I had a few questions: In a Agronomic experiment (for example a factorial experiment with intercropping (at 5 level) and bacteria (in two level) that has been tested in two years, 1- Is it better to use Bartlett's test or Hartley's F-max test for the homogeneity of error variance? Why? Is there a reference? 2- If homogeneity of variance is rejected, and the environments were heterogeneous, what is the solution for this two-year experiment?  How should the analysis be continued, can we do combined analysis, or should we do separate analysis for each year? In this case, how will the interpretation be?  Is there a reference? 3- If we use the mixed model with random effect of the year (or environment), is there a need to test the homogeneity of the variance of the environments? In general, for any number of environments.