General comment

Cf. ‘cat’ and ‘corn’ for similar Dakota, Chiwere, Hoocąk clusters. JEK finds good analogs for these in Algonquian, and there is every indication that they are loanwords in Siouan. The cluster *kw is not found in Siouan, and the sequence *wų is also extremely rare. The Dakota/Chiwere/Hoocąk terms suggest a source with the approximate shape wa•kwan-, wa•kwon-. Cf. †kmą ‘corn’ . In this instance, the prefixes differ, and Lakota ų : Hoocąk ą is also irregular. While we have found a number of cases in which the two back nasal vowels have partially merged, the above factors, plus the rare (historically non-existent in Proto-Siouan?) cluster lead us to believe that these are borrowings in their respective languages, presumably Algonquian. The variation of the prefix in this form between wa- and wi- is also found in corn and cat, as is the alternation between ų and ą, and it could be argued that this variation is part of the behavior of the medial cluster. In fact, the probable source of variability in the vowel is accomodation of AL -me- to SI phonological and morphological norms. me is not a permitted sequence in most of the contemporary languages. The reconstructible PA form is *e•mehkwa•ni. Compare also Kickapoo eemehkwaani ‘spoon’ (Voorhis 1988), Menomini ɛ•mɛhkwan ‘squash, melon, seed’ (Goddard p.c.).