We introduced a methodological criterion that enabled us to test the treatment of leathers without carying out the physico-chemical analytic methods, because we cannot immediately access to them in the museum. The behavioural differences amongst humectant treatments are checked comparing several concentrations and different damage degree of the object. The best results were obtained with 10per thousand sorbitol on leather with low chemical damage, and with 30per thousand glicerol on leather with important chemical damage.