608 CLASSICAL ART 



been urged * with great force that the taking of casts from marble 

 sculptures was impracticable, for the simple reason that Greek mar- 

 ble sculptures were always more or less painted, and the process of 

 making a mould would have injured the coloring. Hence it is inferred 

 that we must draw a sharp line of distinction between two classes of 

 reproductions. On the one hand, from originals of bronze we have 

 copies, in which a high degree of fidelity may be presumed; on the 

 other hand, from originals of marble, and, it may be added, of gold 

 and ivory, we have imitations, whose trustworthiness is much less. 

 Thus, so the inference runs, while we may form a fair idea of 

 the bronze Discus-thrower of Myron or the bronze Doryphorus 

 of Polyclitus, we cannot know, except vaguely, the gold and ivory 

 Hera of Polyclitus or the marble Cnidian Aphrodite of Praxiteles. 



Here is a matter deserving serious consideration. Yet the distinc- 

 tion is perhaps not so important as it at first appears. We have no 

 assurance that the copies of bronze statues were always or even 

 usually made from casts, although that is possible. And even if they 

 were, it must be remembered that the possession of a cast, while 

 it made fidelity in the copy possible, did not by any means necessitate 

 fidelity. On the other hand, Greek marble sculptures may in some 

 instances by the Roman period have so far lost their coloring that no 

 objection would be felt to taking casts from them. And when this 

 was not the case, it must often have been possible to make an accurate 

 model in clay of a marble work, and from this model to make casts, 

 as has recently been done for one of the archaic female figures of the 

 Athenian acropolis. It is conceivable also that a copy was sometimes 

 based upon drawings made in the presence of the original and perhaps 

 accompanied by measurements. However it was done, it is certain 

 that copies much too faithful to have been executed from memory 

 were often made from marble originals. Thus in a caryatid of the 

 Braccio Nuovo of the Vatican we have a Roman copy of one of the 

 caryatids of the south porch of the Erechtheum, in fact, of the particu- 

 lar one which was removed by Lord Elgin and which now stands in 

 the British Museum. Again, there are numerous cases where a work 

 of relief sculpture in marble exists in two or more copies. Take 

 for example the relief representing Orpheus, Eurydice, and Hermes. 

 Whether the Naples example is the actual original or not, the original. 

 as of all such works, was certainly of marble. And in spite of the great 

 inferiority of the Villa Albani example, and the still greater inferiority 

 of the Louvre example, to that in Naples, the differences are not 

 greater than we often find between different copies of a bronze statue. 

 Now it is true that no amount of resemblance between copies affords 

 absolute proof of their resemblance to a lost original. It may con- 

 ceivably be that all derive from a single copy, and that an inexact 

 1 S. Reinach, Revue archMogique, 1900, n, p. 384 ff. 



