Aeschynomene americana L. is a tropical pasture legume, and in Florida
, where it is usually grown with bahiagrass (Paspalum notatum Flugge),
there are about 5000 ha. Limited soil water is a primary reason for s
tand failure of this annual. Soil water in three bahiagrass canopy siz
es: small (2.2 leaf area index [ LAI]), medium (4.5 LAI) and large (7.
7 LAI) was measured at seeding and related to aeschynomene establishme
nt. All canopies were clipped to a 50-mm height when aeschynomene seed
lings emerged. Soil was Eau Gallie fine sand (sandy, siliceous, hypert
hermic Haplaquods). Aeschynomene was drilled into sod on 17 dates betw
een May and July in 1981 to 1985, and seedling density was determined
at emergence and periodically for 21 d. Six of the 17 seedings were fa
ilures because of drought (one because of flooding), and when failure
occurred, it was in all three canopy sizes. Water potential at 75- and
150-mm soil depths was analyzed from 209 dates, and canopy size affec
ted soil water potential on 41 dates. Gravimetric soil-water was analy
zed on 42 dates, and canopy size affected (P less-than-or-equal-to 0.0
5) gravimetric soil-water on five dates. Evapotranspiration (ET) rate
was measured twice and was not different among canopy sizes on 26 June
1984 (avg. 94 mg m-2 s-1). Differences (P less-than-or-equal-to 0.03)
in ET rate were found on 16 May 1986 when the large canopy (128 mg m-
2 s-1) had greater ET rate than the small canopy (100 mg m-2 s-1), wit
h the medium canopy (117 mg m-2 s-1) intermediate. Bahiagrass canopy s
ize before aeschynomene emergence had little effect on the amount of s
oil water and resulting success or failure of aeschynomene. Only rain
and soil water potential were useful for predicting aeschynomene stand
success. Together, rain 0 to 8 d before seeding [descriminant value (
DV) less-than-or-equal-to 22 mm] and soil water potential at seeding (
DV less-than-or-equal-to - 15 kPa) correctly predicted 13 out of 17 se
edings.