80 BUREAU OF AMERICAN ETHNOLOGY [Bunn 44 
Cukra in the interior, midway between the Bluefields and Segovia 
rivers. Reclus (283) locates them well up the Segovia river. This 
author, however, gives the Carca as a different tribe. 
As has been seen, Brinton places the Ulva (Ulua, Woolwa, Walwa, 
Smoos, Sumo) on the headwaters of the Bluefields river; Squier, on 
the middle course of the same river. Squier locates the Melchora 
immediately east of the southern end of Lake Nicaragua. The name 
Sumo (or Smoos) appears to be used rather indefinitely, but more 
generally as an equivalent of the stock name (Ulvan), the people 
embraced being considered as properly forming but one tribe, and the 
above-named supposed tribes as mere minor and local subdivisions, 
It is probable that the Ulvan dialects were related to Chibcha, but 
for the present it has been thought best to keep them distinct. 
RAMA 
As stated by Brinton and Sapper, the Indians speaking this language 
are restricted at present to a small island in the Bluefields lagoon, and 
were confined to the same island at the time Bell lived in the Mos- 
quito territory (1846-1862). There isevidence, however, that formerly 
they occupied a much larger area on the neighboring mainland, but 
whether this region lay along the Bluefields river or farther south it 
is impossible to decide with certainty from the meager data obtainable. 
Bell (259) says: : 
The Ramas inhabit a small island at the southern extremity of Blewfields lagoon. 
They are only a miserable remnant of a numerous tribe that formerly lived on the 
St. Johns and other rivers in that neighborhood. <A great number of them still live at 
the head of the Rio Frio, which runs into the St. Johns river [Rio San Juan] at San 
Carlos fort. . 
Those at the head of the Rio Frio, Costa Rica, are without doubt 
the Guatuso. . 
Squier (4: 366) locates them between the Bluefields and San Juan - 
rivers, indicating, as does Bell, a former more southerly habitat. 
This conclusion agrees with the indications furnished by the very 
brief vocabulary of the language which has been obtained, and which 
shows slight affinity with the Talamancan dialects, but a closer rela- 
tion with those of the Doraskean group of the Chibchan family. 
Following Brinton, the writer has associated it with the latter. 
Bell’s supposition that the Rama are identical with the people living 
on the Rio Frio, Costa Rica—that is to say, with the Guatuso—is, 
however, an error, as appears from comparison of the languages of the 
two peoples and from the great difference in their characteristics so 
far as known, although both belong to the Chibchan stock. 
MosaqutTro 
The mixed race designated by this name inhabits the Gulf coast of 
Honduras and Nicaragua from Cape Gracias southward to a point 
