On the assumption of an additive effect of solute elements in dendrite growth