On my about me and CV pages I’ve mentioned my Master’s work, which was based in the field of archaeology and anthropology. The main focus of my research was to observe and investigate a feature of human bone using micro-CT; this feature was the Harris Line (HL), or growth arrest lines. I’ve been doing this blog for nearly a year now and I haven’t written a post focusing on these lines so I thought it was about time.
This page will explain will be extremely long if I include the background/literature review and a summary of my work so I’ll put them in two separate pages. This one will include what the possible causes of HLs are, the history of their discovery, how HLs form and what relevance they are to archaeology. At a later date (hopefully next week) I’ll write about my Master’s work.
Throughout the literature there are few clear and concise definitions for HLs. Some may refer to specific bones or specific size of a line while others may be more general. One definition that appears to explain Harris lines fairly clearly is Caffey’s description: ‘transversely directed bony trabeculae in a thin stratum which extends entirely across the medullary cavity and exactly parallels provisional zone of calcification’ (1967 p791). A point to note is that although here Caffey describes a transverse line as extending across the bone shaft he also says older lines may be thinner and broken. Another is by Mays (1985 p207) ‘a lattice of dense horizontally orientated trabeculae’. The difference in these definitions can create problems when attempting to identify and record these lines.
Wegner first identified Harris lines in 1874 in relation to rickets (Harris 1931). After experiments of feeding phosphorous to rabbits and chickens transverse lines of bone adjacent to the growth cartilage were observed. These early experiments only focused on a single causation for the transverse lines, however in 1922 Harris (Harris 1933) studied the radiographs of long bones of sick children at University College Hospital. He identified the same bony striations seen by Wegner in children who had not suffered from rickets. This created an interest in the formation and cause of these lines which resulted in investigations into other deficiency diseases and the formation of HLs.
Around the same time Asada in 1924, (Harris 1931) produced HLs experimentally, but not in relation to a disease. His method involved encasing long-bones in plaster causing arrested growth. As this method produced transverse lines Asada proposed that the dense bone was caused when a disturbance in the growth of the bones occurred (Harris 1931). This disturbance was in the activity of osteoblasts and cartilage growth, when cartilage growth had halted but osteoblast activity remained active.
Until this point there had been little attempt made to explain the mechanism of grow for the transverse lines. During Harris’s research he claimed that the development of the lines was due to a disturbance in the proliferative cartilage, which caused abnormal calcification (Silverman 1964). Harris also suggested that these lines were identical to the line which appears at the epiphysis upon reaching adult stature (Harris 1933). However Park (1964) disagreed with this and suggested that there was an alternative explanation. This has been summarised by Caffey (1967). Rather than abnormal calcification the formation of HLs was attributed to the resuming of osteoblast activity after a period of suspended growth. At the beginning of a period of slow or halted growth local osteobalsts form a template on the underside of the prolific cartilage, this is almost invisible on a radiograph. Once growth recommences osteoblast activity resumes causing the template to thicken and become visible on a radiograph. Ogden and Rosen (in Uhthoff & Wiley 1988) suggested that this causes the trabecular bone to thicken and possibly fuse with each other transversely. This means that a line is formed during a period of accelerated growth, or recovery phase, rather than during a period of delayed growth (Caffey 1967). Up until Park the role in recovery of an illness had not been taken into account.
Park also notes that Harris was incorrect in saying that HLs are the same as or similar to the stratum that forms at epiphyseal closure (Park 1964). Park argues that this is not the case as the role of recovery in the formation of the lines is important. He also notes that the structure of a line at epiphyseal closure is different to the structure of a HL (Park 1964). This is because the cells of the epiphysis die out leaving a thin plate of cartilage which is impregnated with calcium-phosphate crystals. This line is also made up of three layers that appear during early development; this is in contrast to a HL where a single event creates a single line.
A study by Alfonso-Durruty (2011) proposed that transverse lines are a result of stress and accelerated growth. To study for this a longitudinal study on rabbits was conducted. To test for stress the rabbits were assigned to groups with varying diets consisting of a control, undernourished and periodically fasted group. Various measurements were then taken including weight, humeral diaphyseal length and diaphyseal growth velocity and the number of HL. It was found that more HL were formed in the control group compared to the under nourished group and these were produced during periods of rapid growth. This point is not a new one with authors assigning the formation of line to rapid growth after illness (Park 1964). However Alfonso-Durruty (2011) suggests that the lines appear with the absence of stress and therefore are a poor indicator of heath.
Relevance to Archaeology
In osteaoachaeology the Harris line (HL) has been used predominately as an indicator of past population health. Many studies have been carried out to determine the frequency of the lines within an individual and/or the age at which they formed. By assessing these ‘stress markers’ the health status of a population can be indicated. Many different sites have been studied differing in both time and place (e.g. Allison et al. 1974, Mays 1995, Piontek et al. 2001, Ribot & Roberts 1996). The majority of these studies used radiographs to observe HLs as they are deemed easy to use (Maat 1984).
An early study by Wells (1967) aimed to compare the morbidity rates of different populations. This was achieved by calculating the average number of HL that occurred in each population. As well as recording HLs that were present Wells also noted the percentage of a population that did not have any lines. He also recorded at which ages the lines occurred, highlighting some intriguing results. It was found that a higher rate of morbidity was found in one middle-Saxon population until the age of ten where then the other middle-Saxon population surpassed it (Wells 1967). Studies like these can produce interesting results that promote further research and thinking, such as what occurred in the others populations to cause a sudden increase in HLs. Wells (1967) also looked at other markers of stress within the different populations including dental caries and hypoplasia and periodontal diseases. This allowed for a full view of the health status of the populations studied.
Many other studies have looked into other markers of stress in correlation with HLs (e.g. Clarke 1982, Maat 1984). McHenry and Schulz (1978) examined the long bones and the teeth of prehistoric Californian Indians. The aim of this paper was to identify whether there was an association between dental enamel hyperplasia (DEH) and HLs, as both have been thought of as indicators of non-specific stress (McHenry & Schulz 1976). DEH has similar aetiologies as HLs causing lines, pits or grooves in the teeth as a response to childhood illness or trauma (Roberts & Manchester 2010). However, no significant association was found between the two features. It was suggested that this was due to the process of bone resorption and therefore the removal of HLs causing DEH to be observed without any DEH. When a HL was found with no associated HL it was thought that DEH form from large traumas whereas HL develop during a quick recovery of a trauma (McHenry & Schulz 1978). This would suggest that HL develop only after small incidences of trauma. In addition to the McHenry & Schulz study a study by Mays (1995) produced contrasting results. He found that a correlation between DEH and HL did exist in juvenile individuals, although he did note that DEH can form without the presence of HL. None the less these conflicting results from the two studies indicate an area for further research to clarify the process of HL formation. A HL identified by Mays can be seen in fig. 5.
Ribot and Roberts (1996) also looked at the evidence of HL and other stress markers. The other indicators of stress studied were: DEH, porosity of the surface of the skull, causing a sieve like appearance, and subperiosteal bone formation on the surface of long bones. This study compared two populations of the same period to determine their health status. By combining many different factors it could be argued that a general idea of health status can be made. As the aetiology is unknown for HL they were used with caution in this study.
However, there is an issue with the use of transverse lines as an indication of health status. As previously mentioned Grindhart (1969) noticed that lines could also form even when no illness occurred. He also remarked that every child that was studied showed at least one line. Due to these reasons HL could be questioned as to whether they can be strong indicators of health, especially when used on their own.
From doing this literature review and getting information about the history, formation and use of Harris Lines I went on to investigate the structure and appearance using micro-CT. I will explain in detail my research involved and discuss the results that were produced. In the mean time you can view the abstract of my Master’s thesis at academia.edu.
Alfonso-durruty, M P (2011) ‘Experimental Assessment of Nutrition and Bone Growth ’ s Velocity Effects on Harris Lines Formation’ in American Journal of Physical Anthropology 145: 169-180
Allison, M. J., Mendozo, D. & Pezzia, A. (1974) ‘A radiographic approach to childhood illness in precolumbian inhabitants of southern Peru’ in American journal of physical anthropology 40:409-415
Ameen, S., Staub, L., Ulrich, S., Vock, P., & Ballmer, F. (2005) ‚‘Harris lines of the tibia across centuries: a comparison of two populations, medieval and contemporary in Central Europe’ in Skeletal Radiography 34: 279-284
Caffey, J. (1967) ‘Transverse lines of Park (Stress lines of Park)’ in Pediatric x-ray diagnosis p.789-795. Year Book Medical Publishers: Chicago
Clarke, S. K. (1982) ‘The association of early childhood enamel hypoplasia and radiopaque transverse lines in a culturally diverse prehistoric skeletal sample’ in Human Biology 54:77-84
Gindhart, P. S. (1969) ‘The frequency of appearance of transverse lines in the tibia in relation to childhood illnesses’ in American journal of physical anthropology 31:17-22
Harris, H. A. (1931) ‘Lines of Arrested Growth in the Long Bones of Diabetic Children’ in British Journal of Radiography 4: 561-588
Harris, H. A. (1933) Bone Growth in Health and Disease. Oxford University Press: London
Maat, G. (1984) ‘Dating and rating of Harris’s lines’ in American journal of physical anthropology 63:291-299
Mays, S. (1985) ‘The relationship between harris line formation and bone growth and development’ in Journal of Archaeological Science 12:207-220
Mays, S. (1995) ‘The Relationship between Harris Lines and other Aspects of
Skeletal Development in Adults and Juveniles’ in Journal of Archaeological Science 22:511-520
McHenry, H. & Schulz, P. (1978) ‘The association between Harris lines and enamel hypoplasia in prehistoric California Indians’ in American journal of physical anthropology 44:507-511
Ogden, J. A. & Rosen, L. C. (1988) ‘Defining the Growth Plate’ in Uhthoff, H. K. & Wiley, J. J. (ed.) Behaviour of the Growth Plate Raven Press: New York
Park, E. A. (1964) ‘The imprinting of nutritional disturbances on the growing bone’ in Pediatrics 33:815-861
Piontek, J., Jerszynska, B., & Nowak, O. (2001) ‘Harris Lines in subadult and adult skeletons from the medieval cemetery in Cedynia, Poland’ in Variability Evol 9:33-43
Ribot, I. Roberts, C. (1996) ‘A study of non-specific stress indicators and skeletal growth in two mediaeval subadult populations’ in Journal of archaeological science 23: 67-80
Silverman, F. (1964) ‘Doctor Park and the growth of bone’ in Pediatrics 33:639-641
Wells, C. (1967) ‘A New Approach to Palaeopathology: Harris’s Lines’ in Brothwell, D. & Sandison, A. T. (eds.) Diseases in Antiquity Thomas Books: Illinois