WETLAND PLANT-RESPONSES TO VARYING DEGREES OF PURPLE LOOSESTRIFE REMOVAL IN SOUTHEASTERN ONTARIO, CANADA

This study examined the response of native vegetation to varying level s of purple loosestrife (Lythrum salicaria) removal and determined its impact on subsequent loosestrife germination and establishment in a s outhern Ontario wetland during 1993 and 1994. The study site was treat ed in 1991 with varying dosages of triclopyr amine herbicide, the trie thylamine salt formulation of triclopyr [[(3,4,6-trichloro-2-pyridinyl )oxy] acetic acid]; 4.0, 8.0, and 12.0 kg active ingredient (a.i.)/ha. Different levels of adult loosestrife control were achieved. The lowe st density of adult loosestrife was observed in the 8.0 and 12.0 kg a. i./ha treatment plots where root kill was most effective. Most new adu lt loosestrife plants in the 8.0 and 12.0 kg a.i./ha treatment plots e stablished from seed. The highest densities of loosestrife seedlings a nd grass (Gramineae) species were observed in the treatment plots with the lowest number of adult loosestrife plants (12.0 kg a.i./ha). Sedg e species (Carer spp.) did not differ between treatment levels during 1993 and 1994. Loosestrife seedling densities decreased from 1993 to 1 994, suggesting that increased native plant species can slow the rate of loosestrife reestablishment from the seed bank. Native vegetation s eems to replace adult loosestrife for a limited time following herbici de application; however, without subsequent treatment, loosestrife wil l slowly reinvade a wetland.