LANGUAGES OF THE KILIMA-NJARO DISTRICT. 495 
to include it in my vocabularies, otherwise than to 
point out an occasional correspondence of terms. 
For the system of orthography used in spelling these 
languages, I would refer the reader to the explanation 
given on page 476 in connection with the Masai 
vocabulary. In the phonology of these languages 
several facts require to be noticed. 
Ki-caga , which has evidently been an isolated tongue 
for several centuries, surrounded by an ocean of Masai, 
has developed rather a peculiar pronunciation. Some 
of the vowels, especially u, have acquired a much duller 
and more indistinct sound than is usual among Bantu 
languages, where ordinarily the vowels have a clear, 
open sound, as in Italian. In Ki-caga it is often 
difficult to distinguish between a and u (u) and i. In 
many words where u is the tonic vowel in other forms 
of Bantu speech, it is changed to i in Ki-caga. 
G often becomes h, and sometimes y. H may re¬ 
place several consonants, but generally stands where 
a g is used in other tongues. L as an initial is very 
frequently changed to n, 9 and sometimes r, while r is 
an exceeding common sound, being constantly sub¬ 
stituted for t, d, and z. Sometimes d stands for t , and 
after the former letter a “ trilled ” sound like an r is 
often heard, so that the listener hardly knows whether 
to write such a word as wanda , ‘ 6 place, 5 ’ ivanda, or 
wandra. F before i often becomes s (si), andp changes 
to w. In Ki-taveita g often becomes y, but never h. 
D sometimes replaces t, and so does s in a few words, 
but otherwise t holds its place and does not change to 
r or h, as is so frequently the case in Ki-caga or 
Kinyika. V has a tendency to become w, and _p does 
9 Where this occurs in the case of the 5th prefix Li, which I 
believe is derived from an older form Ndi , it is probably only a case 
of the n surviving. 
