Generalizing a result of Babenko and Pichugov, it is shown that if T is a weakly compact operator on L1 (It), where 1A is a a-finite nonatomic measure, then III + TIl = 1 + 1ITIh. A characterization of all operators T on L' (p) having this property is also given. In [2] Daugavet proved that if T is a compact operator on C(,i), then I + TiI = 1 + 11ThI, while Babenko and Pichugov [1] subsequently showed the same is true for a compact operator on L'(IL) (see [61 for recent extensions). In general, if X is a Banach space and T E L(X), then T is said to satisfy Daugavet's equation [4] if III TiI = 1 + 1.ITIh. This property of an operator arises naturally in the consideration of problems of best approximation in function spaces where it has been utilized by a number of authors (e.g. [4], and the references cited in [1]). In a recent paper [5] it was shown that Daugavet's equation actually holds for any weakly compact operator on C(,u). The purpose of this note is to give an analogous extension of the theorem of Babenko and Pichugov to weakly compact operators by proving that if , is a a-finite nonatomic measure on a space S, and T is a weakly compact operator on L1(I), then IlI + TII = 1 + IITII (Theorem 1). Related results concerning extensions and generalizations of this theorem, including a characterization of all operators T on L1 (it) for which IIITII = 1+ lT II (Theorem 2) are also given. We begin with a simple result concerning the norming of operators on Ll (,u). As usual we denote by L(X) the space of all bounded linear operators on a Banach space X, by XA the characteristic function of a measurable set A c S, and by IIf! II and IgIk0, the norms of functions f E L1(,V) and g E L??(,i), respectively. The complement of a set A in S is denoted by A'. PROPOSITION 1. Let T E L(L1('i)). Then given any E > 0 there is a measurable set A C S for which 0 ITIh E. PROOF. Since it is a-finite, (Ll(IL))* = L??(,t). Therefore if T E LZ(L'(,it)), then T* E L(LOO (,i)) and IITII = IIT*II = sup11 g 11= =T* g I I Hence given any e > 0 there is a function g E L??(I,) for which higilgo = 1 and IIT*glloo > ITIh e/2. Since IIT*g I I = ess supt I(T*g)(t)I it follows that there is a measurable set A C S for which m(A) > 0 and I(T*g)(t)l > 11TIh E for all t E S. Replacing A (if necessary) by a subset B C A for which 0 < m(B) < E and on which the sign of (T*g)(t) is constant (which may be done since it is nonatomic), and replacing g by (-g) if Received by the editors April 7, 1986. 1980 Mathematics Sject Clssfication (1985 Revision). Primary 47B38, 47A30; Secondary 46E15.