Qualitative and quantitative analysis of DNA methylation in situ at the level of cells, chromosomes, and the chromosomal domain is extremely important in diagnosis and treatment of various pathologies, as well in studies of aging and the effects of environmental factors. Yet, the questions remain unresolved of whether the detectable in situ methylation patterns correspond to the actual DNA methylation per se and/or reflect the accessibility of DNA to antibodies, which depends on the structural features of chromatin and chromosome condensation. Thus, this phenomenon can result in an incorrect determination of the real DNA methylation pattern. In order to eliminate this disadvantage to the extent possible, we modified the commonly used methodology of in situ detection methylcytosine by means of monoclonal antibodies. In this study, we show that the efficiency of immunofluorescent labeling for 5-methylcytosin in centromeric heterochromatin, chromosome arms and sister chromatids is significantly affected by the conditions of pretreatment of chromosome preparations. We used undifferentiated murine embryonic F9 cells to show that variations in the conditions of storage of chromosome preparations can lead to a sharp reduction of labeling intensity and even disappearance of the fluorescence signal in centromeric heterochromatin. Using the developed method, we discovered asymmetric methylation of sister chromatids in F9 cells and in human peripheral blood lymphocytes. This phenomena can lead to asymmetric cell division and asymmetric transcriptional status in daughter cells. Thus, the modified methodology for detection of 5-methyl cytosine in situ can provide for a more precise assessment of methylation of chromosomes and chromosomal regions.