Introduction
Palaeo-analysis of helminth eggs offers valuable insights into the past, including migrations, animal domestication, and the general health of ancient populations (e.g., Reinhard Reference Reinhard1990; Reinhard et al. Reference Reinhard, Slepchenko, Shin and Smith2018; Littleton et al. Reference Littleton, Karstens, Busse and Malone2022). These microscopic eggs can be recovered from soils, sediments, and coprolites through density separation or flotation methods (Anastasiou and Mitchell Reference Anastasiou and Mitchell2013; David and Lindquist Reference David and Lindquist1982; Dufour and Le Bailly Reference Dufour and Le Bailly2013; Reinhard et al. Reference Reinhard, Camacho, Geyer, Hayek, Horn, Otterson and Russ2019).
Over several decades, palaeoparasitological research across North America has documented a wide range of helminth taxa from both pre- and post-European archaeological contexts. Comprehensive details of this work are provided by Reinhard (Reference Reinhard, Ferreira, Reinhard and Araújo2014) and Mitchell (Reference Mitchell2023), as summarised below in the work of Horrocks, Geiger et al. (Reference Horrocks, Geiger, Bickler and Presswell2025).
In Fitz Hugh Sound, British Columbia, eggs of the fish cestode Dibothriocephalus (= Diphyllobothrium) sp., Ascaris lumbricoides, and the fish-eating mammal trematode Nanophyetus salmincola were identified in middens (Bathurst Reference Bathurst2005). In the Aleutian Islands of Alaska, a burial on Adak Island yielded A. lumbricoides and Adenocephalus pacificus eggs (Bouchet et al. Reference Bouchet, West, Lefèvre and Corbett2001), while a midden on Buldir Island contained eggs of a diphyllobothriid species (Bouchet et al. Reference Bouchet, Lefèvre, West and Corbett1999). These findings indicate widespread fish-borne helminth infection among hunter-gatherer communities, largely due to the frequent consumption of raw fish (Mitchell Reference Mitchell2023, 97). Additional Aleutian evidence includes hydatid cysts of Echinococcus sp. associated with a human skeleton (Ortner and Putschar Reference Ortner and Putschar1983).
On the mainland United States, pre-Columbian coprolite studies show that certain parasites were regionally concentrated (Reinhard Reference Reinhard, Ferreira, Reinhard and Araújo2014). For example, cestodes of Dibothriocephalus sp. occur mainly in the north, whereas acanthocephalan infections are characteristic of the Great Basin. These parasitic profiles are typical of hunter-gatherer groups (northern regions) and contrast with hookworm and Strongyloides stercoralis infections, which are more common in agricultural societies in southern North America, with more favourable environmental conditions for their transmission (Reinhard Reference Reinhard, Ferreira, Reinhard and Araújo2014). Enterobius vermicularis infections appear to be ancient among hunter-gatherers (northern) and were widespread among the Ancestral Puebloans of the Southwest (agriculturalists) and in other southern regions (Fry and Hall Reference Fry and Hall1969).
Other helminths commonly found in modern humans, such as Ascaris lumbricoides and Trichuris trichiura, were comparatively rare in pre-Columbian contexts. Additional findings include Taeniidae cestodes at the Schultz site in Michigan (McLary Reference McClary and Fitting1972), Taenia sp. eggs from Oregon coprolites (McDonough Reference McDonough2019), and hydatid cysts of Echinococcus sp. in burials from North Dakota (Williams Reference Williams1985).
Colonial-period studies have examined a range of latrine and refuse deposits, including a latrine from Newfoundland (Horne and Tuck Reference Horne and Tuck1996), barrel latrines from an early Swedish settlement in Philadelphia (Reinhard Reference Reinhard, Ferreira, Reinhard and Araújo2014), surface sediments from the Dutch colony at Albany, New York (Reinhard Reference Reinhard, Ferreira, Reinhard and Araújo2014), and latrines from Williamsburg (Reinhard Reference Reinhard1990). In contrast to pre-Columbian samples, Ascaris lumbricoides and Trichuris trichiura were present at all these sites. Eggs of T. trichiura were also recovered from privies and domestic refuse areas at the company town of Fayette, Michigan (Faulkner et al. Reference Faulkner, Cowie, Martin, Martin, Mayes and Patton2000).
Until recently, notable in this assemblage of North American studies from many different geographic regions and encompassing both pre- and post-Columbian times was the absence of Toxocara canis, the common canid parasite. Given that the domesticated dog (Canis lupus familaris), a carrier of the parasite, was the only animal introduced to the Americas by early people, this helminth species is potentially of value in understanding the timing and nature of early human migration and settlements of these continents, as shown elsewhere (e.g., Horrocks et al. Reference Horrocks, Gumbley, Bickler, Jones and Presswell2024; Horrocks, Denham and Presswell Reference Horrocks, Denham and Presswell2025, Horrocks, Peterson, et al. Reference Horrocks, Peterson, Bickler and Presswell2025).
Eggs of T. canis have been found at archaeological sites in South America (e.g., Civalero and Aschero Reference Civalero, Aschero, Miotti, Salemme and Flegheimer2003; Richardson et al. Reference Richardson, Guillén, Beckett, Kyle, Conlogue and Harper-Beckett2012). Genetic evidence for Toxocara parasites has been identified in canine coprolites from the American Bottom region of the Mississippi Valley dating to the Late Woodland period (AD 600–1050) (Witt et al. Reference Witt, Yarlagadda, Allen, Bader, Simon, Kuehn, Swanson, Cross, Hedman, Ambrose and Malhi2021). The first discovery of T. canis eggs in North America was recently at the Cloudman site in Michigan, encompassing from Early Late-Woodland (AD 500/600–1000) to Historic times, and accompanied by Ascaris lumbricoides and Taeniidae sp. (Horrocks, Geiger et al. Reference Horrocks, Peterson, Bickler and Presswell2025). Notably, North American evidence for T. canis in eastern North America dates relatively late, although dog burials have been found east of the Mississippi River dating as far back as the Archaic period, c. 6400 BC (Peres et al. Reference Peres, Deter-Wolf, Keasler and Hodge2016; Walker Reference Walker2010; Walker et al. Reference Walker, Morey and Relethford2005).
At the time of writing, palaeoparasitological studies appear to have not yet been carried out in the American Southeast region. Recent archaeological excavations at the Letchworth archaeological site, a Woodland Period ceremonial centre in northwestern Florida (Menz Reference Menz2021, Reference Menz2022) (Fig. 1), provided us with the opportunity to initiate palaeoparasitology in the region, shedding light on local human/animal association, and human and domesticated animal health. Little is currently known about these aspects, as there is a lack of faunal macro-remains found at the site due to poor preservation.
Map of Letchworth site, Florida, showing sampling locations: Block 1; EU 1, EU 4, EU 7, and EU 8 (see Table 1 to correlate individual sample numbers).
The study area and site
Letchworth (8JE337) is a Woodland Period (1000 BC–AD 1000) ceremonial center in Jefferson County, Florida, between Tallahassee and Monticello, southwest of Lake Miccosukee (Fig. 1). Analysis of diagnostic ceramics showed that occupation at the site spanned most of the Woodland period, including Deptford, Swift Creek, and Weeden Island components (Newman Reference Newman1997; Tesar et al. Reference Tesar, Swann, Miller, Damour-Horrel and Lavender2003). Letchworth was located near the center of regional, Swift Creek and Weeden Island interaction networks, situated between other large Woodland period centers including Kolomoki, McKeithen, Garden Patch, and Crystal River. The current boundary of Letchworth-Love Mounds Archaeological State Park, which contains most of the site, includes at least seven extant mounds and an extensive artifact scattering to the south of the main platform mound, Mound 1 (Fig. 1).
Excavation into the southern margin of Mound 1 (Fig. 1) produced two 14C dates of 173–172 BC and AD 236–347, the latter providing an earliest date for the construction of the mound (Harper Reference Harper2017). Coring and remote sensing suggested Mound 1 was built in three phases and could have been in use for a considerable time after this date (Seinfeld and Bigman Reference Seinfeld and Bigman2013).
Recent excavations at the site by Menz (Reference Menz2021, Reference Menz2022) focused on the approximately 500-m-wide occupation area south of Mound 1 (Fig. 1). The light construction of a domestic structure and low artifact and feature densities elsewhere at the site suggest that the inhabitants lived at the site on a short-term, temporary, or seasonal basis (Menz Reference Menz2022). Thirty-three new 14C dates from features and other contexts span the Early (1000–200 BC), Middle (200 BC–AD 600), and Late Woodland (AD 600–1000) periods, making the habitation areas at Letchworth among the most comprehensively dated of a large mound center in the region, on par with Crystal River, Kolomoki, and Garden Patch (Pluckhahn and Thompson Reference Pluckhahn and Thompson2018; Pluckhahn et al. Reference Pluckhahn, Menz, West and Wallis2018; Wallis et al. Reference Wallis, McFadden and Singleton2015).
Poor preservation of macro-remains has hindered previous attempts to assess subsistence at Letchworth. Only a small amount of highly fragmented, charred bone was recovered during the recent fieldwork at the site: a combined 26.3 g from 34 m2 of excavation (Menz Reference Menz2022). Preservation of charred macro-botanical material, mostly hickory nutshells, was only marginally better. Light fractions from flotation were exceptionally sparse.
Materials and methods
Samples were collected from various feature contexts, including feature soils, carbonized residue from ceramic sherds, and unwashed hearth rocks with adhered soils. Samples are from four locations across the Letchworth occupation areas, representing the three major components at the site, i.e., Deptford, Swift Creek, and Weeden Island I. Sample locations and contexts/dates are given in Figure 1 and Table 1, respectively, and in the work of Menz (Reference Menz2022). The time periods encompassed by the analysis (Woodland Period, c. 400 BC–AD 800) were determined based on calibrated 14C analysis (Table 1). Most features sampled have at least one corresponding 14C date (Table 1), although in some cases the dates conflict with other chronological markers such as diagnostic artifacts.
Provenance, age, and frequency (+) of helminth eggs from the Letchworth site
Note: 1associated 14C date from same feature or context; *14C date likely anomalous, does not reflect probable occupation). T.c. = Toxocara canis, D.c. = Dipylidium caninum, T. = cf. Taeniidae gen. sp., H. = Hymenolepis sp.
A sample from each of 16 features, and two samples from another feature, giving a total of 18, were analysed for helminth parasite eggs. Samples were prepared by density separation (1.8 specific gravity) with sodium polytungstate (David and Linquist Reference David and Lindquist1982; Horrocks Reference Horrocks2020). Helminth eggs have a specific gravity of approximately 1.130–1.238 (David and Lindquist Reference David and Lindquist1982). The separations were mounted on slides with glycerol jelly. Presence/absence of relevant specimens was noted.
Photomicrographs of helminth eggs were taken with a Canon EOS 600D camera mounted on a Nikon E400 microscope. Measurements were made using a calibration slide.
Results
Despite the previously recorded poor preservation of organic material at the site (Menz Reference Menz2022), four types of helminth egg taxa were recovered and identified. These comprised Toxocara canis, Dipylidium caninum, a cestode of family Taeniidae, and Hymenolepis sp.
Toxocara canis
Nematode of family Ascarididae: A total of 34 specimens of this egg type was found in 12 samples (Table 1, Fig. 2A–F). These eggs are sub-spherical, with intact specimens measuring 70–100 μm in (longest) diameter and exhibiting a pitted surface and thick shell. The usual egg size range of Toxocara canis is considered 75–90 μm. The eggs showed variable preservation, ranging from intact to degraded and/or damaged.
Nematode Toxocara canis and cestode Dipylidium caninum eggs from the Letchworth Site. Charcoal fragments abundant in some samples. B, D, E 400×, remainder 600×; scale bars = 20 μm; feature numbers (in brackets) shown in Table 1. A (20209), B, C (20007), D (20213), E, F (20229) Eggs of T. canis, showing characteristic pitted surface and thick shell. A–D relatively well-preserved; D and E damaged. Larva visible in C (arrows, folded near far-left arrow). Embryo/larval remains visible as brown mass in E (arrow). Degraded, discoloured egg shown in F. G, H (20209) Eggs of D. caninum. G shows egg agglomeration, packet not preserved. H shows degraded packet fragment with two eggs (arrows).
Toxocara canis inhabits the small intestine of canids and is usually asymptomatic, except in pups, where infection is transplacental or transmammary and may be fatal. Eggs are shed in dog faeces, becoming infectious in soil, and are then ingested by dogs to repeat the cycle. Toxocariasis, a zoonotic disease, occurs when humans ingest the eggs from contaminated soil (Harris-Linton Reference Harris-Linton2001). While T. canis cannot mature in humans, larvae may migrate through the tissues, with the rare potential to cause blindness or death. Although humans can be dead-end hosts, T. canis can only complete its life cycle by passing through the intestine of a canid. Therefore, presence of these eggs in archaeological soils/sediments is confirmation of the presence of dogs.
This nematode egg type could potentially be from native wild, or domesticated, host canids (T. canis) and native wild felids (T. cati). Native canids comprise wolves, coyotes, and foxes, and there are several native felid species in North America. Toxocara canis has, on average, larger eggs than T. cati (Panova and Khrustalev Reference Panova and Khrustalev2017; Uga et al. Reference Uga, Matsuo, Kimura, Rai, Koshino and Igarashi2000). Among the samples in this study, most of the eggs are within the larger range, lending more weight to a canid origin. Given (1) the larger egg size, (2) their high frequency at the site, (3) the fact that the American Indian dog was introduced with the first humans and had a domesticated Eurasian ancestor (Leonard et al. Reference Leonard, Wayne, Wheeler, Valadez, Guillén and Vilà2002), and (4) that the early American Indian dog had a variety of uses, e.g., as a food source, herding and hunting aid, and companion (Segura et al. Reference Segura, Geiger, Monson, Flores and Sánchez-Villagra2022), the eggs in this study are most likely, at least in part, from the domesticated dog.
At the time of writing, there is only one previous report of eggs of Toxocara sp. in pre-Columbian contexts in North America, at the Cloudman site in Michigan (Horrocks et al. Reference Horrocks, Peterson, Bickler and Presswell2025), and in South America there appears to be only one record of T. canis associated with dogs in pre-Columbian sites (Richardson et al. Reference Richardson, Guillén, Beckett, Kyle, Conlogue and Harper-Beckett2012).
Dipylidium caninum
Cestode of family Cyclophyllidea: A total of 5 specimens of this egg type was found in three samples (Table 1, Fig. 2G, H). The specimens comprised agglomerations (up to 82 μm longest axis) of up to 8 spherical eggs (21–25 μm). The outer envelope was degraded or missing. The eggs are slightly smaller than expected for this cestode (the usual range is 31–50 mm, with one database giving 25–40 μm (www.troccap.com/feline-guidelines/gastrointestinal-parasites/flea-tapeworm-feline/)), which is probably a result of taphonomic processes. Findings of D. caninum eggs from other archaeological sites have shown a similar smaller size (e.g., Horrocks et al. Reference Horrocks, Brown, Brown and Presswell2023, Reference Horrocks, Gumbley, Bickler, Jones and Presswell2024; Jiménez et al. Reference Jiménez, Gardner, Arújo, Fugassa, Brooks, Racz and Reinhard2012).
Although mainly a canid parasite, Dipylidium caninum has zoonotic potential if infected fleas are ingested by humans, with modern cases nearly always being reported in young children (Sapp and Bradbury Reference Sapp and Bradbury2020). Dipylidium caninum adults live in the small intestine of the definitive host and release gravid segments into the environment via the host’s faeces. The segments are full of egg packets, up to c. 100 μm or more in size, which may contain more than 30 eggs, which are ingested by flea larvae and develop within a cyst in the adult flea. When an infected flea is eaten by a dog, the infective stage is released into the intestine of the host and becomes an adult. Several families of cyclophyllidean cestodes (Anoplocephalidae, Paruterinidae, Nematotaeniidae, Mesocestoididae) have paruterine organs into which eggs move from the uterus before release into the environment, causing some clustering similar to that seen in Dipylidium. However, the morphology, size, and overall context of our specimens provide no evidence that they belong to any of these families; all features are consistent with Dipylidium. Possible eggs of this species have been identified at sites in Arizona and Utah (Reinhard Reference Reinhard, Ferreira, Reinhard and Araújo2014), in Mexico (Jiménez et al. Reference Jiménez, Gardner, Arújo, Fugassa, Brooks, Racz and Reinhard2012), and Puerto Rico (Wiscovitch-Russo et al. Reference Wiscovitch-Russo, Rivera-Perez, Narganis-Storde, García-Roldán, Bunkley-Williams, Cano and Toranzos2020).
Taeniidae gen. sp
These eggs are consistent with the Cestode of family Taeniidae: A total of 16 specimens of this egg type was found in six samples (Table 1, Fig. 3A–C). The eggs are mostly spherical, some ovoid, with a thick, radially striated shell, and measure 27–40 μm in (longest) diameter. The usual egg size range in the Taeniidae is considered 30–50 μm, with one source giving 25–45 μm (WOAH Terrestrial Manual 2021, Chapter 3.10.2, page 3 www.woah.org/fileadmin/Home/eng/Health_standards/tahm/3.10.02_CYSTICERCOSIS.pdf). The characteristic striations of this taxon are clearly visible in Figure 3A, although variably preserved eggs from archaeological contexts do not always reproduce as well in photomicrography as fresh eggs, and appear faint in Figures 3B and C. A similar result can be seen in taeniid eggs from the Michigan Cloudman site (Horrocks, Geiger et al. Reference Horrocks, Peterson, Bickler and Presswell2025).
Cestodes Taeniidae and Hymnenolepis sp. eggs from the Letchworth Site. 600×; scale bars = 20 μm; feature numbers (in brackets) shown in Table 1. A (20020), B, C (20209) Taeniidae eggs, with characteristic striated shell clearly visible on outer edges in A, less so on outer edges in B, C. D, E (20209), F, G (20229) Hymnenolepis sp. eggs, showing variable decay and discoloration. Larval hooks visible in A, B (arrows).
The primary hosts of taeniid cestodes are canids, with a variety of prey species serving as intermediate hosts. Modern taeniids are often associated with domesticated sheep and cattle, which were only introduced to North America following European colonisation (Crosby Reference Crosby1972). Therefore, it is likely that taeniids present in earlier contexts, such as at Letchworth, completed their life cycle using native ungulates and canids. Humans can serve as accidental or intermediate hosts, and encystment of the larval stage in tissues can lead to severe pathology or death; for example, hydatid disease caused by species of genus Echinococcus.
The exact species represented by these cestode finds remains unknown, as eggs of this family cannot be reliably distinguished by microscopy. North American hydatid cyst discoveries reported by Williams (Reference Williams1985) and Ortner and Putschar (Reference Ortner and Putschar1983) led these authors to suggest an association between dogs and pre-Columbian populations. Eggs of taeniid cestodes (Taenia spp. and Echinococcus spp.) have previously been found at pre-and post-Columbian archaeological sites in North America (Horrocks, Geiger et al. Reference Horrocks, Peterson, Bickler and Presswell2025; Reinhard Reference Reinhard, Ferreira, Reinhard and Araújo2014).
Hymenolepis sp
Cestode of family Hymenolepididae: A total of six specimens of this egg type was found in four samples (Table 1, Fig. 3D–G). The eggs are spherical and measure 36–46 μm, with the oncosphere 14–23 μm in diameter. The eggs range from fairly well preserved to degraded with discoloration.
The hymenolepid eggs are likely attributable to Hymenolepis sp. While there are two species closely associated with humans, H. diminuta and H. nana (recently reassigned to Rodentolepis), the size of these specimens suggests they belong to the latter (Fig. 3D–G). The range of sizes reported for H. diminuta is 60–80 μm, whereas the eggs of R. nana are generally smaller (40–60 μm) (Saari et al., Reference Saari, Näreaho and Nikander2019). Both species are principally parasites of rodents, with intermediate hosts being arthropods, often beetles or fleas. The eggs pass into the environment in rat or human faeces where they are ingested by the beetle host, which in turn is eaten by the mammal host, where they mature in the intestine.
There are numerous native rodent species in North America that could potentially complicate determining the origin of this egg type in paleoenvironmental studies. Given the obvious association of the sampled substrate with human occupation, a logical conclusion is that the egg is likely associated with humans or commensal rodents. Beavers, muskrats, squirrels, rabbits, and hares are rodents known to have formed part of Indigenous American diets (Kuhnlein and Humphries Reference Kuhnlein and Humphries2017). Rodent butchering by people would release eggs, as well as their random defecation around the site. Hymenolepid eggs have previously been found at pre-and post-Columbian archaeological sites in North America, in Alaska, Arizona, and Mexico (Reinhard Reference Reinhard, Ferreira, Reinhard and Araújo2014).
Discussion and conclusions
The helminth egg results in this study provide the first such pre-contact study in the American Southeast. The discovery of eggs of Toxocara canis, Dipylidium caninum, a cestode of family Taeniidae, and Hymenolepis sp. provides evidence during the Woodland Period for the probable presence of domesticated dogs, and possibly commensal rodents that may have been hunted or attracted to settlement. The parasites could have affected the health of people and their dogs.
The relatively high frequency of T. canis eggs at Letchworth (12 out of 18 samples) is consistent with that at the Cloudman site in Michigan, where all samples analysed (12) contained these eggs (Horrocks, Geiger et al. Reference Horrocks, Peterson, Bickler and Presswell2025), while the other helminth types were recorded in fewer samples. This consistency could reflect extensive and intensive human-dog association during long periods in Woodland times. The domestic dog was the only pre-Columbian domesticated animal in the Americas and arrived with humans via Beringia (Leathlobhair et al. Reference Leathlobhair, Perri, Irving-Pease, Witt, Linderholm and Haile2019). People are thought to have first arrived in North America c. 16,000 years ago and the earliest dog evidence so far is between c. 9,500 and 10,000 years ago from southern Illinois (Perri et al. Reference Perri, Widga, Lawler, Martin, Loebel, Farnsworth, Kohn and Buenger2019). Faunal remains of dogs have also been identified in Archaic period (6400–4400 BC) contexts in Tennessee and Alabama (Peres et al. Reference Peres, Deter-Wolf, Keasler and Hodge2016; Walker Reference Walker2010; Walker et al. Reference Walker, Morey and Relethford2005). Dogs were deeply integrated into North America’s Indigenous cultures, often ceremonially ritualised, and further used as hunting aids, protectors, draft animals, companions, and sources of food (often primary) and warmth (e.g., Callahan Reference Callahan1997; Oberholtzer Reference Oberholtzer2002). The Letchworth results offer a new line of probable evidence of domesticated dogs in the American Southeast, especially along the Gulf Coast, where bone material is poorly preserved.
Acknowledgements
We are grateful to the supporters of the project, including the Tallahassee resident volunteers for fieldwork. Thanks go to the Panhandle Archaeological Society at Tallahassee, Florida DEP, and State Parks, particularly George Royal, Florida BAR, NPS Southeast Archaeological Center, the Aucilla Research Institute, and the University of Michigan Museum of Anthropological Archaeology.
Financial support
The project was funded in part by the National Science Foundation through Doctoral Dissertation Research Improvement Grant 2214065, and by the Department of State, Division of Historical Resources, and the State of Florida through FL DHR Small Matching Grant 20.h.sm.200.137.
Competing interests
The authors declare none.
