The ya-class stems are more than a hundred and thirty in number, and nearly half of them have forms in use in all periods of the language, about forty occurring only in the earlier, and about thirty only in the modern period.

a. Of the roots making ya-stems, a very considerable part (over fifty) signify a state of feeling, or a condition of mind or body: thus, kup be angry, klam be weary, kṣudh be hungry, muh be confused, lubh be lustful, çuṣ be dry, etc. etc.

b. A further number have a more or less distinctly passive sense, and are in part evident and in part presumable transfers from the passive or yá-class, with change of accent, and sometimes also with assumption of active endings. It is not possible to draw precisely the limits of the division; but there are in the older language a number of clear cases, in which the accent wavers and changes, and the others are to be judged by analogy with them. Thus, √muc forms múcyate once or twice, beside the usual mucyáte, in RV. and AV.; and in the Brāhmaṇas the former is the regular accent. Similar changes are found also in ya-forms from other roots: thus, from kṣi destroy, jī or jyā injure, tap heat, dṛh make firm, pac cook, pṛfill, mī damage, ric leave, lup break, hā leave. Active forms are early made from some of these, and they grow more common later. It is worthy of special mention that, from the Veda down, jā́yate is born etc. is found as altered passive or original ya-formation by the side of √jan give birth.

c. A considerable body of roots (about forty) differ from the above in having an apparently original transitive or neuter meaning: examples are as throw, nah bind, paç see, pad go, çliṣclasp.

d. A number of roots, of various meaning, and of somewhat doubtful character and relations, having present-stems ending in ya, are by the native grammarians written with final diphthongs,āi or e or o. Thus:

e. Roots reckoned as ending in āi and belonging to the a- (or bhū-) class, as gāi sing (gā́yati etc.). As these show abundantly, and for the most part exclusively, ā-forms outside the present-system, there seems to be no good reason why they should not rather be regarded as ā-roots of the ya-class. They are kṣā burn, gā sing, glā be weary, trā save, dhyā think,pyā fill up, mlā relax, rā bark, vā be blown, çyā coagulate, çrā boil, styā stiffen. Some of them are evident extensions of simpler roots by the addition of ā. The secondary roots tāystretch (beside tan), and cāy observe (beside ci) appear to be of similar character.

f. Roots reckoned as ending in e and belonging to the a- (or bhū-) class, as dhe suck (dháyati etc.). These, too, have ā-forms, and sometimes ī-forms, outside the present system, and are best regarded as ā-roots, either with ā weakened to a before the class-sign of this class, or with ā weakened to ī or i and inflected according to the a-class. They are dhā suck,mā exchange, vā weave, vyā envelop, hvā call (secondary, from hū). As of kindred form may be mentioned day share and vyay expend (probably denominative of vyaya).

g. A few roots artificially written with final o and reckoned to the ya-class, with radical vowel lost before the class-sign: thus, do cut, bind, pres. dyáti etc. These, as having an accented áin the sign, have plainly no right to be put in this class; and they are better referred to the á-class (see above, 753 c). Outside the present-system they show ā- and i-forms; and in that system the ya is often resolved into ia in the oldest language.