It is shown that the generally accepted interpretation of experimentally determined widths of the meta-stable zone gives misleading information about the relationship between primary nucleation rate and supersaturation. Evidence is given that nuclei have to reach a certain minimum size before they are detected. The width of the meta-stable zone for unseeded solutions is determined by both nucleation process and measuring technique as opposed to meta-stable zone width for seeded solutions which are mainly determined by the measuring technique. The results show that caution must be exercised in applying experimentally determined width of the meta-stable zone to crystallizer design.