International Journal of Paleopathology xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

International Journal of Paleopathology

journal homepage: www.elsevier.com/locate/ijpp

Research article

The epidemiology of rickets in the 17th–19th centuries: Some contributions

from documentary sources and their value to palaeopathologists
S. Mays
Research Department, Historic England, Fort Cumberland, Fort Cumberland Road, Eastney, Portsmouth PO4 9LD, UK

A R T I C L E I N F O A B S T R A C T

Keywords: This article considers the nature of written sources on the epidemiology of rickets in the post-Mediaeval period,
Biocultural study and examines the value of these sources for palaeopathologists. There is a progression from 17th–18th century
Post-mediaeval sources, which generally make ex cathedra, qualitative statements on rickets frequency to, in the 19th century,
Historical semi-quantitative geographical surveys of its occurrence, through to reports of percentage prevalence in various
Vitamin D
groups. Of course, even these latter cannot be directly compared with prevalences calculated from excavated
skeletal remains, but there are also considerable difficulties in comparing them with one another, and this
effectively precludes synthesis to provide reliable information on geographic and temporal trends at anything
more than a very broad-brush level. Their problematic nature mandates a cautious approach when using written
sources to shed light on the epidemiology of rickets. For palaeopathologists, a useful way of incorporating these
sources into a biocultural approach may be to use them in order to formulate hypotheses that can then be
evaluated using skeletal evidence.

1. Introduction
The dominant paradigm in palaeopathology is currently the bio-
cultural approach, in which skeletal data are harnessed to address
questions of broad historical or archaeological interest (Roberts and
Manchester, 1995: 1; Zuckermann and Armelagos, 2011). Biocultural
study primarily involves integrating palaeopathological data with other
skeletal data and/or other types of archaeological evidence. However,
for the historic period, the potential insights offered by the biocultural
approach may be most fully realised when historical sources are also
integrated within it (Mitchell, in press). This is particularly true for the
post-Mediaeval period, when written texts, including medico-historical
sources, become more abundant. A biocultural approach in palaeo-
pathology normally involves testing of hypotheses using statistical
analyses of data. The focus is often on the study of disease frequencies
in different populations, or in subgroups within a population, so pa-
laeoepidemiology plays a key role. In this paper, I examine some 17th-
19th century written sources on the palaeoepidemiology of rickets, and
consider how they might be incorporated into a biocultural framework
for understanding rickets in this period.
2. Written sources on the epidemiology of rickets
Rickets seems to have been recognised by the second century AD in
ancient Rome (Jackson, 1988: 38), and by the 8th century in China
(Lee, 1967), but the first substantial treatises on the disease come from
the mid 17th century, when Daniel Whistler, Arnold Boot, John Mayow
and (most influentially) Francis Glisson, provided clear clinical de-
scriptions (Whistler, 1645; Boot, 1649; Glisson et al., 1650; Mayow,
1668). These were based on observations on cases in England; Boot also
observed the disease in Ireland and France. As well as clinical de-
scriptions, these authors also offered some epidemiological observa-
tions. Glisson and Whistler both suggest the disease began to be re-
cognised in England in around 1620 (Glisson et al., 1651: 3–4; Whistler,
1950 [1645]: 401). They state that the disease was common at the time
they were writing. Boot claimed it was common in Ireland as well as
England (van Andel, 1927). Glisson stated that rickets had first been
observed in Dorset and Somerset, but since then had spread to other
parts of southern England, including London, but was rare in northern
counties (Glisson et al., 1651: 4). Both Whistler and Glisson state that
rickets was most often seen among the children of the wealthy, with the
offspring of the poor being much less affected (Glisson et al., 1651: 121;
Whistler, 1950 [1645]: 409).
Prior to the above medical treatises, there is evidence that non-
medical writers were aware of rickets. For example, in Suffolk in 1636,
Sir Simonds D’Ewes describes the death of his son, at a year and nine
months old, of convulsions and “the rickets” (Halliwell, 1845: 123,
143–144). Folk remedies for rickets were also circulating at this time.
The domestic receipt books (notes of culinary and medicinal recipes) of
the Fairfax family from Yorkshire contain an entry dated 25th February

E-mail address: simon.mays@historicengland.org.uk.

https://doi.org/10.1016/j.ijpp.2017.10.011
Received 16 August 2017; Received in revised form 26 October 2017; Accepted 31 October 2017
1879-9817/ © 2017 Elsevier Inc. All rights reserved.

Please cite this article as: Mays, S., International Journal of Paleopathology (2017), https://doi.org/10.1016/j.ijpp.2017.10.011
S. Mays
1632 describing five remedies for “the rickets (in children)” (Wedell,
1890: 158). The herbalist John Parkinson describes a concoction for the
treatment of children with rickets in his Theatricum Botanicum, pub-
lished in 1640 (Parkinson, 1640: 980). Seven years later, the
churchman, Thomas Fuller published his tract ‘Good Thoughts in Worse
Times’, the second of a trilogy offering thoughts, primarily from a re-
ligious standpoint, upon the days in which he lived. In one of a series of
entries giving ‘meditations on the times’, he uses the head and limb
defomities seen in the “new disease” of rickets as a metaphor for the
sickness he perceived in people’s souls during the troubled times of the
English Civil War (Fuller, 1863 [1647]: 140). That lay people attributed
the suffering of their offspring to rickets, folk remedies for it had been
devised, and that men such as Fuller expected that his readers would be
familiar with it, supports the notion that rickets was well established in
early 17th century England. Further evidence for this is provided by the
London Bills of Mortality, where, beginning in 1634, rickets begins to
be cited as a cause of death.
The London Bills of Mortality recorded deaths in parishes in London
(Graunt, 1662). They began to be compiled in the 16th century, and
their prime purpose was to help provide warning of plague outbreaks in
the city (Greenberg, 1997). From 1629, cause of death consistently
began to be recorded. The details of deaths were gathered by
‘searchers’. These were elderly women of the parish who had no formal
medical or other scientific training. They would deduce cause of death
from the general appearance of the corpse and from talking to the fa-
mily of the deceased. They would convey this and other information to
the Parish Clerk who was responsible for compiling the Bills of Mor-
tality for his parish. Deaths attributed to rickets peaked at about 3.5%
of the total in the period 1650–60, and declined thereafter, so that by
the mid 18th century they were only 0.4%. Although rickets is not a
lethal disease, its presence increases vulnerability to life-threatening
infections, particularly of the respiratory tract (Hess, 1930: 290; Basit,
2003), hence its appearance in dead infants and children is un-
surprising. The London Bills are a record of how causes of death were
perceived by lay persons rather than providing a reliable source of
epidemiological data. Comparisons between different parishes suggest
that use of rickets as a cause of death descriptor was inconsistent,
searchers in some parishes using it and others not (Newton, 2012). The
decline in the frequency with which rickets was entered as a cause of
death from about 1660 onward cannot be taken as evidence for a de-
cline in the frequency with which the disease was present in dead in-
fants and children. It is inherently improbable that a disease directly
linked to sunlight exposure should decline at a time when industrial
pollution and urban crowding were growing problems. It is more likely
that, for reasons that are obscure, rickets simply passed out of common
usage as a cause of death descriptor during the 18th century (Newton,
2012).
Indeed, in contrast to the evidence of the London Bills, a number of
18th century sources describe rickets as being frequent in London and
elsewhere. In the third quarter of the 18th century, James Nelson, a
London apothecary, recorded that rickets was extremely common in the
capital (Nelson, 1763: 96). Ten years later William Farrar also made
this point (Farrar, 1773: 19), and additionally noted that the disease
could be severe enough to cause pelvic deformity that could lead to
obstetric problems for women in later life (Farrar, 1773: 45–6). Writing
at about the same time, William Fordyce, a physician based in West-
minster, asserted that there must have been more than 20,000 children
in London and its suburbs with the ‘Hectic Fever’ (which, from his
description, clearly denotes rickets) (Fordyce, 1777: 207). By contrast,
George Armstrong, a physician based in Hampstead, then a village
outside London, stated that rickets was rarely met with where he lived
unless it was in those coming out from London who already had the
disease (Armstrong, 1767: 96). Rickets was also held to be common
elsewhere. For example, in the early 18th century, Thomas Floyer, a
physician based in Lichfield, a market town in the English Midlands,
stated that no disease was more frequent in infants than the rickets
2
International Journal of Paleopathology xxx (xxxx) xxx–xxx
(Floyer, 1715: 76). In 1789, Michael Underwood noted a general rise in
rickets which he ascribed to the increase in manufacturing industry
drawing more people into towns (Underwood, 1789: 314).
In contrast to the London Bills of Mortality, from the 18th century
onward, most statements on rickets and its frequency tended to come
from physicians rather than lay people. Nevertheless, difficulties re-
main concerning the reliability with which the disease was recognised.
Some of the most important bony signs of rickets (e.g. enlargement of
sternal rib-ends and long-bone metaphyses, and bowing of long-bones)
had been described by mid 17th century writers (Whistler, 1950
[1645]; Glisson et al., 1651; Mayow, 1926 [1668]), and continued to be
considered key to recognising the disease during the centuries that
followed. However, these early writers also erroneously ascribed a
variety of other features (for example cough, diarrhoea) to rickets. In
later centuries, these, and a variety of further clinical features, such as
fever and loss of appetite, which are likewise not a reflection of vitamin
D deficiency per se but rather of co-occurrent diseases to which the
rickety infant is prey (e.g. respiratory tract and other infections), were
also commonly associated with rickets (Brent and Mitchell, 2008). This
confusion persisted well into the 19th century, and is a serious obstacle
in interpreting epidemiological evidence; it was only in the later dec-
ades of that century that descriptions became more accurate (Brent and
Mitchell, 2008).
The nineteenth century witnessed a growing interest in the epide-
miology of rickets, in particular its geographical distribution, mainly for
the light it might shed on the causes of the disease, which at the time
remained mysterious. In 1855 Augustus Merei, a Hungarian physician
working in Manchester, England, instigated what was perhaps the first
attempt to collate data from different locations to examine the geo-
graphic distribution of rickets. He used his own experience in the north
of England, plus information obtained by letter from practitioners in
other British locations, in order to study the frequency of rickets in
different places (Merei, 1855: 159–186). Some of the results, classified
according to whether Merei’s correspondants felt rickets was rare or
common in their districts are presented in Table 1. The data seem to
indicate that, although many of Merei’s correspondants in urban areas
felt the disease to be rare, when it was common it was generally the
urban poor that were affected. The finding whereby the disease more
Table 1
Rickets prevalence at various locations in Britain according to Merei (1855).
Letters in parentheses denote whether the location was coloured Blue (de-
noted B, rickets common) or Red (denoted R, rickets rare) in the BMA’s 1889
maps (see text).
Rickets
Common Rare
Bath (poor) (B) Aberdeen (B)
Birmingham (B) Bath (wealthy)
London (poor) (B) Blackpool (B)
Manchester (B) Bradford (B)
Newcastle (poor) (B) Bristol (B)
Stockport (B) Cambridge (B)
Cork & SW Ireland (R)
Dublin (B)
Edinburgh (B)
Glasgow (B)
Inverness (R)
Limerick (R)
Liverpool (B)
London (wealthy) (R)
Montrose (R)
Norwich (B)
Perthshire (R)
Scarborough (R)
Scottish Highlands (R)
Southport (R)
Thirsk (R)
S. Mays
often affected the poor in urban areas echoes statements by other mid-
century physicians based on their own experience (Jenner, 1860;
Stephenson, 1865). Although there were exceptions, in urban locations
Merei felt that rickets was generally more common in towns and cities
that had a strong manufacturing base. For example, it was reported as
common in Manchester, but as rare in Liverpool only 35 miles away.
The two were similar sized cities, both with considerable levels of
poverty, but former was a centre of manufacturing industry, the latter a
trading seaport. In Manchester, the disease was seen to some extent
among the wealthy as well as the poor, and its frequency declined with
increasing distance from the city centre. Merei was also familiar with
rickets in his native Budapest: it was common in the crowded tenements
of Pest but rare in wealthy Buda on the opposite bank of the Danube.
Thirty years later, the British Medical Association (BMA) organised
a very large and more systematic survey of rickets in Britain. They sent
out a questionaire to every registered medical practitioner enquiring as
to the frequency of rickets (and several other diseases) in their locales.
Physicians were asked whether rickets was common, that is a physician
would be likely to meet with at least one case per year. An accom-
panying memorandum reminded respondants of the key signs of rickets
(bent bones (usually the leg bones), and enlargement of distal radius
and sternal rib ends). More than 3000 responses were received. The
results were summarised in a series of maps showing disease distribu-
tion, in which locations producing affirmative answers to the above
question were depicted with a blue spot; negative answers were co-
loured red. Maps were compiled for Scotland, England, Wales and
Ireland, and separate detailed maps were produced for some urban
areas and for some of the offshore islands. The results were also sum-
marised in a short publication (Owen, 1889). Perusal of the maps shows
that, in England (Fig. 1), very few substantial settlements appeared in
red. It is only once one gets down to market towns and villages that red
starts to predominate. The largest concentrations of blue correspond to
the chief urban and industrial areas – London, the Black County of the
West Midlands, and a large swathe of south Lancashire and south and
west Yorkshire. In rural areas red predominates but this is less so in the
Midlands and the south than in northern counties. In Wales, the large
towns and cities of Glamorgan and Monmouthshire, and the coal
mining valleys, were mainly coloured blue. Elsewhere, settlements
were mainly red. Scotland produced concentrations of blue in Glasgow
and Edinburgh, but other settlements were mostly red. Ireland was also
mostly red, except Dublin, Belfast and Londonderry. Overall, rickets
was clearly associated with larger towns and cities, and with mining
areas. In Manchester, the entire city and surrounding areas were blue
(there was no trace of the reduction in rickets with distance from the
city centre reported by Merei 30 years earlier). In London, all inner city
areas were blue with the exception of wealthy Mayfair and Belgravia.
In the following decade, Emil Feer, a physician in Basel, conducted a
fairly similar exercise in Switzerland (Feer, 1897). He contacted 270
Swiss physicians. He asked whether, in their districts, they observed
rickets very frequently, frequently, seldom or never (although he did
not define those terms). The responses received, classified by location
(Table 2), showed that rickets essentially paralleled population density,
it being most frequent in the more densely populated lowlands and
rarely encountered in the sparsely populated Alpine regions. Other
writers conducted surveys on an international scale. In the late 1880s,
Theobald Palm, a British physician, wrote to medical missionaries in
China, Japan, India and Morocco, enquiring about the frequency of
rickets (Palm, 1890). Responses indicated that the disease was almost
unknown in these regions. In the 1890s, Léopold Baumel, in Mon-
tpellier, sent out a postal questionaire on rickets to physicians in Europe
and elsewhere (Baumel, 1898). This confirmed that the disease was
common in temperate Europe, but it was almost unknown in tropical
regions; this latter observation had also been made anecdotally by
earlier writers (e.g. Underwood, 1789: 315). By the last quarter of the
nineteenth century, sufficient evidence had accumulated to enable
commentators (Hirsch, 1886; Feer, 1897; Baumel, 1898) to conclude
3
International Journal of Paleopathology xxx (xxxx) xxx–xxx
that, essentially, rickets was rare in southern Europe, the Middle East,
Asia, and in the tropics, but was frequent in cities in central and
northern Europe.
The surveys of the BMA and Feer produced data on rickets fre-
quencies measured on ordinal scales (‘rare’, ‘common’, etc.) so these
data can only be considered semi-quantitative. Back in 1855, Merei had
opined that physicians should ‘exert themselves’ to establish statistical
frequencies of rickets and communicate their findings via the medical
press (Merei, 1855: 160). This call soon began to be answered. As the
century wore on, data on frequency of rickets in the medical literature
was increasingly expressed in terms of percentage prevalence rather
than in semi-quantitative terms. These publications were usually simple
reports of the prevalence of rickets in children examined as outpatients
at the physician’s own institution. The data were obtained by study of
routine medical records rather than gathered specifically for the pur-
pose. In many reports there is little detail on the size and composition of
the group upon which the frequency of rickets was based, but some data
from studies that do give information sample size and/or age compo-
sition are given in Table 3.
3. Discussion
Seventeenth and 18th century sources on the epidemiology of
rickets tend to be ex cathedra statements (opinion carrying weight ac-
cording to the authority of the writer rather than on evidence pre-
sented) based primarily on their authors’ own experience in dealing
with patients with the disease. An exception is the London Bills of
Mortality, but these are fundamentally flawed as sources of epidemio-
logical data, at least in so far as rickets is concerned. Two trends emerge
in the nineteenth century literature. The first was a desire to draw to-
gether the experience of physicians at different locations in order to
build up a more comprehensive picture of the distribution of rickets.
This resulted in surveys at a national or international level, but they
collated data that was generally only semi-quantitative. The second was
a desire to present statistical data instead of subjective opinion re-
garding rickets frequency. This resulted in physicians producing fully
quantitative evidence, in the form of prevalences among children
brought in to their own institutions. Both these trends were primarily
driven by an increasing perception that epidemiological data could help
shed light on the causes of the disease, an issue that became more ur-
gent as rickets became an increasing health concern.
Problems associated with sampling, mortality and diagnosis prevent
simple comparisons between palaeopathological prevalences and epi-
demiological data gathered from living subjects (Waldron, 2009:
253–4). However, potentially, it might be possible to compare geo-
graphic, social and temporal patterning in the two types of data. As
regards the documentary sources, this would require that sources giving
semi-quantitative or quantitative data be synthesised so that trends may
be discerned. There are considerable difficulties in achieving this,
particularly for the former. For example, it is tempting to compare the
British rickets surveys of Merei in the 1850s (Merei, 1855) and the BMA
in the 1880s (Owen, 1889) to try and determine temporal trends over
that 30 year period. In Table 2, Merei’s results are annotated according
to the findings of the 1889 BMA survey at each location. This exercise
indicates that at no locations where the Merei (1855) survey found
rickets to be frequent did the BMA survey find it to be rare, but at many
settlements the reverse was the case. This would seem to be consistent
with rickets becoming more common across the board; in particular,
there are very few urban locations where the disease was described as
rare in 1889 whereas many cities reported it as such back in 1855.
However, unlike the BMA, Merei did not provide his correspondants
with diagnostic criteria for rickets, or a proforma showing how its
frequency should be classified, so it is difficult to make the comparison.
In addition, the BMA, in defining ‘common’ at the level of one case per
year, set the bar rather low, particularly for respondants in urban areas.
For example, Gee (1868a) met with 635 cases in a year in London. This
S. Mays International Journal of Paleopathology xxx (xxxx) xxx–xxx

Fig 1. Distribution of rickets in England in 1889 according to the BMA survey. Blue spot, rickets common; red spot, rickets rare (for details see text). Reproduced by permission of the
BMA. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

was in a hospital specifically for sick children, but even as a physician in
a Manchester dispensary, Ritchie (1871) encountered 219 cases in a
year. In this light, it is perhaps not surprising that even small settle-
ments were often coloured blue on the BMA maps. What is perhaps
more notable is that, occasionally, even fairly large settlements were
red (e.g. Coventry, York and Southampton).
Turning to the prevalence data, Table 3 provides evidence that
rickets was a serious problem in cities in northern and central Europe,
as contemporary commentators asserted (e.g. Hirsch, 1886). In
addition, in locations listed in Table 3 where data are available from
more than one survey, the temporal trend is usually (but not invariably)
upward. Largely because of the influence of Glisson’s work, rickets was
commonly known as the ‘English disease’ in continental Europe, but it
would seem that its frequency in the second half of the 19th century
was as high, if not higher, in many cities in other European countries.
More detailed analysis of these data is however fraught with problems.
These include (inter alia) factors associated with diagnosis, the com-
position and selection of the study-base, and by the practice of using

4
S. Mays International Journal of Paleopathology xxx (xxxx) xxx–xxx

Table 2 extant hospital records to assess disease frequency rather than data
Frequency of rickets in Switzerland (Feer, 1897). gathered specifically for the purpose.
In some of the epidemiological studies in Table 3, there is little on
Region Rickets frequency
the criteria used to diagnose rickets. In these cases, it may have been an
Very Frequent Rare or Absent Total assumption that competent practitioners should have little trouble in
frequent very rare recognising the disease. Compared with earlier times, by the late 19th
century the features of rickets had become more accurately char-
Mittellands & Lakes 8 12 7 0 27
Jura 7 7 7 0 21 acterised (Brent and Mitchell, 2008). Although the extent to which
Transitional zones 1 7 13 2 23 newer, more accurate, published descriptions of the disease influenced
between Mittelland & diagnosis in clinical practice is uncertain, one would expect physicians
Alps in pediatric hospitals, such as the authors and repondents in Table 3, to
Alpine areas 1 5 31 3 40
be more up to date than general practitioners. Consistent with this,
Total 17 31 58 5 111
those sources in Table 3 that do mention diagnostic criteria tend to
emphasise the primacy of skeletal deformities in identifying rickets (e.g.
Gee, 1868a; Quisling, 1888; Cohn, 1894; Mey, 1896). Even if one

Table 3
Some percentage prevalence figures for rickets among children reported by physicians in the later 19th century.

Location Date Frequency Age range Reference Notes

Austria
Vienna 1880s 895/1000 (90%) < 3yrs Kassowitz (1885)b
Britain
Manchester 1855 8/34 (23%) < 4yrs Merei (1855: 171) Children of wealthy families
London 1867 635/2355 (27%) < 4yrs Gee (1868a) Includes slight cases
Manchester 1869–70 219/728 (30%) < 5yrs Ritchie (1871)
Edinburgh 1890s 634/1500 (42%) < 4yrs Thompson (1898)a
Czech Republic
Prague 1860–62 504/1623 (31%) < 5yrs Ritter (1863: 64, 291)
Prague 1890s - (80%) < 2yrs Epstein (1898)a
Denmark
Copenhagen 1862–67 159/1883 (8.4%) < 5yrs Bruenniche (1867)
France
Paris 1884–94 1362/7000–8000 (17–20%) 2–15yrs Leroux (1898)a
Montpellier 1882–97 382/8408 (5%) < 15yrs Baumel (1898)
Germany
Berlin 1841–43 650/4715 (14%) < 5yrs Wollheim (1844)b
Dresden 1858 1654/9000 (18%) Children Küttner (1858)
Berlin 1874–84 N = 4782 (25%) < 2yrs Klein and Schwechten (1883)b
Frankfurt am Main 1883 192/706 (27%) 6mo.–4yr Lorey (1884)
Berlin 1891 858/1303 (66%) < 3yrs Cohn (1894)
Munich 1890s - (63%) < 2yrs Seitz (1892)b
Hungary
Budapest 1891–93 2500/23196 (11%) < 4yrs Bokay (1898) Slight cases not included
Italy
Genoa 1889–90 340/1942 (18%) Children Feer (1897)
Latvia
Riga 1895 865/1000 (87%) < 2yrs Mey (1896) Children of the poor brought for vaccination
Netherlands
Amsterdam 1892–95 979/6721 (15%) Children Voûte (1898)a
Norway
Oslo 1880s 971/4868 (20%) < 3yrs Quisling (1888)
Oslo 1895–96 1148/2374 (48%) < 4yrs Johannessen (1898)
Romania
Iaşi 1889–96 272/15320 (2%) Children Russ (1898)a
Russia
St Petersburg 1880s N = 3188 (62%) Children Van Putteren (1887)c
St Petersburg 1891–93 3055/3225 (95%) 1month–16yrs Schukowski (1894)c Children of the poor
St Petersburg 1890s - (27%) Children Reitz (1895)c
Moscow 1890–95 2148/2530 (80%) < 3yrs Kissel (1897)
Switzerland
Geneva 1889–90 340/1942 (18%) Children Feer (1897)
Basel 1896 43/90 (48%) < 3yrs Feer (1897)
Ukraine
Kiev 1888–91 146/1082 (13%) Children Karnitzky (1898)c
Kiev 1896 679/1209 (56%) < 4yrs Troitzky (1898)a
USA
Philadelphia 1869–71 - (28%) 1mo.–5yrs Parry (1872) Occurs in wealthy as well as the poor

Only instances where sample size and/or the age span of sample are given are included. All study samples are of sick children brought to hospital as outpatients unless stated. For some
further studies that give prevalences but no sample details see Feer (1897) and Baumel (1898).
For references denoted b or c, original sources are identified in Supplementary materials.
a
Data reported in correspondance to Baumel (1898) & presented therein.
b
Figures given in Cohn (1894).
c
Figures given in Kissel (1897).

5
S. Mays
accepts that, for the data in Table 3, accuracy of diagnosis may have
been less of a problem than in earlier times or for authors who did not
specialise in diseases of children, the question of whether early rickets
or only the more advanced cases are included will have an influence on
the frequency reported. This was recognised by contemporary physi-
cians. For example, Gee (1868a) divides his cases according to severity
(albeit without explaining how this was done), and he includes slight
cases in his overall prevalence. He (Gee, 1868b) suggests that a reason
for the low prevalence of rickets reported by Bruenniche in Copenhagen
(Bruenniche, 1867) was because he emphasised bow-leg defomity in
diagnosis, which Gee considered a late sign.
It had been noted since the 17th century (Glisson et al., 1651: 192;
Mayow, 1926 [1668]: 303–4) that rickets most often affects children
between 6 and 30 months old, and this continued to be the case in the
19th century (Bruenniche, 1867; Ritchie, 1871; Bokay, 1898). The age-
span of the sample will therefore have an effect on prevalences re-
ported. This was recognised by 19th century writers. For example, Gee
(1868b) criticised Bruenniche (1867) and other authors for including
the first 5 years of life as the basis for their statistics, implying that it
would tend to depress the prevalence compared with studies using a
younger upper age limit. The bulk of studies in Table 3 have an upper
age limit of 3, 4 or 5 years. The effect this has on statistics will depend
both upon the age profile of rickets cases and the age profile of children
brought to the attention of physicians as outpatients. Gee (1868a),
Ritchie (1871) and Bokay (1898) present data broken down by age, and
this shows that prevalences calculated with an upper limit of 3, 4 or 5
years vary by less than 3 percentage points. If these results are gen-
eralisable, then it would suggest that small differences in upper age
limits do little to compromise comparisons. However, some studies use
15 or 16 years as the upper limit, and others simply include all ‘chil-
dren’ with ages unspecified. Late 19th century data (Quisling, 1888;
Frey, 1898) indicate that < 3% of rickets cases were recorded in chil-
dren over 5 years, so the inclusion of older children could potentially
have the effect of significantly reducing the prevalence figures in these
studies.
The study-bases for the data in Table 3 are children examined by
physicians at the institution where they practiced. Normally these were
children brought along because they were sick, either from rickets or
from some other cause. These hospital-based samples clearly cannot be
considered an accurate guide to prevalence of rickets in the commu-
nities in which they were situated (although figures for children
brought in for vaccination programmes (e.g. Mey, 1896) may be a
better guide). They also cause difficulties for comparing figures be-
tween different authors as we do not know how chronically or severely
sick the children were when they were brought to the author’s atten-
tion. For example, Ritchie (1871) notes that, for some of the children in
his study-base, treatment had been attempted for some months before
they were referred to him.
The way in which records were compiled by individual hospitals is
also likely a factor influencing results. For example, Merei (1855:
178–81) considers reported prevalences of between 3 and 12% for
rickets among children seen by physicians at children’s hospitals in
Dresden, Vienna, Frankfurt and Budapest, as grossly at variance with
the (perceived) high prevalence of rickets in these cities. He argues that
only severe cases of rickets would be likely to be brought to the at-
tention of medical practitioners unless some other disease was present,
in which case the accompanying disease rather than rickets would
probably be entered into the hospital records.
In the light of the above discussion, it would seem that the epide-
miological data in written sources can neither be directly compared
with prevalences from cemetery data, not can the different sources be
compared directly with one another, precluding detailed analysis. By
this stage a reader might be wondering what use, if any, are these
sources to palaeopathologists? One possible answer would be that,
given their problematic nature, they might best be considered as a
source of hypotheses that might be investigated using
6
International Journal of Paleopathology xxx (xxxx) xxx–xxx
palaeopathological study. Although palaeopathological diagnosis has
its own limitations and pitfalls, palaeopathological study may never-
theless be used to assess whether skeletal evidence supports or coun-
termands hypotheses generated from readings of documentary sources.
If the statements of 17th century authorities are taken at face value,
then rickets began as a problem primarily affecting the better-off in
rural locations. It has been suggested that this pattern was real and
reflects a tendency of the wealthy to coddle their infants and young
children indoors (Gibbs, 1994). The rise of the home-based textile in-
dustry, where whole families worked long hours indoors, may also have
played a part (Kellett, 1934; Gibbs, 1994) – the west of England, where
Glisson first noted the disease, was an important textile producing area.
By contrast, the poor at this time largely remained bound to the land. At
some point, most probably during the 18th century, there was a change
so that rickets became predominantly an urban problem that pre-
ponderantly, but not exclusively, affected the poor, a change that re-
flects the increasing industrialisation and urbanisation occurring during
that period (Woods and Woodward, 1984; Stevenson, 1993). However,
even setting aside the question of its general validity, this is a very
broad-brush scenario. One might reasonably expect that different re-
gions or different types of settlements would show differing secular
trends in the disease, depending upon social, environmental, occupa-
tional and other factors. For example, Merei (1855) argued that the size
of the manufacturing base was a key determinant of the frequency of
rickets in urban areas, a suggestion that seems plausible in the light of
the effects that industrial pollution and crowded housing conditions
have on the disease. Palaeopathological study might help, not only to
shed light on the general social, temporal and geographic picture of
rickets, but also to address hypotheses pertinent to regional or local
patterns.
Eighteenth – mid-nineteeth century sources suggest that prevalence
of rickets around urban centres may have tended to decrease with
distance from the city centre; analysis of skeletal remains from towns
and their hinterlands could potentially test this hypothesis. It would
also be intruiging to compare rickets in skeletal assemblages where
there are conflicts between different written sources. For example, in
mid 19th century Edinburgh, Merei’s correspondant felt that rickets was
rare (Merei, 1855: 167–168), whereas another Edinburgh physician
(Stephenson, 1865) opines that it was common. It would also be in-
teresting to investigate Victorian skeletal remains from some of the
settlements coloured red and blue in the BMA survey – for example, was
rickets really rare in 19th century Southampton and York? On an in-
ternational level, the impression of 19th century commentators (e.g.
Hirsch, 1886) that rickets was rare in southern Europe could be usefully
investigated via comparative work between different European post-
Mediaeval assemblages. Given the variety of factors, including, infant
care practices (e.g. swaddling), settlement size, socioeconomic status
and presence of manufacturing industry, that may have potential effects
on rickets prevalence, biocultural understanding of its prevalence is an
inherently multivariate problem. This mandates the use of multivariate
statistical techniques rather than simple latitudinal or spatial analyses if
we are to adequately test hypotheses concerning the role of environ-
mental, social and cultural risk factors.
Given the rapidity of social and technological change during the
18th–19th centuries, well-dated remains are key to studying rickets. In
order to provide adequate statistical power for hypothesis-testing, we
need assemblages of skeletal remains with large numbers of infants and
young children; it is among this cohort that the disease is most often
active (Ritchie, 1871; Mays et al., 2006) and in which skeletal altera-
tions are most readily identified (Brickley and Mays, in press). How-
ever, it is only in recent decades that the value of studying the skeletal
remains from the 18th and 19th centuries has become more widely
appreciated (Mays, 1999), and reburial of such remains, rather than
their curation in museums so that future workers can continue learn
from them, is still depressingly common. Nevertheless, in some regions,
some fairly large skeletal samples have been excavated, studied, and in
S. Mays
some cases retained in museums. For example, in London several
thousand skeletons of individuals under 5 years of age have been re-
covered from various post-Mediaeval burial grounds. Some of these are
detailed in Supplementary Table 1: of 2438 mid 16th- mid 19th century
AD skeletons of infants and young children under 5 years of age, 311
have been diagnosed with rickets using up to date criteria (Ortner and
Mays, 1998; Mays et al., 2006; Brickley and Ives, 2008: 97–107). More
post-Mediaeval assemblages yielding large numbers of child skeletons
are becoming available from elsewhere in Britain (e.g. Brickley et al.,
2006; Waldron, 2007; Proctor et al., 2016; Rando, 2016; Headland
Archaeology, 2016), and these are likely to be augmented by significant
further material as a result of large infrastructure works that will impact
upon some major post-Mediaeval disused burial grounds (High Speed 2,
2014). In other countries in Europe (e.g. Veselka et al. 2015) and
elsewhere (e.g. Angel et al., 1987; Ellis, 2014; Nakayama, 2016), sig-
nificant post-Mediaeval assemblages are also becoming available,
raising the possibility of investigating the distribution of the disease at
an international scale.
4. Conclusion
Seventeenth – nineteenth century written sources on the epide-
miology of rickets are abundant, and cannot be ignored if the potential
of a biocultural approach to understanding the disease in that period is
to be maximised. The sources show a progression from the ex cathedra
statements of 17th and 18th century authorities, through semi-quanti-
tative studies, to statistical reports of prevalence data in the second half
of the 19th century. However, even for these last, there are substantial
obstacles that prevent effective synthesis of the data. Given these
shortcomings, it may be that the best use of these sources in biocultural
studies is to generate hypotheses that can then be investigated using
palaeopathological data.
Acknowledgements
Thanks are due to the British Medical Association for granting ac-
cess to their archives, and for permission to reproduce Fig. 1. I am
grateful to Gill Newton, University of Cambridge, for permission to cite
her unpublished 2012 conference paper. Thanks are due to the re-
viewers of this paper for their helpful comments.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the
online version, at http://dx.doi.org/10.1016/j.ijpp.2017.10.011.
References
Angel, J.L., Kelley, J.O., Parrington, M., Pinter, S., 1987. Life stresses of the free black
community as represented by the First African Baptist church, Philadelphia, 1823–41.
Am. J. Phys. Anthropol. 124, 33–44.
Armstrong, G., 1767. An Essay on the Diseases Most Fatal to Infants. T. Cadell, London.
Basit, S., 2003. Vitamin D in health and disease: a literature review. Br. J. Biomed. Sci. 70,
161–172.
Baumel, L., 1898. Distribution géographique du rachitisme. Comptes-Rendus du XII
Congrès International de Médicine, Moscou, 7 (19) −14 (26) AoÛt 1897. Section VI
Maladie de L’Enfance. Yakovlev, Moscow, pp. 110–124.
Bokay, J., 1898. Influence de l’âge. In Baumel, L. Distribution géographique du rachi-
tisme. In, Comptes-Rendus du XII Congrès International de Médicine, Moscou, 7 (19)
−14 (26) AoÛt 1897. Section VI Maladie de L’Enfance. Yakovlev, Moscow, pp. 123.
Boot, A., 1649. Observationes Medicae, de Affectibus Omissis. Newcomb & Whitaker,
London.
Brent, L., Mitchell, P.D., 2008. Rickets in Victorian London: why treatment was in-
effective for so long. In: Smith, M., Brickley, M. (Eds.), Proceedings of the Eighth
Annual Conference of the British Association for Biological Anthropology and
Osteoarchaeology. Archaeopress, Oxford, pp. 67–78.
Brickley, M., Ives, R., 2008. The Bioarchaeology of Metabolic Bone Disease. Elsevier,
London.
Brickley, M., Berry, H., Weston, G., Hancocks, A., Richards, M., 2006. The people: phy-
sical anthropology. In: Brickley, M., Buteaux, S. (Eds.), St Martins Uncovered. Oxbow,
Oxford, pp. 90–151.
7
International Journal of Paleopathology xxx (xxxx) xxx–xxx
Brickley, M., Mays, S., Metabolic bone disease. In: Buikstra, J.E., (ed.) Ortner’s
Identification of Pathological Conditions in Human Skeletal Remains, 3rd edn.,
Academic Press London (in press).
Bruenniche, A., 1867. Bidrag til Bedómmelsen af Rhachitis i Kjóbenhavn. Jahresbericht
über die Leistungen und Fortschritte in der Gesammten Medicin, 2, 305–306.
Cohn, M., 1894. Zur Pathologie der Rachitis. Jahrbuch Kinderheil Phys Erz 37, 189–248.
Ellis, M.A.B., 2014. A disciplined childhood in 19th century New York City: social
bioarchaeology of the subadults of the Spring Street Presbytarian church. In:
Thompson, J.L., Alfonso-Durruty, M.P., Crandall, J.J. (Eds.), Tracing Childhood.
Bioarchaeological Investigations of Early Lives in Antiquity. University Press of
Florida, Gainesville, pp. 139–155.
Farrar, W., 1773. A Particular Account of the Rickets in Children. Johnson, London.
Feer, E., 1897. Zur geographischen Verbreitung und Aetiologie der Rachitis. Festschrift
Eduard Hagenbach-Burckhardt. Carl Sallmann, Basel, pp. 67–134.
Floyer, J., 1715. The History of Cold Bathing, 4th edn. Innys & Manby, London.
Fordyce, W., 1777. A New Inquiry Into the Causes, Symptoms and Cure of Putrid and
Inflammatory Fevers, With an Appendix on the Hectic Fever and on the Ulcerated a
Malignant Sore Throat, 4th edn. T Cadell, London.
Frey, A., 1898. Influence de l’âge. In: Baumel, L. Distribution géographique du rachitisme.
In, Comptes-Rendus du XII Congrès International de Médicine, Moscou, 7 (19) −14
(26) AoÛt 1897. Section VI Maladie de L’Enfance. Yakovlev, Moscow, pp. 122.
Fuller, T. 1863 [1645]. Good Thoughts in Worse Times. In: Fuller T. (ed.), Good Thoughts
in Bad Times and Other Papers, Ticknor & Fields, Boston, pp. 85–177.
Gee, S., 1868a. On rickets. St Bartholomew’s Hosp. Reps. 4, 69–80.
Gee, S., 1868b. On rickets: further observations. St Bartholomew’s Hosp. Reps. 4,
265–268.
Gibbs, D., 1994. Rickets and the crippled child: an historical perspective. J. R. Soc. Med.
87, 729–732.
Glisson, F., Bate, G., Regemorter, A., 1650. De Rachitide Sive Morbo Puerili. Sadler &
Beaumont, London.
Glisson, F., Bate, G., Regemorter, A., 1651. A Treatise of the Rickets: Being a Disease
Common to Children. Trs. P. Armin. Peter Cole, London.
Graunt, J., 1662. Natural and Political Observations Mentioned in a Following Index and
Made Upon the Bills of Mortality. Thomas Rycroft, London.
Greenberg, S.J., 1997. The dreadful visitation: public health and public awareness in
seventeenth-century London. Bull. Med. Libr. Assoc. 85, 391–401.
Halliwell, J.O., 1845. The Autobiography and Correspondence of Sir Simond D’Ewes,
Bart., During the Reigns of James I. and Charles I. Richard Bentley, London.
Headland Archaeology, 2016. Excavations at St Peter’s Burial Ground, Blackburn. http://
www.headlandarchaeology.com/excavations-at-st-peters-burial-ground-blackburn/.
Hess, A.F., 1930. Rickets, Including Osteomalacia and Tetany. Kimpton, London.
High Speed 2, 2014. High Speed Two Information Paper E12: Burial Grounds, http://
assets.hs2.org.uk/sites/default/files/hb_pdf/E12%20-%20Burial%20Grounds.pdf.
Hirsch, A., 1886. Handbook of geographical and historical pathology. Diseases of Organs
and Parts, vol. III New Sydenham Society, London.
Jackson, R., 1988. Doctors and Diseases in the Roman Empire. British Museum Press,
London.
Jenner, W., 1860. On rickets. Med. Times Gazette 1, 259–262.
Johannessen, A., 1898. Bemerkung über das Vorkommen von Rhachitis in Norvegen.
Comptes-Rendus du XII Congrès International de Médicine, Moscou, 7 (19) −14 (26)
AoÛt 1897. Section VI Maladie de L’Enfance. Yakovlev, Moscow, pp. 128–9.
Küttner, R., 1858. On rickets. Va. Med. J. 10, 237–238.
Kellett, C.E., 1934. Glissonian rickets. Arch. Dis. Child. 9, 233–244.
Kissel, A.A., 1897. Ueber der Häufigkeit der englischen Krankheit in Moskau bei Kindern
unter 3 Jahren. Arch. Kinderheil 23, 279–291.
Lee, T., 1967. Historical notes on some vitamin deficiency diseases in China. In:
Brothwell, D.R., Sandison, A.T. (Eds.), Diseases in Antiquity. Charles C Thomas
Springfield, pp. 417–422.
Lorey, C., 1884. Beobachtungen über Rachitis aus dem Ambulatorium des Dr:
Christ’schen Kinderspitals zu Frankfurt a. M. Jarhbuch Kinderheil Phys Erz 22,
52–59.
Mayow, J., 1668. Tractatus Secundus De Rachitide. Hall & Davis, Oxford.
Mayow, J., 1926 [1668]. Medico-Physical Works. Trs. Brown A.C., Dobbin, L., Old
Ashmolean Reprints, Oxford, pp. 303–331.
Mays, S., Brickley, M., Ives, R., 2006. Skeletal manifestations of rickets in infants and
young children in an historic population from England. Am. J. Phys. Anthropol. 129,
362–374.
Mays, S., 1999. The study of human skeletal remains from English post-Mediaeval sites.
In: Egan, G., Michael, R.L. (Eds.), Old and New Worlds. Historical/Post-Mediaeval
Archaeology Papers From the Societies’ Joint Conferences at Williamsburg and
London 1997 to Mark Thirty Years of Work and Achievement. Oxbow, Oxford, pp.
331–341.
Merei, A.S., 1855. Disorders of Infantile Development and Rickets. John Churchill,
London.
Mey, E., 1896. Eine Studie über das verhalten der Rachitis in Riga. Jahrbuch Kinderheil
Phys Erz 42, 273–283.
Mitchell, P.D., Improving the use of historical written sources in palaeopathology. Int. J.
Palaeopathol. (in press).
Nakayama, N., 2016. Relationship betwwen burial and health status of children in 17th-
19th century Edo, Japan: an integration of burial studies and bioarchaeology, Paper
Presented at the Eighth World Archaeological Congress, Kyoto, Japan 28th
August–2nd September 2016.
Nelson, J., 1763. An Essay on Government of Children, 3rd edn. R&J Dodsley, London.
Newton, G., 2012. Parochial registration and the Bills of Mortality: case studies in the age
structure of causes of death in urban areas between 1583 and 1812. Paper for BSPS
Mortality Past and Present Symposium Celebrating the 350th Anniversary of John
S. Mays
Graunt’s Observations on the London Bills of Mortality, 29th November 2012.
Ortner, D.J., Mays, S., 1998. Dry-bone manifestations of rickets in infancy and early
childhood. Int. J. Osteoarchaeol. 8, 45–55.
Owen, I., 1889. Geographical distribution of rickets, acute and subacute rheumatism,
chorea, cancer and urinary calculus in the British Islands. BMJ 1, 113–116.
Palm, T., 1890. Geographical distribution and aetiology of rickets. Practitioner 4,
270–342.
Parkinson, J., 1640. Theatricum Botanicum Theater of Plants, or a Herball of Large
Extent. Thos. Cotes, London.
Parry, J.S., 1872. Observations on rachitis. Am. J. Med. Sci. 63, 17–52.
Proctor, J., Gaimster, M., Langthone, J.Y., 2016. A Quaker Burial Ground at North
Shields: Excavations at Coach Lane, Tyne and Wear. PCA Monograph 20.
PreConstruct Archaeology, London.
Quisling, N.A., 1888. Studien über Rachitis. Arch. Kinderheil 9, 293–355.
Rando, C., 2016. The human remains collections at the UCL Institute of Archaeology:
recent acquisitions from Eastgate Square, Chichester, Sussex. Archaeol. Int. 19,
79–83.
Ritchie, C.C., 1871. Clinical observations on rickets. Med. Times Gaz 1, 9–11.
Ritter, G., 1863. Die Pathologie und Therapie der Rachitis. August Hirschwald, Berlin.
Roberts, C., Manchester, K., 1995. The Archaeology of Disease, 2nd edn. Sutton, Stroud.
Stephenson, W., 1865. Rickets and its pathology clinically considered. Edinburgh Med. J.
10, 611–621.
Stevenson, J., 1993. Social aspects of the industrial revolution. In: O’Brien, P. (Ed.), The
International Journal of Paleopathology xxx (xxxx) xxx–xxx
Industrial Revolution and British Society. Cambridge University Press, Cambridge,
pp. 229–253.
Underwood, M., 1789. Treatise on the Diseases of Children, vol. 1 Mathews, London.
van Andel, M.A., 1927. Arnoldus Boot author of one of the first descriptions of rickets
(1649). Janus: Arch. Int. Hist. Méd. Geogr. Méd. 31, 346–358.
Veselka, B., Hoogland, M.L.P., Waters-Rist, A.L., 2015. Rural rickets: vitamin D deficiency
in a post-mediaeval farming community from the Netherlands. Int. J. Osteoarchaeol.
25, 665–675.
Waldron, T., 2007. St Peter’s Barton-upon-Humber, Lincolnshire. A Parish Church and Its
Community. Volume 2, The Human Remains. Oxbow, Oxford.
Waldron, T., 2009. Palaeopathology. Cambridge University Press, Cambridge.
Wedell, G., 1890. Arcana Fairfaxiana Manuscripta. Mawson, Swan & Morgan, Newcastle-
Upon-Tyne.
Whistler, D., 1645. De Morbo Puerili Anglorum. Boxius, Leiden.
Whistler, D., 1950 [1645]. De Morbo Puerili Anglorum. Trs. D. Smerdon, in Daniel
Whistler and the English Disease. A Translation and Biographical Note. J. Hist. Med.
5 397–415.
Woods, R., Woodward, J., 1984. Mortality, poverty and environment. In: Woods, R.,
Woodward, J. (Eds.), Urban Disease and Mortality in 19th Century England. Batsford,
London, pp. 19–36.
Zuckermann, M.K., Armelagos, G.J., 2011. The origins of biocultural dimensions in
bioarchaeology. In: Agarwal, S.C., Glencross, B.A. (Eds.), Social Bioarchaeology.
Wiley-Blackwell, Chichester, pp. 15–43.