Growth Characteristics of Ectomycorrhizal Seedlings of Quercus Glauca, Quercus Salicina, Quercus Myrsinaefolia, and Castanopsis Cuspidata Planted in Calcareous Soil

To verify the acclimation capacity of evergreen Fagaceae species on calcareous soil, we compared ecophysiological traits between Quercus glauca Thunb., Q. salicina Blume, Q. myrsinaefolia Blume, and Castanopsis cuspidata (Thunb.) Schottky as typical woody species from southwestern Japan. We also examined the inoculation effects of the ectomycorrhizal (ECM) fungi Astraeus hygrometricus and Scleroderma citrinum, and planted seedlings in calcareous soil collected from a limestone quarry. We measured growth, ectomycorrhizal colonization, photosynthetic rate, and concentrations of nutrients in plant organs for A. hygrometricus-inoculated, S. citrinum-inoculated, and non-ECM seedlings. Six months after planting on calcareous soil, seedlings of the three Quercus species inoculated with A. hygrometricus were larger than non-ECM seedlings, especially Q. salicina, which showed the greatest increase in dry mass. The dry mass of C. cuspidata seedlings was inferior to that of the three Quercus species. In the nutrient-uptake analysis, phosphorus, manganese, and iron uptakes were suppressed in calcareous soil for each Fagaceae species. However, seedlings of Fagaceae species that showed better growth had increased concentrations of phosphorus in roots. We concluded that seedlings of Q. salicina and Q. glauca inoculated with A. hygrometricus were best suited to calcareous soil and were considered as useful species for the reforestation in limestone quarries.


Introduction
Limestone occurs throughout East Asia [1].Calcareous soil, which is formed by weathering of limestone, shows the pH range 7.5-8.5 owing to the buffering effects of the large amounts of calcium carbonate [2,3].Under such conditions, the uptake of micronutrients (e.g., iron (Fe), manganese (Mn), copper (Cu) and zinc (Zn)), which exhibit low availability at high pH, is suppressed [3,4].Additionally, carbonate in calcareous soil causes lime-induced chlorosis as a consequence of Fe deficiency [2,3].Moreover, phosphorus (P) is bound by calcium (Ca), and the uptake of P is also suppressed [3,5].Accordingly, some plant species exhibit poor growth on calcareous soils [5][6][7][8].Limestone has been quarried at many sites worldwide [9], but once quarrying has finished, natural revegetation of such sites is slow [10,11].A limestone quarry site in Japan is currently being rehabilitated to a forest stand [12].The calcareous soil at the quarry has a high pH and large amounts of Ca [12].Seed from a range of plant species from an adjacent natural vegetation site was dispersed across the quarry site; however, growth of germinated plants was slow [13].Thus, limestone quarry sites require artificial reforestation for environmental conservation of such sites.

Soil Collection and Analysis
In January 2009, we collected calcareous soil from the B horizon of a limestone quarry (Todaka Mining Co., Tsukumi, Oita Pref., Japan, 33 • 03 N, 131 • 48 E, 580 m a.s.l.).The color of the soil was dark red (5YR 3/6), and it showed high viscosity owing to a high content of clay [43].Collected soil was sieved (2 mm) at the same site and collected in 32 sandbags.After transportation, soil was kept in a storehouse in Kyushu Research Center, FFPRI.
The pot experiments were carried out by dividing them into two groups, in May 2009 for Q. glauca and Q. salicina and in May 2010 for Q. myrsinaefolia and C. cuspidata.Before planting for each year, the collected soil was placed in autoclave bags and sterilized using a wet cycle of an autoclave at 120 • C for 20 min.(KTS-2346B, Alp Co., Tokyo, Japan) to remove all other mycorrhizal species.After sterilization, five soil samples were collected from five bags for each year.
We measured the soil properties including pH and the concentration of carbon (C), nitrogen (N), exchanged P, base cations (Ca, magnesium (Mg), potassium (K) and sodium (Na)), Fe, and Mn.Soil pH was measured using a pH meter (SG2, Mettler Toledo, Zürich, Switzerland); 10 g of soil was mixed with 25 mL of distilled water and shaken for 1 h before the reading was taken [44].
After measuring the soil pH, samples were dried at 105 • C for 24 h for further analysis.Dried soil samples were used to determine the concentration of C and N using a NC analyzer (Sumigraph NC-220F, Sumika Chemical Analysis Service, Tokyo, Japan).Exchanged P was extracted by the Olsen method [45] by shaking for 1 h.P concentration in the soil solution was determined by the molybdenum blue method [46] using a spectrophotometer (UV-2500PC, Shimadzu, Kyoto, Japan).
The concentrations of exchanged base cations and Mn were determined; 2.5 g of dry soil was mixed with 50 mL of 1 M ammonium acetate solution and shaken for 1 h prior to analysis [45].Solutions for analysis of exchanged Fe were obtained by adding 50 mL of 1 M KCl to 2.5 g of dry soil and shaking for 1 h [45].The base cations, Fe, and Mn concentrations in the soil solution were analyzed using an inductivity coupled plasma analyser (ICPE-9000, Shimadzu, Kyoto, Japan).

Establishment of the Pot Experiment
A total of 48 mycorrhizal seedlings and 24 non-mycorrhizal seedlings of the four Fagaceae species were transported to Kyushu Research Center, FFPRI in May.A total of 24 mycorrhizal seedlings and 12 non-mycorrhizal seedlings each of Q. glauca, Q. salicina, Q. myrsinaefolia, and C. cuspidata were transplanted into free-draining pots (depth: 30 cm, diameter: 15 cm, volume: 3.7 L) filled with calcareous soil.The remaining seedlings, i.e., 24 mycorrhizal seedlings and 12 non-mycorrhizal seedlings of the four Fafaceae species, were sampled at this time.Fagaceae seedlings were raised for 6 months in a naturally illuminated phytotron (Koito Co., Tokyo, Japan) at Kyushu Research Center, FFPRI.The phytotron had three rooms, and seedlings of Q. glauca, Q. salicina, Q. myrsinaefolia, and C. cuspidata were inoculated with A. hygrometricus (abbreviated to AH), S. citrinum (SC), and non-ECM (NE), the seedlings were then put into the three rooms.Four seedlings of each species inoculated with AH, SC and NE were placed in each room.
After establishment, each pot was irrigated with 60 mL of distilled water per day.The average daily photosynthetically active radiation (PAR) during the experimental period was 39.6 mol•m −2 •day −1 in 2009, and 37.5 mol•m −2 •day −1 in 2010, calculated using data on solar radiation from the Japan Meteorological Agency [47] and the ratio of solar radiation to PAR [48].Air temperature in the phytotron rooms was controlled at 30 • C (day) and 25 • C (night) from July to September, and 25 • C (day) and 20 • C (night) from May to June and in October.These temperatures were based on air temperature in Kumamoto city, Japan [47].Relative humidity was controlled at 75%.During the experiment, some contamination by airborne spores was noted.The position of seedlings was changed at regular intervals (every month), and the phytotron rooms were disinfected using ethanol at this time.

Measurement of Seedling Growth
To determine the growth characteristics of seedlings of Q. glauca, Q. salicina, Q. myrsinaefolia, and C. cuspidata, we measured the dry mass of leaves, stems and branches, and roots.Twelve AH, SC, and NE seedlings each for the four tree species were harvested in May (first sampling) and November (second sampling).The roots of harvested seedlings were washed twice with tap water to remove soil, and then washed with distilled water.The washed seedlings were divided into leaves, stems and branches, and root components.Each component was put into its own envelope and oven-dried at 60 • C for 4 days.The dry mass of each component was determined.From the values of dry mass of each component at the first and second samplings, we calculated relative dry mass as follows: Dry mass at second sampling/Dry mass at first sampling × 100 (1) For the relative dry mass, 100 was set as the value at the first sampling.We also measured specific leaf area (SLA [7]) for each species.We used the SLA value to convert the concentration of nutrients into area based values.

Measurement of the Rate of Colonization of ECM Fungi
To measure the proportion of roots with ECM fungal colonization, we selected five large clusters of roots from each of seedling, and over 500 short roots (<5 mm in length) that diverged from them.This proportion was determined before drying.For assessment of ECM fungi, roots of seedlings were harvested at the first and second samplings and carefully washed free of soil under gently flowing water.Cluster roots were soaked in distilled water and observed at 10-40× magnification using a stereomicroscope (STZ-40TBIT, Shimadzu, Kyoto, Japan).ECM short roots of A. hygrometricus and S. citrinum were confirmed by the color, form, and size of the ECM fungi, the presence of rhizomorphs and the absence of root hairs [38,40,49].The numbers of colonized and uncolonized short roots were counted, and we calculated the percentage of ECM fungal colonization from the proportion of colonized short roots to the total number of short roots [50].We also examined the relationship between growth (total dry mass) and ECM fungal colonization.

Photosynthetic Capacity
The photosynthetic rate at light saturation (P sat ) was measured in all seedlings in October.Third leaves counted from the top of the seedling were used for the measurement of P sat and stomatal conductance (g s ).Measurements were made using a portable gas analyzer (LI-6400, LI-COR Biosciences, Lincoln, NE, USA) under steady-state conditions at an ambient CO 2 concentration of 37.0 Pa.The LED light source was adjusted to a saturation light level of 1700 µmol•m −2 •s −1 •PPF.

Analysis of Element Concentrations in Plants
We measured the concentrations of N, P, K, Ca, Mg, Fe, and Mn in leaves and roots at the first and second samplings.Dried samples were ground to a fine powder using a sample mill (WB-1; Osaka Chemical Co., Osaka, Japan).The concentration of N was determined using a CN CORDER analyzer (MT-600, Yanako New Science Inc., Kyoto, Japan).The remaining samples were digested by the HNO 3 -HCl-H 2 O 2 method [51].The concentration of P was determined by the molybdenum blue method using a spectrophotometer.Concentrations of K, Ca, Mg, Fe, and Mn were analyzed using an atomic absorption spectrophotometer (Z-2310, Hitachi High-Technologies Co., Tokyo, Japan).
The concentrations of each nutrient at the first and second samplings were calculated as relative nutrient concentration, as follows: Concentration at second sampling/Concentration at first sampling × 100 For the relative concentrations, 100 was set as the value from the first sampling.Based on relative nutrient concentration and relative dry mass, we examined vector analysis in accordance with Valentine and Allen [52] and Scagel [53].Changes in dry mass and the concentration of nutrients were plotted for AH, SC, and NE seedlings of the four Fagaceae species.The interpretation of vector shifts, as described in Valentine and Allen [52], are shown in Figure 1 as a reference.
We also examined the effects of nutrients on photosynthetic rate.In general, photosynthetic rate is closely related to N [7,54] and P [55,56].Correlation analysis was performed between P sat and the concentrations of N and P in leaves at the second sampling.For the concentrations of N and P, we used the SLA area-based values.[47] and Scagel [48]).The curve indicated by the broken line shows the border to estimate increase or decrease of content.A plot located above the curve shows increase of content.

Statistical Analysis
All parameters were analysed for normality and homoscedasticity using the Shapiro-Wilk test with Sigma Plot 12.3 (Systat Software Inc., San Jose, CA, USA,).Some parameters did not show normality and homoscedasticity; therefore, we used nonparametric tests by Kyplot 3.0 (Kyens Lab.Inc., Tokyo, Japan).The data of soil chemical properties were compared between 2009 and 2010 using Mann-Whitney U test.Dry mass of the whole plant and each organ, ECM fungal colonization, Psat and gs, and concentrations of elements in leaves and roots were examined using a Steel-Dwass test.The mean values of total dry mass and ECM fungal colonization for Q. glauca, Q. salicina, Q. myrsinaefolia, and C. cuspidata were compared among the first and second samplings of AH, SC, and NE seedlings.The values of dry mass of each organ, Psat, gs and concentrations of elements in leaves and roots of the four Fagaceae species were compared among AH, SC and NE seedlings.Different letters are used to indicate statistically significant differences at p < 0.05.The mean values followed by the same letter showed no significant differences.
To examine the relationship between total dry mass and ECM fungal colonization, we conducted a Spearman's rank correlation.Additionally, we examined the relationship between Psat and nutrients  [48] and Scagel [49]).The curve indicated by the broken line shows the border to estimate increase or decrease of content.A plot located above the curve shows increase of content.

Statistical Analysis
All parameters were analysed for normality and homoscedasticity using the Shapiro-Wilk test with Sigma Plot 12.3 (Systat Software Inc., San Jose, CA, USA,).Some parameters did not show normality and homoscedasticity; therefore, we used nonparametric tests by Kyplot 3.0 (Kyens Lab.Inc., Tokyo, Japan).The data of soil chemical properties were compared between 2009 and 2010 using Mann-Whitney U test.Dry mass of the whole plant and each organ, ECM fungal colonization, P sat and g s , and concentrations of elements in leaves and roots were examined using a Steel-Dwass test.The mean values of total dry mass and ECM fungal colonization for Q. glauca, Q. salicina, Q. myrsinaefolia, and C. cuspidata were compared among the first and second samplings of AH, SC, and NE seedlings.The values of dry mass of each organ, P sat , g s and concentrations of elements in leaves and roots of the four Fagaceae species were compared among AH, SC and NE seedlings.Different letters are used to indicate statistically significant differences at p < 0.05.The mean values followed by the same letter showed no significant differences.
To examine the relationship between total dry mass and ECM fungal colonization, we conducted a Spearman's rank correlation.Additionally, we examined the relationship between P sat and nutrients in leaves by the same method.For the data from the second sampling, we also compared each parameter of the seedlings from each of the three phytotron rooms by a Kruskal-Wallis test.There were no significant differences among the three rooms.

Soil Chemical Properties
Table 1 lists the chemical properties of the calcareous soil.The pH of the calcareous soil was almost neutral (7.01).The Ca concentration was high.Comparing the two experimental years for calcareous soil, there was no significant difference.  Extracted in 1 M ammonium acetate; 3 Extracted in 1 M potassium chloride.

Growth Characteristics
At the first sampling stage before the experiment, the total dry mass of three Quercus species showed little difference among AH, SC, and NE seedlings (Figure 2).For C. cuspidata; the total dry mass of AH and SC seedlings showed significant increases compared with NE seedlings.
Forests 2016, 7, 266 7 of 18 in leaves by the same method.For the data from the second sampling, we also compared each parameter of the seedlings from each of the three phytotron rooms by a Kruskal-Wallis test.There were no significant differences among the three rooms.

Soil Chemical Properties
Table 1 lists the chemical properties of the calcareous soil.The pH of the calcareous soil was almost neutral (7.01).The Ca concentration was high.Comparing the two experimental years for calcareous soil, there was no significant difference.  Extracted in 1 M ammonium acetate; 3 Extracted in 1 M potassium chloride.

Growth Characteristics
At the first sampling stage before the experiment, the total dry mass of three Quercus species showed little difference among AH, SC, and NE seedlings (Figure 2).For C. cuspidata; the total dry mass of AH and SC seedlings showed significant increases compared with NE seedlings.At the second sampling, the total dry mass was significantly increased in AH and SC seedlings of Q. glauca, Q. salicina, and Q. myrsinaefolia.For NE seedlings of Q. myrsinaefolia, the total dry mass increased significantly.In contrast, other NE seedlings did not show a significant increase in the total  At the second sampling, the total dry mass was significantly increased in AH and SC seedlings of Q. glauca, Q. salicina, and Q. myrsinaefolia.For NE seedlings of Q. myrsinaefolia, the total dry mass increased significantly.In contrast, other NE seedlings did not show a significant increase in the total dry mass in Q. glauca, Q. salicina, and C. cuspidata.AH seedlings of C. cuspidata did not show significant increases in the total dry mass.
Comparing AH, SC, and NE seedlings at the second sampling, the total dry mass and dry mass of each organ were significantly higher in AH seedlings of Q. glauca, Q. salicina, and Q. myrsinaefolia than those in NE seedlings (Figures 2 and 3).Additionally, SC seedlings had significantly higher dry mass for each organ compared with NE seedlings of Q. glauca, Q. salicina, and C. cuspidata.For the SC seedlings of Q. myrsinaefolia and AH seedlings of C. cuspidata, the dry mass of each organ was not higher than that of NE seedlings.Comparing the AH seedlings of the four Fagaceae species, dry mass was highest for Q. salicina and lowest for C. cuspidata.Comparing AH, SC, and NE seedlings at the second sampling, the total dry mass and dry mass of each organ were significantly higher in AH seedlings of Q. glauca, Q. salicina, and Q. myrsinaefolia than those in NE seedlings (Figures 2 and 3).Additionally, SC seedlings had significantly higher dry mass for each organ compared with NE seedlings of Q. glauca, Q. salicina, and C. cuspidata.For the SC seedlings of Q. myrsinaefolia and AH seedlings of C. cuspidata, the dry mass of each organ was not higher than that of NE seedlings.Comparing the AH seedlings of the four Fagaceae species, dry mass was highest for Q. salicina and lowest for C. cuspidata.SLA of the four Fagaceae species was 85.2 cm 2 •g −1 for Q. glauca, 95.6 cm 2 •g −1 for Q. salicina, 87.7 cm 2 •g −1 for Q. myrsinaefolia, and 162.2 cm 2 •g −1 for C. cuspidata.There were no significant difference among AH, SC, and NE seedlings for each species.

The Percentage of ECM Colonization
At the first sampling, the percentages of ECM colonization in roots of AH seedlings were 69% for Q. glauca and C. cuspidata, 83% for Q. salicina, and 55% for Q. myrsinaefolia (Figure 4).In SC seedlings, these values were significantly lower than those of AH seedlings except for Q. myrsinaefolia.The percentage of ECM colonization of SC seedlings was around 40% for Q. glauca, Q. myrsinaefolia, and C. cuspidate, whereas this value was 54% for Q. salicina.SLA of the four Fagaceae species was 85.2 cm 2 •g −1 for Q. glauca, 95.6 cm 2 •g −1 for Q. salicina, 87.7 cm 2 • g −1 for Q. myrsinaefolia, and 162.2 cm 2 •g −1 for C. cuspidata.There were no significant difference among AH, SC, and NE seedlings for each species.

The Percentage of ECM Colonization
At the first sampling, the percentages of ECM colonization in roots of AH seedlings were 69% for Q. glauca and C. cuspidata, 83% for Q. salicina, and 55% for Q. myrsinaefolia (Figure 4).In SC seedlings, these values were significantly lower than those of AH seedlings except for Q. myrsinaefolia.The percentage of ECM colonization of SC seedlings was around 40% for Q. glauca, Q. myrsinaefolia, and C. cuspidate, whereas this value was 54% for Q. salicina.
significantly to 55% for C. cuspidata.In half of the AH seedlings of C. cuspidata, the percentage of ECM colonization decreased to <1%.For Q. myrsinaefolia, the percentage of ECM colonization of AH and SC seedlings did not change after the experiment.
In NE seedlings of the four Fagaceae species, the percentage of ECM colonization was <1% at the two sampling points.In addition, these values of NE seedlings were significantly lower than those of AH and SC seedlings.
In terms of the relationship between total dry mass and the percentage of ECM colonization, we found a significant positive correlation for each Fagaceae species at the second sampling (Table 2, p < 0.001).Moreover, Q. glauca and C. cuspidata showed a significant positive difference at the first sampling (p < 0.05).The upper row shows the relationships between total dry mass and ectomycorrhizal (ECM) fungal colonization.The lower row shows the relationship between photosynthetic rate at light saturation (Psat) and area-based concentrations of N and P in leaves at the second sampling.* shows a significant correlation at p < 0.05.

Photosynthetic Capacity
The Psat of Q. glauca and Q. salicina was significantly higher in AH seedlings than in SC and NE seedlings (Figure 5).In Q. myrsinaefolia, the Psat in AH and SC seedlings was significantly higher than that in NE seedlings.The Psat of C. cuspidata was significantly higher in SC seedlings than in NE seedlings.Comparing the four Fagaceae species, Psat was highest in AH seedlings of Q. glauca and lowest in NE seedlings of C. cuspidata.
A similar trend was observed for gs as for Psat, and gs of Q. glauca and Q. salicina was significantly higher in AH seedlings than in SC and NE seedlings.gs in AH and SC seedlings of Q. myrsinaefolia  After the second sampling, the percentages of ECM colonization of AH seedlings decreased significantly for Q. glauca, Q. salicina, and C. cuspidata.In addition, the percentages of ECM colonization decreased significantly for SC seedlings of Q. glauca and Q. salicina.The percentages of ECM colonization of the three Quercus species were significantly higher for AH seedlings than those of SC seedlings.In contrast, the percentage of ECM colonization of SC seedlings increased significantly to 55% for C. cuspidata.In half of the AH seedlings of C. cuspidata, the percentage of ECM colonization decreased to <1%.For Q. myrsinaefolia, the percentage of ECM colonization of AH and SC seedlings did not change after the experiment.
In NE seedlings of the four Fagaceae species, the percentage of ECM colonization was <1% at the two sampling points.In addition, these values of NE seedlings were significantly lower than those of AH and SC seedlings.
In terms of the relationship between total dry mass and the percentage of ECM colonization, we found a significant positive correlation for each Fagaceae species at the second sampling (Table 2, p < 0.001).Moreover, Q. glauca and C. cuspidata showed a significant positive difference at the first sampling (p < 0.05).

Photosynthetic Capacity
The P sat of Q. glauca and Q. salicina was significantly higher in AH seedlings than in SC and NE seedlings (Figure 5).In Q. myrsinaefolia, the P sat in AH and SC seedlings was significantly higher than that in NE seedlings.The P sat of C. cuspidata was significantly higher in SC seedlings than in NE seedlings.Comparing the four Fagaceae species, P sat was highest in AH seedlings of Q. glauca and lowest in NE seedlings of C. cuspidata.
Forests 2016, 7, 266 10 of 18 was significantly higher than that in NE seedlings.In C. cuspidata, the gs in SC seedlings was higher than that in AH and NE seedlings.In particular, the gs in AH seedlings of Q. glauca was higher compared with other seedlings.

Element Concentrations in Leaves
The concentrations of elements in leaves at the second sampling are shown in Table 3.The concentration of P was significantly higher in AH seedlings of Q. salicina and Q. myrsinaefolia than in SC seedlings.In particular, AH seedlings of Q. salicina showed the highest P concentration between the four Fagaceae species.In C. cuspidata, the concentration of P was highest for SC seedlings.In contrast, other elements did not show high concentrations in AH seedlings despite their greater growth.The concentration of N in AH seedlings was significantly lower for Q. salicina and C. cuspidata than in SC and NE seedlings.Similarly, the concentration of K was significantly lower in AH seedlings of Q. glauca and Q. myrsinaefolia compared with in SC and NE seedlings.Moreover, NE seedlings had the highest Mg concentration in Q. salicina and C. cuspidata.
For the relationship between the concentrations of N or P and Psat, we found a significant positive correlation between the concentrations of P and Psat of Q. salicina, Q. myrsinaefolia, and C. cuspidata (Table 2, p < 0.05).In contrast, there were significant negative correlations between concentration of N and Psat of Q. salicina and Q. myrsinaefolia (p < 0.05).A similar trend was observed for g s as for P sat , and g s of Q. glauca and Q. salicina was significantly higher in AH seedlings than in SC and NE seedlings.g s in AH and SC seedlings of Q. myrsinaefolia was significantly higher than that in NE seedlings.In C. cuspidata, the g s in SC seedlings was higher than that in AH and NE seedlings.In particular, the g s in AH seedlings of Q. glauca was higher compared with other seedlings.

Element Concentrations in Leaves
The concentrations of elements in leaves at the second sampling are shown in Table 3.The concentration of P was significantly higher in AH seedlings of Q. salicina and Q. myrsinaefolia than in SC seedlings.In particular, AH seedlings of Q. salicina showed the highest P concentration between the four Fagaceae species.In C. cuspidata, the concentration of P was highest for SC seedlings.In contrast, other elements did not show high concentrations in AH seedlings despite their greater growth.The concentration of N in AH seedlings was significantly lower for Q. salicina and C. cuspidata than in SC and NE seedlings.Similarly, the concentration of K was significantly lower in AH seedlings of Q. glauca and Q. myrsinaefolia compared with in SC and NE seedlings.Moreover, NE seedlings had the highest Mg concentration in Q. salicina and C. cuspidata.
For the relationship between the concentrations of N or P and P sat , we found a significant positive correlation between the concentrations of P and P sat of Q. salicina, Q. myrsinaefolia, and C. cuspidata (Table 2, p < 0.05).In contrast, there were significant negative correlations between concentration of N and P sat of Q. salicina and Q. myrsinaefolia (p < 0.05).

Element Concentrations and Contents in Roots
The concentrations of elements in roots at the second sampling are shown in Table 4.The concentrations of P in Q. glauca, Q. salicina, and Q. myrsinaefolia were significantly higher in AH seedlings than in NE seedlings.AH seedlings also exhibited high concentrations of Fe and Mn in Q. myrsinaefolia than in SC seedlings.In C. cuspidata, the concentrations of N and P were highest in SC seedlings.The concentration of Ca was significantly lower in SC seedlings of Q. myrsinaefolia and C. cuspidata.In addition, the concentration of Fe in SC seedlings of Q. salicina and Q. myrsinaefolia was significantly lower.In NE seedlings of Q. myrsinaefolia, the concentration of Mg was significantly higher than in AH and SC seedlings.Across the four Fagaceae species, the concentration of P was the highest for AH seedlings of Q. salicina.

Vector Analysis between Relative Dry Mass and Nutrient Concentration
The relationships between relative leaf dry mass and relative concentrations in leaves are shown in Figure 6.For the relationship of Ca, uptake of each Fagaceae species showed enhanced or luxury consumption.In contrast, Mn showed dilution or efficient use by each Fagaceae species.For N, most plots showed dilution or sufficiency, whereas those for SC seedlings of Q. salicina and SC and NE seedlings of C. cuspidate showed enhanced uptake.In the case of NE seedlings of Q. salicina, the plot showed toxicity.
For P and Fe in the leaves, most plots of each Fagaceae species showed dilution or efficient use for leaves and roots.For P in leaves of NE seedlings of Q. glauca and Q. salicina, and Fe in leaves of NE seedlings of Q. salicina, the plots showed nutrient deficiency.The trend of K in leaves showed various patterns.Plots of AH seedling of Q. glauca and Q. salicina showed sufficiency, but the plots of C. cuspidata and AH and SC seedlings of Q. myrsinaefolia showed dilution.Other plots showed enhanced or luxury consumption.For the relationship of Mg, the plots showed sufficiency or dilution for each Fagaceae species.In the case of NE seedlings of Q. salicina, the plot showed toxicity.
The relationships between relative leaf dry mass and relative concentrations in leaves are shown in Figure 6.For the relationship of Ca, uptake of each Fagaceae species showed enhanced or luxury consumption.In contrast, Mn showed dilution or efficient use by each Fagaceae species.For N, most plots showed dilution or sufficiency, whereas those for SC seedlings of Q. salicina and SC and NE seedlings of C. cuspidate showed enhanced uptake.In the case of NE seedlings of Q. salicina, the plot showed toxicity.For P and Fe in the leaves, most plots of each Fagaceae species showed dilution or efficient use for leaves and roots.For P in leaves of NE seedlings of Q. glauca and Q. salicina, and Fe in leaves of NE seedlings of Q. salicina, the plots showed nutrient deficiency.The trend of K in leaves showed various patterns.Plots of AH seedling of Q. glauca and Q. salicina showed sufficiency, but the plots of C. cuspidata and AH and SC seedlings of Q. myrsinaefolia showed dilution.Other plots showed

Growth and Photosynthesis
After 6 months of cultivation in calcareous soil, we confirmed that symbiosis with ECM fungi was essential for growth acceleration in calcareous soil (Figure 2).Symbiosis with A. hygrometricus contributed to growth acceleration in the three Quercus species compared with that of S. citrinum.In C. cuspidata, only S. citrinum inoculation showed high values of dry mass of each organ (Figure 3).In general, compatibility with ECM fungi and effects on growth acceleration are different among woody species [57,58].Unlike Quercus species, C. cuspidata may have better compatibility with S. citrinum.We have already confirmed that C. cuspidata inoculated with S. citrinum showed accelerated growth [40].
Moreover, AH seedlings of Q. glauca and Q. salicina showed the highest values of P sat and g s (Figure 5).Woody species inoculated with ECM fungi exhibit accelerated water uptake from external hyphae [59].AH seedlings probably had accelerated water uptake, and stomatal closure did not occur.Accordingly, the P sat and g s were high in AH seedlings.Similar trends were shown for AH and SC seedlings of Q. myrsinaefolia, which had high values for the percentage of ECM colonization (Figures 4 and 5).
Comparing AH seedlings, Q. salicina showed remarkable growth acceleration (Figure 2).Q. salicina have high acclimation capacity in calcareous soil, and symbiosis with A. hygrometricus can take place in calcareous soil.Q. salicina grows at high altitudes in the karst region in southwestern Japan [17,18,21].Our results reflected the distribution of Q. salicina.
Compared with Q. salicina and Q. glauca, growth acceleration by inoculation with ECM fungi was not obvious for Q. myrsinaefolia (Figure 2).Acclimation capacity in calcareous soil was lower for Q. myrsinaefolia than for Q. salicina and Q. glauca.In contrast, the total dry mass for C. cuspidata was lowest even in SC seedlings (Figure 2).The leaves of C. cuspidata yellowed after transplantation, indicating the occurrence of lime-induced chlorosis [3].In addition, lime-induced chlorosis can readily appear in calcifuge plants [7].Based on growth and lime-induced chlorosis, C. cuspidata was considered as a calcifuge plant.Conversely, C. cuspidata inoculated with ECM fungi accelerated its growth in acidic soil [40], indicating that this species has a preference for an acidic environment.

Nutrient Relationships
For the nutrients in calcareous soil, the concentration of exchangeable Ca was 2.6 g•kg −1 (Table 1).The values of soil pH and Ca concentration differ among calcareous soils in Japan [12,13,[60][61][62][63], with ranges of 5.8 to 8.3 for pH and 0.8 to 12.1 g•kg −1 for Ca.Compared with other regions, the soil used for our experiment did not have high soil pH, and did not exhibit an extreme excess of Ca.In contrast, the concentration of N in calcareous soil used in our experiment was lower than those in other regions [60][61][62][63].However, the concentration of N in leaves of the four Fagaceae species was not in the deficient range compared with other woody species grown in calcareous soil [13,63].For other nutrients, the concentrations of K and Mg were higher than typical forest soil in this region [64].There was no previous information available on P in calcareous soil in Japan from analyses by the Olsen method.Compared with the data from other countries, the concentration of P was higher in calcareous soil used in our experiment [65,66].Thus, the concentrations of N, P, K, and Mg in calcareous soil in our experiment were not insufficient.
AH and SC seedlings of the four Fagaceae species did not show high N concentrations in leaves (Table 3).Thus, uptake of N in leaves is not related to ECM colonization.Comparing traits among Quercus species, the concentration of N in roots was higher for Q. salicina and Q. glauca than that for Q. myrsinaefolia (Table 4).Q. salicina and Q. glauca had higher adaptability to various poor habitats [19,67].Q. glauca and Q. salicina have high capacity to acquire N from calcareous soil.In contrast, the habitats of natural forests of Q. myrsinaefolia include deep, well-drained soil [68].In the karst region, Q. myrsinaefolia has been found on a site with deep calcareous soil [19].Differences in acclimation capacity for Q. myrsinaefolia may originate from the inherent nutrient requirements.There was a similar case for Picea abies, which has low adaptability to calcareous soil [69], and suppressed uptake of N in roots increased in calcareous soil [70].
On the uptake of P in roots, NE seedlings showed low P concentrations compared with AH seedlings of Q. glauca, Q. salicina and Q. myrsinaefolia (Table 4).Thus, uptake of P is linked with symbiosis with ECM fungi.On calcareous soil, symbiosis with ECM fungi is important for the uptake of P [27,28].Symbiosis between Quercus species and A. hygrometricus produced a high capacity for the acquisition of P.
Meanwhile, we confirmed that the uptake of P, Fe and Mn in leaves was suppressed in each Fagaceae species (Table 3, Figure 6).Uptake of micronutrients (e.g., Fe, Mn, Cu, and Zn) and P is suppressed by low availability at high pH in calcareous soil [3][4][5]70].Our results suggested that uptake of Mn is suppressed for the Fagaceae species in calcareous soil.For P and Fe in leaves, AH seedlings of three Quercus species exhibited increased contents; however, concentration was decreased and showed evidence of dilution (Figure 6).This trend suggests that the uptake of P and Fe in leaves is suppressed even after inoculation with A. hygrometricus.In particular, NE seedlings of Q. glauca and Q. salicina showed a P deficiency in leaves (Figure 6).We considered that NE seedlings of Q. glauca and Q. salicina suffered from serious nutrient suppression due to lower availability.C. cuspidata showed an obvious decrease of the concentration of Fe in leaves, and showed efficient use of little Fe (Figure 6).When calcifuge woody plants are inoculated with ECM fungi, the effects of the fungi are probably quite limited in calcareous soil.
In contrast, Ca in leaves and roots showed a high value (Tables 3 and 4) compared with seedlings of Fagaceae species grown on acidic colluvial soil [40].Moreover, many plots showed enhancement of uptake or luxury consumption (Figure 6).However, the actual concentration of Ca in plant organs was almost the same among AH, SC, and NE seedlings for each Fagaceae species (Tables 3 and 4).Thus, we considered that the differences in Ca were not the main factor for growth differences among AH, SC, and NE seedlings.

Relationships among Variables
In examining the factors affecting differences in growth between AH, SC and NE seedlings, ECM colonization was considered one of the important factors.The percentages of ECM colonization were higher in AH seedlings of the three Quercus species and in SC seedlings of C. cuspidata (Figure 4).Additionally, there were positive relationships between total dry mass and ECM colonization (Table 2).Thus, different percentages of ECM colonization were probably related to differences in growth.
With respect to the relationships between growth and nutrient uptake, N and P in roots showed enhancement of uptake in AH seedlings of Q. salicina (Table 4).The storage of N and P in belowground organs has a significant role in plant growth, and plants with high N and P in roots show high growth rates [71][72][73].Based on these results, it appeared that the amount of stored N and P resulted in the high growth of AH seedlings of Q. salicina.
The photosynthetic rate showed positive correlations between P sat and the concentrations of P in leaves of Q. salicina, Q. myrsinaefolia, and C. cuspidata (Table 2).In general, the photosynthetic rate is closely related to P concentration [55,56].Thus, the concentration of P in leaves is considered as an important factor that affects the photosynthetic rate.In contrast, Q. glauca did not show a positive correlation between P sat and the concentration of P (Table 2).Other factors that affect differences in P sat include high value of g s (Figure 5).AH seedlings of Q. glauca could absorb water through developed hyphae (Figure 4); as a result, the stomata stay open and absorption of CO 2 through the stomata is also increased.
The concentration of N in leaves showed a negative correlation with P sat for Q. salicina and Q. myrsinaefolia (Table 2), and our results contradicted the general trend [7,54].The cause of this negative trend was probably related to the high concentration of N in leaves and the low values of P sat in NE seedlings.In the case of NE seedlings, they had no hyphae of ECM fungi and a lower capacity for water uptake.The stomata of NE seedlings may have closed in order to avoid water stress; as a result, NE seedlings showed lower P sat and g s and decreased absorption of CO 2 through stomata.Similar trends were shown for NE seedlings of Fagaceae species planted on acidic soil [40].
With respect to the relationship among nutrients for C. cuspidata and AH seedlings of Q. myrsinaefolia, the concentrations of K were decreased in leaves (Table 3) and were diluted (Figure 6).High Ca content in soil causes ion-antagonism with K [70].As a result, the suppression of was confirmed for several woody species (Picea abies and several Lauraceae species), in which growth was suppressed on calcareous soil [13,70].Compared with AH seedlings of Q. glauca and Q. salicina, growth levels of C. cuspidata and AH seedlings of Q. myrsinaefolia were inferior.Woody species that show less adaptability to calcareous soil can probably suppress uptake of K.

Conclusions
When seedlings of the four Fagaceae species were planted on calcareous soil, the uptake of various nutrients, such as P, Fe, and Mn in leaves was suppressed.However, Fagaceae seedlings inoculated with ECM fungi showed accelerated growth.In particular, Fagaceae seedlings inoculated with ECM fungi had increased concentration of P in roots, and this trait was important for growth acceleration.Additionally, the concentration of P in leaves was an important factor and affected the differences in photosynthetic rate.The most suitable ECM fungi differed among the four Fagaceae species, and higher growth levels were shown from inoculation with A. hygrometricus for the three Quercus species, and with S. citrinum for C. cuspidata.Growth levels of the seedlings inoculated with ECM fungi were different among the four Fagaceae species, and C. cuspidata, which showed the lowest growth, was considered as a calcifuge plant.From their growth characteristics, we concluded that Q. salicina and Q. glauca inoculated with A. hygrometricus showed the highest growth levels, and these species are most suitable for growth on calcareous soil.
These findings are expected to contribute to new methods for reforestation of limestone quarries in warm-temperature zones in East Asia.In particular, Q. salicina and Q. glauca inoculated with A. hygrometricus may be useful species for reforestation.The effects on growth acceleration without fertilization are advantageous for reforestation in limestone environments.It should be noted that our experiment used sterilized calcareous soil; therefore, further field cultivation experiments with ECM-fungi inoculated plants will be needed.

Figure 1 .
Figure 1.Graphical representation of interpretation comparing changes in dry mass and concentrations of nutrients grown in different media types.The content isograms represent combinations of dry mass and concentration required for constant content per unit dry mass.The arrows indicate the direction in which each interpretation holds (adapted from Valentine and Allen[47] and Scagel[48]).The curve indicated by the broken line shows the border to estimate increase or decrease of content.A plot located above the curve shows increase of content.

Figure 1 .
Figure 1.Graphical representation of interpretation comparing changes in dry mass and concentrations of nutrients grown in different media types.The content isograms represent combinations of dry mass and concentration required for constant content per unit dry mass.The arrows indicate the direction in which each interpretation holds (adapted from Valentine and Allen[48] and Scagel[49]).The curve indicated by the broken line shows the border to estimate increase or decrease of content.A plot located above the curve shows increase of content.

Figure 2 .
Figure 2. Total dry mass for seedlings of four Fagaceae species grown in a calcareous soil (g•plant −1 , mean ± SD, n = 12).Different letters indicate significant effects according to a Steel-Dwass test (p < 0.05).Mean values followed by the same letter showed no significant difference (p > 0.05).AH, seedlings inoculated with Astraeus hygrometricus; SC, seedlings inoculated with Scleroderma citrinum; NE, non-ectomycorrhizal seedlings.

Figure 2 .
Figure 2. Total dry mass for seedlings of four Fagaceae species grown in a calcareous soil (g•plant −1 , mean ± SD, n = 12).Different letters indicate significant effects according to a Steel-Dwass test (p < 0.05).Mean values followed by the same letter showed no significant difference (p > 0.05).AH, seedlings inoculated with Astraeus hygrometricus; SC, seedlings inoculated with Scleroderma citrinum; NE, non-ectomycorrhizal seedlings.

Forests
2016, 7, 266 8 of 18 dry mass in Q. glauca, Q. salicina, and C. cuspidata.AH seedlings of C. cuspidata did not show significant increases in the total dry mass.

Figure 3 .
Figure 3. Dry mass of each organ (leaf, stem and branch, and root) for seedlings of four Fagaceae species grown in a calcareous soil at the second sampling (g•plant −1 , mean ± SD, n = 12).Different letters indicate significant effects according to a Steel-Dwass test (p < 0.05).Mean values followed by the same letter showed no significant difference (p > 0.05).

Figure 3 .
Figure 3. Dry mass of each organ (leaf, stem and branch, and root) for seedlings of four Fagaceae species grown in a calcareous soil at the second sampling (g•plant −1 , mean ± SD, n = 12).Different letters indicate significant effects according to a Steel-Dwass test (p < 0.05).Mean values followed by the same letter showed no significant difference (p > 0.05).

Figure 4 .
Figure 4. Percentage of ectomycorrhizal fungal colonization for seedlings of four Fagaceae species grown in a calcareous soil (%, mean ± SD, n = 12).Different letters indicate significant effects according to a Steel-Dwass test (p < 0.05).Mean values followed by the same letter showed no significant difference (p > 0.05).

Figure 4 .
Figure 4. Percentage of ectomycorrhizal fungal colonization for seedlings of four Fagaceae species grown in a calcareous soil (%, mean ± SD, n = 12).Different letters indicate significant effects according to a Steel-Dwass test (p < 0.05).Mean values followed by the same letter showed no significant difference (p > 0.05).

Figure 5 .
Figure 5. Photosynthetic rate at light saturation (Psat, μmol•m −2 •s −1 ) and stomatal conductance (gs, mol•m −2 •s −1 ) for seedlings of four Fagaceae species grown in a calcareous soil (mean ± SD, n = 12).Different letters indicate significant effects according to a Steel-Dwass test (p < 0.05).Mean values followed by the same letter showed no significant difference (p > 0.05).Psat and gs for Q. glauca and Q. salicina were measured in October 2009, and for Q. myrsinaefolia and C. cuspidata in October 2010.

Figure 5 .
Figure 5. Photosynthetic rate at light saturation (P sat , µmol•m −2 •s −1 ) and stomatal conductance (g s , mol•m −2 •s −1 ) for seedlings of four Fagaceae species grown in a calcareous soil (mean ± SD, n = 12).Different letters indicate significant effects according to a Steel-Dwass test (p < 0.05).Mean values followed by the same letter showed no significant difference (p > 0.05).P sat and g s for Q. glauca and Q. salicina were measured in October 2009, and for Q. myrsinaefolia and C. cuspidata in October 2010.

Figure 6 .
Figure 6.Comparison of relative nutrient concentration in leaves and leaf dry masses for seedlings of four Fagaceae species grown in a calcareous soil.The content isograms represent the combination of dry mass and concentration required for these values at the second sampling relative to the first sampling.A value of 100 represents the dry mass and concentration at the first sampling.

Figure 6 .
Figure 6.Comparison of relative nutrient concentration in leaves and leaf dry masses for seedlings of four Fagaceae species grown in a calcareous soil.The content isograms represent the combination of dry mass and concentration required for these values at the second sampling relative to the first sampling.A value of 100 represents the dry mass and concentration at the first sampling.

Table 1 .
Chemical properties of calcareous soil used in this experiment (mean ± SD, n = 5).

Table 1 .
Chemical properties of calcareous soil used in this experiment (mean ± SD, n = 5).
Mean values of each parameter were analyzed by Mann-Whitney U test (no significance; p > 0.05).1Extractedby the Olsen method;

Table 2 .
Spearman's correlation coefficients (R) for seedlings of four Fagaceae species grown in a calcareous soil (n = 12).
The upper row shows the relationships between total dry mass and ectomycorrhizal (ECM) fungal colonization.The lower row shows the relationship between photosynthetic rate at light saturation (P sat ) and area-based concentrations of N and P in leaves at the second sampling.* shows a significant correlation at p < 0.05.

Table 3 .
Concentrations of elements (N, P, K, Ca, Mg, Fe, and Mn) in leaves of seedlings of four Fagaceae species grown in a calcareous soil sampled at the second sampling (mean ± SD, n = 12).

Table 3 .
Concentrations of elements (N, P, K, Ca, Mg, Fe, and Mn) in leaves of seedlings of four Fagaceae species grown in a calcareous soil sampled at the second sampling (mean ± SD, n = 12).
Different letters indicate significant effects according to a Steel-Dwass test (p < 0.05).Mean values followed by the same letter showed no significant difference (p > 0.05).AH, seedlings inoculated with Astraeus hygrometricus; SC, seedlings inoculated with Scleroderma citrinum; NE, non-ectomycorrhizal seedlings.

Table 4 .
Concentrations of elements (N, P, K, Ca, Mg, Fe, and Mn) in roots of seedlings of four Fagaceae species grown in a calcareous soil sampled at the second sampling (mean ± SD, n = 12).
Different letters indicate significant effects according to a Steel-Dwass test (p < 0.05).Mean values followed by the same letter showed no significant difference (p > 0.05).