1, 2, and 5 are not true anv more. One can construct ACKNOWLEDGMENTS examples which show that &n analog to Lojasiewicz's lemma does not hold in this case. The reason is that one I Want to thank Dr. Uta Vijlkel for many cannot draw any conclusion from the symmetric part of discussions. This work was inspired by discussions on the first derivative of a function on the svmmetric Dart Lorentz transformations and equal-time limits withThe connection between causality and analyticity in scattering theory is formulated in terms of Hilhertspace concepts. The usual rules of nonrelativistic quantum mechanics are assumed to hold for the "in" and "out" states of the scattering system. We show that there is a (physically verifiable) causality condition which implies that each diagonal S-matrix element S,,(E) must he the limit of an analytic function of the energy E , regular in ImE>O. The implications for partial-wave amplitudes and for the forward scattering amplitude in elastic two-body collisions are discussed.