which hierarchical clustering method&metric for dissimilarity matrix?

Dear all, From a Pearson correlation matrix, I call Rpearson (1000x2000) (obtained from 2 matrices X and Y) I want to use the dissimilarity matrix as input for hierarchical clustering. Rdissimilarity = 1 minus Rpearson or Rdissimilarity = 1 minus absolute value of Rpearson.

In that case, I think I can use the following function

Z = linkage(Rdissimilarity,method)

but if I use the Z = linkage(Rdissimilarity,method, metric), is the following appropriate?

Z = linkage(Rpearson,method,'correlation')

(according to the help, 'correlation' metric = 1 minus the sample correlation between points (treated as sequences of values, which explanation is kind of obscure to me. Is the 'correlation' metric appropriate in my case?

Finally, what is the best method (ward, euclidian, ...) to work on dissimilarity matrices?

Thank you for your answers,