" The adjective," wrote Horatio Hale, " when employed in an isolated form, follows the substantive, as Kanonsa, ' house ;' Kanonsa-kowa, ' large house ;' but in general the substantive and adjective coalesce." In some cases the adjective is split in two, and the substantive inserted, as in Tiogen, a composition of Te, " two," and ogeit, " to separate," which is split and the word ononte, " mountain," or hill, inserted, forming Te-ononte-ogen, " Between two mountains," " The local relations of nouns are expressed by affixed particles, such as ke, ne, kon, akon, akta. Thus from Ononta, mountain, we have Onontdkc, at (or to) the mountain; from Akchrat dish, Akehrdtne, in or on the dish," etc. From the variety of its forms and combinations it is a more difficult language than the Algonquian. No European has fully mastered it. No attempt has been made to correct record orthographies further than to give their probable missionary equivalents where they can be recognized. In many cases crude orthographies have converted them into unknown tongues. Imperfect as many of them are and without standing in aboriginal glossaries, they have become place names that may not be disturbed. No two of the early scribes expressed the sound of the same name in precisely the same letters, and even the missionaries who gave attention to the study of the aboriginal tongues, did not always write twice alike. Original sounds cannot now be restored. The diacritical marks employed by Williams and Eliot in the English alphabet, and by Zeisberger and Heckewelder in the German alphabet, are helpful in pronunciations, but as a rule the corrupt local record orthographies are a law unto themselves. In quoting diacritical marks the forms of the learned linguists who gave their idea of how the word was pronounced, have been followed.