It is pointed out in response to Hradil and Bajer [preceding Comment,
Phys. Rev. A 48, 1717 (1993)] that C(M) and S(M) cannot both be measur
ed together by scheme 1, whereas they can by scheme 2. There is no pha
se angle whose cosine and sine both vanish, and putting C + iS equal t
o zero for certain measurement outcomes breaks the connection with pha
se.