studies of several groups or types, some well known, others less common, to interpret them in the light of our present knowledge, and to illustrate them, as far as possible, from hitherto unpublished specimens. There was a temptation to publish all the pieces in American collections, since I have at hand casts or photographs of almost all of them, but this plan, like that of a corpus, is open to the objection that it would entail reproduction of many almost identical examples. The studies of various types are followed by methodical descriptions, in catalogue form, of the objects shown in the accompanying plates; and this catalogue is introduced by an explanation of its method, which would be out of place here. Plates XIX and XXI exhibit several objects for which I could propose no convincing interpretation. The illustrations of them, taken in conjunction with the descriptions — which, in such cases, I have sometimes expanded by pointing out parallels and offering tentative suggestions — may help an ingenious observer to explain designs and inscriptions that I have found obscure.
The footnotes will show that little of importance that has been written on the subject has escaped my attention, and yet I cannot hope to have made use of all the widely scattered literature of magical amulets. With regard to one important work I plead guilty to a neglect imposed by circumstances. Professor Theodor Hopfner's Griechischer Offenbarungszauber has been used only here and there for special topics, and there is little doubt that ideas and interpretations put forward here may be found in his book, and probably supported by wider learning and stronger arguments. Unfortunately, the form in which that work had to be published, lithograph from manuscript, proved too taxing for eyesight which was already strained by the effort of deciphering minutely engraved inscriptions, and by previous work on papyri. For that reason I consulted the book much less than its importance deserved.
Any well-informed student of ancient religion and magic, and many archaeologists, will be able to correct these studies at various points, and to supplement them in many respects. For that matter, I could have gathered more material and used more of what is at hand, and could perhaps have corrected some errors and strengthened some hazardous suggestions, if time were not a consideration. But it is better to send the book out as it is than to strive for an unattainable perfection and in the end leave it unfinished. Such service as it can render to inquirers must make up for defects that are all but unavoidable when one is dealing with a great number of small and often puzzling objects.