A recent literature review reported negative relationships between diet discrimination factors (DDFs = Xfish – Xfood; X = δ15N or δ13C) and the values of δ15N and δ13C in the food of wild organisms but there has been no laboratory-based confirmation of these relationships to date. Laboratory reared guppies (Poecilia reticulata) fed a series of diets with a range of δ13C (−22.9 to −6.6‰) and δ15N (6.5 to 1586‰) values were used to magnify diet-tissue dynamics in order to calculate DDFs once the fish had achieved equilibrium with each of the diets. Values of DDFs range widely for δ15N (7.1 to −849‰) and δ13C (1.1 to −7.0‰) and showed a strong negative correlation with the stable isotope value in the food for δ15N (slope = −0.59 ± 0.02, r2 = 0.95) and δ13C (slope = −0.56 ± 0.02, r2 = 0.94). Based on these relationships, the magnitude of DDF change over environmentally relevant values of δ15N or δ13C would be significant and could confound the interpretation of stable isotopes in the environment. Using highly enriched experimental diets, our study adds to a growing number of studies that undermine the consistent trophic enrichment paradigm with results that demonstrate the currently poor mechanistic understanding of how DDFs arise. The results of our study highlight that the magnitude of the stable isotope values in prey must be considered when choosing DDF values. Future laboratory studies should therefore be directed at uncovering the mechanistic basis of DDFs and, like others before, we recommend the determination of diet-dependent DDFs under laboratory conditions before modeling dietary proportions or calculating trophic positions. Copyright © 2010 John Wiley & Sons, Ltd.