In the first place it may be stated that sufficient linguistic material is preserved from the Apalachee,' Hitchiti, Mikasuki, Alabama, Koasati, Choctaw, Chickasaw, the leading tribes of the Muskogee branch, Natchez, Yuchi, and Timucua, to establish their positions beyond question. The connection of all of the other tribes of the Choctaw group except Pensacola, that of the Chatot, and the tribes of the Natchez branch has been examined by the author in his Indian Tribes of the Lower Mississippi Valley, to which the reader is referred.2 That Hitchiti with but slight variations was spoken by the Apalachicola, Sawokli, and Okmulgee is known to all well-informed Creek Indians to-day, and some of the people of those tribes can use it or know some words of it. The town names themselves are in Hitchiti.

Oconee is placed by Bartram among those towns speaking the "Stinkard" language, and all of the other towns so denominated, so far as we have positive information, spoke Muskhogean dialects belonging to either the Hitchiti or Alabama groups. Oconee, being a lower Creek town, would naturally belong to the first. Further evidence is furnished by the later associations of the Oconee people with the Mikasuki.4

The Tamali, so far as our knowledge of them extends, lived in southern Georgia near towns known to have belonged to the Hitchiti group, and they were among the first to move to Florida and lay the foundations of the Seminole Nation. In Spanish documents a tribe called Tama is mentioned which is almost certainly identical with this, and it may be inferred that the last syllable represents the Hitchiti plural -ati. These facts all point to a Hitchiti connection for the tribe.

Bartram tells us that in his time the language of the Chiaha was entirely different from that of the Kasihta, which we know to have been Muskogee, and in his list of Creek towns he includes it among those speaking Stinkard. As explained above, this latter fact suggests that Chiaha was a Muskhogean dialect, although not Muskogee. By some of the best-informed Creeks in Oklahoma I was told it was a dialect of Hitchiti, and that on account of the common language the Chiaha would not play against the Hitchiti in the tribal ball games, although they belonged to different fire clans, which ordinarily opposed each other at such times. The chief of the Mikasuki told me that Chiaha was the "foundation" of the towns called Osochi, Mikasuki, and Hotalgihuyana, and that anciently all spoke the same language.

Almost confined to one letter published in facsimile, accompanied by its Spanish translation, by Buckingham Smith, in 1860.

Bulletin 43, Bur. Amer. Ethn. The Washa and Chawasha have, however, since been identified as Chitimachan. (See Amer. Jour. Ling., I, no. 1, p. 49.)

3 Bartram, Travels in North America, p. 462.

4 See p. 401.

See p. 181 et seq.

The Tawasa Indians ultimately united with the Alabama, and the living Alabama Indians recall no differences between the languages of the two peoples. Moreover, Stiggins, writing early in the eighteenth century, gives certain episodes in the history of the Tawasa as if he were speaking of the whole of the Alabama.1 Still more ancient evidence is furnished by Lamhatty, a Tawasa, who was taken captive by the Creeks and made his way into the Virginia settlements in 1707. There the historian Robert Beverly met him and obtained from him an account of his travels and a rude map of the region which he had crossed in order to reach Virginia. While the ending of most river names, -oubab, is identical with that which appears in Apalachee, the name of the Gulf of Mexico, Ouquodky, is plainly the Oki hatki, "white water," of the Hitchiti, and is the name still applied by them to the ocean. Since the present Alabama term is Oki hatká we may perhaps infer that Tawasa speech was anciently closer to Hitchiti than to Alabama. Later, however, it was entirely assimilated by Alabama, and therefore it is more convenient and less hazardous to place it in the Alabama group. In either case the Muskhogean connection of the language is assured. It is probable that the "Poúhka" of Lamhatty were the Pawokti later found living with the Alabama, and if so it is a fair assumption that their history was the same as that of the Tawasa.

3

Muklasa is set down by Bartram as a Stinkard town. It was located in the upper Creek country, near the Alabama and Koasati towns, and it has a name taken from either the Alabama or the Koasati language. Gatschet states with positiveness that the Muklasa people were Alabama, and he may have learned that such was the case from some well-informed Indian now dead, for to-day the Creeks have well-nigh forgotten even the name.

The Pensacola disappear from history shortly after their appearance in it, and nothing of their language has been preserved. Their name, however, is plainly Choctaw and signifies "hair people." It may have been given to them because they wore their hair in a manner different from that of most of their neighbors, and Cabeza de Vaca mentions as a curious fact that several chiefs in a party of Indians he and his companions encountered near Pensacola Bay wore their hair long. When we recall Adair's statement to the effect that the Choctaw were called Parsfalaya, "long hair," because of this very peculiarity a connection is at once suggested between the two peoples.

1 See p. 140.

* D. I. Bushnell, Jr., in Amer. Anthrop., n. s. vol. x, no. 4, pp. 568-574.
Ibid., map.

Bartram, Travels in North America, p. 461.

5 Gatschet, Creek Mig. Leg., I, p. 138.

Bandelier, Journey of Alvar Nuñez Cabeza de Vaca, p. 48; also present work, p. 145.

7 Adair, Hist. Am. Inds., p. 192. He spells the word Pas' Pharáàh.

The Tuskegee have spoken Muskogee for more than a hundred years, but from Taitt (1772) and Hawkins (1799) it appears that they once had a language of their own. This statement was confirmed to me by some of the old people and they furnished several words which they affirmed belonged to it. Perhaps these are nothing more than archaic Creek, but in any case the long association of the tribe with the Creeks, Hitchiti, and Alabama points to a Muskhogean connection as the most probable.

The Muskhogean affinities of Yamasee have long been assumed by ethnologists, largely on the authority of Dr. Gatschet, but it can not be said that the evidence which he gives is satisfying. One of the words cited by him as proving this, Olataraca, is Timucua; another, yatiqui, is both Creek and Timucua; and most of the others are not certainly from Yamasee. The traditions of the Creeks are divided, some holding that the Yamasee language was related to theirs, others that it was entirely distinct. This last contention need not have much weight with us, however, because to a Creek Hitchiti is an "altogether different" language. From the statements of Spanish writers it is certain that the language spoken in their territories and those of the adjoining coast tribes, northward of Cumberland Island, was distinct from the Timucua of Cumberland Island and more southern regions. One province is called the "lengua de Guale," the other the "lengua de Timucua."5 More specific evidence as to the nature of that former language is not wanting. In 1604 Pedro de Ibarra, governor of Florida, journeyed from St. Augustine northward along the coast as far as St. Catherines Island, stopping at the important mission stations and posts, and holding councils with the Indians at each place." Until he left San Pedro (Cumberland) Island he employed as interpreter a single Indian named Juan de Junco, but as soon as he passed northward of that point another interpreter named Santiago was added. Moreover, the chiefs met previously were all called "cacique," but afterwards the name "mico" is often appended, the chief of the very first town encountered being called the "cacique and mico mayor don Domingo." It appears in letters written both before and after the one quoted above, as in three by Governor de Canço in 1597, 1598, and 1603, and the report of a pastoral visit to the Florida missions by the Bishop of Cuba in 1606. The earliest of all is in the narrative of an expedition sent from Havana in search of Ribault's Port Royal Colony.

1 Mereness, Trav. in Amer. Col., p. 541; Hawkins, Sketch of the Creek Country, Ga. Hist. Soc. Colls., III, p. 39.

2 See p. 208.

3 See also the Alabama tradition (p. 192) in which Tuskegee, under the name Hatcafaski, seems to be enumerated among the Alabama towns.

The captain of the vessel "landed near the town of Guale and went there, where was the lord micoo (el señor micoo)." A little later "the micoo of a town called Yanahume" 971 came to see him. This word is nothing other than the Creek term for chief.

In 1598 the confessions of Guale Indians, whose testimony was being taken with reference to the revolt of 1597, were communicated by them to a Timucua who understood the language of Guale, and by him to another Timucua who could speak Spanish. In a letter describing his missionary work Fray Baltazar Lopez, who was stationed at San Pedro, states that, while he is himself familiar with the language of his own Indians, he employs interpreters in speaking to the Guale people passing back and forth between their own country and St. Augustine.1

Some supplementary evidence is furnished also by the place and personal names recorded from the Indians in this area, which will be found in the section on the Guale Indians and the Yamasee. The difference between these and Timucua names is apparent when they are compared with the list of names on pages 323-330. The phonetic r does not appear, except in one case where it is plainly not an original sound, while ƒ and 7, which are foreign to the eastern Siouan dialects, are much in evidence. So far as Yuchi is concerned the history of that tribe, as will be seen later, tends to discount the idea of any connection there. Besides, m appears to occur in the Guale language at least-Tumaque, Altamahaw, Tolomato, Tamufa, Ymunapa while it is wanting in Yuchi. To these arguments may be added the positive resemblances to Muskhogean forms in such names as Talaxe (pronounced Talashe), Hinafasque, Ytohulo, Fuloplata, Tapala, Çapala (Sapala), Culupala, Otapalas, Pocotalligo, Dawfuskee. Finally, the relationship is indicated by the speeches of various Creek chiefs at the time of their historic conference with Governor Oglethorpe in 1733.2 Tomochichi, chief of the Yamacraw, a small band of Indians living near Savannah at that time, says "I was a banished man; I came here poor and helpless to look for good land near the tombs of my ancestors." The Oconee chief declares that he is related to Tomochichi, and on behalf of the Creek Nation claims all of the lands southward of the river Savannah. Finally the mico of Coweta thus expresses himself:

I rejoice that I have lived to see this day, and to see our friends that have long been gone from among us. Our nation was once strong, and had ten towns, but we are now weak and have but eight towns. You [Oglethorpe] have comforted the banished, and have gathered them that were scattered like little birds before the eagle. We desire, therefore, to be reconciled to our brethren who are here amongst you, and we give leave to Tomo-chi-chi, Stimoiche, and Illispelle to call their kindred that love

1 Lowery, MSS.

* A True and Hist. Narr. of the Colony of Ga. in Am., &c., Charles Town, S. C., 1741, pp. 31-39.

L

them out of each of the Creek towns that they may come together and make one town. We must pray you to recall the Yamasees that they may be buried in peace amongs their ancestors, and that they may see their graves before they die; and their ow nation shall be restored again to its ten towns.

Here the Yamacraw and the Yamasee seem to be treated as former members of the Creek Confederacy. Unless the Yamasee and the Guale Indians had been so considered the Creeks at this council would not have claimed all of the land on the Georgia coast south of the Savannah River and at the same time have asked that the Yamasee be recalled to inhabit it. It is as guardians of these tribes that they ceded to Oglethorpe the coast between Savannah River and St. Simons Island, with the exception of the islands of Ossabaw, Sapello, and St. Catherines, and a small strip of land near Savannah city.

The particular Muskhogean dialect which these Indians spoke is, however, more difficult to ascertain. Ranjel indicates a connection between the Yamasee and Hitchiti,' and this impression appears to have been shared generally by the Muskogee Indians of later times. On the other hand, the word for chief among the Guale Indians was, as we have seen, miko," the form which it has in Muskogee, whereas the proper Hitchiti term is miki. This means either that Muskogee was already the lingua franca upon the coast of Georgia or else that the languages of the Guale Indians and the Yamasee belonged to distinct groups. According to several traditions the Muskogee at one time lived upon this very coast, and I am inclined to accept the second explanation, but it is not put forward with overmuch confidence.

The name of the Cusabo first appears in the form "Coçapoy" in a letter of Governor Pedro Menendez Marques dated January 3, 1580. It is there given as the name of a big town occupied by hostile Indians and strongly placed in a swamp, about 15 leagues from the Spanish fort at Santa Elena.3 The tribe appears later as one of those accused of fomenting an uprising against the Guale missionaries in 1597, and afterwards among those appealed to for help in putting it down.1 There is every reason to believe that its appellation was connected in some way with that of the Coosa Indians of South Carolina, but how is not certain.

By the English the name is sometimes used to designate all of the coast tribes of South Carolina from Savannah River to Charleston and two on the lower course of the Santee. On the other hand, not only are the latter sometimes excluded, but at least one of the tribes of the neighborhood of Charleston Inlet. Mooney suggests a still more restricted use of the word.5 In its most extended application

1 See p. 95.

2 Or mico; c indicates precisely the same as k.

3 Lowery, MSS.

4 See p. 60.

5 Siouan Tribes of the East, Bull. 22, Bur. Amer. Ethn., p. 86.