Springer 2005

Plant and Soil (2005) 272: 233–244 DOI 10.1007/s11104-004-5048-9

Genotypic differences in manganese efficiency: field experiments with winter barley (Hordeum vulgare L.) C.A. Hebbern1, P. Pedas1, J.K. Schjoerring1, L. Knudsen2 & S. Husted* 1

Plant and Soil Science Laboratory, The Royal Veterinary & Agricultural University, 40 Thorvaldsensvej, DK-1871 Frederiksberg C, Copenhagen, Denmark. 2Danish Agricultural Advisory Service, National Centre Aarhus, Udkaersvej 15, DK-8200 Aarhus N. 3Corresponding author* Received 2 July 2004. Accepted in revised form 19 October 2004

Key words: fluorescence, genotypes, Mn deficiency, winter barley

Abstract Eight different winter barley genotypes were grown in a plot experiment in northern Denmark, on soil where Mn deficiency had been previously demonstrated, to assess field performance and grain yield under two treatments: with foliar Mn fertilisation and without foliar Mn fertilisation. A high Mn efficiency was demonstrated for two of the genotypes and a low Mn efficiency was observed for one of the genotypes in three consecutive years; highly variable results were obtained for other genotypes. Mn efficiency based on grain yields could not be related to whole shoot Mn concentration, biomass or Mn uptake at various vegetative growth stages. Chlorophyll a fluorescence analysis was able to differentiate between the two Mn treatments up to 7 weeks after the last Mn application, whereas Mn concentrations, shoot biomass and Mn accumulation could not. It was not possible to fully alleviate Mn deficiency by repeated foliar spraying on the control plots. Given the growing conditions at the trial site and the low Mn in plant tissue, it is possible that Mn deficiency was extreme enough that Mn efficiency mechanisms broke down. Abbreviations: Mn – manganese; GS – growth stage; Fo – minimum fluorescence; Fm – maximum fluorescence; Fv – variable fluorescence. Introduction Mn deficiency imposes a significant limitation on crop production in many places in the world. High-pH, well-aerated soils are particularly at risk, and correction of deficiency can be difficult to achieve, requiring multiple applications of foliar fertiliser. Danish soils, especially those formed on old marine sediments, are frequently prone to Mn deficiency, in particular affecting winter forms of barley and wheat. Recent years have seen an increase in Mn deficiency problems on formerly problem free soils, possibly as a result of cultivation practices such as liming or * E-mail: [email protected]

slurry application, with subsequent increases in soil pH and soil P (Husted et al., 2004b). Under such conditions, plant available Mn declines (White and Zasoski, 1999), and now Mn deficiency has become the foremost plant nutritional problem in Denmark. It had also been observed that certain genotypes of cereals grown in Denmark yielded markedly better on the low plant available Mn soils than others and were likely therefore to be Mn efficient. Mn efficiency has been defined as a genotype’s ability to produce high yield in a soil whose Mn content is limiting for a standard genotype (Ascher-Ellis et al., 2001). Pioneering work in differences in Mn efficiency between genotypes has been undertaken in Australia

234 (Graham, 1988; Rengel et al., 1994; Sadana et al., 2002), where genotypes, climate and soil conditions are highly different to those of northern Europe. There, genotypes were found to differ markedly in their responses to limited soil Mn. For example barley genotype WI 2585 had a Mn efficiency of only 17% when comparing relative yield between fertilised and unfertilised plots, while the corresponding efficiency of WA 73S276 was 90% in a field trial in Southern Australia (McDonald et al., 2001). To our knowledge, the present paper represents the first study on winter barley genotypic tolerance to low available Mn in northern Europe. The physiological mechanisms that underlie Mn efficiency are not fully understood, but candidates include translocation to the seed (Khabez-Saberi and Graham, 2002), differences in uptake and compartmentation (Huang et al., 1994), improved acquisition of Mn from soil via root exudation (Gherardi and Rengel, 2004; Graham, 1988; Huang et al., 1994; Pearson and Rengel, 1997; Rengel, 1999) and favourable plant root interactions with Mn reducing microorganisms (Posta et al., 1994). There is a necessity for mechanistic studies into the basis of Mn efficiency. Knowledge of the underlying mechanism would improve plant breeding by allowing selection for specific traits and accelerate screening tests for the trait. At present the difference in final yields in plots with and without Mn fertiliser is used effectively to determine Mn efficiency (Graham et al., 1992;

McDonald et al., 2001). Agronomic studies that are both practically and economically relevant to local agricultural practices will assist in improvement and management of Mn deficiency. The field trials detailed in this paper investigated differences in Mn efficiency of various winter barley genotypes commonly grown in Denmark, based on the relative differences in final yield between two treatments on a Mn limited soil. We further examined numerous physiological parameters throughout the growing season of one of the trials. Three years of experimentation are presented. Local agricultural practice was followed for the management of the trials, to allow comment on the applicability of Mn efficiency to local crop production. Genotypic differences in Mn efficiency were positively established for barley genotypes over the repeated trials, although there was considerable variability between the trials. Difficulties in assessing Mn efficiency in field trials are presented and discussed.

Materials and methods Trials were undertaken over 3 years, as detailed in Table 1. In 2001, demonstration plots were sown to compare differences between genotypes based on visual assessments only. This was expanded in 2002 to include eight different genotypes at two locations, Vraa and Aarre, and relative yield differences were used to rank the genotypes by Mn efficiency. The trial was repeated again in

Table 1. Three years of field trials with the analyses performed each year 2001

2002

2003

Four genotypes Two Mn levels Demonstration trial: visual observations only Location: Varde (32;6165794;89467618)

Eight genotypes Two Mn levels Yield comparisons Locations: Vraa (32;561000;6359500) Aarre(32;474395;6157158)

Eight genotypes Two Mn levels Two vegetative samplings: Mn concentrations Biomass Seed Mn Chlorophyll a fluorescence Yield comparisons Location: Stoevring (32;558578;6310172)

The 2001 field trial was a preliminary investigation into Mn efficiency in Denmark. In 2002 the trial was expanded to two locations and eight genotypes with yield comparisons between Mn treatments. The 2003 trial is the main focus of this article and is a repeat of the 2002 trial, expanded to include more analyses throughout the growing season.

235 2003 with eight genotypes of winter barley, four of which were new to this study, at Stoevring in north Jutland. The trials were established at sites where the soil had a history of Mn deficiency and a high likelihood that deficiency would recur. Soil samples from the Stoevring site contained 3.5 (DTPA extractable) and 44.7 (HNO3 extractable) mg Mn kg)1 soil (Husted et al., 2004a). Soil pH was 6.5 (0.01 M CaCl2). The methods section that follows pertains to the 2003 trial only. Experimental design Eight winter barley genotypes were sown, along with a buffer plot of the genotype Carola, which was being grown commercially in the surrounding field. Plots were 30 m2, and arranged in eight blocks of eight genotypes in a fixed order, alternating between Mn-fertiliser treated and Mnuntreated blocks (Table 2). This design produced four replicate plots of each genotype at each treatment, and each Mn treated block was faced directly opposite by an untreated block. The experimental site was treated the same way as the surrounding field, to match the agronomic conditions and practices currently used in this part of Denmark. Plants were sown at a concentration of 325 seeds m2 on 8 September 2002. The Mn-treated blocks were sprayed with 2.5 kg MnSO4 and 0.2 L ha)1 Lissapol Bio as surfactant (Zeneca, Denmark) in the autumn. This was repeated 21 days later, then once more at the Table 2. Experimental design of field trials Plot layout in each block Buffer plot Genotype 1 Genotype 2 Genotype 3 Genotype 4 Genotype 5 Genotype 6 Genotype 7 Genotype 8 Buffer plot

Block layout in field

Mn + 1

Mn ) 1

Mn ) 2

Mn + 2

Mn + 3

Mn ) 3

Mn ) 4

Mn + 4

Eight genotypes were planted in 8 plots in each block with a buffer plot between the blocks (left-hand column). Four blocks of each treatment, Mn+ and Mn) were arranged as shown in the right-hand column. The numbers designate block pairs.

beginning of the spring growing season on April 7. Additional fertiliser treatment for all plots was 225 kg NH4NO3 ha)1 on March 20, and 18 t animal manure ha)1 on April 20. Plants from the plots were sampled during the growing season and biomass, chlorophyll fluorescence and Mn concentrations were analysed from samples collected on 1 May 2003 at Zadok’s growth stage 30, and on 1 June at growth stage 45 (Zadoks et al., 1974). The final crop harvest was taken on 23 July, and grain was subsequently analysed for Mn concentrations and crop yield. Sampling The plots were sampled by cutting five 1 m rows of plants (corresponding to 0.6 m2) from each plot and placing into polypropylene bags for transportation to the laboratory. Plants from different plots were always kept separate to avoid any contamination by Mn fertiliser residues. All samples were stored at +5 C for not more than 3 days prior to washing. After removal from cold storage, the samples were weighed to determine fresh weight. The small amount of material collected at the first measurement date allowed all the material to be processed for analysis. At the second date, it was necessary to take a subsample. Therefore, after weighing the initial sample, the plants were spread out onto a clean bench surface and alternate stems (plants) were selected and frozen at )20 C. The rest of the material was discarded. Grain yields were established by weighing at harvest using a plot harvester. Subsamples of grain were collected from each plot and stored in polypropylene bags. Mn efficiency The relative yield differences at growth stages 30, 45 (vegetative, based on comparisons of biomass between the two Mn treatments) and growth stage 100 (harvest, comparison of grain yield) are expressed in terms of Mn efficiency (Ascher-Ellis et al., 2001): Mn efficiency ¼ 100
ðGY=GYþÞ; where GY is the grain yield or vegetative yield, and + and ) designates sprayed and unsprayed treatments, respectively.

236

Field measurements of chlorophyll fluorescence were made using a hand-held portable fluorescence detector (Handy Plant Efficiency Analyser, Hansatech Instruments, Kings Lynn, UK). Three measurements were made on three different youngest emerged leaves per plot. Two each of the Mn treated and Mn-untreated blocks were used in this part of the study. The leaves were dark-adapted for 30 min using the Hansatech leaf-clips. Fluorescence measurements were recorded by measurement for 10 s after pulse illumination with 3000 lmol photons m)2 s)1 at a wavelength of 650 nm. Fluorescence transients were plotted and analysis performed on Biolyzer software (R. Rodriguez & R. Strasser, University of Geneva, Switzerland).

to obtain fragments of less than 1 mm. The samples from the second harvest date contained a high proportion of stem to leaf. It was necessary to pulverise these samples using liquid nitrogen in order to obtain an equivalent level of homogeneity. A subsample was removed from each, placed into paper bags and freeze-dried. Grain was collected by a plot harvester and subsampled in the field. Approximately 100 g of each sample was homogenised in a titanium grinding mill with a titanium rotor (Retsch motor ZM1, F. Kurt Retsch GmbH, Haan, Germany,) to a fine powder, and the ground material was placed in a freeze-drier (Christ Alpha 2–4, Martin Christ, GmbH, Osterode, Germany) until completely dry. Accuracy of sampling on the homogenised material was tested by taking seven replicates of one freeze-dried subsample from each harvest date, including the harvested grain, and digesting it separately for ICP-MS measurement. The coefficients of variance for Mn concentrations were: 6.8% at growth stage 30; 8.0% at growth stage 45; and 14.0% in harvested grain.

Sample surface decontamination

Digestion of plant material

To eliminate the potential effects of Mn fertiliser residue, dust born contamination or soil disrupting the ICP-MS analysis for Mn concentrations, all samples were carefully cleaned. The initial step was a rinse for 1 min under double-distilled water to remove gross contamination and dirt. This was followed by transferring the plant material to a 4 L bucket of MilliQ water containing two drops of Tween 20 (Sigma–Aldrich). The bucket was sealed and shaken 20 times, drained, refilled with MilliQ water and shaken again. This was repeated four times, sufficient to remove all dirt and detergent from the plant material. Following decontamination, the plant material was shaken to remove excess water, placed into polypropylene bags and stored at )20 C. The decontamination procedure was based on Husted et al., (2004a,b), with minor modifications.

About 0.2 g of freeze-dried, homogenised plant material was digested in 70 mL polyethylene high-density vials (Capitol Vial Corp, Fultonville, NY, USA) on a graphite heating block (Mod Block, CPI International, Amsterdam, Holland). A modification of the EPA method 3050 B was used, as follows. About 5 mL of 35% trace purity HNO3 (J.T Baker Instra-Analyzed Reagent) was added to the dried material and the samples were boiled for 15 min. After cooling, 2.5 mL 70% HNO3 was added and the samples were reheated for 30 min. After cooling, 2.5 mL H2O2 (Extra-Pure, Riedel DeHae¨n, Selze, Germany) was added and the samples reheated until the reaction ceased. During the digestion the samples were covered with plastic watch glasses. Following cooling, the samples were diluted to 50 mL with ultra pure water (MilliQ Element, Millipore, Massachusetts, USA), giving a final acid concentration in the samples of 1.75% HNO3. Blank samples were included in each digestion run. Accuracy of the ICP-MS analysis was determined using apple leaf certified reference material (standard reference material 1515, National Institute

The plots were compared on a pair-wise basis. The plots used in this manner are found opposite to each other in the field trial design (Table 2). Then, the mean was of four efficiency values was calculated. Fluorescence measurements

Sample preparation for ICP-MS The plant material was homogenised by crushing the frozen sample within the polypropylene bag

237

Figure 1. Demonstration plots, Varde (Jutland, Denmark), 2001, where plants were grown on a soil low in plant available Mn. Plots from left to right: Isolde, Vanessa and Antonia. The soil in the left side of each plot had been compressed and this subsequently improved growth in the genotypes Isolde and Antonia.

of Standards and Technology, Gaithersburg, MD, USA). The digestion protocol and final dilution step were carried out in a class 100 Teflon coated laminar flow bench (KR-170s Biowizard, Kojair Tech Oy, Vilppula, Finland). ICP-MS analysis Elements were quantified using external calibration (P/N 4400 ICP-MS, Calibration Standard, CPI-International, Amsterdam, Holland). The ICP-MS (Agilent 7500 C, Agilent Technologies, Manchester, UK) was configured with the octopole reaction system to reduce polyatomic interferences and increase accuracy. Data analysis Data from the ICP-MS was accepted when accuracy was established to be better than 90% of the certified reference material value. Statistical analysis was undertaken using SAS (SAS version 8.2, SAS Institute, Cary, NC, USA) for variance analysis and Student’s t-test for comparison of means. The effect of Mn foliar applications on the growth of genotypes and chlorophyll fluorescence at different growth stages were analysed with two-way ANOVA. SignificancewastestedattheP < 0.05level.

Results It had been noted in preliminary field trials in 2001 and 2002 that the genotype Antonia grew

poorly on soils with low plant available Mn (Figure 1). Visual assessments of plant growth in early spring in 2003 (Figure 2 and Table 3) indicated that it had poor over-wintering survival, with a loss of 30% of seedlings. Antonia and Escape were the most affected genotypes at this stage. At growth stage 30 (formation of tillers) all genotypes growing without Mn foliar fertiliser displayed symptoms of moderate deficiency. The plants that had received foliar Mn also showed deficiency symptoms. However, it was much less pronounced than observed for the untreated plots. First vegetative sampling Mn concentrations in the untreated plots (Figure 3b) at growth stage 30 were all below 11 mg Mn kg)1 DW. Barley plants at this stage (Reuter et al., 1997) are considered to be deficient below 13 mg Mn kg)1 DW. The plots that received Mn treatment were all significantly higher in Mn concentration than the untreated plots by at least 50%. The genotypes Antonia and Cleopatra responded least to foliar Mn in terms of Mn concentration. Only three of the genotypes, Antonia, Escape and Menhir, had a significantly increased biomass as a result of Mn application (data not shown). Expressed as total accumulated Mn (Mn concentration · biomass, Figure 3a), the untreated plots all accumulated around 2.5 mg Mn m)2. There was more variability among the treated plots however, and the genotypes Escape, Clara and Vanessa had the highest uptake of Mn (ranging from 13 to 23 mg Mn m)2 for these three genotypes). Second vegetative sampling By the second sampling stage, the genotypes and treatments could no longer be clearly distinguished from each other in terms of Mn concentration (Figure 3d). However, differences between the two treatments were maintained in measurements of shoot biomass, with the exception of the genotypes Ludo and Cleopatra (data not shown). Only the genotype Ludo had a significantly higher Mn concentration when comparing the treated and untreated blocks (Figure 3d). When compared with published Mn deficiency thresholds

238

Figure 2. Differences between Mn treatments in plant growth at growth stage 30. Treated plots had received three previous foliar applications of MnSO4. (a) Antonia, Mn treated; (b) Antonia, untreated; (c) Vanessa, Mn treated; (d) Vanessa, untreated. Trial location: Stoevring 2003.

Table 3. Overwintering of plants in the 2003 field trial at Stoevring Overwintering (% survival) as 24 Genotype Clara Ludo Cleopatra Carola Vanessa Menhir Escape Antonia

Mn) 100 98 98 100 98 96 78 85

Mn+ 100 100 100 100 100 100 100 93

Plots were assessed for overwintering survival based on the percentage of survived plants. Assessment made at the beginning of spring growth.

for this growth stage (at <5 mg Mn kg)1 DW, plants are deficient for Mn, whereas 25 mg Mn kg)1 DW represents adequate nutrition; Reuter and Robinson 1997), then regardless of treatment, all the plots were close to deficient for Mn (concentrations ranged from 6 to 9 mg Mn kg)1 DW). Ludo had accumulated the most Mn (Figure 3c) in both treated and untreated plots. The untreated plots accumulated similar

total amounts of Mn, with the only significant difference between genotypes Clara and Menhir. More variability between the genotypes was found in the treated plots. There were significant increases in accumulated Mn in the genotypes Ludo, Carola, Vanessa and Escape, with the highest response in Ludo, which increased from 14 to 23 mg Mn m)2 on application of Mn. Seed analysis Seed Mn was measured both on the seed used to sow the field trial and in the seed harvested from the plots at the end of the trial (Figure 4). The Mn concentration for Antonia seed used to sow the trial was very high, as the seed had been coated with a commercial Mn treatment. After digest, the concentration of Mn was 214 ± 38 mg kg)1 DW in this sample. At harvest, all genotypes had a reduced seed concentration of Mn compared to the sown seed, a fall of around 50% over the growing season from the concentration in the seed that was sown. There were no significant differences in final Mn concentrations between the Mn treated and untreated plots, although average Mn concentrations were slightly higher upon Mn treatment.

239 (a) 30

(c) 30

Mn uptake (mg M nm-2)

growth stage 30

growth stage 45

25

25

20

20

15

15

10

10

5

5

0

0 MnMn+

Mn concentration (mg Mn kg DW-1)

(b) 60

(d) 20 growth stage 30

growth stage 45

50 15 40 30

10

20 5 10 0

0 Clara

Ludo Cleopatra Carola Vanessa Menhir Escape Antonia

Clara

Ludo Cleopatra Carola Vanessa Menhir Escape Antonia

Figure 3. (a) Mn uptake at growth stage 30; (b) Mn concentration in whole shoots at growth stage 30; (c) Mn uptake at growth stage 45; (d) Mn concentration in whole shoots at growth stage 45. (– – –): critical concentration of Mn in whole shoots of barley at the respective growth stages (Reuter et al., 1997). Each bar represents mean and standard error of four measurements. Trial location: Stoevring 2003.

Mn efficiency at three growth stages At GS 30, only the genotypes Escape and Antonia differed significantly (P < 0.05) between the Mn treatments in fresh weight. At GS 45, only the genotypes Ludo, Cleopatra and Vanessa were not significantly different between treatments (Table 4). Antonia was the least efficient at 76%. All genotypes were significantly different between treatments when estimated based on grain yield data. The most Mn efficient genotype at harvest was Vanessa (92%) and the least were Cleopatra and Antonia (80%), although the relative yield differences between genotypes were not significant. Grain yield and Mn efficiency Mn efficiencies were calculated for the genotypes and compared with previous trial data from 2002

(Table 5). The trial at Vraa was much more high yielding, in line with national average yields for Denmark, while the other two sites (Stoevring, 2003 and Aarre, 2002) were much less so but typical for these locations. Yields for Antonia were between a half and a third less at Stoevring compared to those obtained at Vraa in the previous year. In 2002, Ludo and Carola had both previously been found to be more Mn efficient (97–100%) than Antonia (91%). The results from the field trial in Stoevring in 2003 put Vanessa as the most Mn efficient genotype, although differences are non-significant. In both 2002 and 2003, Antonia was the poorest performing genotype (68–80% efficiency). Mn efficiency at Aarre in 2002 was non-significantly different for the four genotypes, although mean percentages are markedly different (68–83%). In the 2003 trial, the genotypes least sensitive to

Seed Mn concentration (mg Mn kg-1 DW )

240

250

Pre-trial Mn Mn +

200 20

10

0 Clara

Ludo Cleopatra Carola Vanessa Menhir Escape Antonia

Figure 4. Mn concentration in grains, showing grain used to sow the plots, and both the Mn treated and untreated plots at harvest. Pre-trial seed for Antonia had been treated with a commercial seed coating containing Mn. Each bar represents mean and standard error of four measurements. Trial location: Stoevring 2003. Table 4. Stoevring trial, 2003. Mn efficiency (%) at three growth stages based on the relative yield difference between the two Mn treatments Vegetative (%)

Clara Ludo Cleopatra Carola Vanessa Menhir Escape Antonia LSD

Grain (%)

GS 30

GS 45

GS 100

94.4 106.3 87.0 92.9 84.6 97.3 64.8 57.7 40.9

85.6 86.6 95.7 80.6 85.0 86.9 88.6 75.9 35.6

89.3 83.1 80.3 84.3 92.2 90.1 82.9 80.1 16.2

ab a ab ab ab ab b* b*

l* l l l* l l* l* l*

w* w* w* w* w* w* w* w*

The vegetative index is based on the fresh weight yield. For grain, it is determined from harvested plots. Results for genotypes where biomass increased significantly (P < 0.05) with Mn treatment are labelled with *. Efficiency is calculated for each pair block and the value presented here is the average of four such pairs. Mean values with the same letter are not significantly different.

applied Mn, and thus the most Mn efficient, were Vanessa, Menhir and Carola (92%, 90%, and 89%, respectively). Increases were, however, not significant, although the grain yield of all the genotypes increased significantly in the Mn treated plots (Table 4).

Chlorophyll a fluorescence analysis At growth stage 30, there were few differences between the two Mn treatments for a number of the fluorescence parameters, including Fo, Fm and Fv/Fm (data not shown). At growth stage 45 and based on Fv/Fm ratios, Vanessa was significantly more healthy than Clara, Carola and Escape but equivalent to Ludo in the untreated plots (Table 6). Both Vanessa and Ludo genotypes showed no significant improvement in this parameter as a result of spraying (0.66 at Mn) and 0.62 at Mn+ for Vanessa) whereas all other genotypes did improve, although still not to the optimal value of Fv/Fm 0.83 (Bjorkman and Demmig, 1987; Krause and Weis, 1984). Minimum fluorescence (Fo) declined in all genotypes (except Antonia and Menhir) under Mn fertilisation, and Fm increased with Mn treatment with the exception of Ludo. To illustrate this, the average transients for genotype Carola at two Mn treatments are provided (Figure 5).

Discussion In the field trials conducted in 2001 and 2002, differences in Mn efficiency were very apparent (Figures 1 and 2). It is clear, however, that using

241 Table 5. Yield comparisons (t ha)1) and Mn efficiency (%) of four genotypes in three independent field trials over 2 years Location: Stoevring 2003 Mn treatment

Ludo Carola Vanessa Antonia LSD

Location: Vraa 2002

Mn efficiency

Mn treatment

Mn efficiency

Location: Aarre 2002 Mn treatment

Mn efficiency

Mn)

Mn+

(%)

Mn)

Mn+

(%)

Mn)

Mn+

(%)

4.66 4.42 4.61 3.62 0.61

5.63 5.28 5.02 4.52 0.51

83 w 84 w 92 w 80 w 16.2

7.63 8.84 8.06 7.33 0.77

7.83 8.57 8.45 8.02 0.71

97 wxy 103 wx 95 xy 91 y 10.2

3.48 3.14 4.76 3.62 1.36

5.33 4.13 5.98 5.28 1.25

66 w 81 w 83 w 68 w 29.8

a a a b

l lm mn n

b a b b

m l lm lm

ab a a ab

lm m l lm

Mn efficiency is calculated based on pair wise treatment of data, as in Table 4. Mean yield of four plots at each treatment are shown, along with significant difference (P < 0.05) between genotypes at that treatment level. Means with the same letter are not significantly different.

Table 6. Leaf fluorescence measurements at growth stage 45 Genotype Mn) Clara Ludo Cleopatra Carola Vanessa Menhir Escape Antonia Mn+ Clara Ludo Cleopatra Carola Vanessa Menhir Escape Antonia

Fo

Fv/Fm

Fm

1031 1130 1030 1162 1084 899 1176 825

ab* a* ab* a* ab* ab a* b

2385 2862 2676 2363 2512 2443 2658 2283

m* lm lm* m* m* lm* lm* m*

0.554 0.596 0.607 0.500 0.657 0.627 0.545 0.613

xy* wxy wx* y* w wx* xy* wx*

833 909 878 954 941 929 1032 786

a* a* a* a* a* a a* a

2987 2768 3155 2889 3221 3241 3055 2846

lm* lm l* lm* l* l* lm* lm*

0.695 0.634 0.717 0.657 0.616 0.698 0.652 0.716

w* w w* w* w w* w* w*

Mean values with the same letter are not significantly different. Values marked with * are significantly (P < 0.05) different between Mn treatments. Each value is a mean of six measurements.

an approach that matches field trials for Mn efficiency to local agricultural practice can be inadequate. Typically two to three fertiliser applications are made to a crop locally and this, in the 2003 trial, was shown to be ineffective at correcting deficiency. Under these circumstances Mn-efficient genotypes could be misinterpreted as Mn inefficient, as the relative yield difference between the Mn treated and untreated plots is

not fully attained. It is essential to reach the yield plateau in the Mn treated plots in order to compare genotype response to deficiency. This can only be achieved by careful observation of the crop, and spraying with Mn a sufficient number of times. It could also be expected that at extremes of Mn deficiency, such as that found in the 2003 trial, the mechanism of efficiency is inadequate and Mn efficiency breaks down. Indeed, for the genotypes studied over the 3 years, the magnitude of differences of Mn efficiency was much less than that reported in the literature (McDonald et al., 2001), where in the present trial differences ranged from 80% to 92%, compared with 17–90% in the Australian trial. However, it should be noted that the growth conditions and Mn availability in the soil are highly different, and observations on Mn efficiency cannot be directly compared. There are advantages in adopting local agricultural practices in these types of trials, such as indicating the likely usefulness of Mn efficiency of genotypes in normal farming conditions. However, the magnitude of Mn efficiency can be masked, as the potential differences between treatments will not be attained, leading to misidentification of Mn efficiency. For example, Carola was determined as Mn efficient in previous trials (Vraa 2002; Table 5), but not in this present trial. There are additional considerations to be made regarding trial design. To evaluate relative yield differences between Mn treated and untreated plots, an underlying assumption is made that the trial site is non-heterogeneous for

242 3500

Fluorescence yield (mV)

3000

Fm

2500

Fv

2000

1500

Fo

1000

MnMn+

500 10-3

10-2

10-1

100

101

102

103

104

105

ms Figure 5. Average fluorescence transients for genotype Carola. Measurements made at growth stage 45. Maximum fluorescence yield (Fm) is higher in the Mn fertilised plants and minimum fluorescence (Fo) is lower in the Mn treated plants. Differences between the two treatments are significant at both Fo and Fm. Trial location: Stoevring 2003.

Mn availability, and for other factors that influence yield. A difficulty with the trials presented here is that mean values for four pair-plots are often non-significant, resulting from a considerable spatial variability between pair plots. Not only is the application of sufficient Mn to control plots essential, but so is the location and proximity of paired plots to each other, to enable precise determination of relative yield differences (see Ascher-Ellis et al., 2001 for discussion). Assays for scoring Mn efficiency have long been sought (Graham, 1988). It has been reported elsewhere that scoring based on visual assessment of deficiency symptoms was most useful for seedlings, more so than Mn concentrations, content or chlorophyll a fluorescence (Longnecker et al., 1990). In this trial, Mn concentrations were found to differentiate effectively between Mn treatments, but not between genotypes. After stem elongation (growth stage 45) Mn concentrations fell, a result of rapid structural growth and a subsequent dilution effect. It was also clear that the effect of spraying as measured by Mn concentration had almost completely disappeared 8 weeks after application, as there were few differences between treatments (Figure 3d). The greatest

genotypic variation was in total Mn uptake. However, this variation did not correlate with Mn efficiency when ranked by relative grain yield. Establishing Mn efficiency based on biomass at vegetative stages was also inadequate. At growth stage 30, fresh weights were not affected by Mn applications, except in the genotypes Escape and Antonia. By growth stage 45, after stem elongation, the two Mn treatments could be distinguished more clearly, significantly so in all but three of the genotypes, Ludo, Cleopatra and Vanessa. However, all genotypes were distinguishable between treatments at grain harvest. Potential interactions between Mn deficiency and abiotic or other stresses are highlighted by the over-wintering survival. The Mn-inefficient genotype Antonia was particularly poor at surviving winter. Yield losses due to poor seedling establishment could be expected and it was certainly the case that at the end of the trial, Antonia had the lowest average Mn efficiency between pair plots (80.1%). Escape also performed poorly over the winter period and had a similar low Mn efficiency (82.9%; Table 4). The most useful indicator of plant Mn status during the growing season was obtained by

243 variable chlorophyll a fluorescence detection. At growth stage 45, fluorescence detection could consistently distinguish the different Mn treatments. The characteristic pattern of higher fluorescence yield (Fm) and lower initial fluorescence (Fo) in Mn treated plants compared to untreated plants, was in agreement with previous work on chlorophyll a fluorescence (Hannam et al., 1987; Kriedmann et al., 1985). Fo is thought to increase under Mn deficiency, as the result of poor photon harvesting and the loss of excitation energy via fluorescence (Kriedmann et al., 1985). Declining photosystem II efficiency as a result of insufficient Mn supply also leads to a decline in variable fluorescence. This typical pattern of photosystem II fluorescence was seen in all of the barley genotypes included in this trial. The contribution of seed Mn to Mn efficiency was considered to be minimal in this trial. Vanessa had the lowest Mn concentration (12 mg Mn kg)1 DW) in sown seed, but the highest Mn efficiency at the end of the trial. Antonia had been coated with a Mn preparation, but this apparently had no effect on seedling survival over the winter. This contrasts with Australian experiments on durum wheat, where a strong confounding effect of seed Mn on Mn efficiency was observed (Khabez-Saberi et al., 2000). However, it is worth noting that the seed concentration in the Australian experiments in general ranged from approximately 10–50 mg Mn kg)1 DW, whereas the concentration in the Danish seeds ranged from 10 to 15 mg Mn kg)1 DW. The reduction in Mn concentrations of the harvested seed to approximately 50% of that found in the sown seed, reflects a very poor Mn availability of the soil at the trial site. It is possible that the poor seedling establishment under the growing conditions found in the region could have been improved, if seeds with higher Mn had been available. It is probable that for example growing Vanessa instead of Antonia is a rational choice on Mn deficient soils, but it will not be possible to do away with Mn amendments throughout the growing season. The frequency of Mn applications may however be reduced through use of a Mn efficient genotype, with subsequent economic benefits to the producer. It is concluded that Mn

efficiency of the genotypes screened in this study could not be precisely or consistently described, but genotypic differences are present and bear further investigation. Acknowledgements This work was supported by grants from The Danish Agricultural and Veterinary Research Council (Contract 53-00-0234); The Ellen, Christian and Anders Petersen Foundation; and The Danish Grain Breeding Foundation. The technical assistance of Bente Broeng is gratefully acknowledged. References Ascher-Ellis J S, Graham R D, Hollamby G J, Paull J, Davies P, Huang C, Pallotta M A, Howes N, KhabezSaberi H, Jefferies S P and Moussavi-Nik M 2001 Micronutrients. In Application of Physiology in Wheat Breeding. Eds. M P Reynolds, J I Ortiz-Monasterio and A McNab. CIMMYT, Mexico. Bjorkman O and Demmig B 1987 Photon yield of O2 evolution and chlorophyll fluorescence characteristics at 77 K among vascular plants of diverse origin. Planta 170, 489–504. Gherardi M J and Rengel Z 2004 The effect of manganese supply on exudation of carboxylates by roots of lucerne (Medicago sativa). Plant Soil 260, 271–282. Graham R D 1988 Genotypic differences in tolerance to manganese deficiency. In Manganese in Soils and Plants. Eds. R D Graham, R J Hannam and N C Uren. pp. 261–276. Kluwer Academic Publishers, Dordrecht, Boston, London. Graham R D, Ascher J S and Hynes S C 1992 Selecting zincefficient cereal genotypes for soils of low zinc status. Plant Soil 146, 241–250. Hannam R J, Riggs J L and Graham R D 1987 The critical concentration of manganese in barley. J. Plant Nutr. 10, 2039–2048. Huang C, Webb M J and Graham R D 1994 Manganese efficiency is expressed in barley growing in soil system but not in solution culture. J. Plant Nutr. 17, 83–95. Husted S, Mikkelsen B F, Jensen J and Nielsen N E 2004a Elemental fingerprint analysis of barley (Hordeum vulgare) using inductively coupled plasma mass spectrometry, isotope-ratio mass spectrometry, and multivariate statistics. Anal. Bioanal. Chem. 378, 171–182. Husted S, Thomsen M U, Mattsson M and Schjoerring J K 2004b Influence of nitrogen and sulphur form on manganese acquisition by barley (Hordeum vulgare). Plant Soil. Khabez-Saberi H and Graham R D 2002 Improvement of screening for manganese efficiency by producing seed with similar manganese content in different genotypes and genetic stocks. J. Plant Nutr. 25, 2803–2815. Khabaz-Saberi H, Graham R, Ascher-Ellis J and Rathjen A 2000 Quantification of the confunding effect of seed manganese content in screening for manganese efficiency in

244 durum wheat (Triticum turgidum L. var. durum). J. Plant Nutr. 23, 855–866. Krause G H and Weis E 1984 Chlorophyll fluorescence as a tool in plant physiology. II. Interpretation of fluorescence signals. Photosynth. Res. 5, 139–157. Kriedmann P E, Graham R D and Wiskich J T 1985 Photosynthetic dysfunction and in vivo changes in chlorophyll a fluorescence from manganese deficient wheat leaves. Aust. J. Agr. Res. 36, 157–169. Longnecker N E, Graham R D, McCarthy K W, Sparrow D H B and Egan J P 1990 Screening for manganese efficiency in barley (Hordeum vulgare L.). In Genetic Aspects of Plant Mineral Nutrition. Ed. N E Bassam. pp. 273–280. Kluwer Academic Publishers. McDonald G K, Graham R D, Lloyd J, Lewis J, Lonergan P and Khabas-Saberi H 2001 Breeding for improved zinc and manganese efficiency in wheat and barley. In 10th Australian Agronomy Conference, Hobart, Australia, 2001. Pearson J N and Rengel Z 1997 Genotypic differences in the production and partitioning of carbohydrates between roots and shoots of wheat grown under zinc or manganese deficiency. Ann. Bot. London 80, 803–808. Posta K, Marschner H and Ro¨mheld V 1994 Manganese reduction in the rhizosphere of mycorrhizal and nonmycorrhizal maize. Mycorrhiza 5, 119–124.

Rengel Z 1999 Physiological mechanisms underlying differential nutrient efficiency of crop genotypes. In Mineral Nutrition of Crops Fundamental Mechanisms and Implications. Ed. Z Rengel. Food Products Press, an Imprint of The Haworth Press, Inc., New York, London, Oxford. Rengel Z, Graham R D and Pedler J F 1994 Time-course biosynthesis of phenolics and lignin of wheat genotypes differing in manganese efficiency and resistance to take-all fungus. Ann. Bot. London 74, 471–477. Reuter D J, Edwards D G and Wilhelm N S 1997 Temperate and Tropical Crops. In Plant Analysis: An Interpretation Manual. Eds. D J Reuter and J B Robinson. pp. 81–279. CSIRO Publishing, Victoria. Sadana U S, Lata K and Claassen N 2002 Manganese efficiency of wheat cultivars as related to root growth and internal manganese requirement. J. Plant Nutr. 25, 2677–2688. White J G and Zasoski R J 1999 Mapping soil micronutrients. Field Crop Res. 60, 11–26. Zadoks J C, Chang T T and Konzak C F 1974 A decimal code for the growth stages of cereals. Weed Res. 14, 415– 421. Section editor: P.J. Gregory