In footnote 11 of chapter 3, I claim that P. Ljungløf showed that P is the intersection closure of the class of parallel multiple context-free languages (PMCFL). This is not in fact what Ljungløf has shown. It is not known whether what I have attributed to Ljungløf is true, but if it were, then because PMCFL⊆LOGCFL, and because LOGCFL is closed under intersection, it would imply that LOGCFL = P, which is believed not to be the case. (Thanks to Makoto Kanazawa for correcting me on this.)
In section D-1.1 (page 248) I claim that primacy is not efficiently testable. That this is not true was proven by Agrawal, Kayal and Saxena in their 2004 paper, PRIMES is in P, which appears in the Annals of Mathematics 160, pages 781-793.