Non-Metric Variation of the Infracranial Skeleton

Home , Trochanteric fossa

J. Anat. (1978), 125, 1, pp. 23-37 23 With 3 figures Printed in Great Britain Non-metric variation of the infracranial skeleton

MICHAEL FINNEGAN Osteology Laboratory, Kansas State University, Manhattan, Kansas 66506, U.S.A. (Accepted 28 October 1976)

INTRODUCTION Recently, non-metric (i.e. qualitative) variation has begun to play a more central role in osteological analysis as a result of the work of Berry & Berry (1967, 1971, 1972) in England, Czarnetzki (1971, 1972) in Germany, Knip (1970, 1971) in Holland, Larnach & MacIntosh (1970, 1971, 1972) in Australia, Brunner (1972) in Switzerland, S. Kaul (1973, personal communication) in India, Vargas (1973) in Mexico, and Pucciarelli (1971, 1972) in Argentina. In the United States, recent studies have been undertaken by Sublett (1965), Ossenberg (1969, 1970), Corruccini (1972, 1974), Merbs (1967), Rightmire (1972), Birkby (1973), McWilliams (1970, 1974), Buikstra (1972), Melbye (1969), Zegura (1974), Cybulski (1972, 1973), Gaherty (1974), Korey (1970) and Finnegan (1972, 1974b). Most of these workers have studied regional and continental populations and have utilized cranial non-metrics, population affinity and micro-evolutionary trends. Each justifies the non-metric approach on a quasi-genetic basis, quoting Berry & Berry and Ossenberg, and makes use of the statistics developed by C. A. B. Smith (see Berry & Berry, 1967). Some researchers, however, have queried certain basic assump- tions, asking to what degree the sides, in cases of bilateral traits, and the sexes, can be pooled in making comparisons between populations, and to what extent the traits are age dependent. Thus, Korey (1970) studied side independence, sex independence and age dependency, Buikstra (1972) the age correlates of non-metric traits, while Corruccini (1974) made a detailed investigation of side and sex dimorphism and age dependency in regard to 72 non-metric cranial traits. Finnegan (1974a) compared 30 infracranial non-metric traits (Table 1) with a number of cranial non-metric traits as indicators of population distance between Coast Eskimo, Yukon Eskimo and Aleut skeletons. He found that infracranial trait distances correlated very highly (r = 0-906; P < 0-01) with cranial trait distances. At that time he suggested that infracranial traits might be better suited to non-metric analysis than cranial traits, for the following reasons: (1) All traits considered have the possibility of bilateral expression. (2) Most of the traits are found on heavy bone materials which are most apt to survive prolonged burial and subsequent excavation. (3) Many of these traits have a long history of studies dealing with sex and side dimorphism. However, Finnegan (1972) also indicated that more research would have to be done on these traits, using populations where sex, race and age are better controlled. The purpose of this paper is to present side-to-side differences, sex dimorphism and age dependency in regard to 30 infracranial traits. 24 MICHAEL FINNEGAN

MATERIALS AND METHODS The Terry Collection, housed at the Smithsonian Institution, was used as the basic population from which the sample was taken. The Terry Collection is unique and very useful to present purposes as it is well documented for race, sex, age, and cause of death. Various measurements are listed and geographical and other information is often recorded. The two racial groups best represented in the collection, and so used for this study, are American Whites and American Blacks. The Whites are probably a more homogeneous group: it has been suggested that the White genetic admixture in the Black population ranges from 10-30 % (see Glass & Li, 1953; Roberts, 1955; Glass, 1955). In choosing samples for study, only those specimens which showed accurate ages on their morgue or hospital records were considered. Their sex was certain. In all, 50 white and 50 black males and 50 black and 46 white females were selected. It was hoped to find 10 specimens of the four categories (white male, white female, black male, black female) for each of the five age groups: 20-29, 30-39, 40-49, 50-59 and 60-69, but only six white females could be found between the ages of 20 and 29. In the following description of the infracranial traits studied an effort has been made to group together variants of the same bone. 1. Allen's fossa The cervical fossa (depression) of Allen, when present, is usually located near the anterior superior margin of the femoral neck close to the border of the head. It can vary from a small depression to a large eroded area 1 cm2 where cortical bone has been lost exposing underlying trabeculae. The border of this fossa may have a ridge or thickening around it, reminiscent of an inflammatory response. To be scored as present, the underlying trabeculae must be seen, so that this variant can be differ- entiated from variants 2 and 3 (see below). 2. Poirier'sfacet Poirier's facet is scored as present when there is a noticeable, however slight, bulging of the articular surface of the femoral head toward the anterior portion of the femoral neck. This facet is necessarily smooth, and is not to be confused with plaque formation (see 3 below). 3. Plaque This variant is found in the same region as 1 and 2 above. Plaque formation is scored as present when an overgrowth or bony scar can be defined extending from the area of Poirier's facet on the femoral head down on to the femoral neck where it often surrounds or covers Allen's fossa. In rare cases all of the above variants may be defined on a single femur. 4. Hypotrochantericfossa The hypotrochanteric fossa is located in the superior posterior part of the femoral diaphysis between the gluteal ridge and the lateral margin. This variant is found in close association, but not correlated, with the gluteal ridge and the third trochanter (see below). Although various authors have defined a variety of ways to score the fossa, we use present or absent, realizing that interpretation is often equivocal. Non-metric infracranial variation 25

5. Exostosis in trochantericfossa Rarely there is a tubercle in the trochanteric fossa. More frequently bony spicules or an exostosis are present in an otherwise smooth fossa. The tubercle may be long and have cusps like a molar tooth. We consider the fossa as having two variations: smooth surface (absent), tubercle or exostosis (present). 6. Third trochanter Often a rounded tubercle can be found at the superior end of the gluteal crest. It resembles the lesser trochanter in being an oblong, rounded or conical bony tuberosity. It develops instead of, or above, the gluteal ridge. The difference between a third trochanter and a gluteal ridge is therefore one of kind and not one of degree of development. 7. Medial tibial squattingfacet The lower margin of the anterior surface of the tibia presents a rough transverse depression for the attachment of the articular capsule of the ankle joint. This depression can usually be divided into medial and lateral fossae separated by a raised area. These fossae usually show obvious vascular pitting. Frequently, the inferior articular surface is extended into the medial fossa and this extension is scored as a medial squatting facet. 8. Lateral tibial squattingfacet This occurs when the inferior articular surface extends into the lateral fossa of the transverse depression (see above). 9. Supracondyloidprocess Rarely, a small bony process arising from the medial supracondylar ridge 5-7 cm above the medial epicondyle may be observed. It is usually pointed, with the point directed downward, anteriorly and medially. It may vary in length from 2 to 20 mm. 10. Septal aperture Frequently an aperture is found in the bony septum that separates the olecranon from the coronoid fossa. When it is definitely present it is scored, but caution is necessary in archaeological material because the aperture may be a post-mortem artifact. 11. Acetabular crease Often there is a fold, pleat or crease on the articular surface of the acetabulum. This defect may arise anywhere along a line from the acetabular fossa superiorly to the border of the articular surface. The defect is not a remnant of faulty fusion of the ilium with either the pubis or ischium, as the pleat usually lies between these junctures. 12. Pre-auricular sulcus Frequently there exists a sulcus running inferiorly and posteriorly between the greater sciatic notch and the auricular surface. The sulcus may be, in part, the result of a morphological elevation of the inferior portion of the auricular surface. This 26 MICHAEL FINNEGAN variant is only scored as present when it is defined below the arcuate line, thus con- fining it to the true pelvis. 13. Accessory sacralfacets An articular facet can often be observed either on the sacrum, ilium or both, posterior to the auricular surface. On the ilium, it is located on the iliac tuberosity for the sacroiliac ligament, while on the sacrum it is observed on the sacral tuberosity. In order to be scored as present, the facet must lie below the superior border of the auricular surface, and must be distinguishable and discrete from the auricular surface itself. 14. Acromial articularfacet An articular facet may be present on the inferior surface of the acromial process. This facet is usually oval, with the long axis running posteriorly and inferiorly from the attachment of the coracoacromial ligament. 15. Suprascapularforamen Rarely, the suprascapular notch is converted to a foramen by complete ossification of the suprascapular ligament. Only the complete foramen is scored as present. 16. Circumflex sulcus Often, a sulcus for the circumflex artery can be found on the posterolateral border of the scapula nearly bisecting the origin of the teres minor muscle. 17. The vastus notch A small notch is often observed in the superolateral angle of the patella. This is scored when the area for the insertion of the vastus lateralis tendon shows some concavity. The border ofthis notch must be smooth (see below) to be scored as present. 18. Vastusfossa The vastus fossa, when present, is a small depression just anterior to the vastus notch, although it can be defined with or without the presence of the notch. 19. Emarginate patella Emarginate patella (bipartite patella) is scored when a notch is present in the superolateral border of the patella, as is the vastus notch. However, in this case, the notch is noticeably larger and, unlike the vastus notch, the emarginate notch is very rough, leaving the impression that a portion of the bone had been pulled away from the patella. This is probably the case, for occasionally the superolateral angle has survived as a separate, or as several small, ossicles. Emarginate patella does not coexist with vastus notch or fossa. 20. Os trigonum The posterior border of the talus shows a groove for the tendon of flexor hallucis longus. Immediately lateral to this groove there may be a small process known as Stieda's process. When this process is not fused, or is only partially fused, with the posterior border of the talus, it is known as the os trigonum. In archaeological material this variant can be detected when Stieda's process is present and a fusion Non-metric infracranial variation 27 line can still be ascertained, or when the facet for the articulation of the os trigonum on the talus can be observed. In the latter case, the os trigonum usually remains at the excavation site undetected. 21. Medial talarfacet The upper medial surface of the neck of the talus may present a facet which does not follow the line of curvature of the trochlear surface. Although the position of this facet is variable, it is none the less quite distinct when present. 22. Lateral talar extension An extension of the lateral third of the anterior trochlear margin on to the neck of the talus is often present. It is more common than the medial facet of the talus, but not as common as the lateral tibial squatting facet. 23. Inferior talar articular surface The inferior surface of the head of the talus may present one large articular surface, or may be divided into two surfaces which are either two discrete facets, or are continuous, but are on different planes. This variant is scored as single or double. 24. Anterior calcanealfacet double The anterior and middle calcaneal articular facets may take a variety of shapes: a single ovoid facet, a partially conjoined or hour-glass-shaped facet, or two discrete facets. This trait is scored present when two discrete facets can be observed. 25. Anterior calcanealfacet absent Rarely, when the morphology of the middle calcaneal facet appears similar to the double facet (see above), the anterior calcaneal facet is entirely missing. 26. Peroneal tubercle Three processes can be defined on the lateral surface of the calcaneus: anteriorly, a peroneal tubercle; medially, a raised surface for the calcaneo-fibular ligament; and posteriorly and inferiorly, a lateral process. The anterior or peroneal tubercle may be absent, or conjoined with the area for the calcaneo-fibular ligament, in such a way that the peroneal tubercle cannot be distinguished. When the peroneal tubercle can be defined, it is scored as present. 27. Atlasfacetform The superior articular facets of the atlas may take a variety of forms; there may be a long, more or less oval, facet, or there may be two distinct facets. The distinct facets may be separated either by a groove or a ridge of bone between the facets. This facet is scored as single or double. 28. Posterior bridge The most common variant of the atlas is a bridge of bone extending from the superior articular process posteriorly to the posterior arch. This transforms the sulcus for the vertebral artery to a tunnel. Although bony spicules may project down from the superior articular process or arise from the posterior arch, the variant is not scored unless a complete bridge can be defined (see below). 28 MICHAEL FINNEGAN

7 Figs. 1-3. The various trait states which are considered in this paper. Numbers correspond with non-metric trait descriptions in the text. Non-metric infracranial variation 29 9

.11

13 14

N%%%sl 2 ~~~~12~ ~ ~ ~ 1

,19 16 18

Fig. 2. For legend see opposite. 30 MICHAEL FINNEGAN 23

'*~~ 22 -I

20' 2X2f ~~A

24 25 /~~~~Ih ~~~~26

28 29 Fig. 3. For legend see p. 28. 29. Lateral bridge The lateral bridge is the bridge of bone extending from the superior articular process laterally to the transverse process. This bridge also creates a tunnel for the vertebral artery. Some vertebrae require close observation because combinations of bridges may be present and additional bridges may connect the lateral bridge with the posterior bridge. 30. Transverse foramen bipartite Frequently, some of the transverse foramina in the third to the seventh cervical vertebrae are bipartite. A comprehensive bibliography of these and other traits has now been published Non-metric infracranial variation 31 Table 1. Chi square (x2) valuesfor comparing left and right sidesfor each trait, sex and race Non-metric variable White female White male Negro female Negro male 1. Allen's fossa present 1-124 0-082 0-510 0-000 2. Poirier's facet present 0-105 0-087 0-041 0 053 3. Plaque formation 0091 0-521 0-398 0-040 4. Hypotrochanteric fossa present 0-196 0-118 0-380 0-089 5. Exostosis in trochanteric fossa 2-047 6-127* 0-002 0 191 6. Third trochanter present 0000 1-850 0785 0-213 7. Medial tibial squatting facet 0000 0-154 0354 0-122 8. Lateral tibial squatting facet 0-067 2-292 0-044 0-334 9. Supracondyloid process present 1-092 2-013 0000 0.000 10. Septal aperture present 0074 1-129 0458 1-449 11. Acetabular crease present 0000 0-486 0000 1P355 12. Pre-auricular sulcus present 0-596 0-432 0-552 0-075 13. Accessory sacral facets present 0 074 0 004 0-071 0-063 14. Acromial articular facet present 0-184 0 052 0-161 0-000 15. Suprascapular foramen present 0-102 1-993 0354 0*000 16. Circumflex sulcus present 0011 0-368 0-162 0-368 17. Vastus notch present 1-384 0-258 0000 0-241 18. Vastus fossa present 0-468 0-689 2-984 0-336 19. Emarginate patella 0.000 0-000 0000 1-993 20. Os trigonum present 0 754 0-000 0-272 0-272 21. Medial talar facet present 0-451 1 468 1-804 0 000 22. Lateral talar extension present 0 049 2-524 0.000 0-545 23. Inferior talar articular surface 1 032 2-297 0-668 0-379 24. Anterior calcaneal facet double 0000 0-160 0-273 0-049 25. Anterior calcaneal facet absent 0-215 1-557 1-114 2-292 26. Peroneal tubercle present 0-273 0 000 1-980 0-364 27. Atlas facet form double 0-754 1 208 1-449 0-065 28. Posterior bridge present 0-000 0-000 0000 0-065 29. Lateral bridge present 0-000 2-013 0-000 6.124* 30. Transverse foramen bipartite 0 045 0 040 0-364 0-167 * P < 005.

(Finnegan & Faust, 1974). All 30 traits were scored either as 0 = no observation; 1 = negative observation; 2 = positive observation. The statistics used throughout are based on those of C. A. B. Smith as detailed by Berry & Berry (1967). Smith's x2 statistic was used to test side-to-side differences, sex dimorphism and age dependency. Within each of these groups inter-correlations were generated, and the x2 test was also used to delineate the significant differences when the correlation coefficient showed significant age dependency.

RESULTS AND DISCUSSION Side-to-side differences White females showed no significant difference in 30 side-to-side comparisons (see Table 1). Of 30 left-right side comparisons in white males, only one significant difference was found at the 0 05 level (no. 5 - exostosis in trochanteric fossa). Negro females showed no significant differences in 30 left-right side comparisons. Negro males showed only one significant difference at the 0 05 level in 30 left-right comparisons (no. 29 - lateral bridge present). These results are similar to those found when making side comparisons on the same 30 infracranial traits using the aforementioned archaeological populations 32 MICHAEL FINNEGAN

Table 2. Chi square valuesfor comparing dimorphism between the sexes: sides and race treated separately White Negro d left c' right & left 6' right V. V. V. V. Non-metric variable V left V right 9 left 9 right 1. Allen's fossa present 1-383 0-134 3-027 1-045 2. Poirier's facet present 0-635 0031 2-465 2-549 3. Plaque formation 0-963 1-882 1-847 0855 4. Hypotrochanteric fossa present 0-429 0-587 3-193 4.488* 5. Exostosis in trochanteric fossa 0-896 3-191 3-413 2K134 6. Third trochanter present 0-210 3-203 0-183 0000 7. Medial tibial squatting facet 5-869* 7-790t 3-235 0-729 8. Lateral tibial squatting facet 1-240 0-398 8-396t 4.453* 9. Supracondyloid process present 1-287 2-124 0-000 0-000 10. Septal aperture present 1-054 3-203 6-651t 9648t 11. Acetabular crease present 1522 0-306 8-777t 3-235 12. Pre-auricular sulcus present 42-883t 64-243t 64-262t 81-566t 13. Accessory sacral facets present 1*544 2-138 0-323 0-003 14. Acromial articular facet present 3-888 1724 2-041 1-055 15. Suprascapular foramen present 3-628 8-611t 2-013 4-055* 16. Circumflex sulcus present 0-369 1-182 0-641 3-277 17. Vastus notch present 1-408 3-618 8-367t 5.828* 18. Vastus fossa present 0-060 0-018 0-271 0-387 19. Emarginate patella 0.000 0-000 1-993 0-000 20. Os trigonum present 0-678 2-922 0-272 0-272 21. Medial talar facet present 1-732 0-653 1278 0-045 22. Lateral talar extension present 1-388 0-364 0-545 0-000 23. Inferior talar articular surface 0-210 0.000 0-668 0-379 24. Anterior calcaneal facet double 0-369 0046 0-150 0-482 25. Anterior calcaneal facet absent 2-239 0-575 8-777t 0-154 26. Peroneal tubercle present 0-425 1411 0-361 0-041 27. Atlas facet form double 6.220* 0-283 5.607* 2-013 28. Posterior bridge present 2-864 2-864 2-292 3-129 29. Lateral bridge present 0430 4-181* 6.124* 0.000 30. Transverse foramen bipartite 0-038 0-031 1-993 2-582 * P < 0-05; t P < 0-01.

(Finnegan, 1974a). On the basis of side-to-side comparisons, the infracranial traits would seem to be better suited than cranial traits, as the female cranial traits from the same archaeological populations produce one difference significant at the 0-01 level (see Finnegan, 1972).

Sex difference In comparing the sexes (Table 2) it was found that 30 male-female comparisons of the left side in the White sample generated three significant differences, two at the 0'05 level and one at the 0-01 level. The traits involved were medial tibial squatting facet, atlas facet form and pre-auricular sulcus, respectively. The Negro sample produced many more significant left side differences between the sexes: two at the 005 level (atlas facet form and lateral bridge) and six at the 0-01 level (lateral tibial squatting facet, septal aperture, acetabular crease, pre-auricular sulcus, vastus notch, and anterior facet). Sex differences noted on right side comparisons in the White sample were one at the 0 05 level (lateral bridge present) and three at the 0-01 level Non-metric infracranial variation 33 (medial tibial squatting facet, pre-auricular sulcus and suprascapular foramen). As in left side comparisons, right side comparisons between the sexes in the Negro sample showed six significant differences, four at the ,0 05 level (hypotrochanteric fossa, lateral tibial squatting facet, suprascapular foramen and vastus notch), and two at the 0.01 level (septal aperture and pre-auricular sulcus). Combining left and right side comparisons in the White sample we find seven significant differences are generated in 30 trait comparisons. With cranial traits, however, 12 significant differences were found in 42 variables (Finnegan, 1972; see also Berry, 1975). Again, as with side-to-side differences, it appears that sexual differences are a little less frequent among the 30 infracranial traits. In dealing with these 30 infracranial traits we can assume that side-to-side dimorphism will not jeopardize the statistical nature of the distance statistic used. However, in the case of sex dimorphism we obviously have to make some corrections before the sexes can be summed legitimately and used in our distance statistic. Various ways of doing this have been proposed. Finnegan (1972) has suggested that, if we keep the number of males and females nearly equal in each population sample, we have adequately corrected for frequency differences displaying sexual dimorphism. Gaherty (1974) suggests that males and females be summed for all character variants which do not show sexual dimorphism, and only the male or female samples be used for those traits that do display sexual dimorphism. This seems quite reasonable and justifiable, the only disadvantage being a reduction in the sample size for those traits which show sexual dimorphism. Jantz (1970) attacked the sex dimorphism problem in his analysis of Arikara populations by eliminating all sexual dimorphic traits before his distance statistic was applied. This can be justified, but again much information is lost by the exclusion of these traits. Kellock & Parsons (1970) and Finnegan (1972) made a 'principal component' analysis of non-metric trait variation in human crania, and found that the largest Eigen value accounting for the greatest percentage of total variance was indeed a sex dimorphic trait. It seems clear that if we exclude the sex dimorphic traits from the analysis, we have excluded those traits which best distinguish and define the affinities among the populations in question. It therefore seems necessary to include sex dimorphic traits in analysing population distance. Age dependency Relatively little work has been done on age regression and age dependency in cranial and infracranial traits. Ossenberg (1970) suggests that a relatively small amount of age dependency would not greatly alter the significance ofgenetic distance estimates between populations, at least in anthropological studies. On the other hand, Korey (1970) and Buikstra (1972) suggest that the age dependency of cranial traits must be considered before a distance statistic is applied between populations. Korey suggests that significant frequency correlation over the entire age range for a number of cranial variants would affect an analysis of distance between populations where age profiles in the two groups were dissimilar. In testing the 30 infracranial traits for age dependency, the present author utilized correlation in order to show significance (Table 3), and the question was asked: if by correlation it is found that these traits are age dependent (see Table 3), are there significant differences in trait frequency between any age groups? In other words, these age-dependent traits could be highly correlated, but frequency differences in the 20-29 age group might not be significantly different from any other age group based on the X2 test. This statistical technique, 3 ANA 125 34 MICHAEL FINNEGAN

Table 3. Correlation coefficients between age and each non-metric variable: side, sex and race treated separately The x2 level is based on split-half differences and 20-29 v. 60-69 age group differences. White females White males Negro females Negro males Non-metric(______- variants Left Right Left Right Left Right Left Right 1 -0 4956t -0-3783t -0-2938*t -0-1345 -0 0994 -0-1813 -0-1886 -0-2629 2 0-1491 0*0059 -0-0680 -0-1658 0 1456 0-0663 -0-2709 -0-2512 3 0-2879 0-2784 0-1914 -0-0177 0 1109 0 0794 0-1075 0-0865 4 0-0729 0-0221 -0 0578 -0-1736 -0O0087 -0 0369 -0-1300 -0 0479 5 0-4153tt 0-1364 0*0257 0-1378 0-4959t 0-4676t 0-5936t 0-5591t 6 0-0521 0-1890 0*0059 -0-0872 -0-1431 0 2595 -0-1539 0-2020 7 000000 00000 0-0913 0-0266 0 0325 0-1388 -0 0539 -0-0668 8 0-1426 0 0288 -0-0515 0-1325 -0-1505 -0-1974 -0-0647 -0-2934*t 9 0 0939 0-0806 -0 0527 0*0000 0-0000 0-0000 0 0000 0-0000 10 -0-2212 -0-3310*t -0-0405 0 0675 -0-0236 0 0109 -0-2366 -0-1547 11 0-0924 0-0924 -0-1389 -0-1083 -0-1860 -0-1860 -0-1678 -0-0996 12 0-0979 0 0755 -0-1702 -0-6758 0-2392 0-1415 0-1502 0-0915 13 0.3192*t 0-1570 0 0010 0*0947 0 2058 0-2010 0-3561t1 0-2039 14 0-1607 0-0902 -0-0668 -0 0097 0-2832*t 0-2358 0-1315 0-1084 15 -0-0165 -0-1010 0-1593 0 0000 0-1816 0-1956 0'0000 0 0000 16 0-1858 0-1438 0 0881 0-0762 -0-0741 -0-0087 -0-0443 0-1124 17 0-0177 -0*0137 -0-0143 -0-2187 0-3088* 0.3053* 0 0334 -0-0104 18 0-2070 0 0350 0-0850 0-0864 0-1486 0-1232 0-0462 0-0883 19 0 0000 0 0000 0 0000 0 0000 0 0000 0 0000 0 0000 0 0000 20 -0-1520 0-3104*t -0 0551 0-0222 0-0654 -0-1189 0-1041 0-0954 21 -0-2037 -0-2132 -0-2611 -0-0264 -0-1004 0 0991 0.0970 0-0177 22 0-0348 -0-0366 -0-0820 0-1142 -0-1755 -0-2551 -0-2751*t -0-4389t1 23 -0-1753 -0-1167 -0 1689 -0-0842 0 0051 -0-0001 0 0173 0-0382 24 0-0221 0-0646 0 0479 0-0698 0 0740 0-1716 -0 0107 -0-0184 25 0-0332 0 0579 0 0468 0-0622 0 0567 0-1315 -0-0024 0 0631 26 0 0999 0-0229 0-0947 0 1199 0-2722 -0 0011 -0 0533 -0-1465 27 0-0221 0-0688 -0-1138 -0-0144 0-2453 -0-1047 -0 0857 0 0459 28 -0-2276 -0-2276 0-0655 -0-1135 -0 0404 0-0980 0 0997 -0 0911 29 -0-2576 -0-2576 -0-0527 000000 00000 00000 0-1749 0o0000 30 0-2057 0-2710 0-0291 0*0086 -0-0521 -0 0644 -0-2983* -0-3186*t y: * P < 005; t P < 001; x2: t P < 005. which is more robust than correlation, was employed to find out if the frequency differences between groups were significant. The above supposition was indeed confirmed by the x2 test. Twenty of 240 correlations were significant, 9 at the 0 01 level, and 11 at the 0 05 level. However, in 240 x2 examinations between the youngest and oldest age groupings within each sex, side and race, only 11 significant differences were generated - all at the 0O05 level. At this level we would expect 12 non-significant differences, and so the finding of only 11 significant differences does not exceed chance expectations. These results add credibility to the variables used, and suggest that the age-dependent nature of these traits cannot be fully expressed in terms of correlation coefficients. It is also of interest that some of the traits are more consistent in age correlation than others - both in sex and race. For example, a detailed examination of Table 3 shows Allen's fossa (trait number 1) to be significantly correlated with age in 3 of 4 cases in the White sample, but not at all in the Negro sample. Also, lateral talar extension present (trait number 22), presents a progression between sides, sexes and Non-metric infracranial variation 35 races: reading the Table from the left we find the value of the correlation increasing from the White females to the White males and Negro females to the Negro males, with only the correlations of the Negro males showing significance. However, for most of the significant correlations there exists no apparent patterning. This form ofinquiry is being continued, but already the following recommendations can be made: (1) The age-regressive nature of infracranial traits is not strong enough to warrant any correction in applying the distance statistic unless the combined populations generate a distinct bimodal age distribution. (2) If the mean ages of the groups in question are not significantly different, and if the variants within the populations can be shown to be more or less homogeneous with respect to age, then we need not make correction for age dependency.

CONCLUSIONS (1) Based on statistical techniques employed in both cranial and infracranial trait analysis it appears that side dimorphism is not a concern in the final biological distance equation. (2) Although sex dimorphism is seen in both cranial and infracranial traits, there are available methods for dealing with such differences so that the traits may be used in biological distance statistics. (3) Age dependency, at least in the infracranial skeleton, need not concern us in the distance statistic and its subsequent analysis. Corruccini (1974) has shown that this is not entirely true of cranial traits. (4) It is necessary to pay close attention to the trait descriptions themselves before we make population comparisons based on frequencies gleaned from the literature. The author has found several frequency differences in regard to White and Negro populations between the present work and that of Angel (1964), resulting from slight differences in the notation of the traits considered, and the fact that Angel utilized cadaver material rather than macerated bones. (5) Finally, based on the archaeological samples studied by the author, it appears that there is less trait loss resulting from archaeological conditions and excavation techniques in infracranial than in cranial material. Trait loss is frequent, for example, on the medial wall of the orbit.

SUMMARY 196 skeletons of known age, sex and race from the Terry Collection were studied in order to document 30 non-metric infracranial traits. Each trait had the ability to be expressed bilaterally, although significant side dimorphism was not observed. Sex differences were statistically significant for some of the traits within a racial group, but these differences were not as pronounced as the differences generated by non-metric cranial traits in the same populations, and were not effective in all racial groups. In general, these infracranial traits show some age dependency when correlation statistics are used, but this dependency is lost when the more robust X2 statistic is used. These data suggest that infracranial non-metric traits may be superior to cranial non-metric traits for population comparisons. Infracranial traits may be more durable than cranial traits having regard to the nature of most archaeological material.

3-2 36 MICHAEL FINNEGAN

REFERENCES ANGEL, J. L. (1964). The reaction area of the femoral neck. Clinical Orthopaedics 32, 130-142. BERRY, A. C. (1975). Factors affecting the incidence of non-metric skeletal variants. Journal ofAnatomy 120, 519-535. BERRY, A. C. & BERRY, R. J. (1967). Epigenetic variation in the human cranium. Journal of Anatomy 101, 361-379. BERRY, A. C. & BERRY, R. J. (1971). Origins and relationships of the ancient Egyptians. In The Population Biology of Ancient Egypt (ed. D. R. Brothwell & B. Chiarelli), pp. 199-208. London and New York: Academic Press. BERRY, A. C. & BERRY, R. J. (1972). Origins and relationships of the ancient Egyptians: based on a study of non-metrical variations in the skull. Journal ofHuman Evolution 1, 199-208. BIRKBY, W. H. (1973). Discontinuous morphological traits of the skull as population markers in the prehistoric southwest. Ph.D. dissertation, University of Arizona. BRUNNER, J. A. (1972). Die fruhmittelalterliche Bevolkerung von Bonaduz. Eine anthropologische Unter- suchung. Chur: Schriftenreihe des Ratischen Museums, Heft 14. BUIKSTRA, J. (1972). Techniques for coping with the age regressive nature of non-metric traits. Abstract. American Journal ofPhysical Anthropology 37, 431-432. CORRUCCINI, R. (1972). The biological relationships of some prehistoric and historic Pueblo populations. American Journal ofPhysical Anthropology 37, 373-388. CORRUCCINI, R. (1974). A critical examination of the meaning of discrete traits for human skeletal biological studies. American Journal ofPhysical Anthropology 40, 425-445. CYBULSKI, J. S. (1972). Discrete non-metric skeletal variants and application to the study of early North- west Coast physical variation. Paper Presented at 25th Annual Meeting of Northwest Anthropological Conference, 25 March 1972. Portland. CYBULSKI, J. S. (1973). The Gust Island burial shelter: Physical Anthropology. Mercury Series, Archae- ological Survey of Canada, paper no. 9, National Museum of Man, Ottawa, pp. 62-113. CZARNETZKI, A. (1971). Epigenetische Skelettmerkmale in Populationsvergleich. I. Rechts-links-Unter- schiede bilateral angelegter Merkmale. Zeitschrift fUir Morphologie und Anthropologie 63, 238-254. CZARNETZKI, A. (1972). Epigenetische Skelettmerkmale in Populationsvergleich. 11. Frequenzunterschiede zwischen den Geschlechtern. Zeitschrift fur Morphologie und Anthropologie 63, 341-350. FINNEGAN, M. (1972). Population definition on the Northwest Coast by analysis of discrete character variation. Ph.D. dissertation, University of Colorado. FINNEGAN, M. (1974a). Discrete non-metric variation of the post-cranial skeleton in man. Abstract. Americal Journal ofPhysical Anthropology 40, 135-136. FiNNEGAN, M. (1974b). A migration model for Northwest North America. In International Conference on the Prehistory andPaleoecology of Western North American Arctic and Subarctic (ed. S. Raymond & P. Schledermann). University of Calgary Press. FINNEGAN, M. & FAUST, M. A. (1974). Bibliography of Human and Nonhuman Non-Metric Variation. Research Reports No. 14. Department of Anthropology, University of Massachusetts, Amherst. GAHERTY, G. (1974). Discrete traits, cranial measurements, and non-biological data in Africa. Abstract. American Journal ofPhysical Anthropology 40, 136. GLASS, B. (1955). On the unlikelihood of significant admixture of genes from the North American Indians in the present composition of the Negroes of the United States. American Journal of Human Genetics 7, 368-385. GLASS, B. & Li, C. C. (1953). The dynamics of racial intermixture - an analysis based on the American Negro. American Journal ofHuman Genetics 5, 1-20. JANTZ, L. (1970). Change and variation in skeletal populations in Arikara Indians. Ph.D. dissertation, University of Kansas. KELLOCK, W. L. & PARSONS, P. A. (1970). A comparison of the incidence of minor non-metrical cranial variants in Australian aborigines with those of Melanesia and Polynesia. American Journal ofPhysical Anthropology 33, 235-240. KNIP, A. S. (1970). Metrical and non-metrical measurements on the skeletal remains of Christian populations from two sites in Sudanese Nubia. Part I and lI. Proceedings. Koninklijke nederlandse akademie van wetenschappen, Series C: Biological and Medical Sciences 73, 433-468. KNIP, A. S. (1971). The frequency of non-metrical variants in Tellem and Nokara skulls from the Mali Republic. Part I and Part II. Proceedings. Koninklijke nederlandse akademie van wetenschappen, Series C: Biological and Medical Sciences 74, 422-443. KOREY, K. (1970). Characteristics of the distributions of non-metric variants. Master's thesis, University of Chicago. LARNACH, S. L. & MACINTOSH, N. W. G. (1970). The craniology of the Aborigines of Queensland. The Oceania Monographs No. 15. LARNACH, S. L. & MACINTOSH, N. W. G. (1971). The mandible in Eastern Australian Aborigines. The Oceania Monographs No. 17. Non-metric infracranial variation 37 LARNACH, S. L. & MACINTOSH, N. W. G. (1972). The Keppel Islanders. A supplement to the craniology of Queensland. Archaeology andPhysical Anthropology in Oceania 7, 8-14. MCWILLIAMS, R. (1970). Physical anthropology of Wann and Sam, Two Fourche Maline Focus archaic sites in Eastern Oklahoma. Bulletin ofthe Oklahoma Anthropological Society 19, 101-136. MCWILLiAMS, R. (1974). The utility of the superior axis foramen in the analysis of cremated human skeletons. Abstract. American Journal ofPhysical Anthropology 40, 145. MELBYE, F. J. (1969). An analysis of a late woodland population in the Upper Great Lakes. Ph.D. dissertation, University of Toronto. MERBS, C. F. (1967). Cremated human remains from Point of Pines, Arizona. A new approach. American Antiquity 32, 498-506. OSSENBERG, N. S. (1969). Osteology of the Miller Site. Occasional Paper Number 28 Art and Archaeology, Royal Ontario Museum. OSSENBERG, N. S. (1970). The influence of artificial cranial deformation on discontinuous morphological traits. American Journal ofPhysical Anthropology 33, 357-372. PUCCIARELLI, H. M. (1971). Variaciones Craneanas En Grupos Raciales Aborigenes De La Republic Argentina. Trabajo de Tesis para optar al Doctorado an Ciencias Naturales. University Nacional De La Plata Argentina. PUCCIARELLI, H. M. (1972). Relaciones Entre Huesos Wormianos Y Otros Rasgos Neurocraneanos Sobre Un Grupo Racial Homogeneo. Anales De La Sociedad cientifica argentina 194, 233-243. RIGHTMIRE, G. P. (1972). Cranial measurements and discrete traits compared in distance studies of African Negro skulls. Human Biology 44, 263-276. ROBERTS, D. F. (1955). The dynamics of racial intermixture in the American Negro - some anthropological considerations. American Journal ofHuman Genetics 7, 361-367. SUBLETT, A. J. (1965). The Complater Project: Osteological analysis. Pennsylvania Archaeologist 2, 74-92. VARGAS, L. A. (1973). Estudio De Los Caracteres Craneanos Discontinuous En La Poblacion De Tlatilco. Tesis. Escuela Nacional De Anthropologia L Histuris, Mexico. ZEGURA, S. (1974). A comparison of distance matrices derived from craniometric measurements and cranial observations. Abstract. American Journal ofPhysical Anthropology 40, 157.