Plant Pathology (2004) 53, 770–779 Doi: 10.1111/j.1365-3059.2004.01100.x Rust resistance in Salix to Melampsora larici-epitea Blackwell Publishing, Ltd. M. H. Peia*†, C. Ruiz a, C. Bayona and T. Hunter b a Rothamsted Research, Harpenden, Hertfordshire AL5 2JQ, UK, and b4 Wally Court Road, Chew Stoke, Bristol BS40 8XL, UK A total of 174 Salix (willow) clones belonging to 57 species and 14 interspecific hybrids were inoculated with seven pathotypes of Melampsora larici-epitea using the leaf disc method. Infection types were scored based on the uredinial pustule area data and the inoculum density. A close correlation (R2 = 0·82) was found between the average pustule area and the average number of spores produced. Most of the willows were also assessed in the field for rust in 1999. Most willow clones belonging to the species native to western Europe were infected by the rust. In inoculation experiments, uredinia developed on 46 S. viminalis clones, out of a total of 47. In the field, all the S. viminalis clones were infected by rust. Within the subgenus Vetrix, eight out of the 17 willow species that originated from North and South America produced rust pustules in inoculation experiments. Of these, S. pellita was most susceptible. Salix irrorata and S. lasiolepsis var. bracelinae produced well developed pustules after inoculation but no rust infections were detected in the field. In both leaf disc tests and field assessments, no rust infections were found on S. candida, S. cordata, S. drummondiana, S. eriocephala, S. hookeriana, S. houghtonii, S. humilis, S. rigida var. mackenziana and S. syrticola. Of 12 species of subgenus Vetrix native to northeast Asia and Japan, only S. kochiana was susceptible both in inoculation tests and in the field. Salix rossica produced no symptoms in leaf disc tests but showed low levels of infection in the field. The maximum infection type scores in leaf disc tests were highly significantly correlated with field disease severity scores (Spearman rank correlation coefficient was 0·76, P < 1 × 10−10). Keywords: field disease assessment, leaf disc inoculation, Melampsora larici-epitea, pathotypes, willows Introduction Salix (willow) is one of the largest genera of woody plants in the northern hemisphere. According to different authorities, there are between 300 and 500 Salix species worldwide. The species diversity is richest in China (∼270 species) followed by the former Soviet Union (∼120 species) (Argus, 1997). There are some 65 willow species in Europe and over 100 species in North America. Willows also occur in Japan, the Middle East, northern Africa, India, and Central and South America. Salix species are grouped into three subgenera, subgenus Vetrix (shrub willows), subgenus Salix (tree willows) and subgenus Chamaetia (dwarf willows) (Bean, 1980). Of the three subgenera, Vetrix is the largest, containing about twothirds of the Salix spp. Within each subgenus, the species are further grouped into various sections. Over the past 20 years, willows have been grown as a major crop in short-rotation coppice (SRC) plantations for renewable energy in the UK and western Europe. Almost all the *To whom correspondence should be addressed. †E-mail: ming.pei@bbsrc.ac.uk Accepted 19 July 2004 770 willows grown in SRC plantations belong to the subgenus Vetrix, with the species within sections Vimen and Vetrix among the most important. For example, many SRC plantings in the UK are established with S. viminalis, S. burjatica (syn. S. dasyclados) and hybrids between S. viminalis and S. caprea and/or S. cinerea. SRC plantations are harvested at 2- to 4-year intervals and sustain dense, fast-growing canopy during the growing season. Rust caused by Melampsora is the most serious disease in these plantations. Rust defoliates susceptible plantings prematurely and, when severe, reduces yields by as much as 40% (Parker et al., 1993). The most widespread and destructive Melampsora species on SRC willow is M. larici-epitea. During the life cycle, it produces five spore stages and alternates on larch (Larix) (Pei et al., 1993, 1996). Within M. larici-epitea, there is a large variation in pathogenicity to different willows. Six formae speciales have been recognized in continental Europe (Sydow & Sydow, 1915; Gäumann, 1959) and two ‘races’ have been recognized in Japan (Hiratsuka, 1932; Hiratsuka & Kaneko, 1982). In the UK, several rust pathotypes (a pathotype is defined as one or a group of pathogen genotypes showing the same virulence/avirulence patterns on a set of host differentials) occurring on SRC willows were identified under formae speciales, larici-epitea typica © 2004 BSPP Rust resistance in willow 771 Table 1 Melampsora larici-epitea isolates used in the experiments Isolate Pathotype* Host clone Collection site and date DB DM G K Q ST VM LET3 LD1 LR5 LR1 LET4 LET5 LET1 S. × calodendron ‘De Biardii 445’ S. daphnoides ‘Meikle’ S. burjatica ‘Germany’ S. burjatica ‘Korso’ S. × mollissima ‘Q83’ S. × stipularis S. viminalis ‘Mullatin’ Aberdeen, Scotland, September 1992 Long Ashton, SW England, September 1999 Long Ashton, SW England, September 1997 N Devon, SW England, September 1991 Long Ashton, SW England, September 1992 Long Ashton, SW England, September 1995 Long Ashton, SW England, September 1991 *See Pei et al. (1999). (LET), larici-retusae (LR) and larici-daphnoides (LD) (Pei et al., 1996, 1999). In the UK, the occurrence of Melampsora on biomass willows was investigated using site/clone trials (24 clones and four sites) during 1992– 96 (Pei et al., 1999; MHP & CR, unpublished results). The results suggested that most of the clones were infected by LET and LR of M. lariciepitea. Among the LET pathotypes, LET1 was most prevalent, causing severe infections on several S. viminalis clones each year at all sites. The host range of the pathotypes belonging to LR was restricted to S. burjatica/S. dasyclados clones. For low-input crop systems such as SRC willow, one of the most desirable options in disease management is to deploy resistance genes against pathogens. The first step towards selection and breeding for resistance is to identify the sources of resistance. Previous studies involving 77 clones from the National Willow Collection at Long Ashton (Pei et al., 1996) revealed marked variation in pathogenicity in M. larici-epitea and rust resistance in willows. The present study was carried out to characterize rust resistance in a wider range of Salix species/clones to major pathotypes of M. larici-epitea from the UK. Materials and methods Willow plants and rust isolates Willows grown in the National Willow Collection (NWC) at Long Ashton were used for this study. The NWC contains over 1000 clones belonging to more than 180 species/species hybrids and serves as the germplasm source for willow breeding. Willow cuttings (20 cm in length) were made from 1-year-old stems during December–January and kept at −5°C. Willow plants were grown from the cuttings in pots containing John Innes compost no. 3 in an unheated glasshouse. At the time of inoculation, plants from these cuttings were 60 – 80 cm tall and actively growing. A total of 174 willow clones belonging to 57 species and 14 interspecific hybrids were used in this study. The nomenclature of the willows was adopted from Bean (1980), whose treatment was largely based on Willows of the USSR (Skvortsov, 1968). The classification of sections under subgenera, and the names of those which were not included in Bean (1980), were adopted from Skvortsov (1968), Dorn (1976) and Wang & Fang (1984). The © 2004 BSPP Plant Pathology (2004) 53, 770 –779 parentage or possible parentage of the hybrids, or presumed hybrids, with commonly used names are indicated in Table 3. Seven M. larici-epitea isolates, each derived from a single uredinium, were used in inoculation experiments (Table 1). The species identity of the isolates was confirmed by morphological examination of telia and successful inoculation of the alternate host Larix decidua using the method described previously (Pei et al., 1993). When tested on the willow differentials (see Pei et al., 1996), pathogenicity of four isolates corresponded to LET, two to LR and one to LD. These pathotypes were found to be among the most frequent in UK biomass plantations (Pei et al., 1999). Inoculation experiments Due to the large numbers of willow clones tested, it was necessary to carry out two inoculation experiments. The seven rust isolates were inoculated onto 83 willow clones in the first experiment and 91 clones in the second. Two weeks before inoculation, urediniospores were freshly bulked up on detached leaves of the host clone placed in Petri dishes containing water-soaked filter paper. For the experiments, leaf discs 1·1 cm diameter (95 mm2 area) were cut from the fifth through to the 15th leaves from the first unfurled (furled edge less than one-third of total leaf edge) on actively growing willow shoots. Two or three leaves were used from each plant to obtain the leaf discs. Five leaf discs, each from a different leaf, were used as replicates for each willow clone/rust isolate combination. The discs were placed, abaxial surface up, on blotting paper bridges soaked in tap water in 25 compartments of 10 × 10 cm square Petri dishes. Two Petri dishes, 60 mm in diameter, containing 1·2% water agar were also placed in the spray target area. Urediniospores were suspended in tap water containing 0·004% Tween 20 (one drop in 100 mL) and sprayed on to the target area (1 mL per 10 × 10 cm area) using a Humbrol air brush (Humbrol Ltd, UK). After inoculation, the leaf discs were incubated in a growth chamber at 16°C with 16 h day−1 illumination at an intensity of 80 µE m−2 s−1. Inoculum densities (viable spores per leaf disc) were estimated by counting the number of germinating spores on the water agar in 10 fields of view of a light microscope (10 × 10 magnification = 2·4 mm2 each field) for each Petri dish 24 h after inoculation. 772 M. H. Pei et al. Reactions on leaf discs were recorded 13 days after inoculation using a digital camera (Olympus C-2500 L). An image analysis software, SigmaScan Pro 5·0 (SPSS Inc.), was used to assess the disease. For each leaf disc, pustule numbers were counted and pustule diameters were measured manually with the trace measurement mode function. If there were more than 10 pustules on a leaf disc, only 10 randomly selected pustules were measured to obtain an estimate of average pustule diameter. Urediniospores produced on each leaf disc were counted in the second experiment. After photographs were taken, each leaf disc bearing uredinia was gently transferred into a 20 mL screw-top tube using forceps. The tubes containing leaf discs were stored at −4°C until spores were counted. To suspend the spores, 2-mmdiameter glass beads (∼60 beads) and 0·5 mL distilled water containing Tween 20 were added to each tube and the tubes were vigorously agitated for 30 s using a vortex machine. The spores were counted using a haemocytometer and the number of spores on each leaf disc estimated by averaging eight spore counts. Field disease assessments In September 1999, rust disease assessments were conducted in the NWC at Long Ashton. The NWC planting had a configuration of 10 stools at 1 × 0·5 m spacing for each willow clone. The species of rust was identified by examining the position of telia on the leaf surface (M. larici-epitea forms telia mainly on the lower surface of leaves) and the morphology of urediniospores (see Pei et al., 1993). Disease severity was scored as described by Hunter et al. (1996) using four ordinal scores, as follows: none – 0, no rust; low – 1, occasional leaves with pustules; moderate – 2, leaves frequently encountered with pustules; and severe – 3, many leaves with rust pustules, often present in dense aggregation. Rust severity scores were assigned from observing the overall plot for each clone. Data analyses and disease scoring Recent study suggested that the number of spores applied can greatly influence the number of pustules produced in M. larici-epitea (Pei et al., 2002). In this study, disease was scored based on pustule area and inoculum density data according to the method described by Pei & Hunter (in press) using the following steps: (i) for each clone/isolate combination, a slope factor was calculated by dividing average square root pustule area per leaf disc by the square root of the inoculum density per leaf disc for the isolate; (ii) for each isolate, the clone which produced the largest pustule area (the most susceptible reaction) was chosen and the maximum slope factor was calculated by dividing its average square root of the pustule area by the square root of inoculum density (an average maximum slope factor, AMSF) was calculated by averaging the maximum slope factors for all the isolates); (iii) disease was scored for each clone/isolate combination using 0–4 scales. An assumption was made that the average maximum slope factor represents the midpoint for the scale 4, the most susceptible reaction. Disease scores were given as scale 0, slope factor = 0; scale 1, 0 < slope factor ≤ AMSF × 2/7; scale 2, AMSF × 2/7 < slope factor ≤ AMSF × 4/7; scale 3, AMSF × 4/7 < slope factor ≤ AMSF × 6/7; scale 4, slope factor > AMSF × 6/7. Spearman’s rank correlation test was applied to examine for correlation between the maximum infection type scores (regardless of isolates) and the disease severity scores (0–3 scale) assessed in the field. The analyses were carried out using GenStat 5 Release 4·2 (Genstat Committee, 2000). The relationship between pustule area and spore yield was examined by plotting mean uredinial pustule area (square root transformed) against mean number of spores (square root transformed) for each clone/isolate combination which produced uredinia (combinations which did not produce uredinia were excluded). Results In both inoculation experiments, inoculum densities were in the range of 40 –120 spores per leaf disc. Of the 13 willow clones belonging to subgenus Salix (Table 2), only S. magnifica inoculated with isolates G and ST produced occasional, small (0·3 –0·4 mm diameter) uredinia on the peripheries of leaf discs. No symptoms occurred on the remaining clones. No rust infections by M. larici-epitea were observed in the field on any of the 13 clones belonging to subgenus Salix. For willows belonging to subgenus Vetrix, uredinia developed on most willow species native to western Europe, such as S. caprea, S. cinerea ssp. oleifolia, S. daphnoides and S. viminalis (Table 3). Only one S. viminalis clone, ‘Novosibirsk’, did not produce rust pustules in inoculation experiments. A low-level rust infection was found, however, on this clone in the field. Most S. viminalis Table 2 Tree willows (subgenus Salix) inoculated with Melampsora larici-epitea isolates Natural distribution of willow species Willow accession Sect. Amygdalinae S. triandra L. ‘Brunette Noire’, ‘River Severn’ Sect. Humboldtinanae S. longipes Shuttlew. ex Andersson ‘LA172/01’ S. goodingii Ball. ‘WB 50 0576’ Sect. Magnificae S. magnifica Hemsl. ‘LA056/01’ Sect. Pentandrae S. lasiandra Benth. ‘LA080/03’ S. lucida Muhlenb. ‘LA081/01’ S. pentandra L. ‘LA093/02’, ‘LA093/03’, ‘ Dark French’ S. pseudopentandra Flod. ‘LA177/01’ Sect. Salix S alba L. var. coerulea ‘LA006/020’ S. neotricha Goerz. ‘LA173/01’ Europe and Asia North America North America China North America North America Europe and Asia Northeast Asia Europe and Asia Western Europe © 2004 BSPP Plant Pathology (2004) 53, 770 –779 Rust resistance in willow 773 Table 3 Infection type of Salix clones within subgenus Vetrix after inoculation on leaf discs with Melampsora larici-epitea isolates and field rust assessment Willow accession Sec. Arbuscella S. cantabrica Rech. ‘LA157/01’ S. phylicifolia L. ‘LA095/03’ Sec. Balsamiferae S. pyrifolia Anderss. ‘LA100/02’ S. pyrifolia ‘LA100/03’ Sec. Brewerianae S. irrorata Anderss. ‘LA074/01’ S. lasiolepsis Benth. var. bracelinae Ball. ‘LA171/01’ Sec. Caesiae S. kochiana Traut. ‘LA169/01’ Sec. Canae × Sec. Daphnella S. incana Schrank. × S. daphnoides Vill. var. acutifolia ‘NZ CM31226’ S. incana × S. daphnoideσ var. acutifolia ‘NZ CM31228’ Sec. Cordatae S. adenophylla Hook. ‘Farndon’ S. cordata Michx. ‘LA052/01’ S. eriocephala Michx. ‘ER65’, ‘558’, ‘R1077’, ‘R631’, ‘R800’, ‘R892’ S. houghtonii ‘LA167/01’ S. rigida Muhl. var. mackenziana Cronq. ‘Mackenziana’ S. syrticola Fern. ‘LA184/01’ Sec. Cordatae × Sec. Arbuscella S. eriocephala × S. petiolaris Smith ‘601’ Sec. Cordatae × Sec. Hastatae S. myricoides Muhl. × S. hastata Muhl. ‘LA088/01’ Sec. Cordatae × Sec. Longifoliae S. eriocephala × S. exigua Nutt. ‘611’ Sec. Daphnella S. daphnoides Vill. ‘Meikle’ S. daphnoides ‘LA053/08’ S. daphnoides ‘LA053/17’ S. daphnoides var. acutifolia (Willd.) Doell. ‘LA054/04’ S. daphnoides var. acutifolia (× S. caprea?) ‘Latifolia’ Sec. Glabrella S. reinii Fr. et Sav. ‘WB 21/4’ Sec. Glaucae S. glauca L. ‘Cosmopolitan Mason’ Sec. Hastatae S. hastata L. ‘Kashmir WB 50 0258’ Sec. Helix S. amplexicaulis Bory et Chaub. ‘Bory’ S. caesia Vill. ‘Misurina Belluna’, ‘Piccolo’ S. gilgiana Seem. ‘LA166/01’ S. integra Thumb. ‘LA073/01’ Sec. Helix S. linearistipularis (Franch.) Hao S. miyabeana Seem. ‘LA087/02’ S. purpurea L. ‘LA097/48’, ‘Uralensis’, ‘Siberian 077’ Sec. Humboldtinanae S. longipes Shuttlew. ex Andersson ‘LA172/01’ S. goodingii Ball. ‘WB 50 0576’ © 2004 BSPP Plant Pathology (2004) 53, 770 –779 M. larici-epitea isolate and infection typea Natural distribution of willow species DA DB G K Q ST VM Field rust assessment b SW Europe N Europe 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N L N America 0 0 1 0 0 0 1 0 1 0 0 1 0 0 N N N America N America 0 0 2 0 0 0 0 0 2 1 1 1 1 1 N N NE Asia 0 2 0 1 0 0 0 M 0 1 0 2 1 0 1 L 0 1 0 1 1 0 0 L N America N America N America 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 L N N N America N America N America 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N N N N America 0 0 0 0 0 0 0 N N America 0 0 0 1 0 0 0 N 0 0 0 0 0 0 0 N 4 0 4 4 0 0 0 1 0 0 0 0 0 0 0 1 1 2 1 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 S L M M N NE Asia 0 0 0 0 0 0 0 N N Europe, N Asia to N America 0 0 0 0 0 0 0 – N Europe, N Asia to N America 0 0 0 2 0 0 0 M SE Europe E Europe, Central and N Asia NE Asia Japan 0 0 0 0 1 0 0 1 0 0 0 0 0 0 L L 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N N NE Asia E Asia Europe, Asia, N Africa 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N N L S America S America 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N N Europe Europe 774 M. H. Pei et al. Table 3 Continued Willow accession Sec. Lanatae S. hookeriana Barratt ‘LA066/03’, ‘WB 50 0134’ Sec. Longifoliae S. sessilifolia Nutt. ‘LA183/01’ Sec. Nigricantes S. appenina Skv. ‘LA021/01’, ‘LA021/02’, ‘Pescara’ S. mielichoferi Sauter. ‘LA086/01’, ‘LA086/02’ S. myrsinifolia Salisb. (= S. nigricans Sm.) ‘LA091/11’ S. myrsinifolia ssp. alpicola ‘WB 50 0345’ S. myrsinifolia ssp. cotinifolia ‘WB 50 0351’ Sec. Subviminalis S. gracilistyla Miq. ‘Neko-Yanagi’ (LA064/02) (LA064/02) S. gracilistyla var. melanostachys (Mak.) Schneid. ‘Kuro-Yanagi’ (LA135/01) (LA085/01) Sec. Vetrix S. aegyptiaca L. ‘Cambridge’ S. aegyptiaca ‘D. Scott’ S. appendiculata Vill. ‘LA020/01’ S. aurita L. ‘Malham’ S. caprea L. ‘LA037/05’ S. caprea L. ‘Higher Green’ S. caprea ‘Populus’ S. caprea ‘Silberglans’ S. caprea var. sphacelata (Sm.) Wahlenb. ‘LA037/14’ S. cinerea L. ‘LA045/02 S. cinerea ‘LA045/03’ S. cinerea ‘LA045/04’ S. cinerea ‘LA037/06’ S. cinerea ‘Aquatica’ S. cinerea ssp. oleifolia Macr. ‘LA046/05’ S. discolor Muhl. WB 50 0574 S. disperma Roxb. Ex Don. ‘LA068/01’ S. humilis Marsh. ‘LA067/01’ S. laggeri Wimm. ‘LA170/01’ S. petiolaris Sm. ‘LA094/02’ S. scouleriana Barr. ‘LA108/03’ S. scouleriana ‘WB 50 0116’ S. serissaefolia Kim. ‘Kogome-Yanagi’ (LA182/01) (LA182/02) Sec. Vetrix × Sec. Daphnella S. caprea × S. daphnoides (= S. × erdingeri Kern.) ‘LA042/01’ S. caprea × S. daphnoides ‘LA042/02’ S. caprea × S. daphnoides ‘LA042/03’ Sec. Vetrix × Sec. Helix S. futura Seem. × S. integra ‘Westonbirt 1’ S. futura × S. integra ‘Westonbirt 2’ S. × pontederana Willd. (S. cinerea × S. purpurea) ‘LA176/01’ Sec. Vetrix × Sec. Nigricantes S. caprea × S. nigricans Sm. ‘LA160/01’ Sec. Vetrix × Sec. Vimen S. × calodendron (= S. cinerea × S. viminalis) ‘LA041/01’ S. × calodendron ‘445 De Biardii’ S. × calodendron ‘Spaethii’ M. larici-epitea isolate and infection typea Natural distribution of willow species DA DB G K Q ST VM Field rust assessment b N America 0 0 0 0 0 0 0 N N America 0 0 0 0 1 0 0 M Central Europe Central Europe N and Central Europe to N Asia 0 0 0 0 0 1 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 N N N 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N N NE Asia 0 0 0 0 0 0 0 N NE Asia 0 0 0 0 0 0 0 N SE Europe to Central Asia 0 0 0 0 1 1 1 L 0 0 0 0 0 1 1 0 0 1 0 1 3 0 0 0 0 3 0 0 0 1 0 0 0 0 0 0 0 1 2 0 1 0 0 1 1 0 1 1 0 1 0 0 0 1 0 1 1 0 1 0 0 2 1 0 1 0 0 1 0 0 0 L L L L N L L N L 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 2 2 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 1 0 1 1 1 0 0 0 1 2 0 0 0 1 1 1 1 1 0 0 2 1 0 0 0 0 1 1 0 0 0 0 1 1 0 0 0 0 0 0 N N N N L L L N L N L L 0 0 0 0 0 0 0 N 0 3 0 2 2 2 0 N 4 0 0 0 0 0 1 1 0 0 0 1 0 0 M L 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 N – N 0 0 0 4 0 0 0 L 0 0 0 3 3 3 0 0 0 1 1 1 0 0 0 3 3 3 0 0 0 M M M Central Europe Europe to W Asia Europe to NW Asia Europe Europe, NW Asia, N Africa W Europe N America Central Asia N America Central Europe N America N America Japan © 2004 BSPP Plant Pathology (2004) 53, 770 –779 Rust resistance in willow 775 Table 3 Continued Willow accession Sec. Vetrix × Sec. Vimen S. caprea × S. viminalis (S. × dasyclados Wimm.) ‘Grandis’ S. × hirtei (= S. cinerea × S. viminalis × S. aurita) ‘Reifenweide’ Sec. Villosae S. candida Flugge ex Willd. ‘LA036/01’ Sec. Vimen S. burjatica Nassarov (= S. dasyclados Wimm.) ‘Germany’ S. burjatica ‘Korso’ S. burjatica ‘LA033/18’ S. burjatica ‘LA022/02’ S. burjatica ‘Warburg’ S. burjatica × S. viminalis ‘Ashton Stott’ S. burjatica × S. viminalis ‘Stott 11’ S. burjatica ‘Groene Daggelder’ S. drummondiana Barratt ex Hook. ‘LA164/01’ S. ‘Koten’ S. pellita Anderss. ‘LA175/01’ S. rehderiana Schneid. ‘LA102/01’ S. rossica Nassarov. ‘LA181/01’ S. sachalinensis F. Schmidt ‘LA106/03’, ‘Sekka’ S. schwerinii Wolf ‘K3 Hilliers’, ‘LA109/01’ S. schwerinii × S. viminalis ‘Aage’, ‘Henrik’, ‘Stephan’ S. viminalis L. ‘Ballardiana’ S. viminalis ‘Black Satin’ S. viminalis ‘Bowles Hybrid’ S. viminalis ‘Brittanica’ S. viminalis ‘Brittany Greens’ S. viminalis ‘Brown Mirriam’ S. viminalis ‘Carmen’ S. viminalis ‘CEH78-22’ S. viminalis ‘Cinnamomea’ S. viminalis ‘English Rod’ S. viminalis ‘French Osier’ S. viminalis ‘Gallica’ S. viminalis ‘Gravanchede Touraine’ S. viminalis ‘Gravange Nantaise’ S. viminalis ‘Groenlandensis’ S. viminalis ‘Irish Rod’ S. viminalis ‘Ivybridge’ S. viminalis ‘Kaiser-Weide’ S. viminalis ‘LA115/20’ S. viminalis ‘LA115/31’ S. viminalis ‘LA115/42’ S. viminalis ‘LA115/59’ S. viminalis ‘LA115/89’ S. viminalis ‘LA115/94’ S. viminalis ‘Longifolia’ S. viminalis ‘Mullatin’ S. viminalis ‘Nobilis’ S. viminalis ‘Novosibirsk’ S. viminalis ‘Oostenrijks Groen’ S. viminalis ‘Orm’ S. viminalis ‘Pecher Jaune’ S. viminalis ‘Poland’ S. viminalis ‘Pulchra Ruberrima’ © 2004 BSPP Plant Pathology (2004) 53, 770 –779 M. larici-epitea isolate and infection typea DA DB G K Q ST VM Field rust assessment b 0 0 0 0 1 0 0 L 0 1 0 1 1 3 1 M N America 0 0 0 0 0 0 0 N NE Europe to N Asia 0 0 3 1 0 0 1 M 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 1 4 1 0 1 2 0 1 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 0 0 0 0 0 0 2 0 1 1 2 0 0 1 1 1 0 1 1 1 0 1 1 1 0 0 0 0 0 0 0 0 0 1 0 1 1 0 0 0 0 0 0 1 0 0 0 4 0 0 0 0 0 1 3 0 1 4 1 1 1 3 1 4 1 1 1 2 1 1 1 2 2 0 0 0 0 1 0 2 0 1 0 4 0 1 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 1 1 1 2 1 1 0 2 1 2 2 1 2 1 3 1 2 1 3 2 2 2 3 1 0 3 1 0 2 0 2 3 0 S L M L L M – N M M N – N N N S M L M M M L – S L S S M M S L M L – L S S M L L M M L L L L L L Natural distribution of willow species N America Japan N America E Asia NE Asia NE Asia NE Asia Europe to Asia 776 M. H. Pei et al. Table 3 Continued Willow accession S. viminalis ‘Purpurea’ S. viminalis ‘Rapp’ S. viminalis ‘Reader’s Red’ S. viminalis ‘Regalis/Konigs-Hanfweide’ S. viminalis ‘Rijsenburger Kat’ S. viminalis ‘Riparia/Uferhanfweide 1’ S. viminalis ‘Romanin’ S. viminalis ‘Romarin Brun’ S. viminalis ‘Schijndels Rood’ S. viminalis ‘Stone Osier’ S. viminalis ‘Stricta’ S. viminalis ‘Tenuifolia’ S. viminalis ‘Tom Hunter’ S. viminalis ‘Ulv’ S. viminalis ‘Utelescens’ S. viminalis ‘Winsendra’ Sec. Vimen × Sec. Amygdalinae S. × mollissima (= S. triandra × S. viminalis) ‘Q83’ Sec. Vimen × Sec. Vetrix S. × stipularis (= S. viminalis × S. cinerea) ‘LA029/01’ Natural distribution of willow species Europe M. larici-epitea isolate and infection typea DA DB G K Q ST VM Field rust assessment b 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 0 1 2 0 0 0 1 2 1 1 1 1 1 0 0 2 1 2 4 0 0 3 1 1 1 1 2 1 1 1 0 1 0 2 2 0 1 1 M L L L L L L M M L M M L L M M 0 0 0 0 4 0 1 S 0 0 0 0 1 4 0 S a Rating scale 0 – 4. See text for description of scale. N, no rust; L, low rust; M, moderate rust; S, severe rust, –, missing from the field disease assessments in 1999. b clones were infected by isolates ST and VM. Only S. viminalis ‘Cinnamomea’ and S. viminalis ‘Pecher Jaune’ were infected by DB. Within subgenus Vetrix, eight out of the 17 willow species that originated from the Americas produced rust pustules in inoculation experiments. Of these, S. pellita was highly susceptible to four of the rust isolates tested and showed moderate rust infections in the field. Salix irrorata and S. lasiolepsis var. bracelinae produced well developed pustules in inoculation experiments, but no rust infections were detected in the field. In both laboratory inoculation experiments and field disease assessment, no rust infections were observed on S. candida, S. cordata, S. drummondiana, S. eriocephala, S. hookeriana, S. houghtonii, S. humilis, S. rigida var. mackenziana and S. syrticola. Of 12 species of subgenus Vetrix native to northeast Asia and Japan, only S. kochiana was susceptible both in inoculation tests and in the field. Salix rossica, which showed no symptoms in the leaf disc tests, had low levels of infection in the field. The Spearman rank correlation coefficient was 0·76 between the maximum infection type score (irrespective of particular isolates) and field disease severity (P < 1 × 10−10). A close correlation (R2 = 0·82) was found between the average pustule area and the average number of spores produced in the second inoculation experiment (Fig. 1). Discussion In this study, a wide range of willow species and species hybrids was tested for resistance using several pathotypes of M. larici-epitea. Compared with a previous study (Pei et al., 1996), the experimental procedure was improved by: (i) measuring disease using digital image, and (ii) adjusting disease scores according to inoculum density. Previously, infection types were scored as follows: 0, no pustules; 1, feeble pustule development, < 0·15 mm diameter; 2, restricted pustule development, mostly 0·15– 0·3 mm diameter; 3, many pustules, mostly 0·3 –0·5 mm diameter; 4, many pustules, mostly > 0·5 mm diameter. With the previous scoring system, the number of pustules could not be defined, since inoculum densities may vary between inoculations. Also, estimation of pustule size was subject to the assessor’s judgment and therefore required considerable experience to achieve consistency. In the present study, direct measurement of disease using image software proved to be more efficient and accurate. Also the present method provided the means of weighting differences in inoculum pressure and made it much easier to compare disease data between inoculations. In the present study, uredinial pustule area was closely correlated to the number of spores produced (R2 = 0·82). Spore production is one of the most important driving variables in the development of disease epidemics and is also one of the most accurate and least subjective ways of assessing the growth of pathogens and the susceptibility of hosts (Johnson & Taylor, 1976). Recent results with M. larici-epitea (Pei et al., 2002, 2003b) suggested that there were reasonably good correlations between spore yield and uredinial pustule area, with 61·2% of the variance accounted in Pei et al. (2002) and R2 = 0·77 in Pei et al. (2003b). A close correlation (Spearman rank correlation © 2004 BSPP Plant Pathology (2004) 53, 770 –779 Rust resistance in willow 777 Figure 1 The relationships between spore yield and uredinial pustule area of Melampsora lariciepitea per leaf disc of Salix clones 13 days after inoculation. coefficient 0·94) between the number of spores produced and uredinial pustule area was also found in M. laricipopulina (Pei et al., 2003a). The evidence obtained so far suggests that disease scores based on uredinial pustule area can be indicative of spore production in willow and poplar rusts. As M. larici-epitea varies greatly in its pathogenicity, rust resistance in Salix may differ against different geographical populations of the pathogen. Melampsora lariciepitea is distributed in Europe, Asia and North Africa. The rust also occurs in Australasia (Dingley, 1977), apparently spread from outside sources as no Salix species are native to Australia or New Zealand. From the results obtained by Ramstedt (1999) and Spiers & Hopcroft (1996), M. larici-epitea populations in Sweden and in New Zealand appear to be similar to those in England. In this study, several willow species native to Japan and northeast Asia, such as S. gilgiana, S. gracilistyla, S. miyabeana and S. reinii, were immune to rust both in leaf disc inoculation tests and in the field; however, these willows have been recorded as hosts of M. larici-epitea in Japan (Hiratsuka & Kaneko, 1982). Larch-alternating M. epitea also occurs in North and South America under the name of M. paradoxa (syn. M. bigelowii) (Ziller, 1974). Melampsora paradoxa has been recorded on S. eriocephala, S. cordata and S. glauca, which showed no infections in this study. This rust has been recorded on at least 37 Salix species (see Sydow & Sydow, 1915). However, the actual range of the host species for M. paradoxa is not yet certain due to the lack of experimental data. Willows belonging to Sec. Vimen are favoured by biomass willow growers because of their yield potential and coppicing ability. In Europe, S. viminalis, the most important species for biomass, has a long history of cultivation (Bean, 1980). In this study, almost all the 48 S. viminalis clones were more or less infected by the forma specialis larci-epitea typica, indicating that the rust is well adapted © 2004 BSPP Plant Pathology (2004) 53, 770 –779 to S. viminalis. On the other hand, several Far Eastern species within Sec. Vimen, such as S. schwerinii, S. sachalinensis and S. rehderiana, were free of rust both in the inoculation experiments and in the field. The present results also showed that three clones of the hybrid between S. viminalis and S. schwerinii were also free of rust. Two commercial S. schwerinii × S. viminalis clones, ‘Tora’ and ‘Bjon’, have been widely grown in SRC plantations in Sweden and the UK since the early 1990s and to date have sustained high levels of rust resistance. Field rust assessments carried out in Sweden also suggested that interspecific hybrids of S. viminalis were more resistant than the intraspecific hybrids (Johansson & Alström, 2000). These findings suggest that there is great potential to breed novel varieties from Sec. Vimen by incorporating resistance genes from different species. However, caution is required, given that rust resistance shown in one geographical region may not be effective in another geographical region. In future, testing novel clones for resistance against relevant rust populations before largescale release would be important in evaluating the possible risk of rust outbreaks as a result of the introduction of rust from different geographical regions. Most of the clones that did not support rust pustule development in leaf disc inoculations were also free of rust in the field. However, several willows that showed substantial infections (infection types 2–3) in leaf disc inoculation experiments were free of rust in the field. For example, S. irrorata and S. lasiolepsis var. bracelinae produced well developed uredinia in leaf disc tests, but no rust infections were detected in the field. Leaf disc inoculations were carried out using leaves taken from young, actively growing shoots. From the present authors’ experience, young willow leaves are generally more susceptible to rust and, under the experimental conditions, leaf discs are more susceptible to rust infection than plants grown in the field (Pei et al., 1996). With Puccinia species on cereals, 778 M. H. Pei et al. it was noticed that in a protected environment, the pathogenicity of a given rust isolate is likely to be wider than in its natural habitat (Anikster, 1984). The discrepancies between the laboratory and field data in this study may be explained in two ways: either expression of rust resistance in these willows was affected by the different growing conditions for hosts or rust genotypes having the same virulence were not present in the field when willows were assessed for rust infection. Unpublished disease assessment records (TH & MHP) showed that the latter is unlikely because no rust infections had been detected on both S. irrorata and S. lasiolepsis var. bracelinae in the NWC since 1990. Despite the above-mentioned discrepancies, there was good correlation (Spearman’s rank correlation coefficient = 0·76) between the results from leaf disc tests and those from field disease assessments. Similar observations were made in a previous study with M. larici-epitea (Pei et al., 1996). In the North American poplar rust M. occidentalis, studies showed that the results from leaf disc inoculation tests generally conformed to the extent of disease in the field (Hamelin et al., 1994). Such evidence suggests that the leaf disc inoculation test remains an effective means of characterizing host resistance/pathogen virulence in Melampsora on poplar and willow because of its efficiency and simplicity. Breeding for disease resistance is a priority in most agricultural systems. The deployment of resistance genes against pathogens can be particularly important in lowinput crop systems such as SRC willow. Willows are dioecious (occur as male or female), hybridize with relative ease, and those used for biomass production usually reach sexual maturity within 1–2 years. Furthermore, willows are propagated by cuttings and new genotypes can be multiplied in a relatively short period of time. These attributes make breeding for rust resistance a favourable option. The present study showed that there are abundant natural sources of resistance in Salix against M. lariciepitea. Further work is needed to explore the potential of deploying natural resistance in breeding to limit the impacts of rust disease on SRC willows. Acknowledgements This research was funded by the Department for Environment, Food and Rural Affairs (Defra), UK, and the European Union. Rothamsted Research receives grant-aided support from the Biotechnology and Biological Sciences Research Council of the United Kingdom. References Anikster Y, 1984. The formae speciales. In: Bushnell WR, Roelfs AP, eds. The Cereal Rusts, Vol. I. Origins, Specificity, Structure and Physiology. Orlando, FL, USA: Academic Press, 115–30. Argus GW, 1997. Infrageneric Classification of Salix (Salicaceae) in the New World. USA: The American Society of Plant Taxonomists. (Systematic Botany Monographs, 52.) Bean WJ, 1980. Trees and Shrubs Hardy in the British Isles, Vol. IV, 8th edn. London, UK: John Murray, 246 –312. Dingley JM, 1977. Additions to the rust fungi of New Zealand – VI. New Zealand Journal of Botany 15, 29–37. Dorn RD, 1976. A synopsis of American Salix. Canadian Journal of Botany 54, 2769–89. Gäumann E, 1959. Die Rostpilze Mitteleuropas. Beiträge zur Kryptogamenflora Schweiz 12. Bern, Germany: Buchdruckerei Büchler. Genstat Committee, 2000. Genstat 5 for Windows, Release 4.2, 5th edn. Oxford, UK: Numerical Algorithms Group. Hamelin RC, Ferriss RS, Shain L, Thielges BA, 1994. Prediction of poplar leaf rust epidemics from a leaf disc assay. Canadian Journal of Forest Research 24, 2085–8. Hiratsuka N, 1932. Inoculation experiments with some heteroecious species of the Melampsoraceae in Japan. Japanese Journal of Botany 6, 1–33. Hiratsuka N, Kaneko S, 1982. A taxonomic revision of Melampsora on willows in Japan. Reports of the Tottori Mycological Institute 20, 1–32. Hunter T, Royle DJ, Arnold GM, 1996. Variation in the occurrence of rust (Melampsora spp.) and other diseases and pests, in short-rotation coppice plantations of Salix in the British Isles. Annals of Applied Biology 129, 1–12. Johansson LKH, Alström S, 2000. Field resistance to willow leaf rust Melampsora epitea in inter- and intraspecific hybrids of S. viminalis and S. dasyclados. European Journal of Plant Pathology 106, 763–9. Johnson R, Taylor AJ, 1976. Spore yield of pathogens in investigation of the race-specificity of host resistance. Annual Review of Phytopathology 14, 97–119. Parker SR, Royle DJ, Hunter T, 1993. Impact of Melampsora rust on yield of biomass willows. In: Abstracts of the 6th International Congress of Plant Pathology, Montreal, Canada, 28 July –6 August 1993. Ottawa, Canada: National Research Council Canada, 117. Pei MH, Hunter T, 2004. Disease scoring by taking inoculum densities into consideration in leaf disc inoculations with poplar and willow rusts. In: Pei MH, McCracken AR, eds. Rust Diseases of Willow and Poplar. Wallingford, UK: CABI, in press. Pei MH, Hunter T, Ruiz C, 1999. Occurrence of Melampsora rusts in biomass willow plantations for renewable energy in the United Kingdom. Biomass and Bioenergy 17, 153– 63. Pei MH, Royle DJ, Hunter T, 1993. Identity and host alternation of some willow rusts (Melampsora spp.) in England. Mycological Research 97, 845–51. Pei MH, Royle DJ, Hunter T, 1996. Pathogenic specialisation of Melampsora epitea var. epitea on Salix. Plant Pathology 45, 679–90. Pei MH, Ruiz C, Harris J, Hunter T, 2003a. Quantitative inoculations of poplars with Melampsora larici-populina. European Journal of Plant Pathology 109, 269–76. Pei MH, Ruiz C, Hunter T, Arnold GM, Bayon C, 2002. Quantitative relationships between inoculum of Melampsora larici-epitea and corresponding disease on Salix. Plant Pathology 51, 443–5. Pei MH, Ruiz C, Hunter T, Bayon C, 2003b. Rust resistance in Salix induced by inoculations with avirulent and virulent isolates of Melampsora larici-epitea. Forest Pathology 33, 383–94. © 2004 BSPP Plant Pathology (2004) 53, 770 –779 Rust resistance in willow Ramstedt M, 1999. Rust disease on willows – virulence variation and resistance breeding strategies. Forest Ecology and Management 121, 101–11. Skvortsov AK, 1968. Willows of the USSR. Moscow, USSR: Nauka. [In Russian with English summary.] Spiers AG, Hopcroft DH, 1996. Morphological and host range studies of Melampsora rusts attacking Salix species in New Zealand. Mycological Research 100, 1163–75. © 2004 BSPP Plant Pathology (2004) 53, 770 –779 779 Sydow P, Sydow H, 1915. Monographia Uredinearum 3. Leipzig, Germany: Bornträger. [In German.] Wang C, Fang CF, 1984. Salicaceae. Flora of the People’s Republic of China, Vol. 20. Beijing, China: Science Press. [In Chinese.] Ziller WG, 1974. The Tree Rusts of Western Canada. Ottawa, Canada: Canadian Forestry Service, Environmental Canada. (Publication no. 1329.)