Routine semen examination remains an important tool for the diagnosis and treatment in human subfertility. Of all semen parameters, sperm morphology seems to be one of the most powerful indicators of a man's fertilizing potential in vitro and in vivo. Lack of standardization of sperm morphology assessments remains the main reason for the usefulness of this parameter. The aim of this study was to analyze the agreement between the wet-stained preparations versus those stained with modified Diff-Quik for sperm morphology. A total of 100 unselected semen samples from infertile couples were analyzed. Sperm morphology was evaluated with unstained specimens and following modified Diff-Quik staining according to the strict (Kruger classification) criteria by two different examiners (intralaboratory blind assessment). Mean percentages of morphologically normal spermatozoa were identical on wet and stained preparation slides (4.79 vs. 4.61, p >.05). Wide divergence of results was found with respect to the percentage of sperm with head and midpiece defects with the two different preparations (p>.001). The percentage of sperm tail defects was similar in both methods (p>.05). Simple linear regression analysis between the two methods revealed very good correlation for the morphologically normal spermatozoa (r=.83), but poor correlation for the sperm head, midpiece. and tail defects (r=.25. .25. and .28, respectively). Wet preparation is suitable only for the morphologically normal spermatozoa, but to determine the percentage of the defective spermatozoa, staining the smear is recommended.