How far one ought to establish fundamental propositions in this matter, I can not easily say. Perhaps it would be scientifically most correct to be satisfied for the time with collecting the carefully and keenly observed material and getting the anatomists, who are already in need of material for professional investigations, to take the matter up; in collecting photographs of hands belonging to persons whose characters are well known and in getting a sufficient number of properly equipped persons to make the collection. If we had enough material to draw fundamental principles from, much that has been asserted by Bell, Carus, D’Arpentigny, Allen, Gessmann, Liersch, Landsberg,[124] etc., might be proved and tested. But their statements are still subject to contradiction because their fundamental principles are not sufficient for the development of a system. Probably nobody will doubt some of the more common statements; all will grant with Winkelmann that a beautiful hand is in keeping with a beautiful soul; or with Balzac that people of considerable intellect have handsome hands, or in calling the hand man’s second face. But when specific co-ordinations of the hand are made these meet with much doubt. So for example, Esser[125] calls the _elementary_ hand essentially a work hand, the _motor_ essentially a masculine hand, having less soul and refinement of character than will and purposefulness. So again the _sensitive_ hand implies generally a sanguine character, and the _psychic_ hand presents itself as the possession of beautiful souls and noble spirits. However true this classification may be, the establishment and description of the various significatory signs is very difficult, especially because the forms named rarely appear in clear and sharply defined subdivisions. The boundaries are fluid, like the characters themselves, and where the properties of one group pass almost directly into the other, both description and recognition are difficult. If, then, we can not depend upon a systematic, and at present remote treatment, we still may depend on well-founded observations which appear as reliable presuppositions in the light of their frequent repetition.