[R] post-hoc comparisons following glmm

[Michaël Coeurdassier]

Dear R community,

I performed a generalized linear mixed model using glmmPQL (MASS 
library) to analyse my data i.e : y is the response with a poisson 
distribution, t and Trait are the independent variables which are 
continuous and categorical (3 categories C, M and F) respectively, ind 
is the random variable.

mydata<-glmmPQL(y~t+Trait,random=~1|ind,family=poisson,data=tab)
Do you think it is OK?

Trait is significant (p < 0.0001) and I would like to perform post-hoc 
comparisons  to check  where the difference among  Trait categories but 
I did not find  a solution  in R help list or others.

Thank you in advance for your help

Michael