12 BUREAU OF AMERICAN ETHNOLOGY [BULL. 68 
7 
aspirated sounds in the other languages, and in all three ’ is occasion- 
ally employed after a glottalized sound. Parentheses separate parts 
of a stem which there is reason to think are not essential. 
In copying Duralde’s vocabulary of eastern Atakapa ¢ has been 
substituted for sh and te for ch or tch, and where a vowel has been 
doubled to indicate quantity the proper form has been used instead, 
but some devices, although evidently wrong, it has been impossible 
to correct intelligently and they are allowed to stand. These include 
in particular combinations of k and g, such as kk, kg, gg, probably 
intended for the spirant 2. 
COMPARISON OF THE MATERIAL 
In comparing this material it must be remembered that we are 
dealing with fragments, albeit with large fragments. The frag- 
mentary nature of Atakapa is particularly evident. While the great- 
est body of material has been preserved from Chitimacha that is also 
the most complicated language; probably, taking into consideration 
the relative complexity of the three tongues, that preserved from 
Tunica is the most nearly adequate. 
COMPARISON OF THE PHONETICS 
In my description of the phonetics this ground has practically 
been covered already. The unsatisfactory state im which we find 
them prevents anything like a detailed comparison. The differences, 
however, though striking, are by no means fundamental. The most 
important of these are the presence of a velar r and the absence of % in 
Tunica, the absence of an 7 in Chitimacha, and apparent vestiges of 
Zt and f in Atakapa, both of which are wanting in the other two. 
It is also interesting to observe that in the employment of s and the 
affricatives Chitimacha occupies an intermediate position with 
reference to the others. In Atakapa we find ¢, ts, and dz, and in 
Tunica c, tc, and dj, ts and dz being wanting and s much less employed 
than c. In Chitimacha we have all of these, though there is some 
uncertainty regarding the exact number and quality of the sounds 
represented. 
Chitimacha is much more consonantic than either Tunica or 
Atakapa, being marked, indeed, for its strong tendency to conso- 
nantic clusters. On the other hand, Tunica is distinguished for the 
great number of disyllabic stems. On close examination, however, 
we find that the typical phonetic combination in the principal stems 
is the consonant-vowel-consonant pattern, the highest proportion 
being found in Atakapa, the next in Chitimacha, and the lowest in 
Tunica. Affixes consisting of single sounds are more common in 
Chitimacha and Atakapa than in Tunica. 
