In a recent article’) I was able to show that the vowel system of Old Chinese could be anaiysed as a two-term close/open contrast, s/a. This IF very different from Karlgren’s ‘Archaic Chinese’ but IS remarkably slmllar In esscntlal features to the analyses of Modern Pekingese by Hartmann,?) Hockett3j and Rygaloff.4j (See also the further study by Mart&) and the more recent prosodic anaiysis by I-ialhday.“jj Hartmann adopted the theory of three basic vowels (high, mid, Iowj developed, but not published, by Trager and Kennedy. He wrote these vowels as 1. e, u, recogmzmg as well three semlvowels J, r, w Hackett found It unnecessary to retain a separate sign for the high nuclear vowel, which he regarded Instead as zero vocahsm. By writing I and u Instead of 1 and \t he gave his transcrlptlon the appearance of recogmzmg more ‘vowels’ than Hartmann’s but this IS a mere: matter of orthography because he still called I and II semlvowels, and only e and CI vowels. (More recently’) he ellmmates a separate class of semlvowels and dlstlngutshes four high vowels I, u (previously IU), r, U, apart from the mtd and low vowels a and a This seems to me to be a retrograde step ) Rygaloff also reduced the number of nuclear vowels to two but m a ‘) E G.Pulleyblank, ‘An mterpretatlon of the vowel systems of Old Chmese and of