It is characteristic of Goethe that, on becoming acquainted with Howard's work, he at once gave a warning against subdividing his scale without limit. Goethe foresaw that the attempt to insert too many transitory forms between Howard's chief types would result only in obscuring that view of the essentials which Howard's original classification had opened up. Obviously, for a science based on mere onlooking there is no objection to breaking up an established system into ever more subdivisions in order to keep it in line with an increasingly detailed outer observation. This, indeed, modern meteorology has done with Howard's system, with the result that, to-day, the total scale is made up of ten different stages of cloud-formation.