INTRODUCTION
Palaeoparasitology, or archaeoparasitology, is the study of parasite infections in past populations. Analysis of ancient sediments has revealed parasite infections dating from early prehistory through the medieval period in many regions across the world.Footnote 1 Parasite remains from archaeological sites are a useful tool for understanding diet, human–animal interactions, migration and sanitation in the past, and for tracking infectious diseases throughout history. The diverse life cycles and transmission routes of various parasites are used to gain an understanding of these lifestyle factors.Footnote 2 Such parasites can be present on the surface of the body (ectoparasites, such as fleas and lice) or inside the body (endoparasites, such as intestinal worms).Footnote 3 Endoparasites will be our focus in this study and include both helminths (worms) and protozoa (single-celled organisms). Helminths can be grouped into nematodes (roundworms), trematodes (flukes) and cestodes (tapeworms). There are numerous parasites in each of these groups with varying geographic distributions, intermediate and definitive hosts, and sites of infection in the human body. In the regions conquered by the Romans, the majority of helminths for which we have evidence reside in the intestines or liver and bile ducts.Footnote 4
When intestinal parasites establish an infection in their human host, they mate and release eggs into the faeces. These can then be recovered from archaeological sediment samples that contain faeces, such as cesspits, or in actual preserved pieces of faeces, such as coprolites. In some cases these eggs need a period of time to mature in the environment, as in the case of roundworm and whipworm,Footnote 5 while in others the larvae develop in an animal intermediate host before being transmitted to their human host, where they mature into adult worms.Footnote 6 These animal hosts can be very specific, which limits the geographical distribution of parasites and requires close proximity between the animal host and humans.Footnote 7
As studies undertaken on Roman sediment samples have increased, it is becoming possible to look for trends in parasite infections across the Empire.Footnote 8 The majority of previously published palaeoparasitological work for this period is from British sites and has mainly been published in excavation volumes and site reports.Footnote 9 We are now in a position to pool and analyse the British data, allowing us to determine patterns in infection by site type. While different kinds of samples and different methodological approaches will affect the number of eggs from each taxonomic group recovered, the identification of these taxonomic groups of parasites at each site still provide a comprehensive data set. This level of analysis has not been possible for most other regions in the Empire, as relatively fewer palaeoparasitological studies have been undertaken outside of Britain.
The aim of this study is to contribute new evidence for parasite infection in Roman Britain through microscopic analysis of sediment from the pelvic area of skeletons excavated from the western cemetery of Roman London (Londinium) to understand parasite transmission better at the individual level in this community. These new data are then compared to previously published studies from Roman Britain.
ROMAN LONDON
The Roman settlement of Londinium was established in a.d. 48 and was firmly embedded in trade with the Rhineland, Gaul and the Mediterranean, with the military being a strong presence throughout.Footnote 10 As many parasites are transmitted by the faecal–oral route, it is valuable to understand key aspects of city structure that contribute to (or help limit) disease spread. Remains of wooden pipes indicate that the early town may have had some piped water to its centre, which could then provide clean drinking water to inhabitants.Footnote 11 There is also archaeological evidence for a series of large wells which could have provided water to the population without the need for piped water supply.Footnote 12 Such wells can be easily contaminated with waste in densely populated cities.Footnote 13 Residential areas in Londinium were crowded, which would contribute to spread of infectious diseases including parasites. For example, the Watling Court site (WAT78) provided evidence for multiple residential buildings separated by very narrow alleys and little green space. One of the residential buildings at the site also provided evidence for a second storey, a feature of houses that suggests increased population density.Footnote 14 Estimates of the population of Londinium have suggested that prior to the Boudican revolt (a.d. 60/61) there were around 10,000 people living there, and it grew to about 25,000–30,000 in the second century.Footnote 15
In the first decades of its existence, buildings in the town were mainly constructed of timber, with wattle-and-daub walls and earth floors,Footnote 16 the latter a feature of early Roman buildings in Britain.Footnote 17 Excavations have also produced remains of concrete floors, plaster and marble in debris from the Boudican revolt, indicating that these materials were also being used in buildings.Footnote 18 Mortar and mosaic floors are found in higher-status buildings and some houses in Londinium.Footnote 19 Building materials and urban landscapes can have a strong impact on infectious disease transmission and are thus important considerations in our understanding of parasite transmission across different settlements.Footnote 20
Londinium was surrounded by burial grounds, which show a mixture of inhumations and cremations, and were used by a variety of different status and social groups, including the military.Footnote 21 From the outset, Londinium was inhabited by people from the region and elsewhere in Britain, as well as those from Continental Europe and the Mediterranean, including migrants from North Africa.Footnote 22 Stable isotope and bioarchaeological data show the presence of first- and second-generation immigrants, some of Black African heritage, and two individuals possibly from Scandinavia were found during recent Crossrail excavations (XSM10).Footnote 23 The epigraphic and funerary datasets also provide evidence for the presence of the military community and enslaved people.Footnote 24 Therefore, it should be expected that many of the individuals in our study would have been part of this diverse population.
Environmental and stable isotope evidence for food-ways in Roman London all emphasise how different the settlement was to the rest of Britain, with evidence for the import of numerous exotic foodstuffs and the zooarchaeological data showing the influence of the military and urban dietary trends, with the dominance of cattle and pig remains, and fresh and marine seafoods being consumed. The stable isotope data echo these findings, but also show us that the diets of women and children contained less meat.Footnote 25 This diverse diet would impact the risk of zoonotic parasite infection.
MATERIALS AND METHODS
Here we present evidence for parasite infections from a third- and fourth-century cemetery located on the western outskirts of Londinium, probably lining the road which connected London to Silchester. Excavations in 1989 at 24–30 West Smithfield, 18–20 Cock Lane and 1-4 Giltspur Street (WES89) uncovered 127 burials (see fig. 1). Fourteen of these burials can be defined as ‘chalk burials’ in which the individuals were laid on, or interred within, layers of chalk placed within the grave cuts and individuals generally were orientated north–south or west–east, with adults and children represented, and the deceased often accompanied by jewellery, decorated combs or pottery.Footnote 26 The human remains are curated by London Museum and were recorded using the Wellcome Osteological Research Database (WORD). Age at death in those less than 18 years old was determined using dental development and eruption, long-bone length and epiphyseal fusion. If it was not possible to determine an age-range, individuals were recorded as ‘nonadult’. In adults (>18 years old), degenerative changes to the sternal rib end, auricular surface and pubic symphysis, as well as tooth wear were used to determine age. Individuals were assigned to age-groups or ‘adult’ if it was not possible to determine an age-range due to skeletal completeness or preservation. Sex estimation was only undertaken for adults, and based on morphological differences in the skull and pelvis, with individuals being estimated as male, female, intermediate or undetermined.Footnote 27
Sediment samples were taken during the excavation of WES89, with extant samples from 29 individuals being available for analysis at London Museum Stores. Sampling took place directly anterior to the sacrum, and in most cases, control samples came from above or below the skeleton (table 1).
This sampling information is crucial, because helminths which reproduce in the gastro-intestinal tract deposit eggs within the faeces of the infected individual. Thus, in order to find evidence of these parasites, samples that are likely to contain human faeces must be collected. In the case of skeletal remains, that means collecting sediment from the pelvic area of the skeleton, directly anterior to the sacrum (if the individual was buried in the supine position) where the intestines and any faeces contained within in them would have been located and later decomposed.Footnote 28 In order to ensure that any eggs found within the pelvis represent true infection rather than contamination of the wider burial environment, control samples need to be collected from areas away from the abdomen where we would not expect to find eggs from infection of the individual.
From each sample of pelvic sediment, a subsample of 0.2 g of sediment was processed for analysis. This subsample was disaggregated with 5–6 mL of 0.5 per cent trisodium phosphate solution for at least 2 hours. Once the sediment was fully suspended in the liquid, it was passed through a stack of microsieves of mesh size 300 μm, 160 μm and 20 μm. As the eggs of intestinal parasites that infect humans in northern Europe are generally between 25 and 150 μm in size, this should concentrate the eggs on the 20 μm mesh.Footnote 29 This material was washed from the 20 μm mesh, centrifuged (3184 g for 5 minutes), supernatant removed, and mixed with glycerol. Five slides from each sample were viewed under a digital light microscope at ×400 magnification (Olympus BX40F microscope with GXCAM-9 digital camera). If any eggs were found in the first five slides then the entire sample was viewed to calculate the concentration of eggs in the sample. Eggs were identified by their shape, dimensions, colour and special characteristics in comparison to standard reference texts.Footnote 30 For samples that contained parasite eggs, 0.2 g of sediment from the control sample were also analysed to detect any generalised contamination of the burial soil by human faeces (and so parasites). The minimum prevalence of parasitic infection in the population was calculated from the number of positive individuals as a percentage of the total number of burials studied.
RESULTS
Helminth eggs were found in the pelvic sediment of three of the twenty-nine individuals (10.3 per cent) from WES89. Ascaris sp. (roundworm) was the only species of parasite present (fig. 2). The positive individuals were 761 (adult male), 841 (adult, undetermined), and 935 (adult female). The eggs were decorticated, meaning that they had lost the mamillated surface coat generally found in modern roundworm eggs. This is often the case in pelvic sediment from ancient burials, as the mamillated coat can be stripped away by the action of insects and soil micro-organisms over time. The egg concentrations were low in the pelvic sediment samples. Five eggs were found in the 0.2 g sub-sample from male 761, three in individual 841 and two in female 935. The egg concentrations and mean dimensions of these eggs are listed in table 2. No helminth eggs were found in the control samples from the three individuals; the entire 0.2 g sub-sample was analysed for the control samples. This suggests that the eggs we did find in the pelvic sediment likely originated from the decomposed intestines of that individual, and does not indicate generalised contamination of the burial soil by faeces and parasite eggs. The eggs were identified as Ascaris sp. without a species level identification as it is not possible to differentiate between the species usually found in humans (Ascaris lumbricoides) and that usually found in pigs (Ascaris suum) using the morphology of the eggs themselves.Footnote 31 Their presence in pelvic sediment from human burials would suggest that they are more likely the human-infecting species, but both species are able to infect humans.Footnote 32 Therefore, it is safest to just refer to them as roundworm.
With three of twenty-nine individuals positive for parasite eggs, this gives a minimum prevalence of 10.3 per cent in the population. We write minimum prevalence, as it is possible that more individuals were infected with parasites at the time of their death, but those eggs might have been lost over the centuries by the action of fungi and insects in the soil.
DISCUSSION
How does the intestinal parasite infection found in the London population compare with that from Britain and elsewhere? Previous palaeoparasitological work on British sites reported over a period from 1968 to 2022 is summarised in table 3. Samples derived from sediment from pits, sewer drains, occupation layers, cesspits and latrines, ditches, wells, and pelvic sediment from burials. Each site was assigned to a type: military, urban, or rural, following the definitions of the Rural Settlement of Roman Britain Project.Footnote 33 Parasite diversity was then broadly compared between different site types to look for patterns in parasite transmission. The transmission routes, geographical distribution and anatomical sites of infection for these parasites are listed in table 4.
Note: *in Europe the species present are typically acquired from freshwater fish only.
The studies summarised in table 3 identified six different taxa of intestinal helminths found in Britain. The most common parasites identified in Britain are roundworm (Ascaris sp.), which was found at 86 per cent of sites (13/15), and whipworm (Trichuris sp.), which was found at 67 per cent of sites (10/15). Following that, but much less common, are zoonotic parasites, including beef/pork tapeworm (Taenia sp.) found at 20 per cent of sites (3/15) and fish tapeworm (Dibothriocephalus sp.) found at 13 per cent of sites (2/15). Finally, the liver flukes, including Lancet liver fluke (Dicrocoelium sp.) and common liver fluke (Fasciola sp.) were found at 13 per cent of sites (2/15), though they were never found together. Analysis of these results by site type shows that roundworm and whipworm are the most commonly identified eggs in military sites, and roundworm was the most common parasite identified in urban and rural sites (fig. 3). Urban sites have the highest taxonomic diversity overall with all parasite species found in Roman Britain identified in urban sites. The diversity in military sites was the lowest with only roundworm, whipworm, and Fasciola liver fluke being identified.
As noted above, three (10.3 per cent) of the 29 individuals from London showed evidence for roundworm (Ascaris) eggs in sediment from their pelvis. Pelvic sediment samples studied from Canterbury, York, Bletchingley, Bleadon and Churchill contained a higher prevalence of 38 per cent for roundworm (Ascaris sp.), as well as 4 per cent of individuals positive for beef/pork tapeworm (Taenia sp.), and 1 per cent for fish tapeworm (Dibothriocephalus sp.).Footnote 34 The results from Londinium are similar to Roman Italy, with a combined minimum prevalence of 9 per cent at the sites of Lucus Feroniae, Oplontis, Vacone and Vagnari (first century b.c. to fourth century a.d.),Footnote 35 and of 28 per cent in fourth- to fifth-century a.d. Florence.Footnote 36 A further study of individuals from Neolithic to Roman era Greece found a minimum prevalence of 16 per cent.Footnote 37 We describe this as a minimum prevalence, as eggs that were originally present in other individuals may have washed away or been destroyed by soil fungi and insects, so we regard these individuals as negative, when they may have been infected during life. Furthermore, it is important to keep in mind that these proportions indicate how many individuals were infected at their time of death, and are not a perfect reflection of the living population. School-aged children (5–15 years old) are the most commonly infected age group in modern populations,Footnote 38 and in ancient populations where infant mortality was quite high, those infants who survived past the age of five would have a higher life expectancy and may be found in the cemetery as adults. Thus, we may be missing a number of childhood infections, if those at high risk for infection survived these years and were buried in adulthood without an active infection. However, when considered together, such data start to give perspective on how common intestinal parasite infection may have been in the Roman world.
PARASITE DIVERSITY IN ROMAN BRITAIN
Roundworm (the only parasite found in the individuals studied from Londinium) is a soil-transmitted helminth, often grouped with whipworm (Trichuris trichiura), and hookworm (Ancylostoma duodenale/ceylanicum and Necator americanus). They are all similarly transmitted by ingestion or contact with infective eggs or larvae in soil that has been contaminated with human faeces.Footnote 39 In modern populations, soil-transmitted helminths are estimated to infect about 1.5 billion people worldwide, though their distribution is heavily concentrated in sub-tropical regions and low- and middle-income countries.Footnote 40 Roundworm is the most common parasite found in British sites overall, and the most common parasite recovered from urban sites (fig. 3). Roundworm is transmitted by the faecal–oral route. Infection is acquired when eggs are accidentally ingested from the environment, typically through unwashed fruits and vegetables, unwashed hands, or contaminated water.
Sanitation conditions are strongly linked to the transmission of soil-transmitted helminths including roundworm; thus the recovery of this parasite in 10.3 per cent of individuals studied from Londinium gives an indication of ineffective sanitation infrastructure and the presence of practices that would have allowed for the spread of faecal–oral pathogens. Removal of waste from living areas becomes more challenging with increasing population density, and adequate removal is important for preventing the spread of gastro-intestinal infections. Removing the faecal waste of the 25,000 to 30,000 people estimated to be living in Londinium in the second century required planning and systematic implementation by a regulatory authority within the settlement.Footnote 41 Without regular emptying of cesspits and organised areas for chamber pots to be dumped, faecal waste could easily contaminate water sources such as the wells.
The evidence for latrines in Britain varies based on the type of site. In military camps and forts, latrines were often set up, even on a semi-permanent basis. In temporary camps, latrines would have been trenches dug into the ground that could then be filled over, as opposed to more permanent latrines built as part of bath complexes or as independent buildings in permanent forts.Footnote 42 The pit that was studied from Ambleside (Cumbria) may have been one of these temporary latrines at the fort. Less work has been done looking at the location of private latrines in Britain, but we may expect that many non-elite houses did not have private latrines, and in these cases chamber pots were likely used. There is evidence for the use of chamber pots across the Empire.Footnote 43 Cesspits were also common throughout the Empire, and in homes that did have private latrines, these were likely small cesspits located inside the house or in gardens outside the house, as has been found in the north-western provinces.Footnote 44 These cesspits and chamber pots would have needed to be emptied, and the process for collecting and removing human excrement from the city could have created opportunities for individuals to come into contact with faecal material of others and for wider contamination of settlements.Footnote 45
Flooring and building techniques can influence pathogen transmission, and in particular soil-transmitted helminths like roundworm. As discussed above, there is evidence that early Roman period buildings in Britain were constructed with earth-beaten floors.Footnote 46 Worrell and colleagues found that having finished household floors was a protective factor against soil-transmitted helminths in modern populations in Kenya.Footnote 47 Earth-beaten floors may have been harder to wash compared to those finished with materials such as stone or concrete, and could easily be contaminated with faecal material on shoes or spills from chamber pots. Furthermore, the numerous flies known to be present in Roman settlements could have resulted in helminth eggs being moved around the home and contaminating other areas such as food-processing areas.Footnote 48
Interestingly, no zoonotic parasites (those transmitted from non-human animals to humans) were found in the pelvic sediment from individuals from the western cemetery despite the parasite diversity at urban sites in Britain being higher than any other site type (fig. 3). Zoonotic parasites that have been found in other urban sites in Britain include beef or pork tapeworm (Taenia saginata or Taenia solium, respectively), fish tapeworm (Dibothriocephalus sp.) and liver flukes (Fasciola sp. and Dicrocoelium dendriticum).
In fact, the majority of evidence for tapeworms, both fish and beef/pork tapeworms, comes from urban centres, in particular Londinium. This may be a result of access to a larger variety of foods in major centres. Fish tapeworm infection is acquired in northern Europe through the consumption of freshwater fish that carry larvae.Footnote 49 Analysis of sulphur isotope values in Roman human remains suggests that freshwater and marine fish were consumed,Footnote 50 and faunal remains reveal that the most common species of fish recovered from British sites are eel, herring, plaice, cyprinid and salmonid.Footnote 51 Dibothriocephalus sp. can be found in salmon and cyprinids; thus consumption of these fish could have resulted in infection if they were eaten raw, smoked or undercooked. Marine fish were generally preferred over freshwater fish in the Roman period.Footnote 52 If fish sauce (garum), a popular condiment, was produced from freshwater fish in this area of Britain, it is possible that it could transmit fish tapeworm if salting conditions were not adequate to kill larvae.Footnote 53 Recovery of fish bones and amphora in Britain has provided evidence for importation of fish sauce and pickled fish from Iberia starting in the Iron Age and increasing into the Roman period.Footnote 54 These amphorae become less common in the second century, and it has been suggested that this may be a reflection of fish-sauce production locally within Britain.Footnote 55 The only evidence for fish tapeworm in Britain comes from Londinium and Canterbury. There was possible evidence for garum or liquamen production at Peninsular House, a Roman waterfront site in Londinium.Footnote 56 However, the fish remains found suggest that garum was made from herring and sprat, which are not common intermediate hosts of Dibothriocephalus sp. Thus, it may be more likely that fish tapeworm infection was acquired from ingestion of freshwater fish rather than fish sauce imported into Londinium.
Beef or pork tapeworm have also mainly been found in major urban centres including Londinium and Canterbury (table 3). These eggs can represent infection with beef tapeworm (Taenia saginata) or pork tapeworm (Taenia solium). It is not possible to differentiate the two species based on the morphology of the eggs. Both have very similar life cycles.Footnote 57 Eggs are ingested by grazing cattle or pigs and the larvae develop in the tissues of the animal and can cause infection when the animal is eaten. In order to prevent infection, the meat needs to be cooked to at least 56–65 degrees Celsius.Footnote 58 Pig remains make up a higher proportion of faunal assemblages in Londinium, similar to other urban sites, compared to other sites in Roman Britain.Footnote 59 Numerous cattle scapulae have been found in Britain with a hole through the centre which suggests that shoulders were hung to be dried or smoked.Footnote 60 Smoking and drying are not very effective at killing parasite larvae, and this could be another practice leading to beef tapeworm infection.Footnote 61 Higher levels of pig remains coupled with the practices of preserving cattle meat mean it is possible that the Taenia sp. eggs found in Londinium could have come from infective beef or pork.
Liver flukes found in Britain include Dicrocoelium dendriticum and Fasciola sp. Only Dicrocoelium has been previously found in Londinium. The evidence for Fasciola and Dicrocoelium in Britain was found in occupation layer soil, pits and ditches, none of which can be confidently linked to human ingestion of eggs or human infection (see table 1).Footnote 62 While both Fasciola and Dicrocoelium can infect humans, they can also result in pseudoparasitism, when a person eats the liver or intestines of an infected animal, so that the parasite eggs are mixed in with the human faeces, but without that human actually being infected.
HEALTH AND DISEASE IN ROMAN BRITAIN
The impact on health from roundworm infection in these individuals can be quite varied depending on worm burden (the number of worms one individual carries) and various host factors such as health status, immune function and co-existing infections.Footnote 63 A higher number of worms carried within one's intestines causes more severe symptoms, although it is important to note that in many cases an individual may be asymptomatic with these infections. The health impact of intestinal helminths can be considered in a stepwise fashion. First, health impacts can be reversible. For example, in children infected with roundworm, these infections can negatively impact growth. However, if the infection is cleared, their growth may catch up, so the impacts on health are potentially reversible.Footnote 64 With chronic infections or continual reinfections, however, children can experience stunting of growth from which they do not recover. Infections also cause malnutrition which can be reversible or chronic, and this occurs through multiple mechanisms. The worms themselves result in decreased nutrient absorption and intake as a result of decreased appetite and abdominal symptoms.Footnote 65 Malnutrition has widespread effects from growth loss, decreased physical fitness, impaired cognitive development in children, and increased susceptibility to infection.Footnote 66 Helminth infections including roundworm can also cause acute and life-threatening illness in some cases. Roundworm can cause bowel obstruction, for which children are at higher risk, given their smaller intestinal lumens. Through blocking of lumens within the gastro-intestinal system, roundworm can also cause appendicitis, cholecystitis, and pancreatitis which can be life-threatening if untreated, as would have been the case in the Roman period.Footnote 67
The palaeoparasitological analyses undertaken on British sites so far give us some evidence for the distribution of parasite infections in these communities. One recent British study found that for roundworm alone, Roman females and individuals over 13 years of age showed slightly higher rates of infection; however, neither result was statistically significant.Footnote 68 The pelvic sediment samples studied from Londinium in the current study only identified roundworm infections in adult individuals (one male, female and undetermined sex). While this seems to suggest that adults were commonly infected in this community, only four of the twenty-nine individuals studied were non-adults between 6 and 11 years old. With these types of analyses it is important to remember that these pelvic soil samples are from the non-survivors and they do not necessarily reflect the living population.Footnote 69 Despite these limitations, we would expect similar individual-level distributions of infections to those seen in most endemic communities today, as infection rates are very consistent between communities around the world. In modern communities, often a small proportion of a population carries the largest burden of worms. In particular, school-aged children are at highest risk for infection, which usually decreases in adulthood.Footnote 70 Factors that may contribute to higher rates of infection in this age group include playing outdoors in areas where soil is contaminated, defecation in open areas and decreased hand washing.Footnote 71 Further analysis of larger series of pelvic soil samples will be needed to understand the distribution of parasite infections better at the individual level in Londinium.
CONCLUSION
The excavations of the western cemetery produced new evidence for roundworm (Ascaris sp.) infection in individuals from Londinium. As elsewhere in the Empire, soil-transmitted helminths, roundworm and whipworm, were the dominant parasites in Britain and are the only type of parasite found in the individuals studied in the sample of individuals from the cemetery. Transmission of roundworm in Londinium was probably due to poor sanitation, as well as the management and reuse of faecal material in agriculture.Footnote 72 In Britain, there is also evidence for zoonotic parasites including tapeworms and liver flukes, although these were not found in this study. Our results, coupled with other studies of populations in Europe, suggest that perhaps a minimum of 10–30 per cent of people living in Britain in this period may have been infected by intestinal parasites. The evidence for parasites across Roman Britain suggests that the health impacts of parasites should be taken into consideration in discussions of infectious diseases in individuals living in this time period. Infections with these parasites were likely to cause nutrient deficiencies, stunting of growth and impaired cognitive development, as well as acute mortality from abdominal complications of infection.Footnote 73
ACKNOWLEDGEMENTS
We would like to thank the editors and reviewers for their advice about the manuscript, and we are grateful to London Museum for granting us permission to study the pelvic samples, and for the help and support of Archive staff for facilitating access to the site archive.