A review of the status of the New Zealand flatworm in the UK

The ‘New Zealand flatworm’, Arthurdendyus triangulatus, is a native of the South Island of New Zealand, which has established in the UK, Ireland and the Faroe Islands. In its introduced range, it is a predator of lumbricid earthworms. To assess the impact of A. triangulatus on earthworm species, flatworm distributions were manipulated into ‘high’, control and ‘low’ densities within a replicated field experiment. Earthworm biomass in the ‘high’ flatworm density treatment was significantly lower than the control or ‘low’ treatments. This was due to a reduction in the anecic species Lumbricus terrestris and, to a lesser extent, Aporrectodea longa. There was little evidence of negative effects on other earthworm species, with even a weakly positive relationship between flatworm density and epigeic biomass. Principal components analysis showed a clear separation of anecic species from A. triangulatus, but the epigeic species Lumbricus festivus and Lumbricus rubellus grouped with A. triangulatus, suggesting that they could be benefitting from reduced intraguild competition. Flatworm densities of 0.8 per m2, comparable to natural infestations in grassland, were predicted to give a reduction in total earthworm biomass of c. 20 %. The bulk of this was comprised of a reduction in anecic species biomass. In particular, it is considered that A. triangulatus poses a serious risk to L. terrestris populations, with implications for soil functioning and indigenous earthworm-feeding wildlife.

Invasion of agricultural land by the earthworm predator, Arthurdendyus triangulatus (Dendy) : The 7th international symposium on earthworm ecology · Cardiff · Wales · 2002

Invasion of agricultural land by the earthworm predator, Arthurdendyus triangulatus (Dendy)The 7th international symposium on earthworm ecology · Cardiff · Wales · 2002

The New Zealand flatworm Arthurdendyus triangulatus (=Artioposthia triangulata) is an example of an invasive organism that, by reducing lumbricid earthworm populations, could have a major impact on soil ecosystems in Britain and the Faroe Islands. How it was introduced into the British Isles is not known, but like many invasive species, it is suspected that it was introduced by humans and was associated with the trade between New Zealand and Britain. Once established in Britain it found in the large, readily available earthworm population a niche that it could exploit. The microclimate of the forests in the center and south of the South Island of New Zealand from whence the flatworm came is similar to that in parts of the British Isles and consequently conducive to its survival. Although when compared with many other invertebrate introductions (e.g., insects) the flatworm's rate of increase has been slow, a retrospective study strongly suggested that, in Scotland, they spread from botanic gardens to horticultural wholesalers, then to domestic gardens, and only latterly did they invade agricultural land. As with other invasive species, the application of the CLIMEX computer model and geographic-information-system techniques have been used to try to ascertain its potential distribution within both the British Isles and Europe. The data suggest that A. triangulatus could become established in northwest continental Europe, and Denmark, Iceland, and Sweden have added it to their quarantine pest lists. The fact that A. triangulatus is the only one of 12 alien terrestrial planarians in Britain to be considered a pest suggests that this group of invertebrates is behaving in a fashion similar to other invasive organisms and obeys the "tens rule" suggested by M. Williamson. Since within the period 1995–1997 another three species of predatory terrestrial planarians have been recorded in Scotland it is evident that the existing phytosanitary regulations must be more strongly enforced or new ones introduced if the spread of terrestrial planarians is to be halted. For reprints of this Invited Feature, see footnote 1, p. 1259.

The New Zealand flatworm, Artioposthia triangulata, and the Australian flatworm, Caenoplana alba, have become quite widespread in the British Isles since their introduction, probably in the early 1960s. They are considered as pest species since they eat earthworms and consequently may affect soil structure and fertility. The distribution of the two species has been recorded by two surveys: a Scottish survey between 1991 and 1993, and a national survey, MEGALAB WORMS, in March 1995. The latter was organized as part of National Science Week by BBC TV Tomorrow's World, The Daily Telegraph newspaper and the Office of Public Service and Science. Results of both surveys are presented as distribution maps of confirmed records. New Zealand flatworms are common and widespread in Northern Ireland and Central Scotland, and occur less commonly in other parts of Scotland, and in Northern England. There are a few scattered records from Southern England. The Scottish survey has resulted in retrospective distribution maps. Australian flatworms are widespread in South West England, and are found in several places in North West England. There are also records from South and North Wales, Northern Ireland and one from Southern Scotland. There were many false and negative reports. There are records of both species from the Republic of Ireland.

SummaryThe distribution of the New Zealand flatworm (Artioposthia triangulata) in Scotland was surveyed between July 1991 and February 1993. There were 348 records from domestic gardens, 56 from botanic gardens, garden centres and nurseries, with only 13 from farms. Although most of the records came from around the major cities the flatworm was found to have become established throughout the Scottish mainland and some of the Islands, e.g. Bute, Gigha, Orkney and Skye.The impact of the flatworm on earthworm populations in agricultural land in Scotland was, as yet, found to be minimal but the fact that seven adjacent farms near Dunoon were infected suggested it could be spread from farm to farm and that in the West of Scotland it could become widespread in agricultural land.

The New Zealand flatworm (Arthurdendyus triangulatus), which is an obligate predator of native earthworms, is an alien species to the British Isles and is widely distributed in Scotland. What little is known about its biology under field conditions is mainly from Northern Ireland. Samples taken from single sampling dates have shown A. triangulatus to have an aggregated distribution. To determine the spatio‐temporal distribution of A. triangulatus, a grass field in western Scotland was intensively sampled over a 16‐month period. Data indicate an increase in flatworm numbers and seasonal trends in body weight, the appearance of egg capsules and hatchlings. Results also showed that spatially, the flatworms were aggregated, but this was a transient phenomenon over the period of the experiment. The distribution of egg capsules in the study area was strongly aggregated and related to the appearance of hatchlings. Flatworms may aggregate in areas where soil moisture is optimal for the survival of the New Zealand flatworm.

SummaryThe internal transcribed spacer regions (ITS1 and ITS2) including the 5.8S region of the ‘New Zealand flatworm’,Arthurdendyus triangulates, are 1004 base pairs in length. Restriction fragment length polymorphism analysis of PCR products (PCR‐RFLP) was conducted onA. triangulatesspecimens from 45 locations in Northern Ireland, Scotland, England and New Zealand. Seven restriction endonucleases(Alu I, Rsa I, Sau3A I, Cfo I, Nde I, Dde I, and Mbo I)were used to reveal intraspecific variation. Analysis of molecular variance revealed the presence of population genetic substructuring, with most genetic heterogeneity present between populations rather than between individuals or geographical regions. No distinct differences were found between Northern Irish and Scottish populations but phylogenetic analysis supports the hypothesis of multiple introductions from New Zealand. There was no significant relationship between genetic distance and geographic distance, as would be expected for natural spread, indicating that this species is largely anthropochorous, even in parts of New Zealand.

SummaryThe density of the New Zealand flatworm,Artioposthia triangulata, found between 1993 and 1995 on the surface of the soil under various types of debris scattered in four allotments, was 1–2 m−2. The type of debris did not markedly affect the density of the flatworm and the distribution of the flatworm within the allotment studied was significantly clustered. Fluctuation in numbers of specimens under different sets of compost‐filled plastic sacks varied in a similar manner to one another. The flatworm was most abundant in the upper 100 mm of soil and its egg capsules were most numerous between 100 and 200 mm below the surface. The flatworm was as abundant under the soil surface shelters as it was beneath shelters buried below the soil surface.

Comparison of the earthworm population of a garden infested with the Australian land flatworm (Australoplana sanguinea alba) with that of a non-infested garden

Status and public perception of the ‘New Zealand flatworm’, Artioposthia triangulata (Dendy), in Northern Ireland

SummaryThe flatwormArtioposthia triangulatawas found, from studies using weighted down plastic sheeting, to move predominantly through the soil rather than over it and to use earthworm burrows. Under compost‐filled plastic sacks the flatworm was most active at night although its numbers were similar to those during the day. The transfer of specimens into an area covered with weighted plastic sheeting had no lasting effect on their numbers. The flatworm was regularly removed over 6 years from under paving stones, sheets of newspaper and cardboard placed on the ground in a garden. The rise and fall of the numbers of the flatworm under this debris suggest a predator‐prey periodicity between the flatworm and earthworms of 3 years.

Studies have shown the New Zealand flatworm to be a predator of earthworms and to be widely distributed throughout most of Ireland and Scotland. Except for one record in 1965, it had not been found in England until, in December 1992, it was positively identified from a garden centre near Manchester. Viable and empty egg capsules were also found suggesting that it had reproduced and become established in that nursery. During 1993 other confirmed reports were received from Yorkshire and East Anglia, indicating that it may be able to survive and multiply across large areas of England.

SummaryEarthworm populations from 200 fields (132 pasture and 68 arable) in Scotland are compared. Earthworms populations from two pasture fields at a Scottish farm infested with the New Zealand terrestrial planarian, Arthurdendyus triangulates (a predator of earthworms), are compared with a subset of these 200 fields (none of which had planarians at the time of the survey). A variety of univariate and multivariate methods are used for comparison. Results from the 200 fields show that the number of individuals, number of species, richness and diversity decline eastwards and northwards across Scotland. There is an overall difference between pasture and arable fields at the same farm, with fewer anecic earthworms (Aporrectodea longa and Lumbricus terrestris), but more Aporrectodea rosea, in arable fields. Conversely, species richness and cumulative species diversity is greater in arable fields, and sample similarity is less. The planarian‐infested fields show differences from the subset of western Scottish pasture fields and from each other. Both have fewer Aporrectodea caliginosa (and endogeic total) and A. longa (and anecic total) than the western pasture fields. One field has fewer L. terrestris and fewer earthworms in total. Univariate factors show no significant differences between the infested fields and similar non‐infested fields, but multivariate analysis suggests differences. Waterlogging and recent growth of rushes in the infested fields are discussed in relation to the reduced population of earthworms and to rainfall.

Earthworms are a major food item of many vertebrates, and their high biomass in temperate ecosystems provides an abundant and high-energy food resource. Earthworm tissue has a protein content of 60–70% (dry weight) with a high amino-acid content that is well matched to the requirements of vertebrates (Lee 1985). In the Western Palearctic, earthworms are a regular part of the diet of 186 mammal-, bird-, reptile-, and amphibian-predator species (Granval and Aliaga 1988). Specifically, earthworms of the family Lumbricidae are the main prey of many nocturnal animals, including 8 species of bird (belonging to families Rallidae, Charadriidae, Scolopacidae, and Strigidae), 5 species of mammal: badgers (Meles meles), hedgehogs (Erinaceus europaeus), moles (Talpa europaea), and 2 shrews (Sorex araneus and Neomys fodiens), and 19 species of amphibian (Granval and Aliaga 1988). The predominance of vertebrate earthworm predators foraging at night is likely to be the result of a higher overall abundance of earthworms near the soil surface at night than during the day. Certain earthworms (especially anecic species, known as night crawlers) perform nocturnal vertical migrations, where they emerge from the soil and move on the surface during the night (Lee 1985). Earthworms are very sensitive to bright light and ultraviolet radiation (Edwards and Lofty 1972). Thus, earthworms come to the soil surface at night or during periods of very low light intensity during the day (late in the evening or early in the morning), to mate or in search of organic food litter (Lee 1985). Many ecological and behavioral studies (time–budget studies, energetic studies, optimal foraging theory) require data on the abundance and availability of prey species. However, it is time consuming and logistically difficult to collect such data on subterranean species such as earthworms. Earthworms are usually sampled according to the standardized method described by Bouche and Gardner (1984) and Bouche and Aliaga (1986), which gives an accurate estimate of earthworm biomass. This method combines 2 complementary extraction techniques: 1) a chemical extraction using diluted formalin, which makes earthworms surface, is sprayed either on bare soil or on short-cut grass and 2) a physical extraction of worms that did not respond to the formalin and/or died near the surface from soil cores. However, such a method can only be correctly used in the daylight because worms forced from the soil by the formalin are very difficult to find and collect at night with portable headlamps. Moreover, earthworm populations are known to be highly aggregated in patches (Poier and Richter 1992, Rossi et al. 1997) and samples spaced only a few meters apart can give large differences in abundance of earthworms. Hence, to account for the spatial heterogeneity of worms, it is generally advised to sample several plots (usually 3–6) at the same site (Lee 1985). Because this technique requires 45 min of formalin spraying per m plus 15 min of preparation time (grass cutting), it is extremely time consuming. Thus, sampling can take many hours depending upon the number of plots sampled. Such a long, involved sampling time is generally not compatible with the work of individual wildlife researchers, who need easy and fast methods to estimate the availability of prey to predators, and not necessarily the exact biomass of prey in the habitat. A habitat containing a high biomass of earthworms (measured by standard procedure) may be of little interest to the predator if prey are too deep to be captured. For example, shorebirds, like Eurasian oystercatchers (Haematopus ostralegus), are limited by the size of their bill and choose prey (bivalves) at a well-defined depth in the mud; therefore, knowing the total biomass of bivalves is useless if only the most superficial ones are potential prey (Zwarts et al. 1996). To our knowledge, the availability of earthworms at night has not yet been quantitatively assessed, and no study has focused on the reliability of the diurnal standard method for estimating nocturnal earthworm availability. We describe an original yet simple method to sample earthworms, designed for wildlife biologists studying nocturnal vertebrate predators. We tested this new sampling procedure as part of an ecological study on the Eurasian woodcock (Scolopax rusticola), a bird that mainly forages on earthworms. The few studies on habitat selection in woodcocks (including the closely related American woodcock, Scolopax minor) found a correlation between habitat selection and earthworm availability, in summer and winter (Hudgins et al. 1985, Hirons and Johnson 1987, Granval and Bouche 1993, Duriez et al. 2005c). On winter nights woodcocks use fields extensively, especially meadows (Cramp and Simmons 1983), which are among the richest habitats for earthworms in Europe (Edwards 1983, Binet and Trehen 1990, Binet 1993, Fraser 1994). Because woodcocks have a maximum 1 E-mail: o.duriez@wanadoo.fr