The best-characterized model pathway of protein folding, that of disul
phide bond formation in the small protein BPTI, has been questioned re
cently. A reinvestigation of that pathway, using alternative methods,
concluded that the intermediates with non-native disulphide bonds accu
mulated to lower levels than previously had been observed (17). On thi
s basis, a revised pathway was proposed that simply omitted those inte
rmediates. Even if totally correct, however, the new observations are
not inconsistent with the important characteristics of the original pa
thway and even confirmed many of them. Certain crucial observations th
at were the experimental basis for the original pathway were ignored,
and these observations invalidate the revised pathway.