.194 
ON TIIE MALAYAN AND POLYNESIAN 
undergo in Malay or Javanese, does not exceed twelve; and sulur, 
the root in question, could not he subjected even to one half this 
number, not one of which would correspond in sound or sense with 
any one of the Malagas! compounds. 
The very length of these Malagas! compounds appears to me to 
be good evidence against the allegation that the Malagasi is of Ma¬ 
layan origin. The great majority of Malay and Javanese roots are 
bisyllables; but in the Malagasi they frequently extend to four or 
even five syllables ; and when to these are added, not monosyllabic 
prefixes or affixes, as in Malay and Javanese, but sometimes prefixes 
or affixes, of two, three, and even of four syllables, the monstrous 
length of some compounds may readily be supposed From the root 
suht already mentioned, although only of two syllables, is formed, for 
example, the compound mamprfampanolo, which means, “ to order 
to cause to exchange,” being a word of six syllables, of which the 
languages of the Malayan family afford not one example. But words 
of even double this length may be formed ! 
I come now to the evidence afforded by words. The Malagasi 
Dictionary, already quoted, contains about 8000 words, exclusive of 
compounds. I have gone carefully over it more than once, and can 
discover no more than 140 which are of Malayan origin, which 
would make about A,th part of the language. 
Of 
But to the dictionary is appended a list of words especially called 
roots. These amount to 500; and among them I find just six Ma¬ 
layan words, and no more. 
The nature of the Malayan words found in the Malagasi, is of 
much importance in the inquiry. Sixty are the names of natural 
objects, and thirteen are numerals. There is no preposition among 
them, no auxiliary verb, nor any other word essential to the struc¬ 
ture of a sentence. The language, in a word, might be written or 
spoken without them, with far more ease, and that is not difficult, 
than good English can be written or spoken without the assistance 
of the Norman-French portion of it. 
The Malayan words received into the Malagasi are, with few ex¬ 
ception, corrupted in sound, a result to be expected from the differ- 
