Weeds may augment

biological control of insects

Miguel A. Altieri
Outbreaks of some insect pests are more likely to
occur in weed-free than in weed-diversified crops.
w e e d s have traditionally been considered
totally unwanted plants that reduce yields by
directly competing with crops or by harbor-
ing insect pests and plant diseases. Certain
weeds, however, can be regarded as impor-
tant components of agroecosystems, which
occidentulis. Similarly, in Georgia, soybean
plots with a dense cover of sicklepod (Cassia
obfusifoliu)were less susceptible to attack by
velvet bean caterpillar (Anticursiugemmata-
lis) and green stink bug (Nezuru viridulu)and
had more predators than weed-free plots.
!;
R 54
c

can complement existing insect pest manage- More importantly, these trends were still
36
ment systems. Outbreaks of some insect prevalent in soybean plots weeded for two to c
L
pests are more likely to occur in weed-free four weeks after soybean emergence, but al-
p" 18
than in weed-diversified crop systems. lowed to remain weedy thereafter. No further c
Moreover, crop fields with a dense weed weed removal after the two- to four-week 5
0
cover and high diversity usually have more period was necessary to obtain maximum 20
I
M Davs after40olantino
I
50
I
60
beneficial insects than do weed-free fields. yields. , "
Plants commonly considered weeds in many Weeds can also affect the field perfor- Effect of grass weed borders on leafhoppers
situations are considered desirable wild mance of insect parasites. For example, the in dry-bean fields in Colombia.
plants in this case. extent to which naturally occurring popula-
In spite of the obvious beneficial implica- tions of parasitic wasps (Trichogrammaspp.)
parasitized artificially placed corn earworm
47.8
___
-weedy
weed-free /

ii
tions of these findings for integrated pest
36.24-
management, implementation of weed control (Heliothis zeu) eggs on soybean plants de- (Y

programs compatible with insect manage- pended on the plant species associated with f
ment systems has not yet emerged. However,
work in Colombia (South America) and in
soybean. Higher egg parasitization was
recorded on soybean associated with Desmo-
.
R 28.68-
0)

the southeastern United States has provided dium sp. and Croton sp. than on soybean in- -2
=0
19.12-

encouraging results. Weed manipulation termingled with grasses or in soybean mono- B

m
could be a practical technique to augment cultures. Further studies indicated that the 5
z
9.56-

biological control of insects in California field behavior and efficiency of Trichogram- ,/

agroecosystems. mu spp. could also be manipulated by spray- 0-
50 57 M 71 76 85 92 99 106 llJ
ing crop plants with water extracts of several Days after planting
Experimental evidence plants. For example, parasitization of corn Abundance of velvet bean caterpillar in
In tropical Colombia, establishment of earworm eggs increased significantly in soy- weedy and weed-free Georgia soybeans.
borders of wild grasses (Eleusine indicu and bean, tomato, cowpeas, and cotton plots
Leptochloufiliformis) around small plots of sprayed with water extracts of corn and 17-
dry beans effectively reduced populations of pigweed ( Amurunthus sp.).
leafhoppers (Empouscu kruemeri), the main Weeds growing on crop field borders can
bean pest of the Latin American tropics. Fur-
ther tests indicated that these grasses exerted
also affect the dynamics of beneficial insect
communities in a particular crop system. In .
3
e
u)
n
10.2-
a repellent effect on the leafhoppers. Georgia, predator numbers in soybean rows B
Experiments in northern Florida corn declined sharply the farther the rows were 0
Y)

fields revealed that allowing a strip of native from a weedy ditchbank adjacent to the field. 2 6.8-
b
c
and selected wild plants to grow in every These findings suggest that manipulation of m
tenth row of corn substantially reduced the border vegetation can help maintain a natural E 3.4-

incidence of fall armyworm (Spodoptera balance of beneficial insects in a cultivated

fnrgiperda), and increased the number of
predators.
Leaving a ground cover of johnsongrass
or sudangrass in vineyards of the San Joa-
quin Valley, California, reduced densities of
the willamette mite, presumably by enhanc-
ing activity of a predaceous mite, Metaseiulus
22 CALIFORNIA AGRICULTURE, MAY-JUNE 1981
area. Encouragement of wild plants that pro-
vide shelter, pollen and nectar, or alternate
prey or hosts for predators and parasites
along field margins might prove beneficial.
The effectiveness of Anugrus epos, an egg
parasite of the grape leafhopper, was en-
hanced in California vineyards where patches
0- 1 1 1 1 1 1 l 1 1 1 1 1 1 1 1 1
1 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75
Number of rows into field from weed border
Predator numbers in Georgia soybeans with
increasing distance from weedy ditchbank.
Opposite: Encouraging grass borders can
reduce leafhopper densities in dry beans.
 Director
Agricultural Experiment Sfation
University of California POSTAGE PAID
Berkeley, California 94720 U.S. DEPARTMENT O F
AGRl CU L T U R E
PUBLICATION A G R 101 USMAJL
Penalty for Private Use $300

THIRD CLASS
BULK R A T E

Weeds and biological control, continued

TABLE 3. Parasitization of Corn Earworm Eggs (Heliothis zea) by Naturally
Occurring Populations of Trichograrnma spp. Wasps in Crops Sprayed
with Water Extract of Corn, Pigweed (Amaranthus sp.), or Water
Parasitization of corn earworm eggs*
Treatment Soybean Cowpeas Tomato Cotton
Q/O % % % new interdisciplinaryproject will be started in
Pigweed 21.4 a 45.4 a 24.3 a 13.6 a
Corn 17.4 b 45.8 a 21.1 a -
Water 12.6 c 31.6 b 17.6 b 4.2 b
'Meansfollowed bythesameletterineachcolumnarenotsignificantlydifferent(P = 005)
of wild blackberry (Rubus sp.) were pre- As these examples have indicated, surveys
served along field borders. of cropping systems, including coexisting
At times, special manipulations, such as weeds and associated insect communities,
periodic clipping, might be required to force can help determine relevant biological inter-
beneficial species to move from the borders actions and therefore suggest how agroeco-
into the crop. These practices should be systems could be structured to minimize pest
carefully timed, based on the biology of the incidence while augmenting natural controls.
beneficial species involved. For instance, the Tolerable levels of weeds that support popu-
annual cleanup of weeds along edges of alfal- lations of beneficial insects can be attained
fa fields should be delayed until after the ag- within fields by designing competitive crop
gregations of dormant lady beetles have mixture systems, cover crops, row spacing
largely dispersed. patterns, periods of weed-free maintenance,
mulching, and cultivation regimes. A series
of other cultural practices can be used to
manipulate the vegetation of crop field
borders.
With the active participation of farmers, a
northern California orchards and vegetable
and row crop fields. Weed research will be
concerned with the integrated components of
insect and disease management systems. Pos-
sibilities of exploiting wild vegetation will be
investigated in the hope that, ultimately,
farmers will benefit from the environmen-
tally safe, energy-conserving features of
diverse, pest-stable, self-sufficient agroeco-
systems.
Miguel A . Altieri is Assistant Professor and Assis-
tant Entomologist, University of California,
Berkeley. Research reported here was funded by
Centro Internacionalde Agricultura Tropical (Col-
ombia), Tall Timbers Research Station (Florida),
University of Florida, Soil and Health Foundation
(Pennsylvania), and University of Georgia.

Weed strips within Florida corn fields reduced fall armyworm infestations.