Hi,

I am trying to test the effects of Plant ID (Different flower colours) and Petal Type (each flower was divided into two different petal types) on 5 different pigment concentration measurements (5 response variables).

I used the adonis() function in the Vegan package:

adonis(apc[,4:8]~apc$Plant_ID, nperm=999, strata = apc$Petal_Type)

adonis(apc[,4:8]~apc$Petal_Type, nperm=999, strata = apc$Plant_ID)

Subsequently I checked for multivariate homogeneity of group dispersions with betadisper function and a permutation test for each of the factors. Both were not significant.

Now I want to see which levels within my factors differ between each other, I was thinking to use the pairwise.perm.manova() function (from the RVAideMemoire package) for that.

However, how will I then be able to see the effects of Plant ID and Petal Type on each of my individual 5 response variables? For example, I know that Plant_ID influences my response variables, but I want to see which of the response variables (different pigment types) differ between plant ID and or Petal types.

Anybody ideas about this issue?

Thanks!