It is this older term that is reconstructible from this set. The present Mandan word for grass is xąhe, and the Mandan glottal in the ‘grass’ part of ‘corn’ is unattested in the rest of Siouan. Though ‘grass’ may not have been transparent in this compound in Mandan, ‘gourd’ clearly was; whereas nothing about the compound would have been transparent in Hidatsa where the ‘gourd’ word is kakuwi. Thus it seems likely that the word was transmitted from Mandan to Hidatsa and thence to Crow, where a look-alike form seems to involve an assimilation of initial k to the following x. The number of different terms, their phonological irregularities, and the clear similarities to terms for corn or related cultigens in neighboring languages all tend to confirm what archeologists already know, namely that maize was introduced long after the break up of Proto-Siouan. Note that three of the four corn terms are based on local, earlier terms for cucurbits, the only other generally cultivated vegetable at the time. Curcurbits precede corn in the archaeological record by hundreds of years. The portion of the Mandan and Catawban terms that is alike phonologically is the portion that refers to cucurbits, probably ‘gourd’. The use of gourds is quite ancient, and the morphemes may be cognate (although they contain an element of sound symbolism). Earliest corn in southern Wisconsin is circa A.D. 1000, north and west of there, later. There could have been an old compound but it cannot have referred to corn. The Catawba forms, if they signal anything at all, would have had to have developed independently from some more ancient term with a distinct but related meaning.