1853.] On the Mongolian Affinities of the Caucasians. 61 



Eoot repeated with ba suffix har- 

 monised and serial as in Circas- 

 Chi-chi-bi in Manyak. sian. This feature of the numeral 



serviles is of frequent occurrence. 

 See Essay on Bodo and Dhimali 

 for two good samples. 



P. S. The above paper has been considerably augmented in num- 

 ber of vocables, and in the analysis of them, since it was first pre- 

 sented to the Society, though not to the extent I had hoped and 

 purposed if health had not failed me. If however the principles of 

 the analysis (sufficiently revealed in their application and in the 

 observations of this and the following paper) be sound, they may be 

 easily carried as much further as is desired. 



"With regard to the soundness of those principles I am fully pre- 

 pared for censure of the presumption of attempting to analyse un- 

 known tongues ; — prepared also to see many errors of detail detected, 

 to afford apparent justification of such censure. 



I can but solicit the particular attention of the candid to the per- 

 fect uniformity of the phenomena presented by the vocables, whether 

 nouns, pronouns or numerals, from the very beginning to the very 

 end of my paper, and ask, How this is to be explained, except upon 

 those principles which a comparison of the numerous Himalayan 

 tongues with each other and with that of Tibet led me first to detect, 

 and which my opportunities of novel exploration beyond the Hima- 

 laya afforded me great advantages for testing the more extended 

 application of ? I have to regret that my investigations have been 

 interrupted just when they were beginning to produce their ripest 

 fruit, and to solicit the Society's favourable construction of what is 

 now submitted, as it is, rather than trust to an uncertain future for 

 its improvement. 



Supplement to the paper on the Mongolian Affinities of the 



Since the above paper was hastily written I have obtained through 

 the courteous aid of our Secretary the loan of the Mithridates and 

 Asia Polyglotta. The ampler stock of Caucasian and Mongolian 

 vocables thus placed within my reach (and illustrated too by occa- 

 sional analytical notices) has needed only to be compared with my 



