128 Sif an and Horsok Vocabularies. [No. 2. 



from 1st, a greater or lesser use of the prefixed, infixed and postfixed 

 particles, amounting to nearly constant employment of some or all of 

 them in some tongues, and to nearly total* disuse of some or all of 

 them in others. 2nd, from a preference by one tongue of the pre- 

 fixes, of the infixes by another, and of the suffixes by a third. 3rd, 

 from that transposed position and function of the primary and seconda- 

 ry part of wordsf (root and particle) which is a law of these lan- 

 guages eminently obscurative of identities in its partial operation. 

 4th, from the substitution of a reiterated root, for a root and particle 

 in the composition of words when the various meanings of the root 

 might otherwise transcend the differencing power of the particles, or 

 at all events, not satisfy the demand for an unusually broad distinc- 

 tion. J 5th, from the disjunct or conjunct (elided vowel) method of 



* The disuse or non-use is often only apparent, for the surplus " silent" letters 

 are really prefixes, with a blended instead of a separate utterance. That this is so 

 may be proved to demonstration by identity of function (differential) in the two : 

 and yet the blended or separate utterance makes all the difference between mono- 

 syllabism and its opposite, besides causing other differences that are apt to conceal 

 the essential identity of words. See analysis of Caucasian and Mongolian words 

 in appendix to my last communication. 



f Compare overleap and leap over : what holds good chiefly as to our verbs, 

 holds good equally as to the verbs and nouns of these tongues wherein indeed the 

 two classes of words are but faintly distinguishable, or not at all so. Abundant 

 fresh evidence of the law may be found by comparing Leyden's Indo-Chinese with 

 my Tibeto- Himalayan vocabularies : compare mim-ma and sa-mi, Burmese, with 

 mi-sa, Newari, Root mi ; and ma-nek, Burmese, with Nyi-ma, Tibetan, Root Nyi. 

 Day, sun and morning, when compared speak for themselves. 



J In Gyarung the root pye, bird, is so near to the root pe, father, that they have 

 been segregated by the application to one of the usual prefix, to the other of the 

 iterative principle, or root repeated, whence tape, a father, and pye pye, a bird, 

 forsan et pe pe. I might add, as a 5th cause of difference between these tongues, 

 the different degrees in which each employs the tonic or accentual variant, which 

 principle has been most erroneously supposed to be exclusively Chinese and Indo- 

 Chinese, whereas it prevails far and wide, only more or less developed ; most where 

 the servile particles and so-called silent letters are least in use ; least, where they 

 are most in use; so that the differential and equivalent function of all three pecu- 

 liarities, that is, of " empty words," (see Chinese Grammar) of " silent letters" 

 and of tones is placed in a clear light such as Remusat vainly strove to throw upon 

 one of the three, viewing it separately. See Recherches sur les langues Tartares, 

 p. 355-7, Vol. I. DeCoros strangely enough says nothing about tones or servile 



