Plant Species-Dependent Effects of Liming and Plant Residue Incorporation on Soil Bacterial Community and Activity in an Acidic Orchard Soil

Both liming and plant residue incorporation are widely used practices for the amelioration of acidic soils—however, the difference in their effects is still not fully understood, especially regarding the microbial community. In this study, we took the acidic soils from a subtropical orchard as target soils, and implemented liming and plant residue incorporation with a leguminous and a gramineous cover crop as test plants. After six months of growth, soil pH, total organic carbon (TOC), dissolved organic carbon (DOC) and nutrient contents were determined, soil enzymes involving C, N, P cycling were assayed, and microbial communities were also analyzed using Polymerase Chain Reaction-Denaturing Gradient Gel Electrophoresis (PCR-DGGE). Results showed that liming was more effective in elevating soil pH, while plant residue incorporation exerted a more comprehensive influence—not only on soil pH, but also on soil enzyme activity and microbial community. PCR-DGGE analysis revealed that liming changed the microbial community structure more greatly than plant residue incorporation, while plant residue incorporation altered the microbial community composition much more than liming. The growth responses of test plants to liming and plant residue incorporation depended on plant species, indicating the necessity to select appropriate practice for a particular crop. A further, detailed investigation into the microbial community composition, and the respective functions using metagenomic approach, is also suggested.


Introduction
Acidic soils occupy around 40% of the total agricultural lands worldwide, representing one of the most important limiting factors of agriculture production [1]. In China, acidic soils mainly distribute in the subtropical regions, where hilly areas are predominant with orchards, tea gardens, forest lands as the main land-use types. Traditionally, the farmers use liming to ameliorate the soil acidity [2]. It is reported that liming increased the soil pH in a citrus orchard by 1.0~1.4 units, compared to the adjacent native forests [3]. In the tea gardens of 8-, 50-, 90-year old, liming with CaCO 3 at the rate of 6.4 g·kg −1 increased the soil pH by 3.13, 3.68, 2.24 units, respectively [4]. On the other hand, these subtropical soils are typical of low organic matter content, due to fast mineralization under high temperature and humidity conditions. Consequently, plant residue incorporation is proposed and practiced as an alternative to simultaneously increase the soil pH and the soil organic matter content. Xu et al. [5] reported the increase in soil pH by up to 3.3 units after 42 days of incubation with different types of plant residues. Hue [6] demonstrated that cowpea residue incorporation not only increased soil pH from 4.45 to 5.45, but also promoted the soluble carbon from 0.11 to 2.84 mM in the Ultisol.
It is clear that liming and plant residue incorporation do not share the common mechanisms in ameliorating acidic soils. The primary effects of liming are raising pH, base saturation and Ca content, and its secondary effects include reducing Al activity and P immobilization, modifying microbial activity, and etc. (see review by Fageria and Baligar [7]). Plant residue incorporation increases soil pH as well [5,6], although the amelioration process is time-consuming compared with liming. According to Sakala et al. [8], decomposition of plant residues and the subsequent release of base and mineralized N by soil microbes is the requisite of ameliorating acidic soils by plant residue incorporation. Moreover, a large number of studies demonstrate that plant residue incorporation can directly elevate the organic matter content [9,10] and indirectly promote the microbial activity via increased soil organic carbon [11,12] and enzyme activity [13]. It seems that, apart from neutralizing soil acidity, plant residue incorporation exert more extensive influences on soil properties (e.g., microbial activity) than liming. In this context, it is reasonable to speculate that the outcomes of ameliorating soil with liming or plant residue incorporation can be much different, especially in terms of microbial properties. However, information on this aspect is scarce thus far.
Soil microbe is the essential factor driving most soil biological processes [14], including the decomposition of soil organic matter [15], nutrient (C, N, P, S) cycling [16], soil respiration [17,18], and etc. In turn, soil microbes are strongly affected by environmental cues, with pH, soil organic carbon as the important factors [19][20][21]. This indicates that the increase of soil pH by liming or the increase of soil organic carbon by plant residue incorporation can further regulate the microbial community. Moreover, for plant residue incorporation, the additive effect of increased pH can make the regulation complicated, which differentiates the final effects of liming and plant residue incorporation on soil microbial community. Unfortunately, the difference between them has never been investigated up to date.
We hypothesized that (1) liming would be more effective in neutralizing soil acidity than plant residues, (2) plant residue incorporation would exert a greater influence on microbial community than liming, and (3) plant residues would affect the biological properties more extensively than liming. In this study, we compared the effects of liming and plant residue incorporation in alleviating the acidity of a subtropical orchard soil. Two plant species, one legume species Stylosanthes guianensis and one graminoid species Paspalum natatum, were taken as test plants for comparison, because they are routinely grown in orchards as cover crops in the subtropical areas of China [22]. The soil chemical properties and microbial properties were monitored to evaluate the respective effects of liming and plant residue incorporation.

Soils and Test Plants
Soil was collected from a subtropical orchard at Heshan Hilly Land Interdisciplinary Experimental Station (E112 • 54 , N22 • 41 ), Chinese Academy of Science in Guangdong province [22]. According to USDA soil classification, it is the Ultisol, with the chemical properties as follows (g·kg −1 ): pH 4.98, total organic carbon (TOC) 15.58, dissolved organic carbon (DOC) 0.33, total N 1.64, total P 0.52, total K 10.08, available N 0.137, available P 0.067, available K 0.055. The soils were air-dried and sieved through a mesh of 2 mm pore size.
To evaluate the effects of ameliorating acidic soils by liming or plant residue incorporation, leguminous plant S. guianensis and gramineous plant P. natatum were taken as test plants. These two species are routinely grown in subtropical orchards as cover crops, and the cuttings (the aboveground biomass) are generally incorporated into orchard soil after regular mowing. The seeds of these two species used in this study were commercially obtained.

Experimental Setup
We conducted a pot culture with completely randomized factorial design in the greenhouse. Two factors were included, namely, plant species (two levels: S. guianensis and P. natatum) and soil treatments (three levels-control with no treatment, liming, and plant residue incorporation), and thus, six treatments were produced with each plant species comprising three soil treatments. For liming treatment (named as CaCO 3 ), CaCO 3 of chemical grade was applied to soils at the rate of 2.5 g·kg −1 , equivalent to 6.5 t·ha −1 . For plant residue incorporation treatment (named as OM), oven-dried, and homogenized plant shoot biomass was incorporated into the soil at the rate of 30 g·kg −1 , equivalent to 78 t·ha −1 . To simulate the status in the field conditions, the soil (grown with each of S. guianensis or P. natatum) was incorporated with the plant residues of respective species.
Each plastic pot (18 cm in diameter and 15 cm in height) was filled with 2.0 kg soil. Five replicates were set up for each treatment, thus, totally producing 30 pots with 15 pots for each plant species. Seeds of S. guianensis and P. natatum were surface sterilized with 10% NaClO 2 for 15 min, rinsed with sterilized water for five times, and then germinated at 28 • C until the primary roots of about 2 mm appeared. Approximately 50 germinated seeds were sown into each pot, and the seedlings were thinned to 20 plants of similar size per pot after two weeks of growth. Finally, 30 pots were prepared for two plant species and placed in a greenhouse with a temperature range of 23~30 • C and natural radiation. Pots were irrigated to maintain the stable water content of approximately 18% with the weighing method. To maximize the effects of ameliorating soil acidity, the soil was grown with S. guianensis or P. natatum for two growth cycles, with each lasting for three months. Briefly, after growth of three months at the first growth cycle, plants were carefully removed, and the root fragments in the soil were completely picked out, then, the germinated seeds were sown again for the second growth cycle of three months. Plant residues were incorporated into the soil only once at the beginning of the first growth cycle.
The experiment was finalized after six months of growth. Before the final harvest, the relative chlorophyll content in the leaves was measured using the portable chlorophyll meter SPAD-502 (Minolta Camera Co., Osaka, Japan). Then, plant shoots were separated from roots, and the biomass (fresh weight) of shoots and roots were recorded, respectively. The soil in each pot was homogenized and divided into three aliquots. One aliquot was stored at −80 • C for DNA extraction within two weeks, the second aliquot was stored at 4 • C for soil enzyme assay and DOC determination within three days, and the third aliquot was air-dried at room temperature for the soil chemical analysis.

Measurement of Soil Chemical Properties, TOC and DOC
Soil pH, the total and available nutrient contents (N, P, K) were determined as described previously [23]. Briefly, soil pH was measured in deionized H 2 O (1:2.5 w/v). The contents of total N (TN), total P (TP) and total K (TK) were analyzed using the Kjeldahl method, molybdenum blue colorimetric method and flame photometer, respectively. The contents of available N (AN) was determined after the release and transformation to NH 3 by 1.07 M NaOH and FeSO 4 powder at 40 • C for 24 h, followed by absorption with 2% (w/v) H 3 BO 3 and titration with 0.005 M H 2 SO 4 . Available P (AP) was extracted with the solution of Bray-1 (0.03 M NH 4 F-0.025 M HCl), and then colorimetrically measured. Available K (AK) was extracted with 1.0 M NH 4 OAc (pH = 7.0) and then determined by flame photometer.
TOC was analyzed by wet oxidation with K 2 CrO 4 [24]. The extraction of DOC was according to Mavi et al. [25] with pure water as extractant, and the determination of DOC was performed using a Vario Cube TOC analyzer (Elementar Inc., Hanau, Germany).

Soil Enzyme Assay
Soil enzymes involving C, N, P cycling were assayed, including α-glucosidase (EC 3 . The activities of α-glucosidase, β-glucosidase, β-xylosidase, cellobiosidase and chitinase was assayed using fluorogenic substrates in micro-well plates as described previously [26][27][28]. Briefly, soil suspensions containing 1 g fresh soil in 50 mL sodium acetate buffer (0.5 mol L −1 , pH 5.5) were ultrasonically homogenized, and then 160 µL aliquot of each soil suspension was dispensed into 96-well black microplates (Jet Bio-Filtration Co., Guangzhou, China). Substrate solution (final concentration of 500 µmol L −1 ) were added to each well with a final volume of 200 µL. Negative control and quench control were also applied. Standard curves were produced as well. Microplates were incubated for 3 h at 30 • C with continuous shaking. Fluorescence was measured using a microplate fluorometer (FLx800, BioTek Instruments, Winooski, VT, USA) at 355 nm excitation and 460 nm emission. The activities of urase, nitrate reductase, acid phosphatase and alkaline phosphatase were analyzed according to Hu et al. [29] and Cui et al. [22].
The final PCR products of bacterial community and AMF community were subjected to DGGE analysis using a D-Code Universal Mutation Detection System (Bio-Rad Laboratories, Hercules, CA, USA). For the PCR product of bacteria, DGGE were carried out on an 8% polyacrylamide gel for 10 h at a constant voltage of 80 V and 60 • C in a 45-65% horizontal denaturant gradient [31]. For the PCR product of AMF, DGGE were carried out on a 6% polyacrylamide gel for 15 h at a constant voltage of 70 V and 60 • C in a 30-60% horizontal denaturant gradient [23].

Data Analysis and Statistics
All data were the average of five replicates. Tukey's post hoc test and two-way analysis of variance (ANOVA) were performed using SPSS v21.0 (IBM SPSS In., Chicago, IL, USA). DGGE profiles were quantified with Quantity One ® , and the clustering analysis was also performed with it. To evaluate the effects of pH and DOC on soil chemical properties and enzyme activities, canonical correspondence analysis (CCA) was performed using Canoco for Window 4.5 [36].

The Effects of Liming and Plant Residue Incorporation on Plant Growth
Both CaCO 3 and OM affected plant growth of S. guianensis and P. natatum (Figure 1). Data indicated that OM increased, while CaCO 3 decreased the SPAD value of S. guianensis leaves. In contrast, both OM and CaCO 3 increased the SPAD value of P. natatum leaves with CaCO 3 more effective.

The Effects of Liming and Plant Residue Incorporation on Plant Growth
Both CaCO3 and OM affected plant growth of S. guianensis and P. natatum (Figure 1). Data indicated that OM increased, while CaCO3 decreased the SPAD value of S. guianensis leaves. In contrast, both OM and CaCO3 increased the SPAD value of P. natatum leaves with CaCO3 more effective.
In parallel with the SPAD value, biomass was also affected in a similar pattern (Figure 1), namely, OM increased the biomass of S. guianensis, while CaCO3 did not affect it, and in contrast, both OM and CaCO3 increased the biomass of P. natatum with CaCO3 more effective. In detail, OM increased the biomass of S. guianensis by 70.4%, while CaCO3 decreased by 8.6%. However, both OM and CaCO3 increased the biomass of P. natatum by 23.4% and 44.8%, respectively. This reflects not only the difference between CaCO3 and OM, but also the difference between the two plant species.

The Effects of Liming and Llant Residue Incorporation on Soil Properties
The soil pH was significantly increased by both CaCO3 and OM, and plant species also exerted a significant effect on the regulation of soil pH by different treatments (Table 1). In detail, the increase in soil pH of S. guianensis was slightly less than that of P. natatum (average-0.91 vs 0.99), and the increase by CaCO3 was greater than that by OM (average-1.49 vs 0.41) ( Table 1).
In the soil of S. guianensis, CaCO3 significantly increased TOC, TP and AP, while OM significantly increased TOC, DOC, TN, TP, AP and AK (Table 1). AN and TK were not affected by CaCO3 or OM (Table 1). Differently, in the soil of P. natatum, CaCO3 significantly decreased TN, TP, AP and TK, while OM significantly increased TOC, DOC, TN, AN and AK, but decreased AP and TK (Table 1). In parallel with the SPAD value, biomass was also affected in a similar pattern (Figure 1), namely, OM increased the biomass of S. guianensis, while CaCO 3 did not affect it, and in contrast, both OM and CaCO 3 increased the biomass of P. natatum with CaCO 3 more effective. In detail, OM increased the biomass of S. guianensis by 70.4%, while CaCO 3 decreased by 8.6%. However, both OM and CaCO 3 increased the biomass of P. natatum by 23.4% and 44.8%, respectively. This reflects not only the difference between CaCO 3 and OM, but also the difference between the two plant species.

The Effects of Liming and Llant Residue Incorporation on Soil Properties
The soil pH was significantly increased by both CaCO 3 and OM, and plant species also exerted a significant effect on the regulation of soil pH by different treatments (Table 1). In detail, the increase in soil pH of S. guianensis was slightly less than that of P. natatum (average-0.91 vs 0.99), and the increase by CaCO 3 was greater than that by OM (average-1.49 vs 0.41) ( Table 1). In the soil of S. guianensis, CaCO 3 significantly increased TOC, TP and AP, while OM significantly increased TOC, DOC, TN, TP, AP and AK (Table 1). AN and TK were not affected by CaCO 3 or OM (Table 1). Differently, in the soil of P. natatum, CaCO 3 significantly decreased TN, TP, AP and TK, while OM significantly increased TOC, DOC, TN, AN and AK, but decreased AP and TK (Table 1).

The Effects of Two Practices on Soil Microbial Community
DGGE profiling showed that the bacterial community compositions of three treatments were much different from each other, and the five replicate samples of each treatment grouped well ( Figure 2). Generally, the bacterial community compositions of CaCO 3 and Control were similar to each other, while that of OM was separated apart from CaCO 3 and Control for both S. guianensis and P. natatum. Moreover, the diversity index, species abundance and species evenness were not affected by either CaCO 3 or OM for P. natatum (Table 3). In contrast, for S. guianensis, the diversity index was significantly increased by both CaCO 3 and OM, while species abundance was significantly increased by only CaCO 3 .
while that of OM was separated apart from CaCO3 and Control for both S. guianensis and P. natatum. Moreover, the diversity index, species abundance and species evenness were not affected by either CaCO3 or OM for P. natatum (Table 3). In contrast, for S. guianensis, the diversity index was significantly increased by both CaCO3 and OM, while species abundance was significantly increased by only CaCO3.    3 ) and plant residue incorporation (OM) on soil enzyme activities. α-GLU, α-glucosidase; β-GLU, β-glucosidase; β-XYL, β-xylosidase; CEL, cellobiosidase; URA, urease; NR, nitrate reductase; CHI, chitinase; ACP, acid phosphatase; ALP, alkaline phosphatase. Data followed by the same letter are not significantly different for each plant species (Tukey's post hoc test, p = 0.05, n = 5). Enzymatic activity unit of URA, NR, ACP, ALP is U·g −1 , and that of others is µmol·L −1 MUF·g −1 ·h −1 .  Cluster analysis based on DGGE profiles showed that AMF community composition in the soil of P. natatum was also altered by both OM and CaCO 3 , with OM more effective (Figure 3). The diversity index and species abundance of AMF community were significantly increased by CaCO 3 but not OM (Table 4), which is different from that of the bacterial community.

Plant Species
Appl. Sci. 2020, 10, x FOR PEER REVIEW 10 of 16 Cluster analysis based on DGGE profiles showed that AMF community composition in the soil of P. natatum was also altered by both OM and CaCO3, with OM more effective (Figure 3). The diversity index and species abundance of AMF community were significantly increased by CaCO3 but not OM (Table 4), which is different from that of the bacterial community.  Table 4. Influences of liming (CaCO3) or plant residue incorporation (OM) on arbuscular mycorrhizal fungal community structure in the soil of P. natatum. Data followed by the same letter are not significantly different (Tukey's post hoc test, P = 0.05, n = 5).

CCA of the Effects of Two Practices on Soil Properties
Since the direct effects of CaCO 3 and OM treatments are pH increase and DOC increase, respectively, we performed CCA to explore how CaCO 3 and OM affected soil chemical properties and soil enzymatic activities via pH or DOC. CCA results indicated that pH exerted little influence on soil chemical properties, while DOC showed a strong positive effect on AK and TN, but a weak negative effect on TK, AN, AP and TP ( Figure 4A). For soil enzymatic activities, pH exerted a strong positive influence on ALP activity, while DOC showed a positive effect on β-glucosidase, chitinase, nitrate reductase and negative effect on acid phosphatase ( Figure 4B). In combination with the data in Tables 1 and 2, these results suggest that OM extensively influenced soil chemical properties and soil enzymatic activities mainly via DOC pathway while liming influenced fewer parameters mainly via pH.

CCA of the Effects of Two Practices on Soil Properties
Since the direct effects of CaCO3 and OM treatments are pH increase and DOC increase, respectively, we performed CCA to explore how CaCO3 and OM affected soil chemical properties and soil enzymatic activities via pH or DOC. CCA results indicated that pH exerted little influence on soil chemical properties, while DOC showed a strong positive effect on AK and TN, but a weak negative effect on TK, AN, AP and TP ( Figure 4A). For soil enzymatic activities, pH exerted a strong positive influence on ALP activity, while DOC showed a positive effect on -glucosidase, chitinase, nitrate reductase and negative effect on acid phosphatase ( Figure 4B). In combination with the data in Table 1 and Table 2, these results suggest that OM extensively influenced soil chemical properties and soil enzymatic activities mainly via DOC pathway while liming influenced fewer parameters mainly via pH.

Discussion
Both liming and plant residue incorporation can elevate the pH of acidic soils [3][4][5][6]; however, only a few experiments were designed to compare their effects [37,38]. In this study, we compared the effects of CaCO3 and OM in increasing the pH of subtropical orchard soils with an initial pH of 4.98. It is clear that CaCO3 is more effective in the amelioration of acidic soils than OM. When the effects of CaCO3 and OM are compared, it should be born in mind that their primary effects are much

Discussion
Both liming and plant residue incorporation can elevate the pH of acidic soils [3][4][5][6]; however, only a few experiments were designed to compare their effects [37,38]. In this study, we compared the effects of CaCO 3 and OM in increasing the pH of subtropical orchard soils with an initial pH of 4.98. It is clear that CaCO 3 is more effective in the amelioration of acidic soils than OM. When the effects of CaCO 3 and OM are compared, it should be born in mind that their primary effects are much different. CaCO 3 directly increases the soil pH in via soil chemical pathway, which can always be fast and obvious; however, OM indirectly increases it via soil biochemical pathway with the involvement of soil microbes. Elevating pH with plant residue incorporation involves microbial degradation of organic matter in the plant residues and the subsequent release of alkalinity (excess cations), and thus, can be time-consuming [8]. In this scenario, the effectiveness of plant residue incorporation depends to a certain degree on the compositions and types of plant residues and characteristics of soils [5].
Although some abiotic factors, such as temperature, soil moisture, pH are the important determinants of the decomposition process of plant residues, soil microbes are the key biotic factor contributing much to the decomposition. It is generally accepted that plant residues can be decomposed by a variety of soil microbes with priming effect [39,40]. In fact, the direct effect of OM is increasing DOC due to the decomposition of plant residues [41][42][43][44], and then the promoted microbial activity. In this study, the increased pH together with promoted microbial activity, further mobilized the TN, TP, TK, leading to the increased levels of AN, AP, or AK. It is suggested that the extra-cellular hydrolysis enzymes of soil microbes are mainly responsible for the mobilization of N, P, K [45], with Bacillus the most effective. Our results demonstrate that the microbial community structure in limed soil is closer to that in control soil than that in the soil with plant residue incorporation, indicating a greater regulating effect on bacterial community composition by OM than that by CaCO 3 . This difference suggests that, in parallel with the amelioration of acidic soil, OM can significantly reshape the microbial community structure while CaCO 3 can not. This regulated microbial community structure can further improve the microbial activity, as revealed by the soil enzyme activity in this study. This indicates that CaCO 3 exerted its influence on a limited array of soil chemical properties, while OM exerted its influence on a diverse array of soil chemical properties. Overall, the effects of OM in improving the soil are more comprehensive than CaCO 3 , including not only the increased pH, but also the mobilization of nutrients and the promoted enzyme activity.
In this study, although the manifest difference in the microbial community composition was observed according to DGGE profiling, we did not characterize the dominant species or the modified species. According to Lin et al. [46], liming significantly decreased the diazotroph abundance while plant residue did not, highlighting their difference in regulating specific microbial taxa. Consequently, it is valuable to further put forward this work with 16S rRNA sequencing technique, in order that the dominant or modified species in response to CaCO 3 or OM can be revealed, and their respective functions can be evaluated.
In this study, we also observed the difference between plant species, including the effects on soil pH and plant growth. The soil pH in rhizosphere of S. guianensis was significantly lower than that of P. natatum regardless of treatments, mainly due to the biological N fixation (BNF) in the rhizosphere of legumes [47,48]. It is well acknowledged that the excretion of H + from legume roots is in parallel with BNF [49]. The differences in soil nutrient content (TP, AP, AK) and enzyme activity (α-glucosidase, β-xylosidase, urase) between plant species also existed, but showed complexity due to the interaction of plant species and soil treatments. This complexity was demonstrated previously [6,50]. Taken together, these differences led to the different response of plant growths to treatments, namely, the biomass of S. guianensis was greatly increased by OM, but not by CaCO 3 , while the biomass of P. natatum was increased by CaCO 3 more greatly than by OM. The different growth responses of legumes and graminoids to liming were reported elsewhere. For example, liming at 6.72 t·ha −1 significantly increased the aboveground biomass of barley, but not that of peas (on the average of four years, where the initial soil pH was increased by 1.9 units) [51]. An on-farm experiment indicated that the frequency of a grain yield response to liming was almost twice for corn than for soybean [52], although the different results were also reported [53]. These results highlight the necessity of selecting the appropriate treatments (liming or plant residue incorporation) for ameliorating acidic soils when growing legumes or graminoids.
In this study, we demonstrated that the modification of the microbial community was achieved by both liming and plant residue incorporation, although the effect size of liming was weaker than that of plant residue incorporation. The modification can be the result of changes in pH and organic matter, as induced by liming and plant residue incorporation. It is well accepted that pH and organic matter are the two important factors altering microbial community composition and function [54][55][56][57]. Soil microbe is an essential biotic component in soil ecosystems, playing critical roles in diverse soil functioning. In this study, almost all soil chemical properties except AK were increased by liming or plant residue incorporation, probably due to both elevations of pH and increase in enzyme activity. The availability of most nutrients can be promoted if soil pH shifts from acidic status to neutral status with P being the most sensitive [58]. The activity of β-glucosidase decreases with soil pH increasing from 4.5 to 8.5, while organic matter increases it by providing substrates [59]. This explains well why the activity of β-glucosidase in OM treatment was higher than that in CaCO 3 treatment in this study. In contrast, the activity of alkaline phosphatase in CaCO 3 treatment was higher than that in OM treatment, most probably due to in higher pH in CaCO 3 treatment. Moreover, alkaline phosphatase is mainly secreted by soil microbes [60], and thus, is linked with increased TOC or DOC as demonstrated in this study. Taking together, both liming and plant residue incorporation affect soil chemical properties and biological properties, with the former more effective on chemical processes (especially pH) and the latter more effective on the biological processes. Despite the differences, both practices are able to improve acidic soil with increased nutrient availability and biological activity.

Conclusions
In summary, our study indicates that both liming and plant residue incorporation can ameliorate acidic soils, with liming more effective than plant residue incorporation. In parallel with pH increase, plant residue incorporation also increased the TOC and DOC contents. However, soil nutrient levels and enzyme activities were increased by liming or plant residue incorporation, the effect of plant residue incorporation was more comprehensive than that of liming. PCR-DGGE analysis revealed that liming changed the microbial community structure more greatly than plant residue incorporation, and in contrast, plant residue incorporation altered the microbial community composition much more than liming. The alteration of the microbial community composition by liming and plant residue incorporation should be characterized in detail in the future. When plant species are considered, the growth responses of S. guianensis (legume) and P. natatum (graminoid) to liming and plant residue incorporation are different, with plant residue incorporation more promotive for S. guianensis and liming more promotive for P. natatum. Our results provide a new sight in the amelioration of acidic soils, and help select appropriate practices in a combination of plants.