The field of sacral isation extends to 6 elements in 119 sacra or 2 9,4 of

Transcription

1 (182 spines, Fig. 25a), whereas incomplete individuation of SI occurs in only approximately 1/5 of the spines (50/232, Fig. 25f). In the latter category, this apparent weakness at the cranial extremity of the iliac individuation field may be owing to a relatively less extensive individuation field. The smaller field is represented in the model 'f 1 in Figure 25. Further indirect evidence of a link between a high position of the ilium and incomplete individuation of SI is provided by the contrast in the findings in 37 spines in which the ilium has'apparently settled one element lower than usual at vertebra 26 (25 PSV + 5S): incomplete individuation of SI occurs in only 3 out of the 37 spines (Fig. 25i). Indeed, in those spines with the formula 25 PSV + 5S, 12 have alar facets on SI which indicates transition at the L5/S1 junction (Fig. 25h). These 12 spines have the formula C7, T12, L6, 55. A further four spines with alar facets have six lumbar vertebrae: three of the spines have the formula C7, Til, L6, S5 and are included in the 182 spines in Figure 25a, and one spine has the formula C7, Til, L_6, S4 and is included in Figure 25b. It is possible that the sixth lumbar element is sensitive to the adjacent sacralisation field and reacts by enlargement of its costal element. It is noteworthy that all the remaining 22 spines with the formula 25 PSV + 5S which do not have alar facets on SI, have five lumbar vertebra (C7, T13, L5, S5, Fig. 25g). This sensitivity of the sixth lumbar vertebra may be owing to the fact that the vertebra which is influenced is an extra lumbar vertebra and, possibly, that the ilium in its presumed migration past its usual position to element 26 nay nave settled just slightly cranial to the presumptive SI, that is, somewhat closer than usual to the extra lumbar vertebra in these 12 spines. This suggested closer proximity of the sixth lumbar to the first sacral vertebra is depicted in the model 1h in Figure 25. Three of the 37 spines with the formula 25 PSV + 5S have incomplete individuation of SI (F4g. 25i). Here the caudal migration of the ilium may heve proceeded slightly further than in the 22 spines with 25 PSV+5S (and perfect individuation of SI, model 'g ) cind/or the iliac individuation field may be sn~.* er than usual. Both possibilities are shown in model 'i ' in Figure ^ j. (c) Six Sacral Vertebrae The field of sacral isation extends to 6 elements in 119 sacra or 2 9,4 of

2 292 the total, owing, presumably, to an increase in the cranio-caudal extent of the individuation field. The larger field is indicated diagrammatically in Figure 25 in the models 'j ' to 'o'. Incomplete sacral individuation of SI is present in the majority of such spines (70,0%). In 19 spines with G-piece sacra, the il ium has apparently migrated only to the level of vertebra 24 (Fig. 25j ) with a resultant formula of 23 PSV + 6S. There is incomplete individuation of SI in all but one of the 19 spines and the mechanism is presumed to be similar to that suggested for spines with 23 PSV + 5S} i.e.'the ilium has settled higher, nearly opposite the 24th element, but not a full segment higher, with resultant incomplete individuation of SI. This relationship between the cranial end of the iliac individuation field and the transitional type SI is depicted in model j * of Figure 25. Most of these 19 spines thus support the conmonly held view that a 6-piece sacrum is produced by sacralisation of the fifth lumbar vertebra. However, in most of the spines with 6 sacral vertebrae, the ilium has seemingly migrated to its usual position opposite element 25 and there is an extra element in the total precoccygeal column, giving the formula 24 PSV t- 6S (98 spines, Fig. 25k,l,m,n). The source of the extra element may be a potential 'first coccygeal' or an extra element in the lumbo-sacral region or, rarely, both, for, in a further 2 spines there are two extra elements in the precoccygeal columns with a formula of 25 PSV + 6S (Fig. 25o). In 24 of the 100 spines with 24 PSV (or 25 PSV) + 6S there has been complete and perfect individuation or sacralisation of the extra sacral element, irrespective of the source, as shown by the lack of transitional features in SI or S6 (Fig, 25k). In 50 of the 100 spines, there are transitior.il features only at the S1/S2 junction (Fig. 251), which fact suggests either imperfect individuation of an extra lumbo-sacral element, or possibly, that the field effect of the ilium has shifted slightly caudally to sacralise a.'d assimilate Co I completely, whilst the individuation of SI is incomplete at the rostral end of the field This latter concept is depicted in the model '1' in Figure 25. A further 11 spines with transitional features in only S6 (Fig. 25m), indicate that the field of sacralisation has extended to include the potential first coccygeal element, but that the individuation of Co 1 has been incomplete. There are transitional features in both SI and S6 in 13 spines (Fig. 25n); this combination suggests incomplete individuation at both cranial and caudal extremities of the field,

3 293 irrespective of the source of the extra sacral element. In all the categories of spines with 6-piece sacra and 24 PSV (and possibly also those with 23 PSV), it is suggested that the presumed weakness at the periphery of the sacralisation field which results in incomplete individuation of SI and/or S6, may be related to the size of the field. Thus, in 6-piece sacra, the extent of the field may be smaller than is usual (for 6-piece sacra) relative to the number of elements individuated (six). A similar mechanism may operate to produce transitional features in 5-piece sacra. In addition, the cranial and caudal extremities of the smaller field may lie more caudal and more cranial to SI and S6 respectively than a field which induces six perfect or complete sacral vertebrae. This relationship of the less extensive iliac individuation field to SI and S6 is shown in Figure 25 in model 11'. Thus, two factors may operate to produce the presumed weakness at the cranial extremity of the sacralisation field in both 6-piece and 5-piece sacra. Variation '!n the precise position of the cranial limit of the iliac individuation field is thought to play a more important part in spines with 23 PSV, whereas a less extensive iliac individuation field is considered to be the major factor in spines with 24 PS V. It is noteworthy that transitional features at the S1/S2 junction are present in a high proportion of 6-piece sacra (males 69,2%, females 73,0";) and in relatively few 5-piece sacra (males 15,1%, females 28,2%). If such transitional features are indeed owing to a less extensive iliac individuation field, this may reflect the tendency which is demonstrated in all population groups studied for the sacralisation field to induce a modal number of elements (five) rather than six. Further evidence of this tendency may be the presence of transitional features at the L5/SI junction (alar facets) in 4-piece sacra. Such transitional features are owing, it is thought, to a slightly larger iliac individuation field than is necessary for the differentiation of four sacral vertebrae. The field 'overflows' on to a fifth vertebral element (the last lumbar vertebra). There was no 'overflow' of the field of sacralisation on to a last lumbar vertebra in any of the spines with six sacral vertebrae. This indicates that the sacral differentiation field is restricted to the individuation of six elements at the most. A developmental explanation for the sexual dimorphism in the variations in vertebral number and the frequency of transitional

4 293 irrespective of the source of the extra sacral element. In all the categories of spines v/ith 6-piece sacra and 24 PSV (and possibly also those with 23 PSV), it is suggested that the presumed weakness at the periphery of the sacralisation field which results in incomplete individuation of SI and/or S6, may be related to the size of the field. Thus, in 6-piece sacra, the extent of the field may be smaller than is usual (for 6-piece sacra) relative to the number of elements individuated (six). A similar mechanism may operate to produce transitional features in 5-piece sacra. In addition, the cranial and caudal extremities of the smaller field may lie more caudal and more cranial to SI and S6 respectively than a field which induces six perfect or complete sacral ve> Lcbr^e. This relationship of the less extensive iliac individuation field to SI and S6 is shown in Figure 25 in model '1 1. Thus, two factors may operate to produce the presumed weakness at the cranial extremity of the sacral isation field in both 6-piece and 5-piece sacra. Variation in the precise position of the cranial limit of the iliac individuation field is thought to play a more important, part in spines with 23 PSV, whereas a less extensive iliac individuation field is considered to be the major factor in spines with 24 PS V. It is noteworthy that transitional features at the S1/S2 junction are present in a high proportion of 6-piece sacra (males 69,2%, females 73,0«) and in relatively few 5-piece sacra (males 15,1%, females 28,2%). If such transitional features are indeed owing to a less extensive iliac individuation field, this may reflect ^he tendency which is demonstrated in all population groups studied the sacralisation field to induce a modal number of e l ement (five) rather than six. Further evidence of this tendency may be the presence of transitional features at the L5/S1 junction (alar facets) in 4-piece sacra. Such transitional features are owing, it is thought, to a slightly larger iliac individuation field than is necessary for the differentiation of four sacral vertebrae. The field 'overflows' on to a fifth vertebral element (the last lumbar vertebra). There was no 'overflow' of the field of sacralisation on to a last lumbar vertebra in any of the spines with six sacral vertebrae. This indicates that the sacral differentiation field is restricted to the individuation of six elements at the most. A developmental explanation for the sexual dimorphism in the variations in vertebral number and the frequency of transitional

5 294 features in SI at the S1/S2 junction may be a sex difference in the extent and/or migration tendencies of the iliac individuation field. With regard to the latter, the iliac individuation field presumably settles higher than usual in the spines with 23 PSV and either 5 or 6 sacral vertebrae (26 in all). There are more female (6) than male spines (1) in the category 23 PSV + SS (Fig. 25d,e) but more male (13) than female spines (6) in the category of 23 PSV + 63^ (Fig. 25j). In the total sample, however, there are relatively more female (7,6% of 157) than male spines (5,6% of 248) with 23 PSV (Table 85, p.242). In the spines with 25 PSV (44 in all), the iliac individuation field presumably settles lower than usual and this is present much more often in male (37) than in female spines (7) with 25 PSV, irrespective of whether the presacral/sacral formula is 25+4 (Fig. 25c), 25+5 (Fig. 25g,h,i) or 25+6 (Fig. 25o). There are, too, relatively more male (14,9%) than female spines (4,4%) with 25 PSV. This tendency for males to have more spines with 25 PSV than females, and for females to have more spines with 23 PSV than males has been reported by previous workers incl ng Bornstein and Peterson (1966) and in the S.A. Negro in a larger sample than the present one (de Beer Kaufman 1974). It is thus suggested that there may be a sex difference in the migration tendency of the ilium, namely a trend for it to settle lower in male than in female foetuses. Further, there appears to be a tendency towards more extensive sacral individuation in males. This would account for the trend shown in tne S.A. Negro and other population groups, for a higher number of 6-piece sacra in males than in females. In addition, in males, the increased extent of the individuation field, coupled with the tendency for the ilium to settle lower, may account for the trend for an increased number of PCV in males. In females, however, there is more often a decreased number of PCV than in males and here it is postulated that the ilium settles higher and extends more frequently to only five elements. The smaller size of the sacral auricular surface (in relation to the number of sacral pieces) in female than in male 5-piece sacra may, too, ba owing to the presumed lesser extent of the individuation field. Transitional features at the S1/S2 junction in 5-piece sacra are present significantly more often in female (28,2%) than in male sacra (15,1%). It has already been suggested that such transitional features are owing mainly to a smaller iliac individuation field. Thus, in 5-piece sacra, a sex-difference in the extent of the iliac individuation

6 295 field (smaller in females) may account for the higher frequency of transitional features at the S1/S2 junction as well as a smaller auricular surface in female sacra. Thus, the variations in the S.A. Negro precoccygeal vertebral column are such as might be expected on the basis of the developmental model outlined above. All the variations seem to be explicable in terms of a., iliac individuation field, presumably induced by the ilium, which varies in its final vertebral level, in its cranio-caudal extent and in the strength of its field effect.» 2. Basality Sitsen (1926) suggested a correlation between the exceptionally low frequency of hyperbasality in Malay sacra and squatting. However, a similar low frequency of hyper basality has not been demonstrated in S.A. Negro sacra, in spite of the fact that S.A. Negro people, too, often adopt a squatting position. Gillman (1929) suggested that basality of the sacrum might be influenced by sex since he found that all male sacra in his sample of 32 were hypobasal, whereas the female sacra were more generally homobasal. However, no significant sex differences in the categories of basality have been detected in the present S.A. Negro sample(n = 405). Instead, a relationship is demonstrated between basality and the number of sacral vertebrae. So far as I can ascertain, in the literature available to me, no attempt has previously been made to correlate basality with the number of sacrcl pieces. In addition, the concept of Wells (1963) that hyperbasal, homobasal and hypobasal sacra are produced by variation in the level of the upper surface of the body of SI relative to the cranial margin of the sacro-iliac articulation, is extended and modified to include a further category of marked hypobasality, described in this study. This, in turn, is linked to the concept of the level, extent and strength of an iliac individuation field. More than half of all sacra (55,6%) are moderately hypobasal 5-piece (n = 174) or 6-piece (n = 51) sacra. Homobasal sacra, which resemble most closely the moderately hypobasal sacra, make up a further 13,3% of all sacra and these are most commonly 5-piece (n = 48) sacra, seldom 6-piece (n = 3) or 4-piece (n = 3) sacra. In these homobasal or moderately hypobasal sacra, the ilium seemingly lies in its usual position, level with the upper border of SI (Fig. 26b) or slightly caudal to it(fig.26a) respectively. Furthermore, there is complete and perfect individuation of

7 t 296 iliac individuation SI field Figure 26 Suggested Mechanism of the Production of Basality based on Wells's (1963) Concept and on the Concept of Levei and Strength of the Upper End of the Iliac Individuation Field

8 SI in most of these 5-piece sacra (192/222), half of the 6-piece sacra (27/54) and in all 4-piece sacra. Marked hypobasality is present in 20,7% of all sacra and more of these (n = 65) are 6-piece sacra than 5-piece sacra (n = 19). The great majority of these Sc.cra (69/84), both 6- and 5-piece, show transitional features in SI at the S1/S2 junction. In these sacra, it seems that during development the cranial margin of the auricular surface of the ilium has apparently come to lie slightly more caudal to the centrum of SI than is usual (Fig. 26c); the costal element of SI would then have grown caudo-laterally to reach the cranial margin of the ilium. This results in a markedly hypobasal sacrum. This caudal shift of the iliac individuation field, results, too, in most instances, in the sacralisation of 6 elements, whilst at the rostral end of the field, the individuation of SI is incomplete and this results in transitional features at the S1/S2 junction. Hyperbasal sacra occur less often (10,4%). These sacra seldom show transitional features at the S1/S2 junction, but a high proportion (44,0%) of the 42 hyperbasal sacra show transitional features (alar facets) at the L5/S1 junction. Indeed, a total of 19 sacra have alar facets; 18 of these are hyperbasal sacra and one homobasal. The ontogenetic development of the hyperbasal sacra could be visualized as follows: the cranial margin of the ilium seemingly comes to lie just cranial to the centrum of the definitive SI (Fig 26d). The costal element of SI thus grows rostro-laterally to meet the ilium. It is drawn higher, as it were, and this results in a hyperbasal sacrum. Here, the somewhat closer proximity of the ilium to the last lumbar vertebra, produced by the cranial shifting of the ilium, results in an 'overflow' of the iliac individuation field on to the last lumbar vertebra. This in turn apparently expresses itself by enlargement of the costal element of the last lumbar vertebra with consequent alar facets on SI. No 6-piece hyperbasal sacra are found. The suggested reason for this is that the sacral individuation field does not ever extend to 7 elements and a 6-piece hyperbasal sacrum would connote extension of the cranial margin of the field on to an additional -seventh - element. Thus, the presumed variation in the precisc position of the cranial limit of the ilium relative to SI is suggested as the factor determining both the variation in basality and the transitional features in SI. This hypothesis does not take into account the fact that the rostral extremity of the ilium may have settled at eit.ier element 24, 25 or 26.

9 Wells (1963) states that when SI is the 24th vertebra, the sacra are 'almost invariably hypobasal' and when SI is the 26th, the sacra 'are most commonly hyperbasal' (p. 61). He does not,, however, include variation in sacral number in his concept. In this study, when SI is the 24th vertebra, the sacrum may be composed of either 5 or 6 vertebrae. In the 7 spines with the former, i.e. 23 PSV + 5S, two are markedly hypobasal, three are moderately hypobasal and one is homobasal. In spines with 23 PSV + 6S, all the sacra arc- not only hypobasal but markedly hypobasal. Thus, these findings agree with the correlation of Wells (1963) for 23 PSV spines, which could be modified as follows: 'when SI is the 24th vertebra, the sac^a are almost invariably hypobasal if they are 5-piece sacra, and markedly hypobasal if they are 6-piece sacra'. When SI is the 26th vertebra, the sacrum is usually composed of 5 sacral vertebrae with a formula of 25 PSV + 5S. Rarely, there may be 6 sacral vertebrae with a formula of 25 PSV + 6S. In spines with 25 PSV + 5S, over half (20/37) are indeed hyperbasal but no fewer than 11 are moderately hypobasal and 6 ere homobasal. Further, both of the spines with 25 PSV + 6S are markedly hypobasal. Thus, out of a total of 39 spines in which SI is the 26th vertebra, almost half (18) are not hyperbasal. My findings are not in agreement with Wells's (1963) statement in respect of spines with 26 PSV. We may say, therefore, that when SI is the 26th vertebra, the sacra tend to be hyperbasal but also show homobasality or moderate hypobasality and, rarely, marked hypobasality. It would seem then that basality is influenced not so much by the numerical position at which the ilium settles, as by the precise position of SI in relation to the sacral individuation field of the ilium and by the number of sacral vertebrae induced by that field. 3. Sacrococcygeal Synostosis Synostosis of a first coccygeal element, in which the transverse processes are rudimentary, is clearly distinguishable from a last sacral element which shows transitional features, as in a 5/6-piece sacrum. In 5/6-piece (or, rarely, 4/5-piece) sacra, the transverse processes of the last sacral element are well developed; on one side they complete, and on the other side,they almost complete the last sacral foramen. Synostosis of Co 1 to the sacrum is more common in female (24,8%) than in male S.A. Negro sacra (14,9%). This may be attributed to a

10 higher frequency of secondary synostosis produced by obstetrical trauma. Primary synostosis, it is suggested, is owing to variation in the pattern of ossification of the sacrococcygeal synchondrosis and it occurs approximately equally in male and female sacra. It is suggested, too, that primary synostosis accounts for the finding that synostosis is found in appreciable frequencies in every decade of adult life in both male and female S.A. Negro sacra (if we accept stated ages as reflecting true ages). Sacrococcygeal synostosis is present more frequently in the over-50 age group (males 15,8%, females 30,4") than in the under-50 age group (males 13,4%, females 20,8") in the S.A. Negro. It is noteworthy, however, that synostosis of Co 1 is present in an appreciable number of the younger individuals, for this is mentioned, in the literature available to me, only by Trotter and Lanier (1945). In their study on American Caucasoid and Negroid sacra, they found sacrococcygeal synostosis in all decades from the second to the ninth, though, as reported in this study, they found an increased incidence with age. All other references to the age of the individual and synostosis were found iri Anatomy textbooks, where synostosis is reported as occurring late in life (Quain's Elements of Anatomy 1878, Frazer's Anatomy of the Human Skeleton 1965, Morris's Human Anatomy 1956, Cunningham's Textbook of Anatomy 1972 and Gray's Anatomy 1973). A possible hereditary factor may be involved in sacrococcygeal synostosis for it occurs less often in American Negroes and South Africar Negroes than in Caucasoids. The higher frequency of synostosis in Caucasoids is present in each decade savti i.n the male samples of the second decade. There is, however, a greater mean age in the Caucasoid than in the Negroid samples. ( The higher frequency of synostosis in the Caucasoid sacra may thus be owing, in part, to a greater life expectancy in the American Caucasoid population. E. SUMMARY 1. In the South African Negro, the American Negro and the San, the sacrum may consist of 4, 5 or 6 conjoined vertebrae. Four-piece sacra are rare. Six-piece sacra occur relatively commonly, in 33,1% of male and 23,6% of female S.A. Negro sacra, 29,2% of ma e and 20,4% of female American Negro sacra and 12,9% of male and 18,2% of female San sacra. In the San sample of Duparc (1942), the frequency is 31,3% in male and 16,7% in female sacra. It is possible that the small San

11 300 male sample of the present study is unrepresentative, since it is the only sample of all the populations surveyed which shows a female preponderance of 6-piece sacra. 2. The frequency of 6-piece sacra does not vary significantly among the S.A. Negro sub-groups in either male or female category and so the subsets were combined to compare the frequency of 6-piece sacra with those in other population groups. 3. The highest frequencies of 6-piece sacra in the various population groups are found in males in the S.A. Negro (33,1%), the San (Duparc's sample, 31,3%) and the American Negro (29,2%). These are followed, in males, by the frequencies for the combined Caucasoid sample (27,0%), the Australian aborigine (19,0%^, Japanese (18,8%) and, finally, the Eskimo (6,3%). In female sacra, the frequencies of 6-piece sacra are, with the one exception already stated, lower than in the corresponding male samples. The highest frequencies are found in the two Negro groups and the lowest in the Eskimo and the Australian aborigine. The frequency of 6-piece sacra is significantly higher in the S.A. Negro than in the Eskimo and Australian aboriginal populations (male and female) and in Japanese (among whom the difference from the S.A. Negro is significant in males only). 4. The mean metrical characters of the sacrum of the South African Negro show no major intertribal differences. Mahalanobis's test shows that the 'distances' among the Negro tribal groups are small and not significant. 5. In general, the overall variability of the sacral metrical characters is of the same order in the male and female tribal groups. 6. On the basis of the analysis of the intertribal relationships and the analysis of tribal variability, it was considered justifiable to pool the Negro sub-groups to form a sample representative of the S.A. Negro. 7. Six-piece sacra are, in general, longer and relatively narrower than 5-piece sacra owing to the additional vertebra. Furthermore, they tend to have a greater absolute and relative width of the body of SI. The inclusion of 6-piece sacra in slightly differing frequencies in the ethnic groups studied has not significantly influenced any of the conclusions reached in the metrical analysis of the S.A. Negro and San fema1e sacra but may have obscured the relationships among the corresponding maie piece sacra. 8. The S.A. Negro sacra, are, in general, significantly larger

12 301 than the San sacra. They are longer, broader and have larger mean dimensions of the body of SI, as well as being, in general, more curved and relatively broader in both males and females. Mahalanobis's test shows a significant generalised distance between S.A. Negro and San sacra, as well as a significant distance between each Negro subgroup and the San sample. 9. The differences between S.A. Negro and San sacra may be attributed mainly to the smaller mean stature and generally smaller anthropometric characters of the San as compared with the Negro. The only difference which is not related to the Sari body type is relative sacral breadth: San male sacra are, in general, dolichohieric whilst S.A. Negro male sacra are more often platyhieric or subplatyhieric arid San female sacra are, on the average subplatyhieric whilst the majority of S.A. Negro female sacra are platyhieric. 10. S.A. Negro male and female sacra show a trend to be narrower (absolutely and relatively) than Caucasoid sacra but the evidence is not conclusive. 11. Whilst the metrical characters reveal significant differences mainly of size between Negro and San sacra, the exceptionally high frequency of increased precoccygeal vertebrae, especially in males, is a feature shared by the San and S.A. Negro. The San males have the highest frequency of increased precoccygeal vertebrae for all racial groups surveyed (43,8%), and the South African Negro the second highest (40,3:',). These figures are followed closely by the frequency for American Negro males (34,8%). 12. The high frequency of increased total vertebral number in San, S.A. Negroes and American Negroes is in keeping with the view that Khoisan and Southern African Negroes and the african ancestors of American Negroes sprang from a common proto-negriform stock. The San have the highest frequency of an increased precoccygeal vertebral number and this supports the postulate that the San (Bushman) genotype is closest to that of the last common ancestor from which the Negroes and San have differentiated by isolation, random genetic drift, selective pressures and hybridisation. 13. The closer metrical relationship of San and C. Nguni and of San and Sotho sacra in both 5-piece and piece female sacra and nossibly in male 5-piece sacra may be attributed to the aforementioned hybridisation. The inconsistency in the piece N. Nguni sacra (which shows males closest to San but females furthest from San) may be

13 302 owing, in part to a relatively low frequency of 6-piece sacra in the N. Nguni end San inale samples and partly to the different degrees of sexual dimorphism in the two ethnic jroups (there is a high degree in N. Nguni and an exceptionally low degree of sexual dimorphism in the San), further evidence of a closer relationship between San and C. Nguni (as compared with San-N. Nguni and San-Sotho) is prov ' by the high frequency of increased presacral vertebrae and of inert..sed total precoccygeal vertebrae in both San and C. Nguni males as compared with N. Nguni and Sotho males. ' 14. Variation in the number of presacral and sacral vertebrae is explicable in terms of variation in the level at which the ilium articulates with the vertebral column. It is thought that the ilium exerts a sacral individuating field effect on the adjacent developing vertebral column resulting in the differentiation of sacral vertebrae. Variation in the segmental level, cranio-caudal extent and strength of the field is considered to account for variation in the number of sacral and presacral vertebrae, the occurrence of transitional features and in the basality of the sacrum. 15. Transitional features in the sacrum may occur fit the L5/S1 junction in the form of alar facets on SI which are explained by an 'overflow' of the field of sacralisation on to the last lumbar vertebrae. Alternatively, transitional features may be present at the S1/S2 junction in the form of incomplete fusion or separation of the bodies, vertebral arches; or a second sacral promontory; or a combination of these features. These transitional features at the S1/S2 junction signify imperfect individuation of SI due, it is thought, to a 'weakness' at the cranial extremity of the sacralisation field. 16. Variation in the segmental level of the ilium is considered to explain the seven different patterns of presacral/sacral formulae in the S.A. Negro spinal columns. 17. (i) In the S.A. Negro, when the iliac individuation field extends over only 4 elements (as in 2,4% of all columns), individuation of SI is complete in all the columns. There is, however, an 'overflow' of the field of sacralisation on to the last lumbar vertebra of 3 of the 10 columns. (ii) When the iliac individuation field extends over five elements (as in 68,2% of all spines), the sacral individuation of SI is complete in approximately 75% of spines and individuation of S5 is complete in all but two spines. In the remainder, incomplete sacral individuation of SI occurs relatively frequently in spines with 23 PSV, less often

14 303 in spines with 24 PSV and rarely in spines with 25 PSV. In the latter, there is commonly an overflow of sacralisation (in the form of alar facets) on to the last lumbar element but only in the category of spines with six lumbar vertebrae. (iii) When the iliac individuation field extends over six sacral elements (29,4% of spines), the majority of these spines (70%) show incomplete individuation of SI, i.e. transitional features at the S1/S2 junction irrespective of whether there are 23 PSV or 24 PSV. In contrast, there are no transitional features at the L5/S1 junction in these spines. (iv) Transitional features at the S1/S2 junction are thus the rule in columns with 23 PSV (+ 5S or 6S), are commonly present in columns with 24 PSV + 6S and occur less often in columns with 24 PSV + 5S. In the columns with 23 PSV, incomplete individuation of SI is considered to occur as a consequence of incomplete caudal migration of the iliac individuation field whereas, in the the columns with 24 PSV it is thought that the sacralisation field is smaller than usual relative to the number of sacral vertebrae which are induced. However, in all categories of spines with transitional features at the S1/S2 junction, both factors (variation in the precise position of the cranial limit of the field as well as a smaller field) may operate, to a greater or lesser degree to produce incomplete individuation of 51. (v) Transitional features at the L5/S1 junction are present in spines with 24 PSV + 4ji and 25 PSV + 5S. In the former, i.e. the category of spines with alar facets on a 4-piece sacrum, the 'overflow* of sacralisation may be from an iliac individuation field which is slightly larger than usual relative to the number of sacral vertebrae which are induced (four). In the category of spines with alar facets on a 5-piec.e sacrum, the regional spinal formula almost invariably includes six lumbar vertebrae and it is thought that the ext > lumbar element is sensitive to the adjacent sacralisation field which itself may be closer than usucl lo the last lumbar vertebra. (vi) There is a strong tendency, which is present in all population groups studied, for the iliac individuation field to induce a modal number of five sacral elements. Further evidence of this tendency is reflected by the findings that the field commonly overflows on to the adjacent last lumbar vertebra when only four sacral pieces are individuated, and that incomplete individuation of the sixth sacral element (either SI or S6) is comnonly present when six vertebral elements are induced.

15 In spines with six sacral vertebrae, the additional sacral element is seemingly derived from the 24th element (the potential last lumbar vertebra) when there are 23 PSV + 6S but when there are 24 PSV + 6S, the extra element is derived either from the incorporation of an extra element in the lumbo-sacral region or from Co 1. In some spines, it is not possible to decide among these possible sources of the extra vertebra. 19. (i) Sitsen's suggested correlation of a low frequency of hyperbasal ity and squatting is not supported by the results of the present study, neither is the view of Gillman (1 929) that basal ity is influenced by sex. The concept of Wells (1963) that hyperbasal, homobasal and hypobasal sacra are produced by variation in the level of SI, is modified to include a fourth category of basality, marked hypobasality. Relationships are demonstrated between the four categories of basality, the number of sacral vertebrae, the vertebral numerical position of the first sacral vertebra and Uional features in SI. This is correlated with the concept of the extent and strength of the iliac individuation field. (ii) The majority of S.A. Negro sacra ; 3,9%) are moderately hypobasal or homobasal 5-piece (n = 222) or 6-p ce (n = 54) sacra which usually show no transitional features at the S1/S2 junction. A further 20,7% are markedly hypobasal 6-piece (n = 65) or 5-piece (n = 19) sacra, the majority of which have transitional features at the S1/S2 junction. This is presumed to be owing to a slight caudal shift of the iliac ir.uividuation field relative to the upper border of SI. This shift results in incomplete individuation of the first sacral element but individuation of a sixth sacral element and marked hypobasality. Hyperbasal sacra (10,4-) are either 4-piece or 5-piece sacra and very few have transitional features at the S1/S2 junction. They tend rather to show transitional features at L5/S1 junction, apparently owing to a slight crania 1 shift of the iliac individuation field relative to the upper border of SI. This appears to result in an overflow of sacralisation on to the last lumbar vertebra and the consequent development of alar facets on the superior surface of SI. (iii) When SI is the 24th vertebra, the sacra are almost invariably hypobasal (if 5-piece) or markedly hypobasal (if 6-piece). When SI is the 26th vertebra, just over half of the sacra are hyperbasal but a good proportion are homobasal or moderately hypobasal (if 5-pic-ce) and, when the sacrum is 6-piece (rare), -.hey are marked.y hypobasal.

16 Synostosis of Co 1 to the sacrum in the S.A. Negro is more common in 5-piece than in 6-piece sacra, and in the over-fifties than in the under-fifties, though it is also present in an appreciable percentage of younger individuals. In addition, sacrococcygeal synostosis is more common in Caucasoids than in American and South African Negroes. This Caucasoid excess is present in each decade,save the second However, the higher frequency of synostosis in Caucasoids may be owing, in part, to a greater number of older individuals in the samples. 21. In S.A. Negro female sacra, the measures of length, the measures of the body of SI and the values for the corporo-basal index are, in general, smaller than those of male sacra. Female sacra also tend to be absolutely and relatively broader and to have a more cranially placedposition of maximum curvature than male sacra. Curvature of the sacrum tends to be less in females but this sex difference is not significant. The promontory angle is, on the average, smaller in female than in male sacra. These findings are echoed in San sacra, though only the sex differences for the measure of the width of the body of SI and for the corporo-basal index are significant. 22. The metrical character which best demonstrates sex differences in S.A. Negro and San sacra and, indeed, in all men as well as in all categories of sacra (5-piece, 6-piece and piece), is the corporobasal index. This index carries some of the highest t_ values for differences between male and female means in the series and shows significant differences in the less sexually dimorphic San sacra as well as in all S.A. Negro sub-groups. 23. In piece sacral samples, the S.A. Negro sacra show the highest degree of sexual dimorphism of sacral metrical features of all the ethnic groups available, and the San the least. The N. Nguni subgroup is more sexually dimorphic than the C. Nguni or Sotho sub-group of the S.A. Negro series. Though the San sample size is small, the San sacra show considerably less sexual dimorphism than any one of the S.A. Negro sub-groups. 24. Sexual differentiation of metrical features in S.A. Negro sacra is least in 6-piece sacra, more pronounced in 5-piece sacra and most pronounced in the total sample of piece sacra. The differing relative frequencies of 6-piece sacra in male and female samples have not materially altered the degree of sexual dimorphism in the combined S.A. Negro sample nor in the San sample. In the S.A. Negro sub-group

17 Author De Beer Kaufman P M Name of thesis A study of the sacrum and some aspects of presacral vertebrae in San (Bushmen) and Southern Africa and American Negroes 1975 PUBLISHER: University of the Witwatersrand, Johannesburg 2013 LEGAL NOTICES: Copyright Notice: All materials on the University of the Witwatersrand, Johannesburg Library website are protected by South African copyright law and may not be distributed, transmitted, displayed, or otherwise published in any format, without the prior written permission of the copyright owner. Disclaimer and Terms of Use: Provided that you maintain all copyright and other notices contained therein, you may download material (one machine readable copy and one print copy per page) for your personal and/or educational non-commercial use only. The University of the Witwatersrand, Johannesburg, is not responsible for any errors or omissions and excludes any and all liability for any errors in or omissions from the information on the Library website.