Some plants are more attractive to agricultural pests than the crop to
which they are a pest. The use of these plants as indicator plants to
monitor pest populations is examined. Bean plants were found to be ef
fective indicators for forecasting the growth of carmine spider mite,
Tetranychus cinnabarinus (Boisduval), on greenhouse tomatoes, Lycopers
icum esculentum L. Spider mite population growth rates were the same o
n both host plants, but because they became established up to 5 wk ear
lier on beans, Phaseolus vulgaris L., monitoring carmine spider mite p
opulation growth on beans provides enough time to order and distribute
natural enemies for their control on tomatoes. The oviposition and de
velopmental rates of mites did not differ on the 2 host plants. Howeve
r, the temperature threshold for oviposition was 7 degrees C lower on
beans than tomatoes, resulting in earlier establishment and onset of p
opulation growth in the spring on the indicator plants. The potential
for using indicator plants to establish biological control agents was
investigated with the predatory mite Phytoseiulus persimilis Athias-He
nriot. P. persimilis became established equally well when broadcast on
the crop (with its much lower prey population density) as when releas
ed only on infested indicator plants. Best control of carmine spider m
ites by Phytoseiulus was obtained when the predators were released whe
n spider mite density reached 12 per tomato leaf, which occurred when
the mite density on beans was 100 per leaf, 5 wk after reaching 12 mit
es per bean leaf. Acting on this threshold, a grower would have ample
time to order, receive, and release predatory mites. Predation of spid
er mites by Phytoseiulus was higher on the tomatoes than the beans. Ph
ytoseiulus populations dipped after introduction on both host plants b
ut recovered more rapidly on beans. The different predation and surviv
al rates on the 2 host plants probably served to keep the 2 population
cycles on the 2 hosts out of synchrony, thereby sustaining the popula
tion interactions.