ALTHOUGH MOST linguists generally concede that a great number of languages of southeast Asia probably form one linguistic family-the -this view was not scientifically established but rested until a few years ago only upon a number of common features of the languages composing the family : monosyllabism, tones, and many roots phonetically similar. But in The Vocalism of Sino-Tibetan the phonetic equations for the most important final and medial vowels of the four chief literary languages and of a representative (Lusei) of the most archaic linguistic group of the proposed family were established. In a subsequent article, Problems in Sino-Tibetan, 2 final *-I, *-r, and *-s were discussed, and the rules for the correspondence of final stop and nasal consonants were noted.3 Among the remaining problems for the complete establishment of the Sino-Tibetan family on a scientific basis are the equations for initials, prefixes, and tones. Here we aproach the most difficult problems of Sino-Tibetan. For previous investigators have found that in some of the modern dialects the development of prefixes, initials, and tones are mutually dependent;4 and we may suspect a similar mutual dependence characterized the development of at least some of the principal languages selected for our study. Moreover, if we try to confine our study to initials as much as possible, we note that within the conjugation of the same Old Bodish (classical Tibetan) verb we may find consonantal alternation of initials within the same varga, as dgens (pres.), bkcan (pf.), dgan (fut.), k'on (imper.) fill; and other scholars in the Sino-Tibetan field have noted a similar alternation of initials within the same varga in different languages of