Many studies have drawn attention to the emerging role of MSC (mesenchymal stem cells) as a promising population supporting new clinical concepts in cellular therapy. However, the sources from which these cells can be isolated are still under discussion. Whereas BM (bone marrow) is presented as the main source of MSC, despite the invasive procedure related to this source, the possibility of isolating sufficient numbers of these cells from UCB (umbilical cord blood) remains controversial. Here, we present the results of experiments aimed at isolating MSC from UCB, BM and UCM (umbilical cord matrix) using different methods of isolation and various culture media that summarize the main procedures and criteria reported in the literature. Whereas isolation of MSC were successful from BM (10:10) and (UCM) (8:8), only one cord blood sample (1:15) gave rise to MSC using various culture media [DMEM (Dulbecco's modified Eagle's medium) +5% platelet lysate, DMEM+10% FBS (fetal bovine serum), DMEM+10% human UCB serum, MSCGM] and different isolation methods [plastic adherence of total MNC (mononuclear cells), CD3+/CD19+/CD14+/CD38+-depleted MNC and CD133+- or LNGFR+-enriched MNC]. MSC from UCM and BM were able to differentiate into adipocytes, osteocytes and hepatocytes. The expansion potential was highest for MSC from UCM. The two cell populations had CD90+/CD73+/CD105+ phenotype with the additional expression of SSEA4 and LNGFR for BM MSC. These results clearly exclude UCB from the list of MSC sources for clinical use and propose instead UCM as a rich, non-invasive and abundant source of MSC.