Background. The aim of this research was to observe the changes which take place in the electrophoretic picture of milk proteins after pasteurisation and inoculation with diff erent starter cultures (both traditional and probiotic). After incubation, the yoghurt, kefi r, acidifi ed milk, fermented Bifi dobacterium bifi dum drink and Lactobacillus acidophillus drink were chilled for 14 days to observe the changes which occurred. Materials and methods. The research materials were raw and pasteurised milk, as well as fermented milkbased drinks. The raw milk used for research came from Polish Holstein-Fresian black and white cows. The milk was sampled 3 times and divided into 5 parts, each of which was pasteurised at 95°C for 10 min and then cooled for inoculation: yoghurt to 45°C, kefi r and acidifi ed milk to 22°C and drinks with Bifi dobacterium bifi dum and Lactobacillus acidophillus to 38°C. Milk was inoculated with lyophilised, direct vat starter cultures, in an amount equal to 2% of the working starter. For the production of fermented drinks, the subsequent starters were applied: "YC-180" Christian Hansen for yoghurt, "D" Biolacta-Texel-Rhodia for kefi r, CH-N--11 Christian Hansen for acidifi ed milk, starter by Christian Hansen for the probiotic Bifi dobacterium bifi dum milk, starter by Biolacta-Texel-Rhodia for the probiotic Lactobacillus acidophillus milk. The analyses were conducted in raw, pasteurised and freshly fermented milk as well as in milk drinks stored for 14 days. The total solid content was estimated by the drying method; the fat content by the Gerber method; the lactose content by the Bertrand method; the protein content by the Kjeldahl method with Buchi apparatus; the density of milk was measured with lactodensimeter; acidity with a pH-meter; and potential acidity by Soxhlet-Henkl method (AOAC, 1990). The electrophoretic separation of proteins in raw and pasteurised milk, as well as in freshly produced milk drinks and those stored for 14 days, was performed with SDS-PAGE (on polyacrylamid gel) basing on procedure described by Laemmli (1970). Results. It was shown that, in comparison with raw milk, the pasteurised milk had smaller amounts of αs-, β-and κ-casein, whereas the shares of γ-casein and peptides were greater, and there were no changes in immunoglobulin, α-lactalbumin or β-lactoglobulin levels, which indicated that hydrolysis of caseins had occurred. In all freshly fermented milk drinks, a drop in αs-and β-casein was observed relative to raw milk. An increase in peptides and γ-casein was also noticed (with the exception of acidifi ed milk). There were diff erences in α-lactalbumin and β-lactoglobulin levels between the diff erent drinks: raw, pasteurised or freshly fermented milk. It was shown that kefi r, compared to the other drinks, had the lowest levels of αs-and β-casein, α-lactalbumin and of peptides, as well as the highest level of γ-casein, which is evidence of an increased rate of hydrolysis in that drink. It was stated that, during the storage of fermented milk drin...