Studies were conducted to determine the frequency, intensity, and disp
lacement of migratory flights of adult corn earworm, Helicoverpa zea (
Boddie), form the lower Rio Grande Valley of southern Texas and northe
astern Mexico during the citrus blooming period in February and March
1994. Samples of daily collections of adult male H. zea from pheromone
traps in the valley (10) and southern Texas (21) were examined for th
e presence of citrus pollen using scanning electron microscopy. H. zea
were captured at all locations except Alpine, TX. Based on samples of
the daily collections, citrus pollen-contaminated H, tea were capture
d as far as 661 km from the valley. However, the valley is the nearest
source of commercial citrus production. Insect flight trajectories fr
om the valley were estimated using wind velocity data at 500 m above g
round level and an assumed insect air speed of 5 m/s. Probability of d
aily capture of H. tea at locations that were 112-711 km from the vall
ey were significantly related to duration of the estimated insect flig
ht trajectory; minimum dr temperature at the trap location; and daily
average trap capture in the valley. The independent variables: duratio
n of estimated insect flight trajectory, duration of estimated insect
trap location, and the 1-d lag of minimum air temperature at the fligh
t trajectory were significantly related to the daily capture of citrus
pollen-contaminated H. tea. Wind trajectories measured by mylar ballo
ons (tetroons) and hourly weather data (i.e., wind, precipitation, and
air temperature) may explain some of the residual variance of the log
istic regressions. The logistic regression equations can be used to es
timate the probability of occurrence of daily capture of corn earworm
for use as an advisory to agriculturists, regulatory personnel, and re
search scientists.