Yellow starthistle (Centaurea solstitialis L.) is an introduced Asteraceae that has become established on 10 million acres in the Pacific Northwest and California. This weed functions as an annual or short-lived perennial and depends on seeds for reproduction. Strategies of control that reduce plant fitness or lower seed production or viability may help limit the rate of spread of yellow starthistle. Previous work has shown that grazing and mowing can influence seed production. This study tested the hypothesis that proper timing and frequency of defoliation can reduce the number and viability of seeds produced. The study was conducted in Umatilla County, Oregon using a randomized block design with 4 replications of each of 4 defoliation treatments: (1) single defoliation at the bolting stage; (2) single defoliation at the bud stage; (3) two defoliations, once at the bolting stage and again at the bud stage; (4) non-defoliated control. Each of 4 blocks consisted of a 12 x 12 m area, with 16 plots measuring 3 x 3 m. Plants were defoliated at ground level using a gas-powered string-type mower. Response measurements were collected at the end of the growing season (September) following potential regrowth and included: (1) number of seedheads per plant; (2) number of seeds per seedhead; (3) number of seeds per plant; (4) number of seeds m⁻², (5) seed viability (% germination rates). Supporting measurements included: seedhead diameter; plant height, number of branches per plant; pre-dawn xylem pressure; soil moisture; and documentation of 5 biological control insect species. A single defoliation at bolting resulted in fewer seeds per seedhead, and fewer seeds per plant than non-defoliated controls. A single defoliation at the floral bud stage or repeated defoliation (bolting and again at the bud stage) resulted in equally fewer seeds per plant and fewer seeds m⁻² compared to non-defoliated controls. There was no statistical difference in percent germination of seeds among treatments. Defoliation had no effect on the infestation rates of seedheads by biological control insects. A second study examined nutrient content of yellow starthistle during 6 phenological stages from sites in Union, Baker and Umatilla Counties, Oregon during each of 2 years. Acid detergent fiber, lignin, cellulose and neutral detergent fiber contents increased through phenological development. Crude protein ranged from 16.7 to 5.0%. In Vitro dry matter digestibility ranged from 84.8% to 57.0%. Mineral nutrients P, K, CA, Mg, Mn, Fe, Cu, Zn, and Na were analyzed and determined to be adequate for maintenance needs of ewes. / Graduation date: 2002

Recent management efforts in the Smith and Bybee Lakes Wildlife Area (SBL), a 700-ha preserve in north Portland, Oregon, have included using a water control structure to suppress invasive reed canarygrass (Phalaris arundinacea L.) by flooding during spring and early summer growth periods. For the first year of managed flooding, I sought to determine: (a) the extent and distribution of reed canarygrass at SBL; (b) the effectiveness of the change in water level at suppressing reed canarygrass; and ( c) the effects of the change in water level on other plant species. I established 30 vegetation transects throughout SBL before completion of the water control structure. These transects were randomly distributed, placed perpendicularly to the shorelines of the wetlands, and had a cumulative length of 3.1 km. I measured vegetation on the transects in autumn 2003 and autumn 2004 using the line intercept method at 10-cm intervals. I surveyed the transects to generate elevation profiles, accurate to 0.15 cm, to determine depth and duration of flooding, which I correlated with vegetative changes. I also monitored inundation depth, growth, and phenological response of individual stands of reed canarygrass during the 2004 growing season.

Small broomrape, a holoparasitic weed, is a relatively new weed introduction in the Pacific Northwest that has contaminated a limited number of red clover fields in Oregon. Greenhouse and field studies were conducted to evaluate small broomrape response to common crop and weed species in the Pacific Northwest. Host species in the greenhouse or field study included alfalfa, arrowleaf clover, carrot, celery, common vetch, crimson clover, lettuce, prickly lettuce, red clover, spotted catsear, subterranean clover, white clover, and wild carrot. False-host species included barley, birdsfoot trefoil, creeping bentgrass, cucumber, field corn, fine fescue, flax, Italian ryegrass, nasturtium, oat, orchardgrass, perennial ryegrass, snap bean, sugar pea, sunflower, sweet corn, tall fescue, tomato, and wheat. Non-host species included sugar beet and curly dock. The greenhouse polyethylene bag system provided a rapid and inexpensive screening for plant species host status to small broomrape. Germination and attachment to host roots are initiated by chemical exudates, that may change concentration in response to nutrient availability and microorganisms. Red clover was grown in varying concentrations of ammonium sulfate fertilizer with and without Rhizobium inoculation, and with small broomrape seeds. Neither Rhizobium inoculation nor ammonium concentration influenced the number of small broomrape attachments to red clover roots. A survey was conducted of red clover seed growers with small broomrape-contaminated fields in the Pacific Northwest. Red clover seed from six respondents were cleaned at the same cleaning facility, and the same respondents purchased their seed stock from this cleaning facility. Small broomrape was not identified in red clover fields prior to or during the first clover seed harvest of fall planted red clover in small broomrape-contaminated sites. / Graduation date: 2003