FROM CENTRAL INDIA TO POLYNESIA. 167 



lables. This has usually, as in the " Austroasiatic" languages, 

 been clone by means of some prefix, and it is therefore as a 

 rule the last syllable of a Malayo-Polynesian stem-word that 

 represents the original root. But sometimes an infix and 

 occasionally a suffix appears to have been used. Professor 

 Schmidt's view is that the Malayo-Polynesian roots were 

 originally all monosyllabic and that the modern stem-words of 

 two syllables with which we are familiar have been formed 

 from the original roots by the agglutination of formative ele- 

 ments. These last, he conceives, once had a more or less 

 definite function in the way of differentiating the meaning of 

 the root, but have now become quite fossilised and being no 

 longer separable from the root are regarded as an essential 

 part of the word. Thus it is that the Malayo-Polynesian lan- 

 guages possess very few monosyllabic words but a very large 

 proportion of words of two syllables. 



If this view of the structure of the Malayo-Polynesian 

 languages is correct' (and it is certainly consistent with the 

 results of the study of those languages by several independent 

 scholars of great authority), clearly a great step has been taken 

 towards bridging the apparent gulf between them and the 

 " Austroasiatic" family. It must further be observed that on 

 this old fossilised structure the Malayo-Polynesian tongues 

 have superimposed a newer system of formatives which serve 

 the purpose of differentiating grammatical functions. Thus 

 they deal with their stem-words in much the same w 7 ay as 

 they are supposed to have dealt' (and the " Austrosiatic" lan- 

 guages are known to have dealt) with the original monosyllabic 

 root- words. There is a considerable amount of analogy be- 

 tween these different families of spee&h in the use and even in 

 the form of the prefixes and infixes which they respectively 

 employ. The most striking cases, perhaps, are the prefix pa-, 

 which in the Mon-Khmer languages, Nicobareseand the Malayo- 

 Polynesian family forms causatives, and the infixes -n- and -m- 

 which, with somewhat varying functions, aie found in a good 

 many of these different languages. There are other points of 

 grammatical analogy enumerated in Professor Schmidt's ar- 

 ticle: but they are somewhat less cogent and it would take up 



R. A. Soc, No. 53, 1909. 



