Citation
Prenatal limb growth in humans

Material Information

Title:
Prenatal limb growth in humans linear growth, allometry, locomotion, and skeletal age
Creator:
Warren, Michael W
Publication Date:
Language:
English
Physical Description:
xiv, 106 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Antrhopology thesis, Ph.D ( lcsh )
Dissertations, Academic -- Anthropology -- UF ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1997.
Bibliography:
Includes bibliographical references (leaves 99-105).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Michael W. Warren.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
028639610 ( ALEPH )
38827097 ( OCLC )

Downloads

This item has the following downloads:


Full Text
PRENATAL LIMB GROWTH IN HUMANS: LINEAR GROWTH,
ALLOMETRY, LOCOMOTION, AND SKELETAL AGE
MICHAEL W. WARREN
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA

1997




Copyright 1997 by
Nlichael W. Warren




This work is dedicated to two men that have served as positive influences in my life, both of whom passed away during the preparation of this manuscript.
My father, Bill Warren, was a caring, hard-working man. He is, in part, responsible for my inquisitive nature. Although he was an uneducated man, he had the ability to master skills, answer questions, and find truths that were his own. I hope he instilled in me a small measure of the moral character and sense of fairness that I always admired in him. He was always proud of me, no matter the extent of my accomplishments, and I expect that he would be especially proud to let the neighbors know about his son the anthropologist.
Dr. William R. Maples was, at first, larger than life. Over time he became my mentor, and finally, a friend. The simple fact that he had confidence in my ability has encouraged me to aim high. No one with such a role-model could fail to succeed.
Words cannot be found that express my sorrow at their passing. I was fortunate to have known them, and I dedicate this work to their memory.




ACKNOWLEDGMENTS

I gratefully acknowledge Dr. William Donnelly for access and permission to use the Radiographic Collection of the Department of Pathology, College of Medicine, University of Florida. His assistance in obtaining proper authority with the Institutional Review Board was the first step in this research project. I thank three members of my doctoral committee-Dr. Leslie Lieberman, Dr. Lynn Larkin, and Dr. Sue Boinski for their input, critique and encouragement during my graduate course work and the writing of this dissertation. I also thank Dr. William Leonard for early direction in finding the appropriate growth and development literature. Kendra Smith, Heather Walsh-Haney, Shuala Martin, Phoebe Stubblefield, Cheryl Katzmarzyk and other students, old and new, of the C.A. Pound Human Identification Laboratory are appreciated for the helpful, non-competitive, learning atmosphere that exists in the lab. Thanks are also in order for Karen Jones and Dian Leahy, both of whose guidance and help in navigating the academic waters were invaluable.
Special appreciation is due to Dr. Susan Ant6n, chair of my doctoral committee and my lead dissertation advisor. She provided both professional and personal support above and beyond the call of duty during some difficult times. Special thanks are also due Dr. Anthony Falsetti, whose statistical expertise contributed greatly to this manuscript. Drs. Ant6n and Falsetti and the Department of Anthropology provided much-needed financial support in the form of teaching assistantships during my doctoral studies. I have benefitted greatly from sharing their classrooms.
Finally, I thank my wife, Melinda, and my son, Zachary, for their help and
understanding. They have sacrificed many of their own dreams so that I might pursue




my own. My intellectual curiosity and dreams of ivory towers led me to abandon a well-paying career and move my family from their home. They have not only followed without complaint, but taken the lead in following new dreams of their own.
This research was funded by a Sigma Xi Grant-in-aid and a Forensic Sciences Foundation Lucas Research Grant.




TABLE OF CONTENTS
ACKNOWLEDGMENTS ........................................................................................ iv
LIST O F TA BLES .................................................................................................... viii
LIST O F FIG U RES ................................................................................................... xi
A BSTR A C T ............................................................................................................. xiii
CHAPTERS
1 INTRODUCTION ........................................................................................... 1
Review of the Literature .................................................................................. 2
Statement of Purpose ...................................................................................... 4
2 MATERIALS AND METHODS ..................................................................... 6
Radiographic Measurements .......................................................................... 7
T erm in ology .................................................................................................... 10
Sam ple Profile ................................................................................................ 10
M aternal Profile ............................................................................................. 12
Measuring Relative Long Bone Growth .......................................................... 12
3 CORRELATION BETWEEN CROWN-HEEL LENGTH
AND LONG BONE DIAPHYSEAL LENGTH .............................. 16
Mean Long Bone Lengths ................................................................................. 17
Correlation Between CHL and Length ........................................................... 19
4 RELATIVE GROWTH AND LIMB PROPORTIONALITY ........................... 24
Relative Rates of Growth ................................................................................. 24
Influence of Pathology on Linear Growth and Limb Proportions ................... 27
Sex Differences in Linear Growth and Limb Proportionality ......................... 28
Influence of Self-Identified Racial Category on Linear Growth
and Limb Proportions ....................................................................... 42
5 DETERMINATION OF GESTATIONAL AGE ............................................. 58
6 CONCLUSIONS AND SUMMARY ............................................................... 72
vi




APPENDICES
A INSTITUTIONAL REVIEW BOARD EXEMPT STATUS ............................. 75
B RAW DATA RECORDED FROM AUTOPSY PROTOCOLS ...................... 77
C AUTOPSY PROTOCOL (FACSIMILE) ......................................................... 98
REFERENCES CITED ............................................................................................. 99
BIOGRAPHICAL SKETCH ..................................................................................... 106




LIST OF TABLES

Table page
1. Frequency of pathological conditions determined at autopsy .................. 13
2. M aternal age profile .................................................................................. 14
3. Distribution of all cases with recorded CHL
by CHL group and approximate gestational age ........................... 16
4. Mean humerus length of all fetuses with recorded
CHL by CHL group (n=240) ........................................................... 17
5. Mean radius length of all fetuses with recorded
CHL by CHL group (n=226) ........................................................... 17
6. Mean ulna length of all fetuses with recorded
CHL by CHL group (N=226) ........................................................ 18
7. Mean femur length of all fetuses with recorded
CHL by CHL group (N=191) ........................................................ 18
8. Mean tibia length of all fetuses with recorded
CHL by CHL group (N=137) ........................................................ 19
9. Mean fibula length of all fetuses with recorded
CHL by CHL group (N=118) ........................................................ 19
10. Correlation matrix of r-values for CHL versus long bone
diaphyseal length ............................................................................ 20
11. Conversion from lunar months to weeks ................................................... 21
12. Mean humerus lengths for gestational age in weeks reported
by Warren, Russell et al., and Fazekas and Koza .......................... 22
13. Incremental increase (I1) in millimeters and relative increase in
percent (RIP) between CHL groups Humerus, radius and ulna ...... 26
14. Incremental increase (II) in millimeters and relative increase in
percent (RIP) between CHL groups Femur, tibia and fibula ..... 26
15. Comparison of mean humeral length between all cases,
"normal" cases, and "pathological" cases ..................................... 29




Table page
16. Comparison of mean radial length between all cases,
"normal" cases, and "pathological" cases ..................................... 29
17. Comparison of mean ulnar length between all cases,
"normal" cases, and "pathological" cases ..................................... 30
18. Comparison of mean femoral length between all cases,
"normal" cases, and "pathological" cases ..................................... 30
19. Comparison of mean tibial length between all cases,
"normal" cases, and "pathological" cases ..................................... 31
20. Comparison of mean fibular length between all cases,
"normal" cases, and "pathological" cases ...................................... 31
21. Descriptive statistics for Male sample (upper limb) by CHL group ........... 39
22. Descriptive statistics for Male sample (lower limb) by CHL group ........... 38
23. Descriptive statistics for Female sample (upper limb) by CHL group ......... 37 24. Descriptive statistics for Female sample (lower limb) by CHL group ......... 36 25. Descriptive statistics for "White" sample (upper limb) by CHL group ...... 47 26. Descriptive statistics for "White" sample (lower limb) by CHL group ....... 46 27. Descriptive statistics for "Black" sample (upper limb) by CHL group ........ 45 28. Descriptive statistics for "Black" sample (lower limb) by CHL group ........ 44 29. Proportional indices during ontogeny ...................................................... 48
30. Limb proportion indices for the entire fetal sample and
previously published values for adults .......................................... 48
31. Bonferroni t-test for significant differences in proportional indices ...... 49
32. Ratio of long bone lengths to femur and tibia length for white sample ........ 50 33. Ratio of long bone lengths to femur and tibia length for black sample ........ 51
34. Comparison of values of fetal body length given by
Scam m on and Calkins .................................................................... 61
35. Months of gestation and equivalent CHL ................................................. 62




Table AWg
36. Growth in length and increase in weight with respect to fetal age
(in lunar m onths) ............................................................................. 63
37. Fazekas and K6sa's data: Incremental increase (II) in millimeters
and relative increase in percent (RIP) between CHL groups .......... 64 38. Incremental increase (II) in millimeters and relative increase in
percent (RIP) between CHL groups ................................................ 65
39. Long bone diaphyseal lengths (mm) for Fazekas and K6sa (1978) ...... 66 40. Ranges for upper limb lengths by CHL group and "equivalent"
gestational age ................................................................................. 70
41. Ranges for lower limb lengths by CHL group and "equivalent"
gestational age ................................................................................. 70




LIST OF FIGURES

EVA=page
1 A typical fetal radiograph...................................................... 8
2. Measurements were taken parallel to the long axis of the
bone with a transparent metric scale.................................. 9
3. Comparison of Warren, Fazekas and K6sa, and Russell et al. data ....23
4. Combined distance curves for all bones showing
relative growth trajectories............................................. 25
5. Distance curve for the humerus ............................................... 26
6. Comparison of normal and pathological subsets for humerus length ...32 7. Comparison of normal and pathological subsets for radius length ...... 32 8. Comparison of normal and pathological subsets for ulna length ........ 33 9. Comparison of normal and pathological subsets for femur length....... 33 10. Comparison of normal and pathological subsets for tibia length ........ 34 11. Comparison of normal and pathological subsets for fibula length....... 34 12. Comparison of female and male means for humerus length
by CHL group ........................................................... 35
13. Comparison of female and male means for radius length
by CHL group ........................................................... 40
14. Comparison of female and male means for ulna length by CHL group .... 40 15. Comparison of female and male means for femur length
by CHL group............................................................ 41
16. Comparison of female and male means for tibia length
by CHL group............................................................ 42
17. Comparison of female and male means for fibula length
by CHL group............................................................ 42




18. Comparison of "Black" and "White" means for humerus length
by C H L group ............................................................................... 51
19. Comparison of "Black" and "White" means for radius length
by C H L group ............................................................................... 52
20. Comparison of "Black" and "White" means for ulna length
by CH L group ............................................................................. 52
21. Comparison of "Black" and "White" means for femur length
by C H L group ............................................................................... 53
22. Comparison of "Black" and "White" means for tibia length
by CH L group ............................................................................... 53
23. Comparison of "Black" and "White" means for fibula length
by CH L group ............................................................................... 54
24. Bivariate plot for CHL versus humeral length in raw data space ...... 54 25. Bivariate plot for CHL versus radial length in raw data space ............ 55
26. Bivariate plot for CHL versus ulnar length in raw data space ............. 55
27. Bivariate plot for CHL versus femoral length in raw data space ...... 56 28. Bivariate plot for CHL versus tibial length in raw data space .............. 56
29. Bivariate plot for CHL versus radial length in raw data space ............ 57
30. Comparison of cross-sectional humeral growth curves of the current
study and Fazekas and K6sa (1978) sample .............................. 67
31. Comparison of cross-sectional radial growth curves of the current
study and Fazekas and K6sa (1978) sample .............................. 67
32. Comparison of cross-sectional ulnar growth curves of the current
study and Fazekas and K6sa (1978) sample .............................. 68
33. Comparison of cross-sectional femoral growth curves of the current
study and Fazekas and K6sa (1978) sample .............................. 68
34. Comparison of cross-sectional tibial growth curves of the current
study and Fazekas and K6sa (1978) sample .............................. 69
35. Comparison of cross-sectional fibular growth curves of the current
study and Fazekas and K6sa (1978) sample .............................. 69




Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
PRENATAL LIM~B GROWTH IN HUMANS: LINEAR GROWTH,
ALLOMETRY, LOCOMOTION, AND SKELETAL AGE By
Michael W. Warren
December 1997
Chair: Susan C. Ant6n
Major Department: Anthropology
All human groups correspond to a general species-specific pattern of growth. Modifications in the rate and timing of growth events provide a direct measure of developmental response to the environment. Comparison of the linear growth of the long bones has been shown to be an effective way of demonstrating genetic and secular differences among populations. Most skeletal growth studies, however, have been directed toward the postnatal period. Prenatal studies have generally been clinically oriented and are poorly suited for use by anthropologists working with skeletal populations.
This study analyzes the linear growth and proportionality of the long bones for a sample of 398 fetuses based on radiographic measurements of diaphyseal length and recorded crown-heel length. The data are derived from the full-body radiographs of stillborn and spontaneously- or therapeutically-aborted fetuses delivered between 1976 and 1988. Additional data on crown-heel length, sex, self-identified "race," and pathology are taken from associated autopsy records.




Growth distance curves are plotted using crown-heel length (CHL) as a normalizing datum to reduce size effects. The sample is arbitrarily divided into 10 groups of crownheel length measurements, roughly corresponding to gestational ages from 4.5 lunar months to term. Descriptive statistics for each long bone are provided and relative and absolute growth of the long bones is examined in terms of relative increase in percent and linear incremental increase. All long bone lengths correlate significantly with CHL (r 2 t 0.8375; p < 0.05). Regression formulae are provided that predict CHL from long bone length.
Proportional indices show that relative growth is not significantly affected by
prenatal pathology. Males and females exhibit similar relative growth and proportions, however, "Blacks" had significantly higher brachial and crural indices than "Whites." The current data corresponds closely with that of Fazekas and K6sa (1978), whose data are derived from dry bone measurements of an eastern European population. This shows that (1) relative linear growth is similar between the two reference populations, and (2) radiographic measurements of fetal long bone length are accurate and the radiographic method may be used when skeletal preparation is impossible or undesirable.




CHAPTER 1
INTRODUCTION
Growth rates in humans follow a "mammalian curve" (Gavan and Swindler, 1966). However, the duration of growth and the timing of growth events is species-specific. During phylogeny the primates have modified the "mammalian" curve by altering the duration of growth of specific body segments to adapt to different environments and modes of locomotion (Gavan and Swindler, 1966). Normalization procedures that reduce size effects show that all human groups correspond to a general species-specific pattern of growth (Harrison et al., 1964; Lovejoy, Russell and Harrison,1990). When modifications in the rate and timing of growth events occur, they can provide a direct measure of an individual's developmental response to environmental circumstances. Comparative populational studies have shown that linear long bone growth is a sensitive indicator of differential response to environmental stressors (Acheson and Hewitt, 1954; Hewitt, Westropp and Acheson, 1955; Eleveth and Tanner, 1976; Armelagos et al., 1972; Hummert and Van Gerven, 1983; Mensforth et al., 1978; Merchant and Ubelaker, 1977; Y'Edynak, 1976). Therefore, comparison of relative long bone growth among populations provides data regarding environmental stress and general population health.
Studies of linear long bone growth are also used to assess growth sufficiency in
individuals. Growth curves for individuals that are accelerated or retarded relative to established standards may indicate an array of developmental, pathological or nutritional abnormalities.




2
Review of the Literature
Studies of long bone growth may be placed into 4 categories: (1) general biology and allometry; (2) comparative populational analysis; (3) clinical human growth and development, and (4) determination of developmental age. In general biology, differential growth analysis is used to examine the relationships that exist between growth "events" (Moss, 1954, 1955). Analysis of interspecies differences in mode and tempo of growth contribute to our understanding of the principles of growth and allometry, as well as the ways in which species have modified their growth patterns during phylogeny. (Gavan and Swindler, 1966; Jungers, 1982). Recent studies comparing various primate taxa with prehistoric skeletal populations provide explanations of functional and developmental morphological differences (Simpson et al., 1996; Leigh, 1996).
Anthropological studies of growth, like most general biological studies, utilize a populational or comparative perspective. Populational phenomena can be observed from cross-sectional studies of skeletal populations (Lovejoy et al., 1990). Evaluation of the health of skeletal samples necessarily relies on a normative sample of healthy, wellnourished individuals. Therefore, prehistoric samples are often compared with contemporary samples that are known to be well-nourished and relatively disease-free (Jantz and Owsley, 1984; Hummert and Van Gerven, 1983; Merchant and Ubelaker, 1977; Armelagos et al., 1972).
Most of the numerous studies on human growth and development involve longitudinal studies of clinical interest. Analysis of fetal weight and length at different stages of development are used to assess growth sufficiency and to detect factors that contribute to growth acceleration or retardation. Many of these studies are published as standards which establish normal percentiles of growth for a given age. Other standards based on appearance of ossification centers and/or bone morphology are used to assess "skeletal




3
age." Comparison of skeletal age with chronological age provides an indicator of level of maturity, predicted adult height, and other factors related to healthcare decisions (Tanner et al., 1983; Greulich and Pyle, 1959; Pyle and Hoerr, 1969; Roche, 1988). While these standards are usually used by physicians and other clinicians for growth assessment, they are also used by anthropologists in comparative studies and forensic work (Warren et al., 1997; Jantz and Owsley, 1984; Hummert and Van Gerven, 1983; Merchant and Ubelaker, 1977; Armelagos et al., 1972).
In the prenatal period, growth in length of long bones is linear and several standards have been established for both clinical and non-clinical applications (Halac et al., 1982; Fazekas and K6sa, 1978; Brenner et al., 1976; Birkbeck et al., 1975a, 1975b, 1976; Mehta and Singh, 1972; Russell et al., 1972; Lubchenco et al., 1966). Most studies of fetal growth have been conducted on fetuses aborted during the first 26 weeks of pregnancy (Bagnall et al., 1978, '82). Greater sample size is possible for this group than fetuses of greater gestational age because of an increased frequency of spontaneous and elective abortions during the first trimester of pregnancy.
The close correlation between fetal length (crown-heel length [CHLI or crown-rump length [CRLI) and long bone length to period of growth has been used to determine developmental age in both clinical and anthropological contexts. Most studies utilize CHL because that measurement has been found to contain less interobserver error than CRL (Scammon and Calkins, 1925).
The fetal population of interest to anthropologists is generally a different population than that encountered by medical clinicians. Since anthropologists generally examine skeletal material, it is difficult to know if a condition existed that may have affected the linear growth of the fetus. Therefore, for anthropological purposes, it may be undesirable to conduct a study of stillborn or aborted fetuses by discarding those with non-skeletal pathologic conditions, as the majority display an underlying abnormality. This study




examines all specimens without skeletal abnormalities, even in the presence of an underlying soft-tissue pathology.
Statement of Purpose
Despite quite extensive literature on growth of the fetus, we have yet to establish the extent of populational variation (Ubelaker, 1989). We continue to use standards developed primarily from white, healthy populations in the United States and Europe. Clinical radiography presents an opportunity to gather comparative data from a variety of populations. Although radiography is an invasive procedure, it is used extensively in the postnatal period in cases of suspected trauma, and it should be routinely used in the diagnosis of cause of prenatal abortion and death. This manuscript may serve as a guide for future anthropological research by demonstrating the utility of existing medical data.
This study establishes the correlation between radiographic lengths of long bones with CHL and examines the relative and absolute growth of the long bones in terms of differential growth rates in groups and individuals.
Chapter 3 looks at the relationship between CHL and long bone diaphyseal length and establishes relative growth rates for each of the long bones. The data are placed into groups of similar CHL that roughly correspond to gestational age as a normalization procedure so they can be compared with the results of other fetal studies.
Chapter 4 examines the influence of pathology, sex, and self-identified race on
absolute and relative long bone growth and proportionality for the study sample. Do specific pathologies affect linear growth and limb proportions? If not, then is data derived from the study of aborted fetuses representative of normal growth and development? Differences in limb proportionality have been found to exist between groups of different ancestry. Are these differences apparent during the fetal period or do they occur during later development?




5
Finally, Chapter 5 explores the possibilities and pitfalls of determining
developmental age from linear growth measurements. The results of the current study are compared with those of previous researchers whose data are used in bioarcheological and forensic settings to determine gestational age from crown-heel length.




CHAPTER 2
MATERIALS AND METHODS
This study uses the Radiological Study Collection and autopsy protocols of the
Autopsy Service, Department of Pathology, College of Medicine, University of Florida. This collection is a series of pathology records and radiographs taken of liveborn and stillborn fetuses from spontaneous, therapeutic, and elective abortions delivered af Stands Teaching Hospital in Gainesville, Florida since 1976. The radiographs were initially used to aid in the determination of cause of death. Their retention and use for anthropological study was granted by William Donnelly, M.D., Director of Autopsy Services. Permission to use confidential medical records was obtained from the Health Center Institutional Review Board of the College of Medicine, which granted exempt status to a research protocol submitted under the following category: "Collection or study of existing data, documents, records, pathological or diagnostic specimens, if these sources are publicly available or if the information is recorded by the investigator in such a manner that subjects cannot be identified, directly or through identifiers linked to the subjects" (Institutional Review Board 01, Protocol #033-97; see Appendix A).
The autopsy protocols record a limited maternal history, self-identified racial category of the mother, pathological conditions detected during dissection, and estimated gestational age. The gestational age of the fetuses was based on weight, CHL, last normal menstrual period or stage of organ development. Data were taken from 398 cases that met the following criteria: (1) both autopsy protocols and radiographs existed; (2) no conflicting data were recorded in the autopsy records; (3) at least one




measurable long bone diaphysis was present; and (4) no congenital skeletal abnormalities were detected by the author.
Radiographic Measurements
Three factors determine the accuracy of measurements taken from radiographs. The margins of the bone must be distinct, the bone must be parallel with the film plane, and the degree of x-ray magnification must be considered.
The radiographs were taken in a cabinet-type Hewlett-Packard Faxitron using unscreened Kodak EM-i diagnostic mammography film. This type of film produces extremely sharp images when used in unscreened cassettes (Figure 1). Longer exposure times are required (the purpose of the screen is to limit exposure times for living subjects), but this does not, of course, impose constraints for radiography of non-living subjects. X-ray intensity and exposure times can be adjusted for maximum image quality, as is the case with radiographs used in forensic anthropology case studies. Examination of the radiographs under a lupe demonstrates clear, distinct margins.
Each radiograph was assessed for proper positioning of the fetus. The fetuses were placed in a supine position with the extremities in anatomical position. The radiographs were not marked to indicate the left and right side of the specimen. Only long bones positioned parallel to the film plane were measured. Measurements were taken parallel to the long axis of the bone with a transparent metric scale to within 0.5 millimeters and the maximum length was recorded (Figure 2). When possible, measurements were taken of bones from both sides of the body. When the measurements were unequal the longest measurement was recorded. Trial radiographs taken with a Hewlett Packard Faxitron identical to the one that produced the study radiographs show that inclinations of less than 100 produce errors of less than 15.0%. This is consistent with reports from similar studies which state that a 150 angulation does not introduce significant error when




measuring long bones (Martin and Higginbottom, 1971; Owen, 1971). Recent studies have demonstrated greater error when measuring adult pelves due to absolute greater distances and variation between the x-ray source, the subject, and the film plane (Schroeder, et al., 1997).

Figure 1: A typical fetal radiograph.




Error produced by magnification was found to be less than 2.0% when the bone is 1 centimeter from the film plane. Radiographs of fetuses in utero will have a greater inherent magnification error due to the distance of the fetus from the film plane. Radiographic magnification of up to 19.0% has been recorded in studies of adult pelvic

Figure 2: Measurements were taken parallel to the long axis of the bone with a transparent metric scale.




bones due to varying lengths from x-ray source to subject to film plane. Although preliminary tests indicate that magnification error in this study is minimal, the figures recorded as diaphyseal lengths should be considered"radiographic lengths" and not true anatomical lengths.
Terminology
The sample represents the products of conception from 398 pregnancies dating from 1976 through 1988. Stillborn abortus is most properly referred to as a "specimen" since the remains are considered "tissue samples" and certificates of birth are not recorded. The live-born fetuses and live-born near-term infants included in this sample are issued certificates of birth and disposition of the body is regulated accordingly. These cases are most properly termed "individuals." However, many of the stillborns were given names and afforded the same postmortem treatment as live-born infants they are considered "individuals" in this sense, and the timing of the family's loss makes little difference as a practical (or emotional) matter. This dissertation will, with all proper respect, refer to each death as a "case," and each product of conception as a "specimen," a term often applied to adult cadavers in medical school dissection rooms. No philosophical position is implied by the use of this terminology.
Sample Profile
The data were derived from records of spontaneous abortions, therapeutic/induced abortions, and stillbirths from Shands Medical Center, a tertiary-care center located in Gainesville, Florida. The hospital is affiliated with J. Hillis Miller Health Center, the teaching arm of the hospital and home of the College of Medicine at the University of Florida. Most of the cases were delivered in Gainesville, however, a small number of the autopsies were referrals from hospitals and clinics in the smaller communities surrounding the Gainesville area. It may be assumed, for purposes of comparative




analysis, that all cases were delivered (and in the vast majority of cases, underwent development) between sea-level and 50 meters altitude. No socio-economic data were recorded in the autopsy records.
The total sample of fetuses (n=398) was identified by sex. Males comprised 57.6% of the sample (n=227), and females comprised 42.4% (n=167). Four cases omitted the sex of the fetus, either because sex was ambiguous due to developmental anomalies, or the autopsy record had conflicting information and the actual sex of the specimen could not be determined by the investigator.
In each case, the "race" of the child was based on the "race" of the mother. It is customary for this data to be self-reported when admission records are completed by the mother or her representative. However, in some cases, it has been the personal observation of the author that "race" is often clinician-reported based on perceived phenotype. "Race" was recorded for 387 cases: The category "white" comprised 57.8%/ (n=230); "black" was 40.7% (n=162); and "other" was recorded 1.3% (n=5). The percentage of aborted fetuses among the "black" category are significantly higher than the percentage of self-identified "blacks" in the Gainesville area.
A list of the frequencies of the various pathologies are shown in Table 1. In all cases of less than 7 months gestational age, immaturity was a contributing factor to death. Non-viable immaturity was almost always accompanied by hyaline membrane disease and/or cerebral intraventricular hemorrhage. Several autopsy reports listed immaturity as the primary cause of death, especially in cases in which no underlying pathology was found. In some cases, non-specific sepsis may have been secondary to fetal death in utero with fetal maceration. In addition, chorioamnionitis was almost universally associated with premature rupture of membranes.
It should be noted that the same sampling bias exists with studies of fetal
abortus and stillborn that exists with studies of skeletal populations. The population




does not represent a "normal" healthy population, but a population of non-survivors. This study examines all specimens without skeletal abnormalities, even in the presence of an underlying soft-tissue pathology. The impact of soft-tissue pathology on skeletal growth and development and the implications for anthropologists are addressed in chapter 4.
Maternal Profile
Maternal age was recorded for 346 cases. Maternal age distribution is shown in Table
2. The mean maternal age at abortion/birth is 23.75 years, ranging from 13 to 43 years of age. Only one autopsy record listed advanced maternal age as a contributing factor in fetal death.
Measuring Relative Long Bone Growth
In child growth studies where age is known a number of methods of assessing growth rate have been used: Gain and Shamir (1958) used annual increments, whereas Krogman (1950) and Meredith (1962) preferred to express increments as percentage of growth already attained. Bayer and Bayley (1959) divided the increment by average adult size and this quotient by the interval of time during which the increment was attained. This is similar to the relative growth rate used by Gray (1941). Gavan and Swindler use relative growth rates, that is, change per unit size per unit time. True average of an infinite series of such rates is attained by dividing the natural logarithms of size by the difference in time: (i1n S2- in S1) /(t2 t1), where S equals size and t equals time.
As discussed in later chapters, time intervals during the prenatal period can only be inferred from fetal development and/or maternal history. Therefore, calculation of true growth rates is not possible and the above-mentioned methods of determining growth rates are unavailable. This study draws from the method used by Armelagos et al.




Table 1: Frequency of pathological conditions determined at autopsy.
Pathology n percentage
Abdominal pregnancy 2 <11%
Abruptio placenta / marginal abruption 35 8.8%
Placenta previa 9 2.2%
Immature placenta; placental insufficiency; 29 7.3%
Premature rupture of amniotic membranes 25 6.3%
Amniotic band syndrome 6 1.5%
Chorioamnionitis; chronic villitis 18 4.5%
Anencephaly 15 3.8%
Hydrocephaly 12 3.0%
Congenital heart and/or vascular defects 22 5.5%
Maternal pre-eclampsia 9 2.2%
Maternal diabetes 6 1.5%
Maternal drug use 3 <1%
Maternal syphilis 2 <1%
Congenital syphilis 3 <1%
Potter's syndrome; renal agenesis or dysplasia 14 3.5%
Hydrops fetalis 9 2.2%
Down's syndrome and/or other chromosome 10 2.5%
Sepsis (non-specific) 14 3.5%
Multiple congenital skeletal malformations (i.e. 23 5.8%
Birth trauma and/or fetal anoxia 7 1.8%
Miscellaneous pathology that was a significant 15 3.8%
Note: Percentages do not equal 100%. Each case may have more than one or none of the categorized pathologies.




Table 2: Maternal Age Profile

Maternal Age n Percent Valid % Cumn. %
13 1 0.3 0.3 0.3
14 2 0.5 0.6 0.9
15 8 2.0 2.3 3.2
16 16 4.0 4.6 7.8
17 23 5.8 6.6 14.5
18 29 7.3 8.4 22.8
19 23 5.8 6.6 29.5
20 17 4.3 4.9 34.4
21 22 5.5 6.4 40.8
22 27 6.8 7.8 48.6
23 20 5.0 5.8 54.3
24 18 4.5 5.2 59.5
25 24 6.0 6.9 66.5
26 13 3.3 3.8 70.2
27 15 3.8 4.3 74.6
28 13 3.3 3.8 78.3
29 14 3.5 4.0 82.4
30 13 3.3 3.8 86.1
31 14 3.5 4.0 90.2
32 5 1.3 1.4 91.6
33 3 0.8 0.9 92.5
34 5 1.3 1.4 93.9
35 3 0.8 0.9 94.8
36 6 1.5 1.7 96.5
37 3 0.8 0.9 97.4
39 1 0.3 0.3 97.7
40 6 1.5 1.7 99.4
42 1 0.3 0.3 99.7
43 1 0.3 0.3 100.0
Missing data 52 13.0 ...

Total 398 100.0 100.0

100.0 100.0

Total

398




(1972) in their cross-sectional analyses of prehistoric populations of Sudanese Nubia in which curves of growth distance and growth velocity were used to show the relative increase in percent and the linear incremental increase of each long bone. Instead of using age based on dental development, however, the current study uses CHL to group cases into clusters of similar development. So, like Armelagos et al. (1972), they are not true growth rates, but expressions of proportionality between limb growth and fetal stature.
The data were analyzed on a Power Macintosh 8500/120 using SPSS Graduate PackTM Advanced version for PowerMacintosh, v. 6.1.1, and on an IBM-compatible personal computer using SAS for Windows95. Summary descriptive statistics are provided for the total sample and each sample subset as needed. Bivariate regressions are via the least squares method. Data were generated both log transformed and in raw data space. However, log transformations did not significantly affect the interpretation of the data and is thought to be better suited to analysis of shape variables than linear variables (Jungers et al., 1995). Only plots in raw data space are presented. Post-hoc comparison of the means is used to determine if significant differences exist in long bone proportions between groups. Also, slopes and intercepts of linear regression formulae, and proportional indices are compared using the Bonferroni t-test. Results with Alpha <
0.05 are considered significant.




CHAPTER 3
CORRELATION BETWEEN CROWN-HEEL LENGTH
AND LONG BONE DIAPHYSEAL LENGTH
Three-hundred and ninety-eight cases had both autopsy protocols and measurable radiographs. Of these, 252 cases had a CHL recorded at autopsy and at least one measurable long bone diaphysis. The sample was arbitrarily clustered into 12 groups of CHL, roughly corresponding to gestational ages from 4 lunar months to post-term. The first and last CHL groups are discarded for most of the analysis because they contain an insufficient number of cases. CHL length group distribution is reported in Table 3.
Table 3: Distribution of all cases with recorded CHL by CHL group and approximate gestational age.
CHL CHL in Approx. gestational n
group (mm.) age
0 < 210 > 4 lunar months 3
1 210-239 < 5 lunar months 10
2 240-269 5.0-5.5 lunar months 14
3 270-299 5.5-6.0 lunar months 35
4 300-329 6.0-6.5 lunar months 38
5 330-359 6.5-7.0 lunar months 41
6 360-389 7.0-7.5 lunar months 26
7 390-419 7.5-8.5 lunar months 28
8 420-449 8.5-9.0 lunar months 11
9 450-479 9.0-9.5 lunar months 17
10 480-520 9.5-10 lunar months 24
11 > 520 Post-term 5
Total ... 252




Mean Long Bone Lengths

Mean long bone lengths for each CHL group are reported in Tables 4 through 9. Mean long bone lengths are progressively longer for each successive CHL group, with the exception of the 420-449 CHL group, which introduced a sampling error for the Table 4: Mean humerus length of all fetuses with recorded CHL by CHL group. (n=240)

CHL grp. n mean sd range 210-239 10 27.85 2.46 24.0-31.0 240-269 14 33.18 4.90 26.0-47.0 270-299 35 36.50 2.78 29.0-42.0 300-329 38 39.11 3.40 31.5-47.0 330-359 38 43.21 3.09 37.0-49.0 360-389 26 46.87 2.73 42.0-54.0 390-419 28 53.36 5.35 46.0-69.0 420-449 11 56.14 2.92 52.0-61.0 450-479 16 58.28 5.81 44.0-69.0 480-520 24 64.02 5.07 55.0-78.5

Table 5: Mean radius length of all fetuses with recorded CHL by CHL group. (n=226)

CHL grp. n mean sd range
210-239 9 23.50 2.05 19.5-26.0 240-269 14 26.93 3.92 21.0-37.0 270-299 33 30.52 3.02 24.0-39.5 300-329 38 32.29 2.92 25.5-38.5 330-359 37 35.50 2.47 31.0-40.0 360-389 26 39.00 2.60 34.0-45.0 390-419 26 44.56 4.85 38.0-55.0 420-449 11 45.82 2.69 42.0-50.0 450-479 13 47.62 5.74 36.0-58.0 480-520 19 52.24 2.68 45.5-55.5




femur, tibia and fibula. The growth trajectory for the other bones was affected as well. The error is probably the result of a single case in which the CHL recorded in the autopsy protocol was significantly shorter than the actual measurement.
Table 6: Mean ulna length of all fetuses with recorded CHL by CHL group. (N=226)

CHL grp. n mean sd range
210-239 9 26.00 2.15 22.5-29.0 240-269 14 30.14 4.58 23.5-42.5 270-299 34 33.78 2.67 28.0-39.5 300-329 38 36.33 3.21 29.5-44.0 330-359 37 40.15 2.74 34.5-45.0 360-389 26 43.75 2.47 39.5-51.0 390-419 25 50.38 5.92 43.0-67.0 420-449 11 51.50 2.55 48.0-57.0 450-479 13 54.04 6.43 42.0-67.0 480-520 19 59.37 2.84 54.0-63.5

Table 7: Mean femur length of all fetuses with recorded CHL by CHL group. (N=191)

CHL grp. n mean sd range

210-239
240-269 270-299 300-329 330-359 360-389
390-419 420-449 450-479 480-520

28.60 35.38 39.00 42.71 47.40 51.42 61.19 65.33 64.67 74.20

2.66 5.65 3.59
4.61 3.90 2.99 6.92 6.22 8.26 7.56

24.5-33.0 28.0-51.0 29.0-46.0 32.0-56.0 39.0-55.0 47.0-57.0 52.0-75.5 55.0-73.0 49.0-73.0 62.0-81.0




Table 8: Mean tibia length of all fetuses with recorded CHL by CHL group. (N=137)

CHL grp. n mean sd range
210-239 10 24.80 2.66 20.5-29.0 240-269 12 29.08 3.09 24.5-35.5 270-299 28 33.09 2.96 24.5-38.5 300-329 28 35.23 3.43 29.0-42.5 330-359 27 40.28 3.24 35.0-47.0 360-389 11 44.05 2.48 40.0-49.5 390-419 9 54.22 4.86 47.0-63.0 420-449 3 54.33 5.69 48.0-59.0 450-479 4 53.00 7.57 42.0-58.0 480-520 5 62.60 5.59 55.0-69.0

Table 9: Mean fibula length of all fetuses with recorded CHL by CHL group. (N=118)

CHL grp. n mean sd range
210-239 8 24.38 2.55 20.5-28.0 240-269 12 28.08 2.75 23.5-33.5 270-299 23 31.48 2.92 23.5-37.0 300-329 23 33.17 3.52 27.0-40.0 330-359 25 38.02 3.19 33.0-44.5 360-389 9 41.72 2.33 39.0-47.0 390-419 7 51.21 4.40 47.0-60.0 420-449 3 51.00 4.44 34.0-54.5 450-479 4 49.75 7.37 39.0-55.0 480-520 4 58.38 5.79 51.0-64.5

Correlation Between CHL and Length

Long bone lengths were plotted against CHL. R-values are reported in Table 10. All long bone diaphyseal lengths correlate significantly with CHL (r2 > 0.8375; p < 0.01).




Table 10: Correlation matrix of r-values for CHL versus long bone diaphyseal length Humerus Radius Ulna Femur Tibia Fibula
CHL .9311 .9156 .9246 .9063 .9188 .9088
(249) (232) (232) (196) (142) (123)
Humerus ... .9848 .9876 .9879 .9869 .9881
(231) (231) (194) (141) (122)
Radius ... ... .9928 .9774 .9807 .9809
(231) (189) (137) (119)
Ulna ... ... ... .9854 .9879 .9897
(190) (138) (120)
Femur ...... ... ... .9910 .9911
(142) (123)
Tibia ... ... ... ... ... .9959
(123)
The correlation between fetal long bone length and CHL is consistent with that of adult long bone length and stature (Genoves, 1967; Trotter and Gleser, 1958). Leastsquares linear regression produced the following formulae for predicting CH-IL from radiographic bone lengths:
CHL = 45.571 + variablee) 6.839 7.704 CHL = 47.886 + (variable2) 8.196 8.696 CHL = 51.642 + (variable3) 7.193 8.097 CHL = 90.835 + variablee) 5.188 7.866 CHL = 82.858 + (variable5) 6.308 8.351 CHL = 79.677 + (variable6) 6.896 9.948 where, variable1 is humerus length; variable2 is radius length; variable3 is ulna length; variable4 is femur length; variables is tibia length; and variable6 is fibula length.




In order to compare my raw data with that of other investigators, I convert CHL to
gestational age in weeks as per Table 11. The growth rates from this study are compared with those of Russell et al. (1972), and K6sa (1989) by assuming the validity of relationship between fetal length and gestational age. However, the tenuous nature of determining gestational age based on length will be discussed in Chapter 5. Table 11: Conversion from lunar months to weeks.
Gestational age in lunar months Gestational age in weeks
5.5-6.0 lunar months 22-24 weeks
6.0-6.5 lunar months 24-26 weeks
6.5-7.0 lunar months 26-28 weeks
7.0-7.5 lunar months 28-30 weeks
7.5-8.5 lunar months 30-34 weeks
8.5-9.0 lunar months 34-36 weeks
9.0-9.5 lunar months 36-38 weeks
9.5-10 lunar months 38-40 weeks
The growth trajectories for the Fazekas and K6sa (1978) sample correspond closely with the current data. The slightly longer limb lengths of my sample through most of the curve are most likely a product of slight radiographic magnification error, although genetic or secular differences in growth cannot be ruled out. The humeral length at 22 weeks and term are nearly identical. This reduces the possibility that the curves illustrate a secular difference in size between the two samples.
Long bone lengths from Russell et al. (1972) are significantly higher than both Fazekas and K6sa (1978) and the current study. Table 12 shows the mean lengths for the humerus. Measurements at 28-30 weeks gestation are as much as 21 mm. greater than those reported by Fazekas and K6sa and more than 18 mm. longer than the current




study. This trend continues until term, with mean humeral length 7 mm. longer than the two other studies. However, the 3 Russell curves correspond with one another. Aside from the significantly longer humeral lengths, the slope of the curves is consistent with the slopes from my data and that of Fazekas and K6sa (Figure 3). Only data for the humerus are shown, however, the data for the other 5 long bones show the same trend.
Table 12: Mean humerus lengths for gestational age in weeks reported by Warren, Russell et al. (1972), and Fazekas and K6sa (1978).
Weeks Warren Russell et al. Russell et al. Russell et al. Fazekas &
gestation (LMP) (Osseous (Date of birth) K6sa
devel.)
28-30 46.87 66.33 60.00 n/a 45.00
30-34 53.36 64.57 61.20 61.80 50.40
34-36 56.14 64.97 66.91 65.11 54.30
36-38 58.28 68.84 69.13 67.63 58.40
38-40 64.02 70.14 71.33 71.00 63.10

The Russell et al. data were obtained by radiographic measurements of the long bones of in utero fetuses. The measurements were compared with fetal maturity obtained from 3 other parameters -- Menstrual history (LNMP), osseous assessment of knee and ankle development as reported by Hartley (1957), and extrapolation from the date of delivery. The greater humeral length reported by Russell et al. may be due to greater magnification error due to the distance of the in utero fetus from the film plane, or it may have been introduced by questionable normalization procedures based on estimation of gestational age (see Chapter 5). Using CHL as a normalizing datum allows data sets from different researchers to be compared without relying on additional studies on determination of gestational age.




Figure 3. Comparison of Warren, Fazekas and K6sa, and Russell et al. data

S
---- -..
- -

-I I I I I I I I

I I I I I
22-24 24-26 26-28 28-30 30-34
Weeks gestation

I I I
34-36 36-38 38-40

The data presented in this study are consistent with those of other published studies on fetal long bone growth, with the exception of the Russell et al. data. My data also fit studies of postnatal long bone growth at the natal period. These growth studies are based on longitudinal surveys of large numbers of healthy children and have been found to be extremely accurate in clinical practice.

80.00 70.00

60.00 1
en fin

ZgU. UU 40.00 30.00

20.00
10.00

4-

.-----------1 v are

- VY arren
-Russell et al.
(LNMP)
- - - Russell et al. (bone
development)
.... Russell et al. (by
birthdate)
....... Fazekas and Kosa

!
I
I
1




CHAPTER 4
RELATIVE GROWTH AND LIMB PROPORTIONALITY
Establishing standards of growth for a specific population leads to a better
understanding of populational and individual variation. Growth velocities and limb proportions of a given population can be compared with standards derived from wellnourished populations to better understand the impact of environmental conditions on normal growth and development.
Comparison between postnatal growth studies are normalized based on
chronological age. Chronological age is calculated from birth and, in general, variations in gestational duration are not taken into account. Since chronological age is not known for most skeletal populations, anthropological aging techniques such as dental eruption and wear patterns are used to categorize specimens according to developmental age (Merchant and Ubelaker, 1977; Y'Edynak, 1976; Armelagos et al., 1972). This study is hampered by a similar problem in that the time of conception cannot be known with absolute accuracy. However, the linear growth of the fetus allows the use of CHL as a normalizing datum for comparing limb proportionality between two populations. Error in determining age from dental eruption or other methods is not introduced to this type of analysis of fetal long bone growth.
Relative Rates of Growth
By plotting specific bone lengths against CHL it is possible to examine relative rates of growth for each long bone. Figure 4 illustrates relative growth trajectories for each long bone. Fetuses of short gestational age have humeri and femora of nearly equal length, with the femur only slightly longer than the humerus. By the end of the




gestational period the femora have elongated relative to the humeri. The disparity is caused by a greater relative growth velocity during the entire growth period, as opposed to a short period of rate increase. The tibia and fibula show a sampling error
80 70 60 50
40 30
20 10 0
1 2 3 4 5 6 7 8 9 10
CHL group
Figure 4. Combined distance curves for all bones showing relative growth trajectories. The Y axis represents long bone length in mm.;E0 = humerus; A = radius; = ulna; 0 femur; A = tibia; o = fibula. See table 3 for measurement ranges of CHL groups.
for CHL group 7 (i.e. 390-419 mm.). The femur, tibia, and fibula show an increased rate of growth during the period just prior to birth. Again, the low number of cases with lower extremity measurements in the latter CHL groups may produce a sampling error.
Tables 13-14 show the incremental increase in millimeters and the relative increase in percent for long bone growth by CHL group. The femur and tibia show greater incremental and relative growth than the humerus and radius at almost every CHL group. The exception is the 390-419 to 420-449, and the 420-449 to 450-479 groups, which demonstrate a sampling error of negative absolute and relative growth.




Table 13. Incremental increase (I) in millimeters and relative increase in percent (RIP) between CHL groups Humerus, radius and ulna.
Humerus Radius Ulna
CHL group II (mm) RIP II (am) RIP II (mm) RIP
210-239 to 240-269 5.33 19.1 3.43 14.6 4.14 15.9
240-269 to 270-299 3.32 10.0 3.59 13.3 3.64 12.1
270-299 to 300-329 2.61 07.2 1.77 05.8 2.55 07.5
300-329 to 330-359 4.10 10.5 3.21 09.9 3.82 10.5
330-359 to 360-389 3.66 08.5 3.50 09.9 3.60 09.0
360-389 to 390-419 6.49 13.8 5.56 14.3 6.63 15.2
390-419 to 420-449 1.18 02.2 1.26 02.8 1.12 02.2
420-449 to 450479 3.71 06.8 1.80 03.9 2.54 04.9
450479 to 480-520 5.74 09.8 4.62 06.7 5.33 09.9
Table 14. Incremental increase (II) in millimeters and relative increase in percent (RIP) between CHL groups Femur, tibia and fibula.
Femur Tibia Fibula
CHL group II (mm) RIP 1I (m) RIP II (mm) RIP
210-239 to 240-269 6.78 23.7 4.28 17.3 3.70 15.2
240-269 to 270-299 3.62 10.2 4.01 13.8 3.40 12.1
270-299 to 300-329 3.71 09.5 2.14 06.5 1.69 05.4
300-329 to 330-359 4.69 11.0 5.05 14.3 4.85 14.6
330-359 to 360-389 4.02 08.5 3.77 09.4 3.70 09.7
360-389 to 390-419 9.77 19.0 10.17 23.1 9.49 22.7
390-419 to 420-449 4.14 06.8 0.11 0.20 -0.21 -0.4
420-449 to 450-479 -0.66 -00.1 -1.33 -2.4 -1.25 -2.5
450-479 to 480-520 9.53 14.7 9.60 18.1 8.63 17.3




When the curve for the humerus is plotted with data from a postnatal study
(Maresh,1955), the data fit for the natal period. That is, the rate of growth is consistent through 6 months of age, and then accelerates during the period from 6 months to 1 year of age (Figure 5).

180 160
140 120 100 80 60
40

20].
U L Lfq Lf
LiC C U

L02
Age

Figure 5. Distance curve for the humerus prenatal (Warren) and postnatal (Maresh); reflecting the relative length of the humerus to gestational age.
Influence of Pathology on Linear Growth and Limb Proportions
Studies of long bone growth utilizing data from bioarcheological skeletal samples are necessarily cross-sectional in nature. This study is cross-sectional as well, and as such, is biased in that it does not present longitudinal data from a normal, healthy individual, but many different individuals that did not survive until adulthood (Johnston, 1962). However, the autopsy data contained in this study permit the examination of subsets of the sample categorized by the types and severity of pathology. No pathology was




noted in many cases; other cases exhibited an acute insult during the birthing process (i.e. abruptio placenta, placenta previa, or acute chorioamnionitis), or acute "pathology" associated with premature delivery of a normally developing fetus (i.e. hyaline membrane disease or persistent fetal circulation).- These cases are compared with those in which a significant pathological condition, possibly affecting normal growth, was noted during postmortem examination. Tables 15-20 show average long bone length by CHL group for all cases, "normal" cases, and "pathological" cases. Distance curves (Figures 5-10) compare the growth of the long bones of the "normal" subset and the "pathological" subset.
Linear growth is not significantly affected by most prenatal pathology in this study. Likewise, Brenner et al. (1976) found that data derived from spontaneous abortions in the late second and third trimesters were valid for interpreting normal fetal growth, even when underlying pathology was noted in the fetus. The similarity in proportionality does not mean that the "pathological" group is not growth retarded for gestational age in comparison to the normal group it may be, but this is impossible to determine because CHL is used as a normalizing measurement and true gestational age is unknown. However, long bone length relative to CHL does not differ between pathological and non-pathological sets. Figures 6-11 demonstrate that the proportional relationship between long bone length and CHL is both stable and predictable. The lack of effect of pathology on long bone length is consistent with other studies that find girth but not length to be affected by congenital conditions (Richards and Ant6n, 1991).
Sex Differences in Linear Growth and Limb Proportionality
To consider sex differences in fetal growth, the sample was divided into male and female groups and mean long bone lengths were determined for each CHL group (see Tables 21-24 for, descriptive statistics). Bagnall et al. (1978) found no significant




Table 15. Comparison of mean humeral length between all cases, "normal" cases, and "pathological" cases.
CHL group Means for SD Means for SD Means for SD
all cases "normal" "pathologic"
cases cases
210-239 27.85 2.46 27.06 2.06 31.00 ...
240-269 33.18 4.90 33.50 1.84 32.94 6.48
270-299 36.50 2.78 36.09 2.58 37.19 3.09
300-329 39.11 3.40 39.46 2.82 38.57 4.18
330-359 43.21 3.09 43.88 3.09 41.75 2.65
360-389 46.87 2.73 46.79 2.57 47.07 3.35
390-419 53.36 5.35 52.82 5.51 54.18 5.26
420-449 56.14 2.92 55.44 2.99 58.00 2.00
450-479 58.28 5.81 57.59 5.29 59.80 7.26
480-520 64.02 5.07 62.68 3.63 65.90 6.32
Table 16. Comparison of mean radial length between all cases, "normal" cases, and "pathological" cases.
CHL group Means for SD Means for SD Means for SD
all cases "normal" "pathologic"
cases cases
210-239 23.50 2.05 23.07 2.07 25.00 1.41
240-269 26.93 3.92 27.17 1.63 26.75 5.15
270-299 30.52 3.02 30.35 3.10 30.77 3.00
300-329 32.29 2.92 32.89 2.26 31.37 3.61
330-359 35.50 2.47 36.00 2.59 34.46 1.86
360-389 39.00 2.60 39.11 2.29 38.71 3.49
390-419 44.56 4.85 44.00 4.66 45.32 5.21
420-449 45.82 2.69 45.56 2.69 46.50 3.12
450-479 47.62 5.74 47.17 4.96 48.63 8.01
480-520 52.24 2.68 52.55 2.91 51.81 2.45




Table 17. Comparison of mean ulnar length between all cases, "normal" cases, and "pathological" cases.
CHL group Means for SD Means for SD Means for SD
all cases "normal" "pathologic"
cases cases
210-239 26.00 2.15 25.36 1.95 28.25 1.06
240-269 30.14 4.58 30.08 2.01 30.19 6.01
270-299 33.78 2.67 33.36 2.47 34.46 2.94
300-329 36.33 3.21 36.87 2.54 35.50 3.99
330-359 40.15 2.74 40.74 2.85 38.92 2.12
360-389 43.75 2.47 43.97 2.45 43.14 2.61
390-419 50.38 5.92 49.57 6.13 51.60 5.68
420-449 51.50 2.55 51.00 2.09 52.83 3.69
450-479 54.04 6.43 53.56 5.22 55.13 9.51
480-520 59.37 2.84 60.23 2.69 58.19 2.78
Table 18. Comparison of mean femoral length between all cases, "normal" cases, and "pathological" cases.
CHL group Means for SD Means for SD Means for SD
all cases "normal" "pathologic"
cases cases
210-239 28.60 2.66 27.75 2.17 32.00 1.41
240-269 35.38 5.65 34.83 1.94 35.86 7.76
270-299 39.00 3.59 38.55 3.27 39.77 4.10
300-329 42.71 4.61 43.05 3.84 42.23 5.63
330-359 47.40 3.90 48.40 4.09 45.54 2.78
360-389 51.42 2.99 51.06 2.68 52.29 3.73
390-419 61.19 6.92 60.14 6.49 62.00 7.52
420-449 65.33 6.22 62.50 5.45 71.00 2.83
450-479 64.67 8.26 64.20 9.15 67.00 ...
480-520 74.20 7.56 70.00 11.31 77.00 4.58




Table 19. Comparison of mean tibial length between all cases, "pathological" cases.

normala" cases, and

CHL group Means for SD Means for SD Means for SD
all cases "normal" "pathologic"
cases cases
210-239 24.80 2.66 24.00 2.27 28.00 1.41
240-269 29.08 3.09 29.25 2.36 28.92 3.92
270-299 33.09 2.96 32.56 2.65 33.79 3.31
300-329 35.23 3.43 36.32 3.17 33.55 3.24
330-359 40.28 3.24 40.97 3.65 39.42 2.55
360-389 44.05 2.48 43.69 1.33 45.00 4.77
390-419 54.22 4.86 55.13 5.72 53.50 4.61
420-449 54.33 5.69 53.50 7.78 56.00 0.00
450-479 53.00 7.57 51.33 8.33 58.00
480-520 62.60 5.59 62.00 9.90 63.00 3.61
Table 20. Comparison of mean fibular length between all cases, "normal" cases, and "pathological" cases.
CHL group Means for SD Means for SD Means for SD
all cases "normal" "pathologic"
cases cases
210-239 24.38 2.55 23.50 2.24 27.00 1.41
240-269 28.08 2.75 28.08 2.04 28.08 3.54
270-299 31.48 2.92 30.75 2.69 32.27 3.07
300-329 33.17 3.52 34.65 3.07 31.25 3.24
330-359 38.02 3.19 39.04 3.65 36.92 2.25
360-389 41.72 2.33 40.79 1.15 45.00 2.83
390-419 51.21 4.40 51.63 5.82 50.67 2.52
420-449 51.00 4.44 50.25 6.01 52.50 0.00
450-479 49.75 7.37 48.33 8.33 54.00 ...
480-520 58.38 5.79 56.00 7.07 60.75 5.30




70.00 60.00 50.00 40.00 30.00

20.00 +

10.00

0.00

11 N4 Lf)00C
CI I? I? I ?I
6 0 0 C C C
Nt 0 MO ON(7 N L) cC
N N CfO MO M en 11

CHL group
Figure 6. Comparison of normal and pathological subsets for humerus length.

60.00 50.00
40.00 30.00
20.00 S10.00

0.00~ 4 i
ON O N ON ON ON1
0.0 I I I I I
e~l 4 U) 00
CC Cg
6 C CO .
CHL group

Figure 7. Comparison of normal and pathological subsets for radius length.

' Normal set




70.00 60.00

50.00 40.00

30.00 20.00 10.00 0.00

Normal set

ON ON ON N ON ON (ON N
Cfo .D a*, -I Nr 0
eC4 C I C? C? L
CD 0 0 0 0 0 0 0 Q
- N 0 C N ON Cq M 00
(N rq rq 0 M It 't !V

CHL group
Figure 8. Comparison of normal and pathological subsets for ulna length.

80.00

70.00 60.00

50.00 40.00 30.00

-

- -INormal set

20.00 -1

10.00
0.00

ON N N LO ON ON N N
TO ON CI LI?
0 a a a a0 0 0 0
0 Co N ONNC Lg)rou
(I CY) Mo M C
CHL group

Figure 9. Comparison of normal and pathological subsets for femur length.




70.00 60.00 50.00
40.00 30.00
S20.00
10.00 0.00

all a, ON C;\ ON (ON ON ON ON
CfO ON CA Nn00
6\ CO C C:
L" CD Cf O ON
CHL group
Figure 10. Comparison of normal and pathological subsets for tibia length.

70.00 60.00 50.00 40.00

30.00 20.00 10.00
0.00

Normal set

. LO 00
ON 04 LO 0 ItOI

CHL group

Figure 11. Comparison of normal and pathological subsets for fibula length.

I I E R R

..o--* "




difference in length of the long bones between males and females of 8 to 24 weeks gestation. The current study extends their findings through the entire fetal period.
No significant difference in proportionality or relative growth was found between males and females. Distance curves illustrate the similarity in proportions between males and females for this sample (Figures 11-16). Success at differentiation between males and females during the fetal period, based on skeletal traits, has been confined to the metric and non-metric analysis of the pelvic bones (Boucher, 1957) and the dimensions of the vertebral bodies (Lippert and Lippert, 1960). Thus, sex cannot be distinguished by long bone length relative to CHL.
70.00 60.00 50.00
40.00 30.00
20.00 ,Female : 10.00 Male
0.00
ON, 011 MN 0NN O ON ON O
ON CA LO 00
CHL group
Figure 12. Comparison of female and male means for humerus length by CHL group.




Table 24. Descriptive statistics for Female sample (lower limb) by CHL group.

CHL group n

Femur mean sd

range

Tibia n mean

sA range

Fibula n mean sd

2.13 26.0-31.0 8.14 28.0-51.0 3.19 32.0-45.0 5.26 38.0-56.0
4.07 40.0-55.0 1.27 47.0-51.0 5.48 52.0-66.0 4.57 62.0-73.0 14.14 49.0-69.0 4.24 72.0-78.0

5
5 13
8
14
5
2
2
2
2

25.20 27.00
33.46 36.63
41.46 43.30 51.25 57.50 50.00
64.50

2.25 22.0-27.5
2.21 24.5-30.0 2.44 29.0-38.5 3.38 32.5-42.5 3.63 35.0-47.0 1.99 40.0-45.0 6.01 47.0-55.5 2.12 56.0-59.0 11.31 42.0-58.0 6.36 60.0-69.0

25.00 26.70
31.45 34.71 38.96
41.25 51.00 53.50
47.00 59.00

1.83 23.0-27.0 2.56 23.5-30.0 2.65 28.0-37.0
3.46 30.5-40.0 3.68 33.0-44.5 0.96 40.0-42.0
1.41 52.5-54.5 11.3 39.0-55.0 2.83 57.0-61.0

210-239
240-269 270-299 300-329 330-359 360-389
390-419 420-449 450-479 480-520

28.90 35.50
39.43 44.95 48.50 49.43 57.00 67.75 59.00 75.00

range




Table 23. Descriptive statistics for Female sample (upper limb) by CHL group.

CH-I, group

Humerus
ni mean sd range

Radius
n mean sd range

Ulna
n meanrag

210-239 5 28.30 2.11 26.0-31.0 4 24.25 1.26 23.0-26.0 4 26.50 1.78 25.0-28.5
240-269 6 34.17 6.74 27.5-47.0 6 26.67 5.23 23.0-37.0 6 30.47 6.16 25.5-42.5
270-299 14 36.75 2.64 32.0-42.0 12 30.88 3.49 25.5-39.5 13 33.58 2.41 29.5-38.0
300-329 12 40.38 3.54 35.0-47.0 12 33.08 3.14 29.0-38.5 12 37.38 3.46 33.0-44.0
330-359 20 43.68 3.18 37.0-48.0 18 35.39 2.42 31.0-40.0 18 40.22 2.65 35.0-45.0
360-389 8 45.75 2.38 42.0-50.0 8 37.19 1.96 34.0-40.0 8 42.44 1.95 40.0-46.0
390-419 9 51.56 2.26 48.0-56.0 7 41.71 2.64 38.0-46.0 7 47.21 2.41 43.5-51.0
420-449 7 57.36 2.84 53.0-61.0 7 46.07 3.3 42.0-50.0 7 52.14 2.84 49.0-57.0
450-479 6 55.67 8.57 44.0-69.0 6 45.83 7.84 36.0-58.0 6 52.42 8.98 42.0-67.0
480-520 8 62.75 4.06 57.0-70.0 6 51.75 2.6 48.0-55.0 6 59.00 2.92 54.0-62.0

range




Table 22. Descriptive statistics for Male sample (lower limb) by CHL group.

Femur n mean sd

range

Tibia
n mean sci

range

Fibula n mean sd

2.99 26.0-33.0 2.93 32.0-40.0 3.89 29.0-46.0 4.03 32.0-49.0 3.48 39.0-51.0 3.13 48.0-57.0 6.87 54.0-75.5 7.78 55.0-66.0 3.87 64.0-73.0 10.21 62.0-81.0

25.38 30.57 32.77 34.68 39.00
44.67 55.07
48.00 56.00 61.33

2.75 23.0-29.0 2.83 28.0-35.5 3.40 24.5-38.0 3.36 29.0-41.0 2.25 36.0-42.5 2.86 41.0-49.5 4.66 50.0-63.0
2.83 54.0-58.0 6.03 55.0-67.0

24.83 29.07 31.50 32.50 37.15
42.10 51.25
46.00 52.50 57.75

3.01 22.0-28.0 2.61 26.5-33.5 3.21 23.5-36.0 3.44 27.0-39.0 2.49 34.0-42.0 3.13 39.0-47.0
4.81 47.0-60.0
2.12 51.0-54.0 9.55 51.0-64.5

CHL group

210-239
240-269 270-299 300-329 330-359 360-389
390-419 420-449 450-479 480-520

29.25 35.29 38.71
41.72 46.30 52.24 63.09 60.50 67.50 73.67

range




Table 21. Descriptive statistics for Male sample (upper limb) by CHL group.

Humerus n mean sd

range

2.60 25.0-31.0 3.26 26.0-37.0 2.93 29.0-41.0 3.24 31.5-45.0 3.00 38.0-49.0 2.79 43.0-54.0 6.19 46.0-69.0 1.63 52.0-56.0 2.87 55.0-65.0 5.51 55.0-78.5

Radiu.s n mean

4
8
21 26
19 18 19
4
7 13

23.75
27.13 30.31 31.92 35.61 39.81
45.61 45.38
49.14 52.46

sd range

1.85 22.0-26.0 2.99 21.0-31.0 2.79 24.0-35.0 2.80 25.5-36.0 2.57 31.0-39.0 2.47 35.0-45.0 5.10 38.0-55.0 1.38 44.0-47.0 2.97 46.0-55.0 2.79 45.5-55.5

Ulna n mean

4
8
21 26 19 18 18
4
7 13

26.38
29.94 33.9 35.85
40.08 44.33 51.61 50.38 55.43
59.54

sd range

2.14 24.0-29.0 3.43 23.5-35.0 2.87 28.0-39.5
3.04 29.5-41.0 2.90 34.5-45.0 2.49 39.5-51.0 6.46 43.0-67.0 1.70 48.0-52.0 3.25 52.0-60.5 2.91 54.0-63.5

CHL group
210-239
240-269 270-299 300-329 330-359 360-389
390-419 420-449 450-479 480-520

4
8
21 26 18 18 19
4 10 16

28.25
32.44 36.33 38.52
42.70 47.36
54.21 54.00 59.85 64.66




60.00 50.00 40.00

30.00 S20.00 < 10.00
0.00

I I I I I I I I I I

C\ all ON ON ON all ON 0N
cf 0 ON cl LO~ 00 C4
0 0 0 0 0 0
-N 0 M ON cr LO 0
C4 cf M Nf
CHL group
Figure 13. Comparison of female and male means for radius length by CHL group.

80.00

70.00 60.00 50.00 40.00 30.00

20.00 +

10.00 0.00

.1 I I I I I I

IN
0, 0

CHL group

Figure 14. Comparison of female and male means for ulna length by CHL group.

II I

I




60.00 50.00
40.00 30.00
20.00

0.00 I I I I I I I I I
N 0I 0 0 0 0 0 0
N 0 0O a' N I
0 N N4 LO 0 C
C1 cf cn M c
CHL group
Figure 15. Comparison of female and male means for femur length by CHL group.

70.00 ,

60.00 50.00
40.00 30.00
20.00 10.00

0.00

CA Lf~) ODN
0 0 0 0 0 0
N 0 \-O LO O
CHL group

Figure 16. Comparison of female and male means for tibia length by CHL group.

IU.U




60.00 50.00
E40.00
30.00
20.00 Female
"- 10.00 Male
0.00 IIIII ...
ON~~3 OaOlOlN N ON O ON
ON) M N N
C? I? IT L9~
0 0 D 0 0 0 C
0~ M 0 C ON N In 0
N en Nn M M
CHL group
Figure 17. Comparison of female and male means for fibula length by CHL group.
Influence of Self-identified Racial Category on Linear Growth and Limb Proportions
Skeletal variation exists at the populational level. Despite extensive research into racial differences in the skeleton, however, describing the range of human variation and detecting racial affinity is perhaps the most difficult task of the osteologist. Group values for long bone proportions have been show to be diagnostic of racial affinity in adults, although there is considerable overlap in ranges when the formulae are applied to individuals (Krogman and Iscan, 1986; Modi, 1957; Schultz, 1937). Krogman (1955) found higher brachial and crural indices for American Blacks than for American Whites. Comparison of the data with earlier studies must acknowledge the basic flaw in racial analysis racial typologies poorly represent the dynamic, fluid nature of human variation. Earlier studies of racial differences in adult proportionality use the terms "American Blacks," "American Whites," Europeans, and "Blacks" as descriptors of sample subsets. However, if we understand the limitations of such groupings, we may




make assumptions about differential limb proportions based on the data. For the following analysis, we will assume that the groupings labeled as "American Blacks," "Blacks," and self-identified blacks primarily represent people of West African descent, and that "American Whites," "Whites," and self-identified whites primarily represent people of European descent. Descriptive statistics for white and black samples are given in Tables 25-28.
The intermembral index (arm length to leg length) for the fetal sample is high
compared with published adult indices, reflecting the early stages of development when the humerus is almost as long as the femur. The elongation of the femur towards the end of the prenatal period and during childhood results in a final intermembral index of near
0.700, as reported by Modi (1957) and Schultz (1937). Elongation of the femur and lower leg during development facilitates bipedal locomotion and is a unique feature of hominid ontogeny. The opposite has been noted in primates who utilize their forelimbs for brachiation and foraging on the trunks of large trees. Elongation of the forelimb in these species is seen early in ontogeny as well. Patterns of growth are species-specific as a result of both phylogenetic and functional constraints. Closely-related species that share locomotor and postural behaviors will also share similar limb proportions (Falsetti and Cole, 1992). Table 29 shows proportional indices for blacks and whites by CHL group.
The higher brachial and crural indices of the fetal sample are partially an artifact of measurement of the fetal diaphysis only. The secondary centers of the femora and humeri constitute a greater percentage of the final bone length than the secondary centers of the forearm and lower leg. Therefore, the ratio of diaphyseal length of the distal limb to the proximal limb is relatively longer than measurements taken of the maximum length of mature bones. Populational differences in proportion begin early in development. Table 30 lists values for the entire fetal sample for intermembral, brachial, and rural




Table 28. Descriptive statistics for "Black" sample (lower limb) by CHL group.

CHL group n

210-239
240-269 270-299 300-329 330-359 360-389
390-419 420-449 450-479 480-520

3
6
21 14 17 11
6
2
1

Femur mean

27.33 36.83 39.21
45.00 48.94 50.55 62.25
64.50
62.00

sd range
1.15 26.0-28.0 7.47 30.0-51.0
4.21 29.0-46.0 4.90 34.0-56.0 3.35 40.0-55.0 2.29 48.0-55.0 8.81 52.0-75.5 3.54 62.0-67.0

Tibia n mean

3
5 15 10 11
6
5
1
1

23.50 29.30 33.20 37.25
41.55 43.33 53.70 59.00
55.00

sd range
0.50 23.0-24.0 2.73 26.0-33.0 3.45 24.5-38.5 3.55 29.0-42.5 2.86 35.5-46.5 1.37 41.0-45.0 4.49 47.0-59.0

Fibula n mean

2
5 11
7
9
6
3

sd range

22.50 0.71 22.0-23.0 27.90 2.38 25.5-31.5 31.77 3.64 23.5-37.0 35.14 3.85 28.0-40.0 39.44 3.27 35.0-44.5 40.75 1.25 39.0-42.0 50.17 1.44 48.5-51.0 54.50 .....
51.00 .....




Table 27. Descriptive statistics for "Black" sample (upper limb) by CHL group.

CHL group
210-239
240-269 270-299 300-329 330-359 360-389
390-419 420-449 450-479 480-520

Humerus mean sA

range

26.50 1.32 25.0-27.5 35.25 6.15 30.5-47.0 36.98 3.23 29.0-42.0 40.57 3.21 33.5-47.0 44.47 2.79 37.0-49.0 46.00 1.79 43.0-49.0 53.50 4.98 48.5-63.0 56.30 3.35 53.0-61.0 62.00 4.85 58.0-69.0 65.63 7.39 55.0-78.5

Radius n mean sd

3
6
20 15 17 11
11
5
4
4

range

22.50 0.50 22.0-23.0 28.75 4.42 24.0-37.0 31.08 3.62 24.0-39.5 33.37 2.92 27.0-38.5 36.53 2.41 32.0-40.0 38.68 1.81 36.0-42.0 44.95 4.75 39.0-53.5 46.00 3.39 42.0-50.0 51.88 5.48 47.0-58.0 51.50 4.14 45.5-54.5

Ulna
n mean sd range

3
6
21 15 17 11 11
5
4
4

24.83 32.08
34.12 37.67
41.44 43.45 50.45 51.50 58.50 59.50

0.76 24.0-25.5 5.37 28.0-42.5 3.09 28.0-39.5 3.03 31.0-44.0 2.40 36.0-45.0 1.56 41.0-46.0 5.36 45.0-61.0 2.18 49.0-54.0 6.70 52.0-67.0 3.79 54.0-62.0




Table 26. Descriptive statistics for "White" sample (lower limb) by CHL group.

CHL group

Femur mean

sd range

Tibia
n mean sd range

Fibula n mean sd

3.01 24.5-33.0 3.67 28.0-40.0 2.51 34.0-44.0 3.86 32.0-48.5 3.95 39.0-55.0 3.39 47.0-57.0 5.96 54.0-72.0 7.72 55.0-73.0 8.26 49.0-73.0 3.77 72.0-81.0

7
7 13 18 16
5
4
2
4
4

25.36 28.93 32.96
34.11 39.41 44.90 54.88 52.00 53.00
64.50

3.05 20.5-29.0 3.53 24.5-35.5 2.39 28.0-38.0 2.87 29.0-41.0 3.29 35.0-47.0 3.38 40.0-49.5 5.92 50.0-63.0 5.66 48.0-56.0 7.57 42.0-58.0 4.20 60.0-69.0

25.00 28.21 31.21 32.31 37.22 43.67 52.00
49.25 49.75 60.83

2.66 20.5-28.0 3.17 23.5-33.5 2.20 27.5-36.0 3.11 27.0-39.0 2.94 33.0-43.5 3.06 41.0-47.0 5.94 47.0-60.0 4.60 46.0-52.5 7.37 39.0-55.0 3.75 57.0-64.5

210-239
240-269 270-299 300-329 330-359 360-389
390-419 420-449 450-479 480-520

29.14 34.14 38.68
41.25 46.26 52.15 60.55 65.75 64.67 77.25

range




Table 25. Descriptive statistics for "White" sample (upper limb) by CHL group.

Humerus
CHL group n mean sd

range

28.43 2.68 24.0-31.0 31.63 3.32 26.0-35.5 35.79 1.83 32.5-39.0 38.20 3.30 31.5-43.5 42.19 3.01 38.0-48.0 47.50 3.16 42.0-54.0 53.26 5.73 46.0-69.0 56.00 2.83 52.0-60.0 56.59 5.59 44.0-62.5 63.22 3.44 57.0-71.0

Radius n mean sd

6
8 13
22 20 15 15
6
9 15

Ulna

range

24.00 2.39 19.5-26.0 25.56 3.10 21.0-31.0 29.65 1.52 27.0-33.5 31.57 2.82 25.5-36.0 34.63 2.21 31.0-38.0 39.23 3.09 34.0-45.0 44.27 5.06 38.0-55.0 45.67 2.27 44.0-50.0 45.72 5.01 36.0-50.0 52.43 2.31 48.0-55.5

n mean
6 26.58 8 28.69
13 33.23
22 35.43 20 39.05 15 43.97 14 50.32 6 51.50 9 52.06 15 59.33

sd range
2.44 22.5-29.0 3.57 23.5-35.0 1.80 30.5-37.5 3.15 29.5-41.0 2.57 34.5-43.0 3.00 39.5-51.0 6.59 43.0-67.0 3.03 48.0-57.0 5.55 42.0-58.0 2.70 54.0-63.5

210-239
240-269 270-299 300-329 330-359 360-389
390-419 420-449 450-479 480-520




Table 29. Proportional indices during ontogeny.

Brachial index Crural index Intermernbral index
CHL group White Black White Black White Black
210-239 0.844 0.849 0.870 0.860 0.962 0.960
240-269 0.808 0.816 0.847 0.800 0.907 0.968
270-299 0.828 0.840 0.852 0.847 0.913 0.940
300-329 0.826 0.823 0.827 0.828 0.926 0.899
330-359 0.821 0.821 0.852 0.849 0.897 0.895
360-389 0.826 0.841 0.861 0.857 0.894 0.902
390-419 0.831 0.840 0.906 0.863 0.845 0.849
420-449 0.816 0.817 0.791 0.915 0.863 0.828
450-479 0.808 0.837 0.820 ... 0.869 ...
480-520 0.829 0.785 0.835 0.887 0.816 1.001

Table 30. Limb proportion indices for the entire fetal sample and previously published values for adults.

Indices Race Fetal (n) Adult Adult
(Warren) (Modi, (Schultz,
'57) '37)
Intermembral w 0.895 (119) 0.704 0.705
b 0.907 (82) 0.703 0.703
Brachial w 0.824 (204) 0.755 0.745
b 0.834 (146) 0.785 0.785
Crural w 0.851 (126) 0.833 0.833
b 0.860 (83) 0.862 0.862

Note: Schultz figures are for males; Under "Race," w = American Whites for Schultz, selfidentified "whites" for Warren, and Europeans for Modi; b = American Blacks for Schultz, selfidentified "blacks" for Warren, and "Black" for Modi. Data for Modi and Schultz taken from Krogman and Iscan, 1986, pp. 294-295.




indices by self-identified group. In the adult, no significant difference is found for the intermembral index between whites and blacks. The same is true for the fetal sample. A Bonferroni (Dunn) t-test for the proportional indices show that there is no significant difference in the intermembral index between the white and black sample (Table 31).
Table 31. Bonferroni t-test for significant differences in proportional indices.
"whites" "blacks"
Bonferroni
(Dunn) tVariable n x S.D. x S.D. Pr>F test
cc = 0.05
Intermernmbral index 201 0.895 0.046 0.907 0.037 0.067 ns
Brachial index 350 0.824 0.031 0.834 0.036 0.004 b > w*
Crural index 209 0.851 0.033 0.860 0.024 0.035 b > w*
However, the crural and brachial indices are significant at the 0.05 level, with blacks having proportionally longer distal limbs than whites. The same is reported by Modi (1957) and Schultz (1937) for white and black adults. Again, these are group values, and ranges are wide with considerable overlap.
Forensic identification of a single individual based on long bone indices would be foolish. Tables 32-33 show the ratio of long bone lengths to femur and tibia length for both the white and black samples. Small sample sizes for the 450-479 and the 480-520 CHL groups produce an obvious sampling error. Otherwise, it is clear that the small differences between the white and black samples are predicted by documented differences between black and white adults in earlier limb proportionality studies (Modi, 1957; Schultz,1937).




Distance curves for each long bone show that the black sample has relatively longer limbs than the white sample for most of the prenatal period. The curves are very similar, however, and the bones of the upper arm and forearm of both groups are nearly identical in relative length by birth. The black sample lacked sufficient data for the final two CHL groups. The distance curves for comparison of "Black" and "White" means for long bone length by CHL group are in Figures 18-23. The curves correspond closely until a sampling error occurs for the 420-449 CHL group and above. No data was available for "Black" femoral, tibial, and fibular measurements, and upper extremities measurements were few.
Table 32. Ratio of long bone lengths to femur and tibia length for "white" sample (after Armelagos et al., 1972).

CHL H/F R/F U/F T/F f/F F/T group

210-239 0.976 0.824 0.912 0.870 0.858 1.149 240-269 0.926 0.749 0.840 0.847 0.826 1.180 270-299 0.925 0.767 0.859 0.852 0.807 1.174 300-329 0.926 0.765 0.859 0.827 0.783 1.209 330-359 0.912 0.749 0.844 0.852 0.805 1.174 360-389 0.911 0.752 0.843 0.861 0.837 1.161 390-419 0.880 0.731 0.831 0.906 0.859 1.103 420-449 0.852 0.695 0.783 0.791 0.749 1.264 450-479 0.875 0.707 0.805 0.820 0.769 1.220 480-520 0.818 0.679 0.768 0.835 0.787 1.198

H/T R/T U/T f/T

1.121 0.946 1.048 0.986 1.093 0.884 0.992 0.975 1.086 0.900 1.008 0.947 1.120 0.926 1.039 0.947 1.071 0.879 0.991 0.944 1.058 0.874 0.979 0.973 0.970 0.807 0.917 0.948 1.077 0.878 0.990 0.947 1.068 0.863 0.982 0.939 0.980 0.813 0.920 0.943

H = humerus; F = femur; R = radius; U = ulna; T = tibia; f = fibula. CHL groups in mam.




Table 33. Ratio of long bone lengths to femur and tibia length for "black" sample (after Armelagos et al., 1972).
CHL H/F R/F U/F T/F f/F F/T H/T R/T U/T f/T
group
210-239 0.970 0.823 0.909 0.860 0.823 1.163 1.128 0.957 1.057 0.957
240-269 0.957 0.781 0.871 0.796 0.758 1.257 1.203 0.981 1.095 0.952
270-299 0.943 0.793 0.870 0.847 0.810 1.181 1.114 0.936 1.028 0.957
300-329 0.902 0.742 0.837 0.828 0.781 1.208 1.089 0.896 1.011 0.943
330-359 0.909 0.746 0.847 0.849 0.806 1.178 1.070 0.879 0.997 0.949
360-389 0.910 0.765 0.860 0.857 0.806 1.167 1.062 0.893 1.003 0.940
390-419 0.859 0.722 0.810 0.863 0.806 1.159 0.996 0.837 0.939 0.934
420-449 0.873 0.713 0.798 0.915 0.845 1.093 0.954 0.780 0.873 0.924
450-479 ... ... ... ... ... ... ... ... ... ...
480-520 1.059 0.831 0.960 0.887 0.823 1.127 1.193 0.936 1.082 0.927
H = humerus; F = femur; R = radius; U = ulna; T = tibia; f = fibula. CHL groups in mm.
70.00 60.00 50.00
40.00
30.00
i 20.00 "Blacks"
10.00 "Whites"
allNON ON O ON ON ON ON
0.0 O 00 N NI
IIII I? CI IT I?
CD 0 0 0 0 0 0
CHL group
Figure 18. Comparison of "Black" and "White" means for humerus length by CHL group.




60.00
50.00.40.00 S30.00
. 20.00 a
" "Blacks"
10.00 -------- "Whites"
N C1 N MO MO en C
0.00 1111I I11
0 Ox Ox 0 O 0, 0 0 0 0
t- N 0 C5 \O ONI N", ",
CHL group
Figure 19. Comparison of "Black" and "White" means for radius length by CHL group.
60.00 50.00
40.00
30.00 20.00
2."Blacks"
10.00 "------ Whites"
0.00 I I I I I I I II
0 0 0 C1 0 0o a 0 ,
. 0 O ON N to
CHL group
Figure 20. Comparison of "Black" and "White" means for ulna length by CHL group.




80.00 70.00 60.00 50.00 40.00 30.00 -

20.00 -+

10.00 0.00

J I I I I I I I

CHL group
Figure 21. Comparison of "Black" and "White" means for femur length by CHL group.

70.00
60.00 50.00
E
40.00
S30.00
S20.00
10.00 0.00

I

a\ 0 q 0
'-4 N4 M M
CHL group

Figure 22. Comparison of "Black" and "White" means for tibia length by CHL group.




70.00 60.00 50.00
40.00 30.00
20.00 10.00 0.00

aN CN ON y
071 0 0 -4 0 0
0: M' O N U)
CHL group

Figure 23. Comparison of "Black" and "White" means for fibula length by CHL group.

90.0 80.0
70.0 60.0 50.0
40.0 30.0
20.0 10.0 0.0 100.0

200.0

300.0 400.0 500.0
Crown-heel length in mm.

600.0

700.0

Figure 24. Bivariate plot for CHL versus humeral length in raw data space (n=248)

I I I I I I

I i

- - -

E




60.0 50.0 E 40.0 30.0 20.0
10.0 0.0
100.0

200.0

300.0 400.0
Crown-heel length in mm.

Figure 25. Bivariate plot for CHL versus radial length in raw data space (n=231)

70.0 60.0 50.0
40.0 30.0
20.0 10.0
fn -I

UU J!,. I
100.0 200.0 300.0 400.0 500.0 6'
Crown-heel length in mm.
Figure 26. Bivariate plot for CHL versus ulnar length in raw data space (n=231)

00.0

500.0

600.0

!




0 .

90.0 80.0 70.0 60.0 50.0 40.0 30.0 20.0 10.0
0.0
100.0

200.0

300.0 400.0
Crown-heel length in mm.

Figure 27. Bivariate plot for CHL versus femoral length in raw data space (n=195)

0.
* *
* *e
ge *g
0S
08$e

I I I I I

200.0

300.0 400.0
Crown-heel length in mmnun.

Figure 28. Bivariate plot for CHL versus tibial length in raw data space (n=142)

I I

500.0

600.0

70.0 60.0 50.0
40.0 30.0 20.0 10.0

0.0 1 100

.0

500.0

600.0




70.0 60.0 50.0 40.0
-- 30.0
20.0 10.0

S
0
S 4~ 3..
6

U.U I Ii100.0 200.0 300.0 400.0 500.0 600
Crown-heel length in mm.
Figure 29. Bivariate plot for CHL versus fibular length in raw data space (n=121)

'.0

|




CHAPTER 5
DETERMINATION OF GESTATIONAL AGE
A basic technique in osteology is determination of age at death. There is a vast literature on determination of age from analysis of the skeleton and dentition (e.g., McKern, 1970; Kerley, 1970, Krogman 1962; Todd, 1939). These studies examine the maturation, degeneration, or metamorphosis of various aspects of the skeleton to arrive at a probable age at death. The investigator calculates a skeletal age and chronological age since birth is assumed to fall within an expected range. Techniques for determining the age of fetal or perinatal remains are more difficult because they are designed to calculate the time since conception. In most cases, the time of conception cannot be known with certainty. Many studies calculate gestational age based on the mother's last normal menstrual period (LNMP) with the assumption that conception usually occurs within one week of the last menstrual period, however an error of as much as two weeks is possible (Naeye and Dixon, 1978). Additionally, errors in reporting occur, such as when implantation bleeding is mistaken for the last menstrual period. Even in cases of isolated coitus, fertilization may not take place immediately and one must rely on the history provided by the mother (Peterson et al., 1989; Golbus and Berry, 1976).
The possibility of miscalculating gestational age presents a problem for growth and development studies. Several methods have been used to omit infants that are older than their calculated gestational age. Authors have excluded outliers in bimodal curves and scattergrams (Gruenwald, 1966; Lubchenco et al., 1963), used maternal examinations and histories (Babson et al., 1976; Brenner et al., 1976), or considered clinical tests of a neonate's maturity (Babson et al., 1976; Brenner et al., 1976;




Gruenwald, 1966; Lubchenco et al., 1963) in order to establish the validity of gestational ages based on LNMP. Exclusion of fetuses and infants that are younger than their calculated gestational age is more problematic. Isolated cases may be growth retarded but present no other signs of pathology. Manipulation of data in order to arrive at "valid" gestational ages is a circular process, and data are no doubt excluded that would broaden the variability of normal fetal growth.
The close correlation between CHL or crown-rump length (CRL) and long bone length to period of growth has been used to determine fetal developmental age. Estimation of developmental age in fetuses is used to predict the expected birth date and diagnose the presence of pathologies that affect normal growth. It is also an issue in cases of forensic identification, or when the determination of viability is necessary for legal disposition of death (Weaver, 1986; Kuroda, 1970). In populational studies, differentiation between infants and fetuses is an important factor in paleodemographic reconstructions, providing clues about abortion rates, maternal deaths during childbirth, and early childhood disease (Mensforth, 1985; Jantz and Owsley, 1984; Hummert and Van Gerven, 1983; Merchant and Ubelaker, 1977; Armelagos et al., 1972; Johnson, 1961, '62, '68, '69; Walker, 1968).
Current methods for determining gestational age in living fetuses are based on
sonographic imaging to determine head size, long bone length, or femoral head cartilage diameter (Mashiack, Blankstein and Stem, 1975; Jeanty and Romero, 1984; Lange and Manning, 1984). Earlier studies utilizing radiography (Christie, 1950) have been abandoned because of the risks of radiation exposure to the mother and fetus. In the case of premature delivery or abortion, Olivier and Pineau (1958, 1960) have derived linear regression formulae based on CHL to determine gestational age. These studies are currently used by pathologists to determine developmental age of intact, fleshed fetuses.




Trotter and Gleser (1952, '58, '77), Genov~s (1967), Hoffman (1979) and others have generated stature formulae based on long bone lengths of adults and children. Anthropologists have found secular and genetic differences in proportionality and stature between populations, so formulae for predicting stature have been derived for different populations. Similarly, several factors may affect stature and long bone length in fetuses and differences in growth rates may exist between populations. Genetic diversity, secular trends related to nutrition or access to medical care, or the environment (particularly high altitude) may make a given population unsuitable as a standard for other populations.
Numerous data (Toldt 1879; Zangemeister 1911, '12, '17; Dietrich 1925; Scammon and Calkins 1929; Guthmann and Kn6ss 1939; Girtner 1947; Brock 1954; Olivier and Pinneau 1958, 1960) confirm the correlation between the duration of pregnancy and the development of the fetus. While the exact time of conception is not known with certainty in the vast majority of cases, the preponderance of data using "estimated" time of conception based on different criteria suggest that there is a close correlation between linear length and gestational age. By means of "Haase's rule," the age of the fetus can be determined from its body length. It is known that until the 5th lunar month the fetal body length in centimeters can be closely estimated by squaring the number of the months of pregnancy, and after this time by multiplying the months by 5 (Fazekas and K6sa, 1978; Brock, 1954).
Brock (1954) published data on how the values of body length of fetuses derived from known periods of pregnancy correspond to those obtained by Haase's rule. For comparison Brock used the data of Streeter (1920, 1948, 1949), and Scammon and Calkins (1929). Table 34 (from Fazekas and K6sa, 1978) below shows that there are no significant differences between the actual data and the values obtained by Haase's calculation. Since Haase's rule is curvilinear for the first 5 lunar months the values




obtained are somewhat high for the first 3 months of pregnancy, whereas in the later lunar months they nearly coincide with values of the actual body length of the fetuses.
Table 34. From Fazekas and K6sa's Table 7, pp. 31 Comparison of values of fetal body length given by Scammon and Calkins with the values obtained on the basis of Haase's Rule (after Brock).
Period of pregnancy Length Increase in Increase in Length
calculated from the of length of fetus length of fetus obtained on
last menstruation fetus (cm) in 4-week in 4-week the basis of
(weeks) periods periods Haase's
(.1.) in mm. (R.I.P.) calculation
8 ... ... ... 4
12 7.0 ...... 9
16 15.5 8.5 122 16
20 22.7 7.2 46 25
24 29.2 6.5 29 30
28 35.0 5.8 20 35
32 40.4 5.4 15 40
36 45.4 5.0 12 45
40 50.2 4.8 11 50
.I. = incremental increase in mm.; R.I.P. = relative increase in percent.
The most often used method of calculating developmental age in bioarcheological and forensic contexts requires the measurement of dry bone (Fazekas and K6sa, 1978). However, the Fazekas and K6sa sample is small (n=138) and consists of a homogenous eastern European sample which may be inappropriate for use in determining developmental age in contemporary Americans. Furthermore, the measurement of dry bone requires skeletal preparation and may be unnecessarily invasive. Huxley (1996) has recently addressed the problem of dry bone shrinkage, which may add an additional source of error.
Fazekas and K6sa grouped their fetuses on the basis of body length, not "period of pregnancy," because of their inability to determine the exact time of conception. They




used "Haase's Rule" which they feel has been shown by other researchers (Dietrich, 1925; Scarnmon and Calkins, 1929; and Streeter, 1920, 1948, 1949) to be very accurate when compared with gestational age based on LNMP. They divided the fetuses in different groups as follows: Fetuses 40 cm. long were included in the age group of 8 lunar months. In the same group were fetuses 39 and 41 cm. long, whereas those with a body length of 42 and 43 cm. (born at the beginning of the 9th lunar month) were included in the age group of 8.5 lunar months. So, lunar months were classified by the fetal lengths listed in Table 35.
Table 35. Months of gestation and equivalent CHL (from Fazekas and K6sa, 1978)

Months of gestation
5 lunar months 5.5 lunar months 6 lunar months 6.5 lunar months 7 lunar months 7.5 lunar months 8 lunar months 8.5 lunar months 9 lunar months 9.5 lunar months 10 lunar months

Equivalent CFLL length
24-26 cm.
27-28 cm.
29-31 cm.
32-33 cm.
34-36 cm.
37-38 cm.
39-41 cm.
42-43 cm.
44-46 cm.
47-48 cm.
49-51 cm.

It is unclear how fetuses less than 5 lunar months were categorized according to CHL. Fazekas and K6sa used the following table (Table 36) to compare data from two previous studies to estimations from Haase's rule. The body length as calculated by the




use of Haase's Rule compares favorably to the data of Dietrich (1925), and almost exactly coincides with the data of Scammon and Calkins (1929).
Table 36. Fazekas and K6sa's Table 8, pp. 31: Growth in length and increase in weight with respect to fetal age (in lunar months).
Period of Body length of Body length of Body length of
pregnancy fetuses (cm) fetuses (cm) fetuses (cm)
(lunar months) Dietrich Scammon & Calkins Haase's calculation
2 3.0 ... 4
3 9.8 7.0 9
4 18.0 15.5 16
5 25.0 22.7 25
6 31.5 29.2 30
7 37.1 35.0 35
8 42.5 40.4 40
9 47.0 45.4 45
10 50.0 50.2 50
post-term ... ... 54
The data from the current study have been compared with those of Fazekas and K6sa by converting their data into CHL groups. Table 37 shows Fazekas and K6sa's data in terms of incremental increase and relative percent increase of long bone growth. My data follows in Table 38. 1 have compared the mean long bone lengths to my data (Table 39), as well as plotted long bone growth relative to CHL for their data (Figures 30-35). It is clear that Fazekas and K6sa's data serves as a valid reference population for determining gestational age in the United States. Results obtained by my radiographic method correspond with the dry bone measurements of Fazekas and K6sa and, therefore, the method can be used in instances where skeletal preparation is impossible or undesirable.




Table 37. Fazekas and Kosa's (1978) data: Incremental increase (II) in millimeters and relative increase in percent (RIP) between CHL groups.

CHL group

Humerus II (mm) RIP

Radius II (mm) RIP

Ulna II (mm) RIP

Femur II (mm) RIP

Tibia II (mm) RIP

Fibula I (mm) RIP

210-239 to 240-269 240-269 to 270-299 270-299 to 300-329 300-329 to 330-359 330-359 to 360-389 360-389 to 390-419 390-419 to 420-449 420-449 to 450-479 450-479 to 480-520

6.17 2.23
4.02 3.86
3.40 4.98 3.02 1.93 6.39

23.8 07.0 11.7 10.1 08.1 10.9 06.0 03.6 11.5

4.99 2.11 3.59 2.39 3.01
3.49 2.24 2.47 4.04

23.2 08.0 12.5
07.4 08.7 09.3 05.5 05.7 08.8

5.95
1.60 4.15 3.33 3.37
4.74 2.42 1.79 5.95

24.9 05.4 13.2
09.4 08.7
11.2 05.1 03.6 11.6

6.38 2.69
5.43 3.45 4.17 7.18
4.31 2.43 9.08

24.1 08.2 15.3
08.4 09.4 14.8 07.7 04.1 14.5

5.49 23.5 3.47 12.0 4.06 12.6 3.57 09.8 3.77 09.4 5.29 12.1 4.13 08.4 1.54 02.9 7.57 13.8

5.60 2.60
4.19 3.38 3.97
4.93 3.44 0.96 7.87

24.8 09.3 13.6 09.7 10.4 11.7 07.3 01.9 15.2




Table 38. Incremental increase (11) in millimeters and relative increase in percent (RIP) between CHL groups.

CHL group

210-239 to 240-269 240-269 to 270-299 270-299 to 300-329 300-329 to 330-359 330-359 to 360-389 360-389 to 390-419 390-419 to 420-449 420-449 to 450-479 450-479 to 480-520

Humerus ll(nun) RIP

5.33 3.32 2.61
4.10 3.66
6.49 1.18 3.71
5.74

19.1 10.0 07.2 10.5 08.5 13.8
02.2 06.8 09.8

Radius 11I(mm) RIP

3.43 3.59 1.77 3.21 3.50 5.56 1.26 1.80
4.62

Ulna
ll(mm) RIP

14.6 13.3 05.8 09.9 09.9
14.3 02.8 03.9 06.7

4.14 3.64 2.55 3.82 3.60 6.63
1.12 2.54 5.33

15.9
12.1 07.5 10.5 09.0 15.2
02.2 04.9 09.9

Femur 11 (mm) RIP
6.78 23.7 3.62 10.2 3.71 09.5 4.69 11.0 4.02 08.5 9.77 19.0 4.14 06.8 -0.66 -00.1
9.53 14.7

Tibia Hl(mm) RIP

4.28 4.01 2.14 5.05 3.77 10.17 0.11
-1.33
9.60

17.3 13.8 06.5
14.3 09.4 23.1
0.20
-2.4
18.1

Fibula
1 (nmm) RIP

3.70
3.40 1.69
4.85 3.70
9.49
-0.21
-1.25
8.63

15.2
12.1 05.4 14.6 09.7 22.7
-0.4
-2.5
17.3




Table 39. Long bone diaphyseal lengths (mm) for Fazeka and Kosa (1978): Mean length and standard deviations from the mean (S.D.), after Armelagos, et al. (1972).

CHL group

210-239
240-269 270-299 300-329 330-359 360-389
390-419 420-449 450-479 480-520

Humerus Mean S.D.

25.91 32.08
34.31 38.33
42.19 45.59 50.56 53.59 55.52 61.91

1.43 1.70 1.38 2.81 2.57
2.34 1.84 1.97 2.58 3.07

Radius Mean S.D.

21.51 26.50 28.61 32.20
34.59 37.60
41.09 43.33 45.80 49.84

1.21 1.61 1.92 2.05 1.91 1.72 1.88 1.29
2.14 1.99

Ulna
Mean S.D.

23.91 29.86
31.46 35.61 38.95
42.32 47.06 49.49 51.28 57.23

1.59 1.89 1.39
2.45 1.96
2.20 2.55 1.65 2.55
2.10

Femur
Mean S.D.

26.48 32.86 35.55
40.98 44.43 48.60 55.78 60.09 62.52 71.60

1.55 1.99
2.22 2.99 3.17 2.32 1.92 2.92 2.95
4.06

Tibia
Mean S.D.

23.37 28.86 32.33 36.39 39.96 43.73
49.03 53.16
54.70 62.27

1.91 2.16 3.13 2.87
2.10 1.89
2.42 2.30 2.96 3.29

Fibula Mean S.D.

22.54 28.14 30.74 34.93 38.31
42.28 47.21 50.66 51.62
59.46

1.83 2.28 2.36 2.81 2.29
2.22 2.73 1.88 2.77 2.75




70.00 60.00 50.00
S40.00
30.00 *Warren
E 20.00 - - - Fazekas and
. Koza
0.00
-- q~ q O ON O L C C U
0.00 1 1 1 1 1 1 1 1 1 11I I
Gestational age in lunar months Figure 30. Comparison of cross-sectional humeral growth curves of the current study and the Fazekas and Kosa (1978) sample.

60.00 50.00 40.00 30.00
. 20.00
10.00
0.00

Lf O M* %10 C C U) C0 Lfl

Gestational age in lunar months Figure 31. Comparison of cross-sectional radial growth curves of the current study and the Fazekas and Kosa (1978) sample.




60.00
50.00 S40.00 o 30.00
20.00 10.00 0.00

SWarren
------- Fazekas and
Koza

q L Lfq
Gestational age in lunar months

Figure 32. Comparison of cross-sectional ulnar growth curves of the current study and the Fazekas and Kosa (1978) sample.
80.00
70.00
60.00
50.00
S40.00
3 ..-* Wrre
30.00 Warren
20.. -------- Fazekas and
20.00
Koza
10.00
0.00 II I l l I l lII
q Ltq qt LO q t..- q ~ O
Gestational age in lunar months
Figure 33. Comparison of cross-sectional femoral growth curves of the current study and the Fazekas and Kosa (1978) sample.




70.00
60.00
6 50.00 S40.00
30.00 Warren
S20.00 -- Fazekas and
10.00 Koza
0.00
t" Lf1 Lf) 0. 011 O ON C)
Gestational age in lunar months Figure 34. Comparison of cross-sectional tibial growth curves of the current study and he Fazekas and Kosa (1978) sample.
70.00
60.00
50.00 .-.
40.00 30.00
Warren
a 20.00
---------- Fazekas and
10.00 Koza
0.00Il0 III
Gestational age in lunar months figure 35. Comparison of cross-sectional fibular growth curves of the current study and te Fazekas and Kosa (1978) sample.




Table 40. Ranges for upper limb lengths by CHL group and "equivalent" gestational age. All measurements are in millimeters.
CHL Equivalent Humerus n Radius n Ulna n
group Gestational
Age in lunar
months
210-239 < 5 24.0-31.0 10 19.5-26.0 9 22.5-29.0 9
240-269 5.0-5.5 26.0-47.0 14 21.0-37.0 14 23.5-42.5 14
270-299 5.5-6.0 29.0-42.0 35 24.0-39.5 33 28.0-39.5 34
300-329 6.0-6.5 31.5-47.0 38 25.5-38.5 38 29.5-44.0 38
330-359 6.5-7.0 37.0-49.0 38 31.0-40.0 37 34.5-45.0 37
360-389 7.0-7.5 42.0-54.0 26 34.0-45.0 26 39.5-51.0 26
390-419 7.5-8.5 46.0-69.0 28 38.0-55.0 26 43.0-67.0 25
420-449 8.5-9.0 52.0-61.0 11 42.0-50.0 11 48.0-57.0 11
450-479 9.0-9.5 44.0-69.0 16 36.0-58.0 13 42.0-67.0 13
480-520 9.5-10 55.0-78.5 24 45.5-55.5 19 54.0-63.5 19

Table 41. Ranges for lower limb lengths All measurements are in millimeters.

by CHL group and "equivalent" gestational age.

CHL Equivalent Femur n Tibia n Fibula n
group Gestational
Age in lunar
months
210-239 < 5 24.5-33.0 10 20.5-29.0 10 20.5-28.0 8
240-269 5.0-5.5 28.0-51.0 13 24.5-35.5 12 23.5-33.5 12
270-299 5.5-6.0 29.0-46.0 35 24.5-38.5 28 23.5-37.0 23
300-329 6.0-6.5 32.0-56.0 36 29.0-42.5 28 27.0-40.0 23
330-359 6.5-7.0 39.0-55.0 40 35.0-47.0 27 33.0-44.5 25
360-389 7.0-7.5 47.0-57.0 24 40.0-49.5 11 39.0-47.0 9
390-419 7.5-8.5 52.0-75.5 16 47.0-63.0 9 47.0-60.0 7
420-449 8.5-9.0 55.0-73.0 6 48.0-59.0 3 34.0-54.5 3
450-479 9.0-9.5 49.0-73.0 6 42.0-58.0 4 39.0-55.0 4
480-520 9.5-10 62.0-81.0 5 55.0-69.0 5 51.0-64.5 4

I.I...... ,, ZE

V:I!




71
Tables 40-41 (above) again present the range of measurement of the long bones for each CHL group and its equivalent gestational age. Gestational age may be determined by using the regression formulae in Chapter 3 to estimate CHL, then using prior studies to arrive at a gestational age based on CHL. However, the formulae are only valid for the reference population from which they were derived, and should be applied with caution in forensic or bioarcheological contexts dealing with other populations.




CONCLUSIONS AND SUMMARY

Measurements of the six long bones of the extremities were taken from a series of
fetal radiographs. Additional data was collected from associated autopsy protocols. I divided the sample arbitrarily into groups of similar CHL, and thus, similar developmental age in order to establish the relative growth and proportionality of the limb bones.
All long bone lengths correlated significantly with CHL (r2 > 0.8375; p < 0.05). This relationship is consistent with the relationship of adult long bones to stature (Genoves, 1967; Trotter and Gleser, 1958). I provide least-squares linear regression formulae for predicting CHL from radiographic bone lengths. The relative linear growth of the long bones was also consistent with other published data (K6sa,1989).
Distance curves show that the femora elongate relative to the humeri during
development as a result of greater incremental increase and relative increase in percent over the entire fetal period. The high intermembral index is a result of the relatively short lower limbs of the fetus. The intermembral index decreases for successive CHL groups. A decreasing intermembral index during ontogeny has occurred as an adaptation for bipedal locomotion and is a unique feature of hominid evolution.
The growth velocities for this study are consistent with postnatal studies through the natal period up to 6 months. Growth then accelerates during the period from 6 months to 1 year (Maresh, 1955).
Previous fetal growth studies have failed to adequately address the impact of
pathology on linear growth and proportionality. Extensive autopsy data have enabled me to divide my sample into "pathological" and "normal" sets for comparison. My data




show that long bone proportions and relative growth are not significantly affected by prenatal pathology. The lack of effect of pathology on long bone length is consistent with other studies that find girth but not length to be affected by congenital conditions (Richards and Ant6n, 1991). The proportional relationship between long bone length and CHL is stable and predictable. Therefore, fetal abortus may serve as a model for normal linear growth and proportionality.
Bagnall et al. (1978) found no significant difference in long bone length or limb proportions between males and females of 8 to 24 weeks gestational age. My data confirms that the same is true throughout the fetal period. Sex cannot be determined by proportional indices.
Race differences in fetal long bone proportions are found to mirror those in adults. Self-identified blacks had significantly higher brachial and crural indices than whites, although there is considerable overlap in ranges. The wide overlap of group values demonstrates the range of variability characteristic of "racial" traits. Assigning ancestral group or affinity to a fetus based on brachial and crural indices is tenuous at best.
In order to establish the gestational age of a fetus based on long bone length, one must first accept the validity of earlier studies that correlate CHL with calculated gestational age based on last normal menstrual period (LNMP), isolated coitus, stage of organ development, or backward extrapolation from the birth date. The ability of clinicians to accurately predict, in most cases, the expected birth date within 1 week is fairly convincing, but errors of several weeks do occur in individual cases. Previous studies have shown that actual linear measurements closely correspond with predicted values via Haase's Rule. Dry bone measurements based on the data of Fazekas and K6sa (1978) have traditionally been used by anthropologists working with fetal remains. My data corresponds closely with that of Fazekas and K6sa. This shows that (1) relative linear growth is similar between the two samples, and (2) radiographic measurements of




74
fetal long bone length are accurate and the radiographic method may be used when skeletal preparation is impossible or undesirable.




APPENDIX A
INSTITUTIONAL REVIEW BOARD EXEMPT STATUS




UNIVERSITY OF
9 FLORIDA
Health Center Institutional Review Board PO Box 100173
Gainesville, FL 32610-0173
M E MOR AN D UM (352) 846-1494
Fax (352) 846-1497
DATE: Januai-y 29, 1997
TO: Michael W, Warren
2927 NW 10th Place Gainesvill FL 5
FROM: R. Peter lafrate, PharmD
Chair, IRE
SUBJ: EXEMPTION (IRB# 033-97) entitled: Linear Growth and Proportionality in Fetuses and
S.il.l borns; An American Ccrieal_, rary Population
Your exemption request has been reviewed by the Chair of the IRE and was APPROVED on January 27, 1997. Your project has been assigned an IRE number; please refer to this
number in future correspondence. Enclosed is a copy of your exemption request with the IRE
approval stamped on it. If this project changes or otherwise requires IRE review and
approval, you have the explicit responsibility to pursue IRE review at that time. In
approximately one year you will be contacted by the IRE office and asked to provide some follow-up information (e.g. whether your project is still active; whether any changes have been made since you submitted your initial request).
Your exemption was granted under the following checked category of exempt research using human subjects.
[1 #1 Commonly accepted educational settings involving normal educational practices.
[] #2 Educational tests (cognitive, diagnostic, aptitude, achievement), survey or interviews or observing public behavior.
(1 #3 Research involving the use of educational tests, survey or interview procedures, or observation of public behavior that is not exempt-under 2 above, if -the subjects are public officials or candidates for public office or a federal statute requires that the confidentiality will be maintained throughout the research and thereafter.
#4 Collection or study of existing data, documents, records, pathological or diagnostic specimens, if these
arc ~ va -ulc, tm'o -- ~ eoddb he P! in such a manner that Sllj~z no
/Z cr aepulcl -be rifte normation is recre -identified, directly or through identifiers linked to the subjects.
1] #5 Research and demonstration projects for public benefit.
[] #6 Taste and food quality evaluation and consumer acceptance studies.
Thank you for keeping the IRE informed about your research, thereby allowing us to keep accurate files. If the IRE staff can be of any further assistance, please feel free to
call.
encl
cc Division of Sponsored Research IRB file

Equal Opportunity / Affirsative Action Institution




APPENDIX B
RAW DATA RECORDED FROM AUTOPSY PROTOCOLS




Series # CHL Weight Sex Race Mother's age Cause of death/comments

285.0 325.0
445.0
485.0 360.0 310.0 315.0 270.0 300.0 265.0 230.0
340.0 360.0 235.0

400 425 475 1425 475* 3200 893* 650 750
400 400 168 235 822 850 1575 313

Premature rupture of membranes; sepsis; no congenital malformations Hydrocephalus; induced abortion No congenital malformations; anoxic stess unknown etiology No congenital malformations; fetal death in utero of unknown cause Amniotic band syndrome; no congenital malformations Anencephaly
Placental infarctions
Twin "A" unknown cause of death in utero Twin "B" identical twin to above; no congenital malformations Chorioamnionitis; intraventricular hemorrhage No congenital malformations; intraventricular hemorrhage Abruptio placenta; hyaline membrane disease; immaturity No congenital malformations; fetal death in utero; unknown cause Immature lungs; intraventricular hemorrhage Sepsis; patent ductus arteriosis Premature rupture of membranes; oligohydraminos Ventral wall defect of abdomen; elective abortion Anencephaly
Amniotic band syndrome with multiple deformities Twin congenital syphilis

.. 686
1210




Series # CHL Weight Sex Race Mother's age Cause of death/comments

370.0 750
260.0 275
... 353 300.0 800
315.0 805
205.0 287
325.0 950
... 1050^
... 250
... 484* 335.0 825
... 1688 300.0 525
345.0 811*
260.0 862*
260.0 440
250.0 220
... 150
... 319*
... 675

No congenital malformations; history of maternal polyhydraminos No congenital malformations; marginal placental abruption No congenital malformations; placental infarction Twin "A" abruptio placenta; no congenital malformations Twin "B" abruptio placenta; no congenital malformations Turner's syndrome; nuchal lymphangioma No congenital malformations; acute chorioamnionitis Hydrops fetalis; severe maceration Placental infarction; pulmonary hypoplasia Placenta previa; placental infarction; no congenital malformations Maternal eclampsia; intraventricular hemorrhage No congenital malformations; placental infarction No congenital malformations; chorioamnionitis; pneumonia No congenital malformations; marginal cord insertion Cystic hygroma; hydrops fetalis; induced AB No congenital malformations; placental insufficiency; unknown Caudal regression syndrome; sirenomelia Fetal death in utero of unknown cause; no congenital malformations Oligohydraminos; intrauterine growth retardation Maternal cocaine abuse; fetus positive for cocaine metabolites




Series # CHL Weight Sex Race Mother's age Cause of death/comments

27.0 650
... 500 ... 700
640
... 620* ... 2160
... 780 ... 250 40.0 1060*

318.0 280.0
475.0
...
350.0
410.0 340.0

m b m w f w m w f w m w f w f w f b

580
455 685
2040* 400 501 720 1760
480* 1000

3

Abruptio placenta; no congenital malformations Prematurity; hyaline membrane disease; 30% placental abruption Downs syndrome Trisomy 21; induced abortion Abruptio placenta; premature rupture of membranes No congenital malformations; fetal death in utero unknown cause Congenital cytomegalovirus; hydrops; hydrocephaly Anencephaly
5 p-chromosome abnorm.; no gross abnormalities detected Pulmonary hypoplasia; assymetric growth retardation Anencephaly
Chorioamnionitis; pneumonia; no congenital abnormalities Sacral meningocele 2nd to spina bifida; induced AB "Intrauterine growth retardation"; oligohydraminos No congenital malformations; maternal diabetes No congenital malformations; necrotizing chorioamnionitis Birth trauma from footling breech; non-viable immaturity Materal pre-eclampsia; bronchial dysplasia Abruptio placenta;; no congenital malformations Placental insufficiency; growth deficiency No congenital malformations; history of maternal cocaine abuse

4




Series # CHL Weight Sex Race Mother's age Cause of death/comments

... 830 ... 1110 ... 860 270.0 380*
320.0 720
... 310 345.0 700
230.0 225
370.0 980
480.0 3240
310.0 530
320.0 610
400*
290.0 650
290.0 690
335.0 650
... 1900
... 3030
2600 250

Premature rupture of membranes; non-viable immaturity Prematurity; intraventricular hemorrhage No congenital malformations; non-viable immaturity Immature; macerated; maternal diabetes; hydrocephaly Abruptio placenta; no congenital malformations Maternal hx of habitual ABs; marginal abruption of placenta Twin immaturity; interstitial emphysema Chronic villitis; no congenital malformations; immaturity Twin immature; interstitial emphysema Renal agenesis; hypoplastic lungs; oligohydraminos deformations No congenital malformations; possible placental infarction No congenital malformations; premature labor No congenital malformations; immaturity; macerated Twin immaturity; premature rupture of membranes Twin intraventricular hemorrhage No congenital malformations; hemolytic disease; chorioamnionitis Hydrocephalus
Maternal pulmonary embolism, treated with anticoagulant Cystic kidney disease; anuria; pulmonary hypoplasia Asplenia; no other abnormalities; intraventricular hemorrhage




Series # CHL Weight Sex Race Mother's age Cause of death/comments

375.0 1050
... 720
... 3175
... 660 370.0 830*
250.0 200*
365.0 1050
... 2990
245.0 280
252.0 275*
270.0 550
490.0 2520
340.0 1000

280.0 290.0 340.0 414.0 280.0

No congenital malformations; prematurity; placental infarction Premature rupture of membranes; amniotic band syndrome Anencephalic
No congenital malformations; intraventricular hemorrhage Placental infarction; fetal death in utero Twin no congenital malformations; possible placental infarction Abruptio placenta; cleft palate and other malformations Hydrops fetalis; hypoplastic lungs Non-viable immaturity; unknown cause of death Twin placental thrombosis; monochorionic/diamniotic placenta Anencephaly; pregnancy terminated Hydrocephalus; multiple anomalies; maternal use of antibiotics Moderate hydrocephaly; pregnancy terminated Trisomy 18; multiple congenital anomalies No congenital malformations; fetal death in utero Intraventricular hemorrhage; chorioamnionitis; marginal abruption Massive abruptio placenta Premature rupture of membranes; pulmonary hypoplasia Sepsis; meningitis; pneumonia Mosaic Downs syndrome; elective termination of pregnancy

2270* 400 580
840 2500
490




Series # CHL Weight Sex Race Mother's age Cause of death/comments

275.0 845*
365.0 790*
415.0 1950
... 1430 448.0 3625
405.0 1340
... 570 465.0 2050*
... 740 ... 820 ... 1800
... 3420* 355.0 960
350.0 940*
235.0 310*
275.0 400*
... 1100 315.0 530
290.0 560
550.0 2580

m b ... No congenital malformations; unknown cause of spontaneous AB
m w 29 No congenital malformations; placental infarction
m w 29 Potter's syndrome
m b 20 Abruptio placenta; fetal death in utero; no congenital malformations
f w 22 Severe hydrops fetalis
m w 19 No congenital malformations; premature ruptured membranes; sepsis
f b 19 Maternal hypertension; pre-eclampsia; no congenital malformations
m b 18 No congenital malformations; unknown cause of death
f w 18 Twin "A" Polyhydraminos; prematurity
f w 18 Twin "B" Polyhydraminos; prematurity
m b 22 Maternal schizophrenia; no congenital malformations
m w 22 Maternal hypertension; fetal death in utero; macerated
f w 25 Sepsis; intraventricular hemorrhage
m w 40 Advanced maternal age; marginal placenta; maternal hypertension
m w 30 Twin placental thrombosis; monochorionic/diamniotic placenta
f b ... Abdominal pregnancy; maternal syphilis
f w ... Trisomy 18; hyaline membrane disease; congenital heart disease
f w 23 Bronchopneumonia; immaturity; ventricular hemorrhage
m b 23 Multiple anomalies suggestive of chromosome abnormality
f w 27 Potter's syndrome; polyhydraminos




Series # CHL Weight Sex Race Mother's age Cause of death/comments
259 415.0 1030* m b ... Hemorrhagic pulmonary edema; no congenital malformations
260 335.0 765 f w 16 No congenital malformations; unknown cause of death
261 320.0 510 m w 23 No congenital malformations; placenta previa
262 470.0 2000* m w ... Renal malformations; moderate maceration
263 450.0 2350 f w 31 Rectocloacal malformation; pulmonary hypoplasia
264 365.0 900 m b 19 Twin placenta previa; intraventricular hemorrhage
265 410.0 1350 m b 23 Premature rupture of membranes; chorioamnionitis
266 360.0 925 f w 29 Hydrocephalus
267 470.0 2565 f b 25 Congenital heart disease
268 330.0 550 m w 30 Placental infarction
269 350.0 920 m b 28 Patent ductus arteriosus and foramen ovale
270 ... 1100 f w 19 Hyaline membrane disease; interstitial edema
271 460.0 1675 f b 20 No congenital malformations; unknown cause of death
272 380.0 1090 m w 28 No congenital malformations; abruptio placenta
273 520.0 3000 m w 15 Hemorrhagic pulmonary edema
274 340.0 670 f b 24 Hypotensive event in mother with chronic hypertension
275 225.0 220 m b 22 No congenital malformations; unknown cause of death
276 ... 2160 f w 17 Anencephaly
277 490.0 3100 m w 21 Congenital heart disease
278 425.0 1720 m b 18 Hyaline membrane disease; thyroid hyperplasia




Series # CHL Weight Sex Race Mother's age Cause of death/comments
239 285.0 575 f w 19 Omphalocele with herniated internal organs
240 415.0 2126 m w ... Potter's facies; multiple congenital malformations
241 280.0 310* f w 25 No congenital abnormalities; non-viable immaturity
242 350.0 840 f w 22 Hydrocephalus
243 290.0 540 m w 22 No congenital malformations; hyaline membrane disease
244 285.0 450 m b 31 Premature rupture of membranes; chorioamnionitis
245 335.0 700 m b 28 Acute chorioamnionitis 2nd to premature rupture of membranes
246 415.0 2610 m w 34 Maternal polyhydraminos; generalized edema
247 200.0 200 m w 19 Cystic hygroma; possible Turner's syndrome
248 260.0 400 m b 33 Potter's syndrome; agenesis of kidneys
249 325.0 650 f b 18 Placenta previa; prematurity; hyaline membrane disease
250 ... 975 f b 17 Abruptio placenta; perinatal asphyxia
251 225.0 200 f w 19 Premature rupture of membranes; bronchopneumonia
252 285.0 400 f b 19 Immature placenta; chorioamnionitis
253 265.0 240 m b 16 No congenital malformations; placental fibrosis
254 520.0 3225 f b 24 No congenital malformations; unknown cause of death
255 380.0 985 f w 21 No congenital malformations; unknown cause of death
256 260.0 210 f w ... Hydrocephalus; cleft palate
257 295.0 488 f b 20 No congenital malformations; acute chorioamnionitis
258 285.0 470 m b 24 No congenital malformations; unknown cause of death




Series # CHL Weight Sex Race Mother's age Cause of death/comments

520.0 2675*
... 840* 235.0 275
310.0 720
350.0 815
370.0 1050* 360.0 1220
330.0 640*
430.0 2310
330.0 680*
305.0 510** 310.0 570** 295.0 340*
... 2200
... 1460 335.0 850
420.0 1255
340.0 640
... 1800 360.0 1000

f w w m f b m w f w m b f w m w f w f w f w f w m b f b f w m w m w m w f w
m w

Placental infarction
Maternal DIC; diffuse petechial hemorrhages Maternal pre-eclampsia; no congenital malformations Myocardial fibrosis; moderate hydrops fetalis; syndactyly Twin dissecting hematoma in jejunum Unknown cause of death; no congenital malformations Oligohydraminos; Potter's facies with pulmonary hypoplasia Twin placenta previa
Multiple skeletal anomalies; congenital heart disease Triplet "C" non-viable immaturity Triplet "B" twin to twin transfusion w/ Triplet "A" (donor) Triplet "A" twin to twin transfusion w/ Triplet "B" (recipient) Unknown cause of death; fetal death in utero, macerated Anencephaly; selective termination of pregnancy Potter's facies; oligohydraminos; perinatal asphyxia Hydrocephalus
Abruptio placenta; no congenital malformations Maternal diabetes; agenesis (phocomelia) of upper limbs Potter's facies; multiple congenital malformations Placenta previa; peritonitis




Full Text

PAGE 1

PRENATAL LIMB GROWTH IN HUMANS: LINEAR GROWTH, ALLOMETRY, LOCOMOTION, AND SKELETAL AGE By MICHAEL W. WARREN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNWERSTTY OF FLORIDA 1997

PAGE 2

Copyright 1997 by Michael W. Warren

PAGE 3

This work is dedicated to two men that have served as positive influences in my life, both of whom passed away during the preparation of this manuscript. My father. Bill Warren, was a caring, hard-working man. He is, in part, responsible for my inquisitive nature. Although he was an uneducated man, he had the ability to I master skills, answer questions, and find truths that were his own. I hope he instilled in me a small measure of the moral character and sense of fairness that I always admired in him. He was always proud of me, no matter the extent of my accomplishments, and I expect that he would be especially proud to let the neighbors know about his son the anthropologist. Dr. William R. Maples was, at first, larger than life. Over time he became my mentor, and finally, a friend. The simple fact that he had confidence in my ability has encouraged me to aim high. No one with such a role-model could fail to succeed. Words cannot be foimd that express my sorrow at their passing. I was fortimate to have known them, and I dedicate this work to their memory.

PAGE 4

ACKNOWLEDGMENTS I gratefully acknowledge Dr. William Donnelly for access and permission to use the Radiographic Collection of the Department of Pathology, College of Medicine, University of Florida. His assistance in obtaining proper authority with the Institutional Review Board was the first step in this research project. I thank three members of my doctoral committee— Dr. Leslie Lieberman, Dr. Lynn Larkin, and Dr. Sue Boinski for their input, critique and encouragement during my graduate course work and the writing of this dissertation. I also thank Dr. William Leonard for early direction in finding the appropriate growth and development literature. Kendra Smith, Heather Walsh-Haney, Shuala Martin, Phoebe Stubblefield, Cheryl Katzmarzyk and other students, old and new, of the C.A. Pound Human Identification Laboratory are appreciated for the helpful, non-competitive, leanving atmosphere that exists in the lab. Thanks are also in order for Karen Jones and Dian Leahy, both of whose guidance and help in navigating the academic waters were invaluable. Special appreciation is due to Dr. Susan Anton, chair of my doctoral committee and my lead dissertation advisor. She provided both professional and personal support above and beyond the call of duty during some difficult times. Special thanks are also due Dr. Anthony Falsetti, whose statistical expertise contributed greatly to this manuscript. E>rs. Anton and Falsetti and the Department of Anthropology provided much-needed financial support in the form of teaching assistantships during my doctoral studies. I have benefitted greatly from sharing their classrooms. Finally, I thank my wife, Melinda, and my son, Zachary, for their help and understanding. They have sacrificed many of their ov^m dreams so that I might pursue IV

PAGE 5

my own. My intellectual curiosity and dreams of ivory towers led me to abandon a weU-paying career and move my family from their home. They have not only followed without complaint, but taken the lead in following new dreams of their own. This research was fimded by a Sigma Xi Grant-in-aid and a Forensic Sciences Foundation Lucas Research Grant.

PAGE 6

TABLE OF CONTENTS ACKNOWLEDGMENTS ^^ LIST OF TABLES • ^"^ LIST OF FIGURES "^^ ABSTRACT '^^ CHAPTERS 1 INTRODUCTION ^ Review of the Literature ^ Statement of Purpose 2 MATERIALS AND METHODS ^ 7 Radiographic Measurements Terminology „ Sample Profile t^ Maternal Profile Measuring Relative Long Bone Growth 3 CORRELATION BETWEEN CROWN-HEEL LENGTH AND LONG BONE DIAPHYSEAL LENGTH 16 Mean Long Bone Lengths ^ Correlation Between CHL and Length 4 RELATIVE GROWTH AND LIMB PROPORTIONALITY 24 Relative Rates of Growth ^4 Influence of Pathology on Linear Growth and Limb Proportions // Sex Differences in Linear Growth and Limb Proportionality 28 Influence of Self-Identified Racial Category on Linear Growth and Limb Proportions ^^ 5 DETERMINATION OF GESTATIONAL AGE 58 6 CONCLUSIONS AND SUMMARY 72 VI

PAGE 7

APPENDICES A INSTITUTIONAL REVIEW BOARD EXEMPT STATUS 75 B RAW DATA RECORDED FROM AUTOPSY PROTOCOLS 11 C AUTOPSY PROTOCOL (FACSIMILE) 98 REFERENCES CITED 99 BIOGRAPHICAL SKETCH 106 vn

PAGE 8

LIST OF TABLES Table page 1. Frequency of pathological conditions determined at autopsy 13 2. Maternal age profile 14 3. Distribution of all cases with recorded CHL by CHL group and approximate gestational age 16 4 Mean humerus length of all fetuses with recorded CHL by CHL group (n=240) 17 5 Mean radius length of all fetuses with recorded CHL by CHL group (n=226) 17 6 Mean ulna length of all fetuses with recorded CHL by CHL group (N=226) 18 7. Mean femur length of all fetuses with recorded CHL by CHL group (N-191) 18 8 Mean tibia length of all fetuses with recorded CHL by CHL group (N=137) 19 9 Mean fibula length of all fetuses with recorded CHL by CHL group (N=118) 19 10. Correlation matrix of r-values for CHL versus long bone diaphyseal length 20 11. Conversion from lunar months to weeks 21 12 Mean humerus lengths for gestational age in weeks reported by Warren, Russell et al., and Fazekas and Koza 11 1 3 Incremental increase (H) in millimeters and relative increase in percent (RIP) between CHL groups Humerus, radius and ulna 26 14. Incremental increase (H) in millimeters and relative increase in percent (RIP) between CHL groups Femur, tibia and fibula 26 1 5 Comparison of mean humeral length between all cases, "normal" cases, and "pathological" cases 29 vm

PAGE 9

IX Table ^ 1 6 Comparison of mean radial length between all cases, "normal" cases, and "pathological" cases ^^ 17 Comparison of mean ulnar length between all cases, "normal" cases, and "pathological" cases ^^ 1 8 Comparison of mean femoral length between all cases, "normal" cases, and "pathological" cases ^^ 1 9 Comparison of mean tibial length between all cases, "normal" cases, and "pathological" cases ^^. 20 Comparison of mean fibular length between aU cases, "normal" cases, and "pathological" cases ^^ 21. Descriptive statistics for Male sample (upper limb) by CHL group 39 22. Descriptive statistics for Male sample (lower limb) by CHL group 38 23. Descriptive statistics for Female sample (upper limb) by CHL group 37 24. Descriptive statistics for Female sample (lower limb) by CHL group 36 25. Descriptive statistics for "White" sample (upper limb) by CHL group 47 26. Descriptive statistics for "White" sample (lower limb) by CHL group 46 27. Descriptive statistics for "Black" sample (upper limb) by CHL group 45 28. Descriptive statistics for "Black" sample (lower limb) by CHL group 44 48 29. Proportional indices during ontogeny 30. Limb proportion indices for the entire fetal sample and previously published values for adults 4^ 31. Bonferroni t-test for significant differences in proportional indices 49 32. Ratio of long bone lengths to femur and tibia length for white sample 50 33. Ratio of long bone lengths to femur and tibia length for black sample 51 34 Comparison of values of fetal body length given by Scammon and Calkins 35. Months of gestation and equivalent CHL 62

PAGE 10

Table page 36. Growth in length and increase in weight with respect to fetal age (in lunar months) 63 37. Fazekas and Kosa's data: Incremental increase (H) in millimeters and relative increase in percent (RIP) between CHL groups 64 38. Incremental increase (II) in millimeters and relative increase in percent (RIP) between CHL groups 65 39. Long bone diaphyseal lengths (mm) for Fazekas and Kosa (1978) 66 40. Ranges for upper limb lengths by CHL group and "equivalent" gestational age • 70 41 Ranges for lower limb lengths by CHL group and "equivalent" gestational age 70

PAGE 11

LIST OF FIGURES Figure page 1. A typical fetal radiograph 8 2 Measurements were taken parallel to the long axis of the bone with a transparent metric scale 9 3. Comparison of Warren, Fazekas and Kosa, and Russell et al. data 23 4. Combined distance curves for all bones showing relative growth trajectories 25 5. Distance curve for the humerus 26 6 Comparison of normal and pathological subsets for humerus length 32 7. Comparison of normal and pathological subsets for radius length 32 8 Comparison of normal and pathological subsets for ulna length 33 9 Comparison of normal and pathological subsets for femur length 33 10. Comparison of normal and pathological subsets for tibia length 34 1 1 Comparison of normal and pathological subsets for fibula length 34 1 2 Comparison of female and male means for humerus length by CHL group 35 13. Comparison of female and male means for radius length by CHL group 40 14. Comparison of female and male means for ulna length by CHL group .... 40 15 Comparison of female and male means for femur length by CHL group 41 16. Comparison of female and male means for tibia length by CHL group 42 17. Comparison of female and male means for fibula length by CHL group 42 XI

PAGE 12

Figure page 18. Comparison of "Black" and "White" means for humerus length by CHL group 51 19. Comparison of "Black" and "White" means for radius length by CHL group 52 20. Comparison of "Black" and "White" means for ulna length by CHL group 52 21. Comparison of "Black" and "White" means for femur length by CHL group 53 22. Comparison of "Black" and "White" means for tibia length by CHL group 53 23. Comparison of "Black" and "White" means for fibula length by CHL group 54 24. Bivariate plot for CHL versus humeral length in raw data space 54 25. Bivariate plot for CHL versus radial length in raw data space 55 26. Bivariate plot for CHL versus ulnar length in raw data space 55 27. Bivariate plot for CHL versus femoral length in raw data space 56 28. Bivariate plot for CHL versus tibial length in raw data space 56 29. Bivariate plot for CHL versus radial length in raw data space 57 30. Comparison of cross-sectional humeral growth curves of the current study and Fazekas and Kosa (1978) sample 67 31 Comparison of cross-sectional radial growth curves of the current study and Fazekas and Kosa (1978) sample 67 32. Comparison of cross-sectional ulnar growth curves of the current study and Fazekas and Kosa (1978) sample 68 33. Comparison of cross-sectional femoral growth curves of the current study and Fazekas and Kosa (1978) sample 68 34. Comparison of cross-sectional tibial growth curves of the current study and Fazekas and Kosa (1978) sample 69 35. Comparison of cross-sectional fibular growth curves of the current study and Fazekas and Kosa (1978) sample 69 xu

PAGE 13

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy PRENATAL LIMB GROWTH IN HUMANS: LINEAR GROWTH, ALLOMETRY, LOCOMOTION, AND SKELETAL AGE By Michael W. Warren December 1997 Chair: Susan C. Anton Major Department: Anthropology All human groups correspond to a general species-specific pattern of growth. Modifications in the rate and timing of growth events provide a direct measure of developmental response to the environment. Comparison of the linear growth of the long bones has been shown to be an effective way of demonstrating genetic and secular differences among populations. Most skeletal growth studies, however, have been directed toward the postnatal period. Prenatal studies have generally been clinically oriented and are poorly suited for use by anthropologists working with skeletal populations. This study analyzes the linear growth and proportionality of the long bones for a sample of 398 fetuses based on radiographic measurements of diaphyseal length and recorded crown-heel length. The data are derived from the full-body radiographs of stillborn and spontaneouslyor therapeutically-aborted fetuses delivered between 1976 and 1988. Additional data on crown-heel length, sex, self-identified "race," and pathology are taken from associated autopsy records. xiu

PAGE 14

Growth distance curves are plotted using crown-heel length (CHL) as a normalizing datum to reduce size effects. The sample is arbitrarily divided into 10 groups of crownheel length measurements, roughly corresponding to gestational ages from 4.5 lunar months to term. Descriptive statistics for each long bone are provided and relative and absolute growth of the long bones is examined in terms of relative increase in percent and linear incremental increase. All long bone lengths correlate significantly with CHL (r^ > 0.8375; p < 0.05). Regression formulae are provided that predict CHL from long bone length. Proportional indices show that relative growth is not significantly affected by prenatal pathology. Males and females exhibit similar relative growth and proportions, however, "Blacks" had significantly higher brachial and crural indices than "Whites." The current data corresponds closely with that of Fazekas and Kosa (1978), whose data are derived from dry bone measurements of an eastern European population. This shows that (1) relative linear growth is similar between the two reference populations, and (2) radiographic measurements of fetal long bone length are accurate and the radiographic method may be used when skeletal preparation is impossible or undesirable. xiv

PAGE 15

CHAPTER 1 INTRODUCTION Growth rates in humans follow a "mammalian curve" (Gavan and Swindler, 1966). However, the duration of growth and the timing of growth events is species-specific. During phylogeny the primates have modified the "mammalian" curve by altering the duration of growth of specific body segments to adapt to different environments and modes of locomotion (Gavan and Swindler, 1966). Normalization procedures that reduce size effects show that all human groups correspond to a general species-specific pattern of growth (Harrison et al., 1964; Lovejoy, Russell and Harrison,1990). When modifications in the rate and timing of growth events occur, they can provide a direct measure of an individual's developmental response to environmental circumstances. Comparative populational studies have shown that linear long bone growth is a sensitive indicator of differential response to environmental stressors (Acheson and Hewitt, 1954; Hewitt, Westropp and Acheson, 1955; Eleveth and Tanner, 1976; Armelagos et al., 1972; Hummert and Van Gerven, 1983; Mensforth et al., 1978; Merchant and Ubelaker, 1977; Y'Edynak, 1976). Therefore, comparison of relative long bone growth among populations provides data regarding environmental stress and general population health. Studies of linear long bone growth are also used to assess growth sufficiency in individuals. Growth curves for individuals that are accelerated or retarded relative to established standards may indicate an array of developmental, pathological or nutritional abnormalities.

PAGE 16

Review of the Literature Studies of long bone growth may be placed into 4 categories: (1) general biology and allometry; (2) comparative populational analysis; (3) clinical human growth and development, and (4) determination of developmental age. In general biology, differential growth analysis is used to examine the relationships that exist between growth "events" (Moss, 1954, 1955). Analysis of interspecies differences in mode and tempo of growth contribute to our understanding of the principles of growth and allometry, as well as the ways in which species have modified their growth patterns during phylogeny. (Gavan and Swindler, 1966; Jimgers, 1982). Recent studies comparing various primate taxa with prehistoric skeletal populations provide explanations of fimctional and developmental morphological differences (Simpson et al., 1996; Leigh, 1996). Anthropological studies of growth, like most general biological studies, utilize a populational or comparative perspective. Populational phenomena can be observed from cross-sectional studies of skeletal populations (Lovejoy et al., 1990). Evaluation of the health of skeletal samples necessarily relies on a normative sample of healthy, wellnourished individuals. Therefore, prehistoric samples are often compared with contemporary samples that are known to be well-nourished and relatively disease-free (Jantz and Owsley, 1984; Hummert and Van Gerven, 1983; Merchant and Ubelaker, 1977; Armelagos et al., 1972). Most of the numerous studies on human growth and development involve longitudinal studies of clinical interest. Analysis of fetal weight and length at different stages of development are used to assess growth sufficiency and to detect factors that contribute to growth acceleration or retardation. Many of these studies are published as standards which establish normal percentiles of growth for a given age. Other standards based on appearance of ossification centers and/or bone morphology are used to assess "skeletal

PAGE 17

age." Comparison of skeletal age with chronological age provides an indicator of level of maturity, predicted adult height, and other factors related to healthcare decisions (Tanner et al., 1983; Greulich and Pyle, 1959; Pyle and Hoerr, 1969; Roche, 1988). While these standards are usually used by physicians and other clinicians for growth assessment, they are also used by anthropologists in comparative studies and forensic work (Warren et al., 1997; Jantz and Owsley, 1984; Hummert and Van Gerven, 1983; Merchant and Ubelaker, 1977; Armelagos et al., 1972). In the prenatal period, growth in length of long bones is linear and several standards have been established for both clinical and non-clinical applications (Halac et al., 1982; Fazekas and Kosa, 1978; Brenner et al., 1976; Birkbeck et al., 1975a, 1975b, 1976; Mehta and Singh, 1972; Russell et al., 1972; Lubchenco et al., 1966). Most studies of fetal growth have been conducted on fetuses aborted during the first 26 weeks of pregnancy (Bagnall et al., 1978, '82). Greater sample size is possible for this group than fetuses of greater gestational age because of an increased frequency of spontaneous and elective abortions during the first trimester of pregnancy. The close correlation between fetal length (crown-heel length [CHL] or crown-rump length [CRL]) and long bone length to period of growth has been used to determine developmental age in both clirucal and anthropological contexts. Most studies utilize CHL because that measurement has been foimd to contain less interobserver error than CRL (Scammon and Calkins, 1925). The fetal population of interest to anthropologists is generally a different population than that encountered by medical clinicians. Since anthropologists generally examine skeletal material, it is difficult to know if a condition existed that may have affected the linear growth of the fetus. Therefore, for anthropological purposes, it may be undesirable to conduct a study of stillborn or aborted fetuses by discarding those with non-skeletal pathologic conditions, as the majority display an tmder lying abnormality. This study

PAGE 18

examines all specimens without skeletal abnormalities, even in the presence of an underlying soft-tissue pathology. Statement of Purpose Despite quite extensive literature on growth of the fetus, we have yet to establish the extent of populational variation (Ubelaker, 1989). We continue to use standards developed primarily from white, healthy populations in the United States and Europe. Clinical radiography presents an opportimity to gather comparative data from a variety of populations. Although radiography is an invasive procedure, it is used extensively in the postnatal period in cases of suspected trauma, and it should be routinely used in the diagnosis of cause of prenatal abortion and death. This manuscript may serve as a guide for future anthropological research by demonstrating the utility of existing medical data. This study establishes the correlation between radiographic lengths of long bones with CHL and examines the relative and absolute growth of the long bones in terms of differential growth rates in groups and individuals. Chapter 3 looks at the relationship between CHL and long bone diaphyseal length and establishes relative growth rates for each of the long bones. The data are placed into groups of similar CHL that roughly correspond to gestational age as a normalization procedure so they can be compared with the results of other fetal studies. Chapter 4 examines the influence of pathology, sex, and self-identified race on absolute and relative long bone growth and proportionality for the study sample. Do specific pathologies affect linear growth and limb proportions? If not, then is data derived from the study of aborted fetuses representative of normal growth and development? Differences in limb proportionality have been found to exist between groups of different ancestry. Are these differences apparent during the fetal period or do they occur during later development?

PAGE 19

Finally, Chapter 5 explores the possibilities and pitfalls of determining developmental age from linear growth measurements. The results of the current study are compared with those of previous researchers whose data are used in bioarcheological and forensic settings to determine gestational age from crown-heel length.

PAGE 20

CHAPTER 2 MATERIALS AND METHODS This study uses the Radiological Study CoUection and autopsy protocols of the Autopsy Service, Department of Pathology, College of Medicine, University of Florida. This collection is a series of pathology records and radiographs taken of liveborn and stillborn fetuses from spontaneous, therapeutic, and elective abortions delivered at Shands Teaching Hospital in Gainesville, Florida since 1976. The radiographs were irutially used to aid in the determination of cause of death. Their retention and use for anthropological study was granted by William Donnelly, M.D., Director of Autopsy Services. Permission to use confidential medical records was obtained from the Health Center Institutional Review Board of the College of Medicine, which granted exempt status to a research protocol submitted imder the following category: "Collection or study of existing data, documents, records, pathological or diagnostic specimens, if these sources are publicly available or if the information is recorded by the investigator in such a manner that subjects cannot be identified, directly or through identifiers linked to the subjects" (Institutional Review Board 01, Protocol #033-97; see Appendix A). The autopsy protocols record a limited maternal history, self-identified racial category of the mother, pathological conditions detected during dissection, and estimated gestational age. The gestational age of the fetuses was based on weight, CHL, last normal meristrual period or stage of organ development. Data were taken from 398 cases that met the following criteria: (1) both autopsy protocols and radiographs existed; (2) no conflicting data were recorded in the autopsy records; (3) at least one

PAGE 21

measurable long bone diaphysis was present; and (4) no congenital skeletal abnormalities were detected by the author. Radio graphic Measurements Three factors determine the accuracy of measurements taken from radiographs. The margins of the bone must be distinct, the bone must be parallel with the film plane, and the degree of x-ray magnification must be considered. The radiographs were taken in a cabinet-type Hewlett-Packard Faxitron using unscreened Kodak EM-1 diagnostic mammography film. This type of film produces extremely sharp images when used in imscreened cassettes (Figure 1). Longer exposure times are required (the purpose of the screen is to limit exposure times for living subjects), but this does not, of course, impose constraints for radiography of non-living subjects. X-ray intensity and exposure times can be adjusted for maximimi image quality, as is the case with radiographs used in forensic anthropology case studies. Examination of the radiographs under a lupe demonstrates clear, distinct margins. Each radiograph was assessed for proper positioning of the fetus. The fetuses were placed in a supine position with the extremities in anatomical position. The radiographs were not marked to indicate the left and right side of the specimen. Only long bones positioned parallel to the film plane were measured. Measurements were taken parallel to the long axis of the bone with a transparent metric scale to within 0.5 millimeters and the maximum length was recorded (Figure 2). When possible, measurements were taken of bones from both sides of the body. When the measurements were unequal the longest measurement was recorded. Trial radiographs taken with a Hewlett Packard Faxitron identical to the one that produced the study radiographs show that inclinations of less than 10 produce errors of less than 15.0%. This is consistent with reports from similar studies which state that a 15 angulation does not introduce significant error when

PAGE 22

8 measuring long bones (Martin and Higginbottom, 1971; Owen, 1971). Recent studies have demonstrated greater error when measuring adult pelves due to absolute greater distances and variation between the x-ray source, the subject, and the film plane (Schroeder, et al., 1997). Figure 1: A typical fetal radiograph.

PAGE 23

Error produced by magnification was found to be less than 2.0% when the bone is 1 centimeter from the film plane. Radiographs of fetuses in utero will have a greater inherent magnification error due to the distance of the fetus from the film plane. Radiographic magnification of up to 19.0% has been recorded in studies of adult pelvic Figure 2: Measurements were taken parallel to the long axis of the bone with a transparent metric scale.

PAGE 24

10 bones due to varying lengths from x-ray source to subject to film plane. Although preliminary tests indicate that magnification error in this study is minimal the figures recorded as diaphyseal lengths should be considered"radiographic lengths" and not true anatomical lengths. Terminology The sample represents the products of conception from 398 pregnancies dating from 1976 through 1988. Stillbom abortus is most properly referred to as a "specimen" since the remains are cor\sidered "tissue samples" and certificates of birth are not recorded. The live-bom fetuses and live-bom near-term infants included in this sample are issued certificates of birth and disposition of the body is regulated accordingly. These cases are most properly termed "individuals." However, many of the stillboms were given names and afforded the same postmortem treatment as live-bom infants ~ they are considered "individuals" in this sense, and the timing of the family's loss makes little difference as a practical (or emotional) matter. This dissertation will, with all proper respect, refer to each death as a "case," and each product of conception as a "specimen," a term often applied to adult cadavers in medical school dissection rooms. No philosophical position is implied by the use of this terminology. Sample Profile The data were derived from records of spontaneous abortions, therapeutic /induced abortions, and stillbirths from Shands Medical Center, a tertiary-care center located in GainesvUle, Florida. The hospital is affiliated with J. HiUis Miller Health Center, the teaching arm of the hospital and home of the College of Medicine at the University of Florida. Most of the cases were delivered in Gainesville, however, a small number of the autopsies were referrals from hospitals and clinics in the smaller commimities surrounding the Gainesville area. It may be assumed, for purposes of comparative

PAGE 25

11 analysis, that all cases were delivered (and in the vast majority of cases, underwent development) between sea-level and 50 meters altitude. No socio-economic data were recorded in the autopsy records. The total sample of fetuses (n=398) was identified by sex. Males comprised 57.6% of the sample (n=227), and females comprised 42.4% (n=167). Four cases omitted the sex of the fetus, either because sex was ambiguous due to developmental anomalies, or the autopsy record had conflicting information and the actual sex of the specimen could not be determined by the investigator. In each case, the "race" of the child was based on the "race" of the mother. It is customary for this data to be self-reported when admission records are completed by the mother or her representative. However, in some cases, it has been the personal observation of the author that "race" is often clinician-reported based on perceived phenotype. "Race" was recorded for 387 cases: The category "white" comprised 57.8% (m=230); "black" was 40.7% (n=162); and "other" was recorded 1.3% (n=5). The percentage of aborted fetuses among the "black" category are significantly higher than the percentage of self-identified "blacks" in the Gainesville area. A list of the frequencies of the various pathologies are shown in Table 1. In all cases of less than 7 months gestational age, immaturity was a contributing factor to death. Non-viable immaturity was almost always accompanied by hyaline membrane disease and /or cerebral intraventricular hemorrhage. Several autopsy reports listed immaturity as the primary cause of death, especially in cases in which no underlying pathology was found. In some cases, non-specific sepsis may have been secondary to fetal death in utero with fetal maceration. In addition, chorioamnionitis was almost universally associated with premature rupture of membranes. It should be noted that the same sampling bias exists with studies of fetal abortus and stillboms that exists with studies of skeletal populations. The population

PAGE 26

12 does not represent a "normal" healthy poptdation, but a population of non-survivors. This study examines all specimens without skeletal abnormalities, even in the presence of an imderlying soft-tissue pathology. The impact of soft-tissue pathology on skeletal growth and development and the implications for anthropologists are addressed in chapter 4. Maternal Profile Maternal age was recorded for 346 cases. Maternal age distribution is shown in Table 2. The mean maternal age at abortion/birth is 23.75 years, ranging from 13 to 43 years of age. Only one autopsy record listed advanced maternal age as a contributing factor in fetal death. Measuring Relative Long Bone Growth In child growth studies where age is known a number of methods of assessing growth rate have been used: Gam and Shamir (1958) used annual increments, whereas Krogman (1950) and Meredith (1962) preferred to express increments as percentage of growth already attained. Bayer and Bayley (1959) divided the increment by average adult size and this quotient by the interval of time during which the increment was attained. This is similar to the relative growth rate used by Gray (1941). Gavan and Swindler use relative growth rates, that is, change per vmit size per tmit time. True average of an infinite series of such rates is attained by dividing the natural logarithms of size by the difference in time: ( In S^ In S^)/(t2 1^), where S equals size and t equals time. As discussed in later chapters, time intervals during the prenatal period can only be inferred from fetal development and/or maternal history. Therefore, calculation of true growth rates is not possible and the above-mentioned methods of determining growth rates are unavailable. This study draws from the method used by Armelagos et al.

PAGE 27

13 Table 1: Frequency of pathological conditions determined at autopsy. Pathology n percentage Abdominal pregnancy Abruptio placenta / marginal abruption Placenta previa Immature placenta; placental insufficiency; Premature rupture of amniotic membranes Amniotic band syndrome Chorioamnionitis; chronic villitis Anencephaly Hydrocephaly Congenital heart and /or vascular defects Maternal pre-eclampsia Maternal diabetes Maternal drug use Maternal syphilis Congenital syphilis Potter's syndrome; renal agenesis or dysplasia Hydrops fetalis Down's syndrome and/or other chromosome Sepsis (non-specific) Multiple congenital skeletal malformations (i.e. Birth trauma and /or fetal anoxia Miscellaneous pathology that was a significant Note: Percentages do not equal 100%. Each case may have more than one or none of the categorized pathologies. 2 <1% 35 8.8% 9 2.2% 29 7.3% 25 6.3% 6 1.5% 18 4.5% 15 3.8% 12 3.0% 22 5.5% 9 2.2% 6 1.5% 3 <1% 2 <1% 3 <1% 14 3.5% 9 2.2% 10 2.5% 14 3.5% 23 5.8% 7 1.8% 15 3.8%

PAGE 28

Table 2: Maternal Age Profile 14 Maternal Age n Percent Valid % Cum. % 13 1 0.3 0.3 0.3 14 2 0.5 0.6 0.9 15 8 2.0 2.3 3.2 16 16 4.0 4.6 7.8 17 23 5.8 6.6 14.5 18 29 7.3 8.4 22.8 19 23 5.8 6.6 29.5 20 17 4.3 4.9 34.4 21 22 5.5 6.4 40.8 22 27 6.8 7.8 48.6 23 20 5.0 5.8 54.3 24 18 4.5 5.2 59.5 25 24 6.0 6.9 66.5 26 13 3.3 3.8 70.2 27 15 3.8 4.3 74.6 28 13 3.3 3.8 78.3 29 14 3.5 4.0 82.4 30 13 3.3 3.8 86.1 31 14 3.5 4.0 90.2 32 5 1.3 1.4 91.6 33 3 0.8 0.9 92.5 34 5 1.3 1.4 93.9 35 3 0.8 0.9 94.8 36 6 1.5 1.7 96.5 37 3 0.8 0.9 97.4 39 1 0.3 0.3 97.7 40 6 1.5 1.7 99.4 42 1 0.3 0.3 99.7 43 1 0.3 0.3 100.0 Missing data 52 13.0 ... ... Total 398 100.0 100.0 ...

PAGE 29

15 (1972) in their cross-sectional analyses of prehistoric populations of Sudanese Nubia in which curves of growth distance and growth velocity were used to show the relative increase in percent and the linear incremental increase of each long bone. Instead of using age based on dental development, however, the current study uses CHL to group cases into clusters of similar development. So, like Armelagos et al. (1972), they are not true growth rates, but expressions of proportionality between limb growth and fetal stature. The data were analyzed on a Power Macintosh 8500/120 using SPSS Graduate Pack^M Advanced version for PowerMacintosh, v. 6.1.1, and on an IBM-compatible personal computer using SAS for Windows95. Summary descriptive statistics are provided for the total sample and each sample subset as needed. Bivariate regressions are via the least squares method. Data were generated both log transformed and in raw data space. However, log transformations did not significantly affect the interpretation of the data and is thought to be better suited to analysis of shape variables than linear variables (Jungers et al., 1995). Only plots in raw data space are presented. Post-hoc comparison of the means is used to determine if significant differences exist in long bone proportions between groups. Also, slopes and intercepts of linear regression formulae, and proportional indices are compared using the Bonferroni t-test. Results with Alpha < 0.05 are considered significant.

PAGE 30

CHAPTER 3 CORRELATION BETWEEN CROWN-HEEL LENGTH AND LONG BONE DIAPHYSEAL LENGTH Three-hvindred and ninety-eight cases had both autopsy protocols and measurable radiographs. Of these, 252 cases had a CHL recorded at autopsy and at least one measurable long bone diaphysis. The sample was arbitrarily clustered into 12 groups of CHL, roughly corresponding to gestational ages from 4 limar months to post-term. The first and last CHL groups are discarded for most of the analysis because they contain an insufficient mmiber of cases. CHL length group distribution is reported in Table 3. Table 3: Distiribution of all cases with recorded CHL by CHL group and approximate gestational age. CHL CHL in Approx. gestational n group (mm.) age < 210 > 4 lunar months 3 1 210-239 < 5 lunar months 10 2 240-269 5.0-5.5 Ivmar months 14 3 270-299 5.5-6.0 limar months 35 4 300-329 6.0-6.5 limar months 38 5 330-359 6.5-7.0 lunar months 41 6 360-389 7.0-7.5 lunar months 26 7 390-419 7.5-8.5 lunar months 28 8 420-449 8.5-9.0 lunar months 11 9 450-479 9.0-9.5 lunar months 17 10 480-520 9.5-10 lunar months 24 11 > 520 Post-term 5 Total 252 16

PAGE 31

17 Mean Long Bone Lengths Mean long bone lengths for each CHL group are reported in Tables 4 through 9. Mean long bone lengths are progressively longer for each successive CHL group, with the exception of the 420-449 CHL group, which introduced a sampling error for the Table 4: Mean humerus length of all fetuses with recorded CHL by CHL group. (n=240) CHLgrp. n mean sd range 210-239 10 27.85 2.46 24.0-31.0 240-269 14 33.18 4.90 26.0-47.0 270-299 35 36.50 2.78 29.0-42.0 300-329 38 39.11 3.40 31.5-47.0 330-359 38 43.21 3.09 37.0-49.0 360-389 26 46.87 2.73 42.0-54.0 390-419 28 53.36 5.35 46.0-69.0 420-449 11 56.14 2.92 52.0-61.0 450-479 16 58.28 5.81 44.0-69.0 480-520 24 64.02 5.07 55.0-78.5 Table 5: Mean radius length of all fetuses with recorded CHL by CHL group. (n=226) CHLgrp. n mean sd range 210-239 9 23.50 2.05 19.5-26.0 240-269 14 26.93 3.92 21.0-37.0 270-299 33 30.52 3.02 24.0-39.5 300-329 38 32.29 2.92 25.5-38.5 330-359 37 35.50 2.47 31.0-40.0 360-389 26 39.00 2.60 34.0-45.0 390-419 26 44.56 4.85 38.0-55.0 420-449 11 45.82 2.69 42.0-50.0 450-479 13 47.62 5.74 36.0-58.0 480-520 19 52.24 2.68 45.5-55.5

PAGE 32

18 femur, tibia and fibula. The growth trajectory for the other bones was affected as well. The error is probably the result of a single case in which the CHL recorded in the autopsy protocol was significantly shorter than the actual measurement. Table 6: Mean ulna length of all fetuses with recorded CHL by CHL group. (N=226) CHLgrp. n mean sd 2.15 range 210-239 9 26.00 22.5-29.0 240-269 14 30.14 4.58 23.5-42.5 270-299 34 33.78 2.67 28.0-39.5 300-329 38 36.33 3.21 29.5-44.0 330-359 37 40.15 2.74 34.5-45.0 360-389 26 43.75 2.47 39.5-51.0 390-419 25 50.38 5.92 43.0-67.0 420-449 11 51.50 2.55 48.0-57.0 450-479 13 54.04 6.43 42.0-67.0 480-520 19 59.37 2.84 54.0-63.5 Table 7: Mean femur length of all fetuses with recorded CHL by CHL group. (N=191) CHLgrp. n mean sd 2.66 range 210-239 10 28.60 24.5-33.0 240-269 13 35.38 5.65 28.0-51.0 270-299 35 39.00 3.59 29.0-46.0 300-329 36 42.71 4.61 32.0-56.0 330-359 40 47.40 3.90 39.0-55.0 360-389 24 51.42 2.99 47.0-57.0 390-419 16 61.19 6.92 52.0-75.5 420-449 6 65.33 6.22 55.0-73.0 450-479 6 64.67 8.26 49.0-73.0 480-520 5 74.20 7.56 62.0-81.0

PAGE 33

19 Table 8: Mean tibia length of all fetuses with recorded CHL by CHL group. (N=137) CHLgrp. n mean sd range 210-239 10 24.80 2.66 20.5-29.0 240-269 12 29.08 3.09 24.5-35.5 270-299 28 33.09 2.96 24.5-38.5 300-329 28 35.23 3.43 29.0-42.5 330-359 27 40.28 3.24 35.0-47.0 360-389 11 44.05 2.48 40.0-49.5 390-419 9 54.22 4.86 47.0-63.0 420-449 3 54.33 5.69 48.0-59.0 450-479 4 53.00 7.57 42.0-58.0 480-520 5 62.60 5.59 55.0-69.0 Table 9: Mean fibula length of all fetuses with recorded CHL by CHL group. (N=118) CHLgrp. n mean sd range 210-239 8 24.38 2.55 20.5-28.0 240-269 12 28.08 2.75 23.5-33.5 270-299 23 31.48 2.92 23.5-37.0 300-329 23 33.17 3.52 27.0-40.0 330-359 25 38.02 3.19 33.0-44.5 360-389 9 41.72 2.33 39.0-47.0 390-419 7 51.21 4.40 47.0-60.0 420-449 3 51.00 4.44 34.0-54.5 450-479 4 49.75 7.37 39.0-55.0 480-520 4 58.38 5.79 51.0-64.5 Correlation Between CHL and Length Long bone lengths were plotted against CHL. R-values are reported in Table 10. All long bone diaphyseal lengtiis correlate significantly with CHL (r^ > 0.8375; p < 0.01).

PAGE 34

20 Table 10: Correlation matrix of r-values for CHL versus long bone diaphyseal length Humerus Radius Ulna Femur Tibia Fibula CHL .9311 (249) .9156 (232) .9246 (232) .9063 (196) .9188 (142) .9088 (123) Humerus .9848 (231) .9876 (231) .9879 (194) .9869 (141) .9881 (122) Radius ... .9928 (231) .9774 (189) .9807 (137) .9809 (119) Ulna .9854 (190) .9879 (138) .9897 (120) Femur ... ... ... .9910 (142) .9911 (123) Tibia ... ... ... ... .9959 (123) The correlation between fetal long bone length and CHL is consistent with that of adult long bone length and stature (Genoves, 1967; Trotter and Gleser, 1958). Leastsquares linear regression produced the following formulae for predicting CHL from radiographic bone lengths: CHL = 45.571 + (variablel) 6.839 7.704 CHL = 47.886 + (variable^) 8.196 8.696 CHL = 51.642 + (variable^) 7.193 8.097 CHL = 90.835 + (variable^) 5.188 7.866 CHL = 82.858 + (variableS) 6.308 + 8.351 CHL = 79.677 + (variable^) 6.896 9.948 where, variable^ is humerus length; variable^ is radius length; variable^ is uhia length; variable^ is femur length; variable^ is tibia length; and variable^ is fibula length.

PAGE 35

21 In order to compare my raw data with that of other investigators, I convert CHL to gestational age in weeks as per Table 11. The growth rates from this study are compared with those of Russell et al. (1972), and Kosa (1989) by assuming the validity of relationship between fetal length and gestational age. However, the tenuous nature of determining gestational age based on length will be discussed in Chapter 5. Table 11: Conversion from lunar months to weeks. Gestational age in lunar months Gestational age in weeks 5.5-6.0 lunar months 22-24 weeks 6.0-6.5 lunar months 24-26 weeks 6.5-7.0 lunar months 26-28 weeks 7.0-7.5 lunar months 28-30 weeks 7.5-8.5 lunar months 30-34 weeks 8.5-9.0 limar months 34-36 weeks 9.0-9.5 lunar months 36-38 weeks 9.5-10 lunar months 38-40 weeks The growth trajectories for the Fazekas and Kosa (1978) sample correspond closely with the current data. The slightly longer limb lengths of my sample through most of the curve are most likely a product of slight radiographic magnification error, although genetic or secular differences in growth cannot be ruled out. The humeral length at 22 weeks and term are nearly identical. This reduces the possibility that the curves illustrate a secular difference in size between the two samples. Long bone lengths from Russell et al. (1972) are significantly higher than both Fazekas and Kosa (1978) and the current study. Table 12 shows the mean lengths for the humerus. Measurements at 28-30 weeks gestation are as much as 21 mm. greater than those reported by Fazekas and Kosa and more than 18 mm. longer than the current

PAGE 36

22 study. This trend continues until term, with mean humeral length 7 mm. longer than the two other studies. However, the 3 Russell curves correspond with one another. Aside from the significantly longer humeral lengths, the slope of the curves is consistent with the slopes from my data and that of Fazekas and Kosa (Figure 3). Only data for the humerus are shown, however, the data for the other 5 long bones show the same trend. Table 12: Mean humerus lengths for gestational age in weeks reported by Warren, Russell et al. (1972), and Fazekas and Kosa (1978). Weeks gestation Warren Russell et al. (LMP) Russell et al. (Osseous devel.) Russell et al. (Date of birth) Fazekas & Kosa 28-30 46.87 66.33 60.00 n/a 45.00 30-34 53.36 64.57 61.20 61.80 50.40 34-36 56.14 64.97 66.91 65.11 54.30 36-38 58.28 68.84 69.13 67.63 58.40 38-40 64.02 70.14 71.33 71.00 63.10 The Russell et al. data were obtained by radiographic measurements of the long bones of in utero fetuses. The measurements were compared with fetal maturity obtained from 3 other parameters Menstrual history (LNMP), osseous assessment of knee and ankle development as reported by Hartley (1957), and extrapolation from the date of delivery. The greater humeral length reported by Russell et al. may be due to greater magnification error due to the distance of the in utero fetus from the film plane, or it may have been introduced by questionable normalization procedures based on estimation of gestational age (see Chapter 5). Using CHL as a normalizing datum allows data sets from different researchers to be compared without relying on additional studies on determination of gestational age.

PAGE 37

23 Figure 3. Comparison of Warren, Fazekas and Kosa, and Russell et al. data 80.00 -r70.00 X 30.00 -20.00 -10.00 -0.00 22-24 Warren Russell et al. (LNMP) Russell et al. (bone development) Russell et al. (by birthdate) Fazekas and Kosa 24-26 26-28 28-30 30-34 Weeks gestation 34-36 36-38 38-40 The data presented in this study are consistent with those of other published studies on fetal long bone growth, with the exception of the Russell et al. data. My data also fit studies of postnatal long bone growth at the natal period. These growth studies are based on longitudinal surveys of large numbers of healthy children and have been found to be extremely accurate in clinical practice.

PAGE 38

CHAPTER 4 RELATIVE GROAAHTH AND LIMB PROPORTIONALITY Establishing standards of growth for a specific population leads to a better understanding of populational and individual variation. Growth velocities and limb proportions of a given population can be compared with standards derived from wellnourished populations to better understand the impact of environmental conditions on normal growth and development. Comparison between postnatal growth studies are normalized based on chronological age. Chronological age is calculated from birth and, in general, variations in gestational duration are not taken into account. Since chronological age is not known for most skeletal populations, anthropological aging techniques such as dental eruption and wear patterns are used to categorize specimens according to developmental age (Merchant and Ubelaker, 1977; Y'Edynak, 1976; Armelagos et al, 1972). This study is hampered by a similar problem in that the time of conception cannot be known with absolute accuracy. However, the linear growth of the fetus allows the use of CHL as a normalizing datum for comparing limb proportionality between two populations. Error in determining age from dental eruption or other methods is not introduced to this type of analysis of fetal long bone growth. Relative Rates of Growth By plotting specific bone lengths against CHL it is possible to examine relative rates of growth for each long bone. Figure 4 illustrates relative growth trajectories for each long bone. Fetuses of short gestational age have humeri and femora of nearly equal length, with the femur only slightly longer than the himierus. By the end of the 24

PAGE 39

25 gestational period the femora have elongated relative to the humeri. The disparity is caused by a greater relative growth velocity during the entire growth period, as opposed to a short period of rate increase. The tibia and fibula show a sampling error Figure 4. Combined distance curves for all bones showing relative growth trajectories. The Y axis represents long bone length in mm.; = humerus; = radius; = ulna; = femur; A= tibia; o = fibula. See table 3 for measurement ranges of CHL groups. for CHL group 7 (i.e. 390-419 mm.). The femur, tibia, and fibula show an increased rate of growth during the period just prior to birth. Again, the low number of cases with lower extremity measurements in the latter CHL groups may produce a sampling error. Tables 13-14 show the incremental increase in millimeters and the relative increase in percent for long bone growth by CHL group. The femur and tibia show greater incremental and relative growth than the humerus and radius at almost every CHL group. The exception is the 390-419 to 420-449, and the 420-449 to 450-479 groups, which demonstrate a sampling error of negative absolute and relative growth.

PAGE 40

26 Table 13. Incremental increase (11) in millimeters and relative increase in percent (RIP) between CHL groups Hiimerus, radius and ulna. Humerus Radius Ulna CHL group n(mm) RIP n(imn) RIP II (mm) RIP 210-239 to 240-269 5.33 19.1 3.43 14.6 4.14 15.9 240-269 to 270-299 3.32 10.0 3.59 13.3 3.64 12.1 270-299 to 300-329 2.61 07.2 1.77 05.8 2.55 07.5 300-329 to 330-359 4.10 10.5 3.21 09.9 3.82 10.5 330-359 to 360-389 3.66 08.5 3.50 09.9 3.60 09.0 360-389 to 390-419 6.49 13.8 5.56 14.3 6.63 15.2 390-419 to 420-449 1.18 02.2 1.26 02.8 1.12 02.2 420-449 to 450-479 3.71 06.8 1.80 03.9 2.54 04.9 450-479 to 480-520 5.74 09.8 4.62 06.7 5.33 09.9 Table 14. Incremental increase (H) in millimeters and relative increase in percent (RIP) between CHL groups Femur, tibia and fibula. Femur Tibia Fibula CHL group H (mm) RIP n(mm) RIP n(mm) RIP 210-239 to 240-269 6.78 23.7 4.28 17.3 3.70 15.2 240-269 to 270-299 3.62 10.2 4.01 13.8 3.40 12.1 270-299 to 300-329 3.71 09.5 2.14 06.5 1.69 05.4 300-329 to 330-359 4.69 11.0 5.05 14.3 4.85 14.6 330-359 to 360-389 4.02 08.5 3.77 09.4 3.70 09.7 360-389 to 390-419 9.77 19.0 10.17 23.1 9.49 22.7 390-419 to 420-449 4.14 06.8 0.11 0.20 -0.21 -0.4 420-449 to 450-479 -0.66 -00.1 -1.33 -2.4 -1.25 -2.5 450-479 to 480-520 9.53 14.7 9.60 18.1 8.63 17.3

PAGE 41

27 When the curve for the humerus is plotted with data from a postnatal study (Maresh,1955), the data fit for the natal period. That is, the rate of growth is consistent through 6 months of age, and then accelerates during the period from 6 months to 1 year of age (Figure 5). H 1 1 1 1 1 \ 1 1 1 1 i 1 1 1 1 CN -^ VO Age Figure 5. Distance curve for the humerus prenatal (Warren) and postnatal (Maresh); reflecting the relative length of the humerus to gestational age. Influence of Pathnlo^ nn Linear Growth and T imb Propor tions Studies of long bone growth utilizing data from bioarcheological skeletal samples are necessarily cross-sectional in nahire. This study is cross-sectional as well, and as such, is biased in that it does not present longitudinal data from a normal, healthy individual, but many different individuals that did not survive until adulthood Qohnston, 1962). However, the autopsy data contained in this study permit the examination of subsets of the sample categorized by the types and severity of pathology. No pathology was

PAGE 42

28 noted in many cases; other cases exhibited an acute insult during the birthing process (i.e. abruptio placenta, placenta previa, or acute chorioamniorutis), or acute "pathology" associated with premature delivery of a normally developing fetus (i.e. hyaline membrane disease or persistent fetal circulation). These cases are compared with those in which a significant pathological condition, possibly affecting normal growth, was noted during postmortem exanunation. Tables 15-20 show average long bone length by CHL group for all cases, "normal" cases, and "pathological" cases. Distance curves (Figures 5-10) compare the growth of the long bones of the "normal" subset and the "pathological" subset. Linear growth is not significantly affected by most prenatal pathology in this study. Likewise, Brenner et al. (1976) foimd that data derived from spontaneous abortions in the late second and third trimesters were valid for interpreting normal fetal growth, even when underlying pathology was noted in the fetus. The similarity in proportionality does not mean that the "pathological" group is not growth retarded for gestational age in comparison to the normal group it may be, but this is impossible to determine because CHL is used as a normalizing measurement and true gestational age is imknown. However, long bone length relative to CHL does not differ between pathological and non-pathological sets. Figures 6-11 demonstrate that the proportional relationship between long bone length and CHL is both stable and predictable. The lack of effect of pathology on long bone length is consistent with other studies that find girth but not length to be affected by congenital conditions (Richards and Anton, 1991). Sex Differences in Linear Growth and Limb Proportionality To consider sex differences in fetal growth, the sample was divided into male and female groups and mean long bone lengths were determined for each CHL group (see Tables 21-24 for descriptive statistics). Bagnall et al. (1978) found no significant

PAGE 43

29 Table 15. Comparison of mean humeral length between all cases, "normal" cases, and "pathological" cases. CHL group Means for all cases SD Means for "normal" cases SD Means for "pathologic" cases SD 210-239 27.85 2.46 27.06 2.06 31.00 240-269 33.18 4.90 33.50 1.84 32.94 6.48 270-299 36.50 2.78 36.09 2.58 37.19 3.09 300-329 39.11 3.40 39.46 2.82 38.57 4.18, 330-359 43.21 3.09 43.88 3.09 41.75 2.65 360-389 46.87 2.73 46.79 2.57 47.07 3.35 390-419 53.36 5.35 52.82 5.51 54.18 5.26 420-449 56.14 2.92 55.44 2.99 58.00 2.00 450-479 58.28 5.81 57.59 5.29 59.80 7.26 480-520 64.02 5.07 62.68 3.63 65.90 6.32 Table 16. Comparison of mean radial length between all cases, "normal" cases, and "pathological" cases. CHL group Means for all cases SD Means for "normal" cases SD Means for "pathologic" cases SD 210-239 23.50 2.05 23.07 2.07 25.00 1.41 240-269 26.93 3.92 27.17 1.63 26.75 5.15 270-299 30.52 3.02 30.35 3.10 30.77 3.00 300-329 32.29 2.92 32.89 2.26 31.37 3.61 330-359 35.50 2.47 36.00 2.59 34.46 1.86 360-389 39.00 2.60 39.11 2.29 38.71 3.49 390-419 44.56 4.85 44.00 4.66 45.32 5.21 420-449 45.82 2.69 45.56 2.69 46.50 3.12 450-479 47.62 5.74 47.17 4.96 48.63 8.01 480-520 52.24 2.68 52.55 2.91 51.81 2.45

PAGE 44

30 Table 17. Comparison of mean ulnar length between all cases, "normal" cases, and "pathological" cases. CHL group Means for all cases SD Means for "normal" cases SD Means for "pathologic" cases SD 210-239 26.00 2.15 25.36 1.95 28.25 1.06 240-269 30.14 4.58 30.08 2.01 30.19 6.01 270-299 33.78 2.67 33.36 2.47 34.46 2.94 300-329 36.33 3.21 36.87 2.54 35.50 3.99 330-359 40.15 2.74 40.74 2.85 38.92 2.12 360-389 43.75 2.47 43.97 2.45 43.14 2.61 390-419 50.38 5.92 49.57 6.13 51.60 5.68 420-449 51.50 2.55 51.00 2.09 52.83 3.69 450-479 54.04 6.43 53.56 5.22 55.13 9.51 480-520 59.37 2.84 60.23 2.69 58.19 2.78 Table 18. Comparison of mean femoral length between all cases, "normal" cases, and "pathological" cases. CHL group Means for SD Means for SD Means for SD all cases "normal" "pathologic" cases cases 210-239 28.60 2.66 27.75 2.17 32.00 1.41 240-269 35.38 5.65 34.83 1.94 35.86 7.76 270-299 39.00 3.59 38.55 3.27 39.77 4.10 300-329 42.71 4.61 43.05 3.84 42.23 5.63 330-359 47.40 3.90 48.40 4.09 45.54 2.78 360-389 51.42 2.99 51.06 2.68 52.29 3.73 390-419 61.19 6.92 60.14 6.49 62.00 7.52 420-449 65.33 6.22 62.50 5.45 71.00 2.83 450-479 64.67 8.26 64.20 9.15 67.00 ... 480-520 74.20 7.56 70.00 11.31 77.00 4.58

PAGE 45

31 Table 19. Comparison of mean tibial length between all cases, "normal" cases, and "pathological" cases. CHL group Means for all cases SD Means for "normal" cases SD Means for "pathologic" cases SD 210-239 24.80 2.66 24.00 2.27 28.00 1.41 240-269 29.08 3.09 29.25 2.36 28.92 3.92 270-299 33.09 2.96 32.56 2.65 33.79 3.31 300-329 35.23 3.43 36.32 3.17 33.55 3.24^ 330-359 40.28 3.24 40.97 3.65 39.42 2.55 360-389 44.05 2.48 43.69 1.33 45.00 4.77 390-419 54.22 4.86 55.13 5.72 53.50 4.61 420-449 54.33 5.69 53.50 7.78 56.00 0.00 450-479 53.00 7.57 51.33 8.33 58.00 480-520 62.60 5.59 62.00 9.90 63.00 3.61 Table 20. Comparison of mean fibular length between all cases, "normal" cases, and "pathological" cases. CHL group Means for all cases SD Means for "normal" cases SD Means for "pathologic" cases SD 210-239 24.38 2.55 23.50 2.24 27.00 1.41 240-269 28.08 2.75 28.08 2.04 28.08 3.54 270-299 31.48 2.92 30.75 2.69 32.27 3.07 300-329 33.17 3.52 34.65 3.07 31.25 3.24 330-359 38.02 3.19 39.04 3.65 36.92 2.25 360-389 41.72 2.33 40.79 1.15 45.00 2.83 390-419 51.21 4.40 51.63 5.82 50.67 2.52 420-449 51.00 4.44 50.25 6.01 52.50 0.00 450-479 49.75 7.37 48.33 8.33 54.00 ... 480-520 58.38 5.79 56.00 7.07 60.75 5.30

PAGE 46

32 Normal set Pathological set ON On ON ON ON CTi \0 ON (N in CA S CM ro CO rS 6 o O o T-l •4* rv O CO tS CM CN rr> CO ON 00 CO o VO CO o ON CO ON o C^J ON o in o CM in I o 00 CHL group Figure 6. Comparison of normal and pathological subsets for humerus length. bu.uu g 50.00 s _g 40.00 ^ 30.00 • J .^^^^J,^"-''^ ^ 20.00 -o ^ 10.00 0.00 1 — 1 1 1 Normal set Pathological set 4+ r-l ON On vO o\ ts CM ^ M csl CM 2 ON On ON ON ON o th 00 rH -=!< I>> t
PAGE 47

33 c 30.00 + J 20.00 -10.00 4. 0.00 Normal set Pathological set + 4H ON CO ON NO CN) 8; ON CN CO ON ID CO On 00 CO ON ON 7 1" o CN in O o CNl s o o CO o CO CO 6 NO CO o CO o 5i o in o 00 CHL group Figure 8. Comparison of normal and pathological subsets for ulna length. ON ON ON On ON ON 0\ ON ON o CO NO ON CN in 00 T-H ^ t>. CN) '^ CN CN CO ^ CO *! ^ •* in o o O o o o o 6 o o T-H Tf K o CO NO On CN in 00 CN
PAGE 48

34 CHL group Figure 10. Comparison of normal and pathological subsets for tibia length. 70.00 J 60.00 -50.00 .^ 40.00 -S 30.00 4I 20.00 -10.00 -I0.00 Normal set Pathologic set + Hi ON t cs OS cn I o o CO ON IT) CO I o CO CO On 00 CO I o NO CO ON o ON CO ON o On I o in o CM in o 00 CHL group Figure 11. Comparison of normal and pathological subsets for fibula length.

PAGE 49

35 difference in length of the long bones between males and females of 8 to 24 weeks gestation. The current study extends their findings through the entire fetal period. No significant difference in proportionaUty or relative growth was found between males and females. Distance curves illustrate the similarity in proportions between males and females for this sample (Figures 11-16). Success at differentiation between males and females during the fetal period, based on skeletal traits, has been confined to the metric and non-metric analysis of the pelvic bones (Boucher, 1957) and the dimensions of the vertebral bodies (Lippert and Lippert, 1960). Thus, sex cannot be distinguished by long bone length relative to CHL. .S 70.00 T 60.00 50.00 ^ 40.00 -c •Z 30.00 ^ 20.00 X 10.00 0.00 ON ON ^. tM CM ON On CS ON 7 ON 00 ON 5i ON o CN) ^ 6 o o o o o S CO CO NO CO ON CO CN in "* CHL group Figure 12. Comparison of female and male means for humerus length by CHL group.

PAGE 50

36 o o o o in o in o o 60 CM o CO tN CO o CS in in in o in o in o o in o 6 CO CO rsi S CO CO o •>* tN in OS CO tN in V 00 in in 00 OS CO CO 00 T— 1 CN rg CO CO o l-H T— OJ § o o o IN in T— < IN JQ o o O in O O o o 6 in CM CO CO 00 CO t— 1 in CO in tN ON m c >* in o T— 1 IN eg ^ T-H fN CN CS ir> o in in o o in o q q 01 bO T3 03 S 01 60 in o u CS CS in o CM in CN o CO CS CN CN o o IN tN \0 CS 00 CO cr\ CN ^ SO CO CO (N I in cvi CO 00 CO IN I o in CO CO in o as ON CO CO in in I o IN q so ON in o CNl c4 CO SO CO 00 so o rt; CO tA ccj in o CS in rH IN m m 00 in I o CN CO o o d in CO in 00 CN in >* I o CN CO "* 1—1 I— I CS 00 m o 00 CO VO CS o ON o in CO in CJS 00 CS t?? 2^ CO ^ in vo ^ in in o IN q o in 00 o CS in I o IN IN CS CO NO NO o CN in 00 o o IN in CO tN CN NO t-in in tN CN NO ON NO Os CO (N ON NO CN O T-H fNl O 'J' CN ON ON CN I O IN (N On CN) CO I o o CO On in CO t o CO CO ON 00 CO I o NO CO as o ON CO ON I o CN ON NO o NO NO CO o in NO CN CN CS CN o o ON in 00 tv CS IN CN O o in CN CN cr^ IN I o in o CN in t o 00

PAGE 51

37 aD o to X u a a, 3 Oh en 0) 01 .a, u 0) CO O) H to IT) in O o o O O o o o & s 00 (N 00 CO •** in in in 1^ cs O in in o o o in o o o IH ID IT) cs en in CO o CO a-. Cv) in ? 00 T— 1 so in in ON 00 00 ON CN On I— 1 VO Cvl CO CM T— t CN cs 00 CN en OS 01 01 o u U o in t^ Ti< 00 in ?R CN CM I-H CN CN s NO CN O CO CO CO IN CO o CN CN in CN in ON in o CN CO CN NO CN in cs CN CO NO CS o CO 00 CN NO CO rs 00 IN. NO NO o CO CO cs CO cs in ON CO in in cs ON in 00 CO I o ON cs in CO IN NO NO CN 00 00 d CO 00 o CO CO o d CN CS ON CO in CO o d o CO NO ON In CO o NO 00 CO NO IN o in o CN CO CO tN o NO 00 in NO CO 00 IN in in 00 NO CN in IN CO 00 in i-i in ^ S IN IN NO NO IN >* I in IN tN CN CN) CO NO (N in CO in 00 tN CO o in 00 CO CN NO cs CO CO CN CO in IN NO CO 00 CO d in IN 00 NO CO iri NO in 00 NO cs NO in CO in 00 CN NO CO IN m CJN NO o IN I O IN in Cn no in o 00 '^ Cn NO in in in IN cs NO in NO CS s 00 ON IN NO 00 ON as ON ON ON C3N ON ON 0\ o CO NO On r-J in 00 r—i '^ IN (N CN CN CS CO CO P3 ^ ^ "* "? o O o o o o o o o o T-l ^ IN o CO NO ON CS in 00 CNl CS CN CO CO CO CO <* ^

PAGE 52

38 o in o o o o o o in 01 oo tN CO CO CO ON CO CN IN o in o ifi in o o o O o o u CO CS CM s ON CO IN t— ( in in ^ O 1—1 CN ^ CO T— t 00 CN in in CO 04 CO CO CN CO ^ CN ON § CO 00 o o in o in in o T-H tQ o o O in tQ ^ CN ON I— 1 CO CN CO IN CO CN in •<* CN in tN in C CO t>. CO rH CO I— 1 in VO T— 1 c^ CN q in O q in in o o o On in 00 i-i CM ON CO 00 tN (50 CN CO CO "T 'T T NO in NO q 1 q in o ci q 6 o ci bi CO 00 ^ ON •o rH d * a\ o o IN CN CN tN IN CO CM q in in NO e ON in 00 r-( NO CM CO d IN CO CNl CO CO ^ ^ in NO vO NO c> c '^ IN T-H CN in CM o CN tN rH 1—i rH CM ^ CO &3 ON On 0\ ON c^ ON ON 3i ?• S O CO NO On CN in 00 rH <*< IN ^ & C^ CN O) CO CO CO -* t f "? o O O o o o 6 o O 6 |J tH IN o CO NO ON CM in 00 D CN (N CNl CO CO CO CO ^ •^ ^

PAGE 53

39 to a. o & i-i (A X ;3 •f-H u •T3 ^ oi, i ~ O o in q q q o o in in 0) (N in CO ON CO 1-H in I-H in NO CM in d NO 1 CO NO 1 s 1 q 1 in 6 in in in cb q o q l2 CO CM 00 tN ON CN CO 5^ CO CO 00 CM in "^ 1— < CO 00 q o On ON NO ^ in CM T-H O^ is (N CO (N CO CM CM NO T-H CO CM g 00 On O) in 00 00 o CO CO T-H NO 00 CO CO in 01 as CO CO tri CO T-H in d in in in ON in c -* 00 CM CM CTN T-H 00 1-H 00 T-H '^f b> CO T-H o q o O q o O o o in CO in CO NO CO ON CO in in in K in in in in 1 s p q 1 o 1 in q o q 6 o in }H CM cs CM in CM CO in CO 00 CO NO in •S 00 ON ON ON o 00 IN. in o T-H 00 CO IS. ^ T-i tN (N CM CN CM in r-i CM (N i K CO 1— 1 CO CM ON 1-H NO I-H 00 T-H vO 00 T-H NO S en ^ -H CO in CO ON CO Lfi in On CM in c <* 00 tN NO CM ON 1—^ 00 T-H T-H '^ tN CO rH q o q O o o q o O in bC CO CO rH in Tf "? On NO in S 00 Cn s 6 6 6 in C5 o o Ci q CD I-I in CM (N 0\ (N r-i CO 00 CO CO NO CM in in in in in •^ 2 CM CO ON CN o o K ON T-H CO NO tN 00 in CM CO CN CO CO CN) NO T-H CM in i ^ ^ >* CO CM "1 R ^ T-H CM § in 00 'O M3 6 00 CM CO NO CO 00 CO CN ^ i in ON in NO K •>* 00 T-H CN NO tN 00 T-H 00 T-H ON T-H >* o T-H NO T-H 1 OS CO ON CM 85 c^ ON CN CO On in CO ON ON 1-H On K T in o CM 6 i O o CO CO 6 NO CO ci CO 6 CM 6 ^

PAGE 54

40 60.00 -r CHL group Figure 13. Comparison of female and male means for radius length by CHL group. 80.00 J 70.00 .. i 60.00 S 50.00 % 40.00 .r. 30.00 i 20.00 + ON i OS On ON IT) CO 8 en o CO CO ON c CO I o CO 1< t o ON CO ON o CN ON I o ID O CNl in o 00 CHL group Figure 14. Comparison of female and male means for ulna length by CHL group.

PAGE 55

41 60.00 -r 50.00 40.00 --S ^ 30.00 c 01 20.00 .10.00 0.00 o CTn en ON 7 in o cs o in o CHL group Figure 15. Comparison of female and male means for femur length by CHL group. 70.00 -J 60.00 g 50.00 ;^ 40.00 60 c 30.00 .-.ilS^ 1 20.00 ^ 10.00 n 00 1 c c c ^ ON 5 S n CM ON ON ON in 00 o ^ o ON o On in o in o 00 CHL group Figure 16. Comparison of female and male means for tibia length by CHL group.

PAGE 56

42 60.00 -r 50.00 -40.00 .30.00 -.s 150 c 01 20.00 .10.00 0.00 ON CO 6 On CM On ON CN ON (N CO I o o en ON IT) CO o CO CO ON 00 CO o CO ON T— I I o On CO On I O CN| ON I o LO o cs in o 00 CHL group Figure 17. Comparison of female and male means for fibula length by CHL group. Influence of Self-identified Racial Category on Linear Growth and Limb Proportions Skeletal variation exists at the populational level. Despite extensive research into racial differences in the skeleton, however, describing the range of human variation and detecting racial affinity is perhaps the most difficult task of the osteologist. Group values for long bone proportions have been show to be diagnostic of racial affinity in adults, although there is considerable overlap in ranges when the formulae are applied to individuals (Krogman and Iscan, 1986; Modi, 1957; Schultz, 1937). Krogman (1955) foimd higher brachial and crural indices for American Blacks than for American Whites. Comparison of the data with earlier studies must acknowledge the basic flaw in racial analysis racial typologies poorly represent the dynamic, fluid nature of human variation. Earlier studies of racial differences in adult proportionality use the terms "American Blacks," "American Whites," Europeans, and "Blacks" as descriptors of sample subsets. However, if we understand the limitations of such groupings, we may

PAGE 57

43 make assumptions about differential limb proportions based on the data. For the following analysis, we will assume that the groupings labeled as "American Blacks," "Blacks," and self-identified blacks primarily represent people of West African descent, and that "American Whites," "Whites," and self-identified whites primarily represent people of European descent. Descriptive statistics for white and black samples are given in Tables 25-28. The intermembral index (arm length to leg length) for the fetal sample is high compared with published adult indices, reflecting the early stages of development when the humerus is almost as long as the femur. The elongation of the femur towards the end of the prenatal period and during childhood results in a final intermembral index of near 0.700, as reported by Modi (1957) and Schultz (1937). Elongation of the femur and lower leg during development facilitates bipedal locomotion and is a tmique feature of hominid ontogeny. The opposite has been noted in primates who utilize their forelimbs for brachiation and foraging on the trimks of large trees. Elongation of the forelimb in these species is seen early in ontogeny as well. Patterns of growth are species-specific as a result of both phylogenetic and functional constraints. Closely-related species that share locomotor and postural behaviors will also share similar limb proportions (Falsetti and Cole, 1992). Table 29 shows proportional indices for blacks and whites by CHL group. The higher brachial and crural indices of the fetal sample are partially an artifact of measurement of the fetal diaphysis only. The secondary centers of the femora and hiameri constitute a greater percentage of the final bone length than the secondary centers of the forearm and lower leg. Therefore, the ratio of diaphyseal length of tiie distal limb to the proximal limb is relatively longer than measurements taken of the maximum length of mature bones. Populational differences in proportion begin early in development. Table 30 lists values for the entire fetal sample for intermembral, brachial, and crural

PAGE 58

44 O IT) o o in o o CO 1— 1 en en o •* t CN in M 6 lA in o 6 O in t-4 S en CN 00 r4 in en CJs en 00 T) 3 cn in 00 e^j in o CS en en en t-H T-H s o in o [::: T-l ^ ^ in T— 1 o in s •2 ^ (N ?:5 en ^ en o o m in 01 bo "S H 6 0) bO ^ l-l o U ON ^D en 1— I CN en CN o in CD O in en CM en en CN en in in in 0C3 en I in '^ CN in CN ^D in ON in O in o o On IN in en in in so 00 en On en CM T— 1 <* 00 CM NO CM en en CN en 00 NO en en NO (—1 o in in en o o en CN CN in en tv o o ON en r^ T^ en en ON in CM en en ^ >* in in in in in o 1— 1 T-H NO in T— t [ i-i o o O o o in o NO NO in in K^ Cn, in T in in in t>. ^ o o o o o o o 1— On •^ o 00 CnI CM en C~l eo ^ in NO ON en CN ON NO (N I O '^ CM ON ON CM I O CM o in ON T-H Tj< ON cn CM 00 in •* en CN 00 en ^ in in o o o ON in C-J in o in 00 o 04 ^ c^ "* ^ in NO NO NO >* i>~ r-H NO CN T-H t-i T-H T-H On ON ON o as P" s CN in 00 T-H tv CN •? en en T •* in -A o o 6 o 6 o o m NO ON c^ in S cn en en en >* t ^

PAGE 59

45 S3 a 3 o M en bO 3 •-* hJ TJ u >-. ^ 01 "S cfi .^ U Cfl ^: p pa 1! !S Fi 3 n: en u •J3 in in 0) -t3 o J U in (N tN SO m 00 (N CM OS CO 00 CO OS q in CO 00 o Csj CO CO VO CN CO cs CO o so CO in o so CO so SO in in CO so in so CO in o OS o so in o o csi so in OS in cs so CO in o in o in o in 00 in o in OS in ^ ^ "^ CO cs cs tN o in o in CN CN O CM CM in 00
PAGE 60

46 s X u I I I 01 60 T3 (8 o in O o IT) o o in o in 00 CO CO On CO CO IN o CN in in in S in in LO o o o o o o 6 o CN CO CN IN. CS IN CM CO CO T-H tN so On CO IN in S o . tN CO T-H 00 T-H NO T-H in 'i' CN TJH ^ o o O in O o o o O o CO CO o ^ 00 in in tN in F:i f5 K T-H 00 in o o o o o 6 o o 6 CN 00 CN ^ CN CO ON CO tN S in in ON f^ T-H o tN NO T-H in NO 00 in ON 2; CO NO ON CN tN NO CM tN CO CO (N CO CO CO in tN 00 CO T-H T-H 00 NO in CN NO CN in T-H in in in tN tN NO in CM ON CO 00 CO T-H NO CM in s in NO NO ^ tN IN T-H CN CM CO CN CO T-H o T-H rt< NO CO CN NO CN ON ON CN ON CN CO cy^ in CO ON 00 CO ON T-H C3N in O T-H CM O CN CN 6 o CO o CO CO o >^ CO o CO CN o in o

PAGE 61

47 q q in q q in * Tl^ in ^ in 93 "o) 00 CM CO T-H SO CO CO OS <* • 1-H tn T3 vO CO 00 CO r—f ts 00 in >* ;C 3 tn (N CO T^ CO CO CO iri CM iri CO f 0) u a K r^ CO CO as OS so OS CM <4-l (9 o_ ts Cs| rH in CM in CN ?? g 06 t— < in 06 ts is CO SO so CO CM CO CO CO ^ T}< in in in SO m % K IS. CO T-H CM CM T-H CM in T-l ts so T-H T-H so T-H Q4 Cl, 'h 3 as OS 0\ OS OS 0\ as OS ?^ q w CO \o On CM in 00 rH ts CN 1 CM CN CM CO CO CO "T ^ "1* in 6 6 C5 6 6 ci 6 rH ^ ^N CO so OS CN in 00 K CN CM CN CO CO CO CO ^ ^ s u 0) 2 (S H

PAGE 62

48 Table 29. Proportional indices during ontogeny. Brachial index Crural index Intermembral index CHL group White 0.844 Black 0.849 White Black White Black 210-239 0.870 0.860 0.962 0.960 240-269 0.808 0.816 0.847 0.800 0.907 0.968 270-299 0.828 0.840 0.852 0.847 0.913 0.940 300-329 0.826 0.823 0.827 0.828 0.926 0.899 330-359 0.821 0.821 0.852 0.849 0.897 0.895 360-389 0.826 0.841 0.861 0.857 0.894 0.902 390-419 0.831 0.840 0.906 0.863 0.845 0.849 420-449 0.816 0.817 0.791 0.915 0.863 0.828 450-479 0.808 0.837 0.820 0.869 ... 480-520 0.829 0.785 0.835 0.887 0.816 1.001 Table 30. Limb proportion indices for the entire fetal sample and previously published values for adults. Indices Race Fetal (Warren) (n) Adult (Modi, '57) Adult (Schultz, '37) Intermembral w 0.895 (119) 0.704 0.705 b 0.907 (82) 0.703 0.703 Brachial w 0.824 (204) 0.755 0.745 b 0.834 (146) 0.785 0.785 Crural w 0.851 (126) 0.833 0.833 b 0.860 (83) 0.862 0.862 Note: Schultz figures are for males; Under "Race," w = American Whites for Schultz, selfidentified "whites" for Warren, and Europeans for Modi; b = American Blacks for Schultz, selfidentified "blacks" for Warren, and "Black" for Modi. Data for Modi and Schultz taken from Krogman and Iscan, 1986, pp. 294-295.

PAGE 63

49 indices by self-identified group. In the adult, no significant difference is found for the intermembral index between whites and blacks. The same is true for the fetal sample. A Bonferroni (Dunn) t-test for the proportional indices show that there is no significant difference in the intermembral index between the white and black sample (Table 31). Table 31. Bonferroni t-test for significant differences in proportional indices. 'whites' Variable S.D. 'blacks' S.D. Intermembralindex 201 0.895 0.046 0.907 Brachial index 350 0.824 0.031 0.834 Crural index 209 0.851 0.033 0.860 Pr>F Bonferroni (Dunn)ttest a = 0.05 0.037 0.067 ns 0.036 0.004 b > w* 0.024 0.035 b > w* However, the crural and brachial indices are significant at the 0.05 level, with blacks having proportionally longer distal limbs than whites. The same is reported by Modi (1957) and Schultz (1937) for white and black adults. Again, these are group values, and ranges are wide with considerable overlap. Forensic identification of a single individual based on long bone indices would be fooUsh. Tables 32-33 show the ratio of long bone lengths to femur and tibia lengtii for both the white and black samples. Small sample sizes for the 450-479 and the 480-520 CHL groups produce an obvious sampling error. Otherwise, it is clear tiiat the small differences between the white and black samples are predicted by documented differences between black and white adults in earUer limb proportionality studies (Modi, 1957; Schultz,1937).

PAGE 64

50 Distance curves for each long bone show that the black sample has relatively longer limbs than the white sample for most of the prenatal period. The curves are very similar, however, and the bones of the upper arm and forearm of both groups are nearly identical in relative length by birth. The black sample lacked sufficient data for the final two CHL groups. The distance curves for comparison of "Black" and "White" means for long bone length by CHL group are in Figures 18-23. The curves correspond closely until a sampling error occurs for the 420-449 CHL group and above. No data was available for "Black" femoral, tibial, and fibular measurements, and upper extremities measurements were few. Table 32. Ratio of long bone lengths to femur and tibia length for "white" sample (after Armelagos et al, 1972). CHL H/F R/F U/F T/F f/F F/T H/T R/T U/T f/T group 210-239 0.976 0.824 0.912 0.870 0.858 1.149 1.121 0.946 1.048 0.986 240-269 0.926 0.749 0.840 0.847 0.826 1.180 1.093 0.884 0.992 0.975 270-299 0.925 0.767 0.859 0.852 0.807 1.174 1.086 0.900 1.008 0.947 300-329 0.926 0.765 0.859 0.827 0.783 1.209 1.120 0.926 1.039 0.947 330-359 0.912 0.749 0.844 0.852 0.805 1.174 1.071 0.879 0.991 0.944 360-389 0.911 0.752 0.843 0.861 0.837 1.161 1.058 0.874 0.979 0.973 390-419 0.880 0.731 0.831 0.906 0.859 1.103 0.970 0.807 0.917 0.948 420-449 0.852 0.695 0.783 0.791 0.749 1.264 1.077 0.878 0.990 0.947 450-479 0.875 0.707 0.805 0.820 0.769 1.220 1.068 0.863 0.982 0.939 480-520 0.818 0.679 0.768 0.835 0.787 1.198 0.980 0.813 0.920 0.943 H = humerus; F = femur; R = radius; U = ulna; T = tibia; f = fibula. CHL groups in mm.

PAGE 65

51 Table 33. Ratio of long bone lengths to femur and tibia length for "black" sample (after Armelagos et al, 1972). CHL H/F R/F U/F T/F f/F F/T H/T R/T U/T f/T group 0.970 210-239 0.823 0.909 0.860 0.823 1.163 1.128 0.957 1.057 0.957 240-269 0.957 0.781 0.871 0.796 0.758 1.257 1.203 0.981 1.095 0.952 270-299 0.943 0.793 0.870 0.847 0.810 1.181 1.114 0.936 1.028 0.957 300-329 0.902 0.742 0.837 0.828 0.781 1.208 1.089 0.896 1.011 0.943 330-359 0.909 0.746 0.847 0.849 0.806 1.178 1.070 0.879 0.997 0.949 360-389 0.910 0.765 0.860 0.857 0.806 1.167 1.062 0.893 1.003 0.940 390-419 0.859 0.722 0.810 0.863 0.806 1.159 0.996 0.837 0.939 0.934 420-449 0.873 0.713 0.798 0.915 0.845 1.093 0.954 0.780 0.873 0.924 450-479 ... ... 480-520 1.059 0.831 0.960 0.887 0.823 1.127 1.193 0.936 1.082 0.927 H = humerus; F = femur; R = radius; U = ulna; T = tibia; f = fibula. CHL groups in mm. 70.00 J 60.00 50.00 .. t 40.00 c •^ 30.00 4. s g 20.00 + 10.00 -0.00 0^ CO s OS ON cs i CO ON CO 00 CO I o CO CHL group Figure 18. Comparison of "Black" and "White" means for humerus length by CHL group.

PAGE 66

52 60.00 -r CHL group Figure 19. Comparison of "Black" and "White" means for radius length by CHL group. CHL group Figure 20. Comparison of "Black" and "White" means for ulna length by CHL group.

PAGE 67

53 CO =5 IN 0^ ON CO O CM CS o o CO On ir> fO I o en en ON 00 CO I o CO I o ON CO ON I o Cvt ON IN '^ I o in o tN ITi I O 00 CHL group Figure 21. Comparison of "Black" and "White" means for femur length by CHL group. 70.00 J 60.00 -S 50.00 .6 •^ 40.00 -•5 g 30.00 -3 20.00 -10.00 .0.00 -• "Blacks" "Whites" ON CO CN| 6 On NO fN) On ON On ON ON ON ON o On CM in 00 T— 1 '^ t^ CN| CN CO CO CO Tt" -* m o O o o o o o o p: o CO NO ON CM in 00 CM CO CO CO CO >* •^ CHL group Figure 22. Comparison of "Black" and "White" means for tibia length by CHL group.

PAGE 68

54 70.00 J 60.00 -I 50.00 j: 40.00 + bC J 30.00 4I 20.00 -10.00 -0.00 -"Blacks" "Whites" + ?5 SO CN On CN CO in 9 On X> CO ON I-H 'St in 6 i-i CN 6 CM CN o o CO 6 CO CO CHL O CO group 6 CJ\ CO o CM s 6 00 Figure 23. Comparison of "Black" and "White" means for fibula length by CHL group. 90.0 -r 80.0 .. g 70.0 1 60.0 to 50.0 c £ 40.0 1 30.0 X 20.0 •^ 10.0 0.0 1— 100.0 + + + H 200.0 300.0 400.0 500.0 Crown-heel length in mm. 600.0 700.0 Figure 24. Bivariate plot for CHL versus humeral length in raw data space (n=248)

PAGE 69

60.0 J 50.0 -S 40.0 -.5 X, bc 30.0 + c •I 20.0 10.0 + 0.0 100.0 55 •^•*i X'i • • + 4+ 200.0 300.0 400.0 Crown-heel length in mm. 500.0 H 600.0 Figure 25. Bivariate plot for CHL versus radial length in raw data space (n=231) 70.0 J 60.0 -I 50.0 -•§ 40.0 •5 S 30.0 4^ 20.0 .. 10.0 -0.0 100.0 + + + + 200.0 300.0 400.0 500.0 Crown-heel length in mm. 600.0 Figxire 26. Bivariate plot for CHL versus ulnar length in raw data space (n=231)

PAGE 70

56 90.0 J 80.0 .. 6 70.0 ^ 60.0 .5 £ 50.0 + to Ji 40.0 .o 30.0 6 tS 20.0 -10.0 -. 0.0 100.0 Q + + 200.0 300.0 400.0 Crown-heel length in mm. 500.0 H 600.0 Figure 27. Bivariate plot for CHL versus femoral length in raw data space (n=195) 70.0 J 60.0 -6 50.0 ..S 40.0 -60 S 30.0 a 20.0 .. 10.0 0.0 100.0 14^ + • • •• • • •• 444-\ 200.0 300.0 400.0 500.0 Crown-heel length in mm. 600.0 Figure 28. Bivariate plot for CHL versus tibial length in raw data space (n=142)

PAGE 71

57 70.0 J 60.0 -I 50.0 -.s £ 40.0 + wo ^ 30.0 I 20.0 10.0 -0.0 100.0 •• + + 200.0 300.0 400.0 Crown-heel length in mm. 500.0 600.0 Figure 29. Bivariate plot for CHL versus fibular length in raw data space {n=121)

PAGE 72

CHAPTER 5 DETERMINATION OF GESTATIONAL AGE A basic technique in osteology is determination of age at death. There is a vast literature on determination of age from analysis of the skeleton and dentition (e.g., McKern, 1970; Kerley, 1970, Krogman 1962; Todd, 1939). These studies examine the maturation, degeneration, or metamorphosis of various aspects of the skeleton to arrive at a probable age at death. The investigator calculates a skeletal age and chronological age since birth is assumed to fall within an expected range. Techniques for determining the age of fetal or perinatal remains are more difficult because they are designed to calculate the time since conception. In most cases, the time of conception cannot be known with certainty. Many studies calculate gestational age based on the mother's last normal menstrual period (LNMP) with the assumption that conception usually occurs within one week of the last menstrual period, however an error of as much as two weeks is possible (Naeye and Dixon, 1978). Additionally, errors in reporting occur, such as when implantation bleeding is mistaken for the last menstrual period. Even in cases of isolated coitus, fertilization may not take place immediately and one must rely on the history provided by the mother (Peterson et al., 1989; Golbus and Berry, 1976). The possibility of miscalculating gestational age presents a problem for growth and development studies. Several methods have been used to omit infants that are older than their calculated gestational age. Authors have excluded outliers in bimodal curves and scattergrams (Gruenwald, 1966; Lubchenco et al., 1963), used maternal examinations and histories (Babson et al., 1976; Brermer et al., 1976), or considered clinical tests of a neonate's mattirity (Babson et al., 1976; Brermer et al., 1976; 58

PAGE 73

59 Gruenwald, 1966; Lubchenco et al., 1963) in order to establish the validity of gestational ages based on LNMP. Exclusion of fetuses and infants that are younger than their calculated gestational age is more problematic. Isolated cases may be growth retarded but present no other signs of pathology. Manipulation of data in order to arrive at "valid" gestational ages is a circular process, and data are no doubt excluded that would broaden the variability of normal fetal growth. The close correlation between CHL or crown-rump length (CRL) and long bone length to period of growth has been used to determine fetal developmental age. Estimation of developmental age in fetuses is used to predict the expected birth date and diagnose the presence of pathologies that affect normal growth. It is also an issue in cases of forensic identification, or when the determination of viability is necessary for legal disposition of death (Weaver, 1986; Kuroda, 1970). In populational studies, differentiation between infants and fetuses is an important factor in paleodemographic reconstructions, providing clues about abortion rates, maternal deaths during childbirth, and early childhood disease (Mensforth, 1985; Jantz and Owsley, 1984; Hummert and Van Gerven, 1983; Merchant and Ubelaker, 1977; Armelagos et al., 1972; Johnson, 1961, '62, '68, '69; Walker, 1968). Current methods for determining gestational age in living fetuses are based on sonographic imaging to determine head size, long bone length, or femoral head cartilage diameter (Mashiack, Blankstein and Stem, 1975; Jeanty and Romero, 1984; Lange and Manning, 1984). Earlier studies utilizing radiography (Christie, 1950) have been abandoned because of the risks of radiation exposure to the mother and fetus. In the case of premature delivery or abortion, Olivier and Pineau (1958, 1960) have derived linear regression formulae based on CHL to determine gestational age. These studies are currentiy used by pathologists to deternune developmental age of intact, fleshed fetuses.

PAGE 74

60 Trotter and Gleser (1952, '58, '77), Genoves (1967), Hoffman (1979) and others have generated stahire formulae based on long bone lengths of adults and children. Anthropologists have found secular and genetic differences in proportionaUty and stature between populations, so formulae for predicting stahire have been derived for different populations. Similarly, several factors may affect stahire and long bone length in fehises and differences in growth rates may exist between populations. Genetic diversity, secular hends related to nutrition or access to medical care, or the environment (particularly high altitiide) may make a given population unsuitable as a standard for other populations. Numerous data (Toldt 1879; Zangemeister 1911, '12, '17; Dieti-ich 1925; Scammon and Calkins 1929; Guthmann and Knoss 1939; Gartiter 1947; Brock 1954; Olivier and Pinneau 1958, 1960) confirm the correlation between the duration of pregnancy and the development of the fetiis. While the exact time of conception is not known with certainty in the vast majority of cases, the preponderance of data using "estimated" time of conception based on different criteria suggest that there is a close correlation between linear lengtii and gestational age. By means of "Haase's rule," the age of the fehis can be determined from its body length. It is known that until the 5th lunar month the fetal body length in centimeters can be closely estimated by squaring the number of the months of pregnancy, and after this time by multiplying the months by 5 (Fazekas and Kosa, 1978; Brock, 1954). Brock (1954) published data on how the values of body length of fetiises derived from known periods of pregnancy correspond to tiiose obtained by Haase's rule. For comparison Brock used the data of Stireeter (1920, 1948, 1949), and Scammon and Calkins (1929). Table 34 (from Fazekas and Kosa, 1978) below shows that there are no significant differences between the achial data and tiie values obtained by Haase's calculation. Since Haase's rule is curvilinear for the first 5 lunar months the values

PAGE 75

61 obtained are somewhat high for the first 3 months of pregnancy, whereas in the later lunar months they nearly coincide with values of the actual body length of the fetuses. Table 34. From Fazekas and Kosa's Table 7, pp. 31 Comparison of values of fetal body length given by Scammon and Calkins with the values obtained on the basis of Haase's Rule (after Brock). Period of pregnancy calculated from the last menstruation (weeks) Length of fetus (cm) Increase in length of fetus in 4-week periods (I.I.) in mm. Increase in length of fetus in 4-week periods (R.I.P.) Length obtained on the basis of Haase's calculation 8 12 7.0 ... ... 4 9 16 15.5 8.5 122 16 20 22.7 7.2 46 25 24 29.2 6.5 29 30 28 35.0 5.8 20 35 32 40.4 5.4 15 40 36 45.4 5.0 12 45 40 50.2 4.8 11 50 LI. = incremental increase in mm.; R.I.P. = relative increase in percent. The most often used method of calculating developmental age in bioarcheological and forensic contexts requires the measurement of dry bone (Fazekas and Kosa, 1978). However, the Fazekas and Kosa sample is small (n=138) and consists of a homogenous eastern European sample which may be inappropriate for use in determining developmental age in contemporary Americans. Furthermore, the measurement of dry bone requires skeletal preparation and may be uimecessarily invasive. Huxley (1996) has recently addressed the problem of dry bone shrinkage, which may add an additional sotirce of error. Fazekas and Kosa grouped their fetuses on the basis of body length, not "period of pregnancy," because of their inability to determine the exact time of conception. They

PAGE 76

62 used "Haase's Rule" which they feel has been shown by other researchers (Dietrich, 1925; Scammon and Calkins, 1929; and Streeter, 1920, 1948, 1949) to be very accurate when compared with gestational age based on LNMP. They divided the fetuses in different groups as follows: Fetuses 40 cm. long were included in the age group of 8 lunar months. In the same group were fetuses 39 and 41 cm. long, whereas those with a body length of 42 and 43 cm. (bom at the begirming of the 9th Ivmar month) were included in the age group of 8.5 lunar months. So, lunar months were classified by the fetal lengths listed in Table 35. Table 35. Months of gestation and equivalent CHL (from Fazekas and Kosa, 1978) Months of gestation Equivalent CHL length 5 limar months 24-26 cm. 5.5 lunar months 27-28 cm. 6 lunar months 29-31 cm. 6.5 lunar months 32-33 cm. 7 lunar months 34-36 cm. 7.5 lunar months 37-38 cm. 8 lunar months 39-41 cm. 8.5 lunar months 42-43 cm. 9 lunar months 44-46 cm. 9.5 limar months 47-48 cm. 10 lunar months 49-51 cm. It is xmclear how fetuses less than 5 limar months were categorized according to CHL. Fazekas and Kosa used the following table (Table 36) to compare data from two previous studies to estimations from Haase's rule. The body length as calculated by the

PAGE 77

63 use of Haase's Rule compares favorably to the data of Dietrich (1925), and almost exactly coincides with the data of Scammon and Calkins (1929). Table 36. Fazekas and Kosa's Table 8, pp. 31: Growth in length and increase in weight with respect to fetal age (in lunar months). Period of Body length of Body length of Body length of pregnancy fetuses (cm) fetuses (cm) fetuses (cm) (lunar months) Dietrich Scammon & Calkins Haase's calculation 2 3.0 3 9.8 7.0 4 18.0 15.5 5 25.0 22.7 6 31.5 29.2 7 37.1 35.0 8 42.5 40.4 9 47.0 45.4 10 50.0 50.2 post-term ... 4 9 16 25 30 35 40 45 50 54 The data from the current study have been compared with those of Fazekas and Kosa by converting their data into CHL groups. Table 37 shows Fazekas and Kosa's data in terms of incremental increase and relative percent increase of long bone growth. My data follows in Table 38. 1 have compared the mean long bone lengths to my data (Table 39), as well as plotted long bone growth relative to CHL for their data (Figures 30-35). It is clear that Fazekas and Kosa's data serves as a vaUd reference population for determining gestational age in the United States. Results obtained by my radiographic method correspond with the dry bone measurements of Fazekas and Kosa and, therefore, the method can be used in instances where skeletal preparation is impossible or vmdesirable.

PAGE 78

64 13 C TO 1/1 )-H OJ •4-t 0) a; en O) u C 0! 0) TO T3 00 IN On O -^ OJ N til a, IN o ^ & -O l-TH to h1-( iZ 6 2 ^ H 01 en en 3 E P< k; a, O u CO so O ON o o NO 00 CD ON CO CM ON ON CO CS CO ON 00 CO in CM CO o IN ON CO t^ CO ON CN CO ON IN o T-H ID O T-i NO On d K IN 00 in o NO 00 <*< T— t ^ ON 00 CO CM cs 1— 1 T— < o^ o s cs 00 o o CO T-H ON NO o in IN IN On CM CO LO IN in in CO ^ CO CO in Tti t— ( IN 00 o ON NO CO in CO •<*< 'i* 00 o in ON o NO CN| in 00 00 1—1 IN o >-H in O ^ ^ CO CO * CM c^ rj< CM CM s CO T-H (N rH o in >^ I-H ON O CO CO CO 00 o in CM IN rH O T-H ON T-H in ON CO CO CM 00 o IN CO CO 00 o o CO in CO o o ^ fx CM ON IN CO in ON o in o IN CO CM CM ON CM O IN O O 00 rH O IN CO CM rH CM O NO CM -"t NO 00 o ON d 00 ON CO CO NO o CM o CO CO ON CO o in ON in 00 o o o v,D in CO T-< O T-l ON CO vO ON On ON On On C3N ON ?" o NO On CN in 00 T-H Tf In (N CM CM CO CO CO '^ t •^ in O O o o O o o O 6 ^ t^ o CO NO On CM in 00 ^ CN CO CO CO CO • 6 6 6 o O T-H >* IN o CO NO ON CM in CM CM (N CO CO CO CO >* ^

PAGE 79

65 8 u •a C I a> Ch Q^ HH C CO es 01 re :d ? en QJ 3 CN 1— 1 t SO IN IN '^ in CO in T— 1 o ON o o rji IN T-H o o 00 O "* T-H IT) CNj CO SO CO CO T-H TjH CO On CO 00 m CO T-H o d ^ T-H IN T-H CO so o T-H ON o CO CM cs 00 T-H 00 T-H O T-H in o IN IN tN T-H T-H T-H CO CO O ^ ^ r4 \r> CO o d ttH ON IN CS un q ID o 00 d q IN CO r-4 d T-H ON o T-H T-H 00 o ON •r-< NO o -* 00 SO T-H In ON SO o Cn IN T-H so so CO in so CO CO ^ ^ O^ ^ d ON On in so CO ^H in CNl IN rH O TjH -^jt in T-H tx) in TJH CO CS ON CO CO On in o d CO CO in d CN 00 ON o o so CO in T-H O IN cn4 IN o so csi ON o CO in d rsi in CO so CO CO SO CO 00 OS On On O O in 00 o so SO >* CO CO so in On >* SO CnI O CJN o in 00 ON cs o so CO o o 00 00 CN 00 CO CN SO ^ o o 00 T-H I-H tN T-H CO ON ON o CO CO so o CN SO ON o IN Oh 3 O u bO u ON so CN) o CM On CO ON ON C^ I CN ON SO IN I o CN CO o o CO CJN CJN CN I o IN CN ON in CO 6 CO CO ON o) CO I o o CO ON 00 CO 1 o NO CO CJN in CO I o CO CO o C3N CO OS 00 CO o so CO OS I o (N o ON CO OS IN *< I o in On I O cs o CN| in o 00 ON IN I o in

PAGE 80

66 (0 v. o en C O > CO T3 en C to bO C _q; C (0 00 On D tC en 3 -D S Ui a; S D 5 ^ i § a, O ;-i bO X u 00 00 (N CO 00 ON fN CN CN CO CO 00 in itH CS CN CN (N CN CN T— 1 CNl rg in CO ON T— 1 CO 00 CN 1—1 CM (N v£5 (N 00 (N o CO CO 00 CO CN o in 1—1 in ON in 1— 1 C3N T— 1 CO 00 o ON 00 CN o CO ON ON CM 1— 1 (N CO CNl CN 1—1 CN CN CN CO CO 00 CO CO On CO NT) ON CO CO q NO o IN. IN CN CO oo ("nI NO ON CO ON CO ^ CN CN CN CO CO CO 't ^ un in NO in in ON ON CM CN ON ON 1—1 CN CO (N ON CM ON in On NO o r— 1 1—1 fNj (N CO CN 1—1 CN CM ^ 00 00 in in 00 ON CO O NO 00 ON o Cs| in o NO NO (N CN CO in CO o 00 in in o NO CM NO 1—1 tN ON in ON 00 C?N CO in NO On o CN in in in NO in in o 1—1 1—1 1—1 1-H CM 1— 1 CN CM 1—1 CM CN rH ON NO 00 NO 1—1 NO in CN CO NO o On 00 CM CO CN CO r-1 ON CN 1— ( CO in CO 00 CO CN ON r— 1 in IN in CM in CN H 1—1 4 NO CN ON in o 1—1 ON CN 00 00 On CM T— 1 ON CTs H 1—1 1—1 CNl 1— 1 t-i 1—1 1—1 CM 1—1 H O •> in NO o CN ON in o NO ON o CO CO o 00 1* 00 H NO g CM 00 CN CN CO CO IN. CO CO in On CO o 00 CO 1—1 00 IS. in CO 00 ts ON 00 in ts o 1—1 1—1 1—1 (N CM CN 1—1 1—1 CM CO 1—1 ON 00 o T— 1 CO CO CO ON rH in NO in ON in CN in rH On in CN CN CO CO 00 CO CN in o in CO in in in 1—1 NO On ON ON On On ON ON ON On o CO NO ON CM in 00 1—1 1* In CM CN CM CM CO CO CO 1* "* l^) O O o o o o o o O o 1—1 •<^< rv. o CO NO ON CN in 00 (N CN CM CO CO CO CO 5t ^

PAGE 81

67 70.00 -r 60.00 -Warren 'Fazekas and Koza o On o o Gestational age in lunar months Figure 30. Comparison of cross-sectional humeral growth curves of the current study and the Fazekas and Kosa (1978) sample. 60.00 TGestational age in lunar months Figure 31. Comparison of cross-sectional radial growth curves of the current study and the Fazekas and Kosa (1978) sample.

PAGE 82

68 Warren Fazekas and Koza o ON ITl o o Gestational age in lunar months Figure 32. Comparison of cross-sectional ulnar growth curves of the current study and the Fazekas and Kosa (1978) sample. Gestational age in lunar months Figure 33. Comparison of cross-sectional femoral growth curves of the current study and the Fazekas and Kosa (1978) sample.

PAGE 83

69 in o in o in o in o in o in O "* in in >£) VD t^ t^ 00 00 CTv C3N O Gestational age in lunar months 'igure 34. Comparison of cross-sectional tibial growth curves of the current study and he Fazekas and Kosa (1978) sample. 70.00 -r 60.00 6 50.00 B ;^ 40.00 + J 30.00 .I 20.00 .. 10.00 + 0.00 Warren Fazekas and Koza in o in o in o in o in o in o . 00 00 ON ON o Gestational age in lunar months igure 35. Comparison of cross-sectional fibular growth ciirves of the current study and le Fazekas and Kosa (1978) sample.

PAGE 84

m 70 Table 40. Ranges for upper limb lengths by CHL All measxirements are in millimeters. group and "equivalent" gestational age. CHL group Equivalent Gestational Age in limar months Humenis n Radius n Ulna n 210-239 <5 24.0-31.0 10 19.5-26.0 9 22.5-29.0 9 240-269 5.0-5.5 26.0-47.0 14 21.0-37.0 14 23.5-42,5 14 270-299 5.5-6.0 29.0-42.0 35 24.0-39.5 33 28.0-39.5 34 300-329 6.0-6.5 31.5-47.0 38 25.5-38.5 38 29.5-44.0 38 330-359 6.5-7.0 37.0-49.0 38 31.0-40.0 37 34.5-45.0 37 360-389 7.0-7.5 42.0-54.0 16 34.0-45.0 16 39.5-51.0 16 390-419 7.5-8.5 46.0-69.0 18 38.0-55.0 16 43.0-67.0 15 420-449 8.5-9.0 52.0-61.0 11 42.0-50.0 11 48.0-57.0 11 450-479 9.0-9.5 44.0-69.0 16 36.0-58.0 13 42.0-67.0 13 480-520 9.5-10 55.0-78.5 14 45.5-55.5 19 54.0-63.5 19 :'|" Ml I: i Table 41. Ranges for lower limb lengths by CHL group and "equivalent" gestational age. All measurements are in millimeters. CHL group Equivalent Gestational Age in lunar months Femur n Tibia n Fibula n 210-239 <5 24.5-33.0 10 20.5-29.0 10 20.5-28.0 8 240-269 5.0-5.5 28.0-51.0 13 24.5-35.5 11 23.5-33.5 11 270-299 5.5-6.0 29.0-46.0 35 24.5-38.5 18 23.5-37.0 13 300-329 6.0-6.5 32.0-56.0 36 29.0-42.5 18 27.0-40.0 13 330-359 6.5-7.0 39.0-55.0 40 35.0-47.0 17 33,0-44.5 15 360-389 7.0-7.5 47.0-57.0 14 40.0-49.5 11 39.0-47.0 9 390-419 7.5-8.5 52.0-75.5 16 47.0-63.0 9 47.0-60.0 7 420-449 8.5-9.0 55.0-73.0 6 48.0-59.0 3 34.0-54.5 3 450-479 9.0-9.5 49.0-73.0 6 42.0-58.0 4 39.0-55.0 4 480-520 9.5-10 62.0-81.0 5 55.0-69.0 5 51.0-64.5 4 i^e:

PAGE 85

71 Tables 40-41 (above) again present the range of measurement of the long bones for each CHL group and its equivalent gestational age. Gestational age may be determined by using the regression formulae in Chapter 3 to estimate CHL, then using prior studies to arrive at a gestational age based on CHL. Hov^ever, the formulae are only valid for the reference population from which they were derived, and should be applied with caution in forensic or bioarcheological contexts dealing with other populations.

PAGE 86

CONCLUSIONS AND SUMMARY Measurements of the six long bones of the extremities were taken from a series of fetal radiographs. Additional data was collected from associated autopsy protocols. I divided the sample arbitrarily into groups of similar CHL, and thus, similar developmental age in order to establish the relative growth and proportionality of the limb bones. All long bone lengths correlated significantly with CHL (r2> 0.8375; p < 0.05). This relationship is consistent with the relationship of adult long bones to stature (Genoves, 1967; Trotter and Gleser, 1958). I provide least-squares linear regression formulae for predicting CHL from radiographic bone lengths. The relative linear growth of the long bones was also consistent with other published data (K6sa,1989). Distance curves show that the femora elongate relative to the humeri during development as a result of greater incremental increase and relative increase in percent over the entire fetal period. The high intermembral index is a result of the relatively short lower limbs of the fetus. The intermembral index decreases for successive CHL groups. A decreasing intermembral index during ontogeny has occurred as an adaptation for bipedal locomotion and is a unique feature of hominid evolution. The growth velocities for this study are consistent with postnatal studies through the natal period up to 6 months. Growth then accelerates during the period from 6 months to 1 year (Maresh, 1955). Previous fetal growth studies have failed to adequately address the impact of pathology on linear growth and proportionality. Extensive autopsy data have enabled me to divide my sample into "pathological" and "normal" sets for comparison. My data 72

PAGE 87

73 show that long bone proportions and relative growth are not significantly affected by prenatal pathology. The lack of effect of pathology on long bone length is consistent with other studies that find girth but not length to be affected by congenital conditions (Richards and Anton, 1991). The proportional relationship between long bone length and CHL is stable and predictable. Therefore, fetal abortus may serve as a model for normal linear growth and proportionality. Bagnall et al. (1978) found no significant difference in long bone length or limb proportions between males and females of 8 to 24 weeks gestational age. My data confirms that the same is true throughout the fetal period. Sex cannot be determined by proportional indices. Race differences in fetal long bone proportions are found to mirror those in adults. Self-identified blacks had signiticantiy higher brachial and crural indices than whites, although there is considerable overlap in ranges. The wide overlap of group values demonstrates the range of variability characteristic of "racial" traits. Assigning ancestral group or affinity to a fetiis based on brachial and crural indices is tenuous at best. Li order to establish the gestational age of a fetus based on long bone length, one must first accept the validity of earlier studies that correlate CHL with calculated gestational age based on last normal menstrual period (LNMP), isolated coitiis, stage of organ development, or backward extrapolation from the birth date. The ability of clinicians to accurately predict, in most cases, the expected birth date within 1 week is fairly convincing, but errors of several weeks do occur in individual cases. Previous studies have shown that actual linear measurements closely correspond with predicted values via Haase's Rule. Dry bone measurements based on the data of Fazekas and Kosa (1978) have traditionally been used by anthropologists working with fetal remains. My data corresponds closely with that of Fazekas and Kosa. This shows that (1) relative linear growth is similar between the two samples, and (2) radiographic measurements of

PAGE 88

74 fetal long bone length are accurate and the radiographic method may be used when skeletal preparation is impossible or undesirable.

PAGE 89

APPENDIX A INSTITUTIONAL REVIEW BOARD EXEMPT STATUS

PAGE 90

76 ^%s UNIVERSITY OF 'FLORIDA Health Center Institutional Review Board MEMORANDUM DATK: January 29, 1997 Michael W. Warren y-^ /n 2927 NW 10th Place Gainesville£/'FL yf'4^6^5 R. Peter lafrate, PharmD y S ^' ^ Chair, IRB TO: FROM: PO Box 100173 Gainesville, FL 32610-0173 (352) 846-1494 Fax (352) 846-1497 SDBJ: EXEMPTION ( IRB# 033-97) entitled: Linear Growch and Proportionality in Fetuses and Suxll uorns ; Ai'i American Contemporary Population Your exemption request has been reviewed by the Chair of the IRB and was APPROVED on January 27, 1997. Your project has been assigned an IRB number; please refer to this number in future correspondence. Enclosed is a copy of your exemption request with the IRB approval stamped on it. If this project changes or otherwise requires IRB review and approval, you have the explicit responsibility to pursue IRB review at that time. In approximately one year you will be contacted by the IRB office and asked to provide some follow-up information (e.g. whether your project is still active; whether any changes have been made since you submitted your initial request) Your exemption was granted under the following checked category of exempt research using human subjects. (1 #1 Commonly accepted educational settings involving' normal educational practices [1 #2 Educational tests (cognitive, diagnostic, aptitude, achievement), survey or interviews or observing public behavior [1 #3 Research involving the use of educational tests, survey or interview procedures, or observation of public behavior that is not exemptunder 2 above, if the subjects are public officials or candidates for public office or a federal statute requires that the confidentiality will be maintained throughout the research and thereafter. #4 Collection or study of existing data, documents, records, pathological or diagnostic specimens, if these rose ::ro publicly available or if the inforyjition is recorded by the PI in such a msn-ner that S'.ib-iect£ can.-.ot be identified, directly or through identifiers linked to the subjects. n #5 Research and demonstration projects for public benefit. [] #S Taste and food quality evaluation and consumer acceptance studies. Thank you for keeping the IRB informed about your research, thereby allowing us to keep accurate files. If the IRB staff can be of any further assistance, please feel free to call. end ... cc Division of Sponsored Research IRB file Equal Opportunity / Affirmative Action Institution

PAGE 91

APPENDIX B RAW DATA RECORDED FROM AUTOPSY PROTOCOLS

PAGE 92

Series # CHL Weight Sex Race Mother's age Cause of death/comments 379 ... 400 f b ^ Premature rupture of membranes; sepsis; no congenital malformations 380 285.0 425 f w 27 Hydrocephalus; induced abortion 381 325.0 475 m w 17 No congenital malformations; anoxic stess unknown etiology 382 445.0 1425 f b 16 No congenital malformations; fetal death in utero of unknown cause 383 ... 475* f b 24 Amniotic band syndrome; no congenital malformations 384 485.0 3200 m b 20 Anencephaly 385 360.0 893* f b 29 Placental infarctions 386 310.0 650 m w 26 Twin "A" unknown cause of death in utero 387 315.0 750 m w 26 Twin "B" identical twin to above; no congenital malformations 388 270.0 400 m b ... Chorioamnionitis; intraventricular hemorrhage 389 300.0 400 m b 14 No congenital malformations; intraventricular hemorrhage 390 265.0 168 m b 16 Abruptio placenta; hyaline membrane disease; immaturity 391 230.0 235 m w 17 No congenital malformations; fetal death in utero; unknown cause 3^ ... 822 m w 17 Immature lungs; intraventricular hemorrhage 3^ 340.0 850 m b 19 Sepsis; patent ductus arteriosis 394 360.0 1575 m w 36 Premature rupture of membranes; oligohydraminos 395 235.0 313 f w 23 Ventral wall defect of abdomen; elective abortion 396 ... ... ..• ••• Anencephaly 397 ... 686 f w 23 Amniotic band syndrome with multiple deformities 398 ,.. 1210 f b 23 Twin congenital syphilis oo

PAGE 93

Series # CHL Weight Sex Race Mother's age Cause of death/comments No congenital malformations; history of maternal polyhydraminos No congenital malformations; marginal placental abruption No congenital malformations; placental infarction Twin "A" abruptio placenta; no congenital malformations Twin "B" abruptio placenta; no congenital malformations Turner's syndrome; nuchal lymphangioma No congenital malformations; acute chorioamnionitis Hydrops fetalis; severe maceration Placental infarction; pulmonary hypoplasia Placenta previa; placental infarction; no congenital malformations Maternal eclampsia; intraventricular hemorrhage No congenital malformations; placental infarction No congenital malformations; chorioamnionitis; pneumonia No congenital malformations; marginal cord insertion Cystic hygroma; hydrops fetalis; induced AB No congenital malformations; placental insufficiency; unknown Caudal regression syndrome; sirenomelia Fetal death in utero of unknown cause; no congenital malformations Oligohydraminos; intrauterine growth retardation Maternal cocaine abuse; fetus positive for cocaine metabolites 359 370.0 750 f b 21 360 260.0 275 f w 21 361 353 m b 20 362 300.0 800 m w 25 363 315.0 805 m w 25 364 205.0 287 f w 31 365 325.0 950 m b 18 366 1050^ f w 22 367 250 m w 368 484* f w 20 369 335.0 825 f w 30 370 1688 f b 19 371 300.0 525 f b 36 372 345.0 Sll'^ f b 28 373 260.0 862* f w 30 374 260.0 440 m w 25 375 250.0 220 m w 22 376 ... 150 f b 18 377 ... 319* f b 32 378 675 m b 29

PAGE 94

Series # CHL Weight Sex Race Mother's age Cause of death/comments 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 327.0 440.0 318.0 280.0 475.0 350.0 410.0 340.0 650 500 700 640 620* 2160 780 250 1060* 580 455 685 2040* 400 501 720 1760 480* 1000 m m f m f m f f f f m m f m f f f f m b w w w w w w w b w w w w b b b b b b 18 Abruptio placenta; no congenital malformations 19 Prematurity; hyaline membrane disease; 30% placental abruption 42 Downs syndrome Trisomy 21; induced abortion 29 Abruptio placenta; premature rupture of membranes 20 No congenital malformations; fetal death in utero unknown cause 26 Congenital cytomegalovirus; hydrops; hydrocephaly 15 Anencephaly 35 5 p-chromosome abnorm.; no gross abnormalities detected 22 Pulmonary hypoplasia; assymetric growth retardation Anencephaly 40 Chorioamnionitis; pneumonia; no congenital abnormalities 35 Sacral meningocele 2nd to spina bifida; induced AB "Intrauterine growth retardation"; oligohydraminos 30 No congenital malformations; maternal diabetes 17 No congenital malformations; necrotizing chorioamnionitis 26 Birth trauma from footling breech; non-viable immaturity 30 Materal pre-eclampsia; bronchial dysplasia 24 Abruptio placenta; ; no congenital malformations 29 Placental insufficiency; growth deficiency 21 No congenital malformations; history of maternal cocaine abuse 00 o

PAGE 95

Series # CHL Weight Sex Race Mother's age Cause of death/comments Premature rupture of membranes; non-viable immaturity Prematurity; intraventricular hemorrhage No congenital malformations; non-viable immaturity Immature; macerated; maternal diabetes; hydrocephaly Abruptio placenta; no congenital malformations Maternal hx of habitual ABs; marginal abruption of placenta Twin immaturity; interstitial emphysema Chronic villitis; no congenital malformations; immaturity Tvkfin immature; interstitial emphysema Renal agenesis; hypoplastic lungs; oligohydraminos deformations No congenital malformations; possible placental infarction No congenital malformations; premature labor No congenital malformations; immaturity; macerated Twin immaturity; premature rupture of membranes Twin intraventricular hemorrhage No congenital malformations; hemolytic disease; chorioamnionitis Hydrocephalus Maternal pulmonary embolism, treated with anticoagulant Cystic kidney disease; anuria; pulmonary hypoplasia Asplenia; no other abnormalities; intraventricular hemorrhage 319 830 m b 23 320 1110 f w 27 321 860 m w 322 270.0 380* f b 18 323 320.0 720 m b 29 324 310 m b 28 325 345.0 700 m w 26 326 230.0 225 f w 19 327 370.0 980 m w 26 328 480.0 3240 m w 24 329 310.0 530 m w 17 330 320.0 610 m w 27 331 400* m w 28 332 290.0 650 f b 25 333 290.0 690 m b 25 334 335.0 650 m w 23 335 1900 f w 24 336 3030 m w 16 337 ... 2600 m w 26 338 250 f w 16 00

PAGE 96

Series # CHL Weight Sex Race Mother's age Cause of death /cominents 299 375.0 1050 m b m 300 ... 720 m w 25 30X ... 3175 f w ... 302 ... 660 f w 20 303 370.0 830* f b 16 304 250.0 200* f b 32 305 365.0 1050 f w 37 306 *.. 2990 f w 37 307 245.0 280 f w 22 308 252.0 275* m w 30 309 270.0 550 m w 18 310 490.0 2520 f w 18 311 340.0 1000 n\ w ... 312 ... i w 18 313 ... 2270* m w 24 314 280.0 400 f b 25 315 290.0 580 f b 29 316 340.0 840 f b 27 317 414.0 2500 m w 20 318 280.0 490 m w 28 No congenital malformations; prematurity; placental infarction Premature rupture of membranes; amniotic band syndrome Anencephalic No congenital malformations; intraventricular hemorrhage Placental infarction; fetal death in utero Twin no congenital malformations; possible placental infarction Abruptio placenta; cleft palate and other malformations Hydrops fetalis; hypoplastic lungs Non-viable immaturity; unknown cause of death Twin placental thrombosis; monochorionic/diamniotic placenta Anencephaly; pregnancy terminated Hydrocephalus; multiple anomalies; maternal use of antibiotics Moderate hydrocephaly; pregnancy terminated Trisomy 18; multiple congenital anomalies No congenital malformations; fetal death in utero Intraventricular hemorrhage; chorioamnionitis; marginal abruption Massive abruptio placenta Premature rupture of membranes; pulmonary hypoplasia Sepsis; meningitis; pneumonia Mosaic Downs syndrome; elective termination of pregnancy

PAGE 97

Series # CHL Weight Sex Race Mother's age Cause of death /comments No congenital malformations; unknown cause of spontaneous AB No congenital malformations; placental infarction Potter's syndrome Abruptio placenta; fetal death in utero; no congenital malformations Severe hydrops fetalis No congenital malformations; premature ruptured membranes; sepsis Maternal hypertension; pre-eclampsia; no congenital malformations No congenital malformations; unknown cause of death Twin "A" Polyhydraminos; prematurity Twin "B" Polyhydraminos; prematurity Maternal schizophrenia; no congenital malformations Maternal hypertension; fetal death in utero; macerated Sepsis; intraventricular hemorrhage Advanced maternal age; marginal placenta; maternal hypertension Twin placental thrombosis; monochorionic/diamniotic placenta Abdominal pregnancy; maternal syphilis Trisomy 18; hyaline membrane disease; congenital heart disease Bronchopneumonia; immaturity; ventricular hemorrhage Multiple anomalies suggestive of chromosome abnormality Potter's syndrome; polyhydraminos 279 275.0 845* m b 280 365.0 790* m w 29 281 415.0 1950 m w 29 282 1430 m b 20 283 448.0 3625 f w 22 284 405.0 1340 m w 19 285 570 f b 19 286 465.0 2050* m b 18 287 740 f w 18 288 820 f w 18 289 ... 1800 m b 22 290 3420* m w 22 291 355.0 960 f w 25 292 350.0 940* m w 40 293 235.0 310* m w 30 294 275.0 400* f b 295 1100 f w 296 315.0 530 f w 23 297 290.0 560 m b 23 298 550.0 2580 f w 27 00 00

PAGE 98

Series # CHL Weight Sex Race Mother's age Cause of death /commer\ts 259 415.0 1030* m b 260 335.0 765 f w 16 261 320.0 510 m w 23 262 470.0 2000* m w 263 450.0 2350 f w 31 264 365.0 900 m b 19 265 410.0 1350 m b 23 266 360.0 925 f w 29 267 470.0 2565 f b 25 268 330.0 550 m w 30 269 350.0 920 m b 28 270 1100 f w 19 271 460.0 1675 f b 20 272 380.0 1090 m w 28 273 520.0 3000 m w 15 274 340.0 670 f b 24 275 225.0 220 m b 22 276 2160 f w 17 277 490.0 3100 m w 21 278 425.0 1720 m b 18 Hemorrhagic pulmonary edema; no congenital malformations No congenital malformations; unknown cause of death No congenital malformations; placenta previa Renal malformations; moderate maceration Rectocloacai malformation; pulmonary hypoplasia Twin placenta previa; intraventricular hemorrhage Premature rupture of membranes; chorioamniorutis Hydrocephalus Congenital heart disease Placental infarction Patent ductus arteriosus and foramen ovale Hyaline membrane disease; interstitial edema No congenital malformations; unknown cause of death No congenital malformations; abruptio placenta Hemorrhagic pulmonary edema Hypotensive event in mother with chronic hypertension No congenital malformations; unknown cause of death Anencephaly Congenital heart disease Hyaline membrane disease; thyroid hyperplasia 00

PAGE 99

Series # CHL Weight Sex Race Mother's age Cause of death/conunents 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 285.0 415.0 280.0 350.0 290.0 285.0 335.0 415.0 200.0 260.0 325.0 225.0 285.0 265.0 520.0 380.0 260.0 295.0 285.0 575 2126 310* 840 540 450 700 2610 200 400 650 975 200 400 240 3225 985 210 488 470 f m f f m m m m m m m m w w w w w b b w w b b b w b b b w w b b 19 Omphalocele with herniated internal organs Potter's f acies; multiple congenital malformations 25 No congenital abnormalities; non-viable immaturity 22 Hydrocephalus 22 No congenital malformations; hyaline membrane disease 31 Premature rupture of membranes; chorioamnionitis 28 Acute chorioamnionitis 2nd to premature rupture of membranes 34 Maternal polyhydraminos; generalized edema 19 Cystic hygroma; possible Turner's syndrome 33 Potter's syndrome; agenesis of kidneys 18 Placenta previa; prematurity; hyaline membrane disease 17 Abruptio placenta; perinatal asphyxia 19 Premature rupture of membranes; bronchopneumonia 19 Immature placenta; chorioamnionitis 16 No congenital malformations; placental fibrosis 24 No congenital malformations; unknown cause of death 21 No congenital malformations; unknown cause of death Hydrocephalus; cleft palate 20 No congenital malformations; acute chorioamnionitis 24 No congenital malformations; unknown cause of death CO en

PAGE 100

Series # CHL Weight Sex Race Mother's age Cause of death/comments Placental infarction Maternal DIC; diffuse petechial hemorrhages Maternal pre-eclampsia; no congenital malformations Myocardial fibrosis; moderate hydrops fetalis; syndactyly Twin dissecting hematoma in jejunum Unknown cause of death; no congenital malformations Oligohydraminos; Potter's facies with pulmonary hypoplasia Twin placenta previa Multiple skeletal anomalies; congenital heart disease Triplet "C" non-viable immaturity Triplet "B" twin to twin transfusion w/ Triplet "A" (donor) Triplet "A" twin to twin transfusion w/ Triplet "B" (recipient) Unknown cause of death; fetal death in utero, macerated Anencephaly; selective termination of pregnancy Potter's facies; oligohydraminos; perinatal asphyxia Hydrocephalus Abruptio placenta; no congenital malformations Maternal diabetes; agenesis (phocomelia) of upper limbs Potter's facies; multiple congenital malformations Placenta previa; peritonitis 219 520.0 2675* w 18 220 840* w m 27 221 235.0 275 b 25 222 310.0 720 m w 22 223 350.0 815 w 36 224 370.0 1050^^ m b 25 225 360.0 1220 w 20 226 330.0 640* m w 23 227 430.0 2310 w 32 228 330.0 680* w 31 229 305.0 510** w 31 230 310.0 570** w 31 231 295.0 340* m b 18 232 2200 b 26 233 1460 w 234 335.0 850 m w 27 235 420.0 1255 m w 29 236 340.0 640 m w 24 237 1800 f w 32 238 360.0 1000 m w 24 00 ON

PAGE 101

Series # CHL Weight Sex Race Mother's age Cause of death/comments 199 385 1090 m b 31 Premature rupture of membranes; prolapsed cord 200 ... 2740 f w 31 Syndactyly; double uterus/vagina; diaphragmatic hernia 201 ... 1700 m w 24 Potter's syndrome; congenital absence of kidneys 202 ... 2040 m b 19 Potter's syndrome; renal hypoplasia 203 ... 365 f w 26 Anencephaly b 23 Marginal abruptio placenta; premature rupture of membranes 205 205.0 165 f w 18 Anencephaly 22 Pnevimonia secondary to meconium aspirahon f b 34 Acute fetal anoxia; no congenital malformations 22 Twin twin to twin transfusion syndrome (recipient) 19 Recent ischemic necrosis of heart, no congenital malformations 32 Twin premature; vent, hematoma; no congenital malformations 2^j 485.0 2400 m w 28 Atelectasis; pulmonary hemorrhage; ventricular hemorrhage 212 290 460* m b 21 Maternal pyelonephritis; no congenital malformations 213 220 450'' f w 21 No congenital malformations; "nuchal bleb" 214 305 520 f b 26 No congenital malformations; maternal incompetent cervix 204 265.0 390 m 205 205.0 165 f 206 520.0 2890 m w 207 ... 2210* 208 490.0 2350* m w 209 310.0 580 m w 210 395.0 1240 m w 215 275.0 380 f w 216 510.0 3280 m w 19 No congenital malformations; premature rupture of membranes 24 No congenital mals; meconium aspiration; asphyxia neonatorum 217 285.0 400 m b 18 placental infarct; no congenital malformations 2jg 35Q0 975 f b 23 Abruptio placenta; no congenital malformations 00

PAGE 102

Series* CHL Weight Sex Race Mother's age Cause of death/ comments Congenital heart disease; aortic atresia 18 Meconium aspiration; recent (within 72 hrs) ventricular hemorrhage 22 No congenital malformations; hyaline membrane disease 19 No congenital malformations; sepsis 26 Fetal death in utero; no congenital malformations; maternal drug use 31 Twin immaturity; hyaline membrane disease 30 Rh incompatibility; no congenital malformations 18 Unknown cause of death; no congenital malformations 36 Intrauterine infection; no congenital malformatons 30 Twin hyaline membrane disease; immaturity 30 Prematurity; no congenital malformations; spontaneous AB 25 Anencephaly 29 Fetal death secondary to maternal eclampsia and seizures 17 Abruptio placenta; no congenital malformations 40 Possible abruptio placenta; no congenital malformations 27 Trisomy 18; congenital heart disease 18 Sepsis; no congenital malformations 23 Congenital syphilis 21 Complete abruptio placenta 17 Twin no congenital malformations; immaturity 179 490.0 3110 f w 180 280.0 320 m b 181 360.0 830 m b 182 340.0 940 m b 183 540.0 3070* f w 184 320.0 765 f w 185 1220 m b 186 400.0 1220* f w 187 270.0 150* m w 188 335.0 800 f w 189 210.0 100* w 190 510.0 3925 m b 191 410.0 1100 m w 192 340.0 750 m b 193 225.0 210 m b 194 415.0 1449 m w 195 720 f w 196 410.0 1540* f b 197 510.0 2650 m b 198 380 f w 00 00

PAGE 103

Series # CHL Weight Sex Race Mother's age Cause of death /comments 159 405.0 1200 f b 160 2494 f b 161 450.0 2540 m w 162 380.0 1270 m w 163 580 m b 164 ... 2725 f b 165 2600 f h 166 2382 m w 167 410 m b 168 1160 m w 169 340* f b 170 300.0 420* m w 171 285.0 430 f w 172 390.0 1450 m w 173 390.0 1225 m w 174 465.0 1990 m b 175 392.0 1230* m w 176 1700 m w 177 300.0 560 f b 178 260.0 620* m w 21 Neonatal asphyxia; abruptio placenta 27 Polydactyly in all extremities, no other malformations 22 Potter's syndrome; bilat. hip dislocations, club feet 19 Twin "B"; immaturity; hyaline membrane disease 19 Maternal preeclampsia; no congenital malformations 21 Status post diaphragmatic hernia repair; pulmonary hypoplasia 15 Fetal death in utero, meconium stained, no congenital malformations 18 Tetralogy of Fallot 25 Chorioamnionitis; premature delivery 16 Status post ligation of patent ductus; pulmonary dysplasia Unknown cause of death; severely macerated 33 No congenital malformations 17 No congenital malformations; acute infection 22 Twin twin to twin transfusion syndrome (recipient) 22 Twin twin to twin transfusion syndrome (donor); twin to above 25 Premature; no pathology noted 37 No pathology; maternal diabetes and hypertension 26 Potter's syndrome, club feet 18 Multiple congenital anomalies; poss. Triploidy 25 Pulmonary hypoplasia; cystic hygroma oo

PAGE 104

Series # CHL Weight Sex Race Mother's age Cause of death /comments 156 139 ... 3770 m 140 400.0 ... m 141 490.0 2625 142 ... 2875 143 ... 1330 m w 23 Trai^sposition of great vessels; sudder\ death w 23 Non-viable immaturity; no congenital defects 141 490.0 2625 m w 24 Stahjs post emergency C-section, fetal distrss f w 24 Diaphragmatic hernia, pulmunary hypoplasia; patent ductus b 26 Placenta previa; hyaline membrane disease 144 390.0 1130 m b 17 Asphyxiation by nuchal cord 145 315.0 580 m b 16 Abruptio placenta and placenta previa 146 305.0 660 m w 25 Tw?in "A"; no congenital malformations 147 300.0 410 m w 25 Twin "B"; no congenital malformations 148 370.0 810 m b 21 Scattered placental infarcts 149 528 2940 f b 21 Potter syndrome with multiple anomalies 150 345.0 820 f b 21 Torsion of umbilical cord; no congenital anomalies 15J 1525 m b 17 No congenital malformations; maternal syphilis 152 498.0 2500 m w 28 Congenital heart disease 153 410.0 1360 f w 40 Ventricular septal defect b 22 Maternal syphilis; no congenital abnormalities 154 320.0 625 m 155 ... 620 m w 19 Hyaline membrane disease; uncomplicated pregnancy, 2nd abortion 470.0 2120 m w 22 Sepsis ^57 1120 m w 19 Sepsis, patent ductus 158 ... 2700 m w 24 Fatty metamorphosis of pregnancy; no congenital malformations O

PAGE 105

Series # CHL Weight Sex Race Mother's age Cause of death/comments Persistent fetal circulation, patent ductus, meconium aspiration Potter's syndrome, severe malformations Premature rupture of membranes; hyaline membrane disease Abruptio placenta Hypoplastic left heart Exencephaly (early form of anencephaly before degeneration) Severe immaturity Abnormal fades; premature rupture of membranes Bronchopulmonary dysplasia Hydrocephalus; congenital heart disease Chorioamnionitis and pneumonia Chorioamnionitis, premature rupture of membranes Meconium aspiration Severe immaturity and hyaline membrane disease Severe immaturity Cardiac anomalies Premature rupture of membranes; strep sepsis No congenital malformations; pneumonia Multiple heart malformations Bacterial sepsis 119 2540 f b 25 120 1360 m b 17 121 ... 1140 m b 17 122 350.0 725 m b 22 123 2402 f w 21 124 m w 125 838 f w 19 126 1050 m w 23 127 420.0 1520 f b 25 128 400.0 2140 f b 16 129 280.0 450 m w 130 320.0 820 m w 30 131 3500 m b 16 132 1410 m w 18 133 330.0 610 f b 16 134 ... 4010 f w 135 ... 1460 f b 15 136 510 f b 24 137 2800 m b 20 138 980 m b 16 ^o

PAGE 106

Series # CHL Weight Sex Race Mother's age Cause of death/comments 99 485.0 2755 f b 21 100 380.0 1030 m w 43 101 3620 m w 40 102 820 f b 21 103 365.0 1090 m b 15 104 290.0 510 f w 29 105 1400 m b 18 106 1120 f w 107 330.0 750 m w 28 108 2800 f w 109 660 f b 30 110 325.0 520 m b 19 111 ... 870 m b 22 112 ... 1340 m b 113 500.0 3030 m b 15 114 320.0 880 f b 33 115 277.0 525 m w 22 116 500.0 3130 f w 27 117 2760 m b 17 118 ... 640 f w 25 Massive pulmonary hemorrhage; no congenital malformations No congenital malformations; intraventricular hemorrhage Trisomy 13; tetralogy of fallot Multiple congenital abnormalities; amniotic band syndrome Non-viable immaturity; bronchopulmonary dysplasia No congenital malformations; chorioamnionitis Hyaline membrane disease Non-viable immaturity Immature placenta; no congenital abnormalities Campomelic dwarfism; sex reversal male karyotype, female organs Anencephaly with cranioschisis Hyaline membrane disease Hyaline membrane; immaturity DIC; Klebsiella pneumonia Multiple vasular congenital defects Anencephaly Twin (other had acardiac anomaly); unknown cause of death Diaphragmatic defect with pulmonary hypoplasia Occipital encephalocele Multiple soft-tissue malformations

PAGE 107

Series # CHL Weight Sex Race Mother's age Cause of death/comments 79 495.0 2770 m b 21 Abruptio placenta, no congenital malformations 80 455.0 1825 m w 30 Interruption of aortic arch; multiple anomalies 81 400.0 1440 f b 27 Maternal preeclampsia and Rh negative sensitivity 82 245.0 940 f b 34 Multiple congenital abnormalities 83 470.0 2520 m w 24 Premature rupture of membranes 84 450.0 1525 m w 39 Placental infarction, fetal death in utero 85 ... 2800 m b 17 No congenital abnormalities; fetal death in utero 86 310.0 600 m w 20 Non-viable immaturity; no congenital abnormalities 87 ... 630 m b 21 Non-viable immaturity; hyaline membrane disease 88 ... 500 { w 23 No congenital malformations 89 330.0 740 m b 14 Hyaline membrane disease; non-viable immaturity 90 385.0 920 m w 28 Severe hyaline membrane disease; non-viable immaturity 91 ... ... f b 20 Severe hyaline membrane disease; non-viable immaturity 92 310.0 630 f b 22 Non-viable immaturity, first of twins; hyaline membrane disease 93 310.0 650 f b 22 Non-viable immaturity, second of twins; hyaline membrane disease 94 475.0 2560 m b ... Premature rupture of membranes 95 ... ... m w ... Post status multiple gastrointestinal surgical procedures 96 ... 2610 m w 29 Down's syndrome with congenital heart disease 97 ... 5440 f w 21 SIDS in large (80th percentile) infant 98 ... 710 m b 17 Abruptio placenta OJ

PAGE 108

Series # CHL Weight Sex Race Mother's age Cause of death/comments Abruptio placenta; hyaline membrane disease Abruptio placenta; Bochdalek hernia of diaphragm Premature rupture of membranes; no malformations Non-viable prematurity Multiple congenital deformities; cleft lip /palate, imperf. anus, etc. Anencephalic Premature rupture of membranes Fetal death in utero, unknown etiology Syndactyly with other minor abnormalities, poss. Trisomy E Thanatophoric dwarf type I Anuiiotic band syndrome Amniotic band complex; total abruptio placenta Prematurity; multiple MI; pulmonary emphysema No congenital malformaton; immaturity No congenital malformation; premature rupture of membranes Placental infarction Premature rupture of membranes; pitocin induced delivery Abruptio placenta Post-mature by "dates "; meconium aspiration Transposition of great vessels 59 360.0 900 m b 23 60 350.0 825 m w 17 61 ... 3300 m b 18 62 321.0 750 m w 63 485.0 2475 m w 31 64 365.0 1100 m w 22 65 ... 1134 m w 66 285.0 440 f 67 440.0 2030 f w 35 68 ... 2600 m b 19 69 7 b 16 70 270.0 660 m b 27 71 420.0 1600 m w 29 72 270.0 440 m b 23 73 300.0 580 m w 18 74 420.0 1500 f b 75 2600 m w 22 76 475.0 2350 f b 25 Tl 480.0 2640 f w 78 m w ...

PAGE 109

Series # CHL Weight Sex Race Mother's age Cause of death /comments 39 400.0 1500 m w 18 Abruptio Placenta 40 600.0 4850 m b ... Meningitis 41 ... ... m b ... Multiple minor anomalies. I.e. syndactyly, low ears 42 390.0 1280 m b ... Maternal preeclampsia; truncus arteriosus; absence of ductus arter. 43 ... 2000 m w ... Premature; delivered early 2nd to maternal DM; > B/P 44 ... 1230 m w 36 Prematurity; twin delivered by C-section 45 330.0 740 m b 21 Hyaline membrane disease 46 385.0 1180 f w 21 Abruptio placenta; prematurity with hyaline membrane disease 47 ... f w 28 Idiopathic pulmonary hypertension 48 510.0 ... f w 24 Fetal death in utero; etiology undetermined 49 ... 2000 f w ... Spontaneous delivery; fungal sepsis 50 470.0 1960 m w 25 True knot in umbilical cord 51 280.0 500 f b 17 Non-viable immaturity 52 340.0 730 f b 34 Sacral meningomyelocele 53 350.0 900 f w 31 Maternal eclampsia; no malformations present 54 ... 820 m b ... Severe neonatal asphyxia 55 620 f b 18 Birth trauma; absence of 3rd and 4th toes, left foot 56 380.0 1200 m w 25 Bronchopulmonary dysplasia; patent ductus arteriosus 57 1700 f b ... Multiple minor congenital malformations 58 530 f b ... Non-viable immaturity

PAGE 110

Series* CHL Weight Sex Race Mother's age Cause of death/comments 18 355.0 1150 f w 20 19 1430 m w 27 20 2800 m w 21 i *" 1500 m w 31 22 355.0 650 m w 27 23 470.0 2800 f w 31 24 ... 430 m w 16 25 ... m w 15 26 760 f 27 400.0 1175 m w 18 28 3030 m w 36 29 f b 30 f w 22 31 f b 32 310.0 660 m 16 33 365.0 1620 m w 35 480.0 2350 m b 31 36 465.0 2650 m w 37 1200 m w 23 38 530.0 3480 m w Abruptio placenta No gross abnormalities Multiple congenital malformations Hydrops fetalis; prematurity Hydrocephalus; therapeutic AB Persistent fetal circulation syndrome No congenital abnormalities Fetal death in utero; cause unknown Abdominal pregnancy Intraventricular hemorrhage Non-viable immaturity Normal delivery and child; gastric rupture Down's syndrome; translocation type in a young mother Limited autopsy; hemolytic-uremic syndrome Probable intrauterine pneumonia Partial abruptio placenta; c-section due to fetal distress Fetal death in utero to diabetic mother; no pathology found Aortic atresia; atrial septal defect; mitral valve hypoplasia Non-viable immaturity Anomalous pulmonary venous return; pulmonary infarction 0^

PAGE 111

RAW DATA RECORDED FROM AUTOPSY PROTOCOLS Series # CHL Weight Sex Race Mother's age Cause of death/comments 1 ... 3450 m w 21 2 .„ ... m w ... 3 ... ... m w ... 4 410.0 1890 m b 17 5 390.0 2300 m b 17 6 350.0 1050 m w 20 7 355.0 950 f b 17 8 280.0 600 m b 34 9 335.0 950 f w 27 10 445.0 1710 m w 18 11 340.0 700 f w ... 12 410.0 1800 f w 15 13 460.0 2150 f w ... 14 290.0 640 m w 40 15 400.0 1020 f w 17 16 310.0 1380 m b 13 17 700 m w 20 Eagle Barrett Syndrome; absent abdominal musculature Meningitis; no congenital abnormalities Caudal regression syndrome; duhamel complex of anomalies Severe growth retardation; below 3rd %; Trisomy 18 Diaphragmatic hernia with pulm. hypoplasty Severe immaturity, hyaline membrane disease Severe immaturity, hyaline membrane disease Immature placenta; fetal death at undetermined date Hydrops fetalis Thrombosed placenta/tracheoesophageal fistula Polyspenia complex; abdominal organ situs inversus Uterine infection with normal fetus Placenta previa Surviving twin; atrophic umbilical cord Death possibly secondary to maternal trauma CHF; hydrops fetalis (hence high weight) Placental irifarcts; encephalomalacia

PAGE 112

APPENDIX C AUTOPSY PROTOCOL (FACSIMILE) DOE, Jane A. 49-35-23 A87-094 POST-MORTEM RECORD DEPARTMENT OF PATHOLOGY College of Medicine The J. Hillis Miller Health Center University of Florida SEX: Female RACE: Black AUTOPSY NO: A87-094 DOB: 6/16/87 AGE: Stillborn NAME: DOE, Jane A. DATE OF DEATH: 6/16/87 HOSPITAL #: 49-35-23 DATE OF AUTOPSY: 6/16/87 SERVICE: OB/GYN ATTENDING DR: John Smith, M.D. PROSECTOR: K. Smith, M.D. PERMIT BY: Joan Doe, Mother RESTRICTIONS: None Copies to : Department Box # John Smith, M.D. OB/GYN J-494 S. Martin, M.D. OB/GYN J-478 P. Stubblefield, M.S.W. Social Services J-255 Referring Physician: None FINAL ANATOMIC DIAGNOSES (1) STILLBORN, MACERATED FEMALE (Birth weight 2,040 grams; crownheel length, 47.5 cm; gestational age 36-38 weeks by length), ESTIMATED GESTATIONAL AGE 32 WEEKS; FETAL DEATH IN UTERO DIAGNOSED FOLLOWING ONE DAY OF REPORTED DECREASED FETAL MOVEMENT. (2) MATERNAL HISTORY: 30 YEAR OLD GRAVIDA 2, PARA 1-0-0-1; PAST MATERNAL HISTORY OF SEIZURE DISORDER AND DEPRESSION (medications discontinued August, 1986) RUBELLA IMMUNE, BLOOD GROUP A POSITIVE. (3) DIFFUSE HEMORRHAGE (4) VAGINAL DELIVERY FOLLOWING OXYTOCIN INDUCTION (5) NO CONGENITAL ABNORMALITIES 98

PAGE 113

REFERENCES CITED Acheson RM and Hewitt D (1954) Oxford child health survey: Stature ar^d skeletal maturation in the preschool child. Br J Prevent Med 8:59-65. Aguilar T, Teberg AJ, Chan L, and Hodgman, JE (1995) Intrauterine growth curves of weight, length, and head circumference for a predominantly Hispanic infant population. Public Health Reports, May-June 1995, Vol. 110(3):327-332. Anton SC (1997) Developmental age and taxonomic affinity of the Mojokerto Child, Java, Indonesia. Am J Phys Anthropol 102:497-514. Armelagos GJ, Meilke JH, Owen KH, Van Gerven DP, Dewey JR, and Mahler PE (1972) Bone growth and development in prehistoric populations from Sudanese Nubia 7 Hum Evol 1:89-119. Babson SG, and Benda GI (1976) Growth graphs for the clinical assessment of infants of varying gestational age. / Pediatrics 89(5):814-820. Bagnall KM, Harris PF, and Jones PRM (1978) Studies on the pattern of ossification in the long bones of human fetal limbs with some observations on sex and size differences. / Bone Joint Surg 60(B):284. Bagnall KM, Harris PF, and Jones PRM (1982) A radiographic study of the longitudinal growth of primary ossification centres in long bones of the human fetus. Anat Rec 203:293-299. Bayer LM, and Bayley N (1959) Growth Diag nosis. Chicago: University of Chicago Press. Birkbeck JA, Billewicz WA, and Thomson AM (1975a) Human fetal measurements between 50 and 150 days gestation. Ann Hum Biol 2:173-178. Birkbeck JA, Billewicz WA, and Thomson AM (1975b) Fetal growth from 50 and 150 days of gestation. Ann Hum Biol 2:319-326. Birkbeck JA (1976) Metrical growth and skeletal development of the himian fetus. In The Bioloev of Human Fetal Growth, Roberts DR, and Thomson AM (eds.), pp. 39-68. London: Taylor and Francis. Boucher BJ (1957) Sex differences in the foetal pelvis. Am J Physiol Anthrop 15:581. Brenner WE, Edelman DA, and Hendricks CH (1976) A standard of fetal growth for the United States of America. Am } Obstet Gynecol 126(5):555-564. 99

PAGE 114

100 Brock J (1954) Biologische Paten fur den Kinderarzt I/l. Berlin-GottingenHeidelberg: Springer. Christie A (1950) The estimation of fetal maturity by roentgen studies of osseous development. Amer } Obstet Gynec 60:133. Dietrich HA (1925) Anatomie und Physiologie des Fetus und Biologic der Placenta Halban-Seitz. I. Augl. VI, 1:177. Eleveth PB, and Tanner JM (1976) Worldwide Variation in Human Growth Cambridge, Massachusetts: Cambridge University Press. Falsetti AB, and Cole TM EI (1992) Relative growth of the postcranial skeleton in callitrichines. / Human Evol 23:79-92. Fazekas I, and Kosa K (1978) Forensic Fetal Osteolog y. Akademiai Kiado Publishers, Budapest, Hungary. Garn S (1957) Research in human growth. Human Biology 29:1-11. Gam SM, and Shamir Z (1958) Methods for Research in Human Growth Springfield, 111.: Charles C. Thomas. Gartner H (1947) Bin statistisches verfahren zur berechnung der tragezeit aus den reifemerkmalen des neugeborenen imd seine anwendbarkeit bei gutachtlichen entscheidungen. Arch Gyniik 174:538. Gavan JA and Swindler DR (1966) Growth rates and phylogeny in primates. Am J Phys Anthrop 24:181-190. Genoves S (1967) Proportionality of the long bones and their relation to stature among Meso-americans. Am J Phys Anthrop 26:67-77. Golbus MS, and Berry LC Jr (1976) Human fetal development between 90 and 170 days postmenses. Teratology 15:103-108. Gray H (1941) Individual growth rates: From birth to maturity for 15 physical traits. Hum Biol 13:306-333. Greulich WW, and Pyle SI (1959) Radiographic Atlas of Skeletal Development of the Hand and Wrist, 2nd edition. Stanford University Press. Gruenwald P (1966) Growth of the human fetus. Am J Obstet Gynecol 94(8):1112-1119. Guthmarm H, and Knoss S (1939) Mit welcher sicherheit lasst sich aus der kindsgrosse die tragezeit ermitteln? Zbl Gyniik 63:2636. Halac J, Arias ME, Numa RY, and Halac E (1982) Fetal growth rates in Argentina a national collaborative study. Clin Pediatr 21:342-347.

PAGE 115

101 Harrison GA, Weiner JS, Tanner JM and Bamicot NA (1964) Human Biolog y. New York and Oxford: Oxford University Press. Hartley JB (1957) Radiological estimation of foetal maturity. British J Rad 30:561-576. Hewitt D, Westropp CK and Acheson RM (1955) Oxford child health survey: Effect of childish ailments on skeletal maturation. Br J Prevent Soc Med 9:179-186. Hoffman JM (1979) Age estimation from diaphyseal lengths: two months to twelve years. / Forensic Sci 24:461. Hummert JR, and Van Gerven DP (1983) Skeletal growth in the Medieval population from Sudanese Nubia. Am J Phys Anthropol 60:471-478. Huxley AK (1996) A comparison between the use of diaphyseal length and foot length to calculate gestational age from forensic fetal remains. Proceedings of the American Academy of Forensic Sciences, February 1996, Volume II, p. 180. Jantz RL, and Owsley DW (1984) Long bone growth variation among Arikara skeletal populations. Am } Phys Anthropol 63:13-20. Jeanty P, and Romero R (1984) Estimation of the gestational age. Semin Ultrasound CT MR. 5:121. Johnston FE (1961) Sequence of epiphyseal union in a prehistoric Kentucky population from Indian Knoll. Human Biology 33:66-81. Johnston FE (1962) Growth of the long bones of infants and young children at Indian Knoll. Am J Phys Anthrop 20:249-254. Johnston FE (1968) Growth of the skeleton in earlier peoples. In Brothwell, DR (ed.): The Skeletal Biology of Earlier Human Populations Oxford: Pergamon Press, pp. 57-66. Johnston FE (1969) Approaches to the study of developmental variability in human skeletal populations. Am J Phys Anthopol 31:335-341. Jungers WL (1982) Lucy's limbs: skeletal allometry and locomotion in Australopithecus afarensis. Nature 297:676-677. Jimgers WL, Falsetti AB, and Wall CE (1995) Shape, relative size, and size-adjustments in morphometries. Yearbook of Physical Anthropology 38:137-161. Kerley ER (1970) Estimation of skeletal age: After about 30 years. In Stewart TD (ed.): Personal Identification in Mass Disasters Washington, D.C.: National Museum of Natural History, pp. 57-70. Kosa F (1989) Age estimation from the fetal skeleton. In Age Markers in the Human Skeleton. Iscan MY (ed.). Charles C. Thomas, Springfield, 111., pp. 21-54. Krogman WM (1950) A handbook for the measurement and interpretation of height and weight in the growing child. Monog Soc Res Child Devel 13:iii-68.

PAGE 116

102 Krogman WM (1955) The skeleton in forensic medicine. Postgrad Med 17(2):A48-A62. Krogman WM (1962) A problem in the ageing of human remains. / Forensic Sci 7:255-264. Krogman WM, and Iscan, MY (1986) The Human Skeleton in Forensic Medicine Charles C. Thomas, Springfield, lU. Kuroda S (1970) Medico-legal problems concerning the fetus and infant. Jap J Lev Med. 24:361. Lange IR, and Marming,FA (1984) Fetal biophysical assessment: An ultrasound approach. Semin Ultrasound CT MR. 5:269. Leigh SR (1996) Evolution of human growth spurts. Am J Phys Anthrop 101:455-474. Lippert H, and Lippert E (1960) Geschlechtsunterschiede an den wirbelkorpem menschlicher feten. Z menschl Vererb u Konstit.-Lehre 35:445; cited in Fazekas I, and Kosa K (1978) Forensic Fetal Osteology Akademiai Kiado Publishers, Budapest, Hungary. Lovejoy CO, Russel KF, and Harrison ML (1990) Long bone growth velocity in the Libben population. Am } Hum Biol 2:533-541. Lubchenco LO, Hansman C, and Boyd E (1966) Intrauterine growth in length and head circumference as estimated from live births at gestational ages from 26 to 42 weeks. Pediatirics 37(3):403-408. Lubchenco LO, Hansman C, Dressier M, and Boyd E (1963) Intrauterine growth as estimated from livebom birth-weight data at 24 to 42 weeks of gestation. Pediatirics, November 1963 793-800. Maresh MM (1955) Linear growth of the long bones of extiremities from infancy through adolescence. Am J Diseases Child 89:725. Martin RH, and Higginbottom J (1971) A clinical and radiological assessment of foetal age. / Obst Gyn British Commonwealth 78:155-162. Mashiack S, Blankstein J, and Stem M (1975) The determination of fetal age and growth rate by ultrasonography. Harejuah. 88:499. McKem TW (1970) Estimation of skeletal age: From puberty to about 30 years of age. In Stewart TD (ed.): Personal Identification in Mass Disasters Washington, D.C.: National Museum of Natural History, pp. 41-56. Mehta L and Singh HM (1972) Determination of crown-rump length from fetal long bones:Humerus and femur. Am J Phys Anthrop 36:165-168. Mensforth RP (1985) Relative tibia long bone growth in the Libben and Bt-5 prehistoric skeletal populations. Am J Phys Anthrop 68:247-262.

PAGE 117

103 Mensforth RP, Lovejoy CO, Lallo JW and Armelagos, GJ (1978) The role of constitutional factors, diet, and infectious disease in the etiology of porotic hyperostosis and periosteal reactions in prehistoric infants and children. Med Anthropol 2:1-59. Merchant VL and Ubelaker DH (1977) Skeletal growth of the protohistoric Arikara. Am } Phys Anthrop 46(l):61-72. Meredith HV (1962) Childhood interrelations of anatomic growth rates. Groxvth 26:2340. Modi JP (1957) Medical Jurisprudence and Toxicology Bombay: Tripathi private. Moss ML (1954) Differential growth analysis of bone morphology. Am J Phys Anthropol 12:71-76. Moss ML (1955) Relative growth of the human fetal skeleton, cranial, and post cranial. Ann NY Acad Sci 63:528-536. Naeye RL, and Dixon JB (1978) Distortions in fetal growth standards. Pediatr Res 12:987-991. Olivier G, and Pineau H (1958) Determination de I'age du foetus et de I'embryon. Arch Anat (La Semaine Des Hopitaux). 6:21-28. Olivier G, and Pineau H (1960) Nouvelle determination de la taille foetale d'apres longueurs diaphysaires des os longs. Ann Med Leg. 40:141-144. Owen RH (1971) The estimation of foetal maturity. Brit J Radiology 44:531-534. Peterson DJ, Alexander GR, Powell-Griner E, and Tompkins ME (1989) Variations in the reporting of gestational age at induced termination of pregnancy. Am } Public Health 79:603-606. Pyle SL and Hoerr ML (1969) A Radiologic Standard of Reference for the Growing Knee Charles C. Thomas, Springfield, 111. Richards GD, and Anton SC (1991) Craniofacial configuration and postcranial development of a hydrocephalic child (ca. 2500 B.C. 500 a.d.): With a review of cases and comment on diagnostic criteria. Am J Phys Anthrop 85:185-200. Roche (1988) Assessing the skeletal maturity of the hand-wrist: Pels method Charles C. Thomas, Springfield, lU. Russell JG, Mattison AE, and Easson WT (1972) Skeletal dimensions as an indication of foetal maturity. Brit } Radiol. 45:667. Scammon RE and Calkins LA (1923a) Simple empirical formulae for expressing the lineal growth of the human fetus. Proc Soc Exp Biol 21:353-356.

PAGE 118

104 Scammon RE and Calkins LA (1923b) New empirical formulae for determining the age of the human fetus. Anat Rec 25:148-149. Scammon RE and Calkins LA (1925) Crown-heel and crown-rump length in the fetal period and at birth. Anat Rec 29:372-373. Scammon RE and Calkins LA (1929) Growth in the Fetal Period University of Minnesota Press, Minneapolis. Scheuer JL, Musgrave JH, and Evans SP (1980) The estimation of late fetal and perinatal age from limb bone length by linear and logarithmic regression. Ann Human Biol 7:257-265. Schroeder CF, Schmidtke SZ, and Bidez MW (1997) Measuring the human pelvis: A comparison of direct and radiographic techniques using a modem United Statesbased sample. Am J Phys Anthrop 103:471-479. Schviltz AH (1937) Proportions, variability and assymetries of the long bones of the limbs and the clavicles in man and apes. Human Biol 9:281-328. Simpson SW, Russell KF and Lovejoy CO (1996) Comparison of diaphyseal growth between the Libben population and the Hamann-Todd chimpanzee sample. Am J Phys Anthrop 99:67-78. Streeter GL (1920) Weight, sitting height, head size, foot length, and menstrual age of the human embryo. Contrib. To Embryol, Carnegie Inst. Washington. 11:143. Publ. No. 274. Streeter GL (1948) Developmental horizons in human embryos; description of age groups, XV, XVI, XVU, and XVIU, being third issue of survey of Carnegie Collection. Contrib. To Embryo!., Carnegie Inst. Washington. 32:133. Streeter GL (1949) Developmental horizons in human embryos. A review of the histogenesis of cartilage and bone. Contrib. To Embryol, Carnegie Inst. Washington. 33:149. Tanner JM, Whitehouse RH, Cameron N, Marshall WA, Healy MJR, and Goldstein H (1983) Assessment of Skeletal Maturity and Prediction of Adult Height (TW2 Method). 2nd edition. San Diego: Academic Press, Inc. Todd TW (1939) Skeleton, locomotor system, and teeth. In Cowdry EV (ed.) Problems of Ageing Baltimore: Williams and Wilkins, pp. 278-338. Toldt C (1879) Uber Altersbestimmung menschlicher embryonen. Prag med Wschr 4:121. Trotter M, and Gleser GC (1952) Estimation of stature from long bones of American whites and Negroes. Am J Phys Anthrop 10:463-514. Trotter M, and Gleser GC (1958) A re-evaluation of stature based on measurements taken during life and of long bones after death. Am J Phys Anthrop 16:79-123.

PAGE 119

105 Trotter M, and Gleser GC (1977) Corrigenda to "Estimation of stature from long bones of American whites and Negroes," American Journal of Physical Anthropology (1952). Am J Phys Anthrop 47:355-356. Ubelaker DH (1989) The estimation of age at death from immature human bone. In Age Markers in the Human Skeleton. Iscan MY (ed.). Charles C. Thomas, Springfield, 111., pp. 55-70. Walker PL (1968) The linear growth of long bones in Late Woodland Indian Children. Proc Indian Acad Sci 78:83-87. Warren MW, Smith KR, Stubblefield PR, Martin SS, Walsh-Haney HA and Maples WR (1997) The use of clinical radiographic atlases in the determination of skeletal age in a mass fatality. Proceedings of the American Academy of Forensic Sciences, Volume in. Weaver DS (1986) Forensic aspects of fetal and neonatal skeletons. In Reichs KJ (ed.) Forensic Anthropology Charles C. Thomas, Springfield, 111., pp. 90-100. Y'Edynak G (1976) Long bone growth in western Eskimo and Aleut skeletons. Am J Phys Anthrop 45:569-574. Zangermeister W (1911) Die altersbestimmung des rotus nach graphischer methode. Z Geburtsh Gyndk 69:127. Zangermeister W (1912) Tafeln zur Altersbestimmung menschlicher Friichte Ferdinand Enke, Stuttgart. Zangermeister W (1917) Studien uber die schwangerschaftsdauer und die fruchtentwicklung. Arch Gyndk 107:405.

PAGE 120

BIOGRAPHICAL SKETCH Michael W. Warren received his Bachelor of Arts in history cum laude from the University of North Florida in Jacksonville. He received his Master of Arts and PhD. in anthropology from the University of Florida, specializing in osteology, skeletal biology, human anatomy, evolutionary theory and forensic identification and trauma analysis. He is a member of the American Association of Physical Anthropologists and a student member of the American Anthropological Association and American Academy of Forensic Sciences. 106

PAGE 121

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentatjon'and is fullv^^aeauate, in scope and quality, as a dissertation for the degreeiyf^ctor of I*tilo/ophy ^_^ L Susan Anton, Chair Assistant Professor of Anthropology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation ancl is fully adequate, in scope and quality, as a dissertation for the degree of Doctoriof Phil'ostjphy. Anti|6ny^lFalsetti| Assistant Professfejr of Anthropology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Leslie S. Lieberman Associate Professor of Anthropology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of^Philosophy. Sue Boinski Assistant Professor of Anthropology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctprbf Philospphy. Lynn Larkin Professor of Anatomy and Cell Biology This dissertation was submitted to the Graduate Faculty of the Department of Anthropology in the College of Liberal Arts and Sciences and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. December 1997 Dean, Graduate School


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EOLITJNV3_EGLH9B INGEST_TIME 2017-07-11T21:51:31Z PACKAGE AA00022858_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES