832 PROCEEDINGS OF SECTION F. 
relationship of the Oceanic to the Siamese, Canobodian, and 
Indo-European respectively; for instance, Muller and Kean 
did not find -a single Oceanic numeral word in Siamese and 
Cambodian, and after the most careful investigation of these 
twelve Oceanic numeral words, Bopp could find only six which 
he deemed identical with the corresponding Indo-European 
numeral words ; of the personal pronouns, only those of the first 
and second persons with the Indo-European of these persons ; 
of the formative suffixes, only possibly and doubtfully one with 
an Indo-European suffix, and of the formative prefixes none 
he deemed identical in Indo-European. Now, not to mention 
other things, these formative prefixes are a well known distinctive 
feature of the grammatical structure of the Oceanic family, and so 
important that any method that not only does not account for all 
of them as is done above, but not even for one of them, is plainly 
inadequate; also, in the method of this paper no use of or reference 
to “roots” is made. All the Oceanic words are formed words, and 
as such are compared both as to phonisis and meaning with the 
corresponding formed Semitic words, this being deemed the right 
method. Notsowith Bopp. For instance, the Oceanic word tas, 
the sea, is compared with the Arabic word taé or tasu, the sea ; 
but Bopp compares it with the Sanscrit root si¢ (or stk), to moisten, 
whence sikta, moisten, siéaka, cloud, in which, to say nothing 
more, the meaning is not the same. 
Tt is worthy of special notice that the Indo-European pronouns 
of the first and second persons only five of the six numeral words 
compared by Bopp with the Oceanic, namely, those for 2, 3, 4, 6, 
and 7, are the Indo-European pronouns and numerals, usually 
compared with the corresponding Semitic as remarkably alike, 
while it is admitted that such resemblances, which may be 
casual, do not establish the relationship of the Indo-European to 
the Semitic family. In order to that, the grammatical structure 
of the two families must be taken into account; and as to 
grammatical structure, proof is not found either between the 
Indo-European or the Semitic, or between the Oceanic and the 
Indo-European, but is found, as has been pointed out, between 
the Oceanic and the Semitic. Assuming that the facts sub- 
stantially are as have been set forth in this paper, then it is 
absolutely certain, not only that the Oceanic words and formative 
particles dealt with are purely the words and particles of the 
distinctively Semitic stock, but also that the characteristic 
grammatical structure as a whole of the Oceanic is that which as 
a whole is distinctively and exclusively characteristic of the 
Semitic family of languages. I may here add that the peculiar 
gutturals, or the guttural phonology, of the Semitic family of 
languages is an important part of this whole—that is, of this 
distinctive or exclusively characteristic grammatical structure ; 
