^{page 207 note 1} Neither is the explanation of de Saussure, Mém. Soc. Ling. 7. 86, satisfying on the semantic ride. Brugmann's comparison with Lat. cura (from coisa) lacks any precise morphological confirmation.

^{page 208 note 1} As further material for the examination of the question whether -ῐν- is ever a normal sequent of -isn-, the words τι-ν-θ⋯ς ‘hot,’ τι-ν-τ⋯ν ⋯φθ⋯ν and τῐ-ν-⋯σσει ‘shakes’ might all be referred to a root ^*t(w)is-, extended in Greek by a suffixed -ν-. But this material is all elusive, for back of the root ^*t(w)ei-s- we may set up a root ^*t(w)ei-, citing for it Latin titio ‘fire-brand,’ titubat ‘shakes, totters’ (?from ^*ti-twi-dhā-), and possibly titillat ‘tickles’ (? ‘causes to shake’); are liable to interpretation on the same lines.