Abstract
The normalization correction introduced by McCarthy and Walker is examined. It is concluded that it should be absent.