Jr. Fuxa et Ar. Richter, Classical biological control in an ephemeral crop habitat with Anticarsia gemmatalis nucleopolyhedrovirus, BIOCONTROL, 44(4), 1999, pp. 403-419
Anticarsia gemmatalis nucleopolyhedrovirus (AgNPV) was released as a single
spray in soybean at two sites in Louisiana, near Crowley and Baton Rouge,
after which viral prevalence and population density were monitored for 3-4
years. Each site had a plot with no treatment (control) and two virus-treat
ed plots, one planted with soybean yearly for 3-4 years, and the other plan
ted with soybean for 3-4 years except for rotation to a different crop in y
ear 2. In year 1, the single spray of AgNPV resulted in viral prevalence ra
tes ranging from 25-100% A. gemmatalis mortality over the entire growing se
ason. By the end of this season, viral accumulation in soil averaged 4.1 x
10(4) occlusion bodies (OB)/g at Crowley and 7.4 x 10(3) OB/g at Baton Roug
e, which had a sandier soil than Crowley. At Crowley, prevalence of AgNPV r
eached 49% insect mortality in the unrotated plot in year 2, and 31% in the
rotated plot and 38% in the unrotated plot in year 3, in spite of moderate
to low population densities of A. gemmatalis. At Baton Rouge, AgNPV preval
ence decreased to peak prevalence rates of 25% insect mortality in year 2,
4% in year 3, and 11% in year 4, even though A. gemmatalis population densi
ties were moderate in years 2 and 3. Viral concentration in soil decreased
to 2.7 x 10(2) OB/g (rotated plot) and 1.4 x 10(4) OB/g (unrotated plot) by
the end of year 3 at Crowley and to 10 OB/g (rotated plot) and 31 OB/g (un
rotated plot) by the end of year 4 at Baton Rouge. In forward stepwise mult
iple regressions, the concentration of OB in soil was significantly (p = 0.
0001) and positively correlated with AgNPV prevalence, but correlations of
the latter parameter with host population density and rainfall were not sig
nificant (p > 0.05). Prevalence rates of the fungal entomopathogen Nomuraea
rileyi were significantly (p < 0.01) and negatively correlated with AgNPV
prevalence at Crowley but not in the combined data set. Prevalence of N. ri
leyi was negatively correlated (p = 0.0001) with precipitation and positive
ly correlated (p = 0.0001) with A. gemmatalis population density. The resea
rch demonstrated that AgNPV can be introduced and established for long-term
suppression of A. gemmatalis in an ephemeral crop habitat, but certain sit
e-related conditions, perhaps relating to soil, are necessary for its conti
nued success.