Human X and Y spermatozoa were previously compared by several nonmolecular techniques. Recent studies show that in many of the previous investigations, the methods used to identify the spermatozoa were nonspecific and thus produced contradictory findings. In the present study, the comparison of the 2 germ cell types, X and Y, were performed following fluorescence in situ hybridization (FISH), which is the most reliable genotyping technique currently available. The FISH technique was performed under 3 different treatments: permeabilization with liquid N2, fixation with Carnoy's, and chromatin decondensation with lithium di-iodosalicylate. Mature and immature germ cells (spermatozoa and spermatids) were compared. Lithium showed higher hybridization efficiency, while liquid N2 and Carnoy's fixative maintained better morphological integrity of cells with lower hybridization. The sperm exhibiting hybridization signals were not different in any of the morphometric or qualitative comparisons from those that did not exhibit signals. No significant deviation of the sex ratio from 1:1 was seen in either the mature or immature germ cell population. The spatial distribution of X and Y chromosome-specific signals in the sperm head were identical. The hybridization treatments did not have any preferential effect on the cells of specific genotype (X or Y). Neither head parameters (length, HL; width, HW; area, HA) nor tail length (TL) significantly differed between X and Y populations of spermatozoa under any of the treatments. Similarly, the haploid, X-specific round cells did not differ from Y-specific ones by their size (diameter) and shape. These results indicate that neither mature sperm nor their precursors possess significant morphological differences between X and Y genotypes.