Despite advances in molecular systematics, the taxonomy of tardigrades still depends largely on morphological and morphometric traits. The fact that the variability of any biological trait is determined by the interaction between genetics and environment prompts a very fundamental question: is it possible for tardigrades of the same genotype, but originating from various habitats that differ in environmental conditions, to have phenotypes so different that they would be erroneously classified as different species taxa by means of classical taxonomy? Here, we present the results of a broad and fully controlled laboratory experiment in which we investigated the phenotypic plasticity of a number of traits that are traditionally considered to be taxonomically important. In order to achieve this, we have cultured six tardigrade species belonging to four eutardigrade families (Milnesiidae, Hypsibiidae, Isohypsibiidae, and Macrobiotidae) under five experimental regimes, reflecting key environmental factors that are likely to vary in natural habitats (i.e. temperature and food availability). We then measured a number of key taxonomic traits and compared their dimensions between the treatments. Over two years of experimentation we have obtained more than 28 000 morphometric measurements for over 2300 individuals. Such an extensive data set allowed us to test some of the fundamental assumptions of classic tardigrade taxonomy. We found that in the five parachelan species analysed, the great majority of both absolute and relative traits differed significantly between the treatments, whereas there were no significant differences in the apochelan species. Overall, tardigrades grew largest under the low-temperature treatment, whereas the smallest specimens were observed under high-temperature and low-food regimes. However, the prevalent statistical significance resulted mainly from the considerable statistical power of our analyses, and not from effect sizes, which varied mostly between low and moderate. In other words, the differences, although consistent, were minor in terms of taxonomical significance, and probably would not be considered by classic taxonomists as sufficient to designate animals from different treatments as separate taxa.