The ooriginal publication is available nline.


© The Japanese Forest Society and Springer-Verlag Tokyo 2005

J For Res (2005) 10:93-99

DOI 10.1007/s10310-004-0101-3

Ectomycorrhizal development in a Pinus thunbergii stand in relation to the location on a slope and their effects on tree mortality from Pine Wilt Disease

Tamio Akema

Forest Microbiology Group, Kyushu Research Center, Forestry and Forest Products Research Institute
4-11-16, Kurokami, Kumamoto-shi, Kumamoto 860-0862, Japan
Tel. +81-96-343-3948; Fax +81-96-344-5054

Kazuyoshi Futai

Laboratory of Environmental Mycoscience, Graduate School of Agriculture, Kyoto University, Sakyo-ku, Kyoto, Japan

Received: 29 August 2003 Accepted: 18 June 2004

Abstract

The relationship between ectomycorrhizal development and mortality from pine wilt disease was studied in an artificial Pinus thunbergii Parl. stand on a slope. The development of ectomycorrhizae and the survival of the trees showed the same tendency, which suggests a correlation between the mycorrhizal development and the resistance to pine wilt disease. The development of pine roots and mycorrhizae was greater in the upper part of the slope. The ratio of mycorrhizae to the total of mycorrhizae and fine tap roots was also higher in the upper part of the slope. Tree mortality was clearly biased and more trees survived in the upper part of the slope than in the middle and the lower part. There was no significant difference between the upper and the lower part of the slope in the number of feeding wounds marked by the pine sawyer beetle which show the opportunity of infection with this disease. There were no clear correlation between the development of mycorrhizae and the content of soil substrate such as total carbon, nitrogen and phosphorus. The abundant mycorrhizae in the upper part of the slope which mitigate the drought stress might have decreased the tree mortality.

key words

Ectomycorrhiza, Pine wilt disease, Drought stress, Tree mortality, Pinus thunbergii


Introduction

Japanese red pine, Pinus densiflora Sieb. & Zucc., and Japanese black pine, P. thunbergii Parl., are major components of exploited suburban forests and maritime forests in Japan, respectively. However, such pine forests, especially those in the south-western part of Japan have been devastated by pine wilt disease in the late twentieth century, and the disease is spreading northward (Kishi 1995).

Pines are well-known ectomycorrhizal trees. Ectomycorrhizae perform many functions such as improvement of host resistance to root diseases (Marx 1969, Buscot et al. 1992), enhancement of nutrient and water uptake (Finlay and Read 1986, Bledsoe 1992), and are generally recognized to supplement and/or enhance the functions of roots. Pines are presumed to show virtually absolute dependence upon mycorrhizal fungi (Allen 1991). Pine wilt disease itself is caused by a nematode, Bursaphelenchus xylophilus (Steiner & Buhrer) Nickle and has no direct relation to mycorrhiza, but the tolerance level may be increased indirectly by mycorrhizal colonization since mycorrhizae generally improve the physiological conditions of pine trees.

The purpose of this paper is to examine the abundance and distribution of pine roots and mycorrhizae in relation to their location on a slope, and then to correlate the distribution of trees killed by pine wilt disease. It is empirically said that pine trees growing near the ridge have larger amount of mycorrhizae than the trees on the bottom of the slope, but this has been empirically attributed to the complex effects of the differences in soil characteristics including fertility, water condition, microorganisms, and so on. In an attempt to elucidate the determining factor of the mycorrhizal distribution, we also measured total carbon, total nitrogen, extractable phosphorus and soil water conditions.


Materials and Methods

Study Site

The study site was an artificial pine stand in Tokuyama Experimental Station of Kyoto University Forest, Yamaguchi Prefecture (34°04´N 131°50´E). The annual mean temperature and precipitation in the station were 14.9°C and 1,915 mm, respectively. The rainfall is mainly in June and July, and the drought in early spring and summer is often severe. The altitude was between 240 and 270 m and the average inclination is 25 degrees. The top of the slope is the ridge line and the bottom end is cut vertically to make a forestry road, so the lowermost part of the slope is a bluff of 1 to 2 m in height. This stand was planted with several local strains of Japanese red pine collected mainly from western Japan and 16 half-sib strains of Japanese black pine trees along the slope according to the locality where they are from in 1970.

Until 1980 most trees in this stand had been thriving, but thereafter many trees have been killed by pine wilt disease year after year, and only 28.1% of the initial population was left at the end of 1993 (Nakai et al. 1995). The location of the surviving and died pine trees are shown in Fig. 1. We made two field surveys in July 1993 and May 1997. The first survey were followed by some supplementary samplings in 1993 and 1994.

Fig. 1. The position of living pine trees in 1993 (open circles) and killed ones (black circles) between 1980 and 1993 in Tokuyama Experimental Station. The top of each column is the ridge line. Frames show the area shown in Figs. 2.

We analyzed the spatial data used in Nakai et al. (1995) provided by the authors to find some tendency of mortality depending on the position in a slope. As the trees were planted in columns along the slope, trees in each column were divided into three equal parts simply by the number from the top. If the remainder is one, it was assigned to the middle group. When it was two, they were assigned to the upper and the lower group.

Sampling

First survey (1993 and 1994)

Soil samples were collected mainly from the stand of local strains of P. thunbergii No. 241 and No. 236, both of which survived at exceptionally high percentage. These two strains seems to have higher tolerance than other strains since both of the strains planted in each separated stand showed low mortality.

Root samples were collected from four subplots arranged at various heights on the slope, named 1 for the uppermost, 2 for the upper middle, 3 for the lower middle and 4 for the lowermost. 10 replications for each subplot were collected and the sampling points were allocated in level line for each height (Fig. 2a). We put a 30×30 cm wooden frame on the ground as a jig and cut the roots by thrusting a knife along the frame. For each point, three soil samples were collected from different depths; 0-5 cm, 5-10 cm and 10-15 cm. The soil samples were washed under running water on a 5 mm-meshed sieve and plant roots remaining on the sieve were brought to the laboratory in polyethylene bags. Thickness of A0 and A layers measured in subplots 1 to 4 in Fig. 2a are shown in Table 1.

Fig. 2 The sampling points. a Sampling points in the first survey for roots, mycorrhizae and soil nutrients (dots, subplot 1 to 4) and for water potential (triangles, subplot A and B). Squares, circles and crosses show the position of surviving pine trees at early 1994 of strain 236, 241 and Shimonoseki, respectively. b Sampling areas for roots and mycorrhizae (subplot H and L) in the second survey. Dots and circles are the surviving pine trees at early 1997 of strain 231 and 232, respectively.

Table 1. Thickness of A0 and A layers (cm) measured in the first survey at subplots 1 to 4 in Fig. 2a.
Subplots
1234
A0 layer6.0 ± 0.306.2 ± 0.295.4 ± 0.345.7 ± 0.40
A layer8.8 ± 1.0512.0 ± 1.5014.8 ± 1.7011.3 ± 1.17

Values shown are mean ± SE

To measure the concentration of carbon, nitrogen and phosphorus, we collected supplemental soil samples using 100 ml core sampler in October 1993 from the same points as previous.

We collected further 20 soil samples, 100 ml for each, to determine the water characteristics for study site in July 1994. A ten samples from the upper part of the slope between subplot 1 and 2, and the other ten samples from the lower part, between subplot 3 and 4 (Fig. 2a, subplot A and B respectively). Samples were collected from the depth of 5-10 cm.

Samples to estimate the water potential in the field were collected in October 1993 and July 1994. In 1993, four ten samples were collected in subplot 1 to 4. Two ten samples from subplot A and B were collected in 1994.

Second survey (1997)

The subject of first survey was a single stand and might be insufficient to generalize the result, so we carried out another survey in different stand in Tokuyama Experimental Station. In May 1997, soil samples were collected from upper and lower part of the slope in the same stand but under different local strain of Japanese black pine, No. 231 (Fig. 2b). This strain is slightly less resistant to pine wilt disease than No. 241 and 236 although the difference is not significant (p> 0.1, χ2-test). Other strains hardly survived in the lower part of the slope and this was the only available strain in 1997. Twenty five samples were randomly collected from the upper part (subplot H) where many trees were surviving, and another 30 samples were collected from the lower part (subplot L) around the few surviving trees only for the measurement of roots and mycorrhizae. All soil samples were contained in polyethylene bags, and brought to the laboratory.

Distribution of roots and mycorrhizae

First survey:

We immersed the samples in running water, then sorted using forceps into the roots of pines and other plants. The pine roots were then classified into mycorrhizae, tap roots smaller than 2 mm in diameter which may bear mycorrhizae, and the roots larger than 2 mm in diameter. The biomass in each class of roots was evaluated by fresh weight. Mycorrhizal ratio was evaluated as the proportion of the weight of mycorrhizal roots to that of fine tap roots plus mycorrhizal roots.

Second survey:

All pine roots were sorted out from soil samples and thick roots (over 2 mm in diameter) were discarded. The remaining fine tap roots and mycorrhizae were cut into small pieces (less than 1 cm) by a mixer with 400 ml of water and 100 ml of the suspension was subsampled. The root pieces in subsamples were sorted into tap roots and mycorrhizae under dissecting microscope and dried in 65°C drying oven for one day, and their dry weights were evaluated. When the amount of roots is very small, whole volume of the suspension served for measurement. Mycorrhizal ratio was also evaluated as described above.

Soil nutrients

We conducted all soil nutrient analysis on the samples collected in October 1993 at the subplots 1 to 4. Total amount of carbon and nitrogen were measured by CN coder (YANACO MT-600). A 0.5g of soil served the measurement, and 4g of cobalt oxide was used as oxidizer to combust soil. Extractable phosphorus was extracted with dilute hydrochloric acid and sulfuric acid, and the amount was determined by Molybdate-Vanadate method (Nelson et al. 1953).

Water condition

Soil samples collected in July 1994 at subplots A and B were brought in polyethylene bags to the laboratory, and were loaded into stainless 100 ml soil sampling tubes. After immediate measurement of fresh weight, the water contents at the potential of 0 to -3.1, -9.8 to -98, -1.6×103, -3.1×104 and -3.1×105 kPa (pF 0 to 1.5, 2.0 to 3.0, 4.2, 5.5 and 6.5) were determined by sand column method, pressure plate method, centrifuge method, air drying and oven drying, respectively. The water potential of the samples in the field condition was estimated using those data. Because the sampling points in 1993 (subplot 1 to 4) were different from those in 1994 (subplot A and B), values for 1993 were substituted by the averages of subplots 1 and 2 for subplot A, subplot 3 and 4 for subplot B.

Number of feeding wounds marked by pine sawyer beetle

The number of feeding wounds marked on 1-year and 2-year internodes of the twigs by pine sawyer beetle, Monochamus alternatus Hope, the vector of B. xylophilus, was counted to see if the biased distribution of wilted pine trees could be ascribed to sawyer beetle's preferential feeding behavior bringing greater infection frequency of pine wood nematode to the lower part of the slope. For this purpose five living trees planted both at the upper and lower part of the stand were examined in October 1993 in the stand of strain 241. About 20 twigs were arbitrarily sampled from each of upper and lower branches of the crown.


Results

Distribution of pine trees killed by pine wood nematode

The mortality of each group along the slope (upper, middle and lower) between 1980 and 1997 was calculated. The mortality varied among strains (p< 0.01, χ2-test) but more trees survived in the upper part than in the middle and the lower part (P< 0.001, Friedman's test) regardless of strains (Fig. 3).

Fig. 3. The survival rate of the Japanese black pine strains on the upper, middle and lower part of the slope between 1980 and 1997.

Distribution of roots and mycorrhizae and the mycorrhizal ratio

First survey:

As shown in Fig. 4, the largest amount of pine roots distributed in subplot 1 (the uppermost on the slope), while the other plant roots did not show any tendency in their distribution. In the pine roots, more mycorrhizae and fine tap roots distributed in subplot 1.

Irrespective of the height on the slope, the amount of mycorrhizae was smaller in deep layer (10-15 cm) than those in surface and middle layers though the surface layer of subplot 3 contains smaller amount of mycorrhizae than the middle layer. In subplot 1, mycorrhizal biomass decreased with the soil depth.

Fig. 4. Distribution of roots at different heights on the slope and in different depth from the soil surface after the removal of litter in the samples collected in the first survey at the subplots 1, 2, 3 and 4 in Fig. 2a. Other plants include shrubs, grass, herbs and ferns except tubers. Pine thick roots are thicker than 2 mm in diameter and bare only tap roots. Fine tap roots are pine tap roots which are thinner than 2 mm and bear mycorrhizae. Mycorrhizae are ectomycorrhizae formed on pine roots. Symbols and bars are mean ± S. E.

The proportion of mycorrhizae to the total of fine tap roots and mycorrhizae (mycorrhizal ratio) at various heights on the slope was compared (Fig. 5). Mycorrhizal development was greater in the higher part of the slope (subplots 1 and 2). Mycorrhizal ratios in the surface and the middle layers were different among subplots (Friedman's test, P< 0.05), while those in the deep layer showed no difference among the subplots.

Fig. 5. The proportion of mycorrhizae to the total of fine tap roots and mycorrhizae by fresh weight in each depth and subplot. Values are mean ± S. E.

Second survey:

The amount of fine tap roots and that of mycorrhizae, and the mycorrhizal ratio are shown in Fig. 6. 20 samples out of 25 collected from the upper part of the slope contained pine roots, while only 4 samples out of 30 collected in the lower part contained pine roots. As far as the samples containing pine roots are concerned, the amount of mycorrhizal roots was significantly different between subplots H and L in Fig. 2b (Mann-Whitney's U-test, P< 0.05) though there was no difference in the amount of fine tap roots of pines between the subplots.

Fig. 6. The amount of fine tap roots and mycorrhizae in the upper subplot (H) and lower subplot (L) on the slope shown in Fig. 2b. Mycorrhizal ratio (the proportion of mycorrhizae to the total of fine tap roots and mycorrhizae) are also shown.

Although only 4 samples contained pine roots in the lower part, the mycorrhizal ratios in all of them were lower than any of those in the upper part, and the difference was significant (Mann-Whitney's U-test, P< 0.01).

Total carbon and nitrogen, and extractable phosphorus

Irrespective of height on the slope carbon content was highest in the surface layer (which roughly corresponded to A0 horizon) and decreased with the depth (Fig. 7). Among 4 sampling subplots in Fig. 2a, the carbon content was lowest at subplot 3. The carbon content did not correlate with the amount of any of total pine roots, fine tap roots, or mycorrhizae. C/N ratios showed significant difference (Kruskal-Wallis's test, P< 0.001 for 0-5, 5-10 cm depth and P< 0.05 for 10-15 cm depth) among the 4 subplots and the C/N ratio was highest in subplot 1. Extractable phosphorus content in the surface layer was not uniform (Kruskal-Wallis's test, P< 0.02) and the samples from subplot 3 contained lower concentration of extractable phosphorus but there were no significant difference among the samples from the deeper layers in each subplot.

Fig. 7. The concentration of total carbon, total nitrogen, extractable phosphorus and C/N ratio of the soil samples collected in October, 1993 in each depth and subplot. Values are given in mean ± S. E.

Water condition

Water characteristic curves of the soils in the upper and lower part of the slope are illustrated in Fig. 8. The available water capacities by weight, which are estimated by the difference in soil water content between field capacity (-6.2 kPa) and permanent wilting point (-1.6×103 kPa), were 25% for the upper part and 30% for the lower part, respectively, and the soil of the upper part consistently showed higher water potential over the whole range of the moisture.

Fig. 8. Water characteristic curves in subplots A and B. Values are mean ± S. E.

Using these water characteristic curves, the water potential of the soil in the upper and the lower part of the slope were estimated in October 1993 and July 1994 (Table 2). The water potential in the upper part varied more widely (-10 to -630 kPa) than the lower part (-20 to -320 kPa).

Table 2. Moisture weight percentage (MWP) and water potential of the soil in the upper and the lower part of the slope.
Position on the slopeMWP(%)Water potential (kPa)
October 1993Upper part38.2-10
Lower part45.5-20
July 1994Upper part23.2-630
Lower part33.7-320

Values in October, 1993 are the average of subplot 1 and 2 for the upper part and subplot 3 and 4 for the lower part. Values in July, 1994 are directly measured at subplot A and B for the upper and the lower part. Sampling points are shown in Fig. 2a.

Number of feeding wounds marked by pine sawyer beetle

The numbers of feeding wounds per twig are shown in Table 3. According to Mann-Whitney's U-test, there were no significant difference between the upper and the lower part of the slope.

Table 3. The number of feeding wounds per twig marked on 1-year and 2-year internode by pine sawyer beetle on higher and lower position on trees located at higher and lower part of the slope.
Position on the SlopeNumber of feeding wounds per twig
Higher twigLower twig
1-year2-year1-year2-year
Upper slope0.70 ± 0.230.55 ± 0.230.57 ± 0.220.76 ± 0.18
Lower slope0.35 ± 0.130.55 ± 0.150.36 ± 0.170.64 ± 0.13

In the lower part of the slope, 20 high branches and 14 low branches were sampled. In the upper part of the slope, 20 high branches and 21 low branches were sampled. Values are given in Mean and S.E.


Discussion

Mycorrhizae are widely believed to improve the water relations of associated plants (Brownlee et al. 1983, MacFall et al. 1991). Mexal and Reid (1973) showed that Cenococcum graniforme (now called C. geophilum), which is a very common mycorrhizal fungus, can grow at the water potential of -20 bars (-2.0×103 kPa, lethal level in most plants) or less. Allen (1991) suggested that fungi such as C. geophilum may support the water uptake of the host during dry season. We also sporadically found C. geophilum in our study site regardless of the subplots (data not shown). So, the pine trees planted at the upper part might be tolerant to drought stress with the assistance of such mycorrhizae. Kikuchi et al. (1991) suggested that mycorrhizal fungi may facilitate the tolerance to pine wilt disease from their results of in vitro experiments although the difference was not statistically significant. To evaluate the role of mycorrhizae in reducing the pine wilt mortality, the associated mycorrhizal fungi must be examined strictly as the determinants of the symbiotic relationship.

Pine wilt disease is considered to be caused by cessation of water flow (Ikeda and Suzuki 1984), i.e. it is a kind of disease caused by water deficiency. As known widely, drought stress accelerates the development of pine wilt disease (e.g. Ikeda 1996). However, more trees were killed in the lower part of the slope where the water condition may be moderate compared with that in the upper part. There were more roots of plants other than pine trees in the lower part of the slope, which suggests water competition between pine trees and other plants are severe in lower part. However, as the soil on the lower part was always more humid than the higher part of the slope, water might be sufficient for both of pine trees and other plants in the lower part.

The biased distribution of killed trees is not attributable to less chance of infection in the upper part, since the number of feeding wound of the vector insects did not show significant difference between the upper and the lower part. So, as Miki et al. (2001) suggested, the pine trees in the upper part of the slope might have higher tolerance to pine wilt disease than those in the lower part. We found distinct difference in the amount of mycorrhizae between the pine trees in the upper and lower part.

In natural forests biased distribution of pine roots and mycorrhizae are often observed, and that have been attributed to the differences in soil nutrient level, humidity, vegetation, mycobiota, and so on, and/or to their complexed effect without enough evidence. Negative correlation has been reported between the content of soil nitrogen or phosphorus and the ectomycorrhizal development (Ruehle and Wells 1984, Daft and Nicolson 1969), though in other studies (e.g. Cordell and Marx 1994) fertilizer (both nitrogen and phosphorus) is neutral to mycorrhizal development. In this study, however, we could not explained the correlation between mycorrhizal development and soil substrates such as carbon, nitrogen, and phosphorus. The thickness of organic layer also did not explain the difference in mycorrhizal development.

We found that the water condition was different between upper and lower part of the slope, but could not describe thoroughly the water condition of the subplots, but this seems to be very important issue. Further intensive study is necessary to elucidate the influence of soil humidity on the root and mycorrhizal development, tolerance to drought and the disease. We did not evaluate the water condition of pine trees itself in this study. To elucidate the function of mycorrhizae in dry condition, analysis of the water condition of host plants is important.

The absorptivity of mycorrhizae are higher than tap roots (MacFall et al. 1991) and we found that the mycorrhizal ratio is higher in the upper part of the slope where more pine roots were observed. This suggests that the root systems in the upper part show higher efficiency in water uptake.

In this study, we found that the mycorrhizal development vary with the height in a slope, which suggests the difference in the efficiency of the root system in water absorption. We hypothesize that the abundant mycorrhizae in the upper part enhance the water uptake of pine trees, mitigate the drought stress and thereby decrease the mortality.

Acknowledgment

The authors thank Drs. T. Kosaki and J. Yanai for their help in evaluating the water characteristics. They also thank Drs. I. Nakai, Y. Akita and S. Kitagawa for their help in sampling soils and providing data of the distribution of killed trees.


Literature cited


Return to Akema's top page