If transparency is measured using samples on which a black ink has been preprinted on the reference substrate by the user, care should be taken in the selection of the black ink used. A black must be selected which exhibits little or no bronzing and which exhibits a similar level of gloss to the same sample when overprinted by the test ink. Even the pre-printed contrast card can give rise to anomalous results if the gloss change is significant. Any measure of transparency could be affected by such surface phenomena and care should be taken to select materials which minimize them. If this is not done, erroneous results may well be obtained. In severe cases this may even arise with a 0/45 or 45/0 geometry. Such results are normally obvious from simple visual assessment of the samples.