The bioassay use is one of the more common techniques for evaluating an herbicide potential as well as identifying and quantifying its residues in soil or in water. In these studies it is very important to know the effects exerted by some variables on the biological material and almost always it is possible to relate them by a mathematical expression. A problem in obtaining the expressions is to find that one which better represents the dose-response relation. This work presents a comparative study between the logistic equation [y=a/(1+(X/b)c)] and the normal polynomial equations of 1o, 2o and 3o degrees. In evaluating the more adequate model for the dose-response study with glyphosate and imazapyr, some criteria of theoretical order and practical application were considered, and it was concluded that: a) the logistic function presents a significative estimate of its parameters for herbicide imazapyr and a justifiable biological reason for both herbicides; b) the characteristics of total dry biomass (BST), aerial part (BSA), root (BSR) and root length (CR) are adequate for the doseresponse relation studies; c) the I50 estimate is variable with the function and the evaluated characteristic; d) the tomato is more sensible to imazapyr.
Statistics; herbicide