Current injection testing (CIT) is a powerful method for verifying the performance of earthing systems, but it is not without its challenges. Mutual coupling effects may introduce significant errors when measuring low impedance earthing systems, especially where those systems are complex. Choice of an appropriate injection circuit becomes more critical, though physical or operational limitations result in the need for compromises. This paper presents a case study which demonstrates these effects, where CIT was completed at an underground power station and supporting above-ground infrastructure. Both overland leads and a 220 kV circuit outage were used with above and below ground voltage traverses, and very different results obtained.