Update to Hyperidentities in groups and semigroups, Aequationes Mathematicae 23 (1981) 50-65.
In the last sentence of the paper, I ask whether `every variety of groups is residually finite'; i.e., worded more correctly, whether the free groups in such varieties are residually finite; equivalently, whether such varieties are generated by their finite groups. 

Alexander Olshanskii has pointed out to me that Hanna Neumann [10] noted that varieties of Burnside groups of appropriate prime exponents give negative answers to this, and that another sort of counterexample is given in [11]. 

[10]  Hanna Neumann, Varieties of groups, Springer-Verlag 1967 x+192 pp.  MR0215899

[11]  A. Yu. Olshanskii, Varieties in which all finite groups are abelian, Matem. Sbornik, 126 (1985) N.1, 59-82.  MR0773429