Empirical estimates for the target strength of krill are extracted from publications. These are confined to measurements on aggregations of live euphausiids and should not be affected by a frequent cause of bias in single-animal measurements, namely thresholding. Theoretical estimates for the target strength are derived from the deformed-cylinder scattering model assuming specific sets of physical and orientational parameters, for which there is an empirical basis. The theoretical estimates show a non-monotonic dependence of target strength on both animal size and transmit frequency, notwithstanding admitted shortcomings. Some recent single-animal measurements of target strength for live euphausiids and euphausiid-related species, made under high signal-to-noise-ratio conditions, are consistent with the general pattern. Several specific recommendations are made for future, improved determinations of krill target strength. Based on the comparisons, general prediction curves for the target strength are presented that are applicable to a wide range of lengths, acoustic frequencies and orientation parameters.