Botanical Studies (2010) 51: 451-456.
Growth responses of two Mosla species to soil nitrogen and water supply
Qian-Jin CAO1, Meng WANG1, Ying GE1, Scott X. CHANG2, Jian-Min ZHANG1, and Jie CHANG1*
1 College of Life Sciences, Zhejiang University, 388 Yuhangtang Road, Hangzhou 310058, P.R. China
2Department of Renewable Resources, University of Alberta, Edmonton, Alberta, Canada T6G 2E3
(Received December 22, 2008; Accepted February 9, 2010)
ABSTRACT. The different responses to growth condition changes due to nitrogen deposition and rainfall variability among congeneric species may alter their differentiation pattern. We grew two Mosla species (La­miaceae) under different nitrogen and water conditions. Under waterlogging, the two species showed similar positive growth responses to high nitrogen supply, implying potential chances to extend their populations in very wet habitats. However, under the wet and relatively dry conditions, M. dianthera was less tolerant to high nitrogen (biomass of M. dianthera was suppressed by high nitrogen more remarkably than that of M. cavaleriei) but more tolerant to nitrogen limitation than M. cavaleriei (with the decrease in water supply, root mass ratio of M. dianthera was increased by nitrogen limitation more noticeably than that of M. cavaleriei). Therefore, in relatively dry habitats, soil nitrogen addition might be unfavorable for M. dianthera.
Keywords: Differentiation; Habitat; Morphology; Mosla cavaleriei; Mosla dianthera; N deposition.
INTRODUCTION
Plant species are experiencing global environmental changes including increasing nitrogen (N) input into soils due to deposition (Moore et al., 2004) and altered precipitation regimes (GeBler et al., 2007). Each species has its unique niche and environmental optimum. Thus, the responses to changes in growth conditions possibly differ among species. For instance, rare species are more susceptible to N fertilization than abundant species (Suding et al., 2005). The different responses among species co-occurring in one ecosystem bring changes in ecosystem composition (Suding et al., 2005), which is studied extensively. However, less attention is paid to species belonging to phylogenetic groups, though the different responses among closely-related taxa may change the differentiation pattern. Moreover, for a given species, the combination of current status of existence with its special response may determine its fate in the context of global environment changes.
The genus Mosla Buch.-Ham. ex Maxim. (Lamiaceae) is very suitable for studying the responses of closely-related species to changing growth conditions. Most


*Corresponding author: E-mail: jchang@zju.edu.cn; Tel & Fax: +86-571-8820-6465; Mobil: 133-9658-0523.
species in the genus are mainly distributed in South and East China, the most likely origin of the genus (Zhou, 1995). All species are annual herbs and reproduce through seed dispersal. In the main distribution areas, plants of the genus are experiencing environmental changes. The maximum of N deposition rates in South and East China reaches 64 kg N ha-1 yr-1 (Lu and Tian, 2007). Simultaneously, the frequencies of intense rainfalls and the annual precipitation are increasing, but the annual rainy days are decreasing in these areas (Qian et al., 2007). Mosla cavaleriei Levl. and M. dianthera (Buch.-Ham. ex Roxb.) Maxim. are included in this study. These two species have the greatest morphological and genetic similarities in the genus (Zhou, 1995), but they differ in their current status of existence. M. cavaleriei is limited to wet and fertile habitats, where it is subordinate in its communities; M. dianthera is found in habitats with relatively low N concentration and diverse soil water status and is frequently dominant in its communities. The questions asked in this study are (1) whether the two congeneric species with similar morphology and genetics but different current status of existence respond differently to changes in growth conditions and (2) whether the differences in responses to soil N between species are related to the N status in their natural habitats. This may aid us to predict the effects of global environmental changes on closely-related taxa.
452
Botanical Studies, Vol. 51, 2010
MATERIALS AND METHODS
yr-1 on the base of sufficient N supply. Consequently, the N concentration in the HN treatment increased to 1.67-fold that of the MN treatment. Factor two, water condition, had three levels: waterlogging (WL, water surface was main­tained at about 2 cm above soil surface), wet (water content > 90% of WHC), and diy (water content > 30% of WHC) condition. For the wet and dry treatments, plants were not watered with distilled water unless the soil water content dropped to 90% and 30% of WHC, respectively. To pre­vent water leakage from the bottom of the pots, plants were watered slowly so that water would be fully absorbed by the soil. Watering was performed around 18:00 every day. In total, there were 12 treatment combinations for each of the two species. Each treatment combination had six replicates (six pots). Three individuals were grown in each pot. The extent of soil water depletion depended on the plant growth and weather conditions, so the pots were weighed every day to determine the watering time for wet and dry treatments. The day/night temperature during the experiment was about 35/20°C. The average daily amount of light the plants experienced was 29.3 mol m-2 d-1 on cloudless days.
Plant materials and cultural conditions
Major habitats of the two Mosla species are hilly road­sides, forest edges, and brims of streamlets, where the soils are usually less than 10 cm thick. Therefore, we used pots (12 cm diameter, 9 cm height) to cultivate the plants in this study. The pots should not limit the root growth of the two species substantially compared with fields. In nat­ural habitats of M. cavaleriei and M. dianthera, the alkali-hydrolyzable N concentration in soils is in the range of 150-350 mg kg-1 and 50-270 mg kg-1, respectively. Thus, we designed four levels of N concentration to simulate the different N status in habitats of the two species. Besides, constant waterlogging and discontinuous watering were used to respectively simulate the constant stream at brims of streamlets and intense rainfalls followed by many non-rainfall days in nature.
In October 2006, we collected seeds of M. cavaleriei and M. dianthera from natural populations in Yongjia County (120.78° E longitude, 28.48° N latitude), Zhejiang Province in East China. After the air-dried seeds were placed in a refrigerator (0-4°C) for one month to break dormancy, the seeds were germinated in trays with peat (Metro Mix 290) on 20 April 2007. The trays were placed in growth chambers with a 16 h photoperiod, day/night temperatures of 25/15°C, and 70-80% relative humidity at an irradiance of about 25 [imol photon m-2 s-1 (pho-tosynthetically active radiation, X= 400-700 nm). After emergence, the seedlings were watered with full strength Hoagland's nutrient solution (Hoagland and Arnon, 1950). When the seedlings reached about 5 cm (20 May 2007), they were transplanted to pots mentioned above (filled with 120 g of a peat, vermiculite, and perlite mixture (10:7:3 v:v:v) ). The soil water holding capacity (WHC) was 79.2%. One week later, the N and water treatments were applied.
Measurements and calculations
Seven weeks after the treatments were applied (15 July 2007), when plants of the two species were at the vigorous vegetation growth stages, all individuals were measured for their height. Then, each individual plant was clipped at the soil surface. All green leaves from each plant were collected and scanned with a scanner (ScanMaker 4900, Microtek International Inc., USA) immediately after harvesting. The leaves with severe chlorosis or wilting were not measured. The WinFLORA Pro 2002a software (Regent Instruments INC, Quebec, Canada) was used to determine leaf area from the images scanned. However, because the roots of plants in each pot were so entangled that it was not feasible to separate them by individuals, the biomass of roots, stems, and leaves of all individu­als in each pot were measured totally after being dried in an oven at 65°C for at least 48 h to constant weight. Leaf mass ratio (LMR, leaf mass/total mass), stem mass ratio (SMR, stem mass/total mass), root mass ratio (RMR, root mass/total mass), specific leaf area (SLA, leaf area/leaf mass), and leaf area ratio (LAR, leaf area/total mass) were calculated following Hunt (1978).
Experimental design
The experiment was conducted in a greenhouse at Zhejiang University (120.08° E longitude, 30.30° N lati-tude), where we used a completely randomized factorial design. Factor one, N condition, had four levels: high N (HN, 167% N), medium N (MN, 100% N, as control), low N (LN, 40% N), and severely low N (SN, 10% N). The MN, LN, and SN treatments were 100%, 40%, and 10% N concentration (the amount of compounds including N was reduced in proportion) of the complete Hoagland's solu-tion (Hoagland and Arnon, 1950) with addition of CaCl2, KH2PO4, and KCL to maintain equal Ca2+, K+, and H2PO4-concentrations in solution. For these three treatments, 15 ml nutrient solution was sprayed equably in each pot every week. We designed the HN treatment in reference to stud-ies on N deposition (Bowden et al., 2004). We simultane-ously added 15 ml complete Hoagland's solution and 15 ml solution of 6.48 mmol L-1 NH4NO3 per pot per week for the HN treatment, equivalent to the addition of 15 g N m-2
Data analysis
To evaluate the differences in the growth traits between species and among N and water conditions, a three-way ANOVA was performed after data transformations (if necessary, Table 1). After ANOVA, coefficients of variance (CVs) for the growth traits of each species were calculated from standard deviations and means. The least significant difference (LSD) post-hoc comparisons were used to separate the results among N treatments under the same water conditions. In all of the above tests about significance levels, differences were considered
CAO et al. ― Mosla responses to soil N and water
453
Table 1. The P values of three-way ANOVAs for testing the differences in the growth traits of two Mosla species (M. cavaleriei and M. dianthera) between species and among soil water conditions (wet and relatively dry conditions, and waterlogging), N supply (high, medium, low, and severely low).

Trait
Species (S)
Water (W)
N
SxW
SxN
WxN
SxWxN
(d.f.=1)
(d.f.=2)
(d.f.=3)
(d.f.=2)
(d.f.=3)
(d.f.=6)
(d.f.=6)

Total masslog (g)

***

***

**

***

*

***

NS
Leaf masslog (g)
***
***
*
***
**
**
NS
Stem masslog (g)
***
***
***
***
NS
***
NS
Root masslog (g)
***
***
NS
***
NS
***
NS
LMR
NS
***
NS
NS
***
**
*
SMR
NS
*
***
NS
***
**
NS
RMR
NS
**
***
NS
NS
NS
*
Plant heightlog (cm)
***
***
**
***
NS
NS
NS
SLA (cm2 g-1)
***
NS
NS
NS
NS
***
NS
LAR (cm2 g-1)
***
**
**
NS
**
**
NS
*, P<0.05; **, P<0.01; P<0.001; NS, P>0.05, not significant. The superscripts above parameters denote the functions of data transformation. LMR, leaf mass ratio; SMR, stem mass ratio; RMR, root mass ratio; SLA, specific leaf area; LAR, leaf area ratio.
statistically significant if P< 0.05. All the statistical analyses were conducted using STATISTICA version 6.0 (StatSoft Inc., OK, USA).
Table 2. Coefficients of variance (CVs) of growth traits of two Mosla species supplied with four levels of N (H, high; M, medium; L, low; S, severely low) under three water conditions (wet and relatively dry conditions, and waterlogging).
RESULTS

Trait                                                       CV (%)
Biomass production

M. cavaleriei

M. dianthera

Total masslog (g)

18.1

9.4
Leaf masslog (g)
18.1
13.9
Stem masslog (g)
14.1
11.7
Root masslog (g)
18.0
13.1
LMR
40.5
29.6
SMR
31.9
18.8
RMR
31.7
31.4
Plant heightlog (cm)
25.2
12.7
SLA (cm2 g-1)
35.2
32.4
LAR (cm2 g-1)
52.6
38.8
Biomass production was significantly different between the two Mosla species (Table 1). For these species, water conditions significantly affected all four mass traits while root mass did not significantly differ among N treatments. The CVs of mass traits were greater in M. cavaleriei than in M. dianthera (Table 2). Under the wet and relatively dry conditions, total mass of M. dianthera was significantly lower in the SN and HN treatments than in the MN treatment. For M. cavaleriei the differences of total mass among N treatments were not so large as for M. dianthera (only under the wet conditions. The total mass of M. cavaleriei was significantly lower in the HN treatment than in the LN treatment) (Figure 1). Under the WL conditions, total mass of the two species was significantly greater in the HN treatment than in the other three N treatments.

The superscripts above parameters denote the functions of data transformation and the corresponding CVs of these traits were calculated from the data transformed. LMR, leaf mass ratio; SMR, stem mass ratio; RMR, root mass ratio; SLA, specific leaf area; LAR, leaf area ratio.
Biomass allocation
Although the two species considerably responded to N and water conditions in terms of biomass allocations (Table 1), the allocation traits were not significantly different between these species. The CVs of allocation traits were greater in M. cavaleriei than in M. dianthera (Table 2). Under the three water conditions, the RMR of the two species rose when N supply turned limited. With the decrease in water supply, the differences in RMR between the MN and SN treatments increased for M. dianthera while they decreased for M. cavaleriei (Figure 2). The LMR of M. cavaleriei was much greater in the
low and high N treatments than in the MN treatment under the three water conditions. In contrast, M. dianthera stressed by drought significantly decreased its LMR when N supply was limited. For M. cavaleriei grown under the WL and wet conditions and M. dianthera grown under WL conditions, the SMR decreased while the LMR and RMR increased in the HN treatment compared with the MN treatment.
454
Botanical Studies, Vol. 51, 2010
Figure 1. Biomass (means SE)) of two Mosla species supplied with four levels of N (H, high; M, medium; L, low; S, severely low) under three water conditions (dry and wet conditions and waterlogging (WL)). The different letters above the bars denote significant differences in total mass among N treatments. Note the different scales in the two histograms.
Figure 2. Biomass allocation (means SE) to leaves (LMR), stems (SMR), and roots (RMR) of two Mosla species supplied with four levels of N (H, high; M, medium; L, low; S, severely low) under three water conditions (dry and wet conditions and waterlogging (WL)).
Morphological traits
Both plant height and LAR of the Mosla species were significantly different between species and among N and water conditions while SLA showed no significant response to N or water treatments (Table 1). The CVs of the three traits were greater in M. cavaleriei than in M. dianthera (Table 2). Under the three water conditions, high and low N supply considerably decreased the plant height of M. cavaleriei (except high N under the WL conditions), but the impacts on M. dianthera were minor (Figure 3). Compared with the MN treatment, high and low N supply tended to increase SLA and LAR of the two species under the wet conditions, but the tendency was more remarkable in M. cavaleriei than in M. dianthera (Figure 3). Under the relatively dry conditions, SLA did not respond to N supply for the two species while LAR showed considerable responses.
Figure 3. Plant height, specific leaf area (SLA) and leaf area ratio (LAR) (means SE) of two Mosla species supplied with four levels of N (HN, high; MN, medium; LN, low; SN, severely low) under three water conditions (dry and wet conditions and waterlogging (WL)). The different letters above the bars denote significant differences among N treatments.
DISCUSSION
Similarly positive responses of the two species to high N under waterlogging conditions
The enhanced growth under the WL conditions suggests that the two Mosla species exhibited positive responses to high N supply when the roots were waterlogged. Waterlogging favors the availability and effective diffusion of some nutrients (Teo et al., 1994) and root uptake of plants (Rubio and Lavado, 1999), which might ameliorate relative nutrient limitation in the HN treatment, thus maintaining vigorous growth of the species under supra-optimal N conditions. Moreover, the waterlogged plants of the two species favorably adjusted their biomass allocation and morphological traits (increasing plant height, RMR, and LMR, and decreasing SMR) to contribute to the improved growth.
Stronger tolerance of M. cavaleriei to high N under non-waterlogging conditions
The depressed growth under the wet and dry conditions indicates that the two Mosla species responded negatively to high N supply under non-waterlogging conditions. This may be related to relative (rather than absolute) limitation of other nutrients (such as P) due to supra-optimal N, like other terrestrial plants (grassland species, Johnson et al., 1999; moorland species, Kirkham, 2001). Nevertheless, the suppressive effects on M. dianthera were more remarkable than those on M. cavaleriei, which suggests that the former may be less tolerant to high N
CAO et al. ― Mosla responses to soil N and water
455
supply than the latter under non-waterlogging conditions. This is supported by the adjustments of biomass allocation and morphological traits. For M. cavaleriei the increase of RMR, LMR, SLA and LAR, and the decrease of SMR and plant height in the HN treatment should promote the efficiency of light capture and the absorption ability of roots when nutrient resources other than N are relatively limited. No similar adjustments were observed in M. dianthera.
N levels responded differently, and the differences in ecotype response were correlated to the N content of the soil in their natural habitats (Eriksen and Nordal, 1989). Therefore, for many plants the responses to soil N variations are closely related to adaptations to N concentrations in the natural habitats.
In the main distribution areas of the two Mosla species, they are experiencing similar environmental changes, including N addition and water status variation in soils. For both species, the negative responses to high N supply under non-waterlogging conditions and the positive responses under waterlogging conditions imply that the influences of N addition are dependent on soil water status. In the context of high rates of N input into ecosystems, therefore, the two species may have more potential chances to extend their populations in very wet habitats such as brims of streamlets and wetlands rather than relatively dry habitats. Under drought conditions, the two species responded differently to either high or low N: M. dianthera was less tolerant to high N but more tolerant to N limitation than M. cavaleriei. At present, the relatively dry habitats such as open lands on the top of hills are inhabited only by M. dianthera and not M. cavaleriei. Thus, N addition in soil environments might be unfavorable for the growth of M. dianthera in relatively dry habitats.
Acknowledgements. The project was supported by the National Science Foundation of China (No. 30870235, 30570113), the Research Fund for the Doctoral Program of Higher Education (No. 20060335008), and the National Science Foundation for Post-doctoral Scientists of China (No. 20080431289).

More favorable adjustments of M. dianthera to N limitation than M. cavaleriei
Under the three water conditions, both of the two Mosla species were negatively responsive to limited N supply in terms of biomass production, though the amplitude of responses were different between species and among water conditions. Responsive to N limitation, the two species grown under the three water conditions increased RMR to a different extent, which is consistent with the "functional equilibrium" theory (Poorter and Nagel, 2000). In response to the decrease in water supply, nevertheless, the difference of RMR between N treatments increased in M. dianthera and decreased in M. cavaleriei. This shows that the influence of N limitation on plant growth depends on soil water status, and M. dianthera can favorably adjust biomass allocation to tolerate N limitation despite soil water limitation while M. cavaleriei can do so only under conditions that supply enough soil water. This shows that N and water in soils influence plants in an interactive way, which has been observed in a study on two Picea species (Patterson et al., 1997).
Differentiation between species
The difference in responses to soil N variations between the two species may be explained by the long-term adaptation to different habitats. The strong tolerance of M. cavaleriei to high N independent of soil water supply may benefit from the adaptation to habitats with high N levels. For M. dianthera, however, the intolerance to high N under non-waterlogging conditions and the favorable adjustments to N limitation may be related to its preference for habitats with relatively low N levels. The differences in growth responses and adaptation to habitats indicate a differentiation of optimal and realized N niches between the two species, and their realized N niches are consistent with the optimal. This favors the survival and population extension of the two species. However, the differences in current status of existence between the species may be related to the strong tolerance of M. dianthera to limitations of other environmental factors (e.g., light and soil water, Liao et al., 2006) and its relatively high ability to produce seeds (Jie Chang, unpubl.). Similarly, Patterson et al. (1997) found that two Picea species differentiating in soil N niches exhibit different sensitivities to N limitation. (The species preferring nutrient-poor habitats is less sensitive). To N augmentation, four ecotypes of Cochlearia officinalis occupying habitats with different
LITERATURE CITED
Bowden, R.D., E. Davidson, K. Savage, C. Arabia, and P. Steu-dler. 2004. Chronic nitrogen additions reduce total soil res­piration and microbial respiration in temperate forest soils at the Harvard Forest. For. Ecol. Manag. 196: 43-56.
Eriksen, A.B. and I. Nordal. 1989. Ecotypic differentiation in re­lation to soil nitrogen in Northern Scandinavian Cochlearia officinalis. Holarctic Ecol. 12: 31-38.
GeBler, A., C. Keitel, J. Kreuzwieser, R. Matyssek, W. Seiler, and H. Rennenberg. 2007. Potential risks for European beech (Fagus sylvatica L.) in a changing climate. Trees 21:
1-11.
Hoagland, D.R. and D.I. Arnon. 1950. The water-culture method for growing plants without soil. Calif. Agr. Exp. St. Circ. 347: 1-32.
Hunt, R. 1978. Plant Growth Analysis. London, Edward Arnold.
Johnson, D., J.R. Leake, and J.A. Lee. 1999. The effects of quantity and duration of simulated pollutant nitrogen depo­sition on root-surface phosphatase activities in calcareous and acid grasslands: a bioassay approach. New Phytol. 141:
433-442.
456
Botanical Studies, Vol. 51, 2010
Kirkham, F.W. 2001. Nitrogen uptake and nutrient limitation in six hill moorland species in relation to atmospheric nitrogen deposition in England and Wales. J. Ecol. 89: 1041-1053.
Liao, J.-X., X.-Y. Zou, Y. Ge, and J. Chang. 2006. Effects of
light intensity on growth of four Mosla species. Bot. Stud.
47: 403-408.
Lu, C.Q. and H.Q. Tian. 2007. Spatial and temporal patterns of nitrogen deposition in China: synthesis of observational data. J. Geophys. Res.- Atmospheres 112: D22S05, doi: 10.1029/2006JD007990.
Patterson, T.B., R.D. Guy, and Q.L. Dang. 1997. Whole-plant
nitrogen- and water-relations traits, and their associated trade-offs, in adjacent muskeg and upland boreal spruce species. Oecologia 110: 160-168.
Poorter, H. and O. Nagel. 2000. The role of biomass allocation in the growth response of plants to different levels of light, CO2, nutrients and water: a quantitative review. Aust. J.
Plant Physiol. 27: 595-607.
Qian, W.H., J.L. Fu, and Z.W. Yan. 2007. Decrease of light
rain events in summer associated with a warming environ­ment in China during 1961-2005. Geophys. Res. Lett. 34:
L11705, doi:10.1029/2007GL029631.
Rubio, G. and R.S. Lavado. 1999. Acquisition and allocation of resources in two waterlogging-tolerant grasses. New Phytol.
143: 539-546.
Suding, K.N., S.L. Collins, L. Gough, C. Clark, E.E. Cleland,
K.L. Gross, D.G. Milchunas, and S. Pennings. 2005. Func­tional- and abundance-based mechanisms explain diversity loss due to N fertilization. Proc. Natl Acad. Sci. USA 102: 4387-4392.
Teo, Y.H., C.A. Beyrouty, B.J. Norman, R.J. Norman, and E.E. Gbur. 1994. Nutrient supplying capacity of a paddy rice
soil. J. Plant Nutr. 17: 1983-2000.
Zhou, S.L. 1995. Systematics and evolution of the genus Mosla (Labiatae). Ph.D. dissertation. Institute of Botany, Chinese Academy of Sciences, Beijing, China. (In Chinese with English summay)
薺葶屬的兩個物種對壤氮和水的
曹前進1 王 猛1 葛 瀅1 張小川2 張建民1 常 傑1
1中國杭州浙江大學生命科學學院
2加拿大阿爾伯塔大學可更新資源系
由於氮沉降和降水的變異,同屬物種對生長條件改變的不同回應可能改變物種間的分化格局。我
們將石薺擊屬的兩個物種培養在不同的氮和水條件下。兩物種在淹水條件下對高氮供應表現出相似的正
面回應,預示著它們有潛在的機會在非常濕潤的生境拓展居群。但是,在中度濕潤和相對乾旱條件下, 與小花薺擊相比,小魚仙草不耐受高氮供應(小魚仙草的生物量被高氮抑制的程度比小花薺擊的更大) 卻更耐受限制氮(隨著水分供應的減少,小魚仙草的根生物量比被氮限制增加的程度比小花薺擊的更 大)。因此,在相對乾旱生境,土壤氮增加可能不利於小魚仙草的生長。
關鍵詞:生境;形態;小花薺擊;小魚仙草;氮沉降;分化。