The service performance of anodised aluminium in architecture is dependent upon the control of the anodising, colouring and sealing processes. The quality of the anodic oxide film is determined by assessment tests. The use of a nitric acid ‘predip’ on soft films was shown to dissolve a soft outerlayer. A new admittance test was devised in which the ‘drift’, Y, was measured using 33%, nitric acid. This test threw light on the sealing quality of films exposed to the environment and those sealed in nickel salt solutions. Well-sealed, hard films gave a Y value of 0.1 to 0.2. A sealing mechanism for nickel acetate solutions is proposed in which ageing of a Ni-A1 complex is thought to play a significant role. The admittance-abrasion (Y-?) profile was used to supplement other test data. An accelated sulphur dioxide test was investigated, the optimum conditions for which were 40°C for 6 hours. The presence of a sealing bloom was determined by a simple go/no go three paper abrasion test. The assumption is made that Yt (25°C ) = a + bt and linear regression has been used to find b. It has been found that for batches having a range of thicknesses but a common sealing time, the linear relation is a reasonable assumption when the gradient is < +10. The admittance values of electrocoloured tin black film s have been found to be anomalous. The gradients are negative due to the presence of tin in the pores. Linear regression has been used to obtain an average value of b over the whole thickness range, which gives information as to the degree of sealing. Colouring and soaking in sulphuric acid at elevated temperatures (before sealing) reduces the abrasion resistance. The abrasive wheel test was also used to show the depth of tin within the pores. The use of the addition agent ‘Tribrite’ in tin colouring solutions suppresses the outward, whiskery growth of tin in the pores.