148 
Wisconsin Academy of Sciences , Arts and' Letters. 
He consequently set to work to find some explanation which should be 
more satisfactory, and the result of his lucubrations is given us in a 
lengthy chapter of the work above cited, pp. B24-890. 
Osthoff’s theory is that we have in the Greek -xa- perfect the particle 
xsv; that SeScoxa is dedoo + xsv. 
Let us proceed to the details. And first a few words on roots and the 
different forms they assume under different circumstances. 
It is a generally accepted theory that roots appear under different forms 
in consequence of accentual conditions. Thus we explain 
Skrt. as- mi 
as- i 
as- ti 
as having the full form of the root as- because accented on the root sylla¬ 
ble; whereas the dual and plural forms 
svds smas 
sthds stha 
stds santi 
take the weakened form of the root -s because the accent is on the personal 
ending. So also in Greek we had origin ally * 8 i- 8 g 5 -jui, *dz-da?- 61, * Si-doo-ri 
*8i-do-r6v } *8 i-6o-tov 9 *di-8o-ju£v, etc., with the strong form of the root 
Sod- in the singular, because the accent stands on the root syllable, 
and the weak form do- in the dual and plural, because the accent is on 
the personal ending. Of course this primitive accentuation was lost in 
Greek, before we know it historically as an independent language; the 
primitive accentual conditions disappeared with the rise of the three sylla¬ 
ble law and of the recessive accent of the verb. 
So also in the inflection of the noun in Greek. In words of the type of 
Ttarpp, prjTrjp, etc., we have the strong form of the stem Ttarep-, pprep- 
where the stem syllable is accented, as in the accusative singular nave pa; 
nominative plural Ttarepe^; but we have the weak form of the stem itarp- 
where the accent falls upon the case ending, as Ttarp- 65 , Ttarp-i ; %a- 
rpd6i{r) also is formed from the weak stem Ttarp-. 
This variation of the form of the root in two, three and sometimes even 
four ways (called by the Germans Stammabstufung ) extends to all parts of 
speech. It is not limited to nouns and verbs, in which the accent origin¬ 
ally shifted, standing now upon the root syllable, now upon the ending, 
but may occur also in monosyllables, which by their varying force some¬ 
times receive a decided stress and at other times are enclitic. 
All parts of speech are equally liable to this change of form dependent 
upon stress-accent. The phenomenon is witnessed most frequently incase 
of nouns, adjectives and verbs; yet it occurs also (and this is at present the 
important point for us) in the case of particles; and we shall soon see that 
in xa we have simply another form of xev, a form bearing the same relation 
to xsv, as our unaccented “ the ” does to our accented “ the or as “ come ” 
in “ cm in” to our “come” in the phrase “ you come.” 
