Who says that social media cannot provide some impulses to study again what one believes one already knows! Recently someone asked a numeral-related question on Tibetan (or rDzong-kha) orthography followed by many clever answers. I resisted the temptation to answer and desisted from answering it because I wished to take a look at Beyer 1991 again. The initial question is, if I recall correctly, something like why do we write sum stong and not stong gsum. My short random answer would have been, that as a rule, an elision of prescript (sngon ’jug) would take place in the case of certain numbers if they happen to be the first member of disyllabic numeral compound. Thus, sum stong (but stong gsum), chig stong (but stong gcig), and so on. Cf. also sum rtsen, nyis ldab, etc. But the rule does not apply to all numeral words that have prescripts. For example, one cannot write zhi stong (as opposed to stong phrag bzhi). The reason may be that orthography such as zhi stong is bound to have ambiguity whereas sum stong not. Actually several insightful observations have already been made in Beyer 1991: 221–225, and any attempt to say anything on it seems like trying to reinvent the wheel. There are, however, some points to rediscover. Note, for example, rtsa in nyi shu rtsa gcig and brgya rtsa gcig. Thus nyer gcig is a contraction of nyi shu rtsa gcig (Beyer). Similarly, note so in so gcig, zhe in zhe gcig, nga in nga gcig, re in re gcig, don in don gcig, gya in brgyad bcu gya gcig, go in dgu bcu go gcig, and so forth. Actually one may ask why nyer, zhe, nga, re, don, gya, and go. There seem to infinite grammatical, syntactical, and orthographical rules to describe and to prescribe!