J. Phytopathology 155, 497–504 (2007)  2007 The Authors Journal compilation  2007 Blackwell Verlag, Berlin doi: 10.1111/j.1439-0434.2007.01269.x ETH Zurich, Institute of Integrative Biology, Zürich, Switzerland Effects of the Harvest Method on the Infestation of Chestnuts (Castanea sativa) by Insects and Moulds T. N. Sieber1, M. Jermini2 and M. Conedera3 AuthorsÕ addresses: 1ETH Zurich, Institute of Integrative Biology, Forest Pathology and Dendrology, 8092 Zürich, Switzerland; 2 Agroscope RAC Changins, Swiss Federal Research Station for Plant Production of Changins, Centre of Cadenazzo, 6594 Contone, Switzerland; 3WSL Swiss Federal Institute for Forest, Snow and Landscape Research, Sottostazione Sud delle Alpi, 6504 Bellinzona, Switzerland (correspondence to T. N. Sieber. E-mail: thomas.sieber@env.ethz.ch) Received November 20, 2006; accepted March 4, 2007 Keywords: fruit spoilage, orchard, phytosanitary measures, biodiversity, fungi, Ciboria batschiana, chestnut weevil (Curculio elephas), chestnut moth (Cydia splendana), vertical transmission of tree endophytes Abstract Chestnuts were collected either every 7 days from suspended nets used to intercept the fruits, every 2 days from the ground or every 7 days from the ground. Nuts were visually inspected after collection for the presence of exit holes of the chestnut weevil (Curculio elephas) and the chestnut moth (Cydia splendana), and 20 nuts per sampling and tree were bisected to assess fungal colonization. Apparently healthy nuts were incubated at 24C and 70–80% relative humidity for 21 days. All nuts were bisected after incubation and examined for the presence of insects and moulds. The harvest method did not have a statistically significant effect on either moulding or insect infestation except on Amphiporthe castanea. This vertically transmitted fungal endophyte was less frequently isolated from fruits collected after 7 days from the ground. The black rot fungus Ciboria batschiana did not occur in chestnuts intercepted in nets, but the difference to chestnuts collected from the ground was statistically not significant. The frequency of nuts colonized by C. batschiana was low in general probably due to the hot and dry summer in 2003. Big, marketable fruits appeared to be less frequently colonized by insects and moulds right after collection. This difference disappeared after incubation except for the chestnut moth. The mother tree had the greatest effect on fungal and insect infestation, indicating the importance of the genetic disposition and/or the phytosanitary situation of each tree. The chestnut weevil preferred chestnuts of the variety ÔLüinaÕ to those of the variety ÔTorcionÕ, whereas the fungi A. castanea, Trichothecium roseum, Clonostachys rosea and Penicillium spp. preferentially colonized chestnuts of the variety ÔTorcionÕ. Introduction Chestnuts are typical seasonal fruits that maintain their optimal commercial quality, turgescence and health for only a comparatively brief period. One of the major difficulties is the high perishability of the product. The major factors in postharvest depreciation are moulding or rotting caused by fungi and the larval development of insects (Wells and Payne, 1980; Breisch, 1993; Bassi et al., 2001). Insect damage is usually due to infestations of the chestnut moth Cydia splendana and the chestnut weevil Curculio elephas, which attack the fruits while still on the trees. Damage increases concomitantly with the development of the larvae (Giacalone and Bounous, 1993). Fungal infections often start in the larval galleries of insects (Wells and Payne, 1980), and many nuts become infected on the ground before picking. Some moulds are considered endophytes that colonize the fruits at various stages during their development but do not cause any symptoms of disease until after fruit fall (Washington et al., 1997, 1998). Expansion of fungal mycelia in the fruits and degradation of the cotyledons occur mainly during storage (Rutter et al., 1990; Giacalone and Bounous, 1993). At early infection stages, it is not easy to differentiate slightly mouldy or parasitized nuts from the good ones until they are processed or consumed (Rutter et al., 1990). As a consequence, various nut-treatment techniques have been developed. The most common are hydrotherapy (cold bath) and thermo-hydrotherapy (warm bath), which was developed in Italy during the 1930s. Thermo-hydrotherapy is nowadays the standard method used for its proven ability to control insect infestations without influencing nut quality (Jermini et al., 2006). Similarly, hydrotherapy prior to cold storage reduced moulding significantly from 60% to 30% (Delatour, 1978; Jermini et al., 2006). However, a disease incidence of one-third of mouldy nuts is still high, and further control strategies are needed for commercial chestnut production. In areas with traditional subsistence chestnut culture www.blackwell-synergy.com Sieber et al. 498 such as in southern Switzerland, where chestnut cultivation has been abandoned for more than half a century, the incidence of parasite attacks is usually high (Jermini et al., 2001). The chances of success in the recovery and renewal of chestnut groves depends on the application of treatment and conservation techniques that are both effective and ecologically, economically and technologically sustainable (Conedera et al., 1997). The local product must be improved by guaranteeing a quality matching that of imported nuts. Various studies have analyzed the effects of treatments and diverse conservation techniques on the quality of the fruit (Wells and Payne, 1980; Fadanelli et al., 1995; Tian and Bertolini, 1997; Washington et al., 1997; Bassi et al., 2001; Jermini et al., 2006), but none has thoroughly examined the impact of the harvest methods on fungal postharvest deterioration. The traditional method of harvest is to allow the nuts to fall to the ground, and employ people to pick them up. This is not a bad system for a few nuts for your own table, but commercial orchards need more efficient methods (Anagnostakis, 2003). Interception of chestnuts in nets either on the ground or suspended above the ground is made in Italy and France mainly to ease and accelerate chestnut collection (Breisch, 1995; Bounous, 2002). In addition, suspended nets are considered to reduce moulding considerably, especially infestations by Ciboria batschiana (Zopf) Buchwald (syn. Sclerotinia batschiana Zopf, S. pseudotuberosa Rehm; anamorphic form Rhacodiella castaneae (Bainier) Peyr., syn. Myrioconium castaneae (Bainier) M. Morelet), the most serious pathogen of chestnut fruits, but proper experimental proof of effectiveness of this method is still pending (Breisch, 1995). Therefore, we performed an experiment in southern Switzerland to compare the suitability of three harvest methods for the reduction of mouldy nuts. Materials and Methods The study was performed in a chestnut (Castanea sativa) orchard at Toricella, Ticino, Switzerland (grid reference 4604¢08¢¢N; 854¢00¢¢E) in autumn 2003. Eight trees were chosen within an area of approximately 1 ha. The crown diameter varied between 10 and 16 m among the trees. Trees 1, 2, 3, 6 and 7 belonged to the variety ÔLüinaÕ, while trees 4, 5 and 8 to the variety ÔTorcionÕ. Three harvesting methods were tested. Method 1: Fruits were collected every 7 days from the nets fixed below the trees; the nets were fixed as high as possible to prevent soil contact when the nets were full of chestnuts (five sampling dates). Method 2: Fruits were collected every 2 days from the ground (11 sampling dates). Method 3: Fruits were collected every 7 days from the ground (five sampling dates). All fruits were collected at each sampling date. Each of the three methods was installed under each tree in a triangular sector of similar size. The fruits were separated in marketable (<100 fruits/kg) and small fruits, counted and checked visually for the presence of exit holes of both the chestnut weevil (Curculio elephas) and the chestnut moth (Cydia splendana) and for fungal postharvest damage. In addition, 20 chestnuts were randomly selected per sampling date, method and tree, bisected and inspected for the presence of moulds and insect larvae. All apparently healthy fruits were incubated at 24C and 70–80% relative humidity for 21 days. All fruits were bisected after incubation and examined for the presence of fungi and insects. Only those fungi were identified which colonized at least one-third of the cut surfaces of the cotyledons. Fungi were identified by microscopic examination at 400· or 1000·. C. spendana and C. elephas were identified on the basis of both the morphology of the larvae present and the type of damage (Della Beffa, 1961; Bonnemaison, 1962; Bovey et al., 1975; Pollini, 1998). The overall number of mouldy chestnuts and the number of chestnuts colonized by a given fungus or insect were expressed as percentages or proportions of the number of collected or incubated chestnuts. Proportions were arcsine-square-root transformed and subjected to the analysis of variance (anova). The model tested was as follows: Yijkl ¼ l þ ai þ bj þ ck þ ðabÞij þ ðacÞik þ ðbcÞjk þ ðabcÞijk þ eijkl Yijkl ¼ dependent variable, i.e. arcsine-square-root transformed proportion of mouldy chestnuts, chestnuts attacked by either Curculio elephas or Cydia splendana or chestnuts colonized by a certain fungal species in the lth sampling unit; l ¼ overall mean; ai ¼ main effect of the ith mother plant individual (i ¼ 1, 2, . . . , 8) or the ith chestnut variety [i ¼ 1 (Lüina) or 2 (Torcion)]; bj ¼ main effect of the jth method [j ¼ 1 (up to 7 days in the net), 2 (up to 2 days on the ground), 3 (up to 7 days on the ground)]; ck ¼ main effect of the kth fruit size [k ¼ 1 (marketable fruits), 2 (small fruits)]; (ab)ij, (ac)ik, (bc)jk and (abc)ijk ¼ two- and threefold interactions between main effects; eijkl ¼ random error. Significant main effects with ‡3 character states were subjected to Tukey’s pairwise comparisons following anova. Regression analysis was employed to compare the colonization of nuts before and after incubation. LOWESS smoothing was used to draw robust Ôregression linesÕ if a relationship proved to be statistically significant (Cleveland, 1981). The F (0 £ F £ 1) parameter was set to 0.5. Results Experimental design All trees produced similar amounts of chestnuts per m2. The proportion of small chestnuts was much larger than that of the marketable ones for all trees. anova did not reveal any statistically significant interactions, and residual analyses showed that the data complied with the requirements of anova. Before incubation The size of the fruits had a statistically significant effect on colonization by moulds and the two insects Harvest Method and Infestation of Chestnuts 499 Curculio elephas and Cydia splendana (Table 1). Small (not-marketable) fruits got more frequently damaged than the marketable ones. The method of harvest did not have a measurable influence on colonization by fungi and C. splendana, but had a statistically significant effect on the frequency of chestnuts attacked by C. elephas (Table 1). Chestnuts with exit holes of this insect were significantly more frequent when the chestnuts were collected from the ground after 7 days than when they were collected after 2 days only. The figures presented in Table 2 suggest that the frequency of the chestnuts parasitized by this insect was higher when the nuts were intercepted in nets than when they were collected from the ground. However, the among-sampling-date variability of the ÔnetÕ data for C. elephas was four times greater than that of the ÔgroundÕ data. Thus, the difference between the net and the ground could not be statistically confirmed. The tree individual from which the fruits originated had a significant effect on colonization by C. splendana. Trees 5 and 6 were more heavily attacked than trees 2 and 4 (Table 2). In addition, C. splendana larvae were more frequently observed in chestnuts of tree 6 than in those of trees 3 and 7. Similarly, chestnuts of tree 1 were more frequently parasitized by the larvae of C. splendana than those of tree 4. The differences in the frequency of fruits infested by insects were, however, not attributable to the variety. Depending on the tree, up to nine chestnuts per one thousand fruits were mouldy right after collection (Table 2). Species of Penicillium and Mucor, Acrospeira mirabilis, Amphiporthe castanea, Trichothecium roseum, Wardomyces columbinus and Clonostachys rosea were most frequently observed in fruits examined immediately after collection. The low frequency of colonization exhibited by these fungi did, however, not allow detecting differential effects of the fruit size, the tree individual, the chestnut variety or the harvest method. After incubation After incubation, fruit size had a distinct effect on colonization by the chestnut moth C. splendana only (Table 3). The method of harvest did not have a significant influence on either fungal or insect attack. In contrast, the tree individual, from which the fruits Table 1 anova table for colonization of chestnuts before incubation. Bold-typed P-values indicate significant effects Dependent variable Curculio elephas Moulds Cydia splendana Sum-ofMeanSum-ofMeanSum-ofMeansquares Df square F-ratio P-value squares Df square F-ratio P-value squares Df square F-ratio P-value Effects Tree (a) Method (b) Fruit size (c) Tree · method (ab) Tree · fruit size (ac) Method · fruit size (bc) Tree · method · fruit size (abc) Error 0.153 0.151 0.319 0.272 0.082 0.025 0.134 4.513 7 2 1 14 7 2 14 0.022 1.289 0.256 0.075 4.44 0.013 0.319 18.832 <0.001 0.019 1.144 0.319 0.012 0.69 0.681 0.012 0.728 0.484 0.01 0.564 0.892 266 0.017 2.414 0.107 5.407 0.695 0.172 0.093 0.423 7 2 1 14 7 2 14 0.345 8.686 <0.001 0.054 1.348 0.262 5.407 136.178 <0.001 0.05 1.25 0.239 0.025 0.62 0.739 0.047 1.175 0.31 0.03 0.762 0.71 10.561 266 0.04 0.032 0.001 0.036 0.038 0.009 0.005 0.031 0.934 7 2 1 14 7 2 14 0.005 1.293 0.254 0.001 0.158 0.854 0.036 10.384 <0.001 0.003 0.778 0.692 0.001 0.356 0.927 0.002 0.655 0.52 0.002 0.622 0.846 266 0.004 Table 2 Number of chestnut fruits collected and frequency of fruits colonized by larvae of Curculio elephas and Cydia splendana and by fungi before incubation. Data presented for each mother plant and each harvest method Plant or method of collection Plant no. 1 2 3 4 5 6 7 8 Harvest method 7 days net 2 days ground 7 days ground a Variety Lüina Lüina Lüina Torcion Torcion Lüina Lüina Torcion Frequency (%) of fruits with larvae of Curculio elephas Frequency (%) of fruits with larvae of Cydia splendanaa Frequency (%) of fruits with moulds 5514 6022 6412 7330 3222 5298 8216 9014 2.7 5.6 4.7 2.7 2.5 3.7 2.8 3.3 18.8 14.4 17.4 11.1 24.3 25.2 13.6 20.3 0.181 0.133 0.172 0.900 0.341 0.529 0.438 0.255 15 224 18 709 17 095 4.4 2.8 3.3 19.4 17.5 15.7 0.361 0.283 0.497 No. fruits collected Fruits of trees 5 and 6 were significantly more frequently colonized than those of trees 2 and 4 (P < 0.05); in addition, fruits of tree 6 were significantly more frequently colonized than those of trees 3 and 7 (P < 0.05), and fruits of tree 1 were significantly more frequently colonized than those of trees 4 (P < 0.05). Sieber et al. 500 Table 3 anova table for colonization of chestnuts after incubation. Bold-typed P-values indicate significant effects Dependent variable Curculio elephas Effects Tree (a) Method (b) Fruit size (c) Tree · method (ab) Tree · fruit size (ac) Method · fruit size (bc) Tree · Method · fruit size (abc) Error Moulds Cydia splendana Sum-ofMeanSum-ofMeanSum-ofMeansquares Df square F-ratio P-value squares Df square F-ratio P-value squares Df square F-ratio P-value 1.116 0.116 0.066 0.406 0.134 0.034 7 2 1 14 7 2 0.159 0.058 0.066 0.029 0.019 0.017 5.415 1.966 2.244 0.986 0.649 0.574 <0.001 0.142 0.135 0.468 0.715 0.564 2.066 0.271 0.792 0.453 0.149 0.026 7 2 1 14 7 2 0.121 14 0.009 0.293 0.994 0.163 14 0.012 7.714 262 0.029 16.408 262 0.063 originated, had a significant effect on colonization by moulds and both insects (Tables 3 and 4). Regarding C. splendana and moulds, the tree effect was, however, not attributable to the variety. In contrast, the variety had an effect on the attack by the chestnut weevil C. elephas, the fruits of the variety ÔLüinaÕ being on average more heavily infested. Fruits from trees 2 and 3 showed the highest incidence of C. elephas. Attack by C. splendana was most severe on trees 5 and 6, and that by moulds on trees 1, 4, 7 and 8 (Table 4). Between 27% and 57% of the chestnuts were colonized by fungi (Table 4). Although not significant, the chestnuts that stayed for only 2 days on the ground showed the least rate of fungal colonization (Table 4). The harvest method did not have a statistically significant effect on the frequency of colonization by any of the fungal taxa except Amphiporthe castanea. anova of the A. castanea data resulted in a significant interaction term between the effects of the harvest method and the mother tree (i.e. the trends for the method effect depended strongly on the mother tree and viceversa). Method and tree effect on the frequency of colonization by this fungus had to be analyzed, therefore, for each tree and method separately. The harvest method had a significant effect on the chestnuts of trees 6 (P ¼ 0.037) and 8 (P ¼ 0.025). Chestnuts from tree 6 were significantly more frequently colonized by A. castanea when they were intercepted in nets than when they were collected from the ground every 7 days, whereas chestnuts from tree 8 were more frequently colonized by A. castanea when they were collected after 2 days than after 7 days, independently of whether they were retained in nets or laid on the ground. The mother tree had a significant effect on colonization by A. castanea for chestnuts that stayed on the ground for 2 days but not for those that were collected every 7 days from the nets or from the ground. Chestnuts of the variety Torcion proved to be more frequently colonized by A. castanea than those of the variety Lüina (P ¼ 0.013). Ciboria batschiana did not occur in chestnuts intercepted in nets, but the difference to chestnuts collected from the ground was statistically not significant. 0.295 0.135 0.792 0.032 0.021 0.013 4.714 <0.001 2.161 0.117 12.643 <0.001 0.517 0.923 0.34 0.935 0.209 0.811 0.186 1 2.013 0.094 0.106 0.521 0.396 0.048 7 2 1 14 7 2 0.288 0.047 0.106 0.037 0.057 0.024 4.195 0.684 1.541 0.543 0.825 0.348 <0.001 0.506 0.216 0.906 0.567 0.706 0.327 14 0.023 0.341 0.988 17.959 262 0.069 The effect of the mother tree was statistically significant also for Acrospeira mirabilis (P ¼ 0.001) and almost significant for Wardomyces columbinus (P ¼ 0.084) and Trichothecium roseum (P ¼ 0.056) (Table 4). A. mirabilis was observed significantly more frequently in chestnuts from trees 1, 7 and 8 than in those from tree 5. In addition, fruits infested by A. mirabilis were less abundant on tree 3 than on tree 7. Chestnuts of the variety Torcion proved to be more frequently colonized by Penicillium spp. (P ¼ 0.021), Trichothecium roseum (P ¼ 0.019) and Clonostachys rosea (P ¼ 0.012) than those of the variety Lüina. The number of fruits with exit holes of larvae of C. elephas prior to incubation was not useful to predict the number of fruits colonized by this insect. Similarly, the bisection of fruits prior to incubation to assess infestation by moulds was not useful to predict infestation by moulds after incubation. For C. splendana existed, however, a weak but significant relationship between the number of fruits with exit holes prior to incubation and the frequency of colonized fruits after incubation (Fig. 1). Discussion The harvest method had no effect on moulding and infestation by the chestnut moth and the chestnut weevil. This result was expected for the two insects because they deposit their eggs directly in or close to the fruits while these are still on the tree. This result was, however, not expected in respect to the infestation by fungi, since interception of chestnuts in nets is considered to provide an effective control of moulds, especially of the black rot fungus Ciboria batschiana (Breisch, 1995), which is considered to cause the most severe losses of all postharvest diseases of the chestnut (Ridé and Gudin, 1960). Although C. batschiana did not occur in chestnuts intercepted in nets (Table 4), the differences among harvest methods were statistically not significant. The frequency of nuts colonized by C. batschiana was low in general, probably due to the hot and dry summer in 2003. Optimum temperature for the growth of C. batschiana is 18.5C and no growth occurs above 31C. In addition, wet weather is Plant or method of collection Frequency (%) of fruits with larvae of No. fruits Curculio Variety incubated elephasa Plant no. 1 Lüina 2 Lüina 3 Lüina 4 Torcion 5 Torcion 6 Lüina 7 Lüina 8 Torcion Harvest method 7 days net 2 days ground 7 days ground Frequency (%) of fruits colonized by the ten most frequently observed fungal taxad Frequency (%) of Frequency fruits with (%) of larvae of fruits Penicillium Mucor Acrospeira Amphiporthe Trichothecium Aspergillus Fusarium Wardomyces Clonostachys Ciboria Cydia with spp.e spp.f mirabilisg castanea roseum spp.h oxysporum columbinus rosea batschiana splendanab mouldsc 4314 4810 4986 6254 2345 3738 6830 6862 5.2 8.4 10.8 5.3 2.3 3.2 2.5 2.4 13.4 17.1 13.4 8.8 23.1 27.2 13.1 10.1 56.3 36.2 27.1 56.1 43.9 38.3 55.1 56.6 39.5 26.7 20.0 42.6 26.0 28.2 36.1 33.1 11.1 3.8 7.8 11.0 12.9 13.8 11.6 10.5 7.0 3.0 0.3 3.7 0 2.0 13.5 8.5 1.7 1.2 1.2 4.0 2.0 2.9 3.1 2.9 0.7 1.2 3.0 1.5 0 0.5 0 3.9 3.4 1.8 0 0.6 0.9 0 0.3 0 0.3 2.0 0.6 1.2 0.6 1.5 0.3 0 0.7 0.6 1.8 0.3 0.3 0.2 0 0 1.0 1.0 0.4 1.2 0.6 0 0.3 1.3 0 0.6 0.3 0 0 0.2 0 3.3 11 591 14 767 13 781 3.7 5.5 5.6 12.6 18.4 11.5 49.6 42.3 51.8 34.1 26.9 39.9 12.3 8.0 13.7 5.5 3.7 4.2 2.1 1.9 3.5 2.5 1.5 0.7 0.3 0.7 1.7 0.9 1.1 0.6 0.3 0.6 1.3 0.4 0.8 1.0 0 0.2 1.8 Harvest Method and Infestation of Chestnuts Table 4 Number of chestnut fruits incubated and frequency of fruits colonized by larvae of Curculio elephas and Cydia splendana and by the most frequently observed fungal taxa after incubation. Data presented for each mother plant and each harvest method a Fruits of trees 2 and 3 were significantly more frequently colonized than those of trees 5, 6, 7 and 8 (P < 0.05). Fruits of trees 5 and 6 were significantly more frequently colonized than those of tree 4 (P < 0.05), and fruits of tree 5 were significantly more frequently colonized than those of trees 1, 3, 7, and 8 (P < 0.05). c Fruits of tree 3 were significantly less frequently colonized than those of trees 1, 4, 7 and 8 (P < 0.05). d rarely observed taxa were Alternaria tenuissima, Aureobasidium pullulans, Cephalotrichum sp., Chloridium sp., Cladorrhinum sp., Geomyces pannorum, Phomopsis spp., Talaromyces wortmannii. e mostly Penicillium expansum. f Mucor heterosporus, M. globosus. g Fruits of trees 1, 7 and 8 were significantly more frequently colonized than those of tree 5 (P < 0.05); Fruits of tree 7 were significantly more frequently colonized than those of tree 3 (P < 0.05). h Aspergillus ochraceus, A. niger, A. parasiticus. b 501 Sieber et al. 502 After incubation 2.0 1.3 0.6 0.0 0.0 0.6 1.3 2.0 Before incubation Fig. 1 Relationship between the colonization of chestnuts by the chestnut moth Cydia splendana before and after incubation (arcsinesquare-root-transformed data). The curve was fitted using LOWESS smoothing with F ¼ 0.5 [P < 0.001; P ¼ probability of the slope of the regression line to be 0 (zero)] needed for the successful dispersal and germination of this fungus (Ridé and Gudin, 1960). Temperatures between June and August 2003 were 2.9C above the mean temperature 1982–2001 [i.e. the maximum temperature was above 30C at 56 days in this period of time (7.7 days in average in the period 1982–2001), and the amount of rain was only half of the 30-year mean (Bader, 2004; Schär et al., 2004) in southern Switzerland]. A similar situation is reported from eastern France where a drought in 1976 almost completely inhibited C. batschiana on acorns of Quercus robur, although inoculum was present abundantly (Delatour and Morelet, 1979). Right after collection, the frequency of nuts with the exit holes of C. elephas was lowest for nuts that were picked from the ground after maximally 2 days. The difference in respect to other harvest methods disappeared during incubation, indicating that the estimation of chestnut-weevil infestations by visual assessment of the nuts 2 days after falling from the trees is not adequate to estimate the true extent of infestation by this insect. Fruit size had a significant effect on attack by moulds and both insects as assessed immediately after collection. Small fruits were more frequently attacked than marketable ones. This effect disappeared during incubation in regard to moulds and C. elephas, but not in regard to C. splendana. Whereas the female beetle of C. elephas perforates cupula, pericarp and testa of the infructescence with its proboscis and deposits eggs (one per fruit) with its telescopic ovipositor directly onto the cotyledons, C. splendana deposits its eggs on leaves close to infructescences where the larvae hatch and from where they migrate towards the bottom of the cupula, perforate the cupula and the hilum of the pericarp to reach the cotyledons (Breisch, 1995; Conedera et al., 2004). Perhaps the hilum of big, marketable fruits is tougher than that of small fruits and consequently more difficult to perforate. This might explain the higher frequency of small fruits attacked by this insect. The mother tree had the greatest effect on the infestation by insects and fungi, but this effect was related to the tree variety only in regard to colonization by the chestnut weevil C. elephas. On average, the fruits of the variety ÔLüinaÕ got more frequently attacked by this insect. This may be due to both morphological and phenological reasons. From the morphological point of view, although shorter in length, the density (number of spines per surface area) of spines on the cupulae of the variety Torcion is higher than that of the variety Lüina and, thus, access of female beetles to the surface of the cupulae for oviposition may be easier on trees belonging to the variety Lüina (Conedera, 1994). From the phenological point of view, Lüina and Torcion display different flowering times, being Lüina 6–10 days earlier in flowering, which may cause a different proneness to beetle attack at time of oviposition (Rudow and Conedera, 2001). The relative position of trees in the orchard had an effect on colonization by C. elephas only. Trees 2 and 3, which are close neighbours and relatively far away from the other trees, were most severely attacked by C. elephas (Table 4), indicating that the place where these two trees grow supports a larger population of this beetle than other places in the orchard. The mobility of many species of beetles is low [i.e. the beetles deposit their eggs not far from the place where they metamorphosed into adults (Schaper, 1992; Scherzinger, 1996)]. The larvae of C. elephas hibernate in the soil not far from the nuts in which they developed. The beetles emerge from soil during one of the following summers. If the mobility of C. elephas were low, the probability would be high that this beetle deposits its eggs on the same tree on which it was born and, thus, the incidence of the insect would increase on and in the vicinity of this tree over the years if no phytosanitary measures were taken. Chestnuts bisected and inspected right after collection were most frequently colonized by Penicillium and Mucor species. Wells and Payne (1980) reported similar results in regard to Penicillium spp. In the present study, colonization by Penicillium and Mucor species greatly increased during incubation at 24C indicating either that the fruits had already been colonized latently before incubation or that the contamination occurred during incubation. Infestations by Penicillium and Mucor species are common during seed storage and are suspected to reduce germinability (Ridé and Gudin, 1960; Breisch, 1995; Conedera et al., 2004; Knudsen et al., 2004; Schröder et al., 2004). Thermohydrotherapy was shown to enhance infestation by these species probably as a consequence of the Harvest Method and Infestation of Chestnuts elimination of antagonists (Delatour, 1978; Knudsen et al., 2004). Most isolates of Penicillium found in the present study were identified as P. expansum based on morphology using the key and description of Domsch et al. (1980), but it could not be determined to which of the two varieties, P. expansum var. crustosum or P. expansum var. expansum, the isolates belong to. P. crustosum Thom is the basionym of P. expansum var. crustosum (Thom) Fassat. According to Overy et al. (2003), P. crustosum was isolated from two-thirds of the marketed chestnuts purchased in seven Canadian provinces and was shown to produce the tremorgenic toxin penitrem-A. If P. expansum found in this study and P. crustosum in the Canadian study were conspecific, penitrem-A contamination of the Swiss chestnuts would be a matter of concern. Of course, commercially produced chestnuts are stored at 0–2C and not at 24C as in the present study. However, P. expansum still grows satisfactorily at 0C and stops growing only below )3C (Ridé and Gudin, 1960; Domsch et al., 1980). Acrospeira mirabilis was the third most frequently observed taxon (following Penicillium spp. and Mucor spp.) before and after incubation. Its biology is not well-known, and it remains to be ascertained whether it should be considered an endophyte that is able to colonize the seeds at an early embryonic state already. The fungus was isolated from acorns of Quercus robur and inflorescences and infructescences of Castanea species (Ellis, 1971; Knudsen et al., 2004). It is widely distributed in the northern temperate regions where it can cause significant losses to chestnuts. In addition, it was also reported from Chile and New Zealand on introduced Castanea spp. (Pennycook and Galloway, 2004). Amphiporthe castanea was the only fungus that was influenced by the harvest method. However, the method effect was statistically significant for the nuts of only two mother trees. Nuts either intercepted in nets or collected within 2 days from the ground were more frequently colonized by A. castanea. This indicates that colonization by A. castanea decreases concomitantly with increasing intensity or the duration of exposure to more competitive fungal colonizers. Most fungi found during this study are soil-borne saprotrophs, whereas A. castanea is a well-know facultatively biotrophic fungus that colonizes the bark of chestnut trees as a harmless endophyte (Bissegger and Sieber, 1994). A. castanea was present in nuts inspected within 2 days after their falling from the trees in the present study, indicating that this endophyte probably is transmitted vertically by seed to the next generation of chestnut trees. Consequently, this fungus colonizes the fruits long before fruit falls, probably in spring and early summer. A. castanea is exposed to more competitive soil-borne fungi only after fruit fall and is either replaced or suppressed by them (i.e. the longer and more intense exposure after fruit fall the less probable is detection of A. castanea). Trichothecium roseum has a worldwide distribution, and was isolated from various types of soils and plant 503 debris but is also known to colonize the seeds of various plant species (Domsch et al., 1980). Schröder et al. (2004) were able to isolate T. roseum from Quercus robur acorns. The fungus has antifungal properties, produces several toxins and was demonstrated to colonize sclerotia of Sclerotinia sclerotiorum and Claviceps purpurea (Domsch et al., 1980). This fungus may, therefore, be considered a candidate biocontrol agent against sclerotia-forming Ciboria batschiana. Clonostachys rosea is another well-known mycoparasite with a worldwide distribution (Moller et al., 2003; Nobre et al., 2005). It occurs in various types of soils and was isolated from the seeds of several cereal species (Domsch et al., 1980). Knudsen et al. (2004) used it as seedcoating of Q. robur seeds after thermo-hydrotherapy and were able to increase survival and germination rate considerably. C. rosea is also effective against Fusarium spp. (Jensen et al., 2000; Knudsen et al., 2004), and it was recently shown to be insecticidal (Toledo et al., 2006). Fusarium species are regularly found to attack the seeds of various plant species (Domsch et al., 1980; Knudsen et al., 2004; Schröder et al., 2004). F. roseum often colonizes the cotyledons in chestnut seeds (Breisch, 1995). We never detected this species, but F. oxysporum was regularly observed. Wardomyces columbinus is known from soils and decaying plant material in Europe (Hennebert, 1968; Domsch et al., 1980), but has never been isolated from chestnuts, so far. Mycelia that formed colonial morphologies typical for Phomopsis spp. were isolated from a few nuts. Unfortunately, species identification was not possible because the cultures remained sterile. Phomopsis castanea (Sacc.) Höhn. and P. endogena (Speg.) Cifferi (syn. P. viterbensis Camici, Phoma endogena Speg.) are described to be associated with chestnuts. Whereas P. castanea is a harmless endophyte in aerial tissues of C. sativa (Washington et al., 1999), P. endogena causes a brown–white rot and later mummification of the cotyledons (Ridé and Gudin, 1960). Botrytis cinerea was never observed during this study. This fungus usually enters into the fruits via insect galleries. Fruits with galleries were abundant in this study, but probably the inoculum was missing due to the exceptionally dry summer. Interception of chestnuts in nets does not have an effect on colonization by moulds at least when autumns are dry. In contrast, the mother tree and the cultivation practices (tillage operations, phytosanitary measures) applied to the tree and the soil below significantly affect the colonization of nuts by insects and fungi. To study the influence of the genetic disposition of trees and various cultivation techniques on infestations by insects and fungi will be exciting topics for future research. Acknowledgements We would like to thank to Viviana Chiesi, Alberto Sassella, Franco Fibbioli, Roberto Rigoni, Agroscope, Cadenazzo, for fruit harvest 504 and control and Angelo Duò, ETH Zurich, Forest Pathology and Dendrology, Zürich, for excellent technical assistance. References Anagnostakis SL. (2003) Growing chestnut trees. Available at: http://www.icserv.com/nnga/faqchest.html Bader S. Die extreme Sommerhitze im aussergewöhnlichen Witterungsjahr 2003 – Arbeitsbericht Nr. 200. Zürich, MeteoSchweiz, 2004. Bassi D, Casiraghi MC, Mignani I, Vercesi A, Delaidelli G. Effetto dei trattamenti postraccolta e dei metodi di conservazione sulla qualità delle castagne. In: Bellini E (ed.), Atti del Convegno Nazionale sul Castagno, Marradi, Italia, 25–27 October 2001. Florence, University of Florence, 2001, pp. 244–249. Bissegger M, Sieber TN. (1994) Assemblages of endophytic fungi in coppice shoots of Castanea sativa. Mycologia 86:648–655. Bonnemaison L. Les ennemis animaux des plantes cultivées et des foreˆts. Paris, Editions Sep, 1962. Bounous G. Raccolta. In: Bounous G (ed.), Il castagno. Bologna, Edagricole, 2002, pp. 125–129. Bovey P, Linder A, Müller O. (1975) Recherches sur les insectes des châtaignes au Tessin (Suisse). Schweiz Z Forstw 126:781–820. Breisch H. Harvest, storage and processing of chestnuts in France and Italy. In: Antognozzi E (ed.), International Congress on Chestnuts, Spoleto, Italy, 20–23 October 1993. Perugia, University of Perugia, 1993, pp. 429–436. Breisch H. Châtaignes et marrons. Paris, Centre Technique Interprofessionnel des Fruits et Légumes (Ctifl), 1995. Cleveland WS. (1981) LOWESS: a program for smoothing scatterplots by robust locally weighted regression. Am Stat 35:54–54. Conedera M. (1994) Inventario e carratterizzazione genetica delle varietà nostrane di castagno da frutto. Boll Soc Ticinese Sci Nat 82:39–50. Conedera M, Jermini M, Sassella A. (1997) Nouvelles perspectives pour la culture du châtaignier au sud des Alpes. Rev Suisse Viticult Arboricult Horticult 29:337–344. Conedera M, Jermini M, Sassella A, Sieber TN. Ernte, Behandlung und Konservieren von Kastanienfrüchten - Merkblatt für die Praxis Nr. 38. Birmensdorf, Switzerland, Eidgenössische Forschungsanstalt WSL, 2004. Delatour C. (1978) Research on a curative method of control of Ciboria batschiana (Zopf) Buchwald on acorns. Eur J For Pathol 8:193–200. Delatour C, Morelet M. (1979) La pourriture noire des glands. Rev For Fr 31:101–115. Della Beffa G. Gli insetti dannosi all’agricoltura. Milan, Edizioni Hoepli, 1961. Domsch KH, Gams W, Anderson TH. Compendium of Soil Fungi. London, UK, Academic Press, 1980. Ellis MB. Dematiaceous Hyphomycetes. Kew, Surrey, UK, Commonwealth Mycological Institute, 1971. Fadanelli L, Chemolli M, Conci S. La conservazione delle castagne: sistemi e formule applicative. Atti del Convegno ÔIncontro Tecnico sul Castagneto da FruttoÕ, Roncegno, 29 October 1994. San Michele all’Adige, Quaderni di esperienze e ricerche, 1995, pp. 43–50. Giacalone G, Bounous G. (1993) Tradizione ed innovazione nella trasformazione e nell’utilizzo delle castagne. Monti Boschi 44: 33–41. Hennebert GL. (1968) Echinobotryum, Wardomyces and Mammaria. Trans Br Mycol Soc 51:749–762. Jensen B, Knudsen IMB, Jensen DF. (2000) Biological seed treatment of cereals with fresh and long-term stored formulations of Clonostachys rosea: biocontrol efficacy against Fusarium culmorum. Eur J Plant Pathol 106:233–242. Jermini M, Rigotti S, Brunetti R, Conedera M. (2001) Investigation on fruit pests affecting chestnut trees in southern Switzerland. Ital Hort 8:17–18. Sieber et al. Jermini M, Conedera M, Sieber TN et al. (2006) Influence of fruit treatments on perishability during cold storage of sweet chestnuts. J Sci Food Agric 86:877–885. Knudsen IMB, Thomsen KA, Jensen B, Poulsen KM. (2004) Effects of hot water treatment, biocontrol agents, disinfectants and a fungicide on storability of English oak acorns and control of the pathogen, Ciboria batschiana. For Pathol 34:47–64. Moller K, Jensen B, Andersen HP, Stryhnz H, Hockenhull J. (2003) Biocontrol of Pythium tracheiphilum in Chinese cabbage by Clonostachys rosea under field conditions. Biocontrol Sci Technol 13:171–182. Nobre SAM, Maffia LA, Mizubuti ESG, Cota LV, Dias APS. (2005) Selection of Clonostachys rosea isolates from Brazilian ecosystems effective in controlling Botrytis cinerea. Biol Control 34:132–143. Overy DP, Seifert KA, Savard ME, Frisvad JC. (2003) Spoilage fungi and their mycotoxins in commercially marketed chestnuts. Int J Food Microbiol 88:69–77. Pennycook SR, Galloway DJ. Checklist of New Zealand ÔfungiÕ. Fungi of New Zealand. Nga Harore o Aotearoa 1. Hong Kong, Fungal Diversity Press, 2004. Pollini A. Manuale di entomologia applicata. Bologna, Edizioni Edagricole, 1998. Ridé M, Gudin C. (1960) Sur la biologie de quelques champignons parasites de la châtaigne et plus particulèrement de Phoma endogena (Speg.). CR Acad Agric Fr 46:536–543. Rudow A, Conedera M. (2001) Blüte und Sortenerkennung bei der Edelkastanie (Castanea sativa Mill.) auf der Alpensüdseite der Schweiz. Bot Helv 111:1–23. Rutter PA, Miller G, Payne JA. Chestnuts. In: Moore JN, Ballington JR (eds), Genetic Resources of Temperate Fruits and Nut Crops. Wageningen, the Netherlands, International Society for Horticultural Science, 1990, pp. 761–788. Schaper C. (1992) Totholz im Wirtschaftswald. Allg Forst Zeitschr 47:598–599. Schär C, Vidale PL, Lüthi D et al. (2004) The role of increasing temperature variability in European summer heatwaves. Nature 427:332–336. Scherzinger W. Naturschutz im Wald: Qualitätsziele einer dynamischen Waldentwicklung. Stuttgart, Ulmer, 1996. Schröder T, Kehr R, Prochazkova Z, Sutherland JR. (2004) Practical methods for estimating the infection rate of Quercus robur acorn seedlots by Ciboria batschiana. For Pathol 34:187–196. Tian SP, Bertolini P. (1997) Biology and pathogenicity of Rhacodiella castaneae in chestnut stored at low temperatures. J Plant Dis Protect 104:23–28. Toledo AV, Virla E, Humber RA, Paradell SL, Lastra CCL. (2006) First record of Clonostachys rosea (Ascomycota: Hypocreales) as an entomopathogenic fungus of Oncometopia tucumana and Sonesimia grossa (Hemiptera: Cicadellidae) in Argentina. J Invertebr Pathol 92:7–10. Washington WS, Allen AD, Dooley LB. (1997) Preliminary studies on Phomopsis castanea and other organisms associated with healthy and rotted chestnut fruit in storage. Aust J Plant Pathol 26:37–43. Washington WS, Hood V, Goubran F, Hepworth G, Stewart-Wade S. (1998) Effect of fungicides applied as foliar sprays and trunk injections on nut rot of chestnuts caused by Phomopsis castanea in Victoria. Aust J Exp Agric 38:295–303. Washington WS, Stewart-Wade S, Hood V. (1999) Phomopsis castanea, a seed-borne endophyte in chestnut trees. Aust J Bot 47: 77–84. Wells JM, Payne JA. (1980) Mycoflora and market quality of chestnuts treated with hot water to control chestnut weevil. Plant Dis 64:999–1001.