Effects of Soybean Row Spacing on Spray Penetration and

Efficacy of Insecticides Applied with

Aerial and Ground Equipment

SCOTT H. HUTCHINS AND HENRY N. PITRE

Department of Entomology, MississippiAgricultural and Forestry Experiment Station,
MississippiState, Mississippi39762

Downloaded from https://academic.oup.com/ee/article-abstract/13/4/948/2393306 by ESA Member Access user on 01 November 2019

Environ.Entomol.13: 948-953 (1984)
ABSTRACT Kromecote® or water-sensitive cards were placed at three or four vertical
levels within the canopy of wide (96.5-101.1 cm) and narrow (17.8-25.4 cm) rows of soy-
beans, Glycine max (L.) Merrill, to measure spray penetration and droplet characteristics
of insecticides applied with aerial or ground equipment. Larval mortality data were used
to measure the efficacy of the insecticides at three vertical strata within the two row-spacing
systems. The size of the droplets deposited was generally smaller within all levels of the
narrow-row canopy compared to the wide-row canopy. The total number of drops per cm2
and mean percentage of coverage was reduced in the narrow-row spacing compared to the
wide-row spacing, but only within the median one-third of the canopy. Similarly, the per-
centage of larval mortality was reduced within the median one-third of the narrow-row
plots compared to the wide-row plots at the same level. The variable combination of drops
per cm2 and number median diameter was significantly correlated (R = 0.6535) with the
percentage of reductions of lepidopterous larval populations of the soybean looper, Pseu-
doplusia inc/udens (Walker), velvetbean caterpillar, Anticarsia gemmatalis Hubner, and
green cloverworm, Plathypena scabra (Fab.).

INCREASEDINTERESTIN double cropping and ero- etration and biological effectiveness of the insec-
sion control, combined with increased fuel costs, ticide when applied to each row-spacing system.
has created an agronomic trend toward narrow-
row production of soybeans, Glycine max (L.)
Materials and Methods
Merrill. Many arthropods tend to prefer the more
dense narrow-row soybean plantings (Sprenkel et Field studies were conducted in 1981 and 1982
al. 1979, Mayse 1978, Buschman et al. 1981). Un- to determine if the dense foliage canopy of nar-
derstanding existing interactions between pest row-row soybean plantings adversely affects the
density and row spacing is essential for effective physical penetration and efficacy of selected insec-
insect pest management. ticides.
The complex of beneficial organisms may in 1981 Field Study. 'Bragg' soybeans were plant-
many instances hold pest populations on soybeans ed in two row-spacings within a 0.2 ha block on
in check (Shepard et al. 1977), but if economically the Mississippi Agricultural and Forestry Experi-
damaging infestations develop, then proper insect ment Station Plant Science Research Farm near
and crop management tactics must be employed. Starkville, Miss. The wide-row (96.5 cm) plots con-
This often requires the application of an insecti- sisted of four rows 12.2 m long, and the narrow-
cide for immediate pest suppression. Chemical in- row (17.8 cm) plots encompassed an equal area.
secticides that are highly effective in controlling The narrow rows were planted with a 17.8 cm
pest outbreaks in conventional wide-row soybean grain drill planter. Recommended crop manage-
plantings may not be as effective when applied to ment practices (Anonymous 1982) for this area
the more dense canopy of narrow-row plantings. were used and all experimental plots were man-
Therefore, there is a need for accurate estimation ually weeded to prevent erroneous interpretation
of insecticide penetration into the foliage canopies of penetration data. The experimental design was
of both wide- and narrow-row soybean plantings. a three-way split-plot design with whole plots (row
This study was designed to investigate differ- spacings) confounded and subplots (insecticides)
ences in pesticide spray penetration and biological arranged in a randomized complete block with
effectiveness between wide- and narrow-row soy- four replications for each treatment. The data were
bean plantings. The objectives were: (a) to identify subjected to an analysis of variance and Duncan's
differences in insecticide efficacy at three vertical multiple range test.
levels within each of the two row-spacing systems; Two synthetic insecticides, methyl parathion
(b) to quantify differences in spray deposits within (1.14 kg [AI]/ha) and methomyl (0.51 kg [AI]/ha),
and between the two row-spacing systems; and (c) and one biological insecticide, Bacillus thurin-
to determine the relationship between spray pen- giensis Berliner val. kurstaki (Bt, 1.14 kg formu-

948
August 1984 HUTCHINS AND PITRE: INSECTICIDE EFFECTIVENESS IN SOYBEANS 949

lation/ha)J were applied in water at a dosage of ments) were arranged in a completely random
94.5 liters of final mixture per ha and at 2.23 kg/ manner with four replications for each treatment.
cm2 pressure using a tractor-driven boom sprayer Water-sensitive' rather than Kromecote cards were
with TX-4 hollow cone nozzles spaced 45.7 cm placed at four vertical positions (sub-subplots)
apart. The plants were in the R-4 (full pod set, within each subplot to record spray deposition and
Fehr et al. 1971) stage of development at the time characteristics at terminal (ca. 120 cm), median
of application and were approximately 135 cm in (ca. 81 cm), lower (ca. 42 em), and soil (ca. 3 cm)

Downloaded from https://academic.oup.com/ee/article-abstract/13/4/948/2393306 by ESA Member Access user on 01 November 2019

height. The spray boom was 45.7 cm above the levels for each row-spacing.
foliage canopy. Methyl parathion (0.28 kg [AI]/ha) was applied
Soybean looper, Pseudoplusia includens Walk- with a fixed-wing Cessna 188-B aircraft in 11.4
er, populations were sampled 1 day pretreatment liters of water /ha using D-8 orifice disks and no.
and 1 and 2 days after application. A I-m2 ground 45 cores over a 15.3-m spray swath (across rows)
cloth was used to sample the insect population in covering both wide- and narrow-row plots. Plants
the wide-row planting (Hillhouse and Pitre 1974). were in the R-4 stage of development at the time
It was difficult to use a ground cloth in the narrow- of application.
row plots without disturbing the arthropod popu- A modified sampling technique was designed
lations before sampling. Therefore, these plots were and utilized to sample defoliating lepidopterous
sampled with a 38-cm sweepnet (Kogan and Pitre larvae (soybean looper, velvetbean caterpillar, An-
1980) using 12 sweeps per sample through the ttcarsta gemmatalts Hiibner, and green clover-
middle of each plot. Differences in sampling pro- worm, Plathypena scahra [F.J) as a pest complex
cedures were accounted for by converting the on days 2, 5, 10, and 25 after application of the
number of live larvae sampled to the percentage insecticide. A I-m2 board, slotted and fitted with
of larval reduction based on samples from the un- rubber gaskets, was placed around the sample
treated control. plants. The main stem of each soybean plant was
Before the application of insecticides, three clipped at each card level (except terminal) and
Kromecote®2 cards were attached to a pole with the vegetative and reproductive plant parts above
alligator clips and positioned in an intermediate the sample board were carefully placed on a
row of each subplot replicate. Each card (7.6 by ground cloth where the insects were identified and
12.7 cm) was evenly spaced and positioned hori- recorded. Data obtained were analyzed as in the
zontal with the soil surface at a different vertical 1981 study.
level (terminal ca. 135 cm, median ca. 90 cm, low-
er ca. 45 cm) within each canopy to monitor spray
Results and Discussion
penetration. Ag-Mark R-2 purple dye (Huitink
1980) was added to the spray mixture (1.25 ml 1981 Field Study. Pretreatment samples taken
dye/liter mixture) so that the spray pattern could 1 day before application indicated uniform pop-
be detected on the cards. The cards were stored ulations of the soybean looper throughout the test
for analysis with an image analyzer". area. Comparisons of insecticide effectiveness
1982 Field Study. Based on results obtained in within the wide-row planting showed methomyl
1981, techniques were modified to identify the lo- to be more effective against the soybean looper
cation within the two soybean canopies where spray than either methyl parathion or Bt when applied
penetration and insecticidal activity was effected. with a boom sprayer (Table 1). Methomyl was more
'Coker 338' soybeans were planted on 30 June effective than the other two insecticide treatments
in two row-spacings within a large commercial field within the narrow row planting; whereas Bt and
at Wiggins, Miss. The designated sampling area methyl parathion treatments had equal effects on
for each wide-row (101.6 cm) plot consisted of looper populations. Because of the confounded ex-
eight rows 15.3 m long; narrow-row plots were perimental design, direct comparisons between
sampled from an equal area planted with a 25.4 row-spacings could not be made (Cochran and Cox
cm grain drill planter. To avoid drifting of spray 1957). However, the relative effects of row-spac-
into untreated plots, treatment swaths were sepa- ing (analysis within row-spacing) on insecticide ef-
rated by a planted buffer zone 13.2 m wide. Con- ficacy in terms of the percentage of reduction of
ventional soybean production practices for this area the larval population indicate that the effective-
of the state were employed (Anonymous 1982), ness of the two synthetic insecticides was some-
and, as in 1981, weeds were manually removed what reduced in the narrow-row soybean planting.
from the test plots to eliminate interference with The opposite response was observed for the bio-
spray penetration. The experimental design was logical insecticide (Bt). These results suggest rela-
expanded to a four-way split-plot design. Whole- tive differences between classes of insecticides
plots (row spacing) and subplots (insecticide treat- within two soybean row-spacing systems, particu-
larly with regard to one pest species, the soybean
I Wettable powder formulation containing 16,000 international looper. This observation should be studied further.
units/mg. The half-life of Bt spores on soybean foliage,
• Kromecote paper manufactured by Champion Papers.
'Image analyzer manufactured by Optomax, Inc. of Hollis,
N.H. • Water-sensitive cards manufactured by Ciba-Ceigy Corp.
950 ENVIRONMENTAL ENTOMOLOGY Yol. 13, no. 4

Table l. Effect of insecticides on the soybean looper, Pseudoplusia inc/udens (Walker), in wide-row (96.5 cm)
and narrow-row (17.8 cm) 'Bragg' soybean plantings

% Reduction in larvae
Insecticide
Row- compared to untreated
spacing Dosage (kg
Type Form. [All/ha) 1 Day 2 Days

Downloaded from https://academic.oup.com/ee/article-abstract/13/4/948/2393306 by ESA Member Access user on 01 November 2019

Wide Methomyl 0.21 L 0.51 74.9a 90.6a
(96.5 cm) Methyl parathion 0.47 E 1.14 51.0b 56.4b
Bt" 3.2 WI' 1.14 20.9c 4.9c
Narrow Methomyl 0.21 L 0.51 66.8a 64.4a
(17.8 cm) Methyl parathion 0.47 E 1.14 29.5b 28.1b
Bt 3.2 WI' 1.14 22.3b 32.2b

Means, within the same column and row-spacing, followed by the same letters are not significantly different (P < 0.05; Duncan's
multiple range test).
• Bt (Bacillus thuringiensis Berliner var. kUTstaki) treatment is expressed in terms of kilograms of formulation which contains
16,000 international units/mg.

exposed to either natural or artificial sunlight, has 1.22% for wide and narrow rows, respectively)
been shown to be less than 24 h (Ignoffo et at when compared to the median level. There was
1973). Raun et al. (1966) demonstrated that ultra- significantly less coverage at each consecutively
violet irradiation of a spore powder of Bt for 72 h lower level within the wide-row spacing, whereas
eliminated its pathogenicity to Ostrinia nuhilalis the median and lower level cards received similar
(Hubner) and Spodoptera frugiperda (]. E. Smith). spray coverage in the narrow-row planting.
The reduced UY-light penetration into the dense The total number of drops per cm2 was also
narrow-row canopy may act to extend Bt spore helpful in making general comparisons of insecti-
viability and result in a prolonged period of con- cide penetration differences between the two row
trol. Before this can be substantiated, however, systems. General trends for drop density within
further studies with more intensive sampling are the two row-spacing systems were proportionately
required to evaluate the long-range effect of Bt on similar to trends in the mean percentage of cov-
pest populations within and between soybean row- erage (Table 2). These drop-density trends agree
spacing systems. with studies by Bouse (1969) who, although work-
Based on data from Kromecote cards placed ing with a different crop, reported that approxi-
within the foliage canopies, there were differences mately 80% of the spray drops deposited were in-
in spray coverage between vertical levels within tercepted by the upper one-half of an oak canopy.
each row-spacing system and relative differences 1982 Field Study. Significant differences (P :=;;
between row-spacing systems; but, there were no 0.05) in the control of lepidopterous larvae be-
differences in penetration between insecticide for- tween wide-row and narrow-row soybeans over all
mulations. There was identical spray coverage (avg. sample dates and vertical levels within the soybean
37.97%) on cards located at the terminal position canopy are presented in Fig. lA. No differences
of both row plantings. We anticipated this because were observed between spacings on days 2 or 25
there was no foliage interference at this level. At after application, but on days 5 and 10 after ap-
the intermediate card position differences in the plication, insect mortality was greatest in the wide-
amount of spray coverage between the two sys- row system. The relatively late (day 5) expression
tems (9.24 and 2.62% for wide and narrow rows, of differences in mortality between wide and nar-
respectively) was apparent. At the lower level, row rows may be a function of the pest complex
however, there was a relatively small difference in sampled. For example, the soybean looper is com-
spray coverage between row spacings (1.87 and paratively tolerant of methyl parathion (Chiu and
Bass 1978), and would require a greater exposure
period before the overall pest population was re-
Table 2. Total number of drops per cm2 of insecticide duced. Additionally, pretreatment samples indi-
deposited on Kromecote cards at vertical levels within cated that each species was represented by a range
wide-row (96.5 cm) and narrow-row (17.8 cm) 'Bragg' of larval stadia. It seems reasonable that the larger
soybean plantings
stages would require a longer exposure time, which
Row-spacing would delay mortality. The degree of larval mor-
Card
position tality within each row-spacing system remained
96.5 cm 17.8 cm
relatively constant over the first three postappli-
Terminal level 150.17a 144.03a cation intervals. This would indicate that methyl
Median level 95.62b 38.88b
parathion reached its maximum effectiveness
Lower level 28.21c 24.07b
within 2 days of application and that a significant
Means, within a column, followed by the same letters aTe not recolonization of the experimental plots did not
significantly different (P < 0.05; Duncan's multiple range test~. occur until 10 days after application. An alterna-
August 1984 HUTCHINS AND PITRE: INSECTICIDE EFFECTIVENESS IN SOYBEANS 951

100 a a
C\I 33.0
..l
~ a eu a~ , • .WIDE
::>
0:0
Z 80
a
• WIDE
27.0 ,
~~ 60 b b ~ NARRCM'
0:
1&.1
a..
\ ,, .••.----- •••.NARROW
I-a
ZI&.I 40 A II)
21.0
\\ A
~o: a.. 15.0
20 0

Downloaded from https://academic.oup.com/ee/article-abstract/13/4/948/2393306 by ESA Member Access user on 01 November 2019

0:
~ C
9.0 ~ ...
2 5
10 ..•.....• d
0 ......
DAYS POST APPlCATION Z 3.0 d'A---
UPPER MEDIAN LOWER
VERTICAL LEVEL
• WIDE
C
1&.1
E;INARROW
ffi 5.00

B ~
0 4.00
a.o\ ,
, .••.---- ...•••NARROW
cs: \\
1&.1 3.00
~ \\ B
I-
2.00 ,
\
Z
100 1&.1 c.•••......
..l 0
0:
1.00 .•....
• WIDE
~~ 80 1&.1
a.. ~-----
~t= 60 f::iaNARROW
UPPER MEDIAN LOWER SOIL
0 ab
I-a
zl&.I
40
C
VERTICAL LEVEL

~o: Fig. 2. (A) Total number of drops per em', (B) Per-
0: 20
centage of area covered by insecticide deposited on
Ie water-sensitive cards at four vertical levels within wide-
row (101.6 cm) and narrow-row (25.6 cm) soybean
plantings. Means followed by the same letters are not
significantly different (P < 0.05; Duncan's multiple range
Fig. 1. Effect of wide-row (101.6 cm) and narrow- test).
row (25.4 cm) soybean spacing on the percentage of
reduction of lepidopterous larvae (soybean looper, vel-
vetbean caterpillar, and green c1overworm) treated with
methyl parathion (0.28 kg [AI]/ha). (A) Over all sample deposited on cards at the median level in the wide-
dates and vertical levels within the canopy and at each row spacing than in the narrow-row spacing (Fig.
of three postapplication intervals. (B) At each of three 2A). Similarly, more area was covered at the me-
vertical levels (upper, median, lower) within the canopy
dian level of the wide-row spacing system than at
on day 2 after application. (C) On day 5 after applica-
ton. Means within A, B, and C followed by the same the same level of the narrow-row planting (Fig.
(analysis within row spacing) letters are not significantly 2B). The decrease in droplet density and coverage
different (P < 0.05; Duncan's multiple range test). may be attributed to interception of the spray by
the dense canopy within the narrow-row system.
Smith and Burt (1970) observed similar effects with
tive explanation might be that the rate of recolo- cotton foliage. They noted that spray interception
nization equaled (or nearly equaled) the rate of at a given level directly affects the number of
larval decline, thus resulting in no net increase or droplets deposited at each lower level. The reduc-
decrease in larval mortality within the 2- to 10- tion in spray deposition at the median level of the
day sampling interval. Larval mortality was less narrow-row system may explain the concomitant
within the median level of the canopy of the nar- reduction in larval mortality. This seems plausible
row-row system on days 2 and 5 after application when considering studies by Fisher and Menzies
than at this same level within the wide-row system (1976), who reported that the number of drops per
(Fig. 1 B and C). Apparently the dense foliage cm2 and percentage of area covered are inversely
canopy inherent in narrow-row plantings limited related to the time required for oriental fruit moth,
droplet deposition in the median one-third of the Grapholitha molesta (Busck), neonate larvae to re-
canopy and thus reduced larval mortality at that ceive a lethal dose of carbaryl.
height. Droplet size is commonly expressed as either
Data obtained from the water-sensitive cards volume median diameter (VMO) or number me-
show differences between the two row-spacing sys- dian diameter (NMD) (Qoutts and Yates 1968,
tems with respect to droplet density, spray cover- Shankland and Tucker 1980). Using a water spread
age, and droplet sizes. More spray droplets were factor of two and information on droplet size and
952 ENVIRONMENTAL ENTOMOLOGY Vol. 13, no. 4

Table 3. Volume median diameter (VMD, /ol) and number median diameter (NMD, II) of insecticide deposited on
water-sensitive cards at four vertical levels within wide-row (101.6 cm) and narrow-row (25.3 cm) soybean plantings

VMD NMD
Card
Row-spacing Row-spacing
position
101.6 cm 25.4 cm 101.6 cm 25.4 cm
321.0bc

Downloaded from https://academic.oup.com/ee/article-abstract/13/4/948/2393306 by ESA Member Access user on 01 November 2019

Terminal level 768.Oa 710.0ab 295.Ocd
Median level 768.0a 581.0cd 358.0ab 271.0d
Lower levcl 728.Oa 531.Od 355.Oab 261.Od
Soil level 635.0bc 560.Ocd 372.Oa 268.Od

Means. within VMD and NMD, followed by the same letter are not significantly different (P < 0.05; Duncan's multiple range test).

density as provided by the image analyzer, VMD let increases as the cube of the diameter. The higher
was calculated to be the drop diameter (IL) at which relative humidity within the lower portions of the
one-half of the total spray volume is in larger drops narrow-row canopy may act to reduce the evap-
and one-half is in smaller drops. Similarly, but orative effects on drop volume and explain the
without regard to drop volume, NMD was calcu- fairly constant VMD and NMD throughout this
lated to be the diameter at which one-half of the row-spacing system.
drops are larger and one-half are smaller. The When efficacy and penetration data are com-
VMD and NMD values at the terminal level (no bined in regression analysis (Neter and Wasser-
foliage interference) did not differ significantly man 1974), existing correlations between two or
between row-spacing systems, which indicates that more parameters can be determined. We used the
a uniform application was made over all plots (Ta- linear least squares selection technique (Daniel and
ble 3). The VMD and NMD were smaller at the Wood 1980) to determine which measures of
median and lower levels within the narrow-row droplet penetration (drops per cm2 or avg. per-
spacing than in the wide-row spacing (Table 3). centage of coverage) and droplet size (VMD or
The narrow-row canopy apparently intercepted the NMD) could account for the highest degree of
majority of large drops before they could pene- variation in larval mortality at 2 days after treat-
trate into the lower levels of the foliage. ment. The combined effects of drops per cm2 and
Akesson and Yates (1964) reported that drops of NMD were significantly correlated (H = 0.6535)
volatile substances decrease in size through evap- with larval population reduction, and together they
oration, and this size reduction can have an im- accounted for approximately 43% of the inherent
portant effect on the distance a drop will travel variation (Table 4). While much of the variation
under field conditions. This evaporation is be- is accounted for by these variables, much more is
lieved to be responsible for the smaller VMD val- not. Therefore, other factors that were not ac-
ues obtained at the soil level compared to the ter- counted for in this study are affecting the pene-
minallevel of the wide-row planting; as the volume tration/mortality relationship. Evidence for this
of each drop decreases, the VMD is reduced. Con- can be seen when qualitative variables that indi-
versely, the NMD at the soil level is larger than at cate row-spacing (11), vertical level within the can-
the terminal level because smaller droplets evap- opy (12' Is), and interactions between row-spacing
orate at a faster rate; that is, the volume of a drop- and vertical level (14, Is) are fixed into the regres-
sion model. These indicator variables appear to
account for the gross differences between row-
spacing and significantly (P < 0.05) increase the
Table 4. Results of multiple regression analyses of the
predictive power of the regression model (H2 =
percentage of larval mortality at 2 days after treatment
(Y) versus dill erent independent variables representing 0.7198). Further experiments should be conducted
droplet density (drops per cm2, Xl) and droplet size to identify the biotic or abiotic conditions which
(number median diameter, X2), without (A) and (8) fixed affect response of larvae to insecticides for each
indicator variable (I.-Is)
row-spacing system.
Mod- Wide-row soybean systems allowed for superior
Regression equations (SE)a control of the defoliating pests considered in this
el
study, but only within the median one-third of the
A Y = -18.23 + 2.20X. + 0.26X2 0.4271
canopy. The wide-row foliage canopy hindered
B Y = 117.26 + 76.6Oftb - 0.8312 + 3.5713 0.7198 spray penetration less than the narrow-row plant-
- 53.8414 - 8.9315 + 5.46XI - 0.32X2
(7.03) (0.19) ing. The variable combination of drops per cm2
and NMD accounted for a significant portion (R2 =
a Standard error of the regression coefficients are designated 0.4271) of the variation affecting the mortality of
parenthetically for independent variables. larvae of the soybean looper, green cloverworm,
b Variables II through Is represent indicator variables: It, 1 for
wide-row spacing and 0 for narrow-row spacing; 12. 1 for upper
and velvetbean caterpillar. In situations where in-
level and 0 for median and lower levels; h 1 for median level sect pests are limiting to soybean production and
and 0 for upper and lower levels; 14 = II x 12; Is = I. x 13' where chemical control is anticipated, wide-row
August 1984 HUTCHINS AND PITRE: INSECTICIDE EFFECTIVENESS IN SOYBEANS 953

spacing (planting) should be considered to facili- mobilization of newly-hatched oriental fruit moth
tate the efficient use of insecticides. larvae. J. Econ. Entomol. 69: 438.
Hillhouse, T., and H. N. Pitre. 1974. Comparison of
sampling techniques to obtain measurements of in-
sect populations on soybean. Ibid. 67: 411-414.
Acknowledgment Huitink, G. 1980. Pattern your Ag spray plane. Coop.
Ext. Ser., Univ. Ark., Bull. MP 183.
We thank Edward Howe for his generous help with IgnofFo, C. M., D. L. Hostetter, and R. E. Pinnell.

Downloaded from https://academic.oup.com/ee/article-abstract/13/4/948/2393306 by ESA Member Access user on 01 November 2019

land and equipment, Bill Kitten for his help and exper- 1973. Stability of Bacillus thuTingiensts and Bacu-
tise with the image analyzer, and W. J. Drapala, H. W. lOviTUSheliothis on soybean foliage. Environ. Ento-
Chambers, R. G. Luttrell, and A. W. Cole for their re- mol. 3: 117-119.
views of the manuscript. This is Miss. Agric. and For. Kogan, M., and H. N. Pitre, Jr. 1980. General sam-
Exp. Stn. publication no. 5614. pling methods for above-ground populations of soy-
bean arthropods, pp. 30-60. In M. Kogan and D. C.
Herzog [eds.], Sampling methods in soybean ento-
References Cited mology. Springer-Verlag, New York.
Mayse, M. A. 1978. Effects of spacing between rows
Anonymous. 1982. Agronomy handbook. Mississippi on soybean arthropod populations. J. Appl. Ecol. 15:
Cooperative Extension Service. 439-450.
Akesson, N. B., and W. E. Yates. 1964. Problems Neter, J., and W. Wasserman. 1974. Applied linear
relating to application of agricultural chemicals and statistical models. Richard D. Erwin, Homewood, III.
resulting drift residues. Annu. Rev. Entomol. 9: 285- Raun, E. S., G. R. Sutter, and M. A. Revelo. 1966.
318. Ecological factors affecting the pathogenicity of Ba-
Bouse, L. F. 1969. Aerial-spray penetration through cillus thuTingiensis var. thuTingiensis to the Euro-
foliage canopies. Trans. A.S.A.E. 12: 86. pean corn borer and fall armyworm. J. Invertebr.
Buschman, L. L., H. N. Pitre, and H. F. Hodges. Pathol. 8: 365-367.
1981. Soybean cultural practices: effects on popu- Shankland, D. L., and J. Tucker. 1980. Aerial ap-
lations of green c1overworm, velvetbean caterpillar, plicaton of pesticide sprays (basic and applied con-
loopers and He/iothts complex. Environ. Entomol. siderations). Miss. Agric. and For. Exp. Stn. Tech.
10: 631-641. Bull. No. 104.
Chiu, P. S.-B., and M. H. Bass. 1978. Soybean loop- Shepard, M., G. R. Carner, and S. G. Turnipseed.
er: minimum rates of insecticides for control. J. Ga. 1977. Colonization and resurgence of insect pests
Entomol. Soc. 13: 151-155. of soybean in response to insecticides and field iso-
Cochran, W. G., and G. M. Cox. 1957. Experimental lation. Environ. Entomol. 6: 501-506.
designs. John Wiley and Sons, New York. Smith, D. B., and E. C. Burt. 1970. Effects of the
Coutts, H. H., and W. E. Yates. 1968. Analysis of size of UL V droplets on deposits within cotton fo-
spray droplet distributions from agricultural aircraft. liage both inside and immediately downwind from
Trans. A.S.A.E. 11: 25. a treated swath. J. Econ. Entomol. 63: 1400-1405.
Daniel, c., and F. B. Wood. 1980. Fitting equations Sprenkel, R. K., W. M. Brooks, J. W. van Duyn, and
to data. John Wiley and Sons, New York. L. L. Dietz. 1979. The effects of the cultural vari-
Fehr, W. R., C. E. Caviness, D. T. Burmood, and J. S. ables on the incidence of NomUTaea Tileyi on phy-
Pennington. 1971. Stage of development descrip- tophagous Lepidoptera and their predators on soy-
tions for soybeans, Glycine max (L.) Merrill. Crop beans. Environ. Entomol. 8: 334-339.
Sci. 11: 929-930.
Fisher, R. W., and D. R. Menzies. 1976. Effect of Received fOT publication 2 NovembeT 1983; accepted
spray droplet density and exposure time on the im- 2 ApTil 1984.