Distribution, population structure and habitat use of the endangered Saint Francis Satyr butterfly, Neonympha mitchellii francisci

American Midland Naturalist, The, April, 2008 by Daniel Kuefler, Nick M. Haddad, Stephen Hall, Brian Hudgens, Becky Bartel, Erich Hoffman

LARVAL FEEDING EXPERIMENTS

We did not find any Neonympha mitchellii francisci larvae in the field, despite intensive searches. Occasionally, we observed other insects feeding on the plants, including larvae of the Appalachian eyed brown butterfly (Satyrodes appalachia), which is known to be a generalist feeder on Carex.

Our larval feeding experiments support the hypothesis that Carex mitchelliana is an important larval food resource. They also indicate, however, that other species may be used. In a test of initial acceptance, all neonates (n = 4) offered C. mitchelliana fed upon it immediately. In contrast, acceptance rates for the other species offered varied from 3 out of 4 on C. atlantica, 2 out of 4 on C. lurida and C. stricta and only 1 out of 4 on C. glaucescens, C. lonchocarpa and C. turgescens. Three larvae that failed to eat these other species were subsequently transferred to C. mitchelliana. All four larvae started on C. mitchelliana survived to the 6th instar and two survived to adulthood, the only ones to do so. Two of the three larvae transferred to C. mitchelliana also survived to the 6th instar. In contrast, only a single larvae feeding on C. stricta, lurida and glaucescens survived to the 6th instar, with none entering pupation.

Although these results are consistent with the habitat correlations we have observed, growth rates were much slower in the lab than the), must be in the wild, as the larvae that reached adulthood took weeks longer to do so than their wild cohort. We suspect that food quality may have been a factor--we fed the larvae on cut leaf segments rather than on whole plants--and we cannot rule out the possibility that this biased our results in other ways as well

[FIGURE 1 OMITTED]

PHENOLOGICAL PATTERNS

Butterflies were consistently more active in the afternoons than in the mornings and less active on sunny days. We observed an additional butterfly for every 3 h later a survey was conducted between 09:00 and 17:30 (F = 14.8, P > 0.002, df = 159). Temperature did not influence butterfly counts (F = 2.3, P = 0.147, df = 159). However, fewer butterflies were observed on sunny days than on overcast or partly cloudy days (F = 3.44, P = 0.035, df = 159). The mean difference between observed-expected butterfly counts was -1.4 ( /- 0.11 SE) on sunny days, 0.04 ( /- 0.024 SE) on overcast days and 0.21 ( /- 0.064 SE) on partly cloudy days.

Seasonal flight phenologies differed in their onset and duration between sites and from year to year (Fig. 1). In four years of study, Neonympha mitchellii francisci emerged between May 18 and May 26 in the first flight period, and between Jul. 12 and Jul. 28 in the second flight period. On average, the time span between emergence of the first flight period and the second was 61 (4.8 s.d.) d. The date on which we recorded highest counts in the first flight period ranged between May 28 and Jun. 8, and in the second flight period between Jul. 24 and Aug. 10. On average, there were 61 (4.4. s.d) d between the peak of the first flight and the peak of the second flight. The first flight period lasted 19 (4.3 s.d.) d within a site and 24 (4.4 s.d.) d among all sites. The second flight period lasted 24 (5.4 s.d.) d within a site and 33 (4.2 s.d.) d among all sites.