The presence of the laryngeal is confirmed by the -h- in Sanskrit ahám. Instead of the unknown medial laryngeal *-H-, *-h₂- is often reconstructed here, but there is no secure evidence that would prove such assumption.
Three forms can be reconstructed formally for the nominative singular, using the comparative method:
*éǵ (Hittite ūg, Old Lithuanian eš, Slavic ja, Avestan azə)
There seem to be no grounds for assuming an initial *h₁ in the nominative singular (although it is not impossible) or a final *-H in the form *éǵ. The form *éǵ is obviously the most archaic one, with *eǵHóm and *eǵóh₂ being younger, dialectal variants. The sandhi form of *éǵ - *éḱ, with a final devoiced plosive, is attested in Baltic (Old Prussian as, es) and apparently in the Hittite variant [script needed] (ú-uk).
Oblique pl. *n̥s- possibly continues earlier **m̥s- (same element of sg. oblique stem with pl. s). This n was made common to the whole paradigm, with even enclitic *nos for *mos, but the verb endings *-me(dʰh₂), *-mos(dʰh₂) were not disturbed.
Dat. *mégʰi for *mébʰi is attested in Italic and Indic, but not in Iranian, as 𐬨𐬀𐬌𐬠𐬫𐬁(maibyā). Also Sanskritमह्यम्(máhyam) may be from original -bʰ-, as this sometimes becomes Indic -h- (even more here by dissimilation from initial m-).