Sex Determination
Accuracy in sex determination for non-human primates approaches 100% given the significant degree of sexual dimorphism present in our close relatives. Although differences in human male/female skeletal anatomy is not nearly as marked,  when multiple indicators are used, accuracy in determination approaches 90 percent.

Standard osteological aspects of the cranium, appendicular skeleton, and pelvis were employed in the determination of adult sex.  Table 1 lists some of the features used for the St. Stephenâs collection, with a brief definition of male/female differences for each landmark.   Although there is a clear range of overlap between the sexes for many measures, the trend for Homo sapiens is a pattern of greater robusticity in males compared to females.

Table 1:  (coming soon)

Skull:  Beginning with the cranium, robusticity of the browridge, mastoid process, nuchal crest, temporal lines, and mandible were employed.  While the female skull retains gracile attributes seen in the pre-pubescent skull, the male cranium becomes markedly rougher in adulthood at the sites of muscle attachment.  Although distinct, the ability to quantify measures of the skull for sex determination has met with limited success.   Hence, the non-metric characteristics of the cranium were more heavily weighted for this phase of the analysis.  The results for each feature, analyzed independently for the disarticulated remains, showed considerable consistency as seen in Table 2.   The five features of the skull demonstrated a 95% male ratio.  In addition, the 6 intact cranium from this collection also showed distinctively male attributes for all features, in a manner where more than one indicator could be tested per individual.

Table 2:
Long Bones:  In the postcranial skeleton, differences in the diameter of the humeral head, femoral head, and bicondylar breadth were used for sex determination, and sites of attachment for the deltoid and hamstring muscles were analyzed for degree of robusticity.  Metric analyses consistently provided a lower ratio of males to females when compared to their non-metric counterparts.  For example, humeral head diameter indicated a 71% male presence in contrast to the 97% ratio found for deltoid tuberosity robusticity.   A similar pattern was found in the femur, with ratios of 63% and 68% using femoral head diameter and biocondylar breadth (respectively), in contrast to linea aspera robusticity which yielded 96% males.  Krogman succinctly addressed this problem at the beginning of his chapter on sex determination when he stated "here is the problem of subjectivity versus objectivity, of description versus measurement, of Îexperienceâ versus statistical Îstandardization.â   Meindl and coworkers later analyzed metric versus non-metric observation and found no significant difference in accuracy "so long as care and skill [were] employed".

As mentioned in the age determination analysis, the disarticulated and commingled nature of the collection precluded analysis of multiple measures of sex in a single individual.  Also, the number of broken long bones further complicated analysis by making it difficult to analyze several measures (such as femoral head diameter and bicondylar breadth) in the same bone.  For each of the metric methods employed, it should be noted that the number of indeterminate specimens which preserved the requisite characteristics were quite high.  For the humeral head, the indeterminate number equaled the number of males (n=27), and was more than double the number of females (n=11).  For the femoral condyles, the indeterminate specimens were quite high (n=33) compared to the male (n=58) and female (n=27) counts.

Innominates:   Factors contributing to the overall shape of the pelvis are constrained by both the demands of bipedal locomotion, as well as those for perpetuating the species.  This region of the skeleton accordingly shows considerable sexual dimorphism, principally in relation to the requirements of childbirth.  Morphology of the pelvis is thus an especially diagnostic tool for sex determination, since female pelvic geometry permits a greater outlet for passage of the neonate head and shoulders during birth.  In particular, features of the innominate such as shape of the pubic bone, elevation of the auricular surface, size/shape of the sciatic notch, size/shape of the acetabulum, presence of a preauricular sulcus,  and curvature and width of sacrum, were useful in this study.

In females, the body of the pubis is longer, thereby increasing the size of the pelvic outlet.  Although 47 pubic symphyseal faces were preserved (as previously discussed for the Todd aging method), only 36 full pubic bodies with a portion of ischiopubic ramus  were available for sex determination.  Of those, 89% were characteristically male, demonstrating a triangular shape with a broad medial aspect and no evidence of a ventral arc.  The female pattern for these features is a rectangular pubis, pronounced ventral arc, and sharp, narrow medial aspect of the ischiopubic ramus.  According to Bass, the presence of a ventral arc is the most diagnostic of the female pubic features.

The greater sciatic notch is an often-used preliminary tool for sex determination, though not the most accurate.   The dimorphic pattern is a wide notch in females, narrow opening in males.  The size and shape of the acetabulum can also be utilized, with males showing greater robusticity in this feature corresponding to the generally greater size of the male femur head with which it articulates.  In the St. Stephenâs collection, 76% of the specimens preserving a sciatic notch were classified as males.  Of those with a complete acetabulum, 98% appeared male.

Elevation of the auricular surface (a landmark previously mentioned for age determination) is another dimorphic feature of the innominate, and one of the few that can be reliably used even on subadults (over age 10).   Females present a completely raised face at this sacroilliac articulation.  Male auricular surfaces are not elevated around the entirety of the articular margin.   Over 75% of the innominates demonstrated the male pattern for the St. Stephenâs collection.
Finally, the presence of a preauricular sulcus is a female characteristic, associated with the elevated auricular surface.  In the past this features was linked to parturition.   Although not present in all females, when it is present, it is considered a distinctively female trait.  Eight individuals presented this characteristic, or 7% of the collection.

Sacrum:  As a rule, the female sacrum is wider and flatter than that of the male, a morphology again permitting a greater outlet for the birth canal.  In the St. Stephenâs collection, 97% of the sacra showed the narrowness and curvature of males.  However, this is a rather subjective observation, and should only be used in conjunction with other techniques of sex determination.

Sex Determination Summary:    The average percentage of males using all features discussed above was 87%.  However, this gives equal credence to each method employed, a potentially unwarranted assumption since not all the methods are evenly weighted for accuracy.  In particular, features of the innominate are far more diagnostic than others.   Krogman ranked accuracy of sex determination using the pelvis at 95%, followed by the skull at 92%, the mandible alone at 90%, and long bone measures at 80%.   Stewart indicated slightly lower yields, however the order of accuracy was the same with the long bones again the least accurate.   For the St. Stephenâs collection, the pelvis produced an 87% male rate, the mandible 99%, the skull 95%, and the combined long bone features 79%.   If the long bone metric and non-metric values are excluded, the average percentage of males represented in this collection is 93%.

Regardless of the method utilized, the vast majority of individuals represented in the Byzantine St. Stephenâs collection are male.  This is a pattern consistent with the copious historical literature indicating a male monastic community at the site during the Byzantine period.


1.   T. WHITE, Reconstruction, 320-329.

2.   The criteria in Table 5 were compiled from descriptions in: W. BASS, Human Osteology:  A Laboratory and Field Manual, 4th ed., (Columbia, Missouri:  Missouri Archaeological Society, 1995); T. ROGERS and S. SAUNDERS,  "Accuracy of Sex Determination Using Morphological Traits of the Human Pelvis", JFS 39 no 4 (1994), 1047-56, esp. 1049; D. MITTLER and S. SHERIDAN, "Sex Determination in Subadults Using Auricular Surface Morphology:  A Forensic Science Perspective", JFS 37 no 4 (1992), 1068-75; W. KROGMAN, The Human Skeleton in Forensic Medicine (Springfield, Illinois:  Charles Thomas, 1962), 112-52.

3.   T. STEWART, Essential of Forensic Anthropology (Springfield, Illinois:  Charles Thomas, 1979), 87-93.

4.   Analysis of non-metric traits was conducted by the author.  Measurements for the metric indicators were collected by:  Colin OâNeil (currently with the US Navy), John Cheadle (University of Maryland), Rebeccah Sanders (Fulbright Fellow), and Anthony Schafer (University of Wisconsin).

5.   W. KROGMAN, The Human Skeleton, 112.

6.   R. MEINDL, C. LOVEJOY, R. MENSFORTH, and L. DON CARLOS, "Accuracy and Direction of Error in the Sexing of the Skeleton: Implications for Paleodemography", AJPA 68 (1985), 85.

7.   It should be noted that sex determination for the innominate is based upon differential growth of the female pelvis.  Thus, it has been found that borderline cases have a tendency to be classified as male (ROGERS and SAUNDERS, "Accuracy", 1047-56).

8.   This allowed using the Phenice method [T. PHENICE, "A Newly Developed Visual Method of Sexing the Os Pubis", AJPA 30 (1969), 297-302] which utilized the shape of the pubis, presence/absence of the ventral arc, and the medial aspect of the ischio-pubic ramus.  T. WHITE (Reconstruction, 325) stated a 96-100% success rate for the method, but cautioned that accuracy is reduced in older individuals.

9.   W. BASS, Human Osteology, 202.

10. G. LETTERMAN, "The Greater Sciatic Notch in American Whites and Negroes", AJPA 28 (1941), 99-116; S. WASHBURN, "Sex Differences in the Pubic Bone of Bantu and Bushman", AJPA 7 (1949), 425-32;  B. BOUCHER, "Sex Differences in the Foetal Pelvis", AJPA 15 (1957), 581-600.

11. MITTLER and SHERIDAN, "Sex Determination", 1068-75.

12. T. STEWART , Essentials, 107.

13. T. STEWART, Essentials, 103.

14. W. BASS, Human Osteology, 200.

15. W. KROGMAN, The Human Skeleton, 112

16. T. STEWART, "Medico-legal Aspects of the Skeleton. I. Age, Sex, Race, and Stature", AJPA 6 no 3 (1948), 315-21.

17. It should be noted that there was an apparent taboo against placing women in monastic burial grounds as discussed by J. PATRICH, Sabas, Leader of Palestinian Monasticism:  A Comparative Study in Eastern Monasticism, Fourth to Seventh Centuries.  (Washington DC:  Dumbarton Oaks, 1995), 35;  and H. GOLDFUS, B. ARUBAS, and E. ALLIATA, "The Monastery of St. Theoctistus", LA 45 (1995), 247-92, esp. 274.  Goldfus et. al. mention Choziba as an exception to the rule during this period.

Home | Introduction | History | Monastery Tour | Tomb Tour Students | Faculty | Field Seasons | Collections | Publications | Presentations | Funding | Future | Related Links