Some observations on the experimental behaviour of a complex soil