Latin before 170-150 B.C. shows regularly I in some terminations and EI in others; ī (except where of perspicuous origin) comes from pre-Latin ei, while ei is from earlier ai or oi. Notably the gen. sg. in - ī is (with Ehrlich) an original locative in -ei; the passive infinitive in -ī had prim. IE -ei, being a dative of a consonant-stem or a locative of an - o -stem. The Latin dative in -ī, older -EI, is not identical in ending with the Oscan dative in -ei. Older ie contracts to Latin ī . Defense of the older view that the active infinitive in -re is an -s-stem with locative -i, not (as Hirt thinks) with locative -ai .
Roland G. Kent (Mon,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: