The City University test (TCU test) aims to identify people with significant colour deficiency and to Classify the type of defect. 222 people with congenital red-green colour deficiency, diagnosed with the Nagel anomaloscope, were examined with the TCU test (2nd Edition). All deuteranopes and 44% of deuteranomalous trichromats failed the TCU test. Deutans who failed could be subdivided into two categories of severity depending on whether errors were made on five or more plates. 96% of protanopes and 26% of protanomalous trichromats failed. Protans made fewer errors than deutans and subcategories of severity could not be distinguished according to the number of errors made. The Farnsworth D15 test was found to be more effective than the TCU test in identifying significant protan colour deficiency. Detection and classification rates varied on all the plates of the TCU test. Mixed protan and deutan classification errors were made by 61% of subjects with the majority result correct in 80%. The most efficient plates are identified and recommendations are made for the optimum use of the TCU test in clinical practice.