Streaking artifacts in tomographic images reconstructed by the filtered-backprojection algorithm are caused by aliasing errors in the projection data. To show this a computer simulation study was performed in which the transforms of undersampled projections were subtracted from the corresponding transforms when the projection data were taken with a very large number of rays. This yielded the aliased spectrum for the undersampled case. An image was reconstructed from the difference transforms. Streaks present in this image exactly matched those present in the undersampled reconstruction. (The number of projections used in this study was large enough to preclude any artifacts caused by their insufficient number.) We have derived a theoretical upper bound for the energy contained in these aliasing artifacts. In this paper we have also briefly touched upon the artifacts caused by other algorithmic aspects of a tomographic system.