# Source paper archive - Source title: Higher stomatal density and smaller stomatal size associate with higher grain yield in winter wheat - Author(s): Elena Ivandi, Pirko Jalakas, Ingmar Tulva, Mati Koppel, Kajal Samantara, Banafsheh Khalegh Doust, Anne Ingver, Merlin Haljak, Reine Koppel, Ebe Merilo, Hanna Hõrak - Source landing URL: https://www.biorxiv.org/content/10.64898/2026.11.04.764249v1 - PDF URL: https://www.biorxiv.org/content/10.64898/2026.11.04.764249v1.full.pdf - PDF SHA-256: d86967a8d4072a59a5acefad24e46189bb0ed737f40c01fc3c91632cd4948d90 - Converter: pymupdf4llm 0.0.27 - Limits: Automated PDF text extraction; reading order, equations, tables and formatting may be incomplete or incorrect. Images are omitted, and no OCR was performed. Markup and links are neutralized for safe downloading. Consult the original PDF for the authoritative version. --- [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) # 1 Higher stomatal density and smaller stomatal size associate with 2 higher grain yield in winter wheat 3 Elena Ivandi [1,2], Pirko Jalakas [1], Ingmar Tulva [1], Mati Koppel [3], Kajal Samantara [1], 4 Banafsheh Khalegh Doust [4], Anne Ingver [2], Merlin Haljak [2], Reine Koppel [2], Ebe 5 Merilo [1], Hanna Hõrak [1 ] 6 1 University of Tartu, Institute of Technology, Nooruse 1, Tartu, Estonia, 2 Centre of Estonian Rural 7 Research and Knowledge, Aamisepa 1, Jõgeva, Estonia, [3] Estonian University of Life Sciences, 8 Institute of Agricultural and Environmental Sciences, Kreutzwaldi 5, Tartu, Estonia, [4] Estonian 9 University of Life Sciences, Institute of Forestry and Engineering, Kreutzwaldi 56, Tartu, Estonia 10 11 **Highlight** 12 We found coordinated adjustment of wheat stomatal densities, but independent stomatal 13 responses to environmental factors on the adaxial and abaxial leaf surfaces, whereas 14 stomatal density positively associated with grain yield. 15 **Abstract** 16 In wheat ( _Triticum aestivum L_ .), flag leaf is crucial in determining grain-filling rate and final 17 grain yield. Stomata, microscopic pores on plant surfaces, balance CO 2 uptake and 18 evaporative water loss; their numbers and aperture sizes depend on environmental factors. 19 Previous studies have reported associations between stomatal traits and grain yield; mostly 20 under controlled conditions. Our study aimed to investigate flag leaf stomatal anatomical and 21 physiological traits, and their relationships with winter wheat grain yield under field 22 conditions. We measured stomatal density and size, and stomatal conductance on adaxial 23 and abaxial flag leaf surfaces in a three-year field study with twenty-five winter wheat 24 varieties. Stomatal density ratio remained constant despite different stomatal densities 25 occurring in different years, indicating coordinated adjustments of stomatal numbers between 26 leaf surfaces in response to environmental conditions. Stomatal conductance ratios varied 27 widely, indicating leaf-surface specific stomatal responses to environmental conditions. Grain 28 yield was positively associated with stomatal density and negatively with stomatal size. Flag 29 leaf removal reduced grain weight per spike and did not affect grain numbers per spike, 30 unlike in spring wheat. Our field study provides valuable insights into the relationships 31 between winter wheat stomatal traits and grain yield that can be used in breeding. 32 **Keywords:** flag leaf, glaucousness, pathogens, stomata, stomatal density, stomatal size, 33 wheat, yield 34 **Abbreviations** 35 FL- flag leaf 36 FLA- flag leaf area 37 GCL- guard cell length 38 GL- glaucousness 39 g s - stomatal conductance 40 g smax - anatomical maximum stomatal conductance 41 PL- pore length 42 SD- stomatal density 43 SR- adaxial to abaxial stomatal density ratio 44 TGW- thousand grain weight 1 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 45 46 **1. Introduction** 47 Wheat ( _Triticum aestivum L_ .) is one of the most widely grown crops and plays a crucial role 48 in global food security (Erenstein _et al._, 2022). In 2025, the global wheat production was 49 843.8 Mt (‘Wheat | USDA Foreign Agricultural Service’, 2026). The flag leaf is the major 50 organ in wheat canopy, determining grain-filling rate, yield components such as grain 51 numbers per spike and grain weight per spike, and overall grain yield by supplying 52 photoassimilates to developing grains after anthesis (Melahat Avci, 2005; Ali _et al._, 2010; 53 Racz _et al._, 2022). As the topmost youngest leaf, the flag leaf absorbs more light compared 54 with lower, shaded leaves. Removal of the flag leaf negatively affects grain yield in spring 55 and winter wheat and forces lower leaves to compensate and to supply photoassimilates to 56 grains (Patterson and Ohm, 1975; Mahmood _et al._, 1991; Melahat Avci, 2005; Racz _et al._, 57 2022). Since most of the previous studies have been conducted in warm climates, the 58 importance of the flag leaf for the formation of grain yield components under cooler 59 conditions of Northern Europe remains unclear. 60 Glaucousness, a characteristic associated with adaptation to water-limited and heat-stressed 61 conditions, manifests as a visible waxy epicuticular layer with a dull white to bluish color on 62 leaves and spikes (Johnson _et al._, 1983; Merah _et al._, 2000; Guo _et al._, 2016; Bi _et al._, 2017). 63 Epicuticular waxes of glaucous cultivars serve as a barrier preventing non-stomatal water 64 loss, leading to higher water use efficiency, increased leaf longevity, and lower leaf 65 temperature, potentially due to increased reflectance (Johnson _et al._, 1983; Richards _et al._, 66 1986). Glaucous cultivars can have higher yields both in drought conditions (Merah _et al._, 67 2000) and when water is not limiting (Johnson _et al._, 1983). 68 Stomata are microscopic pores on plant surfaces (leaves, stems) that balance CO 2 uptake 69 and evaporative water loss, exerting significant control on the world’s water and carbon 70 cycles (Hetherington and Woodward, 2003). Depending on species, stomata can be present 71 on one leaf surface or be distributed between both leaf surfaces. Most plants are 72 hypostomatous, meaning that they have stomata only on the lower (abaxial) leaf surface 73 (Muir, 2015). In amphistomatous plants, stomata are formed on both leaf surfaces, whereas 74 they develop and function at least partly independently (Hõrak, 2026). Stomatal density (SD) 75 in most amphistomatous species tends to be higher on the abaxial (lower) compared with the 76 adaxial (upper) surface (Pemadasa, 1979). The exception is the _Poaceae_ family: their SD is 77 often equal on both leaf surfaces, or higher on the adaxial surface, as in wheat (Pemadasa, 78 1979; Wall _et al._, 2022; Samantara _et al._, 2025), where adaxial surface also contributes more 79 to leaf gas-exchange (Wall _et al._, 2022; Chaplin _et al._, 2026a; Chaplin _et al._, 2026b). Adaxial 80 and abaxial stomata in amphistomatous plants respond independently to environmental 81 stimuli, such as light intensity (Wang _et al._, 2008; Wall _et al._, 2022; Wei _et al._, 2025). Higher 82 adaxial to abaxial stomatal density ratio (SR) has been associated with faster stomatal 83 response kinetics (Woning et al., 2026; Yang et al., 2026). Having more stomata on the 84 adaxial leaf surface increases photosynthetic efficiency and CO 2 uptake (Xiong and Flexas, 85 2020), but it also may increase evaporative water loss and susceptibility to stomata-entering 86 pathogens (e.g., rust fungi) (McKown _et al._, 2014; Muir, 2015; Drake _et al._, 2019). Yield 87 losses caused by pathogens are associated with lodging, decreased nutrient availability, and 88 lower thousand grain weight, and numbers of grain per spike (Figueroa _et al._, 2018; Khanfri 89 _et al._, 2018). Higher SDs may lead to increased susceptibility to stomata-entering pathogens 90 (Dutton _et al._, 2019; Tateda _et al._, 2019). Thus, leaf surface-specific stomatal traits may 91 influence CO 2 uptake, water loss, and pathogen entry, ultimately affecting yield under field 92 conditions. 2 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 93 Wheat stomatal traits are associated with grain yield, but specific relationships may depend 94 on growth stage and environment. For example, a positive correlation was found between 95 abaxial SD and grain yield (Yang _et al._, 2022; Du _et al._, 2026), or both adaxial and abaxial 96 SD and grain yield (Yousufzai _et al._, 2009) in wheat. Other studies have reported a negative 97 relationship between adaxial SD and grain yield (Samantara _et al._, 2025), or found no 98 correlation between stomatal traits and grain yield (Ledent and Jouret, 1978; Shahinnia _et al._, 99 2016; Chaplin _et al._, 2026a). Thus, stomatal traits may have breeding potential in wheat but 100 further characterisation of leaf surface-specific stomatal anatomy and physiology under field 101 conditions is needed to better understand the variation in wheat stomatal traits and their 102 contribution to yield determination. 103 This study aimed to (i) investigate winter wheat flag leaf stomatal anatomical and 104 physiological traits under field conditions over three years, (ii) examine the relationships 105 between flag leaf adaxial and abaxial stomatal traits and grain yield of winter wheat, and (iii) 106 determine the effect of flag leaf removal on yield components in winter wheat and spring 107 wheat. # 108 2. Material and methods 109 **2.1. Plant material and field experiments** 110 Field trials were conducted at the Centre of Estonian Rural Research and Knowledge 111 (METK, Estonia), Jõgeva, Estonia (58°45'N, 26°24'E) in 2023-2025 seasons. Twenty-five 112 winter wheat varieties (Table S1) in three replications were sown (plot size of 9 m [2], and row 113 spacings of 12.5 cm) on 9 September 2022; 11 September 2023, and on 9 September 2024. 114 Fungicide treatment was not applied. Half of the varieties included in the trial were early115 maturing and the other half were late-maturing (Table S1). In 2025, we included ten spring 116 wheat varieties with three replications (three plots per genotype) in the flag leaf removal 117 experiment as a comparison to winter wheat. Plot size was the same as for winter wheat. 118 Spring wheat was sown on 28 April 2025. Field trials were carried out following a randomized 119 complete block design. 120 **2.2. Meteorological data** 121 In 2023–2024, May-June were relatively dry, followed by rainy July (Fig. 1). The wettest year 122 was 2025, when precipitation was above long-period average (LPA, 1991-2020) for most of 123 the spring and summer months. In 2023–2024, temperatures were high from May to August, 124 while in 2025, temperatures were below LPA during the major growth period in May and 125 June, and exceeded the LPA in the main grain filling period in July (Fig. 1). 126 **2.3. Flag leaf traits** 127 Since half of the varieties included in the study were early-maturing and the other half were 128 late-maturing (Table S1), the flag leaf measurements were taken twice, about one week 129 apart (9 and 16 June in 2023; 7 and 11 June in 2024; 16 and 26 June in 2025; flag leaves of 130 early varieties were measured and detached for anatomical measurements on the first and 131 those of late varieties on the second date each year) at the end of the heading stage (BBCH 132 59), when flag leaves had reached full size. Six plants from each plot were randomly 133 selected for flag leaf stomatal measurements and flag leaf removal, and an additional six 134 plants with intact flag leaves were chosen as control plants for yield and yield component 135 determination at the final harvest. The abaxial and adaxial stomatal conductance (g s ) of the 136 central part of flag leaves (avoiding the midrib) was measured using a LI-600 (LI-COR) hand137 held porometer. In further data analysis, total g s was calculated as the sum of abaxial and 138 adaxial g s and g s ratio as the ratio of adaxial g s to abaxial g s . After g s determination, 3 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 139 measured flag leaves were removed and taken to the laboratory for flag leaf area and 140 stomatal anatomical characteristics measurements. In the laboratory, the flag leaves were 141 photographed, and flag leaf areas (FLA, cm [2] ) and widths were measured using ImageJ 142 software (Schneider _et al._, 2012). At late wax maturity (BBCH 89), before mechanical 143 harvesting, 12 marked plants (six with flag leaf removed and six control plants) were 144 harvested manually from each plot for determination of plant yield components of the main 145 shoot (grain number per spike and grain weight per spike). After that, winter wheat was 146 harvested mechanically between 28 July and 8 August, 2023, between 21 July and 2 August 147 in 2024, and between 31 July and 8 August in 2025. After mechanical harvest, the thousand 148 grain weight (TGW, g) of winter wheat was measured, grain yield (kg ha [-1] ) was expressed on 149 a 14% grain moisture level, and grains smaller than 2.2 mm were sorted out and considered 150 as fine grains. 151 For spring wheat, flag leaf measurements were taken on 23 July 2025. Si x plants from each 152 plot were randomly selected for flag leaf removal, and an additional six plants were chosen 153 as control plants with intact flag leaves. A t late wax maturity (BBCH 89), before mechanical 154 harvesting, 12 marked plants (six with flag leaf removed and six control plants) were 155 harvested manually from each plot for determination of plant yield components of the main 156 shoot (grain number per spike and grain weight per spike). Spring wheat was harvested 157 between 27 August and 8 September in 2025. Since spring wheat was a comparison with 158 winter wheat in the flag leaf removal experiment, we did not measure final grain yield in 159 spring wheat. 160 **2.4. Stomatal anatomical analyses** 161 For stomatal anatomical measurements, the central part of the flag leaf was used. Sampled 162 leaf parts were cut in half widthwise (Fig. 2), and epidermal impressions of adaxial and 163 abaxial surfaces were made using dental silicone (Speedex light body, Coltene/Whaledent 164 AG for adaxial and oranwash L, Zhermack for abaxial). Then, nail varnish was applied to the 165 sample impressions and transferred to microscope slides with transparent tape. An area of 166 approximately 0.28 mm [2] in the central part of the sample was captured using the Kern OBF 167 133 microscope and ODC 832 camera (Kern & Sohn GmbH) at 200x magnification. Stomatal 168 density (SD, stomata mm [-2] ), guard cell length (GCL, µm), guard cell width (GCW, µm) and 169 pore length (PL, µm) were measured using ImageJ software (Schneider _et al._, 2012). For 170 GCL, GCW and PL measurements, five stomata per sample were measured and average of 171 these measurements was calculated to gain one representative value per plant. Total SD 172 was calculated as the sum of adaxial and abaxial SDs. Stomatal density ratio (SR) was 173 defined as the ratio of adaxial SD to abaxial SD (McAusland _et al._, 2021). 174 Theoretical maximal anatomical stomatal conductance g smax was calculated with the following 175 equation (Caine _et al._, 2023): ���� � ���� 176 ������ ���� ~~�~~ [�] � [� ] ~~[�]~~ [����] � ~~�~~ 177 where d is diffusity of water in air at 25°C (0.0000249 m [-2] s [-1] ) and � is molar volume of air 178 (0.024464 m [3] mol [-1] ). SD is stomatal density, ���� (µm [2] ) is maximum pore area, calculated 179 as abaxial or adaxial PL*PL/2, and pore depth � (µm) was considered to be equivalent to the 180 width of guard cell at the centre of the stoma (Caine _et al._, 2023). 181 **2.5. Agronomical traits and plant analyses** 182 According to CPVO methodological guidelines (Community Plant Variety Office, 2019), 183 plants were evaluated for glaucousness (GL) on a scale of 1–9 (1- absent or very weak, 9 4 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 184 very strong) about two weeks after heading. GL was evaluated on abaxial and adaxial flag 185 leaf surfaces, and on spike. 186 Plant disease severity was assessed from the beginning of heading stage (BBCH 51) at 187 approximately two-week intervals on three occasions during the growing season. Disease 188 severity was rated on a scale of 1–9, where 1 indicated no visible signs of disease and 9 189 indicated that the entire leaf was infected. In 2024, conditions were favourable for the spread 190 of plant diseases, and disease symptoms were visible even before heading. Therefore, 191 assessments began earlier, at the end of stem elongation (BBCH 35), and were conducted 192 on four occasions during the growing season. Plants were evaluated for the following 193 diseases: yellow rust ( _Puccinia striiformis_ ), tan spot ( _Pyrenophora tritici-repentis_ ), powdery 194 mildew ( _Blumeria graminis_ ), and leaf blotch ( _Zymoseptoria tritici_ ). The three-year cumulative 195 disease index (CDI) was calculated as the summed incidence of all four diseases averaged 196 over the three study years. The cumulative disease index for each year was calculated as 197 the sum of disease incidences for all four diseases within one year. The disease index (DI) 198 for each disease was calculated as the mean of its annual incidence values over the three 199 years. 200 **2.6.** **Statistical analyses** 201 Analyses of variance (ANOVA) General Linear Models (GLM) procedure was used to 202 evaluate differences in traits between genotypes and years. Homogeneity of variances was 203 tested with the Levene’s test. Factorial analysis of variance (two-way ANOVA) was used to 204 evaluate the main effects of genotype, year and their interactions for measured traits. 205 Comparisons between individual means were performed using Tukey’s HSD post hoc test. 206 Simple linear regression analyses were used to detect significant relationships between 207 stomatal characteristics and grain yield. For nonparametric traits (disease index and 208 glaucousness) the Kruskal-Wallis one-way ANOVA by ranks and multiple comparison of 209 mean ranks were used. Spearman correlation was used for correlating nonparametric traits. 210 For statistical analyses, measurements from six plants within each plot were averaged to 211 obtain a single plot-level mean to be used in statistical analyses (n=3 plots per genotype 212 within each year). For diseases, statistical analyses were performed using genotype-level 213 means obtained by averaging values from 3 plots per genotype; for glaucousness, statistical 214 analyses were performed using 3 plots per genotype within each year. For regression and 215 correlation analyses, genotype-level means for each year were used. All effects were 216 considered significant at p < 0.05. Statistical analyses were performed with Statistica (version 217 7.0, StatSoft Inc., Tulsa, OK, USA). All the graphs in the results section were made using 218 GraphPad Prism, version 8.0 (GraphPad software, San Diego, California, USA, 219 www.graphpad.com). 220 The manuscript was reviewed for language and clarity using Grammarly. # 221 3. Results 222 **3.1. Variation of agronomical traits of studied winter wheat varieties** 223 **3.1.1. Grain yield** 5 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 224 Grain yield was lowest in 2025 (average for early varieties 4140 kg ha [-1] and late varieties 225 5097 kg ha [-1] ), while in 2023 and 2024 it was higher, mean values for early varieties 7167 and 226 6149 kg ha [-1 ] and for late varieties 8435 and 8878 kg ha [-1], respectively (Fig. 3A,B). Grain 227 yields of both early and late varieties were significantly affected by year, genotype, and their 228 interaction (Table 1). The variability in grain yield was greater in 2024 than in 2023 and 2025 229 for early varieties, but more uniform in late varieties (Fig. 3A,B). In 2025, TGW was 1.5 times 230 lower compared with 2023 in both early and late varieties (Fig. S1A,B). In 2024, Fredis and 231 Ruske had the lowest TGW among early varieties (Fig. S1A). The proportion of fine grain in 232 both early and late varieties was affected by genotype, year and their interaction (Table 1). In 233 both early and late varieties, the proportion of fine grain was lowest in 2023 (mean 0.49 and 234 0.52%, respectively) and highest in 2025 (mean 16.3 and 13.7%, respectively) (Fig. S1C,D). 235 In 2024, grain yield was remarkably low for early varieties Donskaya, Fredis, and Ruske (Fig. 236 3A), likely due to their susceptibility to yellow rust. We found no significant differences in 237 cumulative disease index between years (Fig. S2A), but the variability in yellow rust infection 238 scores in 2024 was high, likely due to the few susceptible early varieties (Fig. S2B). 239 Cumulative disease index correlated negatively with grain yield in all individual years and in 240 the pooled dataset of 2023 – 2025 (Fig. S3A,C − E). We also found a negative correlation 241 between disease index of three diseases and grain yield (Fig. S3B). 242 **3.1.2. Glaucousness** 243 Adaxial glaucousness (GL) was generally low, but slightly higher in 2024 and 2025 than in 244 2023 (Fig. S4). In warmer years (2023 and 2024), the abaxial GL was higher than in cooler 245 year (2025) (Fig. S4). GL on the spike was highest in 2024 (Fig. S4). In the pooled dataset of 246 2023 − 2025, both abaxial and adaxial GL as well as GL on the spike were positively 247 correlated with grain yield (Fig. S5A). Spike GL was significantly positively correlated with 248 grain yield also in all individual years (Fig. S5A − D). Adaxial GL was positively correlated with 249 grain yield in 2024 and 2025 (Fig. S5C,D), whereas abaxial GL was positively correlated with 250 grain yield only in 2024 (Fig. S5C). 251 **3.1.3. Flag leaf removal** 252 For both early and late varieties, genotype, year and their interaction had significant effects 253 on flag leaf area (FLA), but only in early varieties was the flag leaf width affected by these 254 factors (Table 1). An average FLA was nearly twice as large in 2025 as in 2023 (average for 255 early varieties 33.95 and 16.82 cm [2] and for late varieties 29.67 and 15.38 cm [2], respectively) 256 (Fig. S6A,B). 257 At the heading stage, we removed flag leaves from six randomly selected plants per plot to 258 determine whether flag leaf removal affects grain yield components. Results showed a 259 reduction in grain weight per spike of approximately 18% due to flag leaf removal in 2025 260 (Table 2), but there was no significant effect in 2023, 2024, nor in the pooled dataset (Table 261 2). We detected no effect of flag leaf removal on grain number per spike in any year in winter 262 wheat. 263 In spring wheat in 2025, flag leaf removal had a stronger effect than in winter wheat: grain 264 number per spike decreased by approximately 6.5%, and the grain weight per spike by 24% 265 (Table 2). 266 **3.2. Stomatal anatomical traits and their association with yield** 267 Wheat considerably differs in stomatal density (SD) between leaf surfaces (p= 0.00, Table 3), 268 with higher SD on the adaxial than on the abaxial surface (Fig. 4A-D). SD of both leaf 269 surfaces was affected by genotype and year (Table 1); however, the GxY interaction was 6 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 270 observed only for the adaxial surface of early varieties, where SDs were smaller in 2025 than 271 in 2023 in most and smaller than 2024 in half of the studied varieties (Table 1, Fig. 4A-D). 272 Despite large annual variation in SD, adaxial to abaxial stomatal density ratio (SR) was 273 similar between years, and it was significantly affected by genotype only in early varieties 274 (Table 1, Fig. 4E,F). 275 Adaxial and abaxial GCLs were higher in 2025 compared with 2023 and 2024 for many early 276 and few late varieties (Fig. S7A − D). We observed a positive relationship between adaxial SD 277 and grain yield in 2024 and in the pooled dataset (Fig. 5A), and between abaxial SD and 278 grain yield in 2024, 2025, and in the pooled dataset (Fig. 5B). We found a negative 279 relationship between adaxial GCL and grain yield in two years out of three and in the pooled 280 dataset (Fig. 5C), but abaxial GCL was negatively associated with grain yield only when the 281 data from three years were pooled (Fig. 5D). SR was not significantly associated with grain 282 yield in any individual years nor in the pooled data of 2023 − 2025 (Fig. 5E). 283 We found a negative relationship between FLA and grain yield in the pooled dataset of three 284 years (Fig. 5F), and plants with larger flag leaves had lower SD on both leaf surfaces (Fig 285 S8A,B). SD was negatively related with GCL in all individual years as well as in pooled data 286 (Fig S8C,D). We also observed a negative correlation between both adaxial and abaxial SD 287 and disease index for powdery mildew, and adaxial SD and disease index for yellow rust in 288 the pooled dataset of 2023 − 2025 (Fig. S3B). 289 **3.3. Anatomical maximal and operational stomatal conductance** 290 To address maximum gas exchange potential of leaf surfaces, we calculated maximum 291 anatomical stomatal conductance (g smax ). Adaxial g smax was affected by genotype, year and 292 the GxY interaction in early varieties and by genotype and year in late varieties (Table 1), 293 with lower values recorded in 2025 than in 2023 and 2024 in few early varieties (Fig. S9A-D). 294 Abaxial g smax was affected by genotype and year in both early and late varieties (Table 1). 295 We observed a positive relationship between adaxial SD and adaxial g smax in 2024 and 2025, 296 and in the pooled dataset (Fig. S10A). Similar relationship for the abaxial surface was found 297 in all individual years and in the pooled data of 2023 − 2025 (Fig. S10B). A negative 298 relationship between g smax and GCL for both leaf surfaces was evident only in the pooled 299 data (Fig. S10C,D). We found positive relationships between adaxial and abaxial g smax and 300 grain yield in the pooled data, but not in individual years (Fig. S10E,F). 301 Operational adaxial g s and total g s of early varieties depended only on the year, but in other 302 cases both the genotype, year and their interactions affected operational g s and g s ratio 303 (Table 1). In 2025, adaxial g s was higher than in 2023 in most and higher than in 2024 in 304 some late varieties, whereas abaxial g s was higher compared with 2023 in few varieties (Fig. 305 S11A − D). The value of g s ratio was over 2 (Fig. S11E,F), indicating that adaxial leaf surface 306 is more important in gas exchange than abaxial surface. The g s ratio of early varieties was 307 more variable in 2025 compared with late varieties (Fig. S11E,F). Both early and late 308 varieties exhibited the lowest abaxial and adaxial g s /g smax ratios in 2023 and the highest in 309 2025 (Fig. S12). We found a negative relationship between adaxial g smax and operational g s 310 in the pooled dataset, but a positive relationship in 2024 (Fig. S13A), whereas abaxial g smax 311 was negatively related with operational g s in the pooled dataset (Fig. S13B). We observed a 312 positive relationship between adaxial g s and grain yield only in 2023, whereas both adaxial 313 and abaxial g s and yield were negatively related in the pooled dataset (Fig. 6A,B). Similarly, 314 the relationship between total g s and grain yield was positive in 2023 and negative in the 315 pooled dataset (Fig. 6C). Adaxial and abaxial g s associated positively with each other in the 316 pooled dataset but not in any individual years (Fig. 6D). 7 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 317 # 318 4. Discussion 319 **4.1. Winter wheat grain yield and effects of flag leaf removal on yield** 320 **components** 321 Precipitation and pathogens had major effects on yield in our study. Grain yield was lowest in 322 2025 due to heavy rains and lodging (Fig 1,3), in line with a previous report, where 10 mm 323 increase in precipitation during the grain-filling period and close to maturity decreased winter 324 wheat grain yield by 91 − 92 kg ha [-1 ] (Peltonen-Sainio _et al._, 2011). Lodging occurred shortly 325 after heading and affected several varieties, two of which could not be harvested. Such 326 extensive lodging may have negatively affected grain filling by reducing leaf light exposure 327 and canopy photosynthesis (Setter _et al._, 1997). Pathogens can account for 50 − 60% yield 328 losses in susceptible varieties (Moldenhauer _et al._, 2006; Figueroa _et al._, 2018). Accordingly, 329 disease severity was negatively correlated with yield in all years and in 2024, yellow rust was 330 responsible for major yield losses in the few susceptible early varieties (Fig. 3, S2-3). 331 Flag leaf adaxial and abaxial and spike glaucousness (GL) were positively correlated with 332 grain yield in the pooled dataset and often also in individual years (Fig. S5). Higher yields in 333 glaucous plants irrespective of growth conditions have been found before (Johnson _et al._, 334 1983; Merah _et al._, 2000; Monneveux _et al._, 2004). Advantages of glaucousness may be 335 associated with the ability of cuticular wax to lower leaf temperature via increased reflectance 336 (Johnson _et al._, 1983), and reduce leaf water loss and increase water use efficiency 337 (Richards _et al._, 1986). In case of spikes, high GL may lead to reduced ear wetting and less 338 in-ear sprouting (King and von Wettstein-Knowles, 2000). 339 Flag leaf largely determines grain-filling rate, and may contribute up to 25% to final grain 340 yield in spring wheat (Mahmood _et al._, 1991), so flag leaf removal at the end of the heading 341 stage (BBCH 59) would be expected to lower grain yield components also in winter wheat. 342 Contrary to our expectations, we found that removing the flag leaf significantly decreased 343 grain weight per spike only in 2025 (Table 2) whereas no effect on grain number per spike 344 was detected (Table 2). Other studies have reported a decrease in winter wheat grain weight 345 per spike by ~20-30%, in grain numbers per spike by 10 − 12% (Melahat Avci, 2005; Racz _et_ 346 _al._, 2022), and in grain yield by ~12% (Patterson and Ohm, 1975). We also tested whether 347 winter and spring wheat respond differently to flag leaf removal. Flag leaf removal may lower 348 grain yield by up to 25% and TGW by up to 20%, whereas in some cultivars and studies it did 349 not affect grain yield (Duwayri, 1984; Mahmood _et al._, 1991). Here, the flag leaf removal 350 decreased spring wheat grain number per spike 6.5% and grain weight per spike 24%, which 351 is more than in winter wheat (Table 2). These findings suggest that winter and spring wheat 352 differ in their responses to flag leaf removal, highlighting differences in their physiology and in 353 the contribution of the flag leaf to final grain yield. Winter wheat tends to have more shoots 354 and greater leaf area (Watson _et al._, 1963; Lorenzo _et al._, 2015), and therefore can provide a 355 larger residual photosynthetic source, which can help compensate for the loss of the flag leaf 356 from the main shoot. Indeed, removing the flag leaf of the main shoot reduces the primary 357 source of grain filling (Melahat Avci, 2005), but this deficiency may be compensated by the 358 contributions of assimilates from other leaves (Song _et al._, 2025). Canopy shading may limit 359 such compensation in cool and wet years that support the production of large foliage 360 (Aggarwal _et al._, 1990), which might explain the reduction of grain yield components in winter 361 wheat only in the cool and wet 2025 (Table 2). Spring wheat typically produces fewer tillers 362 (Lorenzo _et al._, 2015), and therefore is more dependent on the main shoot flag leaf during 363 grain filling. When the flag leaf in spring wheat is removed, the remaining tillers and leaves 8 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 364 may be less able to compensate fully for the lost photosynthetic capacity, leading to a 365 reduction in grain yield. Winter wheat may also accumulate more stem reserves during 366 vegetative growth in spring-early summer compared with spring wheat, which has less time 367 for foliage development. Stem reserves can then be used for grain growth and support higher 368 individual grain weight and total yield (Dreccer _et al._, 2009). Considering the potential 369 influence of weather conditions on the observed responses to flag leaf removal, further 370 experiments across multiple environments are necessary to determine how weather 371 conditions affect the impact of flag leaf removal on yield components and whether spring and 372 winter wheat respond in a similar manner. 373 Unexpectedly, FLA was negatively associated with grain yield in the pooled dataset of three 374 years, driven by data of 2025 (Fig. 5F). As fine grain proportion was particularly high in 2025 375 (Fig S1C,D), this relationship is likely explained by inability of plants to realise their yield 376 potential due to cool and wet weather (Fig 1), as well as lodging, which likely hindered 377 canopy photosynthesis. Although the flag leaf may contribute up to quarter of final grain 378 yield, lodging-induced shading may have made actual canopy photosynthesis under field 379 conditions, rather than the maximum photosynthetic capacity of individual leaves, more 380 important for assimilate supply and grain filling as found in previous studies (Driever _et al._, 381 2014; Carmo-Silva _et al._, 2017). 382 **4.2. Stomatal anatomical traits and their association with grain yield** 383 We found higher adaxial than abaxial SD in winter wheat, resulting in SR values consistently 384 above 1, in line with several previous works (Wang and Clarke, 1993; McAusland _et al._, 385 2021; Wall _et al._, 2022; Samantara _et al._, 2025). A recent study comparing wild ( _T._ 386 _dicoccoides_ ) and cultivated wheat ( _T. durum_ ) (Cinti et al., 2026) found similarly higher 387 adaxial than abaxial SD in cultivated but not in wild wheat, hinting at the possibility that 388 higher adaxial SDs have been selected for in breeding. In eudicots, SR depends on growth 389 environment, and increases under elevated vapour pressure deficit (VPD) in cotton (Devi and 390 Reddy, 2018) and in _Arabidopsis thaliana_ (Tulva _et al._, 2024). Although winter wheat SD was 391 affected by both, genotype and year, SR was not affected by year (Table 1), indicating that 392 SR was relatively stable across different environments. Similar stability of SR in different 393 environments was found for rice ( _O. sativa_ ) (Wang _et al._, 2023) and in CO 2 elevation 394 experiments for maize ( _Zea mays_ ) (Driscoll _et al._, 2006). On the other hand, a significant 395 GxY interaction effect on SR was found in spring wheat (Samantara _et al._, 2025), and few 396 wheat varieties showed changes in SR when grown under elevated CO 2 (Wall _et al._, 2023), 397 suggesting environmental plasticity in SR in at least some wheat genotypes. Our findings 398 suggest that in winter wheat, adjustments of stomatal density on the flag leaf surfaces in 399 response to growth conditions are largely coordinated, leading to similar SRs even when 400 SDs vary (Fig. 4A − E). 401 Having more stomata on the adaxial leaf surface may make the plant more susceptible to 402 pathogens (Muir, 2015). All fungal pathogens evaluated in our study enter the plant through 403 stomata (Schilder and Bergstrom, 1994; Moldenhauer _et al._, 2006; Solanki _et al._, 2019; 404 Bernasconi _et al._, 2023; Rebaque _et al._, 2025), except _B. graminis,_ which infects plant 405 epidermal cells (Prats _et al._, 2007). Typically, higher SD is associated with increased 406 susceptibility to pathogen invasion (Dutton _et al._, 2019; Tateda _et al._, 2019). We found no 407 correlation between cumulative disease index and SD in the pooled dataset, but there was a 408 negative correlation between adaxial SD and cumulative disease index in 2024 and abaxial 409 SD and cumulative disease index in 2025 (Fig. S3D,E). We also observed no correlation 410 between cumulative or individual disease indices and SR in any individual year nor in the 411 pooled data (Fig. S3A-E), finding no evidence for the hypothesis that the cost of adaxial 9 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 412 stomata is associated with susceptibility to pathogens (Muir, 2015, 2020). As variation of SR 413 in our study was minor (Fig. 4), the effects of stomatal traits may have been outweighted by 414 other mechanisms influencing disease susceptibility. 415 Higher SD is often associated with smaller stomata (Hetherington and Woodward, 2003; 416 Drake _et al._, 2013) and in our dataset, SD was also negatively associated with GCL (Fig. 417 S8C,D). Both small stomatal size and high SR are associated with faster stomatal responses 418 (Hetherington and Woodward, 2003; Franks and Farquhar, 2007; Drake _et al._, 2013; Woning 419 _et al._, 2026; Yang _et al._, 2026), potentially allowing for increased sensitivity of gas-exchange 420 to abiotic environment. For example, rice plants with smaller stomata and higher SD closed 421 their stomata faster in response to rising temperature and VPD, thereby improving water use 422 efficiency (WUE) (Caine _et al._, 2023). SD together with GCL and PL determine maximal 423 potential for gas exchange (g smax ). Our study showed that although g smax in both early and 424 late varieties was higher in warm 2023 compared with cooler 2025 (Fig. S9A-D), driven by 425 higher SDs in 2023 (Fig. 4, S10), operational g s values were lowest in 2023 and highest in 426 2025 (Fig. S11A-D), leading to a negative relationship between g smax and g s in the pooled 427 dataset (Fig. S13. Similar decoupling between anatomical capacity for gas-exchange and 428 realised conductance was also found in spring wheat, indicating strong plasticity of 429 operational g s (Chaplin _et al._, 2026b). We also found that abaxial g smax was positively related 430 with SD in all years and adaxial g smax in 2024 and 2025 (Fig. S10A,B), whereas negative 431 correlation between g smax and GCL was found only in the pooled data (Fig. S10C,D). These 432 results are in line with previous findings suggesting that SD and not stomatal size is the 433 primary factor affecting g smax (Ochoa et al., 2024; Wall et al., 2022). 434 Stomatal development in grasses occurs in distinct rows along veins, with the earliest stages 435 occurring basally, and during the development, cells move upwards as the leaf expands 436 longitudinally (Hepworth _et al._, 2018; Nunes _et al._, 2020), leading to larger but less dense 437 stomata during flag leaf expansion (Zanella _et al._, 2023). In line with this, we observed a 438 negative correlation between SD and FLA on both leaf surfaces (Fig. S8A,B). A similar 439 relationship has been noted before in winter wheat (Morgan and LeCain, 1991), and also in 440 e.g. eucalyptus (Franks _et al._, 2009), mangroves (Peel _et al._, 2017), barley (Samantara _et al._, 441 2026), and spring wheat (Samantara et al., 2025), and is line with adaptation of cell 442 expansion growth to environmental conditions, leading to altered stomatal spacing (Granier 443 and Tardieu, 2009; Tulva _et al._, 2024). 444 We found a positive association between both adaxial and abaxial SD and grain yield in the 445 pooled data and in some individual years (Fig. 5A,B). Similarly, other studies with winter 446 wheat have found a positive relationship between abaxial SD and grain yield (Yang _et al._, 447 2022; Du _et al._, 2026) and between adaxial SD and grain yield (Baruah _et al._, 2017). 448 However, in spring wheat, a negative relationship between adaxial but not abaxial SD and 449 grain yield was found in warm years 2022 and 2023 and was interpreted as costly water loss 450 via adaxial stomata (Samantara _et al._, 2025). Our study included two warm years 2023 and 451 2024, and relatively cool, wet 2025. Even in 2023, we found no negative association between 452 adaxial SD and grain yield, perhaps due to the different timing of development in spring and 453 winter wheat that may have made winter wheat less susceptible to the dry spring and early 454 summer of 2023. Spring and winter wheat may respond differently to contrasting climate 455 conditions, as demonstrated also by the effects of flag leaf removal on yield components in 456 winter and spring wheat in our experiment (Table 2). We found no significant relationship 457 between SD ratio and grain yield in any individual year or in the pooled data, perhaps due to 458 the minor variation in SR between years (Fig. 4, 5E). Stomatal size on both leaf surfaces was 459 negatively associated with grain yield in the pooled data (Fig 5C,D), in line with another 460 recent wheat field study (Chaplin _et al._, 2026b). In two out of three individual years, adaxial 10 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 461 stomatal size also negatively associated with grain yield (Fig. 5C,D), perhaps suggesting that 462 smaller, potentially faster adaxial stomata are particularly important for plasticity of flag leaf 463 gas-exchange. Indeed, plasticity of adaxial stomatal traits was suggested to be particularly 464 important for wheat adaptation to growth environment in the field (Chaplin _et al._, 2026b). 465 While some studies suggest that smaller stomata are not necessarily faster in wheat (Woning 466 _et al._, 2026) or across species (Elliott-Kingston _et al._, 2016), together, our findings support 467 the hypothesis that smaller, denser stomata that allow for faster responses and higher WUE 468 (Hetherington and Woodward, 2003; Franks and Farquhar, 2007; Drake _et al._, 2013; Caine 469 _et al._, 2023; Yoshiyama _et al._, 2024), are favourable under variable field conditions. 470 **4.3. Adaxial and abaxial stomatal conductance** 471 Stomatal conductance is dependent on stomatal density, size and aperture, and is an 472 important component of leaf CO 2 conductance, whereas surface-specific differences in 473 conductance may lead to different contribution of leaf surfaces to photosyhthesis (Muir, 2015; 474 Ouyang _et al._, 2017; Wall _et al._, 2022; Ochoa _et al._, 2024). We observed a positive linear 475 relationship between adaxial and abaxial g s in the pooled data but not in any individual years 476 (Fig. 6D). The absence of significant relationship within individual years suggests that 477 stomatal responses to environmental factors are independent in the two leaf surfaces in short 478 timescales, but coordination of g s between surfaces is evident in the broad range of g s values 479 measured across the years. Independent stomatal responses between leaf surfaces were 480 found before in wheat, whereas adaxial stomata responded faster to changes in light 481 intensity than abaxial stomata (Wall et al., 2022). Alternatively, our results may be explained 482 by differences in abiotic conditions between leaf surfaces at measurement time. 483 Stomatal conductance often positively correlates with grain yield (Fischer _et al._, 1998; 484 Aminian _et al._, 2011; Roche, 2015). In our study, adaxial and total g s were positively 485 associated with grain yield only in 2023 (Fig. 6A,C), confirming that in warm and dry years, 486 maintaining higher gas-exchange may be beneficial for yield in Nordic conditions (Jalakas _et_ 487 _al._, 2018). In the pooled dataset, the relationships between g s and grain yield were negative 488 (Fig. 6A-C). These results were driven by the data of the cool, wet 2025, where grain yield 489 was low due to lodging (Fig. 1,3). Single timepoint measurements of g s in June were likely 490 not sufficient to predict the contribution of stomata to grain yield throughout the growing 491 season. Stomatal anatomical traits may better explain variation in grain yield, as they capture 492 the environmental variation throughout plant growth and development in the field. 493 # 494 5. Conclusions 495 Contrary to expectations, flag leaf removal in winter wheat had little effect on grain yield 496 components, which is different from spring wheat and winter wheat grown in warmer 497 climates. Thus, in winter wheat, the function of the flag leaf may be compensated by other 498 leaves, as winter wheat usually has more tillers than spring wheat, and by longer time to 499 accumulate stem reserves for further translocation to grains. 500 The major aim of this study was to examine winter wheat stomatal traits and their 501 relationships with grain yield under field conditions across three years. We found that in 502 warm, dry years, plants produce smaller leaves with higher SD. Despite significant variation 503 in growth environment and SD, SR remained stable, indicating that wheat plants increase or 504 decrease the number of stomata proportionally on leaf surfaces. At the same time, stomata 505 on both leaf surfaces may respond differently to environmental factors, as we observed no 506 correlation between adaxial and abaxial g s values in individual years. 11 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 507 Higher stomatal densities and smaller stomatal sizes are associated with higher grain yield, 508 potentially reflecting the benefit of faster stomatal responsiveness achieved by smaller 509 stomata. Correlations between g s and grain yield were inconsistent, highlighting the variability 510 and plasticity of g s . Varieties with higher cumulative disease scores had lower grain yield as 511 expected, but higher SD was not accompanied by stronger infection by stomata-entering 512 pathogens, suggesting that traits other than SD largely determined disease susceptibility in 513 the field. 514 Our results provide insight into variation of leaf-surface specific stomatal traits under field 515 conditions in winter wheat, and on their associations with grain yield. Our findings suggest 516 that breeding towards varieties with smaller, faster stomata at high densities may be 517 beneficial for wheat yield under field conditions. 518 519 **Supplementary data** 520 **Table S1.** Winter wheat varieties used in field trials in 2023 − 2025 521 **Table S2.** Primary data supporting the study 522 **Figure S1.** Winter wheat thousand grain weight (TGW, g) during 2023 − 2025 in early and late 523 varieties 524 **Figure S2** . Fungal diseases severity in 2023 − 2025 525 **Figure S3.** Spearman correlation matrix of winter wheat cumulative disease index (CDI) and 526 disease index (DI) with winter wheat grain yield and stomatal traits 527 **Figure S4.** Winter wheat glaucousness scores across trial years of 2023 − 2025 528 **Figure S5.** Spearman correlations between winter wheat glaucousness, grain yield, and 529 stomatal traits 530 **Figure S6.** Winter wheat flag leaf area (FLA, cm [2] ) of early and late varieties in 2023 − 2025 531 **Figure S7.** Stomatal anatomical traits of winter wheat in trial years 532 **Figure S8.** Linear regressions between flag leaf area and stomatal traits 533 **Figure S9.** Anatomical maximum stomatal conductance (g smax, mmol m [-2 ] s [-1] ) of early and late 534 varieties of winter wheat in 2023 − 2025 535 **Figure S10.** Linear regression between stomatal anatomical maximum conductance (g smax, 536 mmol m [-2] s [-1] ) and stomatal density (SD, stomata mm [-2] ), guard cell length (GCL, µm), and 537 grain yield (kg ha [-1] ). 538 **Figure S11.** Stomatal conductance (g s, mmol s [-1] mm [-2] ) in early and late varieties of winter 539 wheat in 2023 − 2025 540 **Figure S12.** Utilization of anatomical maximum stomatal conductance (g s / g smax *100, %) in 541 early and late varieties of winter wheat in 2023−2025. 542 **Figure S13.** Linear regression between anatomical maximum stomatal conductance (g smax, 543 mmol m [-2 ] s [-1] ) and operational stomatal conductance (g s, mmol s [-1] mm [-2] ) 544 545 **Acknowledgements** 12 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 546 We thank Mikk Välbe for helping us with sample collection and support in laboratory. We are 547 also grateful to Made Ratassepp, Külli Jõessar, Jane Suvvi, Veste Vapper, and Merilin Saar 548 for the help with field works. 549 **Authors contribution** 550 H.H., E.I., E.M. and R.K. conceived and designed the study; E.I., P.J., I.T., M.K., K.S., 551 B.K.D., A.I., M.H., R.K., E.M., and H.H. collected and analysed data; E.I and H.H wrote the 552 manuscript with input from all authors. 553 **Conflict of interest** 554 No conflict of interest declared. 555 **Funding** 556 The research was funded by the Estonian Research Council (PSG404, PRG1620, 557 MOB3ERC114), Plant Breeding Program 2020 − 2030 PA1-SORT(funded by Estonian 558 Ministry of Regional Affairs and Agriculture), and by the revenue of the European Union's 559 Greenhouse Gas Emissions Trading Scheme (ETS) via Estonian Ministry of Regional Affairs 560 and Agriculture (MARTA). The research was conducted using the research infrastructure 561 “TAIM” funded by the Estonian Research Council (TARISTU24-TK28). 562 **Data availability** 563 All primary data to support the findings of this study are available in Supplementary Table S2 564 and will be deposited in a data repository upon acceptance of the manuscript. 565 13 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) # **References** **Aggarwal PK, Fischer RA, Liboon SP** . 1990. Source-sink relations and effects of postanthesis canopy defoliation in wheat at low latitudes. doi: 10.1017/s0021859600071045. **Ali MA, Hussain M, Khan M, Zulkiffal M, Anwar J, Sabir W, Zeeshan M** . 2010. SourceSink Relationship between Photosynthetic Organs and Grain Yield Attributes during Grain Filling Stage in Spring Wheat (Triticum aestivum). International Journal of Agriculture and Biology **12**, 509–515. **Aminian R, Mohammadi S, Hoshmand S, Khodombashi M** . 2011. Chromosomal analysis of photosynthesis rate and stomatal conductance and their relationships with grain yield in wheat (Triticum aestivum L.) under water-stressed and well-watered conditions. Acta Physiologiae Plantarum **33**, 755–764. **Baruah KK, Bharali A, Mazumdar A, Jha G** . 2017. Genotypic variation in carbon fixation, d13C fractionation and grain yield in seven wheat cultivars grown under well-watered conditions. Functional Plant Biology **44**, 809–819. **Bernasconi A, Lorrain C, Flury P, Alassimone J, McDonald BA, Sánchez-Vallet A** . 2023. Virulent strains of Zymoseptoria tritici suppress the host immune response and facilitate the success of avirulent strains in mixed infections. PLoS Pathogens **19**, e1011767. **Bi H, Kovalchuk N, Langridge P, Tricker PJ, Lopato S, Borisjuk N** . 2017. The impact of drought on wheat leaf cuticle properties. BMC Plant Biology **17**, 85. **Caine RS, Harrison EL, Sloan J, Flis PM, Fischer S, Khan MS, Nguyen PT, Nguyen LT,** **Gray JE, Croft H** . 2023. The influences of stomatal size and density on rice abiotic stress resilience. New Phytologist **237**, 2180–2195. **Carmo-Silva E, Andralojc PJ, Scales JC, Driever SM, Mead A, Lawson T, Raines CA,** **Parry MAJ** . 2017. Phenotyping of field-grown wheat in the UK highlights contribution of light response of photosynthesis and flag leaf longevity to grain yield. Journal of Experimental Botany **68**, 3473–3486. **Chaplin E, Tanaka E, Merchant A, Sznajder B, Trethowan R, Salter W** . 2026a. QTLs for heat-induced stomatal anatomy underpin gas exchange variation in field-grown wheat. Frontiers in Plant Science **17** . **Chaplin E, Tanaka E, Merchant A, Sznajder B, Trethowan R, Salter W** . 2026b. Field � Based Dissection of Stomatal Anatomy and Conductance Reveals Stable QTL Under Drought and Heat in Wheat. Plant, Cell & Environment **0:** **1–20.** https://doi.org/10.1111/pce.70974. **Cinti E, Carli A, Cerri M, Reale L, Centritto M, Zuluaga DL, Onofri A, Atzori G, Marino G,** **Haworth M** . 2026. Responses of Wild and Cultivated Tetraploid Wheat to CO2 Enrichment: Stomatal Optimization of Photosynthesis and Water Use Efficiency. Physiologia Plantarum **178**, e70946. **Community Plant Variety Office** . 2019. Protocol for distinctness, uniformity and stability tests: Triticum aestivum L. emend. Fiori et Paol. (Wheat). CPVO-TP/003/5. 2019 _in press_ . **Devi MJD, Reddy VR** . 2018. Transpiration Response of Cotton to Vapor Pressure Deficit and Its Relationship With Stomatal Traits. Frontiers in Plant Science **9**, 1572–1584. 14 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) **Drake PL, de Boer HJ, Schymanski SJ, Veneklaas EJ** . 2019. Two sides to every leaf: water and CO2 transport in hypostomatous and amphistomatous leaves. New Phytologist **222**, 1179–1187. **Drake PL, Froend RH, Franks PJ** . 2013. Smaller, faster stomata: scaling of stomatal size, rate of response, and stomatal conductance. Journal of Experimental Botany **64**, 495–505. **Dreccer MF, van Herwaarden AF, Chapman SC** . 2009. Grain number and grain weight in wheat lines contrasting for stem water soluble carbohydrate concentration. Field Crops Research **112**, 43–54. **Driever SM, Lawson T, Andralojc PJ, Raines CA, Parry MAJ** . 2014. Natural variation in photosynthetic capacity, growth, and yield in 64 field-grown wheat genotypes. Journal of Experimental Botany **65**, 4959–4973. **Driscoll SP, Prins A, Olmos E, Kunert KJ, Foyer CH** . 2006. Specification of adaxial and abaxial stomata, epidermal structure and photosynthesis to CO2 enrichment in maize leaves. Journal of Experimental Botany **57**, 381–390. **Du P, Sun Z, Zhu Y-H, Li F-M** . 2026. Linking root and leaf anatomical traits to yield variation among dryland wheat cultivars. Field Crops Research **344**, 110538. **Dutton C, Hõrak H, Hepworth C, Mitchell A, Ton J, Hunt L, Gray JE** . 2019. Bacterial infection systemically suppresses stomatal density. Plant, Cell & Environment **42**, 2411– 2421. **Duwayri M** . 1984. Effect of flag leaf and awn removal on grain yield and yield components of wheat grown under dryland conditions. Field Crops Research **8**, 307–313. **Elliott-Kingston C, Haworth M, Yearsley JM, Batke SP, Lawson T, McElwain JC** . 2016. Does Size Matter? Atmospheric CO2 May Be a Stronger Driver of Stomatal Closing Rate Than Stomatal Size in Taxa That Diversified under Low CO2. Frontiers in Plant Science **7** . **Erenstein O, Jaleta M, Mottaleb KA, Sonder K, Donovan J, Brown HJ** . 2022. Global trends in wheat production, consumption and trade. Wheat improvement: food security in a changing climate, 47–66. **Figueroa M, Hammond-Kosack KE, Solomon PS** . 2018. A review of wheat diseases—a field perspective. Molecular Plant Pathology **19**, 1523–1536. **Fischer RA, Rees D, Sayre KD, Lu Z-M, Condon AG, Saavedra AL** . 1998. Wheat Yield Progress Associated with Higher Stomatal Conductance and Photosynthetic Rate, and Cooler Canopies. Crop Science **38**, 1467–1475. **Franks PJ, Drake PL, Beerling DJ** . 2009. Plasticity in maximum stomatal conductance constrained by negative correlation between stomatal size and density: an analysis using Eucalyptus globulus. Plant, Cell & Environment **32**, 1737–1748. **Franks PJ, Farquhar GD** . 2007. The Mechanical Diversity of Stomata and Its Significance in Gas-Exchange Control. Plant Physiology **143**, 78–87. **Granier C, Tardieu F** . 2009. Multi-scale phenotyping of leaf expansion in response to environmental changes: the whole is more than the sum of parts. Plant, Cell & Environment **32**, 1175–1184. 15 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) **Guo J, Xu W, Yu X,** _**et al.**_ 2016. Cuticular Wax Accumulation Is Associated with Drought Tolerance in Wheat Near-Isogenic Lines. Frontiers in Plant Science **7** . **Hepworth C, Caine RS, Harrison EL, Sloan J, Gray JE** . 2018. Stomatal development: focusing on the grasses. Current Opinion in Plant Biology **41**, 1–7. **Hetherington A, Woodward I** . 2003. The role of stomata in sensing and driving environmental change. Nature 424, 901-908. Nature **424**, 901–8. **Hõrak H** . 2026. Functional and developmental differences between adaxial and abaxial stomata in amphistomatous leaves. New Phytologist **249**, 1675–1686. **Jalakas P, Tulva I, Kangor T, Sooväli P, Rasulov B, Tamm Ü, Koppel M, Kollist H,** **Merilo E** . 2018. Gas exchange-yield relationships of malting barley genotypes treated with fungicides and biostimulants. European Journal of Agronomy **99**, 129–137. **Johnson DA, Richards RA, Turner NC** . 1983. Yield, Water Relations, Gas Exchange, and Surface Reflectances of Near-Isogenic Wheat Lines Differing in Glaucousness. Crop Science **23**, 318–325. **Khanfri S, Boulif M, Lahlali R** . 2018. Yellow Rust (Puccinia striiformis): a Serious Threat to Wheat Production Worldwide. Notulae Scientia Biologicae **10**, 410–423. **Ledent JF, Jouret MF** . 1978. Relationship between stomatal frequencies, yield components and morphological characters in collections of winter wheat cultivars. Biologia Plantarum **20**, 287–292. **Lorenzo M, Assuero S g., Tognetti J a.** 2015. Low temperature differentially affects tillering in spring and winter wheat in association with changes in plant carbon status. Annals of Applied Biology **166**, 236–248. **Mahmood A, Alam K, Salam A, Iqbal S** . 1991. Effect of Flag Leaf Removal on Grain Yield, Its Components and Quality of Hexaploid Wheat. Cereal Research Communications **19**, 305–310. **McAusland L, Smith KE, Williams A, Molero G, Murchie EH** . 2021. Nocturnal stomatal conductance in wheat is growth-stage specific and shows genotypic variation. New Phytologist **232**, 162–175. **McKown AD, Guy RD, Quamme L, Klápšt** ě **J, La Mantia J, Constabel CP, El-Kassaby** **YA, Hamelin RC, Zifkin M, Azam MS** . 2014. Association genetics, geography and ecophysiology link stomatal patterning in Populus trichocarpa with carbon gain and disease resistance trade-offs. Molecular Ecology **23**, 5771–5790. **Melahat Avci B** . 2005. Effects of Removal of Some Photosynthetic Structures on Some Yield Components in Wheat. Tarım Bilimleri Dergisi **11**, 1. **Merah O, Deléens E, Souyris I, Monneveux P** . 2000. Effect of Glaucousness on Carbon Isotope Discrimination and Grain Yield in Durum Wheat. Journal of Agronomy and Crop Science **185**, 259–265. **Moldenhauer J, Moerschbacher BM, Van Der Westhuizen AJ** . 2006. Histological investigation of stripe rust (Puccinia striiformis f.sp. tritici) development in resistant and susceptible wheat cultivars. Plant Pathology **55**, 469–474. 16 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) **Monneveux P, Reynolds MP, González-Santoyo H, Peña RJ, Mayr L, Zapata F** . 2004. Relationships between Grain Yield, Flag Leaf Morphology, Carbon Isotope Discrimination and Ash Content in Irrigated Wheat. Journal of Agronomy and Crop Science **190**, 395–401. **Morgan JA, LeCain DR** . 1991. Leaf Gas Exchange and Related Leaf Traits among 15 Winter Wheat Genotypes. Crop Science **31**, 443–448. **Muir CD** . 2015. Making pore choices: repeated regime shifts in stomatal ratio. Proceedings of the Royal Society B: Biological Sciences **282**, 20151498. **Muir CD** . 2020. A Stomatal Model of Anatomical Tradeoffs Between Gas Exchange and Pathogen Colonization. Frontiers in Plant Science **11**, 518991–519003. **Nunes TDG, Zhang D, Raissig MT** . 2020. Form, development and function of grass stomata. The Plant Journal **101**, 780–799. **Ochoa ME, Henry C, John GP, Medeiros CD, Pan R, Scoffoni C, Buckley TN, Sack L** . 2024. Pinpointing the causal influences of stomatal anatomy and behavior on minimum, operational, and maximum leaf surface conductance. Plant Physiology **196**, 51–66. **Ouyang W, Struik PC, Yin X, Yang J** . 2017. Stomatal conductance, mesophyll conductance, and transpiration efficiency in relation to leaf anatomy in rice and wheat genotypes under drought. Journal of Experimental Botany **68**, 5191–5205. **Patterson FL, Ohm HW** . 1975. Compensating Ability of Awns in Soft Red Winter Wheat [1] . Crop Science **15**, 403–407. **Peel JR, Mandujano Sanchez MC, Lopez Portillo J, Golubov J** . 2017. Stomatal density, leaf area and plant size variation of Rhizophora mangle (Malpighiales: Rhizophoraceae) along a salinity gradient in the Mexican Caribbean. Revista de Biología Tropical **65** . **Peltonen-Sainio P, Jauhiainen L, Hakala K** . 2011. Crop responses to temperature and precipitation according to long-term multi-location trials at high-latitude conditions. The Journal of Agricultural Science **149**, 49–62. **Pemadasa MA** . 1979. Movements of Abaxial and Adaxial Stomata. New Phytologist **82**, 69– 80. **Prats E, Carver TLW, Gay AP, Mur LAJ** . 2007. Enemy at the Gates. Plant Signaling & Behavior **2**, 275–277. **Racz I, Hiri** ş **c** ă **u D, Berindean I, Kadar R, Muntean E, Tritean N, Russu F, Ona A,** **Muntean L** . 2022. The Influence of Flag Leaf Removal and Its Characteristics on Main Yield Components and Yield Quality Indices on Wheat. Agronomy **12**, 2545. **Rebaque D, Carrasco-López C, Krishnan P,** _**et al.**_ 2025. Zymoseptoria tritici stealth infection is facilitated by stage-specific downregulation of a β -glucanase. New Phytologist **248**, 3191–3207. **Richards RA, Rawson HM, Johnson DA** . 1986. Glaucousness in Wheat: Its Development and Effect on Water-use Efficiency, Gas Exchange and Photosynthetic Tissue Temperatures*. Australian Journal of Plant Physiology **13**, 465–473. **Roche D** . 2015. Stomatal Conductance Is Essential for Higher Yield Potential of C3 Crops. Critical Reviews in Plant Sciences **34**, 429–453. 17 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) **Samantara K, Ivandi E, Tulva I,** _**et al.**_ 2025. Higher adaxial stomatal density is associated with lower grain yield in spring wheat. New Phytologist **248**, 454–460. **Samantara K, Ivandi E, Tulva I, Jalakas P, Morozova D, Kärp M, Tamm Ü, Tamm I,** **Hõrak H, Merilo E** . 2026. Understanding Leaf Surface-Specific Stomatal Traits and Their Effects on Barley Grain Yield. Plant, Cell & Environment **n/a** . **Schilder AMC, Bergstrom GC** . 1994. Infection of wheat seed by _Pyrenophora tritici-_ _repentis_ . Canadian Journal of Botany **72**, 510–519. **Schneider CA, Rasband WS, Eliceiri KW** . 2012. NIH Image to ImageJ: 25 years of image analysis. Nature Methods **9**, 671–675. **Setter TL, Laureles EV, Mazaredo AM** . 1997. Lodging reduces yield of rice by self-shading and reductions in canopy photosynthesis. Field Crops Research **49**, 95–106. **Shahinnia F, Le Roy J, Laborde B, Sznajder B, Kalambettu P, Mahjourimajd S, Tilbrook** **J, Fleury D** . 2016. Genetic association of stomatal traits and yield in wheat grown in low rainfall environments. BMC Plant Biology **16**, 150. **Solanki S, Ameen G, Borowicz P, Brueggeman RS** . 2019. Shedding Light on Penetration of Cereal Host Stomata by Wheat Stem Rust Using Improved Methodology. Scientific Reports **9**, 7939. **Song K, Hong S, Shim S** . 2025. Responses of Photosynthetic Activity in Flag Leaves and Spikes as well as Seed Development of Wheat (Triticum aestivum L.) to Artificial Shading. Agronomy **15**, 2577. **Tateda C, Obara K, Abe Y, Sekine R, Nekoduka S, Hikage T, Nishihara M, Sekine K-T,** **Fujisaki K** . 2019. The Host Stomatal Density Determines Resistance to Septoria gentianae in Japanese Gentian. Molecular Plant-Microbe Interactions® **32**, 428–436. **Tulva I, Koolmeister K, Hõrak H** . 2024. Low relative air humidity and increased stomatal density independently hamper growth in young Arabidopsis. The Plant Journal **119**, 2718– 2736. **Wall S, Cockram J, Vialet-Chabrand S, Van Rie J, Gallé A, Lawson T** . 2023. The impact of growth at elevated [CO2] on stomatal anatomy and behavior differs between wheat species and cultivars. Journal of Experimental Botany **74**, 2860–2874. **Wall S, Vialet-Chabrand S, Davey P, Van Rie J, Galle A, Cockram J, Lawson T** . 2022. Stomata on the abaxial and adaxial leaf surfaces contribute differently to leaf gas exchange and photosynthesis in wheat. New Phytologist **235**, 1743–1756. **Wang H, Clarke JM** . 1993. Genotypic, intraplant, and environmental variation in stomatal frequency and size in wheat. Canadian Journal of Plant Science **73**, 671–678. **Wang Y, Noguchi K, Terashima I** . 2008. Distinct light responses of the adaxial and abaxial stomata in intact leaves of Helianthus annuus L. Plant, Cell & Environment **31**, 1307–1316. **Wang T, Zheng L, Xiong D, Wang F, Man J, Deng N, Cui K, Huang J, Peng S, Ling X** . 2023. Stomatal Ratio Showing No Response to Light Intensity in Oryza. Plants **12**, 66. **Watson DJ, Thorne GN, French SAW** . 1963. Analysis of Growth and Yield of Winter and Spring Wheats. Annals of Botany **27**, 1–22. 18 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) **Wei J, Hu K, Liu M, Liu Y, Tian W, Zhou Y, Fan L-M, Wang Y, Wang Y** . 2025. Guard cells on the adaxial and abaxial leaf surfaces use different compositions of potassium ion channels to drive light-induced stomatal opening. Nature Plants **11**, 1260–1269. **Wheat** **|** **USDA** **Foreign** **Agricultural** **Service** . 2026. https://www.fas.usda.gov/data/production/0410000. Accessed February 2026. **Woning N, Al-Salman Y, Kaiser E,** _**et al.**_ 2026. Revisiting the relationship between stomatal size and speed across species – a meta-analysis. New Phytologist **249**, 2338–2354. **Xiong D, Flexas J** . 2020. From one side to two sides: the effects of stomatal distribution on photosynthesis. New Phytologist **228**, 1754–1766. **Yang Y, Li N, Wu Y, Liu B, Li S, Tao L, Zhan Y, Yang Y, Ni X** . 2022. Key phenotypes related to wheat grain yield in a two-site multicultivar test. Agronomy Journal **114**, 2874– 2885. **Yang Y, Yoka AD, Wang S, Zhaoo W, Zhou Z, Hu W** . 2026. Higher adaxial-to-abaxial stomatal density ratio in C3 cereals than dicots facilitates stomatal kinetics and water use efficiency. Plant Physiology **201**, 1–13. **Yoshiyama Y, Wakabayashi Y, Mercer KL, Kawabata S, Kobayashi T, Tabuchi T,** **Yamori W** . 2024. Natural genetic variation in dynamic photosynthesis is correlated with stomatal anatomical traits in diverse tomato species across geographical habitats. Journal of Experimental Botany **75**, 6762–6777. **Yousufzai M-NK, Siddiqui KA, Soomro AQ** . 2009. Flag leaf stomatal frequency and its interrelationship with yield and yield components in wheat (Triticum aestivum L.). 41 **2**, 663– 666. **Zanella CM, Rotondo M, McCormick-Barnes C,** _**et al.**_ 2023. Longer epidermal cells underlie a quantitative source of variation in wheat flag leaf size. New Phytologist **237**, 1558– 1573. 19 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) **Table 1** . Results of factorial analyses of variance (Two-way ANOVA, p-values, n=3 plots per genotype per year) for genotype (G), year (Y), and GxY interaction in early and late varieties of winter wheat. Empty cells indicate no significant effects of studied factors (G, Y) or their interaction (GxY) on plant traits. |Early varieties|Late varieties| |---|---| |**Trait**<br>G <br>Y <br>GxY|G <br>Y <br>GxY| |**FLA**<br>0.00<br>0.00<br>0.00<br>**Flag leaf**<br>**width**<br>0.00<br>0.00<br>0.04<br>**SD, adaxial**<br>0.00<br>0.00<br>0.00<br>**SD, abaxial**<br>0.00<br>0.00<br>**SR**<br>0.03<br>**GCL, adaxial**<br>0.00<br>0.00<br>0.04<br>**GCL, abaxial**<br>0.00<br>0.00<br>0.04<br>**gs, adaxial**<br>0.00<br>**gs, abaxial**<br>0.00<br>0.00<br>0.00<br>**Total gs**<br>0.00<br>**gs ratio**<br>0.00<br>0.00<br>0.00<br>**gsmax, adaxial**<br>0.00<br>0.01<br>0.04<br>**gsmax, abaxial**<br>0.00<br>0.00<br>**Grain yield***<br>0.00<br>0.00<br>0.00<br>**TGW***<br>0.00<br>0.00<br>0.00<br>**Fine grain***<br>0.00<br>0.00<br>0.00|0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.02<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.00<br>0.04<br>0.00<br>0.04| - Statistical analyses were performed without early varieties Dykanka and Shanzhara due to lodging in 2025 **Table 2.** Factorial analysis of variance (Two-way ANOVA) of the effect of flag leaf (FL) removal on winter and spring wheat yield components (using genotype and treatment as factors, n=3 plots per genotype per year and n=23 varieties in the pooled dataset). **Spring wheat** **2023** **2024** **2025** **2023-2025** **2025** **Control** **No FL** **Control** **No FL** **Control** **No FL** **Control** **No FL** **Control** **No FL** **Grains** 45.10 45.16 38.81 39.66 42.07 42.75 42.22 42.96 42.93 40.14* **spike** **[-1]** **Grain** **weight** **spike** **[-1]** 2.44 2.38 1.82 1.77 1.31 1.07* 1.87 1.76 1.51 1.14* ***** Significant difference (p< 0.05) between control and flag leaf removal (No FL) treatment within one year 20 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) **Table 3.** Results of factorial analyses of variance (Two-way ANOVA, n=25) of winter wheat flag leaf stomatal density (SD) for year, leaf surface, and year x leaf surface interaction on the abaxial and adaxial leaf surfaces of winter wheat. Degr. of SS MS F p Freedom Intercept 710848.1 1 710848.1 19816.90 0.00 year 14424.2 2 7212.1 201.06 0.00 leaf surface 9568.6 1 9568.6 266.75 0.00 Y x leaf surface 158.1 2 79.1 2.20 0.11 Error 5165.4 144 35.9 # **Figure legends and Alt text for accessibility** **Figure 1.** Weather data in Jõgeva in 2023–2025. (A) Total precipitation from January to August in 2023–2025 in Jõgeva in comparison with the respective long-period average (LPA, 1991–2020); (B) Temperatures in 2023–2025 in Jõgeva in comparison with respective longperiod average (LPA, 1991–2020). Weather data from Estonian Environment Agency (Station location in Jõgeva N 58°44´59´´, E 26°24´54´´) Alt text: Graphs labelled A–B, A is showing precipitation in Jõgeva in 2023–2025 from January to August, and B is showing temperatures in Jõgeva in 2023–2025 from January to August. **Figure 2** . Figure illustrating the preparation of leaf impressions from a wheat flag leaf. First, we cut out the central part of the leaf. After that, we cut it in half crosswise from the centre and made epidermal impressions from the adaxial (upper) surface of one half and from the abaxial (lower) surface of the other half (figure created with BioRender.com) Alt text: Figure showing preparation of leaf impressions of wheat flag leaf. **Figure 3** . Grain yield (kg ha [-1] ) of early and late varieties of winter wheat in 2023–2025; (A) Grain yield of early varieties in 2023–2025; (B) Grain yield of late varieties in 2023–2025. In 2025, two varieties (Dykanka and Sanzhara) were lodged in all replicates and, therefore, were not included in the harvest data. Error bars represent standard error of the mean and different letters indicate statistically significant differences between groups. Tukey’s post-hoc test was performed following two-way analysis of variance (ANOVA, p<0.05, n=3). Alt text: Graphs labelled A–B, showing winter wheat grain yield across three trial years in early and late varieties with statistical annotations. **Figure 4.** Stomatal densities (SD, stomata mm [-2] ) and stomatal ratios of winter wheat flag − leaves of early and late varieties in 2023 2025. (A) Adaxial SD of early varieties; (B) Adaxial SD of late varieties; (C) Abaxial SD of early varieties; (D) Abaxial SD of late varieties; (E) Stomatal ratio (SR, no unit) of early varieties; (F) Stomatal ratio of late varieties. Error bars represent standard errors of the means, and different letters indicate statistically significant differences between groups. Tukey’s post-hoc test was performed following two-way analysis of variance (ANOVA, p<0.05, n=3). 21 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made [available under a CC-BY 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) Alt text: Graphs A − F, showing stomatal anatomical traits in 2023 − 2025. A is a bar chart of winter wheat adaxial stomatal density in early varieties. B is a bar chart of adaxial stomatal density in late varieties. C is a bar chart of abaxial stomatal density in early varieties. D is a bar chart of abaxial stomatal density in late varieties. E is a bar chart of adaxial to abaxial stomatal density ratio in early varieties. F is a bar chart of adaxial to abaxial stomatal density ratio in late varieties. All graphs are with statistical annotations. **Figure 5.** Linear regressions between grain yield and flag leaf stomatal traits of winter wheat. (A) Relationship between adaxial stomatal density (SD) and grain yield in 2023-2025; (B) − Relationship between abaxial SD and grain yield in 2023 2025; (C) Relationship between − adaxial guard cell length (GCL) and grain yield in 2023 2025; (D) Relationship between − abaxial GCLand grain yield in 2023 2025; (E) Relationship between adaxial to abaxial stomatal density ratio (SR) and grain yield; (F) Relationship between flag leaf area (FLA) and grain yield. Graphs include linear regression lines for statistically significant relationships (p < − − 0.05, n= 23 25 genotypes within each year and 69 75 in the pooled dataset, error bars show standard errors of the means). − Alt text: Graphs A F, showing linear regressions between winter wheat grain yield and stomatal densities, stomatal size, stomatal ratio, and flag leaf area in individual years and in the pooled data. Stomatal traits and flag leaf area are on the x-axis and grain yield on the yaxis. All the graphs with significant relationships are with regression lines, statistical significances and p-values. **Figure 6.** Linear regressions between stomatal conductance (g s, mmol m [-2] s [-1] ) and grain yield (kg ha [-1] ) on winter wheat. (A) Relationship between adaxial stomatal conductance and grain − yield in 2023 2025; (B) Relationship between abaxial stomatal conductance and grain yield in 2023 − 2025; (C) Relationship between total stomatal conductance and grain yield; (D) Relationship between abaxial and adaxial stomatal conductance. Graphs include regression − lines for statistically significant relationships (p < 0.05, n= 23 25 genotypes within each year and n=69 − 75 in the pooled dataset, error bars show standard error means). − Alt text: Graphs labelled A D. A is showing a linear relationship between winter wheat grain yield and adaxial stomatal conductance in individual years and in the pooled dataset of 2023 − 2025, B is showing a linear relationship between winter wheat grain yield and abaxial − stomatal conductance in individual years and in the pooled dataset of 2023 2025, C is showing a linear relationship between winter wheat grain yield and total stomatal − conductance in individual years and in the pooled dataset of 2023 2025, D is showing a linear relationship between abaxial and adaxial stomatal conductance in individual years and − in the pooled dataset of 2023 2025. All the graphs with significant relationships are with regression lines, statistical significances and p-values. 22 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made – – **Figure 1.** Weather data in Jõgeva in 2023 2025. (A) Total precipitation from January to August in 2023 – 2025 in Jõgeva in comparison with the respective long-period average (LPA, 1991 2020); (B) – – Temperatures in 2023 2025 in Jõgeva in comparison with respective long-period average (LPA, 1991 2020). Weather data from Estonian Environment Agency (Station location in Jõgeva N 58°44´59´´, E 26°24´54´´) – – Alt text: Graphs labelled A B, A is showing precipitation in Jõgeva in 2023 2025 from January to August, – and B is showing temperatures in Jõgeva in 2023 2025 from January to August. [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made **Figure 2** . Figure illustrating the preparation of leaf impressions from a wheat flag leaf. First, we cut out the central part of the leaf. After that, we cut it in half widthwise from the centre and made epidermal impressions from the adaxial (upper) surface of one half and from the abaxial (lower) surface of the other half (figure created with BioRender.com) Alt text: Figure showing preparation of leaf impressions of wheat flag leaf. [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made **Figure 3** . Grain yield (kg ha [-1] ) of early and late varieties of winter wheat in 2023 – 2025; (A) Grain yield – – (kg ha [-1] ) of early varieties in 2023 2025; (B) Grain yield (kg ha [-1] ) of late varieties in 2023 2025. In 2025, two varieties (Dykanka and Sanzhara) were lodged in all replicates and therefore were not included in the harvest data. Error bars represent standard error of the mean and different letters indicate statistically significant differences between groups. Tukey’s post-hoc test was performed following twoway analysis of variance (ANOVA, p<0.05, n=3). – Alt text: Graphs labelled A B, showing winter wheat grain yield across three trial years in early and late varieties with statistical annotations. [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made **Figure 4.** S tomatal densities (SD, stomata mm [-2] ) and stomatal ratios of winter wheat flag leaves of early and late varieties in 2023−2025. (A) Adaxial SD of early varieties; (B) Adaxial SD of late varieties; (C) Abaxial SD of early varieties; (D) Abaxial SD of late varieties; (E) Stomatal ratio (SR, no unit) of early varieties; (F) Stomatal ratio of late varieties. Error bars represent standard errors of the means, and different letters indicate statistically significant differences between groups. Tukey’s post-hoc test was performed following two-way analysis of variance (ANOVA, p<0.05, n=3). Alt text: Graphs A−F, showing stomatal anatomical traits in 2023−2025. A is a bar chart of winter wheat adaxial stomatal density in early varieties. B is a bar chart of adaxial stomatal density in late varieties. C is a bar chart of abaxial stomatal density in early varieties. D is a bar chart of abaxial stomatal density in late varieties. E is a bar chart of adaxial to abaxial stomatal density ratio in early varieties. F is a bar chart of adaxial to abaxial stomatal density ratio in late varieties. All graphs are with statistical annotations. [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made **Figure 5.** Linear regressions between grain yield and flag leaf stomatal traits of winter wheat. (A) Relationship between adaxial stomatal density (SD) and grain yield in 2023-2025; (B) Relationship − between abaxial SD and grain yield in 2023 2025; (C) Relationship between adaxial guard cell length − (GCL) and grain yield in 2023 2025; (D) Relationship between abaxial GCL and grain yield in 2023−2025; (E) Relationship between adaxial to abaxial stomatal density ratio (SR) and grain yield; (F) Relationship between flag leaf area (FLA) and grain yield. Graphs include linear regression lines for statistically significant relationships (p < 0.05, n= 23−25 genotypes within each year and 69−75 in the pooled dataset, error bars show standard errors of the means). − Alt text: Graphs A F, showing linear regressions between winter wheat grain yield and stomatal densities, stomatal size, stomatal ratio, and flag leaf area in individual years and in the pooled data. Stomatal traits and flag leaf area are on the x-axis and grain yield on the y-axis. All the graphs with significant relationships are with regression lines, statistical significances and p-values. [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.04.764249; this version posted November 9, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.11.04.764249) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made **Figure 6.** Linear regressions between stomatal conductance (g s, mmol m [-2] s [-1] ) and grain yield (kg ha [-1] ) − on winter wheat. (A) Relationship between adaxial stomatal conductance and grain yield in 2023 2025; − (B) Relationship between abaxial stomatal conductance and grain yield in 2023 2025; (C) Relationship between total stomatal conductance and grain yield; (D) Relationship between abaxial and adaxial stomatal conductance. Graphs include regression lines for statistically significant relationships (p < 0.05, n= 23−25 genotypes within each year and n=69−75 in the pooled dataset, error bars show standard errors of the means). − Alt text: Graphs labelled A D. A is showing a linear relationship between winter wheat grain yield and − adaxial stomatal conductance in individual years and in the pooled dataset of 2023 2025, B is showing a linear relationship between winter wheat grain yield and abaxial stomatal conductance in individual − years and in the pooled dataset of 2023 2025, C is showing a linear relationship between winter wheat − grain yield and total stomatal conductance in individual years and in the pooled dataset of 2023 2025, D is showing a linear relationship between abaxial and adaxial stomatal conductance in individual years − and in the pooled dataset of 2023 2025. All the graphs with significant relationships are with regression lines, statistical significances and p-values.