Newton, Charles T. [Editor]; Pullan, Richard P. [Editor]: A history of discoveries at Halicarnassus, Cnidus and Branchidae (Band 2, Teil 2) (London, 1863)

GREEK INSCRIPTIONS. 731

The occurrence of the word dvafiaivco in the two
first of the passages just quoted from, makes it not
improbable that the words dvafiairj l£ fytoSouAeov were
a customary formula in the public sale of slaves.

If any such rite as I am here assuming was
really practised, the devoted person would pro-
bably be sold in effigy, and the forms and usages of
a real sale more or less imitated, as, in the magic of
the Middle Ages, compacts with the devil are always
described as instruments drawn up in legal form.

1. 30. 'Ao-^Xa7r<oa for 'Aa-x'Ka.Tnd/M.

1. 31. 7ra^Jo-Tav£Ta[»]. The form Trapirrrdvco for
Tra.pi(TT7}{u is used by late authors ; but I am not
aware of any example of Trafnarrdvo^ai. All the
letters, however, of Trapta-ravsTai are visible except
the final i, which has, I suspect, been confused
with the vertical stroke of the Y immediately
above. If we take ira^Krrdvsrai here as the equiva-
lent of the middle form 7ra.pia-Tarai, the sense would
be, " If any one produces the woman as a witness,
giving her money ; " or perhaps " suborns " would
more nearly express the meaning here.

1. 33. Scum seems to be a mere error for $oua-a, like
Tp-s^av, 1. 40, for rpdirz^av.

1. 31. rau ? 8' spot. In the original the first of
these letters is only a vertical stroke, but the cross
bar of the T may have been obliterated ; the A, as
will be seen by the facsimile, is very like an A;
but this may be from carelessness. At all events,
dva. hpou would make no sense. After Iju.o5 letters
are faintly visible, which may be either M or KA.

1. 35. s[/.o) 8' bV<a. A similar deprecatory formula
occurs in Nos. 82, 81, 85, 86, 87, 90, 91, 92.