Objectives: The purpose of this study is to determine if it is possible to estimate the sex of skeletal remains in the Croatian population using morphological and osteometric analysis of orbital mesurements from CT imaging. Additionally, this study aims to determine if orbital measurements are accurate and reproducible with virtual representation as well as to compare results from the Croatian population to other populations in order to improve the information in the worldwide database. Methods: Two hundred CT scans of the craniums of Croatian patients were included, with equal proportion of male and female (each 100) and ages raging from 18 to 92. Intact orbital margins were used for measurements based on a total of six different variables (orbital breadth, orbital height, biorbital breadth, interorbital breadth, bimaxillary breadth, and zygoorbitale breadth) for metric and morphological analysis. Normality of distribution was tested with Shapiro-Wilk test and performed in SPSS with the statistical significance set at P≤0.05. The precision of the discriminant functions and sectioning points was calculated on the basis of standardized and unstandardized coefficients, as well as the structural matrix. The accuracy of the discriminant functions was evaluated in the original and cross-validated sample (only cross-validated results using leave-one-out rule are shown). Results: Inter-observer error produced values of rTEM and R that caused the exclusion of interorbital breadth, bimaxillary breadth, zygoorbitale breadth, and the left-sided orbital breadth. Weighed Kappa (0.41520) was calculated for inter-observer error morphology and demonstrated a moderate agreement. No significant statistical differences were calculated between the right and the left side of orbital width and height. All measurements tabulated proved to be of significant statistical difference between males and females except for interorbital breadth (P=0.114). Morphology demonstrated high sexual dimorphism (P<0.001) and classification rate (94.5%). The orbital measurements varied in their overall accuracy ranging from 55% (zygoorbital breadth) to 72.5% (biorbital breadth) and 73.5% (left orbital breadth), but with discriminant functions of classification rates 69-70.5%. Conclusion: The more valuable orbital measurements for sex estimation in the Croatian population are the orbital width and height as well as the biorbital breadth, whereas the interorbital breadth does not show a significant difference in Croatian males and females. However, these measurements provide a classification rate/accuracy with the highest value amongst all of 70.5% which is not sufficiently reliable for the orbit to be used alone in sex determination analysis.