On the birth of boundary layers by Grasman