192:^1 THE GREEK MAGICAL PAPYRI IN THE BRIT. MUS. 
2 a 
Pap. CXXIII. 
This papyrus is not a fragment, as Kenyon believed, but 
a complete am ulet, as Wessely took it to be. It is folded just 
like a letter (all details as to folding and writing across the 
fibres and as to the drawing are given accurately by Wessely 
in line 62), and no doubt the writer or the possessor of this 
amulet has used it as a cpL[j(,ojT!,/,6v xal u7roTaxTLx,ov x-ai. yA^oyoq 
(pap. Lond. 121, 396) against all sorts of enemies. The amulet 
is simply a copy from a magic book (v. Wess.) A book of 
magic had consequently at that time already in itself great magic 
power, and you might copy it just as the sorcerers in later times 
copied subjects or psalms or anything out of the Bible on their 
scraps of paper or tablets. In other words, you might use the 
copy of a with just the same effect as if you realised 
the magician’s prescript in practice yourself. Here the writer 
has left out the beginning of the magical procedure, something 
like: ypacps iizX Xajxvri; (p.oXLjBTjg, etc.) voutov vov Aoyov' 
Aoyo; etc. But the Aoyog itself we probably still read in full 
in our papyrus. Then the receipt continues v. 8: åvi^ypacps) 
OTTLilov Xa[jt.[[[jt-]lv '/]5 — and v. 11: xal zlg vo p.évojvov (sc. tv); 
Aarxvvjc) (^ ZTzi - Ypyj ^ z ) to ovo[v.a auTob oTroxELvaL) “at the head 
you write’’; for this name we just read v. 14, where it forms 
part of the vpoloyog to be recited before the magic procedure: 
SITTE Be <(ii.sTa)> tooto to ovo[jt.a to sTravoj too ttstxAoo <(ysypa[jL- 
pivov) laoj [xop[jt,opoToxojjiat (as Ken. gives the name in the note, 
not as he gives it in the text). 
The figure, which is given here according to a photograph 
which I procured through the kindness of Mr. Bell, shows a 
male deity, with hawk\s head surmounted by a crescent, and in 
his right hand holding an anch symbol — and opposite to this 
mighty deity a human figure or a txeAsto^, as you might call 
it, apparently subdued to the omnipotent demon. The copy of 
Kenyon is everywhere reliable, but on the following points I 
venture to disagree with him or to supplement him. 
V. 2 we probably have to read ;(EVT'/]P'.oj/i>co, v. 4 
[XEVEjjayuy, v. 5 ttxv ysvog appsvov Bs xal hvjAoxcov (i. e. appsvcov te 
