unguis.

According to De Vaan, from the Proto-Indo-European u-stem *h₃n̥gʰ-u-, itself from *h₃negʰ- (“nail”). The u-stem was secondarily remade into an i-stem later in Italic. The derivation of this term from the zero-grade *h₃n̥gʰ could serve as an example of Rix's Law, according to which *#h₃NC- would yield #oNC-. Höfler, who rejects the existence of Rix's Law in Latin, alternatively suggests schwebeablaut in the root as an explanation, in which case the term would continue *h₃enbʰ-.In support of this theory, Höfler cites Sanskrit वण्क्रि (váṇkri, “rib”) (< *wénk-ri-) beside व॒क्र (vakrá, “crooked”) (< *wn̥k-ró-) as evidence for the existence for the existence of secondary full-grades in i-stem derivatives. He also cites Sanskrit अङ्घ्रि (áṅghri) (perhaps from *h₃engʰwi-) as evidence for schwebeablaut in the particula root *h₃negʰ- itself. Schrijver, however, rejects schwebeablaut as a possible explanation, arguing that this phenomenon cannot occur within the same paradigm. Cognates include Ancient Greek ὄνυξ (ónux), Old Irish ingen, Sanskrit नख (nakhá, “claw, nail”), Old Armenian եղունգն (ełungn), Old Church Slavonic ногъть (nogŭtĭ), Lithuanian nagas, Persian ناخن (nâxon), Albanian nyell, and Old English næġl (English nail).