effects from graminicides, with variation by chemical, species, and hostplant. E. colon larvae in sethoxydim treatments were less likely to survive to diapause, while those treated with fluazifop-p-butyl and NuFilm were not notably impacted, and results for clethodim were inconclusive. The results for iridoid glycosides were surprising, with the percent dry weight of aucubin almost doubled in the sethoxydim, clethodim, and fluazifop-p-butyl treatments as compared to NuFilm and the control. In the graminicide treatments there was about twice as much aucubin as catalpol, which was reversed in the NuFilm and control with substantially more catalpol than aucubin. The authors note that catalpol is more toxic, so changes in the relative amounts of these compounds could affect how the caterpillars interact with their natural enemies. As aucubin is the precursor to catalpol, these results suggest a change in iridoid biosynthesis in P. lanceolata in response to the graminicides since sequestration of iridoid glycosides typically reflect the ratios in the hostplant. Graminicide exposure reduced larval growth rates, leading to smaller mass at diapause for all three species on P. lanceolata, which can be a predictor for survival through diapause. Graminicide application also reduced group size in E. colon and E. phaeton, which can have unpredictable consequences for foraging behavior.
While some other studies have demonstrated sensitivity to fluazifop-p-butyl, these results suggest that it could be a potential management tool to reduce non-native grasses in habitat. The authors conclude that their results guardedly support the use of fluazifop-p-butyl because it is less damaging to native butterfly populations than the other graminicides evaluated. Nonetheless, it does cause changes in iridoid glycoside profiles and in gregarious feeding behavior, which could influence palatability to predators and increase exposure to predation, respectively. It is unclear if these detrimental effects would outweigh potential benefits of habitat improvement, indicating a need for more research to evaluate the costs and benefits of fluazifop-p-butyl use throughout a species’ life cycle.
