Patterns of skeletal histologic change through timeComparison of an archaic native american population with modern populations.код для вставкиСкачать
THE ANATOMICAL RECORD 226:307-313 (1990) Patterns of Skeletal Histologic Change Through Time: Comparison of an Archaic Native American Population With Modern Populations DAVID E. BURR, CHRISTOPHER B. RUFF, AND DAVID D. THOMPSON Department of Anatomy and Orthopedic Research Laboratory, West Virginia University Health Sciences Center, Morgantown, West Virginia (DJ3.B.); Department of Cell Biology and Anatomy, The Johns Hopkins University School of Medicine, Baltimore, Maryland (C.B.R.); Bone Biology and Osteoporosis, Merck, Sharp and Dohme Research Laboratories, West Point, Pennsylvania (D.D.T.) ABSTRACT This paper compares patterns of histologic change in a n archaic Native American population with those in modern white populations. Histologic sections were removed from core biopsies taken from the anterior femoral cortex of a n archeologic sample of Pecos Indians. The data demonstrate many microstructural similarities between the Pecos and modern populations, even though they were genetically and culturally distinct. Pecos women had small Haversian canals and large osteon mean wall thickness, with no clear evidence of a n intracortical bone volume deficit even in the older age groups, although significant marrow cavity expansion occurred in both males and females with age. No striking relationships were found between bone tissue changes and gross geometric changes with age. The data suggest that a more active life-style is associated with greater osteon mean wall thickness or osteon population density, but that it alone does not protect against significant bone loss on the cortical-endosteal surface. Patterns of cortical bone loss in aging men and women have been well-documented in modern populations (Garn, 1970; Martin and Atkinson, 1977; Martin et al., 1980; Riggs et al., 1982,1983; Riggs and Melton, 1986). There is substantial evidence that such changes occurred in pre- and protohistoric communities a s well (Van Gerven e t al., 1969; Van Gerven, 1973; Carlson et al., 1976; Eriksen, 1976), although the rate, pattern, and severity of bone loss apparently varies significantly by population (Moldawer et al., 1965; Nordin, 1966; Armelagos, 1969; Dewey et al., 1969a,b; Solomon, 1979; Perzigian, 1973; Mazess and Mather, 1975; Eriksen, 1976; Ruff and Hayes, 1982, 1988; Harper et al., 1984; Evers et al., 1985; Martin et al., 1985). Populational differences in the maintenance or loss of bone mass partly reflect the interaction of environmental and cultural factors with the Basic Multicellular Unit-based remodeling system (Armelagos, 1969; Van Gerven, 1973; Larsen, 1981; Richman et al., 1979; Ruff, 1987). The effects of these nongenetic factors can be observed in differences in the size, geometry, and number of osteons. As such, interpopulation comparison of histological structure can shed light on differences in rates and patterns of loss that may result from culturally based causes and allows us to define the limits of human population variability. Bone loss in modern populations is compensated by geometric changes that maintain bone ridigity in males but do not fully compensate for bone loss in females (Smith and Walker, 1964, 1980; Epker and Frost, 1965; Martin and Atkinson, 1977; Ruff and Hayes, 1988). Structural compensatory changes have 0 1990 WILEY-LISS, INC. been documented in both males and females in archaic populations (Ruff and Hayes, 1982). Burr and Martin (1983) suggested that material property changes may supplement these structural adaptations. Material property compensations would involve alterations in histologic structure, especially differences in the size, distribution, or rate of accumulation of osteons and their canals and in the additional porosity resulting from this history of remodeling activity. Because these changes have their basis in the remodeling system, i t is a significant question whether the geometrical compensations that have been observed are related to the histological changes that affect tissue material properties. The purpose of this study is to compare changes in a n archaic Native American population with those in modern white populations. This study 1)compares agerelated patterns of histologic change in a n archaic human population with modern U S . and British populations; 2) assesses sexual dimorphisms in histologic structure; and 3) examines whether micro- and macrostructural adaptations to aging are related. MATERIALS AND METHODS Study Population The Pecos Indians were a n agricultural community that lived in north-central New Mexico between the Received April 19, 1989; accepted J u n e 20, 1989. Address reprint requests to David B. Burr, Department of Anatomy, Indiana University Medical Center, Indianapolis, IN 46223. 308 D.B. BURR ET AL. 14th and 19th centuries. As a study population, this group has the advantage over modern populations of being a closed and stable society in which everyone had equal access to environmental and cultural resources. This increases the uniformity of skeletal growth and remodeling and limits the large interindividual variability found in modern groups. No evidence of serious malnutrition or food shortages has ever been detected in the Pecos Indians. The study population is described more fully by Ruff and Hayes (1983a). Comparative data for modern white populations was derived from the literature. These sample populations were composed of individuals from the United States and Great Britain. One sample population (Georgia e t al., 1982) was of Eastern European extraction. (The number, sex, and age range for individuals in each population is listed in Table 1.) 7. porosity, (Nh X &)/A, 8. percent osteonal refilling = (A, - Ah)/A, An active surface area electronic digitizer (RP 622B, Talos Systems, Inc.) was used to collect boundary coordinates from complete cross sections cut from the femoral midshaft adjacent to the core sample. Cross-sectional geometric properties were calculated from these coordinates using the SLICE Fortran program (Nagurka and Hayes, 1980). This program divides a total area into a series of trapezoids or rectangles, then adds or subtracts these areas to determine the areas and area moments of the entire cross section. The method of sectioning, orienting, and measuring these sections is described by Ruff and Hayes (1983a). Geometric measurements included 9. cortical area, mm2, the area of bone within the periosteum Femora from 27 females (22-60 years) and 28 males 10. total area, mm’, the entire area (bone and mar(21-60 years) were selected for analysis. Sex and age at row cavity) within the periosteum 11. principal cross-sectional moments of inertia (or death were determined from pelvic morphology, dental wear, and endocranial suture closure (Ruff, 1981). second moments of the area), I,, and Imin,mm4. Any There was no significant difference in mean age be- individual cross-sectional moment is defined by the tween males (x = 40.1 s 2.4) and females (% = 39.6 & general equation 2.7). A diaphyseal bone corer mounted in a high-speed I, = n2 d ~ = n12AA,, Dremel drill was used to remove cores 4.0 mm in diameter from the entire thickness of the anterior cortex of the femur exactly at midshaft. Bone density (g/cm3) where n is the perpendicular distance from a defined and I,,, are the largest was calculated from the wet core weight and the core axis to a unit of area, A. I,, moments that can be calculated about any axis through volume, the latter derived from core diameter and the section and are proportional to the maximum and length (Thompson, 1979). Using a Buehler Isomet saw, sections approximately minimum bending ridigity of a bone a t the cross-sec90 pm thick were cut from the core in a plane parallel tional level analyzed 12. polar moment of inertia, in mm4, defined by the to the core’s length (i.e., transverse to the long axis of the femur). The sections were hand ground to 80 pm equation thick, mounted on glass slides, and left unstained. Sample Preparation I, 1 Measurements Osteonal dimensions were measured on the entire sectioned core at 125 x using a HiPad digitizing tablet and Bioquant image analysis software (R & M Biometrics, Nashville, TN). Only secondary osteons with complete canals were counted. The following variables were measured: where r is the distance from the centroid of a cross section to a unit of area, A. This is proportional to the torsional rigidity of a bone at the cross-sectional level analyzed and is equal to the sum of any two crosssectional moments of inertia. Because geometric variables are related to body weight, size adjustments were made by dividing area 1.the number (No),mean individual cross-sectional measurements by femoral length2 and cross-sectional area (A,, mm2), and mean individual perimeter (PO, moments by length4, following a previous study (Ruff, mm) of secondary osteons 1984). Non-normalized “raw” data have been published 2. the number (Nh), mean individual cross-sectional elsewhere (Ruff and Hayes, 1983a,b). area (Ah, mm’), and mean individual perimeter (Ph, RESULTS mm) of Haversian canals of secondary osteons Osteon Dimensions in Pecos and Modern Populations 3. sample area (As, mm2) Table 1 compares mean individual osteon dimenOsteon fragments were not counted, so the absolute sions of the femoral midshaft in the Pecos population rate of past bone turnover will be underestimated for with those collected from modern populations. Mean the Pecos population. Several variables were derived individual osteon areas and Haversian canal areas are from the measurements: on the low end of the modern range, but do not differ much from values found in modern populations even 4. osteonal mean wall thickness (MWT) = 2 x though the average age of the Pecos population is (A, - &)/(PO + Ph), mm younger than that of the modern populations (Table 1). 5. osteon population density (OPD) = N,/A,, When compared with individuals of similar age (i.e., no./mm2 <60 years; Table 1)the magnitude of A, and Ah in the 6. percent of cortex composed of secondary osteons, Pecos population shows even greater similarity to modpercent osteonal bone, (A, x NJA, ern white populations. Osteonal mean wall thicknesses 309 SKELETAL HISTOLOGIC CHANGE THROUGH TIME TABLE 1. Comparison of osteonal dimensions in femoral midshaft: Pecos vs. modern populations' Reference Frost (196 1) Currey (1964) A0 (mm2) Ah (mm2) .05l4 .0344 Ml 1 ,035 ,036 .0474 .045 .0290 .00504 .00454 ,0050 .0032 .00274 ,0021 ,043 ,0051 ,041 ,028 .034' ,041' M/42 - .0304 Jowsey (1966) Martin et al. (1980) Evans (1976, 1977) Pecos population, this study Singh and Gunberg (1970) Thompson (1980) Georgia et al. (1982) - .0023' ,0024' - .00154 - .0060 ,0070 .0041 .0063 .00374 - SexINo. Ml8 F111 M16 Fi5 ?I26 ?I? Ml35 Ml35 Ml28 Fl23 Ml33, F17 Ml54 Fl36 range (years) 53 36-81 23-89 <60 <60 20-90 <60 37-96 36-75 36-75 21-60 22-60 40-88 age (years) 53.0 52.8 60.9 43.5 40.0 ? ? Femur, midshaft 69.3 41.5 Femur, midshaft Femur, region? 71.0 40.1 39.6 62.3 30-97 72 Fl7 50-59 50-59 M/? ? ? ? ? M/4 Bone, region3 Femur, region? Femur, midshaft Femur, midshaft ant. cort. Femur, midshaft ant. cort. Femur, midshaft ant. cort. Femur, mid 113 'Where possible, age subgroups have been extracted from published data to provide a better comparison with data from the Pecos sample opulation. For clarity, standard errors for data taken from the literature are not included in this table. !Standard errors: male A,, .001846; male Ah, ,000171; female A, .001893; female A, .000171. 3Except where specified, the entire cortex at the level examined was included in the measurements. 4Calculated from diameter. osteal apposition (P < .01; Table 41, both of which would tend to reduce OPD. In contrast to males, osteon size did not change with age in females. These observations suggest a n age-associated increase in remodeling activity (i.e., greater osteonal activation frequency) Sex Differences in Osteon Dimensions coupled with greater individual cell-level osteoclastic Females in the Pecos population had larger osteons activity in females to replace large osteons lost by cor(P < .02) and greater osteonal MWT (P < .02) than tical-endosteal resorption with as many equally large males, though Haversian canals were not larger (Table osteons. 2). Because of the smaller osteons, OPD is greater in Osteon population density was inversely related (P < males than in females (P < .06).This implies a history .05) in males to the external shape of the bone, as deof greater remodeling activity in the male femoral cor- termined by the ratio ImaX/I,in. In other words, the tex. No sex differences were found in percent osteonal rounder the cross section, the greater the OPD. In ferefilling. males, a rounder cross section was associated with smaller Haversian canals. Percent osteonal bone was Aging Changes in Osteon Dimensions negatively correlated with the polar moment of inertia Mean Haversian canal size increased by 36% in fe- and with total cross-sectional area in Pecos females (P males over the 40 year age range, resulting in a 48% < .08 and P < .07, respectively). Porosity was posiincrease in porosity (Table 3). Because of large individ- tively associated with I,, (P < .02). ual variation, statistical significance could not be demonstrated. There was no significant decline in either Relationships Among Microstructural Measurements density or percent osteon refilling with age (Table 3). In males osteons became smaller by 26%with age (P Imbalances between intracortical bone resorption < .025) but this did not result in greater osteon density and formation were not found at the tissue level in this in older males. The percent of cortex remodeled to sec- sample, although a n imbalance did exist on the cortiondary osteonal bone declined by nearly 25% with age cal-endosteal surface (Table 4). Smaller osteons con(Table 3). Resorption along the marrow cavity both de- tained smaller Haversian canals in both males and fecreases OPD and skews the osteon population toward males (r2 = .26 and .19, respectively, P < .05). As smaller sizes a s larger osteons are generally found expected, larger Haversian canals were significantly (P closer to the cortical-endosteal surface. There was sig- < .01) correlated with greater porosity, as was the pernificant cortical-endosteal resorption in males with age centage of cortex that was remodeled (males, r2= .23, P (P < .01; Table 4), perhaps partially accounting for the < .02; females, r 2 = .48, P < .01). Both canal size (a measure of relative bone formation) and canal number observed changes in osteon size and number. In females, OPD increased even in the face of signif- (a measure of activation frequency) contributed to icant cortical-endosteal resorption (P < .01) and peri- changes in bone porosity. in the femurs of individuals younger than 60 years from modern populations averages between 72 pm (Currey, 1964) and 84 pm (Jowsey, 1966). MWT in Pecos males and females falls within this range (Table 2). 310 D.B. BURR ET AL TABLE 2. Sexual dimorphism in femoral osteonal dimensions Males x 34,345* 2,267 668.10* 161.03 76.10* 93.26 7.26** 24.40 1.61 1.98 Mean osteon area, A, (pm2) Mean canal area, Ah @m2) Mean osteon perimeter, PO,(km) Mean canal perimeter, P h , (km) Mean wall thickness, MWT (km) Percent osteon refilling Osteon pop. density, OPD (/mm2) Percent osteonal bone Porosity Densitv (n/cm3) Females S.E. t 40,778* 2,404 733.26* 167.84 84.15" 93.99 6.40** 25.79 1.54 1.98 1.846 171 17.32 4.75 12.42 2.08 0.29 1.30 0.12 0.10 S.E. 1.858 171 17.29 5.39 11.56 1.99 0.35 1.59 0.13 0.15 *P < .02. **P < .06 TABLE 3. Age cohort means: Microstructural data Sex Female Age (years) No. 20-29 7 30-39 10 40-49 3 250 7 20-29 6 30-39 10 40 -49 6 250 6 Change' Canal area (w2) Osteon area (km2) Osteon density (no./mm2) Percent osteonal bone Porosity (%I Density (gicrn) 2,004 (227) 2,404 (840) 2,852 (1,138) 2,613 (959) + 36% 36,354 (7,582) 45,817 (10,635) 36,096 (2,669) 40,010 (7,634) + 2% 6.33 (1.35) 5.53 (0.82) 7.06 (2.01) 6.93 (2.18) + 11% 22.88 (6.75) 25.97 (8.02) 25.01 (5.33) 27.50 (9.04) + 9% 1.26 (0.24) 1.45 (0.70) 1.84 (0.53) 1.82 (0.82) + 48% 2.02 (0.14) 2.02 (0.12) 1.86 (0.16) 1.95 (0.17) -3% 2,079 (463) 2,374 (925) 2,278 (529) 2,265 (1,258) + 6% 40,736 (9,346) 34,935 (10,957) 30,432 (4,687) 30,882 (6,000) - 26% 6.80 (1.20) 7.43 (1.50) 7.45 (1.66) 7.26 (1.49) + 2% 27.33 (6.75) 25.24 (7.28) 22.30 (4.53) 22.19 (5.69) -23% 1.38 (0.29) 1.73 (0.67) 1.69 (0.50) 1.57 (0.71) + 3% 2.05 (0.06) 1.98 (0.09) 1.93 (0.08) 1.98 (0.14) - 7% Male Change' 'Percent change calculated as (b X age range)/(20-29 years cohort), where b is the regression slope, and the age range is 40 years. DISCUSSION Sexual Dimorphism in Bone Microstructure In modern populations females have larger Haversian canals in the femoral midshaft, but males have more canals (Thompson, 1980). The same was found in the Pecos population, although the differences between males and females were not statistically significant. The observation that canal size is larger throughout life in women than men argues against the possibility raised by Thompson (1980) that sex differences in canal size detected in modern populations reflect sampling distributions, which are generally weighted toward the older age ranges. No sex-related differences in osteon size have been reported for modern populations, though this was the strongest microstructural sexual dimorphism in Pecos individuals. Although osteon sizes in Pecos Indians and modern white populations are comparable, Haversian canal area in Pecos men and women averaged about half that found in modern groups (Table 1). Part of the reason for this may have been the relatively younger age a t death of individuals in the Pecos sample population. Although the smaller canals imply a greater volume of bone formed per remodeling unit in Pecos Indians than in modern whites, a feature consistent with the more active life-style of the Pecos society (Ruff and Hayes, 1983a,b), MWT in the Pecos population fell within the modern range. The number of osteons in a section is a function of both the remodeling rate and the mean tissue age (MTA) of a region of bone (Frost, 1987a,b). Assuming equivalence of MTA in Pecos males and females, the greater OPD in males may imply more frequent activation of remodeling and more rapid bone turnover and may also reflect the more physically active tasks required of males in this population. It is also possible that sex-specific differences in types of behaviors, e.g., more extensive long distance running in males (Ruff and Hayes, 1983b1, would differentially load the ante- 31 1 SKELETAL HISTOLOGIC CHANGE THROUGH TIME TABLE 4. Age cohort means: Geometric data Sex Female Age (yea r s1 No. 20-29 7 30-39 10 40-49 3 250 7 20-29 6 30-39 10 40-49 6 250 6 Change3 Cortical area' Medullary area' Total area' Inlax2 2.17 (0.29) 2.24 (0.21) 1.78 (0.20) 2.11 (0.25) -6% 0.69 (0.12) 0.78 (0.14) 1.29 (0.42) 1.13 (0.25) + 80% 2.85 (0.36) 3.03 (0.24) 3.07 (0.22) 3.24 (0.16) + 15% 7.15 (1.82) 7.73 (1.31) 7.11 (1.04) 8.17 (1.27) + 15% 5.51 (1.36) 6.21 (1.00) 5.57 (0.19) 6.70 (0.78) + 22% 1.29 (0.11) 1.25 (0.16) 1.28 (0.22) 1.22 (0.18) - 5% 2.15 (0.12) 2.33 (0.28) 2.07 (0.16) 2.09 (0.29) - 8% 0.75 (0.09) 0.81 (0.17) 0.91 (0.18) 1.02 (0.28) + 43% 2.90 (0.12) 3.14 (0.25) 2.98 (0.08) 3.11 (0.09) + 5% 7.42 (1.05) 8.84 (1.36) 7.74 (0.89) 8.59 (1.32) + 9% 5.33 (0.82) 6.27 (1.18) 5.30 (0.48) 5.55 (0.55) - 2% 1.42 (0.32) 1.44 (0.22) 1.47 (0.26) 1.56 (0.30) + 9% Irnin' Imdmin Male Change3 'Normalized for body weight by dividing by femoral length2. All values x 103. 'Normalized for body weight by dividing by femoral length4. All values x 107. 3Percent change calculated as (b x age rangeM20-29 years cohort), where b is the regression slope, and the age range is 40 years. rior cortex of the femoral midshaft, supporting Ruff's hypothesis (1988) that greater AP bending of the femur occurred in males. Aging Changes in Bone Microstructure Porosity is a function of the number of Haversian canals and their size (Atkinson, 1964). Various investigators have shown t h a t increased intracortical femoral porosity in males and females is the result of increased numbers of canals (Currey, 1964; Singh and Gunberg, 1970; Kerley, 1965; Evans, 1977; Martin et al., 19801, increased mean canal area (Aoji, 1959; Jowsey, 1966) or both (Thompson, 1978,1980). A large but statistically insignificant age-related intracortical porosity increase was observed in Pecos females, but not in males. No significant change in density was found in either sex. Haversian canal dimensions increased with age in Pecos females, but MWT did not change. Thus the volume of bone formed per bone structural unit (BSU) (Jaworski, 1976; Parfitt, 19831, or osteon, did not change with age in women. Significant intracortical bone loss in modern populations usually does not occur prior to the seventh decade in females and the ninth decade in males (Martin et al., 1980; Riggs et al., 1983). It is not surprising, therefore, that there is no clear evidence of intracortical bone loss in the Pecos population in the age ranges studied (2060 years). On the other hand, significant age-associated bone loss along the cortical-endosteal surface of the femur occurred in both men and women (Table 4). Significant loss on this bone surface is generally associated with postmenopausal osteoporotic change, whereas intracortical loss is mainly a function of aging (i.e., senile or type I1 osteoporosis) (Riggs et al., 1983). It is unusual in modern populations for significant marrow cavity expansion at the femoral midshaft to occur in active men in middle age, although expansion a t the metaphyses of the long bones occurs slowly. This observation suggests that elements of life-style or environment do not prevent some cortical-endosteal loss of bone. The size of osteons decreases with age in the femora of Pecos Indian males, resulting in decreased MWT with age. This is consistent with observations made by others for modern male femora (Evans, 1977; Martin e t al., 1980). Currey (1964) observed reduced osteon size with age in a population including both men and women, and the observation has been made for other bones as well (Landeros and Frost, 1964; Jowsey, 1966; Hattner et al., 1964). This may be due to the greater probability that large osteons will be remodeled, eventually resulting in a population of smaller osteons (Takahashi et al., 1965). Curiously, however, no change in osteon size with age was found in Pecos females. This observation argues against the hypothesis that the change in osteon size with age is dependent solely on the probability of remodeling larger osteons, because average osteon size in women was larger than that in men, except for the youngest age cohort. Martin et al. (1980) suggest that a decreased osteon area without a n accompanying change in Haversian canal area implies that both osteoclasts and osteoblasts are less active or fewer in older individuals. Compensatory Relationships Among Histologic and Geometric Changes Many significant or near-significant correlations between microstructural morphology and gross geometry suggest the possibility that compensatory relationships exist between the two. In some cases, the relationship may be related to the aging process. The positive association of porosity and I,, in females suggests that loss of bone mass with age is compensated by increased 312 D.B. BURR ET AL. bending rigidity as a function of periosteal expansion (Table 4).The inverse association of percent osteonal bone with J and total area reflect the same process. Total area and J increase by addition of primary lamellar bone to the periosteal surface, which occurs simultaneously with the loss of bone cortical-endosteally. Because the oldest and most highly remodeled bone is along the cortical-endosteal surface, the net effect is to reduce the amount of cortex that is composed of secondary osteons. All of these changes reflect the overall adaptation of the skeleton to peri- and postmenopausal bone loss in women in this population. The relationship among these variables is complex, but confining the examination of bone only to gross geometric or only to microstructural factors may overlook the complete spectrum of bone adaptation to its environment. Material Property Compensations With Age The effects of age on bone material properties are generally considered the result of increased porosity, changes in mineralization, or more subtle ultrastructural characteristics such as collagen cross-linking. Burr and Martin (1983) calculated the shear modulus for the distal radius and proposed that as women lose bone with age, the shear modulus increases in partial compensation. The increased shear modulus was difficult to explain solely on the basis of changes in porosity or mineralization of the bone. There is evidence that changes in osteon numbers and dimensions also affect material properties. Reduced osteon diameters and increased osteon density (if sufficient to offset the effects of the increased porosity that results from more osteons) should improve the fatigue properties of bone (Martin and Burr, 1989). A large number of small diameter osteons increases energy absorption prior to fracture in a bone (Moyle et al., 1978; Moyle and Bowden, 1984). As osteon diameter increases, the work-to-fracture declines, but increases again when osteons reach diameters greater than 200 pm. Fatigue processes in bone are inhibited by large osteons and by increased osteon population density (Corondan and Haworth, 1986). According to these data, osteons are large enough in the Pecos population to impart significant fatigue-inhibiting properties to the bone. The slightly larger osteons in women might provide enhanced energy absorption, but men could make up for slightly smaller osteons by having a greater osteon population density. The significant decrease with age in osteon size coupled with increased OPD might further enhance the bone’s energy absorbing qualities. Both men and women in this population have bone tissue that is microstructurally well-suited to intense physical activity. CONCLUSIONS Although the Pecos Indians were genetically, culturally, and nutritionally distinct from white populations residing in the United States today, they demonstrate many skeletal microstructural similarities with modern populations in the age range 20-60 years. These include similarities in osteon size and mean wall thickness, larger Haversian canals in women but increased osteon population density in men, and age-related reduction of osteon size in men without increased Haversian canal size. Differences between the populations are relatively small and contain substantial statistical uncertainty. The most remarkable difference between populations was the small Haversian canal size in Pecos women, perhaps indicating a greater volume of bone per remodeling unit. The greater OPD in males and the small canal size in females both may have been the result of the active life-style of members of this society. A more active life-style did not protect this population from significant age-associated cortical-endosteal bone loss. However, bone loss in females was compensated geometrically so that the structural ridigity increased even in the presence of a loss of bone mass. Moreover, changes in osteon dimensions with age may have enhanced the fatigue properties of the femur, further compensating for reduced mass. Although cause and effect relationships are impossible to determine in a static study such a s this one, the results suggest that some variations a t the microstructural and macrostructural levels are related. ACKNOWLEDGMENTS The authors thank Nina Clovis, Barbara Roberts, and Dorian Williams for their help in data collection. LITERATURE CITED Aoji, 0. 1959 Metrical studies on the lamellar structure of human long bones. J . Kyoto Prefect., 65t941-965. Armelagos, G.J. 1969 Disease in ancient Nubia. Science, 163:255259. Atkinson, P.J. 1964 Quantitative analysis of cortical bone. Nature, 201 t373-375. Burr, D.B., and R.B. Martin 1983 The effects of composition, structure and age on the torsional properties of the human radius. J . Biomech., 16:603-608. Carlson, D.S., G.J. Armelagos, and D.P. Van Gerven 1976 Patterns of age-related cortical bone loss (osteoporosis) within the femoral diaphysis. Hum. Biol., 48t295-314. Corondan, G., and W.L. Haworth 1986 A fractographic study of human long bone. J. Biomech., 19t207-218. Currey, J.D. 1964 Some effects of ageing in human Haversian systems. J. Anat. [Lond.], 98:69-75. Dewey, J.R., G.J. Armelagos, and M.H. Bartley 1969a Femoral cortical involution in three Nubian archaeological Hum. _populations. _ Biol., 41t13-28. Dewev, J.R., M.H. Bartlev. Jr., and G.J. Armelaeos 1969b Rates of femoral cortical bone-loss in two Nubian populations. Clin. Orthop. Rel. Res., 65:61-66. Epker, B.N., and H.M. Frost 1965 A histological study of remodeling at the periosteal, Haversian canal, cortical endosteal, and trabecular endosteal surfaces in human rib. Anat. Rec., 152:129-135. Ericksen, M.F. 1976 Cortical bone loss with age in three Native American populations. Am. J. Phys. Anthropol., 45t443-452. Evans, F.G. 1976 Mechanical properties and histology of cortical bone from younger and older men. Anat. Rec., 185:l-12. Evans, F.G. 1977 Age changes in mechanical properties and histology of human compact bone. Yrbk. Phys. Anthropol., 20t57-72. Evers, S.E., J.W. Orchard, and R.G. Haddad 1985 Bone density in postmenopausal North American Indian and Caucasian females. Hum. Biol., 57t719-726. Frost, H.M. 1961 Human Haversian system measurements. Henry Ford Hosp. Med. Bull., 9t145-147. Frost, H.M. 1987a Secondary osteon population densities: An algorithm for estimating the missing osteons. Yrbk. Phys. Anthropol., 3Or239-254. Frost, H.M. 1987b Secondary osteon populations: An algorithm for determining mean bone tissue age. Yrbk. Phys. Anth. 30r221238. Garn, S.M. 1970 The Earlier Gain and Later Loss of Cortical Bone in Nutritional Perspective. Charles C. Thomas, Springfield, IL. Georgia, R., I. Albu, M. Sicoe, and M. Georoceanu 1982 Comparative aspects of the density and diameter of Haversian canals in the diaphyseal compact bone of man and dog. Rev. Roum. Morphol. Embryol. Physiol., 28r11-14. Harper, A.B., W.S. Laughlin, and R.B. Mazess 1984 Bone mineral content in St. Lawrence Island Eskimos. Hum. Biol.. 56t63-77. SKELETAL HISTOLOGIC CHANGE THROUGH TIME Hattner, R., 0. Landeros, and H.M. Frost 1964 Are there cell behavioral modes that are unaffected by aging? Gerontology, 4 (Part II):34 (Abstract). Jaworski, Z.F.G. 1976 Parameters and indices of bone resorption. In: Bone Histomoruhometry. P.J. Meunier, ed. Armour Montague, Paris. Jowsey, J . 1966 Studies of Haversian systems in man and some animals. J. Anat., 100:857-864. Kerley, E.R. 1965 The microscopic determination of age in human bone. Am. J . Phys. Anthropol., 23t149-164. Landeros, O., and H.M. Frost 1964 The cross section size of the osteon. Henry Ford Hosp. Med. Bull., 12:517-525. Larsen, C.S. 1981 Functional implications of postcranial size reduction on the prehistoric Georgia coast, U.S.A. J . Hum. Evol., 10: 489-502. Martin, R.B., and P.J. Atkinson 1977 Age and sex-related changes in the structure and strength of the human femoral shaft. J . Biomech., 10:223-231. Martin, R.B., and D.B. Burr 1989 Structure, Function and Adaptation of Compact Bone. Raven Press, New York. Martin, R.B., D.B. Burr, and M.B. Schaffler 1985 Effects of age and sex on the amount and distribution of mineral in Eskimo tibiae. Am. J . Phys. Anthropol., 67:371-380. Martin, R.B., J.C. Pickett, and S.Zinaich 1980 Studies of skeletal remodeling in aging men. Clin. Orthop. Rel. Res., 149t268-282. Mazess, R.B., and W.E. Mather 1975 Bone mineral content in Canadian Eskimos. Hum. Biol., 47:45-63. Moldawer, M., S.J. Zimmerman, and L.C. Collins 1965 Incidence of osteoporosis in elderly whites and elderly Negroes. J.A.M.A., 194; 117-120. Moyle, D.D., and R.W. Bowden 1984 Fracture of human femoral bone. J. Biomech., 7:203-213. Movle, D.D., J.W. Welborn, and F.W. Cooke 1978 Work to fracture of canine femoral bone. J . Biomech., 11t435-440. Nagurka, M.L., and W.C. Hayes 1980 An interactive graphics package for calculating cross-sectional properties of complex shapes. J. Biomech., 13r59-64. Nordin, B.E.C. 1966 International patterns of osteoporosis. Clin. Orthop. Rel. Res., 45:17-30. Parfitt, A.M. 1983 The physiologic and clinical significance of bone histomoruhometric data. In: Bone Histomomhometrv: Techniques and Interpretation. R.R. Recker, ed. CRCPress, Inc., Boca Raton, FL, pp. 143-223. Perzigian, A.J. 1973 Osteoporotic bone loss in two prehistoric Indian populations. Am. J . Phys. Anthropol., 39r87-96. Richman, E.A., D.J. Ortner, and F.P. Schulter-Ellis 1979 Differences in intracortical bone remodeling in three aboriginal American populations: Possible dietary factors. Calcif. Tissue Int., 28209214. Riggs, B.L., L.J. Melton, and H.W. Wahner 1983 Heterogeneity of involutional osteoporosis: Evidence for two distinct osteoporosis syndromes. In: Clinical Disorders of Bone and Mineral Metabolism. B. Frame and J.T. Potts, eds. Excerpta Medica, Amsterdam, pp. 337-342. Riggs, B.L., H.W. Wahner, E. Seeman, K.P. Offord, W.L. Dunn, R.B. 313 Mazess, K.A. Johnson, and L.J. Melton 1982 Changes in bone mineral density of the proximal femur and spine with aging. J. Clin. Invest., 70:716-723. Riggs, B.L., and L.J. Melton 1986 Involutional osteoporosis. N. Engl. J. Med., 314:1676-1686. Ruff, C.B. 1981 A reassessment of demographic estimates for Pecos Pueblo. Am. J . Phys. Anthropol., 54:147-151. Ruff, C.B. 1984 Allometry between length and cross-sectional dimensions of the femur and tibia in Homo supiens supiens. Am. J. Phys. Anthropol., 65:347-358. Ruff, C.B. 1987 Sexual dimorphism in human lower limb bone structure: Relationship to subsistence strategy and sexual division of labor. J . Hum. Evol., 16t391-416. Ruff, C.B., and W.C. Hayes 1982 Subperiosteal expansion and cortical remodeling of the human femur and tibia with aging. Science, 21 7r945-948. Ruff, C.B., and W.C. Hayes 1983a Cross-sectional geometry of Pecos Pueblo femora and tibiae-A biomechanical investigation: I. Method and general patterns of variation. Am. J. Phys. Anthropol., 60:359-381. Ruff, C.B., and W.C. Hayes 1983b Cross-sectional geometry of Pecos Pueblo femora and tibiae-A biomechanical investigation: 11. Sex, age, and side differences. Am. J . Phys. Anthropol., 60:383400. Ruff, C.B., and W.C. Hayes 1988 Sex differences in age-related remodeling of the femur and tibia. J. Orthop. Res., 6t886-896. Singh, I.J., and D.L. Gunberg 1970 Estimation of age a t death in human males from quantitative histology of bone fragments. Am. J . Phys. Anthropol., 33:373-381. Smith, R.W., and R.R. Walker 1964 Femoral expansion in aging women: Implications for osteoporosis and fractures. Science, 145: 156-157. Smith, R.W., and R.R. Walker 1980 Femoral expansion in aging women. Implications for osteoporosis and fractures. Henry Ford Hosp. Med. J., 28:168-170. Solomon, L. 1979 Bone density in ageing Caucasian and African populations. Lancet, pp.1326-1330. Takahashi, H., B. Epker, and H.M. Frost 1965 Relation between age and size of osteons in man. Henry Ford Hosp. Med. Bull., 13: 25-31. Thompson, D.D. 1978 Age-Related Changes in Osteon Remodeling and Bone Mineralization. Ph.D. dissertation, University of Connecticut, Storrs. Thompson, D.D. 1979 The core technique in the determination of age at death in skeletons. J . Forensic Sci., 24t902-915. Thompson, D.D. 1980 Age changes in bone mineralization, cortical thickness, and Haversian canal area. Calcif. Tissue Int., 315-11. Van Gerven, D.P. 1973 Thickness and area measurements as parameters of skeletal involution of the humerus, femur, and tibia. J. Gerontol., 28.40-45. Van Gerven, D.P., G.J. Armelagos, and M.H. Bartley 1969 Roentgenographic and direct measurement of cortical involution in a prehistoric Mississippian population. Am. J . Phys. Anthropol. 31; 23-38.