Copulae is a growing field of interest and application for dependency modelling. There is however no predominant way of choosing the copula model that best fits a given data set. We introduce a new goodness-of-fit test, based on the probability integral transform. The test is consistent, numerically efficient and incorporates a weighting functionality. Results show that the test performs well and that the weighting functionality is very powerful. Applied to stock portfolios the test strongly rejects the Gaussian and the Clayton copulae, while the Student's t copula provides a good fit.