cenysor wrote:
But how do I know what the nominative singular ending of the suffixed stem looks like, whether the noun is declined thematic or athematic and what gender it has?
A root might form different words through suffixing different stems. For example, the root *wlk "wolf" was a thematic noun for a male wolf but could have a *-iH or *-eh2 suffix for a female wolf. The related root *wlp "fox" likewise has forms based on *-iH and *-e/o, but Latin has *-eh1.
It's often simply not clear what the gender was or which stem types were used. Some stem types were extended through analogy. The thematic paradigm was often extended to roots that didn't originally use it. In German, some root nouns were analogically made into u-stems based on the accusative in -um (< *-m).
Starostin's database
http://starling.rinet.ru/cgi-bin/query. ... g&morpho=0
can be a big help because words with the stem(s) are given as examples.
Beekes' book also has a table which relates stem types to gender and accent class:
Hysterodynamic m/f: (k), t, nt, s, h1, h2, r, (l), (m), n, i, u
Proterodynamic n: s, r/n, (l/n), n, (i), u
Proterodynamic m/f: (h2), i, u
Static m/f: (t), nt, r, (i)
Static n: r/n
Stems in parentheses are rare.
