diazinon 29.4%, and untreated check 46.6%. Although the
results show that insect control can reduce the rate of spread of
LY, losses of palms in treated areas were unacceptably high.
 Since Myndus crudus adults are highly active flying insects, it
would be difficult to maintain continuous protection of palms
with insecticides alone. Modifications of the described spray pro-
gram that could result in more effective protection of palms
might include more frequent spraying, higher dosages of the in-
secticides or more toxic or more persistent insecticides, spraying of
alternate hosts of the vector, and treatment of larger areas. Also,
more effective protection of a palm planting could be provided if
a spray program were started prior to the initial infection of the
planting. Incorporating all of these modifications, a hypothetical
2 ha coconut farm threatened by LY could be protected by begin-
ning weekly or perhaps daily applications of a highly toxic and
persistent insecticide to all palms and herbaceous plants within a
kilometer or so of the farm. The program would be continued
until there were no more cases of LY in the province -perhaps a
period of 100 years!
 As more is learned concerning the disease cycle of LY, the life
history and behavior ofM. crudus and as improvements are made
in insecticides and application methods, chemical control of the
vector of LY may become feasible. For now, it does not seem
promising.

Control by Natural Enemies
 In Florida, we have identified a number of natural enemies of
M. crudus, including spiders of several species (Howard and Ed-
wards, 1984), ants, lizards, and three frogs. A fungus, Hirsutella
citriformis Speare and parasitic mites, Leptus sp. and Erythaeus
sp. are occasionally found on M. crudus. These natural enemies
occur in LY-affected areas. Thus, under our conditions they have
been ineffective in preventing the spread of the disease. Myndus
crudus is regarded as native to the neotropics and is distributed
from the extreme southern United States through northern South
America (Kramer, 1979; Meyerdirk and Hart, 1982). Possibly, ef-
fective natural enemies may be present somewhere within this
geographical area.


Cultural Control and Host Plant Resistance
 Considering the large number of palm species visited by M.
 crudus (Howard and Mead, 1980) and the abundance of these in-
 sects on several varieties of coconut palm in Florida, it seems
 doubtful that varieties of coconut can be found that are not at-
 tractive to this insect. Still, the host plant relationships of M.
 crudus are known superficially at best. A coconut palm variety
 that is resistant to this insect may one day be discovered.
 At the present, we are focusing our attention on the grass hosts
 of M. crudus. The females of this insect lay their eggs on grasses,
 and the nymphs develop in the root zones (Fig. 1). Adults are
 found on grasses and on palms (Tsai and Kirsch, 1978; Zenner de
 Polania and Lopez, 1977). We have examined roots of many
 palms without finding M. crudus. Assuming that grasses are a
 vital link in the life cycle of M. crudus, it is theoretically possible
 to eliminate this insect from an area by eliminating its grass hosts.
 Grass or other ground cover is desirable in a coconut planting
 for erosion control or where cattle are reared on the same farm


FIGURE 1: Myndus crudus nymph on stolon of St. Augustine grass.


Also, keeping a farm and its adjacent areas free of herbaceous
vegetation would be overly costly.
 Progress has been made in developing ground cover manage-
ment methods that reduce the infestation of peach orchards by
leafhopper vectors (McClure et al., 1982). This approach should
be investigated as a possible means of controlling LY.
 Eden-Green (1978) reported that St. Augustine grass,
Stenotaphrum secundatum (Walt.) Kuntze, was the most
reliable grass for rearing M. crudus for experiments. Reinert
(1980) reported that higher numbers of adults were collected in
sweep net samples from St. Augustine grass than from Bermuda
grass, Cynodon dactylon Persoon, or Bahia grass, Paspalum
notatum Fluegge. In Florida, high incidence of LY coincides both
with high populations of M. crudus and extensive use of St.
Augustine grass as a turf (Howard, 1980).
 It is thus disconcerting that St. Augustine grass is recommend-
ed in islands of the Pacific as a forage grass compatible with the
coconut-cattle farming system (Smith and Whiteman, 1983). The
extensive use of St. Augustine grass on coconut farms should be
avoided since this could increase the risk of an LY epidemic
should this disease be introduced. But we don't yet know which
other grasses or dicotyledenous plants would reduce this risk.
 A more complete knowledge of the host plant relationships of
M. crudus nymphs is needed. Can a grass be found that is
tolerant of the shade of coconut canopies, that is unacceptable as
a host of M. crudus, and aggressive in competing with grasses that
do serve as hosts? Leguminous and other dicotyledenous ground
covers should also be investigated, since they are unlikely to serve
as hosts of M. crudus.
 In addition, we need to develop knowledge of the flight
behavior of M. crudus. The minimum area over which ground
cover management must be practiced in order to be effective in
controlling LY would be determined largely by the distances that
adult M. crudus fly from their nymphal hosts to palms.
 Most insect-vectored plant diseases are not successfully con-
trolled by controlling the vector. Nevertheless, the possibility of
controlling LY by management of the nymphal hosts of M.
crudus seems promising, and warrants research efforts.


Acknowledgements
 I thank Mr. J.V. DeFilippis, Agricultural Research Technician, for providing Fig.
 1, and Dr. Ted Center, U.S. Department of Agriculture, for reviewing the
manuscript.

References
 1. Beakbane, A.B., C.W.H. Slater, and A.F. Posnette. 1972. Mycoplasmas in
the phloem of coconut, Cocos nucifera L., with lethal yellowing disease (abstract). J.
Hortic. Sci. 47:265.

VOL. XX-PROCEEDINGS of the CARIBBEAN FOOD CROPS SOCIETY


 2. Dabek, A.J. 1975. The incubation period, rate of transmission and effect of
growth on lethal yellowing disease in Jamaica. Phytopathol. Z. 84:1-9.
 3. Eden-Green, S.J. 1978. Rearing and transmission techniques for Haplaxius
sp. (Hom. Cixiidae), a suspected vector of lethal yellowing disease of coconuts. Ann.
Appl. Biol. 89:173-176.
 4. Heinze, K., H. Petzold, and R. Marwitz. 1972. Beitrag zur Atiologie der
Todlichen Vergilbung der Kokospalme. Phytopathol. Z. 74:230-237.
 5. Harries, H.C. 1970. The Malayan Dwarf supercedes the Jamaican Tall
coconut. 1. Reputation and performance. Oleagineux 25:527-531.


145