Correlated errors of experimental data are a common but often neglected problem in physical sciences. Various tools are provided here for thorough propagation of uncertainties in cases of correlated errors. Discussed are techniques especially applicable to three-isotope plots common in geo- and cosmochemistry, where common denominators in isotope ratios, as well as instrumental effects, such as mass-dependent isotope fractionation, could lead to significant correlation of errors. Furthermore, various techniques for calculating linear regressions are compared to each other, showing that the method suggested by Mahon (1996)  gives the best results after correcting for some typographical errors therein and eliminating a minor mistake. We also provide a method for calculating linear regressions through a fixed point, avoiding previously made mistakes.