Although the European medicinal leech (Hirudo medicinalis L. 1758) is one of the best-known members of the Hirudinea due to its use in phlebotomy, this species has been confused with the Mediterranean taxon H. verbana Carena 1820. Here we describe the morphology of adult and juvenile H. medicinalis and document its genetic distance to H. verbana, using newly acquired mitochondrial DNA-sequence (cytochrome c oxidase subunit I, CO-I)-data from specimens collected in Germany. Our CO-I analysis shows that H. medicinalis and H. verbana differ by 9.4 %. Hence, the original Hirudo-population diverged ca. 10 million years ago so that today two geographically separated biospecies exist that co-occur in only a few natural habitats. We analyzed the behaviour of adult H. medicinalis, but could not find differences with respect to its sister taxon H. verbana. Finally, we summarize the occurrence of H. medicinalis in Central Europe and conclude that this once widely distributed freshwater species largely disappeared in many countries. We suggest that the loss of natural freshwater ecosystems, with flat, warm banks, and amphibians (frogs, newts and toads) as preferred host organisms for the juveniles, are largely responsible for the decline of H. medicinalis in Northern Europe.

Endangered species, evolution, Hirudo medicinalis , medicinal leeches, taxonomy

Among the currently ca. 14 000 accepted species of Annelida (segmented worms) found worldwide in freshwater, marine and terrestrial ecosystems, Linnaeus (1758) first described the two most well-known forms: the earthworm (Lumbricus terrestris) and the medicinal leech (Hirudo medicinalis). However, this Linnaean system of the “classis Vermes”, with the three orders “Intestina, Mollusca and Testacea”, was soon replaced by Lamarck’s more detailed account of classification of the “lower animals” (Vinarski 2014). In Vol. 5 of a series of monographs on the systematics of invertebrates, Lamarck (1818) introduced the class Hirudinea (leeches), and listed two European species, Hirudo medicinalis, and “H. sanguisorba”. While the first taxon, the type species of the class Hirudinea Lamarck, 1818 (sangsue médicinale, i.e., the medicinal leech), is still accepted today as a valid species, Lamarck’s “H. sanguisorba” remains an enigma.

During subsequent decades, notably when the use of leeches in phlebotomy (bloodletting) became very popular throughout Europe (ca. 1850), numerous “varieties” of “H. medicinalis” were distinguished by naturalists as well as practitioners (Herter 1936, 1937).

In his classic monograph on leeches, Mann (1962) adopted this interpretation and wrote that H. medicinalis must be regarded as a highly variable species. Accordingly, Herter (1968) and Sawyer (1986), in their influential books, argued that all colour variants of European medicinal leeches should be assigned to the taxon H. medicinalis Linnaeus, 1758 (syn. H. officinalis Savigny, 1822; see also Hechtel and Sawyer 2002). However, based on detailed analyses of pigment patterns and DNA-sequences, it was documented that H. officinalis is not a “colour variant of Linnaeus’ type species”, but a separate taxon, the Mediterranean medicinal leech H. verbana Carena, 1820 (Nesemann and Neubert 1999; De Salle et al. 2005; Kutschera 2004, 2006, 2007, 2012a, 2012b; Trontelj et al. 2004; Siddall et al. 2007; Phillips and Siddall 2009; Elliott and Kutschera 2011; Elliott and Dobson 2014). Utevsky and Trontelj (2005) provided a key to all known European species in the genus Hirudo, and Kutschera (2012a, 2012b) summarized their geographical distribution.

In the present article, we describe the morphology of juvenile and adult H. medicinalis-individuals, add information on its evolutionary distance to its sister taxon H. verbana, and summarize observations on the behaviour, ecology and distribution of this endangered species.

Adult and juvenile European medicinal leeches (H. medicinalis) (plus cocoons) were obtained from undisturbed habitats of eastern Germany (Elliott and Kutschera 2011), and specimens of Mediterranean medicinal leeches (H. verbana) were purchased from a commercial supplier (Sudak, Tr-59560 Murefte Tekirdag, Turkey) (Kutschera and Roth 2005). The leeches were kept in aqua-terraria (90 x 40 x 60 cm, depth of the pond water ca. 10 cm; temperature 22 to 26 °C), and observed/photographed alive. Specimens of H. medicinalis were killed by adding 80 % ethanol to the water, so that the animals were preserved in their non-contracted, natural shape, and photographed. Extraction of DNA from part of the posterior sucker, sequencing of a fragment of the mitochondrial gene cytochrome c oxidase subunit I (CO-I), and phylogenetic analyses, based on newly acquired (and deposited) GenBank-data were performed as described (Kutschera et al. 2007, 2013; Wirchansky and Shain 2010).

The historical use, ecology, genetics and conservation of medicinal leeches was recently summarized (Elliott and Kutschera 2011; Elliott and Dobson 2014), and the present discussion is based in part on these extensive reviews. As noted in the Introduction, Hirudo medicinalis was once abundant in Northern Europe, from Ireland in the west to the Ural Mountains in the east, and from Southern Scandinavia to the countries bordering the Mediterranean (Elliott and Tullett 1984, 1986, 1992), where the Southern species H. verbana occurs (Fig. 8A, B). It is now rare throughout Western Europe, and endangered in many countries (Wells and Coombes 1987; Utevsky et al. 2008, 2010; Kovalenko and Utevsky 2012; Petrauskiené et al. 2011; Westendorff et al. 2008; Sawyer 2013a, 2013b; Trontelj and Utevsky 2005, 2012).

Large numbers of H. medicinalis were obtained from the wild in the 18^th and early 19^th centuries, and towards the end of this period, they were already scarce in many countries. This demand for medicinal leeches was not restricted to Europe. Hirudo medicinalis does not occur naturally in North America, and large numbers were imported from Europe into the United States in the 18^th and 19^th centuries. Several attempts were made to rear this species in the US, without positive results (Elliott and Kutschera 2011). As medicinal leeches became more difficult to find in the 19^th century, the indigenous supply was supplemented by importations of other species of medicinal leeches from outside Western Europe. There was also the development of ‘leech farms’, especially in France and Germany. As late as 1890, a leech farm near Hildesheim in Germany was breeding between three and four million individuals per year (Herter 1968). Leech farms still exist today but, unfortunately, they often rely on imported leeches from Southeastern Europe and Turkey. These imports are often not H. medicinalis, but the closely-related species, H. verbana, which has been confused with the ‘true’ medicinal leech (Michalsen and Roth 2006).

H. verbana was first described from Lago Maggiore in Northeast Italy (Latin: Lacus Verbanus) by Carena (1820) and later regarded as a sub-species of the European medicinal leech (H. medicinalis ssp. officinalis). Living, adult H. medicinalis and H. verbana (as well as the juveniles) have very distinct markings (Figs 1, 5). These two closely related leech species, which diverged ca. 10 million years ago from a common ancestor, were confused or both labelled as “Hirudo medicinalis”, until a close examination of their morphology, combined with breeding studies and DNA-sequencing experiments, yielded unequivocal proof that they are distinct, reproductively isolated taxa (Nesemann and Neubert 1999; Trontelj et al. 2004; De Salle et al. 2005; Kutschera 2004; 2006, 2007, 2012a, 2012b; Siddall et al. 2007; Phillips and Siddall 2009).

Earlier reviews of the literature on the ecology of Hirudo medicinalis showed that there was surprisingly little quantitative information on medicinal leeches in the wild and most of the numerical values were from laboratory studies (Mann 1962; Herter 1968; Elliott and Mann 1979; Sawyer 1986; Petrauskiené et al. 2011). Representative data, compiled from Davies and McLoughlin (1996), are summarized in Table 2. The typical natural habitat is a eutrophic pond with a muddy substratum, littoral vegetation, and a high summer temperature. It should also be a breeding site for amphibians (frogs, toads and newts). Although H. medicinalis and H. verbana are often reported as feeding almost exclusively on the blood of mammals (cattle, horses, deer, humans), they will also suck the blood of fish, water birds, and especially amphibians, both the adults and their larvae. Hoffman (1960) observed adult leeches feeding on toads (Bufo bufo L.) in April when the latter returned to ponds to breed. The leeches attacked both male and female toads, often when the male was grasping the female in the ‘nuptial embrace’. Large numbers of toads died from loss of blood and their corpses provided food for another leech species, Haemopis sanguisuga. Tadpoles as well as juvenile newts are especially important for young medicinal leeches that are unable to pierce mammalian skin for the first two feedings. We corroborated Hoffman’s (1960) findings that were extended by Benecke (2009), Manzke and Winkler (2012) and Winkler and Manzke (2014) (Fig. 7). These authors provided, together with Jueg (2009), evidence for the occurrence of as yet undiscovered relict-populations of H. medicinalis in Eastern Germany.

Six decades ago, laboratory studies showed that the preferred temperature of H. medicinalis in a gradient of 7 to 43 °C was 21 °C (Kaiser 1954), a value midway between the 50 % and 90 % active leeches, as documented in a more recent study (Elliott 2008). Optimum temperature ranges for growth (22 to 25 °C) and breeding (25.5 to 27.5 °C) in the laboratory were similar to predicted values for maximum activity in the field. At 39 to 43.5 °C, the upper lethal range is remarkably high for this species (Kaiser 1954). These high temperature requirements have important implications for the survival of H. medicinalis in the wild: the leeches were unable to reproduce and survive in many water bodies simply because of the low water temperatures (Elliott and Tullett 1986).

A number of explanations have been proposed for the loss of many populations of H. medicinalis in Northern Europe, and these should all be considered in combination. Extensive over-collecting for blood-letting in the nineteenth century is frequently blamed, but used leeches were regularly discarded into the nearest pond or stream and thus may have enabled the survival of this species in the countryside. Contemporary collecting for experimental biology, medical use and pharmaceutical needs is probably a serious threat because the leeches are destroyed, often in large numbers (Shain 2009). Although leech farms offer an obvious solution, this only works if the commercial suppliers actually rear leeches, rather than importing them and thereby reducing populations in the wild (Michalsen and Roth 2006).

A reduction in the availability of suitable vertebrate hosts is another possible reason for the decline in countries where troughs are now used instead of ponds for the watering of cattle and horses. Changes in land use not only caused the loss of ponds but also isolation of the remaining freshwater ecosystems, even to wild animals such as deer, and this may have contributed to a reduction in blood meals from this source. However, there are still many parts of Europe where wild animals such as deer are plentiful, and therefore the almost complete absence of H. medicinalis in these areas is not due to a lack of mammalian hosts.

Davies and McLoughlin (1996) proposed the plausible hypothesis that the declining abundance of field populations of the European medicinal leech could be the result of lower available energy for growth, reflecting leeches now feeding predominantly on amphibian blood of lower energetic value than mammalian blood. This conclusion was supported by slow-growing wild populations of leeches from Dungeness, UK. A serological test was positive for 128 blood meals and showed that most leeches were feeding on amphibian blood with smaller numbers feeding on fish and birds, and only one leech sucking mammalian blood (Wilkin and Scofield 1990, 1991a, 1991b). In a Lake District tarn, only the larger mature leeches (>3.5 g) had fed on mammalian blood, and the proportion of mature leeches feeding on mammals varied from 19 to 26 % among years (Elliott 2008). The most important sources of blood for all leeches in the tarn were probably amphibians, such as newts, frogs, toads and their tadpoles (Fig. 7). Therefore, the slow growth of the leeches could be partially caused by the scarcity of mammalian blood in their diet. Leeches were observed feeding on horses that had waded into the tarn. They never fed to satiation, as seen in the laboratory when offered bovine blood in a sausage skin. Soon after a horse left the water, the leech detached and rapidly crawled back into the water. In wild populations, satiated leeches were never found and it was concluded that the annelids were feeding a little and often, rather than to satiation. A similar conclusion was reached for the population at Dungeness (Wilkin and Scofield 1990, 1991a, 1991b).

Water temperature will also affect the growth of H. medicinalis. Fast-growing leeches that attained maturity after only 289 days were kept at a constant 20 °C (Davies and McLoughlin 1996) (Table 2). This is just above the threshold temperature of 19 °C for most leeches to be swimming and searching for a host in a Lake District tarn (Elliott and Tullett 1986). Water temperature in the tarn exceeded this value on only 100 to 120 days from April to September and was thus a limiting factor for feeding and growth. The high temperature requirements of medicinal leeches (H. medicinalis and H. verbana) impose limitations on their distribution and occurrence. Therefore, the absence of these species from many water bodies may be due partially to the relatively high temperatures required for swimming activity, feeding, growth and breeding, as well as the scarcity of mammalian hosts. It should be noted that there has been a loss of many small, shallow ponds throughout Western Europe, and these are often the ideal habitat for medicinal leeches, especially if they contain amphibian species and water birds. In South-eastern Poland, Buczyńsky et al. (2014) found adult, breeding H. medicinalis-individuals, inclusive of cocoons, in the nests of water birds (Fig. 8B). This finding is in accordance with earlier observations of Herter (1936), who suggested that medicinal leeches may feed on nestlings of aquatic birds, notably when they fall into the water and attract hungry leeches (Young et al. 1981). Buczyńsky et al. (2014) suggest that nests may be important secondary habitats for medicinal leeches, but more observations in aquatic ecosystems, where natural leech populations occur, are necessary to support this hypothesis. These observations indicate that man-made losses of habitat could be also responsible for the demise of some populations of H. medicinalis (Grosser 2004; Elliott and Kutschera 2011; Elliott and Dobson 2014; Utevsky et al. 2008, 2010; Kovalenko and Utevsky 2012).

Finally, we want to point out that, although the distinctive features between H. medicinalis and H. verbana are obvious (Fig. 1) and have been described repeatedly in the literature (Nesemann and Neubert 1999; Trontelj et al. 2004; De Salle et al 2005; Kutschera 2004, 2006, 2007, 2012a, 2012b; Siddall et al. 2007; Phillips and Siddall 2009; Elliott and Kutschera 2011), Hirudo medicinalis is still confused with H. verbana and other leech species. For instance, Reece et al. (2011) depicted an adult H. verbana, but labelled the individual as “H. medicinalis”. In March 2012, the company Leeches U.S.A. Ltd. (Westbury, NY) sold “Hirudo medicinalis” for a research project to be conducted in the Weisblat-lab at the University of California (Berkeley). However, an inspection of the first author of this article (U. K.) revealed that all the specimens used for behavioural studies were unequivocally H. verbana Carena 1820 (see De Salle et al. 2005 for a DNA barcoding-identification for these mis-labelled medicinal leeches). Starr et al. (2011), Sartor et al. (2013) and Russel et al. (2014) depicted leech species, and described the individual as “H. medicinalis”, but in reality the taxa were the distantly related Hirudinaria mallinensis and Malagabdella sp., respectively (see Table 1).

Hence, despite the fact that the European medicinal leech is, in addition to the taxonomically diverse earthworm Lumbricus terrestris (James et al. 2010), one of the most popular known annelids, the unique phenotype of this beautiful “annelid with character” is still widely unknown, even among biologists.

We thank Mr. M. Aurich (Biebertaler Blutegelzucht, Germany) for the provision of living medicinal leeches and Mr. U. Manzke for the photograph depicted in Fig. 6. This project was supported by the Bundesministerium für Wirtschaft und Technologie (Förder-Kz.: 03 ESFHE 021) and the Alexander von Humboldt-Stiftung (Bonn, Germany; AvH-Fellowship Stanford/2012 to U. K.).