with no evidence of screen blinding. Although it appears contrary to good practice to add sugar to a stream from which sugar is to be recovered, it is the overall pol loss that counts, and treatment is justified if a reduction in pol loss can be demonstrated.