We see that if we take two prints from an engraved copper-plate, or cast two images in a mould, very many points may immediately be found whereby they may be distinguished one from another. If, then, it cometh thus to pass in things made by processes the least liable to error, much more will it happen in other things which are made by the free hand. This, however, is _not the kind of Difference_ whereof I here treat; for I am speaking of a difference (from the mean) which a man specially intendeth, and which standeth in his will, of which I have spoken once and again.... This is not the aforesaid Difference which we cannot sever from our work, but, such a difference as maketh a thing fair or foul, and which may be set forth by the "Word of Difference" dealt with above in this Book. If a man produce "different" figures of this kind in his work, it will be judged in every man's mind according to his own opinion, and these judgments seldom agree one with another.... Yet let every man beware that he make nothing impossible and inadmissible in Nature, unless indeed he would make some fantasy, in which it is allowed to mingle creatures of all kinds together.... Any one who leads this carefully cannot fail to see that it is not only that Dürer is not "desirous of laying down rules applicable to all cases," or even of "proposing a definite canon for the relative proportions of the human body," as Thausing indeed points out (p. 305, v. 11): but that he does not conceive the proportions he gives as even approximately capable of these functions; and considers it indeed the very nature and special use of a canon of proportions to be wilfully deviated from, pointing out that, though the deviations of which he is speaking are slight and subtle, they are not to be confused with the accidental ones that can but appear even in work done by mechanical processes. Rather they are such variation as a man "specially intendeth, and which standeth in his will;" and again, "such a difference as maketh a thing fair or foul;" for the use of these normal proportions is that they may enable an artist to deviate from the normal without the proportions he chooses having the air of monstrosities or mistakes or negligences. He does not insist that either of the scales he gives is the best that could be, even for this purpose, but that they are sufficiently good to be used; and he would have marvelled at the wonder that has been caused in innocent critical minds that in his own work he adhered to them so little. He never intended them to be adhered to. V It may be objected that Dürer certainly sometimes thought of a Canon of Proportion as a perfect rule, because he wrote on a MS. page as follows:--