Isolation and characterisation of Heterorhabditis spp. (Nematoda: Heterorhabditidae) from Hungary, Estonia and Denmark

Nematology, 1999, Vol. 1(3), 321-332 Isolation and characterisation of Heterorhabditis spp. (Nematoda: Heterorhabditidae) from Hungary, Estonia and Denmark Christine T. GRIFFIN, Ilona DIX, Susan A. JOYCE, Ann M. BURNELL and Martin J. DOWNES Department of Biology, National University of Ireland, Maynooth, Co. Kildare, Ireland Accepted for publication:31 May 1998 Summary – Targeted surveys were conducted for the entomopathogenicnematode Heterorhabditis in areas of Denmark, Estonia and Hungary.Isolateswere identi ed by IEF, PCR and cross-fertilitytestsas belongingto three distincttaxonomicgroups: H. bacteriophora, the north-west European (NWE) type of H. megidis and the Irish type of Heterorhabditis. The Irish and NWE types of Heterorhabditis were both present in Denmark (at six and four sites, respectively),while only the NWE type was recoveredin Estonia. H. bacteriophora was the dominant heterorhabditididenti ed in Hungary (ten sites),but the Irish type was also detectedat two sites. This is the  rst report of the Irish type of Heterorhabditis on continental Europe. Co-occurrence of two Heterorhabditis types at a single site was noted in Denmark (Irish and NWE) and in Hungary (Irish and H. bacteriophora). Heterorhabditis was recovered at 38.5% of sites (n = 26) in Denmark (north coast of Sjælland), 27.3% of the coastal sites (n = 22) in Estonia, and 32.6% of sites (n = 46) in Hungary. Résumé – Isolation et caractérisation d’espèces d’Heterorhabditis (Nematoda: Heterorhabditidae) originaires de Hongrie, d’Estonie et du Danemark –Des prospectionsciblées ont été effectuées dans certaines régions du Danemark, d’Estonie et de Hongrie pour rechercher les nématodes du genre Heterorhabditis. Les souches, identi ées par les méthodes de concentration isoélectrique, de PCR et d’hybridation, appartiennent aux trois groupes taxinomiques d’Heterorhabditis:H. bacteriophora, le groupe de l’Europe du nord-ouest (NWE) de H. megidis et le groupe irlandais d’Heterorhabditis. Le groupe irlandais et le groupe NWE sont tous les deux présents au Danemark (dans six et quatre sites,respectivement),maisseul ledernier groupea été rencontréen Estonie. H. bacteriophora, présent dans dix sites, est l’espèce dominante d’Heterorhabditis en Hongrie, mais le groupe irlandaisa été également détecté dans deux sites. C’est la première fois que le groupe irlandais est rencontré en Europe continentale. La présence simultanée de deux types d’Heterorhabditis est signalée au Danemark (groupe irlandais et groupe NWE) et en Hongrie (groupe irlandais et H. bacteriophora). Des Heterorhabditis ont été collectés sur 38.5% des 26 sites du Danemark (côte nord de Sjælland), 27.3% des 22 sites du littoral estonien et 32.6% des 46 sites de Hongrie. Keywords: biologicalcontrol, entomopathogenicnematode, geographicdistribution,identication, soil survey. Entomopathogenic nematodes of the families Heterorhabditidae and Steinernematidae have considerable potential for the control of insect pests. The non-feeding infective juvenile (IJ) can survive in the soil for several months until susceptible insects are encountered. The IJs enter insects and, together with their symbiotic bacteria, kill them within days. Surveys for entomopathogenic nematodes have been conductedin many partsof the world, includingAustralia (Akhurst & Bedding, 1986), the United States (Akhurst & Brooks, 1984; Hara et al., 1991) and Europe (Mrácek, 1980; Deseö & Miller, 1985; Burman et al., 1986; Blackshaw, 1988; Vänninen et al., 1989; Hominick & Briscoe, 1990; Grif n et al., 1991; 1994b), for the purposes both of recovering potentially useful isolates and of gaining an insight into the ecology of the nematodes. Smits et al. (1991) identied Heterorhabditis isolates from Europe, on the basis of restriction length polymorphisms, as being of three types: Heterorhabditis bacteriophora, the north-west European (NWE) group and the Irish group. Current evidence suggests that the H. bacteriophora group is the dominantHeterorhabditis in southern and central Europe. Thus, Smits et al. (1991) identi-  ed isolates from Italy, Spain, central Germany (Darmstadt), and Moldova as H. bacteriophora, and the species has also been isolated in the south of France (Grenier et al., 1996). Mrácek and Jenser (1988) identi ed a heterorhabditid isolated near Budapest as H. heliothidis, which is now considered to be conspecic with H. bacteriophora (Poinar, 1990). Isolates from the Netherlands, Poland and the north of Germany were assigned by Smits et al. (1991) to the NWE type of Heterorhabditis. This type has also been isolated in the south of England (Hominick et al., 1995) and in Belgium (Miduturi et al., 1996). While most of the records of it have been from northern Europe, the type was also recovered in northc ®Koninklijke Brill NV, Leiden, 1999 321 C.T. Grif n et al. ern Greece (Menti et al., 1997). The Irish type of Heterorhabditis is, to date, the only heterorhabditid to have been isolated in Ireland, where it is relatively common in sandy coastal grasslands(Grif n et al., 1994a).It was also found associated with coastal sites in Britain (Grif n et al., 1994b; Hominick et al., 1995), but has not previously been recorded outside of these islands. The NWE type of Heterorhabditisand H. megidis share the same rDNA restriction pro les for all the restriction enzymes tested by Joyce et al. (1994a). There is strong evidence, however, that the Irish type represents a separate species. Irish type isolates possess a distinctive IEF protein electrophoregram(Joyce et al., 1994b) and a distinctive repetitive DNA restriction pro le (Smits et al., 1991). Also, they can be distinguished from NWE type Heterorhabditis by mtDNA andrDNA ITS restrictionpro-  les for selected enzymes (Joyce et al., 1994a), and they are reproductively isolated from H. megidis and NWE isolates (Dix et al., 1992; Grif n et al., 1994b; Joyce et al., 1994b). Adams et al. (1998) sequenced 716 bases of the rDNA ITS1 region of nine putative species of Heterorhabditis and observed 35 nucleotide substitutions between the Irish type isolate K122 and H. megidis, two of which are autapomorphiesfor K122. There is thus cumulative evidencethat H. megidis and NWE typeare conspeci c but the Irish type is a distinct species. Here we report the results of a sampling programme in Hungary, Denmark and Estonia. There are no previous reports of entomopathogenic nematodes from Denmark or Estonia. In Hungary, limited sampling in the vicinity of Budapest recovered a single H. bacteriophora isolate (Mrácek & Jenser, 1988). The present sampling programme was part of a broader search in Europe aimed at the recovery of novel wild type strains of Heterorhabditis with useful traits. Therefore, sampling was directed at sites at which it was considered likely that the genus was present, viz sandy soils, and especiallycoastal sites, which we (Grif net al.,1994b)andothers(e.g. Hara et al., 1991; Amarasinghe et al., 1994) have found to be a suitable habitat for heterorhabditids. Materials and methods SURVEY AND SAMPLING Potential sampling sites in Denmark, Estonia and Hungary were initially identied from maps and geological data and subsequently by local inspection. Sites in Denmark and Estonia were all coastal (Fig. 1A, B). In Denmark, sampling was restricted to the north coast of Sjælland (Fig. 1A). Sampling in Hungary was conducted in two regions where soils are relatively sandy, near Kecskemét and Debrecen, respectively (Fig. 1C, D). Between the rivers Danube and Tisza, in and around the Kiskungsági National Park (south of Kecskemét; BácsKiskun county), there are extensive sand hills consisting of river deposited sand reworked by the wind. The second area in Hungary was to the north and east of Debrecen (Hajdú-Bihar and Szabolcs-Szatmár counties). Soil sampling and baiting with Galleria mellonella were as described by Grif n et al. (1994b) with the addition that samples found to be positive for Heterorhabditis were baited a second time. The main sampling programme was conducted in October-November, 1991. Targeted sampling to con rm some of the original  ndings was carried out in Sjælland in December, 1992 and in Hungary in June, 1994. MOLECULAR CHARACTERISATION The isolates from the main sampling programme were identi ed from protein isoelectric focusing electrophoregrams (pH range 3-10) of soluble proteins from IJ as described by Joyce et al. (1994b). This diagnosis was subsequently con rmed by restriction enzyme digest of the internal transcribed spacer (ITS) region of rDNA as described by Joyce et al. (1994a), using the MboI restriction enzyme. Isolates from later targeted sampling programmes were characterised using rDNA ITS restriction pro les only. In the case of isolates expressing the HP88 restriction pro le, the rDNA non transcribed spacer (NTS) region was ampli ed as described by Joyce et al. (1994c) and the ca 1.4 kb ampli cation product DNA was digested using the enzymes HaeIII and HpaII. CROSS-BREEDING STUDIES Interstrain crosses were performed on lipid agar plates as described by Dix et al. (1992). At least 60 second generation virgin females were used for each interstrain cross. Reference strains used for the identication of newly isolated nematodesare shown in Table 1. Results Heterorhabditis was isolated in Sjælland (Denmark), Estonia and Hungary. There was no difference between 322 Nematology Heterorhabditisspp. from Hungary, Estonia and Denmark Fig. 1. Sites sampled for Heterorhabditis. A: Estonia; B: Sjælland, Denmark; C: Kecskemét region (Bács-Kiskun county), Hungary and D: Debrecen region (Hajdú-Bihar and Szabolcs-Szatmár counties), Hungary. (Each symbol may represent more than one site; scale bar =10 km). Vol. 1(3), 1999 323 C.T. Grif n et al. Table 1. Origin and source of reference strains of Heterorhabditis. Name Origin ITS restriction pro le Source HB1 Brecon, Australia H. bacteriophora R.J. Akhurst* HP88 Utah, USA H. bacteriophora HP88 P. Westerman** Darmstadt Darmstadt, Germany H. bacteriophora HP88 T. Jackson*** K122 Wexford, Ireland Irish type Own collections HF85 Flevopolder,The Netherlands H. megidis NWE type P. Westerman** *CSIRO, Canberra, Australia. ** Van Hall Institute, Leeuwarden, The Netherlands. *** New Zealand Pastoral Agriculture Research Institute Ltd., Lincoln, New Zealand. Fig. 2. MboI restriction digests of the PCR ampli cation products of the rDNA internal transcribed spacer region of Heterorhabditis isolates, separated on a 2% agarose gel. 1: EU333; 2: EU335; 3: EU347; 4: EU348; 5: EU349; 6: EU94; 7: EU106; 8: K122; 9: EU85; 10: HF85; 11: EU339; 12: EU362; 13: EU367; 14: EU368; 15: EU369; 16: HP88; 17: Darmstadt; 18: HB1; M: 1 kb marker. Strains EU85 (lane 9) and EU106 (lane 7) are from Denmark, all other strains with the EU designation are from Hungary. the three countries in the proportion either of sites or of samples positive for Heterorhabditis (Table 2). Overall, the genus was recovered at 33% of sites, and in 16% of samples. Three species were detected (Table 2): H. megidis NWE type in Denmark and Estonia; Irish type Heterorhabditis in Denmark and Hungary, and H. bacteriophora (HP88 type restriction pro le) in Hungary. The MboI restriction pro les of the rDNA ITS region from a representativesample of these isolatesare shown in Fig. 2. The faint bands visible on this gel represent incomplete digestionproducts. SJÆLLAND (DENMARK)AND ESTONIA All of the sites sampled in Sjælland and Estonia (and hence all of the sites positive for Heterorhabditis) were within 1.5 km of the sea; the majority were within 200 m of it. In Estonia, Heterorhabditis was detected at six of the 22 sites sampled, and all isolates were identied as the NWE type (Table 2; Fig. 1A). Positive sites included meadow, recreational grassland areas, open coniferous forest with extensive grass cover and the margins of a cereal  eld (Table 3). Both the Irish type and NWE type of Heterorhabditis were isolated in Sjælland (Table 2; Fig. 1B). Most of the sites from which Heterorhabditis was recovered were dune or turf recreational areas, often with shrubs and/or conifers present (Table 3). Both Irish and NWE types occurred together at one site, Gudmindrup Lyng. Six samples were taken at this site; the NWE and Irish types were each recovered from one sample. The Irish type 324 Nematology Heterorhabditisspp. from Hungary, Estonia and Denmark Table 2. Occurrence of three types of Heterorhabditis (H. bacteriophora [H.b] Group, the North West European type of H. megidis [H.m] and Irish type Heterorhabditis) at sites in Hungary, Denmark and Estonia, October-November 1991. Country Number of Number (& %) Number of sites with Heterorhabditis sites and of sites and identied to species samples samples with Total H.b H.m Irish Heterorhabditis Estonia 22 6 (27.3%) a6 0 6 0 81 9 (11.1%) a Sjælland 26 10 (38.5%) a9 0 4 6 (Denmark) 85 15 (17.6%) a Hungary Kecskemét 22 10 (45.5%) A8 6 0 2 39 12 (30.8%) A Debrecen 24 5 (20.8%) A4 4 0 0 57 6 (10.5%) B Total 46 15 (32.6%) a12 10 0 2 96 18 (18.8%) a Entries followed by the same letter do not differ signi cantly (P<0.05) from each other. Lower case letters: between countries (chi-squaretest); upper case letters: between regions of Hungary (Fisher’s exact test). Normal letters: sites; bold letters: samples. was recovered from the landward side of a dune ridge, with mainly shrubs (Rosa rugosa,Empetrum nigrum and heathers) and some grass, while the NWE type came from a grass/clover roadside verge bordering that area. An examination of the site descriptions for other Heterorhabditis-positive sites in Sjælland suggested a possible association between the presence of Irish Heterorhabditis and Rosa rugosa. ExcludingGudmindrupLyng, where both nematode types occurred, roses were present at four of the  ve Irish-positive sites, but at none of the three NWE-positive sites. A test of association between nematode type and presence of roses at nematodepositive sites was signi cant at P= 0.07 (Fisher’s exact test). HUNGARY H. bacteriophora was the dominant type present in Hungary, but the Irish type was identi ed at two sites in the Kecskemét region (Table 3; Fig. 1C). Heterorhabditispositive sites in Kecskemét included meadow, pasture, a grassy hillside with juniper, a rye  eld with associated verge, a wooded roadside verge, and the bank of a driedup alkaline lake (Table 3). In the Debrecen area, only H. bacteriophora was detected (Fig. 1D). Sites in this region had the heaviest soils of all those sampled in any of the three countries, being sandy loams or sandy clay loams rather than sands. Positive sites here were pasture, waste grassland and the marginsof a maize  eld (Table 2). Overall, one third of the sites sampled in Hungary were positive for Heterorhabditis (Table 2). Heterorhabditis was detected at a higher proportionof sites in the Kecskemét region (46%) than in the Debrecen region (21%), and the difference was signi cant at P= 0.054 (Fisher’s exact test). The difference between these two regions of Hungary reached signicance at P<0.05 when the proportion of positive samples was compared (31 vs 11%) (Table 2). When only the data for H. bacteriophora (which occurred in both regions) were considered, the prevalence remained higher in Kecskemét (6/22 sites, 6/39 samples) than Debrecen (4/24 sites, 4/59 samples) but the difference was not signi cant for either sites or samples. In a restricted sampling programme in June 1994, several soil samples were taken at each of the two Kecskemét sites at which the Irish type Heterorhabditis isolates (EU167 and EU176) had been recovered in 1991. These were respectively, a wide roadside verge with hedgerow broadeningto deciduouswoodland, and a low-lying pasture with reeds (Phragmites australis) adjoiningthe sampled area. Irish type Heterorhabditis was again isolated in Vol. 1(3), 1999 325 C.T. Grif n et al. Table 3. Locations of sites where Heterorhabditis was detected in October-November 1991, the habitat from which positive samples were taken, and the identity of the isolated Heterorhabditis (H.m.: H. megidis NWE type; H.b.: H. bacteriophora; Irish: Irish type of Heterorhabditis; Het.: identied to genus only). Site location Sample/isolate Identity Habitat at sample code of isolate Estonia Tallinn EU6 H.m. Park: turf with Elymus Keile Joa EU17 H.m. Open conif. forest with grass cover Lohusalu EU21 H.m. Meadow Taagupi EU36 H.m. Grass EU38 H.m. Weedy margin of cereal  eld Aa EU70 H.m. Turf grass EU71 H.m. Grassy slope Tsitre EU80 Meadow EU81 Het. Meadow Denmark: Sjælland Mørdrup EU85 H.m. Grass (waste ground) Hornbæk EU94 Irish Open coniferousforest with ground oak, roses and grass Gilleleje EU103 Het. Grassy slope EU106 ** Irish Shrubs and long grass Rågeleje EU114 Irish Meadow W. of Rågeleje EU116 Irish Grass EU121 Irish Grassy slope Liseleje EU132 Irish Dunes, some roses EU133 Het. Dunes, some roses Nyrup EU140 H.m. Long grass, scattered conifers EU141 Het. Long grass, scattered conifers Lumsas EU143 Het. Grass (parking place) Gudmindrup Lyng EU158 Irish Grass/rose/heather/Vaccinium EU162 H.m. Grass verge Vindekilde EU166 H.m. Old dunes, long grass Hungary: Kecskemét Near Kecskemét EU167 ** Roadside verge with trees *** Irish *KNP IV EU173 *** Het. Short grass, bank of dry alkaline lake near KNP V EU175 Het. Pasture KNP V(1) EU176 *** Irish Pasture KNP V(2) EU178 H.b. Rye  eld including verge KNP V(3) EU185 H.b. Sand hills; grass and juniper KNP V(4) EU188 H.b. Meadow on hillside EU190 Het. Pasture KNP V(5) EU193 H.b. Meadow KNP V(6) EU195 H.b. Meadow EU196 Het. Meadow KNP VI (Bugac) EU201 H.b. Cut pasture 326 Nematology Heterorhabditisspp. from Hungary, Estonia and Denmark Table 3. (Continued). Site location Sample/isolate Identity Habitat at sample code of isolate Hungary: Debrecen Létavértes EU222 H.b. Sheep pasture Nyírbéltek EU228 H.b. Pasture on hill EU229 Het. Pasture Téglás EU246 H.b. Grass with shrubs (waste ground) Téglás-Újfehértó EU249 H.b. Margins of maize  eld Újfehértó EU254 Het. Grass (waste ground) *KNP IV-VI: Kiskunsagi National Park, Regions IV-VI. ** Irish type re-isolatedin Nov. 1992 (Sjælland) or June 1994 (Kecskemét: EU333, EU335, EU347, EU348 and EU349). *** Hb. Re-isolated in June 1994 (EU339 near Kecskemét; EU369 at KNP IV; EU362 & EU367 at KNP V(1)). the verge (site of EU167), where it was detected in 5/20 samples, but it was not isolated in any of seventeen samples from the pasture (site of EU176). This time, H. bacteriophora was present at both sites (verge: 1/20 samples; pasture: 2/17 samples), though it had not been detected at either site in the 1991 sampling programme. The average number of bait insects parasitised (luminescent) in the Heterorhabditis-positive samples from Kecskemét was similar to that for Denmark and Estonia, while signi cantly more insects were parasitised in the Debrecen samples (Fig. 3). To test whether the different parasitisation levels in the Debrecen and Kecskemét samples was due to differences in species composition between the two areas, a comparison was made testing only those samples in which H. bacteriophora was identi-  ed. H. bacteriophora-positive samples from Kecskemét yielded less than half as many parasitised cadavers per sample as the H. bacteriophora-positive Debrecen samples (Kecskemét: x = 2.3±0.76, n = 6; Debrecen: x = 6.0±2.04, n = 4) and the difference was signi - cant at P= 0.086 (Student’s t test). CROSS-BREEDING STUDIES Hungarian and Danish isolates which display an Irish type pro le are interfertile with each other and with the Irish type isolate K122, from Ireland (Table 4). When the females of the Irish type isolate EU114 (Denmark) were crossed with males of the NWE type isolate EU85 (also from Denmark), embryoniclethal progenyresulted (not in table). Crosses between Hungarian isolates with the Irish type pro le and Hungarian isolates with the HP88-type pro le gave rise to embryonic lethal progeny (data not shown). Fig. 3. Mean number of luminescent Galleria mellonella cadavers per Heterorhabditis-positive soil sample from Estonia, Denmark (Sjælland), and Hungary (Kecskemét and Debrecen). Data are for  rst baiting and total ( rst and second) baiting. Bars accompanied by the same letter are not signicantly different (P <0.05, Duncan’s multiple range test); lower case letters:  rst baiting; upper case letters: total baiting. Hungarian isolates having the same ITS region MboI restriction pro le as the H. bacteriophora HP88 strain were not interfertile when crossed with it. The cross of HP88 females and Hungarian HP88 type males (EU362 and EU369) gave sterile F1 progeny while the reciprocal cross resulted in embryonic lethals. Crosses set up with Darmstadt (a European isolate with the HP88 type pro-  le and interfertile with HP88) gave the same results with these isolates as did the cross with HP88 itself (Table 5). None of the restriction enzymes used to digest either the ITS region or the NTS region of the rDNA gene complex revealed a difference between the Hungarian HP88 type Vol. 1(3), 1999 327 C.T. Grif n et al. Table 4. Crosses between Heterorhabditis isolates sharing the “Irish type” rDNA ITS restriction prole and resulting in fertile progeny (+). Female Male K122 EU94 EU106 EU333 EU349 (Ireland) (Denmark) (Denmark) (Hungary) (Hungary) K122 + + + + + EU94 + + + + + EU349 + + + + + Table 5. Crosses between isolates sharing the H. bacteriophoraHP88 type rDNA ITS restriction prole. Female Male HP88 Darmstadt EU362 EU369 (USA) (Germany) (Hungary) (Hungary) HP88 + + F1 F1 Darmstadt + + n.d. F1 EU369 e.l. e.l. + + + : cross resulted in fertile progeny. e.l.: embryonic lethal. F1: sterile F1 adults. n.d.: not determined. isolates and HP88 itself. Fig. 4 shows a HaeIII digestion pro le of the rDNA NTS region of  ve HungarianH. bacteriophora HP88 type isolates. All of these isolates share the same NTS restriction pro le as the HP88 isolate, having four restrictionfragmentsofca 620, 332,197 and 148 bp. The faint bands visible on this gel most probablyrepresent incompletedigestion products. Discussion Isoelectric focusing electrophoregramsof soluble proteins from IJ and the pattern of the restriction enzyme digest of the ITS region of rDNA revealed that both NWE type H. megidis and Irish type Heterorhabditis occur in Denmark. Isolates showing the Irish type PCR pro le were also recovered from soil samples collected in the Kecskemét region of Hungary. The cross-breeding studies showed that these Hungarian and Danish Irish type isolates are interfertile with each other and with the Irish type isolate K122 (from Ireland). This indicates that the Irish typeof Heterorhabditisis not restricted to the coastal region of Ireland and Britain (Grif n et al., 1994b; Hominick et al., 1995), but occurs in central and northern Europe as well. Fig. 4. HaeIII restriction digests of the PCR ampli cation products of the rDNA non-transcibed spacer region of Hungarian isolates of the HP88 (H. bacteriophora) type separated on a 2% agarose gel. 1: HP88; 2: EU339; 3: EU362; 4: EU367; 5: EU368; 6: EU369; M: 1 kb DNA ladder (Gibco BRL). 328 Nematology Heterorhabditisspp. from Hungary, Estonia and Denmark NWE type H. megidis predominates on the northern continentalfringe,beingthe only Heterorhabditis isolated to date from the Netherlands (Hominick et al., 1995), Belgium (Miduturi et al., 1996), northern Germany and Poland (Smits et al., 1991) and Estonia (present study). The NWE type has also been identied from the south east of England(Hominicket al., 1995) where the climate of Britain most resembles that of the continental fringe. In contrast, the Irish type Heterorhabditis is the only one recovered in Ireland and in all but the south east of Great Britain. This might suggest that it is a species of mild, wet climates. However, its discovery at central European sites in Hungary and on a portion of the continental fringe in Denmark, where it occurs with the NWE type, suggests that the picture is more complicated: further sampling effort will clearly be necessary to delineateits distribution range. The apparent absence of the Irish type in Estonia, with its relatively long cold winters, would support the thesis that the Irish type is a species of equitable climates. Moreover, bearing in mind that soils sampled were of a similar, sandy texture, it is striking that Irish type Heterorhabditis occurred at a lower prevalence toward the climatically more extreme centre of the continent (Hungary, 4% of sites) than in Denmark (23% of sites) or in Ireland and Britain, where it was detected at an average of 12% of sites (Grif n et al., 1994b). This suggests that it is less successful in the conditionsprevailing in Hungary, which would accord with the poorer heat toleranceof Irish type isolates than of Hungarian H. bacteriophorareported by Finnegan et al. (1999). The site at which the Irish type was detectedin bothof the Hungariansamplings(November, 1991 and June, 1994) was a roadside verge shaded by trees — virtually a small woodland; few other wooded sites were included in this study. In the Negev region of Israel, Heterorhabditis densities were higher in plots with a high degree of shade (Glazer et al., 1996). Perhaps in Hungary, Irish type Heterorhabditis is favoured by woodland, where lower temperatures and higher soil moisture would be expected than in the neighbouring grassland. While the other Hungarian site at which the Irish type was recovered was open grassland, it was bordered by reeds, indicating a high water table. The Irish type was not recovered from this site during the repeat sampling in June 1994; however, at that time of the year the nematodesmay have been below the 10 cm soil depth to which it was sampled. In warm regions, Heterorhabditis tends to be found in the deeper soil layers during the summer, especially in soil which is more exposed to solar radiation (Glazer et al., 1996). The Irish type isolates from Denmark and Hungary have not yet been fully characterised for biological traits; given the diversity amongst isolates of the type from Britain and Ireland (Hass, 1996) it would not be surprising if locally adapted ecotypeswere detectedamongst them. A number of Heterorhabditis isolates collected in Hungary show rDNA ITS and rDNA NTS DNA restriction pro les identicalto that of the American isolateof H. bacteriophora, HP88, using key diagnostic restriction enzymes. However, the Hungarian isolates are not interfertile with HP88. HP88-compatible nematodes do occur in Europe, however,as evidencedby the cross-breedingsuccess with the Darmstadt strain. The incompatibility between HP88 and the HungarianHP88-like isolatesis at the post-mating level, and results in either hybrid inviability or hybrid sterility dependingon the directionof the cross. Embryos were formed in all of the crosses that were carried out. These embryos grew into apparently normal but sterile adultsif the maternal partner for the cross was from either the HP88 or Darmstadt strain. Whether genetic or cytoplasmicmaternal factors play an important role in the reproductiveisolation between these isolates has not been determined. The results of this survey provide further evidence of a geographicseparation in Europe between the H. megidislike nematodes (the Irish and NWE types of Heterorhabditis), found predominantly in northern Europe, and the H. bacteriophora group found in southern and central Europe. However, the occurrence of the Irish group in Hungary (present study) and of H. megidis in Greece (Menti et al., 1997) clearly indicates that a division between a northern ‘megidis‘ group and a southern H. bacteriophora group is too simplistic. The recovery rate of Heterorhabditis was much higher in this study, in which sites with sandy soils were specifically targeted for sampling, than in non-targeted surveys conducted in neighbouring areas of Europe. Thus, Heterorhabditis was recovered at one third of sites in Hungary, but was not recovered in the survey of Czechoslovakia (Mrácek, 1980). Similarly, the recovery rate of Heterorhabditis at 38.5% of sites in Sjælland is considerably higher than in West-Flanders (Belgium) (5% of sites) (Miduturiet al., 1996)or in the Netherlands(13% of sites) (Hominick et al., 1995). The ef cacy of targeted sampling in recovering strains of Heterorhabditis was previously demonstrated in Ireland and Britain: Grif n et al. (1994b) recovered Heterorhabditis from 4-45% of sandy sites, while in general surveys, it either was not detected (Blackshaw, 1988; Boag et al., 1992; Gwynn & RichardVol. 1(3), 1999 329