The idea of a canon for human proportions has proved a great stumbling-block for so-called classical or academic artists. It is usually taken to mean an absolutely right or harmonious proportion, any deviation from which cannot fail to result in a diminution of beauty. According to their thoroughness, the devotees of this idea seek to arrive at such a scale of proportions for a varying number of different ages in either sex; often even modifying this again for diverse types, as tall or short, fat or lean, dark or blonde, but allowing no excessive variation for these causes; so that abnormally tall people and dwarfs are not considered. This is, I take it, what the great artist Albert Dürer is generally taken to have been aiming at in his books on proportion. It will not be difficult, I think, to show that Dürer had quite a different idea of what a canon of proportion should be, and how it should be applied. And certainly, had it been possible to study Greek practice more closely, and in a larger number of examples, when this idea (supposed to be drawn from that source) was chiefly mooted, a very different notion of the canon of proportion would have been forced on the most academical of theorists. Dürer's great superiority over such academical masters is, that his idea of a canon of proportion and its use agrees far better with what was apparently Greek practice. Any one who has followed at all the interesting attempts made by Professor Furtwängler and others to group together, by attention to the measurements of the different parts of the figure, works belonging to the different masters, schools, and centres, will have perceived that he is led to assume a traditional canon of proportion from which a master deviates slightly in the direction of some bias of his own mind towards closer knit or more slim figures; such variations being in the earlier stages very slight. Again, it is supposed that from the canon followed by a master, different pupils may branch off in opposite directions according to the leanings of their personal sentiment for beauty. The conception of these ramifications has at least created the hope that critics may follow them through a great number of complications, since a master may modify his canon--after certain pupils have already struck out for themselves, and new pupils may start from his modified canon; and so on into an infinite criss-cross of branches, as any sculptor may be influenced to modify his canon by his fellows or by the masters of other schools whose work he comes across later. In any case, this main fact arises, that the canon appears as what the artist deviated from, not what he abided by: and any one who has any feeling for the infinite nicety of the results obtained by Greek sculptors will easily apprehend that each masterpiece established a new and slightly different canon, and was then in the position to be in its turn again deviated from, as Flaubert says: