The ratio between 2nd and 4th digit length (2D:4D) may be a negative correlate of prenatal testosterone. This possibility has led to a number of studies of 2D:4D and its relationship with sexual orientation and other sex-dependent traits. At first, 2D:4D ratio was calculated from measurements made directly on the fingers but recently a number of studies have used measurements from photocopies of the hands. Here, we compared finger lengths (2D, 3D, 4D, and 5D) and ratios obtained from these two measurement techniques. Our sample consisted of 30 homosexual men and 50 men and 70 women who were not selected for their sexual orientation. We found evidence that (1)2D:4D from photocopies tended to be lower than that from direct measurements, (2) there were differences in finger lengths such that 2D from photocopies tended to be shorter or equal in length to direct measurements, while 4D from photocopies tended to be longer or equal in length to direct measurements, (3) the sex differences in 2D:4D tended to be stronger for photocopy measurements, and (4) the pattern for length differences across 2D to 5D appeared to be different for homosexual men compared to men and women recruited without regard to sexual orientation. We conclude that there are differences in digit ratios obtained from photocopies and direct measurements, and these differences arise from length differences recorded from the different protocols. Therefore, 2D:4D ratios obtained from photocopies and direct measurements should not be combined within one study nor should they be used together in comparative studies. We suggest that finger length differences between the two techniques could result from the shapes of fat-pads at the tips of the fingers and these may be dependent on sex and sexual orientation.