Thanks for the scanned page, Alex. Nicetas Choniates' mention of <zoúmpros> can be ignored. It's a late source (ca. AD 1200), and the pronunciation indicated by the Greek spelling is /zu(m)bros/, which can hardly be anything else than a word of Slavic, presumably early East Slavic, origin.
The first example is also suspect, since <tragelapHos> means the goat-stag, a mythical beast, and whenever the word was used with reference to a real animal, it was something goat-like, like an Arabian gazelle or, in Europe, the chamois. The latter is probably what's meant in this case (I wish I knew what "Morelli Bibl. mscr. I" is, and in particular how old it is).
The most intriguing part is the reference in "AP. IX, 300" (what is it? can you check it up in Tomaschek's list of abbreviations?) to "a bull called zombros" killed by Peucestes with a spear, according to Addaeus the Macedonian. It's intriguing, because Addaeus was a contemporary of Alexander the Great, and the word, which can't be Slavic by any stretch of the imagination but _might_ be Thracian, in this case at least refers to some kind of bull. Even if we don't know for sure that it was a bison, there's some justification for speculating that Thracian may have had a word related to Slavic *zo~brU. If that were true, then the reconstruction *g^ombHros would be the only possibility, and the East Baltic forms would stand apart (Old Prussian wissambras could be German wis(ent)- combined with _native_ [rather than Slavic] sambras = /zambras/). I'll do some more checking.
Piotr
----- Original Message -----
From: "alex_lycos" <altamix@...>
To: <cybalist@yahoogroups.com>
Sent: Monday, January 20, 2003 6:34 AM
Subject: Re: [tied] bison