Stable isotope analysis (SIA) has emerged as an important tool for understanding consumer diets and diet shifts. However, although the general idea behind SIA is clear, the interpretation of data is often fraught with problems because tissue turnover and fractionations are not known. We investigated shifts in stable isotope composition of spiders following a diet shift, using mealworms fed either maize (C4) or wheat (C3) flour. Mealworms had different carbon isotope composition depending on their diet and this difference was reflected in spider body parts. In the experiment, we first fed the spiders on a diet of either maize-fed or wheat-fed mealworms and then switched diet at the time of the second molt. Spiders were then sampled repeatedly until the next molt. We sampled both legs and abdomens, as these are presumed to have different turnover of tissue, and also molt remains were sampled when this was relevant. The data indicated that the spider legs had a turnover of about 20 days, whereas the spider abdomens had a turnover of about 8 days. Molt remains had the slowest turnover and reflected the diet at the previous molt, when the exoskeleton was formed. Both these observations indicate that SIA may be successfully used for elucidating diet shifts. More problematic was the fact that fractionation of carbon isotope ratios varied with body parts and diets. When spiders were fed maize-mealworms then the fractionation was larger for abdomens, but when the spiders were fed wheat-mealworms then the fractionation was larger for legs. The mechanisms underlying this pattern are unclear and deserve further attention.