1Dept. Biology, NSAC, Truro, Nova Scotia, Canada B2N 5E3; 2Dept. Environmental Biology, University of Guelph, Guelph, Ontario, Canada N1G 2Wl; 3Dept. Plant Science, MacDonald Campus, McGill University, 21, 111 Lakeshore Road, Ste-Anne-de-Bellevue, Quebec, Canada H9X 3V9

Background and objectives
Dandelion is a perennial plant which occurs worldwide, but is a problem weed in golf courses, municipal parks, home gardens and agricultural crops [1]. A collaborative project was established in Canada involving three academic institutions and industrial partners, DowAgro Sciences, the Bioproducts Centre and the Saskatchewan Wheat Pool, with the aim of developing a bioherbicide for biological control of dandelion in turfgrass. This was initiated primarily in response to increasing legislation to restrict herbicide use in municipalities across Canada. The objective of this study was to evaluate the efficacy of eight fungal isolates, in various liquid and solid formulations, for control of dandelion under field conditions in the provinces of Ontario, Quebec and Nova Scotia.

Materials and methods
Field efficacy trials were conducted at three collaborating institutions, University of Guelph (UG) (Guelph, Ontario), McGill University (MU) (Montreal, Quebec), and the Nova Scotia Agricultural College (NSAC) (Truro, Nova Scotia), in June, July and September of 1996. At each location, dandelions were grown from seed in the greenhouse, transplanted into turfgrass at 10 plants per plot and allowed to establish for at least 4 weeks before applying treatments in a randomised, complete block design. The eight fungal isolates included spore andlor mycelial liquid formulations of G5/2 (Phoma herbarum), Gill (Phoma exigua), G961.16 (Phomasp.) (produced by UG), AC133 (Myrothecium roridum), AC9530 (Piectosphaerelia cucumerina) (produced by NSAC), Mac2 (Curvularia inaequalis), Mac41H (Colletotrichum sp.) and two solid formulations (sodium alginate and barley grit) of isolate Mac1 (name withheld due to confidentiality reasons) (produced by MU). Isolates were produced at the institutions indicated above and shipped to each site within 24 h of trial inoculation. Isolate performance was compared with a water control and a standard herbicide treatment (Killex) applied at the recommended field rate. Treatments were applied to dandelion in the early evening. Dandelion control was assessed at 1, 3, 7, 14, 28 and 42 days after treatment using a biomass reduction scale of 0-10, where 0=0% biomass reduction and 10=100% biomass reduction in comparison to the control. At 42 days after treatment, shoots were harvested, dried and weighed. Weather data (temperature, humidity and rainfall) were recorded at each site.

Results and conclusions
Mac1, when tested in either formulation, provided a high level of control of dandelion across all three locations in the June, July and September trials. Dandelions treated with Mac1 in the sodium alginate formulation scored 8-10 in 7 of 9 trials, while those treated with Mac1 in the barley grit formulation scored 8-10 in 5 of 6 trials. These scores were significantly greater than, or equivalent to, the level of control achieved by the standard Killex (P=0.05). No damage to turfgrass was observed. Some regrowth of dandelion occurred by 28 days after inoculation with Mac1 at all sites but, in general, shoot dry weights at 42 days after treatment were significantly reduced compared with the control (P=0.05). Mac2 caused severe infection (a score of 7 or greater) on dandelion at McGill and NSAC in the July and September trials when applied as an oil-in-water formulation. However, at NSAC, this formulation was phytotoxic to turfgrass. All other isolates caused generally low levels of infection in all trials.

Mac1 was the most consistently effective isolate at controlling dandelion despite varying climatic conditions due to location andlor season. Based on these results and considering that sodium alginate is more expensive than barley, Mac1 as a barley-based formulation has been selected for further study.

1. Riddle GE, Burpee LL, Boland GJ, 1991. Weed Science 39, 109-118.