Sperm morphology has been suggested to be one of the most reliable factor in predicting male fertility potential both in spontaneous cycles and in assisted reproduction. In this work the morphology of spermatozoa selected with three different techniques (layering, swimup from pellet and mini-percoll) has been assessed and compared in 20 infertile patients. All the techniques allowed the recovery of sperm populations of a better quality than in basal samples, with a higher percentage of normal forms (p < 0.001). No significant differences were observed among the three different techniques in selecting sperm populations. The scatter diagrams comparing pairwise differences between the methods against their means regard the percentage of normal forms selected show a great (over 20%) inter and intra-individual variability. These data demonstrate that the three techniques select different percentages of normal spermatozoa, even in the same patient. Since it is crucial to obtain the highest number of morphologically normal and motile spermatozoa and each technique may provide different results in a given patient, we suggest that at least two techniques of sperm selection be performed before assisted reproduction procedures. This combined testing could also be relevant in the prognostic evaluation of the infertile male, since it explores the different characteristics of sperm function.