When you have too many variables, it is almost impossible to single out why your dependent variable changed. If in one trial, the temperature surrounding the plant is 34° and in the second trial the temperature is 54° your data will be different each time even if all of the other variables the same