Crop Science
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (17)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Brück, H.
Right arrow Articles by Sattelmacher, B.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Brück, H.
Right arrow Articles by Sattelmacher, B.
Agricola
Right arrow Articles by Brück, H.
Right arrow Articles by Sattelmacher, B.
Crop Science 40:120-125 (2000)
© 2000 Crop Science Society of America

CROP PHYSIOLOGY & METABOLISM

Effects of Phosphorus and Water Supply on Yield, Transpirational Water-Use Efficiency, and Carbon Isotope Discrimination of Pearl Millet

H. Brücka, W.A. Payneb and B. Sattelmachera

a Institute of Plant Nutrition and Soil Science, Univ. of Kiel, Olshausenstr. 40, 24118 Kiel, Germany
b Oregon State Univ., Columbia Basin Agric. Res. Center, P.O. 370, Pendleton, OR 97801 USA

hbrueck{at}plantnutrition.uni-kiel.de


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 Materials and methods
 NOTES
 Results and discussion
 REFERENCES
 
Several studies have identified low soil P and water availability as major constraints to pearl millet [Pennisetum glaucum (L.) R. Br.] production in semi-arid West Africa. To evaluate the effects of phosphate and water supply on yield, transpirational water-use efficiency (WUET), and carbon-isotope discrimination ({Delta}), two varieties of pearl millet were cultivated in pots in a glasshouse at the ICRISAT Sahelian Centre, near Niamey, Niger. Phosphate and water supply had significant effects on yield, WUET, and {Delta}. Compared with the control plants, which had adequate water and P availability, yield was reduced 34% by low water supply and 48% by low P supply. Under high P-supply, water stress increased WUET by approximately 37%. Under low P-supply, no effect of water supply on WUET was observed. Water stress increased {Delta} by approximately 0.6{per thousand} for low P plants, and 0.9{per thousand} for high P plants. Added P increased {Delta} by 0.3 to 0.4{per thousand}. WUET and {Delta} did not differ significantly between varieties. Differences in {Delta} between green and necrotic leaves were found within both P treatments under low water supply. We attribute changes in {Delta} to changes in the ratio of external to internal concentration of CO2, (pi/pa), leakage rates of CO2 out of bundle-sheath cells, respiration rates, or chemical composition of the plant material.

Abbreviations: DAS, days after sowing • DM, dry matter • Rubisco, ribulose-1,5-bisphosphate carboxylase-oxygenase • PEPC, phosphoenolpyruvate carboxylase • WUET,S, transpirational water-use efficiency, based on shoot dry matter • WUET,SR, transpirational water-use efficiency, based on shoot and root dry matter • {Delta}, carbon-isotope discrimination • pi/pa, ratio of external to internal concentration of CO2


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
Materials and methods
 NOTES
 Results and discussion
 REFERENCES
 
PEARL MILLET production systems of the West African Semi-Arid Tropics (WASAT) are characterized by low soil productivity and chronically low water supply (Payne, 1997). On-farm yields of pearl millet are usually very low (400–600 kg ha-1 of grain), but fertilization may increase yield to as much as 2500 kg ha-1 (McIntire and Fussell, 1989; Christianson et al., 1990). Generally, low yields are accompanied by low evapotranspirational water-use efficiency, which is brought about by a combination of low leaf area index and, for many environmental stresses, changes in WUET.

The dependence of WUET on water- and nutrient supply has been demonstrated for various C3 plant species, including wheat (Triticum aestivum L.; Farquhar and Richards, 1984), barley (Hordeum vulgare L.; Hubick and Farquhar, 1989), peanuts (Arachis hypogaea L.; Hubick et al., 1986), and sunflower (Helianthus; Virgona and Farquhar, 1996). Although less pronounced, WUET of C4 plants also appears to be affected by water and nutrient supply (Schenk and Barber, 1979; Onken and Wendt, 1989; Payne et al., 1992, 1995). Changes in WUET reflect changes in stomatal conductance and/or internal capacity for CO2 fixation, the latter being affected by enzyme activity (von Caemmerer et al., 1997) and plant nutrient status (Payne et al., 1992; Ranjith et al., 1995).

A correlation has been observed between discrimination against 13C ({Delta}) and WUET. According to Farquhar (1983), {Delta} is mainly caused in C3 species by (i) fractionation due to CO2 diffusion (a); (ii) changes in stomatal resistance or assimilation rate, which affect the ratio of internal to ambient concentration of CO2 (pi/pa); and (iii) fractionation (b3) by the enzyme ribulose-1,5-bisphosphate carboxylase-oxygenase (Rubisco).

For C4 species, {Delta} is further affected by initial fixation of CO2 (b4) by the enzyme phosphoenolpyruvate carboxylase (PEPC), and "leakiness" ({Phi}) of CO2 from bundle-sheath chloroplast cells. Compared with the high discriminative capacity of Rubisco, discrimination by PEPC is relatively small. Farquhar (1983) has used the following equation to describe {Delta} in C4 plants:

(1)

For C3 species, a decrease in pi/pa has been associated with lower {Delta}. However, Farquhar (1983) postulated that the relationship between pi/pa and {Delta} can be either positive or negative for C4 plants, depending on the magnitude of {Phi}. Variation in {Delta} among C4 plants has been attributed to morphological features of bundle sheath cells that can result in different rates of leakage (Hattersley, 1982). Salt stress or changes of leaf water status, which decrease C3 pathway activity, have been cited as environmental factors which affect {Phi} (Bowman et al., 1989). On the other hand, {Phi} appears to be little affected by leaf temperature, irradiance, or variation in CO2 supply (Henderson et al., 1992). Plant nutrient status may also affect {Phi} (von Caemmerer et al., 1997).

Carbon isotope discrimination in C4 plants is likely to be like that in C3 plants, which varies according to sampling technique and among organs (Hubick and Farquhar, 1989; Condon and Richards, 1992) and plant parts of different maturity (Knight et al., 1994). For example, Hubick et al. (1990) found differences among leaf positions in sorghum [Sorghum bicolor (L.) Moench].

Carbon-isotope discrimination has been successfully used to identify varietal differences in WUET in several C3 plants (Condon and Richards, 1992; Lu et al., 1996; Kirda et al., 1992; Sayre et al., 1995). In C4 plants, however, the CO2 concentrating mechanism might mask the potentially high discriminative effects of Rubisco. Therefore, utility of carbon-isotope discrimination as a feasible selection tool for greater WUET in pearl millet and other C4 crop species is less clear, although positive evidence of genotypic variation in sorghum (Henderson et al., 1998) has been reported.

To our knowledge, there is no information on the effects of the two major environmental constraints in WASAT on carbon-isotope discrimination in pearl millet, or on the relation between {Delta} and WUET. The purpose of this study was to compare WUET and 13C measurements of two pearl millet varieties under varying water- and phosphate supply.


   Materials and methods
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
NOTES
 Results and discussion
 REFERENCES
 
A pot experiment was conducted in 1994 in a glasshouse at the ICRISAT Sahelian Centre in Niger, West Africa (13°15' N latitude, 240 m altitude). Treatments consisted of two varieties, two P levels and two water levels, with five replications. The pearl millet cultivars were Sadore, a widely used landrace in this region of Niger, and ICMV IS 89305, a promising composite developed by ICRISAT. To reduce systematic effects due to potential temperature and humidity gradients, pots of treatments were arranged along the draft of electrical air humidifiers and systematically rearranged after every watering event. Pots had a radius of 0.20 m at the surface, and a volume of 20307 cm3. Pots were filled with 26 kg air-dried soil, and sealed with plastic sheets. Sheets were afterwards covered with a thin layer of soil such that almost all water loss over time could be attributed to transpiration. The water treatments were based on the assumption that 16% of volume represents field capacity. Low and high water supplies (50 and 100% of field capacity, W1 and W2, respectively) were maintained by weighing the pots every 2 to 6 d (depending on plants' water use) and, on the basis of weight loss, rewatering them to W1 and W2.

For basal dressing of nutrients, for each pot, fertilizer was mixed with 2 kg of soil. This mixture was covered with 6 cm of unfertilized soil. Phosphate fertilizer (powdered single-super phosphate, SSP) was applied at rates of 69 mg P pot-1 and 920 mg P pot-1, equivalent to 8.3 kg and 110.9 kg SSP per ha. A total of 5.5 g potassium (as KCl) and 2.1 g magnesium (as MgSO4) was applied before planting and at 30 d after sowing (DAS). A total of 5 g of N was applied as calcium ammonium nitrate (CAN) before planting, and at 30 and 49 DAS. Symptoms of Zn deficiency within the +P treatments ceased after fertilization with Fertrilon (BASF, Germany).1

Seeds were planted on 16 July. Thirty seeds were placed into a single hill in each pot. All pots were watered with 2 L of water and sealed. After emergence, holes were made to allow plants to grow. At 19 DAS, plants were thinned to three plants per pot. At 52 DAS, plants were treated with Bifenthrin—[1{alpha}, 3{alpha}(Z)]-(±)-3-(2-chloro-3,3,3-trifluoro-1-propenyl)-2,2-dimethylcyclopropanecarboxylic acid (2-methyl[1,1'-biphenyl]-3-yl)methyl ester— because of heavy attack of spider mites.

Mean night temperature ranged from 26 to 28°C, and mean day temperature varied between 34 and 40°C. The mean temperature increased towards the end of the experiment. Relative humidity varied between 50 and 80%.

Plants of the +P treatment were harvested at DAS 76, and those of the -P treatment at DAS 81. Shoot parameters recorded were number of leaves and tillers and height of main stem. Stems, leaves, and heads were separated. Roots were gently washed over a sieve (mesh size 0.6 mm). Fresh weight (for shoot) and dry weight for root and shoot were determined after drying the samples at 60°C. Transpiration ratios were calculated both in terms of shoot dry matter (WUET,S) and total dry matter (WUET,SR) as biomass (g) per water transpired (l). The total amount of water transpired was corrected for the water remaining in the soil at the date of harvest.

The three youngest leaves were harvested and separated into "green" and "necrotic" tissue, and used for determination of 13C discrimination. All plant material was ground and thoroughly mixed. Three replications of each sample were analysed for isotopic composition with a Carlo Erba elemental analyser (Carlo Erba, Milan) interfaced to a Delta S ratio mass spectrometer (ThermoQuest, Egelsbach, Germany). The internal standard was CO2, which was calibrated against a urea reference (isotopic ratio: -49.44{per thousand}). The working standard was a powdered, commercially purchased C4 sugar, of which two samples were combusted for all 20 samples. If the standard deviation for the sample was > 0.16{per thousand}), two replications of the sample were measured again. Results were calculated as {Delta} defined by

(2)
where {Delta}p and {Delta}a are the isotopic composition with respect to Pee Dee belemnite of the plant material and air, respectively. We assumed {Delta}a to have a value of -8{per thousand}, a value which is widely used for free atmospheric CO2 (Farquhar et al., 1989).

All data from this three factorial experiment with fixed effects were analysed with SAS procedure PROC GLM (SAS, 1996). Data were checked for normal distribution and, where necessary, log transformed.


   Results and discussion
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
 NOTES
 Results and discussion
REFERENCES
 
There were significant differences in the amount of water transpired among the four treatments (Fig. 1) . Cumulative transpiration ranged from 10 for the -P-W treatment to more than 30 l for the +P+W treatment. Plants of the +P-W and -P+W treatments had almost identical amounts of transpiration. Phosphate and water supply had significant effects on all shoot parameters (Table 1) . Except for leaf dry matter, the interaction between phosphate and water supply was significant for all shoot parameters. Differences between varieties were only detected for leaf dry matter, for which Sadore had significantly greater dry matter. As indicated by non-significant VxP and VxW interaction, the response to changes in water and phosphate supply was very similar for all shoot parameters for both cultivars.



View larger version (14K):
[in this window]
[in a new window]
  		Fig. 1 Cumulative water consumption (L) of pot grown pearl millet from 28 DAS to date of harvest as affected by low (-) and high (+) supply of phosphate and water. Results are pooled for two varieties

 

View this table:
[in this window]
[in a new window]
  		Table 1 Mean square values and level of significance of ANOVA for dry matter of different plant organs at harvest, transpirational water-use efficiency (WUET), and carbon-isotope discrimination of necrotic leaves ({Delta}-necrotic). WUE is based on shoot dry matter (WUET,S) or total biomass (WUET,SR)

 
Significant differences were found for root DM production due to water supply, but there was no effect associated with P-level (Table 2) . Payne et al. (1991) reported significant effects of phosphate supply on root dry matter of pearl millet for six consecutive harvests, but pots were much larger (volume 75 L, height 0.75 m) than those of this experiment. For pot-grown plants, treatment effects on root parameters depend on pot size (Barrett and Gifford, 1995). In the present experiment, root DM of the +P-W treatment was only slightly less than root DM of the +P+W treatment. The smallest root DM was observed in the -P-W treatment.


View this table:
[in this window]
[in a new window]
  		Table 2 Dry matter of shoot, leaves, stem, heads, roots and total biomass as affected by low (-) and high (+) supply of phosphate (p) and water (w) s.e.: standard error of estimates, varieties Sadore (Var 1) and IVMV IS 89305 (Var 2), Level of significance: {alpha} = 0.05. Varietal differences were significant for leaf DM only. Data for stem, head, root and total dry matter were pooled across varieties

 
Shoot DM of the +P-W and –P+W treatments were 34 and 48% less than that of the +P+W treatment, respectively (Table 2). Shoot and total DM of the -P+W was smaller than that of the +P-W treatment, but stem DM was similar. Compared with low water supply, P-deficiency had greater effects on dry matter formation of heads and leaves. Data in Table 2 illustrate the strong response of millet to improved water and phosphate supply, and correspond well to published data on shoot yield and shoot parameters of pearl millet (Payne et al., 1991).

WUET was clearly affected by water and phosphate supply (Table 1). As indicated by significance of the PxW interaction term, changes in WUET due to water supply depended on P level. Varietal effects were not detected. The trends were essentially the same when root mass was included (WUET,SR), but the interaction between P supply and water was more pronounced. The small, but significant effect of VxPxW was of only minor importance in explaining observed effects in this experiment. Under high P supply, WUET increased because of to lower water supply by roughly 27% for WUET,S and 30% for WUET,SR (Table 3) . Under low P supply, no effect of water supply on WUET was detectable. Under high water supply, WUET,S was greater under conditions of low phosphate supply (-P+W) compared with +P+W. This effect was more pronounced for WUET,SR. Under low water supply, however, greater WUET was realized under improved P supply. These findings are partially in contrast to Payne et al. (1992), who reported significant increases in WUET due to phosphate application irrespective of water supply. Because there is no supplementary information, e.g., from harvests at earlier date, and only two P levels were used, this discrepancy remains unexplained.


View this table:
[in this window]
[in a new window]
  		Table 3 Transpirational water-use efficiency of pearl millet based on shoot dry matter (WUET,S) or total dry matter including roots (WUWT,SR). s.e. is standard error of estimates for test of PxW means. Significance at 0.05 probability level

 
Carbon isotope discrimination of the necrotic leaves was significantly affected by both phosphate and water supply (Table 1). As was the case for WUET, a major part of variation was attributed to water supply. In contrast to shoot yield and WUET data, the interaction between phosphate and water was not significant. Discrimination against 13CO2 was increased by 0.36{per thousand} by low P supply, and by roughly 0.75{per thousand} by low water supply (Table 4) .


View this table:
[in this window]
[in a new window]
  		Table 4 13C discrimination ({Delta}) of necrotic leaves as influenced by phosphate and water supply. Comparison of means is based on main-effects and pooled for varieties. Minimum significant difference (MSDTukey): 0.14

 
Greater discrimination against 13CO2 was found in necrotic leaves compared with green leaves (Table 5) . This difference was significant for both water stress treatments (+P-W and -P-W). The effect was most obvious for the -P-W treatment. Comparable effects have been reported for field grown sorghum by Hubick et al. (1990), who found that {Delta} differed among leaf positions within the same plant.


View this table:
[in this window]
[in a new window]
  		Table 5 Influence of phosphate and water supply on differences in 13C discrimination between necrotic and green leaf material (difference = {Delta}green - {Delta}necrotic) of pearl millet as affected by low (-) and high (+) supply of phosphate and water. Significance is based on t-tests of pairwise comparisons. For the -P+W treatment, no green leaf material was present

 
As shown in Fig. 2 , the relationship between WUET and carbon-isotope discrimination in pearl millet depended on P supply. For the +P treatments, {Delta} increased under low supply of water (increasing WUET). The inclusion of root DM improved the correlation between {Delta} and WUET from R2 = 0.744 for WUET,S to 0.802 for WUET,SR



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 2 Relationship between transpirational water-use efficiency and carbon-isotope discrimination based on shoot dry matter (WUES) and total biomass (WUESR) as affected by low (-) and high (+) supply of phosphate and water. Discrimination ({Delta}) is calculated according to Eq [2]. If present, {Delta}13C of green leaf material was used for calculating {Delta}. Linear regression is pooled for both varieties and R2-values are calculated for +P-treatments (solid line) and all treatments excluding treatment -P-W (broken line)

 
Under limited phosphate supply, data of the -P+W treatment partly aggregated near the observed linear relationship for data of the +P-treatments. However, the relationship between WUET and {Delta} was less clear if -P+W was included into the regression. Interestingly enough, discrimination against 13C tended to be higher for similar values of WUET under conditions of low P supply (-P+W) compared with conditions of adequate P (+P+W).

From Eq. [1], it is evident that, in C4 plants, various factors affect discrimination against 13CO2. In the case of inadequate water supply, stomatal conductance and pi/pa decreases (Squire et al., 1984; Madhavan et al., 1991). Henderson et al. (1992), for example, showed proportional changes in {Delta} when pi/pa varied from 0.2 to 0.6 for sorghum using short-term gas exchange measurements.

According to Farquhar (1983), the negative slope of regression for the +P treatments (Fig. 2), indicates that {Phi} values were in the range of ~0.21, which is in good agreement with published data. When data of the –P+W treatment were included in regression analysis (Fig. 2; dotted line), the linear regression gave an intercept value of -4.4 and a less negative slope, suggesting increased leakage of CO2 from the bundle sheaths cells. In principle, an increase in leakiness represents a branch point in the metabolic pathway (O'Leary, 1981) and would lead to greater overall carbon discrimination because of the greater discrimination by Rubisco.

The extent of recycling of CO2 is influenced by the coordinated control of enzyme activities, namely Rubisco and PEPC. Nutrient deficiency can affect this internal regulation by inhibiting phosphorylation of PEPC, and has resulted in a shift from C4 to C3 photosynthesis in sorghum (Bakrim et al., 1993), and decreased Rubisco activity relative to PEPC activity (under conditions of low nitrogen supply) in sugarcane (Ranjith et al., 1995). In both cases, {Delta} was affected. We speculate that pertubations of enzyme activities of PEPC and Rubisco under low phosphate supply caused {Delta}-values of the -P+W treatment to be higher compared with treatment +P+W. We attribute this effect to increased leakage.

Plants of the -P-W treatment exhibited a distinct departure from the linear trend of WUET and {Delta} (Fig. 2). Based on Farquhar's (1983) model (Eq. [1]), this implies that {Phi} increased substantially under low P and low water supply. This model was mainly verified by short-term gas exchange measurements, from which discrimination ({Delta}g) data were plotted against pi/pa. By contrast, in our study, no gas exchange measurements were made, and carbon-isotope discrimination of leaf dry matter ({Delta}d) was plotted against WUET to compare treatment effects.

WUET is itself a variable that depends on the manner of calculation. For our data, consideration of root dry matter somewhat increased the correlation between {Delta}d and WUET (see WUET,S versus WUET,SR; Fig. 2) for treatment –P+W, but not for treatment –P-W. This observed discrepancy between WUET and {Delta}d implies that either leakage increased considerably or that disrimination, if based on {Delta}d, is affected by processes not considered in Farquhar's model. A comparison of four C4 species revealed a low correlation between {Delta}d and pi/pa (Henderson et al., 1992). Possibly, non-photosynthetic processes, such as respiration, could have influenced discrimination.

As is evident from Table 5, {Delta}d depends on the physiological age of plant material. This aspect includes two components, namely (i) treatment effects on senescence, and consequent shifts in chemical composition of the plant material, perhaps reflecting changes in partitioning of biomass, and (ii) treatment effects on post-photosynthetic discrimination. It is possible that physiological age has substantial influence on the extent of discrimination. Henderson et al. (1998), for example, found that {Delta}d increased by 0.6{per thousand} between harvests 44 DAS and 58 DAS in sorghum. Our sampling did not allow for further investigation on the relative importance of senescence and post-photosynthetic disrimination.

Only a small fraction of total variance could be assigned to varietal differences. Our limited results indicate small differences in leaf traits associated with greater WUET in pearl millet. However, we compared only two varieties with almost the same maturity, and with comparable yield potential (Buerkert, 1995). The more extensive work of Henderson et al. (1998) identified genotypic variation in sorghum under both controlled and field conditions. Additional work would be required to elucidate the extent of genotypic variability in WUET in pearl millet, and whether {Delta} has potential as a selection tool.


   ACKNOWLEDGMENTS
 
This work was supported by ICRISAT, Sahelian Centre, Niamey, Niger, Section Plant Physiology and funded in part by the DAAD, grant No. 413-afr-3-hi. We thank Dr. Anand Kumar, Pearl millet improvement programme of ICRISAT for kindly providing seed material.


   NOTES
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
 NOTES
Results and discussion
 REFERENCES
 
1 Mention of trade names does not constitute any endorsement. Back

Received for publication February 1, 1999.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
 NOTES
 Results and discussion
 REFERENCES
 




This article has been cited by other articles:


Home page
 Agron. J.Home page
W. A. Payne
Optimizing Crop Water Use in Sparse Stands of Pearl Millet
Agron. J., September 1, 2000; 92(5): 808 - 814.
[Abstract] [Full Text]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (17)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Brück, H.
Right arrow Articles by Sattelmacher, B.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Brück, H.
Right arrow Articles by Sattelmacher, B.
Agricola
Right arrow Articles by Brück, H.
Right arrow Articles by Sattelmacher, B.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Vadose Zone Journal
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome