In a previous study we analyzed the possible relationship between dairy product consumption and multiple sclerosis (MS) worldwide. We showed that a good correlation (Spearman rank ρ = 0.836), statistically strongly significant (p < 0.0001), existed between liquid cow milk consumption and MS prevalence. The interpretation of this strong correlation between MS and milk consumption is still unclear: fresh milk could be considered as a cofactor, but it could also reflect a much stronger association with MS of another unstudied factor, well correlated with milk consumption (yet, this is not the case for latitude). Obviously, the bovine population in each country and, particularly milk cows, has to be considered. In the present study, we analyze the correlations existing between the figures of national, cow milk production and MS prevalence in 20 countries. We also analyze the correlations with the whole bovine, ovine, caprine, porcine, horse, poultry, cat and dog populations. Here again we find significant correlations between (i) cow milk production per inhabitant, (ii) national bovine density per inhabitant, and (iii) local bovine geographic density, and MS prevalence. However, these correlations are relatively weaker than that found with fresh liquid milk consumption in our previous study. No correlation is found with other farm animals or with pets in the same countries. The epidemiological significance of these results, suggesting a preponderant role of fresh cow milk, is discussed.