AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 7523-30 (1988)

Possible Case of Hyperparathyroidism in a Roman Period Skeleton From the Dakhleh Oasis, Egypt, Diagnosed Using Bone Histomorphometry

MEGAN COOK, EL MOLTO, AND COLIN ANDERSON Department of Pathology, University HospitaUUniversity of Western Ontario, London N6A 5C1 (M. C., C.A.), and Department ofdnthropology, Lakehead University, P7B 5E1, Thunder Bay (E. M.), Ontario, Canada

KEY WORDS surface

Histomorphometry, Archaeological bone, Resorption

ABSTRACT A histomorphometric study of thin femoral head sections of a skeletal sample from the Dakhleh Oasis, Egypt, dated from circa 36 B.C. to 400 A.D., identified an adult female (Dk31-Al) in her mid-50s with a high percentage resorption surface with tunneling resorption as is typically found in hyperparathyroidism. Five static histomorphometric bone parameters were measured with the following results for this individual: 1) mean wall thickness, 41.94 pm, 2) trabecular bone volume, 18.54%, 3) surface volume, 4,070 mm2/ cm3, 4) mean trabecular diameter, 132 pm, and 5) total resorption surface, 12.31%. The overall histomorphometric features and differential diagnosis support the diagnosis of hyperparathyroidism. We conclude that histomor- phometry of dried bone, particularly in this case where preservation is ideal, is a valuable investigative technique for palaeopathology.

Histomorphometry deals with the quanti- fication of microscopic structures. When ap- plied to the skeleton, using the concept of the bone remodeling system first proposed by Frost (19661, histomorphometry becomes an important diagnostic tool for assessing the presence of metabolic bone disease, altered bone balance, and turnover. The principle upon which histomorphometry is based is the fact that bone is consistently remodeling. Re- sorption and formation work as a unit (the bone remodeling unit or B.R.U.) and are the- oretically in equilibrium so that bone volume and shape more or less remain constant throughout life. Gain or loss of bone is the direct consequence of the cumulative bone balance occurring at the individual B.R.U. level. The values obtained from the bone his- tomorphometric parameter can determine if this balance is disrupted and if the B.R.U. is “uncoupled,” which in turn can be indicative of some type of pathology.

A number of recent studies have applied histomorphometry to archaeological bone samples in an attempt to elucidate bone dy- namics, both in normal and abnormal states, in past populations (Frost and Wu, 1967;

Stout and Teitaebaum, 1976; Wu et al., 1970; Weinstein et al., 1981; Saunders and Cook, 1985). The fundamental assumption under- lying the application of modern histomor- phometric techniques to the interpretation of archaeological bone is that the dynamics of bone physiology have essentially remained unchanged in Homo sapiens. The above stud- ies support this assumption. However, as is usual when dealing with archaeological bone samples, considerable caution is required, as only static bone parameters can be measured and postmortem diminution of osseous mi- crostructure usually occurs (Lengyel, 1968). These problems lessen as the quality of bone preservation improves. That histomorphom- etry has potential as an investigative tool in palaeopathology is now being recognized. In this paper this potential is shown, as histo- morphometry is used to diagnose hyperpara- thyroidism in a recently excavated skeleton from Egypt. The unique characteristics of this skeleton were revealed during the pro-

Received February 13, 1987; revision accepted June 16,1987.

@ 1988 ALAN R. LISS, INC

24 M. COOK, E. MOLTO, AND C. ANDERSON

Fig. 1. Location of the Dakhleh Oasis, Egypt. The tomb (Dk31) is located approximately 10 km from the village of Bashendi, which is the headquarters of the Dakhleh Oasis Project.

cessing of histomorphometric data for trabec- ular and cortical age estimates for the sample the skeleton was a member of.

Hyperparathyroidism is basically due to an excess of parathyroid hormone (PTH) se- creted by the parathyroid glands, and it may be primary or secondary in nature (Snapper, 1957). PTH is one of the hormones regulating calcium levels in the blood. Greatly simpli- fied, a lowered calcium blood level causes increased amounts of PTH to be released by the parathyroids. PTH stimulates osteoclast precursors; osteoclasts arrive at the bone sur- face and begin to resorb bone. Bone forma- tion is also increased. Calcium is released from the bone matrix, then into the blood. When the normal calcium blood level is reached, excess PTH secretion is halted. In hyperparathyroidism the feedback system is impaired, so excess PTH continues to be se- creted. The resorption goes on unchecked

leading to the increase of osteoclasts, fibro- blasts, and capillaries and may lead to the eventual formation of cysts and brown tu- mours. Histologically, hyperparathyroid pa- tients present unequivocal characteristics because of this unchecked resorption. As will be pointed out several factors have facili- tated the diagnosis made herein.

SKELETAL SAMPLE

The skeleton (Dk31-A1) is from Dakhleh, Egypt, a large Oasis located 660 km SSW of Cairo, at 25" 33"; 28" 55'E (Fig. 1). The Oasis is approximately 100 km east-west by 25 km north-south and has been isolated un- til recent times. Since 1977 Dakhleh has been intensively studied for evidence of past hu- man occupation. The Dakhleh Oasis Project (D.O.P.) is a multidisciplinary study directed by Anthony Mills, under the auspicies of the Department of Egyptology, Royal Ontario Museum, Toronto, Canada. The skeletal bi- ology component is one facet of this study. During the 1986 field season 40 individuals were recovered from Dk31, a Roman period tomb dated circa 36 B.C. to A.D. 400. These people lived in small agricultural communi- ties. Their tombs likely represent individuals who share some kinship (Melbye, 19831, al- though the exact nature of the within-group relationships have yet to be determined (Molto, unpublished). The hyperaridity of the Oasis produced the near perfect preservation of the bone noted above. In fact, the bone condition is comparable to that found in bi- opsy or autopsy material.' Standard obser- vations of morphology and pathology were recorded in the field, and femoral bone sam- ples from the heads and midshafts were ob- tained for a future aging study.

METHODOLOGY

The burials were sexed using the Phenice Method (Phenice, 1969) and other pelvic cri- teria outlined by Stewart (1979:102-116). Adults were aged by symphysis pubis mor- phogenesis following McKern and Stewart (1957) for males and Gilbert and McKern (1973) for females. These data are to be com- pared to the results of osteon age estimates from the femoral midshaft sections and the trabecular bone ages for the femoral heads. The femoral heads were selected for aging since in archaeological bone the trabeculae

'A histopathologist a t the University Hospital at the Univer- sity of Western Ontario mistook this specimen as a modern clinical sample.

HYPERPARATHYROIDISM IN -

Fig. 2. Location (A) on the posterior femoral head that is sampled for the trabecular aging technique. B indicates the specific area measured.

are best preserved in this region (Saunders and Cook, 1985), while osteon aging of the femoral midshaft is now an investigative standard (Ahlqvist and Damsten, 1969; Kerley, 1965; Kerley and Ubelaker, 1978; Thompson, 1978). While the latter methodol- ogy is well documented in the literature, tra- becular aging is less well known, and the method will be briefly outlined here.

A large wedge was taken from each femo- ral head, and sections from specific areas deep in the trabecular structure were sam- pled (Fig. 2), reconstituted in Sandison’s fluid (1955), dehydrated, and embedded in methyl methacrylate (Cook, 1982, unpublished). Measurements are not taken near the sub- chondral bone plate as the trabeculae are generally thickened in this region.

The following five static bone histomor- phometric parameters were measured using a semiautomatic image analyser (Videoplan of Carl Zeiss, Toronto) and “Osteoplan” soft- ware of Malluche et al. (1982). Table 1 sum- marizes the basic nomenclature, abbrevia- tions, and methodology used in this study. Note that for each of the parameter descrip- tions, population data for normal and control groups, adjusted for age, are provided in Ta- ble 2. In the interpretation of published his- tomorphometric values for normal or control groups, age means and age ranges have to be taken into account, since age influences the values (Lips et al., 1978; Courpron et al., 1980; Delling, 1973; Dequeker, 1975). The age ranges published for normal subjects generally vary from 18 to 100 years, but the mean age of these subjects is 2 5 0 years,

A ROMAN PERIOD EGYPTIAN 25

which reflects the greater availability of ob- taining samples from older individuals.

Mean wall thickness This is defined as the mean distance be-

tween the cement line (which marks the limit between bone resorption and formation) and the endosteal surface. The mean wall thick- ness is only measured on surfaces that show no signs of resorption or formation. Lips et al. (1978) in a study of the iliac crest of nor- mal subjects with an age range of 18-90 years and a mean age of 50.9 years found the mean wall thickness to be 49.70. Values for the femoral head studied at this hospital (Uni- versity Hospital) were higher. The mean wall thickness was 53.37 pm, with an age range of 50-100 years and an age mean of 72.85 years (Cook and Gibbs, 1986).

Trabecular bone volume The trabecular bone volume is the percent-

age of marrow space that is occupied by tra- becular bone. Femoral head values measured showed a mean trabecular bone volume of 21.18% in a population with an age range of 50-100 years and a mean age of 72.85 years (Cook and Gibbs, 1986). The trabecular bone volume decreases with age until it reaches its lowest level in women aged 80-100 years (5.5-16%) (Ellis, 1981).

Surface volume ratio This is the ratio of trabecular surface per

unit volume of bone expressed as square mm per cubic cm. The surface volume ratio is always higher in trabecular bone than in cortical and increases progressively as bone becomes more porous. Fazzalari et al. (1985) found the controls in their study of the femo- ral head histomorphometry had a surface1 volume ratio of 3,900 mm2/cm3, with an age range of 17-98 years and a mean age of 63.6 years. The University Hospital values were 2,521 mm2/cm3 (Cook, unpublished).

Mean trabecular diameter This is expressed in microns and is the

mean thickness of the trabeculae. Fazzalari et al. (1985) in the study quoted above found the mean trabecular diameter in normal sub- jects to be 150 pm. The University Hospital values were 255 pm (Cook, unpublished).

Total resorption surface This is the fraction expressed as a percent-

age of the trabecular surface that shows scal- loped or crenated surfaces. Vedi et al. (1984)

Fig. 3. This shows lamellar bone (A), fibrous lamellar bone (B), and tunneling resorption (C). x 100, polarized light.

Fig. 4. Tunneling resorption shown in subperiosteal region of the femur (midshaft). ~ 6 3 , polarized light.

HYPERPARATHYROIDISM IN A ROMAN PERIOD EGYFTIAN

TABLE 1. Nomenclature, abbreviations, and methodology used in the present histomorphometric analysis

27

Parameters Units Abbreviation Method of calculation

Mean wall thickness Microns M.W.T. No. of completed walls; semi- automatic analyser

Trabecular bone volume Percentage T.B.V. 50 fields counted with “Osteoplan” software

Surface/volume mm2/cm3 S N Measured with “Osteoplan” software; 50 fields counted

Mean trabecular diameter Microns D-Trab 50 fields measured with “Osteoplan” software

Total resorption surface Percentage T.R.S. 100 fields counted on semi- automatic image analyser

found a total resportion surface of 2.27% in a group of healthy individuals aged 19-80 years with a mean age of 46.2 years. Eriksen et al. (1986) compared primary hyperpara- thyroid patients with normal controls and found the normal to be 4.3% while the hyper- parathyroid patients had a total resorption surface of 6.5%. Femoral head values at Uni- versity Hospital were 4.17% (Cook, unpub- lished).

RESULTS AND DIAGNOSIS

Dk31-A1 is a female based on the criteria outlined by Phenice (1969). Gilbert and McKern (1973) scores for symphyses were each 15, giving her an age estimate of 55.71 f 3.24. An osteon age of 53 k 5 years using the Thompson (1978) technique provides strong support of the symphysis pubis age. It is important to note that these ages were estimated in a “blind” research design, which further strengthens the congruent results. Thus, Dk31-A1 falls within the age range for which the clinical data are most appropriate. Apart from the histomorphometric data to be discussed, the only noteworthy pathology is widespread demineralization (extremely light bones).

The histomorphometric values for Dk31- Al, the total Dakhleh sample, a recent sam- ple of normal (healthy) individuals and hy- perparathyroid patients are shown in Table 2. Dk31-A1 has a high T.R.S. value in com- parison with the Dk31 sample and the mod- ern controls. Her S N value of 4,070 is also high because of the increase in resorption spaces. Her T.B.V. is also comparatively low. According to Jaworoski (1983) the effects of hyperparathyroidism on bone volume de- pend on the age of the patient. The bone volume is reduced because of the expanding remodeling process.

Her trabecular bone sections show lamel- lar bone, fibrous lamellar bone, and tunnel- ing resorption. The subchondral bone plate also shows scalloping and crenation of the surfaces. These features are shown in Figure 3. Also noteworthy is the fact that sections of her cortical bone (femoral midshaft) show a slight increase in subperiosteal resorption, an example of which is shown in Figure 4. These histomorphometric features, though based on static bone, are consistent with those typically characteristic of hyperpara- thyroidism. The most diagnostic parameter is the high total resorption surface value. In Figure 5 the trabeculae of Dk31-A1 and a hyperparathyroid patient are shown to dem- onstrate the similarities in the T.R.S. parameter.

DIFFERENTIAL DIAGNOSIS

The following disorders must be considered in the differential diagnosis of hyperpara- thyroidism.

Postmenopausal osteoporosis As noted, Dk31-A1 had very light bones

that were diagnosed grossly in the field as osteopenia but could very well be considered postmenopausal osteoporosis given her skel- etal age and sex. Meunier (1980) concludes that for the histomorphometric diagnosis of osteoporosis the most discriminative param- eter is T.B.V., and no clinically diagnosed osteoporotic had a bone volume of over 16%. The T.B.V. of 18.59% for Dk31-A1 is very close to this threshold value and is support- ive of the diagnosis of osteoporosis. However, while osteoporotics show low T.B.V. scores, their T.R.S. values are not usually high and are within the normal range of the control samples (Meunier, 1980). Thus, the rate of osteoclastic versus osteoblastic activity in os-

28 M. COOK, E. MOLTO, AND C. ANDERSON

Fig. 5. The similarities shown between Dk31-A1 (1) and a hyperparathyroid patient (2). A: illustrates the tunneling resorption. x 100, ultraviolet light.

teoporosis is more homeostatic than is hyper- parathyroidism since it does not have an endocrinological impairment of the deminer- Healing fractures show woven bone, active alization process. The excessively high T.R.S. resorption, and formation (Sevitt, 1981). This value (12.31%) in Dk31-A1 likely indicates a female had no fractures of the femur or fem- defect in the feedback mechanism that is oral neck. AS noted the osteoporosis in Dk31- likely independent of osteoporosis. A1 is likely to be postmenopausal rather than

Fracturdtraumddisuse

HYPERPARATHYROIDISM IN A ROMAN PERIOD EGYPTIAN 29

one caused by disuse, as there were no signs of atrophy in the limbs.

Osteomalacia The following histological characteristics

of osteomalacia are taken from Meunier (1983) and Teitelbaum (1980).

1. An osteomalacia that is caused by a Vi- tamin D deficiency in the early stages in- duces a compensatory hyperparathyroidism, with increased resorption and formation. However, the mineralization defect is in ef- fect very soon. As the mean thickness and extent of the osteoid seams increase, the sur- face resorptive features subside. Eventually the seams replace the mineralized bone af- fecting the mechanical competence of the skeleton. Deformity, pain, and fracture result.

The F.D.A. daily recommended dose of Vi- tamin D is 400 IU. Dk31-A1 could be diag- nosed as having early Vitamin D deficiency but this is unlikely for the following reasons: a) the climate of Egypt, b) the women were not then in purdah, an Islamic practice where the women’s bodies, including faces, are en- tirely clothed, c) the whole grain cereals con- sumed contained Vitamin D, and most importantly, d) tunneling resorption is not usual in Vitamin D deficiency osteomalacia.

2. Hypophosphatemic osteomalacia ex- hibits no evidence of hyperparathyroidism. The hypophosphatemia affects the mineral- ization of the osteoid. There is no marked activation of the bone remodeling unit with no extensive resorption as is found in

Paget’s disease Jaworoski (1983) notes that this disease

also shows extensive resorption and forma- tion with woven bone and contiguous areas of bone formation. There is an increase of the T.B.V., and the architecture is abnormal. The characteristic mosaic pattern of cement lines is found. Little or no lamellar bone makes the M.W.T. difficult to measure. In Dk31-A1 there is no abnormal cement line pattern, no increased bone volume, and no other morpho- logical evidence of Paget’s disease.

Dk31-A1.

SUMMARY DISCUSSION

Recent advances in biomedical technology have expanded our ability to diagnose clini- cal disease, and many of these techniques can be used in palaeopathology. Advances in bone histological techniques (i.e., plastic embedding and cutting of calcified tissue) and

their analysis (semiautomatic image analy- sis computer systems) have facilitated the use of histomorphometry as a potential tool in studying bone dynamics in past popula- tions. The application of histomorphometry in this study, to provide a presumptive diag- nosis of hyperparathyroidism in a 1,600- 2,000-year-old burial, demonstrates the value of this technique. However, this diagnosis is accountable to other unique factors. First, the disease in question presents distinct his- tological changes (e.g., high T.R.S. values), which is not the norm for most diseases. This is one reason histology has traditionally had a limited role in palaeopathology (Sandison, 1968). Second, bone preservation in the Dk31 sample is ideal in that diagnosis of the inor- ganic tissue is not a problem. Most static histomorphometric parameters that are mea- sured on “clinical bone” could be measured on the Dk31 bones. Of note is the fact that for most archaeological samples T.R.S. can rarely be measured, thus eliminating the possibility of diagnosing hyperparathy- roidism.

In summary, while histomorphometry has inherent limitations, it can, in certain cir- cumstances, be a useful investigative tool for palaeopathology. If our diagnosis is correct, then histomorphometry has facilitated the diagnosis of the first case of hyperparathy- roidism in ancient bone. In the future, i t may be an aid in diagnosing other diseases in ancient bone that, to date, have eluded detection.

ACKNOWLEDGMENTS

We would like to warmly acknowledge the following individuals: Professor Anthony Mills, Director of the Dakhleh Oasis Project, for providing the opportunity to study the skeletal remains; President Robert Rosehart and Vice-president (Academic) G. Weller, Lakehead University, for giving Professor Molto permission to travel to Egypt during the academic term; Mr. Ben Kaminski, se- nior graphics designer, Lakehead Univer- sity, for the figures; Cindy Lamontagne, Anthropology Secretary, Lakehead Univer- sity, for typing the manuscript; and Mr. Luis Pugliese and Mei-Shu Shih, University of Western Ontario, London, for the histomor- phome’cric computer analysis.

The travel to Egypt was supported by an NSERC grant awarded to Professor Molto by the Senate Research Committee of Lakehead University.

30 M. COOK, E. MOLTO,

LITERATURE ClTED

Ahlqvist, J , and Damsten, D (1969) A modification of Kerley's method for the microscopic determination of age in human bone. J. Forensic Sci. 14(2):205-212.

Cook, ML, and Gibbs, M (1986) Age and sex identifica- tion in the Stirrup Court Cemetary. In WA Fox (ed): Occasional Publications No. 1. Studies in Southwest- ern Ontario Archaeology. London, Ontario: London Chapter of the Ontario Archaeological Society (Inc.),

Courpron, P, Meunier, P, and Lepine, P (1980) Mecha- nisms underlying the reduction with age of the mean wall thickness of trabecular basic structure unit (B.S.U.) in human iliac bone. In W Jee and M Parfitt (eds): Bone Histomorphometry: Third International Workshop. Paris: Societe Nouvelle de Publications Medicales et Dentaires, pp. 323-329.

Delling, G (1973) Age Related Bone Changes. Hamburg: Institute of Pathology, University of Hamburg, Monograph.

Dequeker, J (1975) Bones and aging. Ann. Rheumatic Dis. 34:lOO.

Ellis, HA (1981) Metabolic bone disease. In P Anthony and R MacSween (eds): Recent Advances in Histopa- thology. London: Churchill Livingstone, pp. 185-202.

Eriksen, EF, Mosekilde, L, and Melsen, F (1986) Trabec- ular bone remodeling and balance in primary hyper- parathyroidism. Bone 7:213-221.

Fazzalari, NL, Darractt, J, and Vernon-Roberts, B (1985) Histomorphometric changes in the trabecular struc- ture of a selected stress region in the femur in patients with osteoarthritis and fracture of the femoral neck. Bone 6:125-133.

Frost, HM (1966) The Bone Dynamics in Osteoporosis and Osteomalacia. Springfield, IL: Charles C. Thomas.

Frost, HM, and Wu, K (1967) Histological measurement of bone formation rates in unlabelled contemporary, archaeological and paleontological compact bone. In WD Wade (ed): Miscellaneous Papers in Paleopathol- ogy. Museum of Northern Arizona, Flagstaff, AZ, pp. 9-22.

Gilbert, GM, and McKern, TW (1973) A method of aging the female 0s pubis. Am. J. Phys. Anthropol. 38r31-38.

Jaworoski, ZF (1983) Bone histomorphometric character- istics of the metabolic bone disease. In R Recker (ed): Bone Histomorphometry Techniques and Interpreta- tion, pp. 241-263.

Kerley, ER (1965) The microscopic determination of age in human bone. Am. J. Phys. Anthrop. 23(2):149-163.

Kerley, ER, and Ubelaker, D (1978) Revisions in the microscopic method of estimating age at death in hu- man cortical bone. Am. J. Phys. Anthrop. 49:545-546.

Lengyel, I (1968) Biochemical aspects of early skeletons. In DR Brothwell (ed): The Skeletal Biology of Earlier Human Populations. London: Pergamon, pp. 271-288.

Lips, P, Courpron, P, and Meunier, P (1978) Mean wall thickness of trabecular bone packets in the human iliac crest Changes with age. Calcif. Tissue Res. 26:13-

pp. 107-116.

,v

AND C. ANDERSON

1 1 .

Malluche, HH, Sherman, D, Myer, W, and Massry, SG (1982) A new semiautomatic method for quantitative static and dynamic bone histology. Calcif. Tissue Int. 34:439448.

McKern, TW, and Stewart, TD (1957) Skeletal age changes in young American males. Analysed from the standpoint of age identification. Environmental h o - tection Res. Div. (Quartermaster Res. & Dev. Center, U.S. Army, Natick, MA), Tech. Rep. ER-45.

Melbye, FJ (1983) Human remains from a Roman period tomb in the Dakhleh Oasis, Egypt: A preliminary analysis. JSSEA 13-3:193-201.

Meunier, P (1980) Bone biopsy in diagnosis of metabolic bone disease. In D Coh, R Talmate, and M Lees (eds): Proceedings of 7th International Conference on Cal- cium Regulating Hormones. Amsterdam: Excerpta Medica, pp. 109-117.

Meunier, P (1983) Histomorphometry of the skeleton. In WA Peck (ed): Bone and Mineral Research. Annual 1 Excerpta Medica, pp. 202-222.

Molto, JE Skeletal biology and palaeoepidemiology of the people of Ancient Dakhleh in the Roman period. Unpublished.

Phenice TW (1969) A newly developed method of sexing the 0s pubis. Am. J. Phys. Anthropol. 30:297-301.

Sandison, AT (1955) The histological examination of mummified material. Stain Technol. 30:277-283.

Sandison AT (1968) Pathological changes in the skele- tons of earlier populations due to acquired disease and difficulties in their interpretations. In DR Brothwell (ed): The Skeletal Biology of Earlier Human Popula- tions. London: Pergamon, pp. 205-243.

Saunders, SR, and Cook, M (1985) Trabecular and corti- cal bone parameters in a small sample of protohistoric Iroquois. Paper presented at the Canadian Association of Physical Anthropologists, Thunder Bay.

Sevitt, S (1981) Bone Repair and Fracture Healing in Man. Chapter 19. London: Churchill Livingstone, pp. 296-305.

Snapper, T (1957) Bone Diseases in Medical Practice. New York: Grune and Stratton.

Stewart, TD (1979) Essentials of Forensic Anthropology. Springfield, n: Charles C. Thomas.

Stout, SD, and Teitelbaum, SL (1976) Histomorphome- tric determination of formation rayes of archaeological bone. Calcif. Tissue Res. 21:163-169.

Teitelbaum, S (1980) Histopathology of Osteomalacia Bone Histomorphometry. 3rd International Workshop, Sun Valley. W Jee and A Parfitt (eds). Paris: sOCiBt6 Nouvelle de Publications Medicales et Dentaires, pp. 475-482.

Thompson, DD (1978) Age-Related Changes in Osteon Remodeling and Bone Mineralization. Ph.D. thesis. Farmington: University of Connecticut.

Vedi, S, Tighe, JR, and Compston, JE (1984) Measure- ment of total resorption surface in iliac crest trabecu- lar bone in man. Metab. Bone Dis. Rel. Res. 5:275- 280.

Weinstein, RS, Simmons, DJ, and Lovejoy, CO (1981) Ancient bone disease in a Peruvian mummy revealed by quantitative skeletal histomorphometry. Am. J. Phys. Anthropol. 54:321-326.

Wu, KS, Schubeck, KE, Frost, HM, and Villanueva, A (1970) Haversian bone formation rates determined by a new method and in a mastodon, and in human dia- betes mellitus and osteoporosis. Calcif. Tissue Res. 6t204-219.