At present, there are three ways to determine effectively the quality of the cryopreservation procedure using ovarian tissue before the re-implantation treatment: evaluation of follicles after post-thawing xenotransplantation to SCID mouse, in-vitro culture in a large volume of culture medium under constant agitation and culture on embryonic chorio-allantoic membrane within a hen's eggs. The aim of this study was to compare the two methods, culture in vitro and culture on embryonic chorioallantoic membrane (CAM) of cryopreserved human ovarian medulla-contained and medulla-free cortex. Ovarian fragments were divided into small pieces (1.5-2.0×1.0-1.2×0.8-1.5) of two types, cortex with medulla and medulla-free cortex, frozen, thawed and randomly divided into the following four groups. Group 1: medulla-free cortex cultured in vitro for 8 days in large volume of medium with mechanical agitation, Group 2: medulla-containing cortex cultured in vitro, Group 3: medulla-free cortex cultured in CAM-system for 5 days, Group 4: medulla-containing cortex cultured in CAM-system. The efficacy of the tissue culture was evaluated by the development of follicles and by intensiveness of angiogenesis in the tissue (von Willebrand factor and Desmin). For Group 1, 2, 3 and 4, respectively 85%, 85%, 87% and 84% of the follicles were morphologically normal (P>0.1). The immunohistochemical analysis showed that angiogenesis detected by von Willebrand factor was lower in groups 1 and 3 (medulla-free cortex). Neo-vascularisation (by Desmin) was observed only in ovarian tissue of Group 4 (medulla-contained cortex after CAM-culture). It appears that the presence of medulla in ovarian pieces is beneficial for post-thaw development of cryopreserved human ovarian tissue. For medical practice it is recommended for evaluation of post-warming ovarian tissue to use the CAM-system as a valuable alternative to xenotransplantation and for cryopreservation of these tissues to prepare ovarian medulla-contained strips.