New Forests

, Volume 36, Issue 1, pp 53–65

Growth, biomass partitioning, and water-use efficiency of a leguminous shrub (Bauhinia faberi var. microphylla) in response to various water availabilities


  • Fanglan Li
    • Chengdu Institute of BiologyChinese Academy Sciences
    • Graduate School of the Chinese Academy of Sciences
    • Chengdu Institute of BiologyChinese Academy Sciences
  • Ning Wu
    • Chengdu Institute of BiologyChinese Academy Sciences
  • Chen You
    • Chengdu Institute of BiologyChinese Academy Sciences
    • Graduate School of the Chinese Academy of Sciences

DOI: 10.1007/s11056-008-9081-z

Cite this article as:
Li, F., Bao, W., Wu, N. et al. New Forests (2008) 36: 53. doi:10.1007/s11056-008-9081-z


Responses of the endemic leguminous shrub Bauhinia faberi var. microphylla, to various soil water supply regimes were studied in order to assess water stress tolerance of seedlings. Two-month-old seedlings were grown under water supply regimes of 100, 80, 60, 40, and 20% water field capacity (FC), respectively, in a temperature and light-controlled greenhouse. Plant height and leaf number were measured monthly over a 4-month period, while water use (WU), water-use efficiency (WUE), leaf relative water content (RWC), biomass production and its partitioning were recorded at the end of the experiment. Seedlings exhibited the greatest biomass production, height, basal diameter, branch number, leaf number, and leaf area when soil content was at 100% FC, and slightly declined at 80% FC. These parameters declined significantly under 60% FC water supply, and severely reduced under 40 and 20% FC. RWC, WU and WUE decreased, while the ratio of root mass to stem mass (R:S) increased in response to decreasing water supply. Water stress caused leaf shedding, but not plant death. The results demonstrated that B. faberi var. microphylla seedlings could tolerate drought by reducing branching and leaf area while maintaining a high R:S ratio. However, low dry mass and WUE at 40 and 20% FC suggested that the seedlings did not produce significant biomass under prolonged severe water deficit. Therefore, before introducing B. faberi var. microphylla in vegetation restoration efforts, water supply above 40% FC is recommended for seedlings to maintain growth.


Drought stressWater-use strategyLegumeShrubArid region


Water stress is a key factor limiting plant survival and growth, therefore adversely affecting vegetation restoration in arid and semiarid regions. A deeper knowledge of water-use strategies and drought tolerance of seedling is essential for restoration programs in these regions (Oliet et al. 2002; Sánchez-Coronado et al. 2007). Raising the probability of producing vigorous seedlings and successful establishment is one of the first steps in improving revegetation.

Physiological and morphological changes occur on plants in response to various water availabilities (Becker et al. 1994; Franca et al. 2000; Royo et al. 2001; Anyia and Herzog 2004; González-Rodríguez et al. 2005; Dias et al. 2007). The immediate responses to water deficit are the declines of leaf water potential and stomatal closure, which led to decreases in CO2 uptake and photosynthesis. Furthermore, prolonged drought can limit growth and biomass production, and cause plant death (Ogbonnaya et al. 1998; Anyia and Herzog 2004; Schrader et al. 2005; Rodiyati et al. 2005). The relationship DM = WUE × WU indicates that the dry mass production (DM) increased with increasing water-use efficiency (WUE) for constant water use (WU) (Martin et al. 1999; Rytter 2005). Thus, when we plant trees in the arid regions, both productivity and WUE should be considered as important characters, however patterns of WUE response are under debate because WUE shows a large variation between species (Ismail and Hall 1992; Li 1999; Rytter 2005). Therefore, additional research is needed in order to understand the relationships between biomass production and WUE.

For the dry valleys of the Hengduan Mountains in southwestern China low precipitation and high evaporation generate soil and climate drought. Mean soil water content was 3.8 and 9.5% in dry and rainy seasons, respectively (Wang 2002; Ma et al. 2004). This region is very unfavorable to regeneration of tree species (Bao et al. 1999; Ma et al. 2004). Revegetating with native shrubs may be ecologically reasonable. Bauhinia faberi var. microphylla is a perennial leguminous shrub that natively distributed in the dry valleys and other arid regions of China (Chen 1988). This species is often predominant on slopes (CIB 1962) where it plays a vital role in soil remediation and erosion control. Its leaves, flowers, and fruits have been used as fodders and medicines by local people (YPBH 1989; CIB 1962). Moreover, small (up to 1.5 m in height) and branchy, mature shrub is adapted to the dry environment and shows efficient sprouting and highly competitive behavior, thus it appears to have important potential for ecological restoration of degraded or arid regions. However, this multipurpose shrub alleviates pressures on remaining vegetation in recent decades, especially in areas where habitat undergoes destruction. Restoration of the shrub population in arid and semi-arid regions of China is crucial. Although seeds of B. faberi var. microphylla can germinate under drought conditions, i.e, with germination percentage of 60% at 1.20 MPa water potential (W.K. Bao, unpublished data), poor regeneration has been observed in arid region, seedlings are rarely observed in natural habitats according to present investigation in dry valleys of the Hengduan Mountains. We speculate that drought limits seedling emergence and further results in early dieback. However, until our study drought tolerance of the seedlings had not been investigated.

Thus, the present study conducted in the dry valleys of Minjiang River with hopes of planting B. faberi var. microphylla in the arid region as a reforestation species. The objectives were (1) to investigate the effects of water stress on growth, biomass production and its partitioning, and WUE of B. faberi var. microphylla seedlings; (2) to assess the degree to which seedlings tolerate to water stress; and (3) to obtain implications for population restoration of B. faberi var. microphylla.

Materials and methods

Plant materials and growth conditions

Seeds of B. faberi var. microphylla were collected in October 2004 from the typical dry valleys of Minjiang River in Maoxian County, Sichuan Province, China (103°54′–103°56′E, 31°37′–31°44′N, 1,700–1,900 masl) where mean annual temperature, rainfall and evaporation are 11.2°C, 494.8, and 1,332.4 mm, respectively (Bao et al. 1999). After drying for 1 week in open sunlight, deformed and damaged seeds were discarded, and apparently healthy were allowed to air-dry, then stored at ambient laboratory temperature (2–25°C) until the experimental pretreatment in March 2005. Surface soil was also obtained from the dry valleys of Minjiang River where B. faberi var. microphylla natively distributed. The soil collections were mixed thoroughly. The soil is clay loam (CSTRGISSCAS 1995; Pang et al. 2006; Ma et al. 2004) with rich gravel, pH 7.82, soil water content 9.51% (Wang 2002), organic matter 52.6 g kg−1, total N 1.61 g kg−1, total P 1.14 g kg−1, and total K 11.65 g kg−1. Ten days later 3.5 kg soil was placed in each of 50 plastic pots (25 cm diameters × 28 cm height), to a bulk density of 0.97 g/cm3. A total of 3.5 g slow release fertilizer (13% N, 10% P and 14% K) was added to each pot in order to ensure that all plants could obtain sufficient nutrients. Six weeks later 10 pots were watered and allowed to drain freely until the weight was constant. The difference between this weight and soil dry weight was used to calculate water field capacity (FC). At the level of full water FC, mean water content of soil was 39%.

Before sowing, seeds were treated with 2.5% of sodium hypochlorite (NaOCl) for 1 h. Seeds of similar size were sown four per pot in each of 20 pots on April 5, 2005. Another 20 pots without seeds were arranged as controls. All pots were moved into in an open sided, plastic covered greenhouse at the Maoxian Station for Ecosystem Research, Chinese Academy of Science (103°53′E, 31°41′N, 1,816 masl), where temperature was 30°C/11°C of day/night, and the relative humidity was 45–85%. At midday, photosynthetic photon flux density was about 1,600 μmol−1 m−2 s−1. Pots initially were well-watered (soil moisture was about 80% FC) in order to ensure seed germination. Shortly after emergence, seedlings were thinned to one plant per pot, and the watering treatment was initiated on June 22, 2005. Average dry mass of the thinned seedlings was determined.

Experimental design and management

The experimental layout was a completely random design with five water supply regimes (100, 80, 60, 40 and 20% FC) and four replications. Each treatment pot was paired with a pot without a seedling that served as a control to correct soil evaporation when determining WUE. Transpiration water loss was measured gravimetrically by weighing all pots and calculating the weight loss between watering. The amount of water to add to each pot was the amount lost to transpiration. During the experimental period, pots were watered every other day at 18:00, soil water content was always maintained at 38.8 ± 0.3, 31.6 ± 1.7, 25.6 ± 1.3, 16.5 ± 0.7, and 8.1 ± 1.1% under 100, 80, 60, 40, and 20% FC water supply regimes, respectively. Evaporation from the soil surface was minimized by covering with a 3-cm layer of quartz gravel. The experimental layout was surrounded with a single row of border plants to protect the experimental seedlings from external influences. All pots were put on bricks and randomized weekly. Following periods of rapid growth, increases in seedling weight were estimated from the regression of relationship between seedling fresh weight (Y, g), plant height (X1, cm) and seedling basal diameter (X2, mm): Y = 1.496 + 0.117 X1 + 0.998 X2, R= 0.848, P < 0.001 (F. Li, unpublished data). Weight of the pot plus seedling in each treatment was adjusted weekly. The experiment was terminated on September 22, 2005.

Parameter measurement and calculation

Plant height, measured from the base of stem at soil level to the terminal bud of the main stem, and leaf number were recorded monthly from June to September. The seedlings were harvested at the end of the experiment, and basal diameter, branch number, leaf area, and leaf relative water content (RWC) were measured. Images of compound leaves were recorded with a scanner (Model F6580, Founder Electronics Co., Ltd, Beijing), and images were digitized by the Arcview 3.2a [Environmental Systems Research Institute (ESRI), Inc., New York] software in order to determine both total and individual leaf area. Each seedling was divided into roots, stems, and leaves and dried in an oven for 48 h at 70°C for biomass determination. Total plant biomass was the sum of root, stem, and leaf mass. Root:stem (R:S) ratio was calculated as the root dry mass divided by stem dry mass. Leaf area ratio (LAR) was obtained as total leaf area divided by total dry mass of each seedling. Specific leaf area (SLA) was calculated as the total leaf area divided by the dry mass of leaves.

Water-use efficiency for each seedling was determined by the ratio of dry mass production to total water transpired during the experiment. Average dry mass of thinned seedlings at the beginning was subtracted from final seedlings dry mass for WUE calculation. While calculating the amount of water transpired during the experiment, evaporative loss from the pots was taken into account by subtracting the average amount of water loss from the control pots without plants from each watering treatment.

Leaf RWC was determined gravimetrically pre-dawn. Five fully expanded compound leaves were selected per seedling at the mid-canopy position, placed in dishes containing wet filter paper and weighed immediately, and recorded as fresh weight (FW). Turgid weight (TW) was determined for leaves were floated in distilled water in a closed container at 4°C in the dark for 24 h. Dry weight (DW) was determined for the same leaves after oven drying for 48 h at 70°C. RWC was calculated as: RWC (%) = [(FW – DW)/(TW – DW)] × 100 (Gindaba et al. 2004; Jeon et al. 2006).

Statistical analysis

All the variables from measurements were analyzed using one-way ANOVA with water supply regimes as a factor (n = 4). LSD multiple comparison tests were used to separate significant differences in mean values at the 0.05 level. Relationships among variables were determined using the Pearson’s correlations coefficient test at the 0.05 level. Regression lines were employed for model analysis of variables significant correlations. One-way ANOVA was also performed on the leaf number and plant height with water supply regimes as a factor at the each of 4 months. All of the statistical analyses were performed using SPSS (Standard released version 11.5 for Windows, SPSS Inc., IL, USA) software package.


Variation in growth

Height and leaf number of B. faberi var. microphylla seedlings displayed responses to water availability within 1 month of initiating treatments, and significant cumulative responses to prolonged drought (Fig. 1a, b). All the seedlings exhibited continuous growing until the end of experiment; the increments of height and leaf number were more rapid under 100 and 80% FC compared to other regimes. There was no mortality during the experiment, but leaf shedding was found after 2 months of treatment, more senescent leaves were observed at 20% FC regime (Fig. 1c). Compared with 100% FC water supply regime, 80 and 20% FC water supply caused 22 and 70% reduction in final height, and 19 and 81% in final leaf number, respectively (Fig. 2a, d); total leaf area was also reduced by 40, 78 and 91% under the 80, 60 and 20% FC conditions, respectively (Fig. 2e). Below 60% FC water stress significantly retarded branch number and basal diameter (Fig. 2b, c).
Fig. 1

Plant height (a) and number of green leaves (b) and number of senescent leaves (c) of Bauhinia faberi var. microphylla seedlings under five water supply regimes. Bars represent standard error of mean (n = 4). Different letters in the same month indicate significant differences across water supply regimes according to LSD test (P < 0.05)
Fig. 2

Responses of plant height (a), basal diameter (b), branch number (c), total leaf number (d) and total leaf area (e) of Bauhinia faberi var. microphylla seedlings to various water availabilities. Bars represent standard error of mean (n = 4). Different letters indicate significant differences across five water supply regimes according to LSD test (P < 0.05)

Compared with 100% FC water supply regime, total dry mass was reduced by 21, 73, and 91% when seedlings were given 80, 60 and 20% FC, respectively (Table 1). There was no significant difference in biomass production between 40 and 20% FC. Patterns in leaf mass were similar to those in total dry mass. With decreasing water supply from 100 to 60% FC regime, stem mass, and root mass decreased by 85 and 59%, respectively. Differences in R:S ratio among the 60, 40, and 20% FC regimes were not statistically significant, but all were higher than those at 80% and 100% FC (Table 1).

Leaf RWC was also altered in response to water supply regimes (Table 2). Compared with 100% FC regime, seedlings reduced RWC by 11, 26, and 62% at the 60, 40, and 20% FC, respectively. There was no significant difference in RWC between 100 and 80% FC. SLA was significantly increased at 20% FC regime, but the differences among other regimes were not statistically significant (Table 2). Equivalent decreases in leaf area and plant mass with water stress caused LAR to stay constant.

Variation in water use and water-use efficiency

Water use did not differ between the seedlings of 100 and 80% FC, but it was significantly low for the seedlings in other three regimes (Table 1). WUE was also reduced by decreasing water supply though it did not exhibit significant differences between 100 and 80% FC, and between 40 and 20% FC (Table 1).

Discussion and conclusion

Water stress was a very important limiting factor for seedling growth. The observed changes in growth, biomass production, and partitioning of B. faberi var. microphylla under various water availability supported findings of other studies on tree seedlings (Rose et al. 1993; Yin et al. 2004), shrubs (Niu et al. 2005; Martin and Stephens 2006; Villagra and Cavagnaro 2006), and herbs (Greco1 and Cavagnaro 2002; Anyia and Herzog 2004; Liu and Stützel 2004; Rodiyati et al. 2005). The present study indicated that although there was no mortality during the experiment, the B. faberi var. microphylla seedlings were vulnerable under severe drought condition because of significant leaf shedding and low growth.

Reduction of individual leaf area and formation of new leaves, together with leaf shedding resulted in decrease of total leaf area (Table 2 and Fig. 1). As in other species (Muchow 1985; Buchanan-Wollaston 1997; Maggio et al. 2005), it is a morphological adaptation to reduce water loss. However, the positive correlation between total leaf area and biomass production (Fig. 3a) revealed that decreasing leaf area limits productivity. Significant reduction of height, branch and leaf numbers of the seedlings under water stress (Fig. 2) is partially due to decreases in both apical and intercalary growth (Ogbonnaya et al. 1998; Sardans et al. 2005). High branching would create more surface area, it would be wasteful of soil water (Ogbonnaya et al. 1998; Bañon et al. 2004). Therefore the inhibition of branching under drought conditions observed here in B. faberi var. microphylla is also assumed to be an adaptive mechanism to conserve water.
Fig. 3

Regression of total leaf area on (a) total dry mass and (b) root mass of Bauhinia faberi var. microphylla seedlings (P < 0.001)

Plant productivity is strongly related to the processes of biomass partitioning under water stress (Klepper and Rickman 1990; Kage et al. 2004). Our study was consistent with those on other species (e.g., Ares and Fownes 1999; Bargali and Tewari 2004; Liu and Stützel 2004; Singh and Singh 2006), in that water stress decreased total dry mass and their components, and altered biomass allocation to root systems resulting in higher R:S ratio in stressed seedlings (Table 1). The positive correlation between root mass and WUE of B. faberi var. microphylla seedlings implied that a developed root system could reflect capacity of water uptake supporting previous findings (Garnier 1991; Kozlowski and Pallardy 2002). The significant positive relationship between root mass and total leaf area (Fig. 3b) further reflected the balance between plant organs that govern water uptake and loss. Root growth depends on the supply of carbohydrate from aboveground parts (Greco1 and Cavagnaro 2002), so the reduction of leaf area would limit root growth.

The increased SLA resulting from severe water stress observed here in B. faberi var. microphylla (Table 2) agrees with a finding of Bargali and Tewari (2004) who suggested that plant in low water supply requires large leaf area per leaf weight. It seems that a low RWC would result in stomatal closure and reduced net photosynthetic rate per unit leaf area. In addition a high SLA implies the thin leaves (Carter et al. 1997; Cordell et al. 1998; Castro-Diez et al. 2000). By using data for leaf anatomy from F. Li (unpublished work) concerning seedlings similar to the ones used here, we found that leaf thickness decreased with water stress. However, there are not other reports on this phenomenon in other similar species.

Reports for crops (Ogbonnaya et al. 1998; Gindaba et al. 2004) and woody plants (Gratani and Varone 2004; Masinde et al. 2006) indicate that RWC rarely falls below 50% in well-watered conditions, but commonly reaches 50–40% and occasionally 30 to 20% during severe drought. Similarly, the response of RWC in B. faberi var. microphylla seedlings depended on the intensity of the stress imposed. Mild stress (60 to 40% FC) induced little or no decline and seedlings retained more than 60% of RWC after a 3-month water deficit. The seedling subjected to stress (20% FC) displayed sharp decline in RWC. It is possible that soil water level of 40% FC was a critical water content, below which the seedling would be dehydrate severely and the leaves would senesce. Our result also showed that RWC were positively associated with biomass production (Fig. 4a). This is consistent with the findings of Gindaba et al. (2004), who reported that decreased RWC under drought would result in decline in photosynthate accumulation.
Fig. 4

Regression of (a) total dry mass on leaf relative water content (RWC); (b) root mass on water-use efficiency (WUE) of Bauhinia faberi var. microphylla seedlings (P < 0.001)

The decrease in WUE of B. faberi var. microphylla seedlings under water stress, contradicts the finding with some species (Hubick et al. 1986; Liu et al. 2005; Maggio et al. 2005), and agrees with reports on crops (Zhao et al. 2004; Clavel et al. 2005) and on Acacia ampliceps (Akhter et al. 2005). Engelbrecht et al. (2005) suggested that decreased WUE in water-stressed seedlings might be related to their root systems which are not well-established to acquire water from lower soil layers. In addition, water stress probably resulted in more lignified roots which could decline hydraulic conductivity and water uptake (Jupp and Newman 1987; Mapfumo et al. 1992; North and Nobel 1992). When soil was dried, therefore, both low growth and lignification of the roots led to very low rate of water uptake resulting decrease in WUE. It means that B. faberi var. microphylla seedlings employed a prodigal water-use strategy. WUE decreased because WU decreased less than biomass accumulation with decreasing soil moisture.

In conclusion, it appears the B. faberi var. microphylla seedlings could tolerate water stress through morphological and physiological responses. Although drought had a strong negative effect on production, some growth occurred even at the most severe stress. However, significant leaf shedding, low biomass, and WUE at 40 to 20% FC conditions revealed that this species cannot generate vigorous seedlings under severe drought. Therefore, before introducing B. faberi var. microphylla in restoration efforts, water supply above 40% FC could be recommended for this species to raise vigorous seedlings in the first year of growth. Given our results we conclude poor regeneration in the natural habitats may be attributed to more than just low initial survival of the seedlings under drought conditions. We propose that there could be other factors caused by more severe drought (e.g., soil water content is only 3.8% in dry season) and lack of effective soil seed bank, which constitute reason for poor regeneration in the field. Therefore, further study combining with quantities in the field determine effects of these factors on the seedling survival.
Table 1

Biomass production, water use (WU), water-use efficiency (WUE) and root:stem ratio (R:S ratio) (mean ± SE) of Bauhinia faberi var. microphylla seedlings under five water supply regimes

Water supply regimes (%)

Biomass (g)

R/S ratio

WU (g)

WUE (g kg−1)






12.50 ± 0.66 a

3.20 ± 0.13 a

5.42 ± 0.54 a

3.88 ± 0.38 a

0.73 ± 0.09 a

2.30 ± 0.18 a

5.49 ± 0.37 a


9.84 ± 0.90(79) b

2.02 ± 0.29(62) b

3.81 ± 0.44(70) b

4.00 ± 0.38(103) a

1.08 ± 0.06 b

2.16 ± 0.25 a

4.63 ± 0.36 a


3.40 ± 0.65(27) c

0.97 ± 0.11(30) c

0.84 ± 0.10(15) c

1.60 ± 0.44(41) b

1.81 ± 0.29 c

0.93 ± 0.12 b

3.59 ± 0.34 b


1.64 ± 0.14(16) d

0.47 ± 0.06(15) d

0.54 ± 0.03(10) c

0.61 ± 0.07(16) c

1.12 ± 0.10 bc

0.96 ± 0.08 b

1.73 ± 0.21 c


1.11 ± 0.04(9) d

0.35 ± 0.02(11) d

0.32 ± 0.05(6) c

0.44 ± 0.04(11) c

1.52 ± 0.31 c

0.61 ± 0.02 b

1.81 ± 0.12 c

Different letters within a column indicate significant differences across five water supply regimes according to LSD test (n = 4, P < 0.05); bracket values indicate the percentages on 100% FC water supply regimes

Table 2

Leaf area (LA), leaf area ratio (LAR), specific leaf area (SLA) and leaf relative water content (RWC) (mean ± SE) of Bauhinia faberi var. microphylla seedlings under five water supply regimes

Water supply regimes (%)

LA (cm2)

SLA (cm2 g−1)

LAR (cm2 g−1)

RWC (%)


3.00 ± 0.03 a

162.51 ± 6.47 ac

45.67 ± 3.08 ab

89.48 ± 2.28 a


2.74 ± 0.23 a

173.20 ± 2.21 a

35.00 ± 4.80 a

86.72 ± 1.82 a


1.72 ± 0.06 b

143.10 ± 5.52 b

38.38 ± 4.80 ab

80.44 ± 0.96 b


1.62 ± 0.04 bc

168.07 ± 2.66 c

48.94 ± 5.03 b

66.55 ± 1.53 c


1.34 ± 0.10 c

191.00 ± 6.66 d

49.47 ± 3.52 b

34.09 ± 1.82 d

Different letters within a column indicate significant differences across five water supply regimes according to LSD test (n = 4, P < 0.05)


This study was funded by Chinese Academy of Sciences Action-Plan for West Development (KZCX2-XB2-02) and the Western Light Talent Training Plan of the Chinese Academy of Sciences. We extend our thanks to Maoxian Station for Ecosystem Research of the Chinese Academy of Sciences for providing facilities and technical assistance, and to Prof. Kenneth A. Albrecht for revision of the English text.

Copyright information

© Springer Science+Business Media B.V. 2008