Common knotweed, Polygonum aviculare L. (Polygonaceae), a summer annual occurring in agricultural and urban settings in the Sacramento Valley, was attended by numerous predatory and parasitic insects, many of which fed on the exposed floral nectar. Representatives of 36 insect taxa were observed feeding at the flowers; 29 of these groups contain entomophagous species. Other entomophagous insects were associated with a honeydew-producing, host-specific aphid, Aphis avicularis Hille Ris Lambers. Among predators frequently observed feeding at the flowers were bigeyed bugs, Geocoris spp.

Surveys indicated that common knotweed plants harbored higher densities of bigeyed bugs than did hay alfalfa (Medicago sativa L.), prostrate pigweed (Amaranthus graecizans L.), or field bindweed (Convolvulus arvensis L.). Geocoris punctipes (Say) and Collops vittatus (Say), two predators which are also nectarivorous, survived longer when caged on common knotweed as opposed to hay alfalfa, as did G. punctipes caged on common knotweed as opposed to prostrate pigweed. No difference in longevity was found when G. punctipes was caged on common knotweed with flowers and without flowers.

A further study featured replicated monocultural plots of common knotweed, common purslane (Portulaca oleracea L.), prostrate pigweed, and control plots of bare ground, arrayed in a completely randomized design. Common knotweed plots harbored the highest densities of bigeyed bugs, aphidophagous ladybeetles, and total numbers of predators. Tests of predation efficiency yielded mixed results. Predator discovery of simulated prey (dead vinegar flies glued to cards) was higher on common knotweed than on prostrate pigweed foliage. Similar studies employing egg masses of beet armyworm (Spodoptera exigua Hubner) yielded less definitive results, with rates of discovery by predators being higher on both common knotweed and prostrate pigweed than on common purslane, while the former two rates were not significantly different. Recruitment of predatory ants (Tetramorium caespitum L.) to tuna baits was not significantly different among plots containing the three weed species, despite high numbers of ants attending aphid colonies in the common knotweed plots.

In the next experiment, flats of radish (cv “Sparkler” and “Scarlet Globe,” Raphanus sativus L.), maintained amid wild stands of common knotweed harbored higher densities of bigeyed bugs than did flats in nearby stands of other low-growing weeds. No treatment effect on other predators was evident. Egg masses of an armyworm, Pseudaletia unipuncta (Hayworth) (Lepidoptera: Noctuidae), stapled to radish foliage, were damaged or occupied by predators in slightly greater proportions on radish grown amid common knotweed stands.

In another experiment, predator densities and efficiencies were assessed amid commercial organically-grown bell pepper (cv “Yolo Wonder,” Capsicum annuum [L.]) grown in plots with and without common knotweed. Although the weed harbored higher densities of bigeyed bugs than did bare ground afforded by control plots, visual inspection and shake sampling yielded no evidence for increased densities of predators at the immediate bases or on the foliage of pepper plants. On the other hand, studies employing armyworm egg masses showed enhanced predation by bigeyed bugs and other predators at the bases of pepper plants with adjoining common knotweed. Similar studies involving egg masses stapled to foliage of pepper plants failed to indicate a difference. Common knotweed apparently had no effect on crop vigor and yield.

The final experiment involved replicated plots of hay alfalfa (cv “Amador,” having borders of either common knotweed, prostrate pigweed, common purslane, or uncut alfalfa (control). Proportions of vegetational cover in plot borders differed among several weed regimes: knotweed and control plots featured denser cover than pigweed, and these three regimes exceeded purslane. Purslane borders were devastated by the combined action of herbivorous insects and a fungal pathogen, rendering that regime of little interest. Samples taken on the ground beneath border vegetational canopies yielded an effect due to weed regime for Geocoris densities, with pigweed exceeding knotweed, which in turn exceeded control. Data for total predators, including Geocoris, indicated that knotweed exceeded pigweed, which in turn exceeded control. Visual inspection of borders, which detected Geocoris running about on exposed ground, showed knotweed surpassing purslane and pigweed, which were indistinguishable, followed by control. Despite differences observed in borders, vacuum samples from alfalfa portions of plots indicated no effect by weed regime on densities of bigeyed bugs, Nabis spp., Lygus spp., or lycosid spiders. Similarly, counts of Geocoris taken at ground level amid alfalfa, both before and after mowing the crop, failed to show an effect. Based on shake samples, density of Geocoris was higher on pigweed than on knotweed, whereas the converse was true for Orius tristicolor (White). Nonetheless, based on sweepnetting, Orius densities in alfalfa were not different among the four weed regimes. Predation studies involving beet armyworm egg masses glued to apical foliage of alfalfa failed to show differences among weed regimes. Also, studies employing chunks of tuna placed on cards at ground level amid the alfalfa did not yield significant differences in rates of attendance by Geocoris or by pavement ant. Weed regime had no effect on yield of adjoining alfalfa.

Common knotweed, despite its attractiveness to numerous species of entomophagous insects, may have limited use in enhancing biological control on adjoining crops. The weed may provide alternate prey as well as floral resources, and thus represent such a hospitable habitat that some predators may have no tendency to forage on nearby crop plants. Thus, a weed originally selected to provide a complementary resource (nectar) may also inadvertently afford supplementary resources (alternate prey). Alternate prey could conceivably detract from biological control of pests on adjoining crops. Nonetheless, fieldside common knotweed appears to provide favorable habitat for various predators of agricultural importance. At this point, perhaps the weed should be tolerated in many settings, as it may provide breeding sites and reservoirs from which predators might colonize agricultural fields. Further research should involve multiyear studies on manipulation of common knotweed stands to influence microclimate, soil nutrient levels, and quality of the habitat. These could clarify whether this weed may yet be useful in the enhancement of biological control of insect pests.