The effects of short-term waterlogging on the lint yield and yield components of cotton with respect to boll position

Europ. J. Agronomy 67 (2015) 61–74

Contents lists available at ScienceDirect

European Journal of Agronomy journal homepage: www.elsevier.com/locate/eja

The effects of short-term waterlogging on the lint yield and yield components of cotton with respect to boll position Jie Kuai b , Zhiguo Zhou a,∗ , Youhua Wang a , Yali Meng a , Binglin Chen a , Wenqing Zhao a a b

Key Laboratory of Crop Physiology Ecology, Ministry of Agriculture, Nanjing Agricultural University, Nanjing, 210095 Jiangsu Province, PR China College of Plant Science & Technology, Huazhong Agricultural University, Wuhan 430070, PR China

a r t i c l e

i n f o

Article history: Received 18 June 2014 Received in revised form 25 March 2015 Accepted 27 March 2015 Available online 16 April 2015 Keywords: Cotton Waterlogging days Fruiting positions Boll formation Biomass distribution

a b s t r a c t The objectives of this study were to determine the influence of waterlogging on the lint yield and yield components, biomass accumulation and distribution in the cotton boll with respect to boll position. Cottons were subjected to waterlogging 66 days after the seedlings were transplanted into ponds created by maintaining 1–2 cm of water on the soil surface for 0, 3, 6, 9 or 12 d. The ponds were then drained to allow recovery. The tap root and main stem biomass were significantly reduced and the plant biomass decreased resulting from decreased biomass in fruiting branch 1–8 (FB1–8 ) after waterlogging. The vegetative and reproductive biomass of FB9–16 increased by altered fruiting dynamics resulted from previous waterlogging, and the highest biomass was measured in 6 days of waterlogging (WL6 ). Waterlogging of 3, 6, 9 and 12 d resulted in a 16.0%, 24.1%, 39.5% and 50.2% reduction in lint yield, due to decreased boll number. Altered fruiting dynamics after waterlogging increased the contribution of bolls at position 3 on FB9–16 to the total yield due to an increase in boll number. The proportion of the boll wall and the seed biomass increased, while the proportion of the fiber biomass and the fiber/seed ratio decreased progressively with waterlogging duration. Insufficient assimilates were preferred compensation in boll number to boll biomass. These findings demonstrate that the bolls at various positions differed in their response to waterlogging and that even short periods (3 d) of waterlogging can have considerable long-term effects on the growth of cotton. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Waterlogging is a major abiotic stress that severely constrains crop growth and productivity in many regions (Ahsan et al., 2007; Jackson and Colmer, 2005). During waterlogging, O2 in the soil is rapidly depleted, resulting in hypoxic or anoxic conditions within a few hours. This results in a severe decrease in the energy status of the root cells, affecting important metabolic processes of plants (Sairam et al., 2009). Cotton (Gossypium hirsutum L.) is an important economic crop with poor adaptation to waterlogging (Hodgson and Chan, 1982). Waterlogging is considered to be a major problem in global cotton production (Gillham et al., 1995). Waterlogging occurs frequently in the Yangtze River Valley in China, especially

Abbreviations: CV(%), coefficient of variation; DPA (d), days post anthesis; FB, fruiting branch; MDT (◦ C), mean daily temperature; MDTmin (◦ C), mean daily minimum temperature; MDTmax (◦ C), mean daily maximum temperature; Pn, net photosynthetic rate; RSWC (%), relative soil water content; WL, waterlogging. ∗ Corresponding author at: Department of Agronomy, Nanjing Agricultural University, Nanjing 210095, PR China. Tel.: +86 25 84396813; fax: +86 25 84396813. E-mail address: [email protected] (Z. Zhou). http://dx.doi.org/10.1016/j.eja.2015.03.005 1161-0301/© 2015 Elsevier B.V. All rights reserved.

during the flowering and boll formation stages (Yang et al., 2012), and can greatly influence the growth and yield of cotton (Bange et al., 2004; Hodgson and Chan, 1982). Previous researches demonstrated that waterlogging can reduce cotton yield by 10% (Bange et al., 2004) to 40% (Hodgson, 1982). Field experiments with waterlogged cotton (>16 h) reported reductions in lint yield resulting from a decrease in boll number (Bange et al., 2004; Hodgson, 1982; Hodgson and Chan, 1982). The decreased boll number under waterlogged occurred mainly in response to a decreased overall growth (i.e., height, nodes, leaf area) and lower radiation use efficiency (Bange et al., 2004). In addition, waterlogging can cause significant reductions in cotton stem elongation, shoot mass, root mass and leaf number (Christianson et al., 2010) and can alter the foliar nutrient concentrations (Conaty et al., 2008). Waterlogging has been observed to promote fruit shedding because of inadequate aeration of the roots (Longnecker and Erie, 1968). Studies on cotton revealed that waterlogging significantly decreased the photosynthetic rate (Conaty et al., 2008; Meyer et al., 1987). Milroy and Bange (2013) observed that the radiation use efficiency (RUE) did not recover from a single large waterlogging event early in cotton development and remained low for the rest

62

J. Kuai et al. / Europ. J. Agronomy 67 (2015) 61–74

of the season. However, the response of the lamina net photosynthetic rate (Pn) to repeated waterlogging suggested some degree of acclimation. The indeterminate growth habits of cotton result in the formation of bolls on different fruiting branches at different times and under different environmental conditions and therefore different rates of growth and development (Bondada and Oosterhuis, 2001). Previous reports indicated that fruiting position 1 (the first sympodial position closest to the main stem) contributed more to the total cotton yield than other fruiting positions on the same sympodial branch (Boquet and Moser, 2003; Heitholt, 1993; Jenkins et al., 1990; Pettigrew, 2004). However, in the Yangtze River Valley of China, where high seed cotton yields (7657 kg ha−1 ) were measured, the boll retention rate on the distal sites (i.e., fruiting position 3 and higher) was as high as 58.8% (Gu et al., 2010). Thus, the high boll retention rate of the distal fruiting positions may be key to improving the fiber yield of cotton under conditions of high yield potential. Because cotton is an indeterminate crop, bolls on different fruiting branches and positions exist at different stages of development. Therefore, should waterlogging occur during flowering or boll development, the indeterminate growth of cotton may result in differences in the response to waterlogging. Many previous studies have focused on the effects of waterlogging on the yield of the entire cotton plant. There is no information available on the effects of waterlogging on cotton yield or yield components with respect to boll position. Therefore, the objectives of this study were to determine the influence of waterlogging on the lint yield and yield components, biomass accumulation and distribution in the cotton bolls with respect to boll position. 2. Materials and methods 2.1. Plant materials and growth conditions Cotton (G. hirsutum L.) (cv. Siza 3) seeds were sowed on 8 April 2011 and 2012 at the experimental station of Nanjing Agricultural University in Nanjing (32◦ 02 N and 118◦ 50 E), Jiangsu Province, China. Individual healthy and uniform seedlings with three true leaves were transplanted into 4-m-long, 4-m-wide and 1.5-m-high ponds on 10 May. The ponds were covered with a transparent waterproof film above the crop canopy to exclude the effects of rainfall during waterlogging. Each pond contained five rows of seedlings; the row and plant spacing was 75 × 25 cm. 2.2. Experimental design and treatments The experiment consisted of five waterlogging treatments (i.e., 0, 3, 6, 9 and 12 days of waterlogging) with three replicates laid out in a randomized complete block design. The irrigation of the experimental ponds was controlled manually and was determined using the relative soil water content (RSWC) method outlined by Weatherley (1950). The plants were well-watered before and after the waterlogging event, and the soil water content was maintained between 70% of field capacity (the lower soil water limit) and 80% of field capacity (the upper soil water limit). Five soil water treatments were established on 15 July, 66 days after transplanting the seedlings into the ponds before the plants had cut-out. Hence, the plants were still actively growing and producing new flowers while boll development was occurring at the lower nodes. The groups consisted of a well-watered control (WL0 ) with RSWC maintained at 70–80% of field capacity and four soil waterlogging treatments comprised of waterlogging for 3, 6, 9 and 12 d (i.e., WL3 , WL6 , WL9 and WL12 , respectively). Waterlogging was achieved by maintaining a 1–2 cm water layer on the soil

surface until the evening of the 3rd, 6th, 9th and 12th day, when the water was removed by draining the ponds. The soil surface temperature was measured using a thermometer at noon before the waterlogging treatments were imposed and on the day the waterlogging was terminated. Three thermometers were installed in each pond at a soil depth of 5–10 cm located i) adjacent to a row of plants, ii) in the middle of two rows of plants, and iii) between positions i) and ii). At the same time and positions, the soil oxidation–reduction potential (Eh) was measured using a combined platinum–calomel electrode (FJA-4; Nanjing Zhuandi Instrument Cor.-Ltd., Nanjing, China). 2.3. Morphological indices The bolls were mapped by fruiting branch and position. The first sympodial position closest to the main stem was designated as fruiting position 1, and successive boll positions were designated as fruiting position 2 and fruiting position 3. The bolls with position numbers higher than three were classified as fruiting position 3. The plant height, number of fruiting branches, number of fruiting positions and number of bolls were determined on eight plants per pond every 7 days from the initiation of squaring to boll opening. The rate of boll shedding was also calculated for these eight plants. 2.4. Lint yield, yield components, biomass accumulation and distribution Measurements related to yield and biomass were conducted on the same day across the waterlogging treatments at maturity, and thus, a different number of days following the termination of waterlogging occurred in different treatments and different fruiting branches. In addition, measurements were all taken after the waterlogging event had been terminated, and thus, the plants may have been able to recover from the waterlogging event. Eight mature plants from each pond were slowly uprooted, and the taproot and large lateral roots were retained. Next, the plants were separated into the root, main stem, vegetative organs (leaves, petioles and branches) and reproductive organs (squares, flowers and bolls). The vegetative and reproductive organs were classified according to the fruiting branches (i.e., FB1-4 , FB5–8 , FB9–12 and FB13–16 ). The samples were placed in an electric fan-assisted oven at 105 ◦ C for 30 min and then dried to a constant mass at 80 ◦ C before being weighed. The lint yield of FB1–4 , FB5–8 , FB9–12 and FB13–16 was determined using data obtained from the biomass harvest above and boll number from each corresponding fruiting branch. All indices were determined on eight representative plants of each replicate. The lint yield (g m−2 ) was calculated as the total lint biomass in one square meter. Using the same plants, tagged cotton bolls on FB2–3 , FB6–7 , FB10–11 and FB14–15 were hand-harvested according to the fruiting branch and position and were separated into boll walls, seed and fiber. The boll walls and seed were placed in an electric fan-assisted oven at 105 ◦ C for 30 min and were dried at 70 ◦ C to a constant mass before weighing. The fiber was dried at 50 ◦ C to a constant mass before weighing. The total biomass and the proportion of the boll walls, seed and fiber at different fruiting branches and positions were calculated. The fiber and seed biomass were measured to calculate the lint percentage. 2.5. Weather data The mean daily temperature (MDT), mean daily maximum temperature (MDTmax), mean daily minimum temperature (MDTmin) and the day degrees from May to October during the two growing seasons were collected from a local weather station adjacent to the

J. Kuai et al. / Europ. J. Agronomy 67 (2015) 61–74

experimental site (100 m). A base temperature of 15 ◦ C was used to calculate the day degrees accumulation.

63

Interestingly, it was noticed that the shedding rate in waterlogged cotton was lower in 2012, compared with the shedding rate of WL0 (Table 2).

2.6. Statistical analysis 3.5. Lint yield and yield components The data were subjected to an analysis of variance using the SPSS statistical package version 17.0. Differences at the p < 0.05 level are considered statistically significant using the least significant difference (LSD) test and are indicated by different letters above the bars or numbers. The coefficient of variation (CV, %) was calculated as the ratio of the standard deviation (of all treatments) to the mean. 3. Results 3.1. Weather data The monthly mean daily temperature (MDT), mean daily maximum temperature (MDTmax), mean daily minimum temperature (MDTmin) and number of day degrees measured during the growing season from May to October differed between the two years (Fig. 1). The air temperature (MDT, MDTmax and MDTmin) and number of day degrees were higher in 2012 than in 2011. In particular, a higher MDT and MDTmax occurred in July and August in 2012. The weather data recorded during the waterlogging indicated that the MDT, MDTmax and MDTmin in 2012 were 2.0 ◦ C, 2.7 ◦ C and 0.7 ◦ C higher, respectively, compared with 2011 (Fig. 1A). A similar trend in daily solar radiation of 2011 and 2012 was observed during the growing seasons and the waterlogging event (Fig. 1B). 3.2. Relative soil water content, soil temperature and soil oxidation–reduction potential The relative soil water content was maintained at a saturated level during the waterlogging events and was progressively reduced to the soil water status of the control after drainage. The waterlogged conditions for a period of 12 days caused a significant increase in the soil surface (5–10 cm depth) temperature of 2.68% and 2.62% in 2011 and 2012, respectively, compared with WL0 . The average reduction in the soil oxidation–reduction potential (Eh) over the two years in WL3 , WL6 , WL9 and W12 was 162.8, 267.8, 389.0 and 460.2 mV, respectively, compared with WL0 , indicating that the plots were well waterlogged and O2 in the soil was depleted with waterlogging duration (Fig. 2). 3.3. Biomass accumulation in cotton plants The tap root, main stem and total plant biomass were significantly reduced by waterlogging. The reduction in the total plant biomass after waterlogging resulted from a decrease in the FB1–4 and FB5–8 biomass. In contrast, the biomass of the vegetative and reproductive organs of FB9–12 and FB13–16 were increased by waterlogging, and the highest biomass was measured in the WL6 treatment. There were significant interactions between years and waterlogging for the biomass of the root, main stem, total plant, and the vegetative and reproductive organs. In addition, the effects of waterlogging on the biomass of the vegetative and reproductive organs differed in different years and fruiting branches (Table 1). 3.4. Morphological indices of cotton Short-term waterlogging during flowering and boll formation had a significant effect on the growth of cotton. The plant height, total number of fruiting branches and total number of fruiting positions were reduced by waterlogging relative to the control, even after waterlogging ceased (Fig. 3). In contrast, the shedding rate of cotton increased during the waterlogging event (Fig. 3).

The lint yield, which was calculated from the boll biomass, lint percentage and number of bolls, was significantly affected by the waterlogging (Table 2). The number of bolls was the component most sensitive to waterlogging. The lint yield in 2011 was reduced by 19.2%, 21.5%, 39.1% and 53.5% in WL3 , WL6 , WL9 and WL12 , respectively, primarily because of reductions of 12.8%, 15.2%, 28.0% and 40.8% in the number of bolls. Similarly, the lint yield in 2012 was reduced by 12.0%, 27.0%, 39.8% and 46.7%, respectively, in WL3 , WL6 , WL9 and WL12 , primarily because of reductions of 4.7%, 12.9%, 24.0% and 25.8% in the number of bolls and reductions of 5.9%, 11.8%, 13.7% and 15.7% in the boll biomass. Significant Y × WL interactions were observed for the boll biomass (P < 0.01) and lint percentage (P < 0.05) for the bolls at position 3 or higher. Meanwhile, significant interactions between the years, fruiting branches and waterlogging days were observed for the number of bolls (P < 0.01) and boll biomass (P < 0.01) for the bolls at position 3 or higher. However, these interactions were not significant for the bolls at position 1–2 (Table 3). Further analyses were conducted on the changes in the lint yield and yield components with respect to boll position, which indicated that the number of bolls, boll biomass, lint percentage, lint yield and contribution of the bolls on FB1–4 and FB5–8 to the total yield decreased as the duration of waterlogging increased (Table 4a). In contrast, the lint yield of FB9–12 and FB13–16 increased in the waterlogged treatments compared with WL0 , especially for bolls located at position 3 or higher, whereas the lint percentage for bolls on different branches decreased as the duration of waterlogging increased. The increase in the number of bolls and boll biomass in 2011 and the increase in the number of bolls in 2012 resulted in a higher lint yield on FB9–12 and FB13–16 ; the highest lint yield was recorded in the WL6 . Furthermore, the contribution of the bolls at position 3 or higher on FB9–12 and the bolls on FB13–16 to the total lint yield increased with an increase in the duration of the waterlogging. 3.6. Biomass accumulation and distribution within the cotton boll The accumulation of biomass in the cotton boll after waterlogging differed among the fruiting branches and fruiting positions (Table 5). The biomass of the boll wall and the fiber of the bolls at position 1–2 on FB2–3 and FB6–7 were significantly decreased after waterlogging, and these values progressively decreased as the duration of the waterlogging increased. Waterlogged cotton bolls at position 3 or higher on FB2–3 and FB6–7 had a higher boll wall biomass, whereas the fiber biomass was lower compared with WL0 . The highest boll wall biomass was measured in WL12 in 2011 and WL9 in 2012. The pattern of the boll wall biomass and fiber biomass of the bolls on FB10–11 and FB14–15 was similar to the bolls at position 3 or higher on FB2–3 and FB6–7 . The proportion of the boll wall and seed biomass increased while the proportion of the fiber biomass and the fiber/seed ratio progressively decreased as the duration of the waterlogging increased for the bolls on different fruiting branches and at different positions (Fig. 4). The “source–sink” correlations were studied after the waterlogging ceased. A significant positive correlation was observed between the vegetative and reproductive organs and boll number. There was a significant positive correlation between the net photosynthetic rate [Pn, data presented in Kuai et al. (2014)] and the boll and fiber biomass. In addition, the Pn and biomass of the reproductive organs were negatively correlated with the proportion of the

64

Table 1 The statistical significance of the effects and interactions of short-term waterlogging on biomass of the whole plant, root, stem and vegetative organs, reproductive organs at different fruiting branches in 2011 and 2012. Year

Days of waterlogging

Stem (g)

Root (g)

Total biomass (g plant−1 )

Vegetative organs (g)

Reproductive organs (g)

2011

Significance Year (Y) Fruiting branches (FB) Waterlogging (WL) Y × FB Y × WL FB × WL Y × FB × WL a b c d e * **

Vegetative organs (g)

Reproductive organs (g)

FB9–12

Vegetative organs (g)

Reproductive organs (g)

FB13–16

WL0 WL3 WL6 WL9 WL12 CV%c LSD0.05

15.83a 14.69ab 13.77b 14.69ab 13.08b 7.24

41.35a 35.96b 28.90c 26.32 cd 22.63d 24.33

242.16a 212.24b 206.46b 176.17c 151.13d 17.71

45.94a 32.01b 26.92c 21.07d 17.00e 39.36

11.18a 10.52a 8.14b 7.61b 5.11c 28.59

41.47a 30.16b 24.82c 20.85d 17.18e 35.19

9.36a 8.82a 7.58b 5.39c 4.75c 28.45

39.08b 36.67c 42.73a 35.53c 38.87b 7.16

4.66c 6.28b 8.07a 7.22ab 8.08a 20.91

26.83c 30.22b 37.86a 30.04b 26.74c 14.92

6.46c 6.92bc 7.68a 7.46ab 7.69a 7.43

*,d

**

**

**

**

**

**

**

**

**

*

WL0 WL3 WL6 WL9 WL12 CV% LSD0.05

18.06a 16.01b 13.57c 12.02d 10.77e 20.99

50.58a 44.32b 34.25c 28.24d 22.63e 31.78

286.21a 263.73b 238.76c 212.23d 187.61e 16.55

64.69a 51.88b 46.76c 40.75d 32.64e 25.47

15.83a 14.51a 8.14b 7.41b 6.99b 40.08

54.97a 43.37b 44.05b 38.56c 30.53d 21.05

14.91a 13.67b 11.70c 9.04d 7.96e 25.82

7.01c 7.29bbc 7.86ab 7.94ab 8.06a 5.98

19.20c 26.37ab 27.77a 24.97b 24.52b 13.27

8.53c 10.70a 10.94a 9.62b 9.25bc 10.29

**

**

**

**

**

**

**

32.44b 35.61a 33.71ab 33.67ab 34.26ab 3.38 NS

*

**

**

1.95 NDe 54.68** ND 17.00** ND ND

47.40** 621.65** 138.02** 1261.78** 6.11** 35.38** 21.89**

844.04** 238.95** 576.47** 38.10** 6.19** 11.59** 7.38**

545.57**

239.81**

561.96**

110.52**

748.95**

238.31**

WL0 , WL3 , WL6 , WL9 and WL12 stand for waterlogging days of 0, 3, 6, 9 and 12 d, respectively. Values followed by different letters within the same column are significantly different at P = 0.05 probability level Each data represents the mean of three replications (eight plants in each replication). CV%, cofficient of variation, LSD, least significant difference. NS, not significant. ND, no data. Significant differences at 0.05 probability levels. Significant differences at 0.01 probability levels.

J. Kuai et al. / Europ. J. Agronomy 67 (2015) 61–74

2012

b

Reproductive organs (g)

FB5–8

FB1–4 a

Vegetative organs (g)

J. Kuai et al. / Europ. J. Agronomy 67 (2015) 61–74

65

Fig. 1. A(a) Monthly average of mean daily temperature (mean, maximum and minimum) and day degreees at the experimental site during the cotton growing period (May–October) and A(b) changes of daily temperature (mean, maximum and minimum) during waterlogging in 2011 and 2012; B(a) Daily solar radiation from May to Octorber and B(b) Daily solar radiation during waterlogging. MDTmin, MDTmax stand for mean daily temperature, mean daily minimum temperature and mean daily maximum temperature. A base temperature of 15◦ C was used to calculate day degrees accumulation.

seed and positively correlated with the proportion of the fiber and the fiber/seed ratio (Table 6). 4. Discussion Previous studies have measured the effect of waterlogging on photosynthesis at the single leaf level and on lint yield of the entire cotton plant (Bange et al., 2004; Hodgson, 1982; Hodgson and Chan, 1982). In most studies, the boll number and biomass were determined from bolls collected from the entire plant (Yang et al., 2012), and the effect of waterlogging on the number of bolls and boll biomass at different fruiting branches and positions was averaged. The results from the current study indicated that there was compensation in the number of bolls on the upper fruiting branches

and higher boll positions, indicating that this compensation is limited to long-season environments. Brook et al. (1992a, b, c) showed a strong interaction between yield level and compensation in cotton. This finding may explain why there was no difference when boll biomass and sometimes yield were calculated on a whole-plant basis (Bange et al., 2004; Conaty et al., 2008). Waterlogging inhibits respiration in the root as a result of an insufficient supply of oxygen (Das et al., 2009). An earlier paper demonstrated that high temperatures increase a plant’s oxygen demand through increased respiration, and reduced oxygen availability in the saturated soil can cause severe damage to plants even under short-term waterlogging (Wang et al., 2009). As observed in this study, the monthly daily temperature (mean, maximum and minimum), day degrees and daily solar radiation at the

66

J. Kuai et al. / Europ. J. Agronomy 67 (2015) 61–74

Fig. 2. Effects of short-term waterlogging on soil relative water content, surface temperature and the oxidation–reduction potential (5–10 cm depth). BW, before waterlogging. TW, the day terminating waterlogging. WL0 , WL3 , WL6 , WL9 and WL12 stand for waterlogging days of 0, 3, 6, 9 and 12 d, respectively. Data (n = 3) are mean values ± SD, calculated from three replications.

Fig. 3. Dynamic changes of morphological indices for cotton growth after short-term waterlogging (a) plant height; (b) total fruiting branches number; (c) fruiting positions number; (d) boll number; (e) shedding rate. Data (n = 3) are mean values ± SD, calculated from three replications (eight plants in each replication). Open diamond, closed triangle, open triangle, closed circle and open circle represent WL0 , WL3 , WL6 , WL9 and WL12 (waterlogging for 0, 3, 6, 9 and 12 days), respectively. The arrows signify the first of each month. * and ** , significant differences at 0.05 and 0.01 probability levels, respectively.

experimental site during the cotton growing period, especially from July to October in 2012 were higher, compared with those of 2011 (Fig. 1). A similar trend for those indictors was observed during waterlogging event. These differences resulted in a much greater reduction in Eh during waterlogging in 2012 (Fig. 2), which might bring more severe harm to the cotton. Waterlogging can induce several physiological disturbances including a reduction in growth, photosynthesis and boll formation, which result in a decreased yield (Bange et al., 2004; Christianson et al., 2010; Guang et al., 2012; Hodgson and Chan, 1982; Meyer

et al., 1987). The results from the current study agree with these findings. The biomass of the tap root, main stem and whole plant was significantly reduced by waterlogging; similar results were observed by Bange et al. (2004), who found a reduction in biomass accumulation and shoot dry matter caused by waterlogging. Aerial environment determines the intensity of the response to waterlogging (Grassini et al., 2007). In the present study, a greater reduction in the tap root and stem biomass was observed after waterlogging in 2012 compared with 2011, and the difference in weather conditions between 2011 and 2012 (higher temperatures and daily

J. Kuai et al. / Europ. J. Agronomy 67 (2015) 61–74

67

Table 2 Morphological indices, lint yield and yield componentsa for cotton at harvest after short-term waterlogging in 2011 and 2012. Years

Days of waterlogging

Height (cm)

Fruiting branches number

Fruiting positions number

Shedding rate (%)

Boll number

Boll biomass Lint percentage (g) (%)

Lint yield (gm−2 )

2011

WL0 b WL3 WL6 WL9 WL12 CV%d

104.9ac 90.7b 88.1bc 85.1 cd 82.5d 9.7

19.4aa 17.2b 16.2c 15.4d 14.9d 10.7

70.4a 64.2b 62.0b 55.8c 53.6c 11.0

65.9 67.3 65.2 67.7 72.3 4.1

25.0a 21.8b 21.2b 18.0c 14.8d 19.3

4.5a 4.5ab 4.4b 4.3c 4.3c 2.46

41.8a 40.3b 38.7c 37.5d 35.8e 6.03

256a 205 b 201b 156 c 119 d 27.92

2012

WL0 WL3 WL6 WL9 WL12 CV%

94.1a 90.4a 85.6b 80.7c 83.5bc 6.2

16.8a 15.6b 15.0bc 14.2c 14.2c 7.2

71.4a 64.2b 58.2c 53d 50.8d 14.1

67.4 60.6 62.6 66.6 66.5 4.5

23.3a 22.2b 20.3c 17.7d 17.3d 13.2

5.1a 4.8b 4.5c 4.4d 4.3e 7.68

41.3a 40.1b 39.3c 37.5d 36.2e 5.17

274a 241a 200b 165c 146c 25.72

a

Indices were determined at maturity. WL0 , WL3 , WL6 , WL9 and WL12 stand for waterlogging days of 0, 3, 6, 9 and 12 d, respectively. c Values followed by different letters within the same column are significantly different at P = 0.05 probability level. Each data represents the mean of three replications (eight plants in each replication). d CV%, cofficient of variation. b

solar radiation were recorded in 2012) resulted in a significant interaction between years and waterlogging, suggesting that conditions in the aerial environmental during waterlogging (i.e., high radiation and/or temperatures) could determine higher transpiration rates, thus increasing plant stress. Grain yield in sun flower (Helianthus annuus L.) grown in plastic pots was reduced by 20% under waterlogging without shading, while 5% reduction in grain yield was observed under waterlogging and shading condition. Previous research showed a greater reduction in the photosynthesis, chlorophyll concentration and root viability of creeping bentgrass (Agrostis palustris Huds.) maintained in growth chambers subjected to waterlogging in combination with high temperatures (Huang et al., 1998a,b). However, our research was not performed in a controlled environment. Further investigation in growth chambers is needed before we can determine the predominate factor among the weather conditions that affect waterlogging. Fruit development from different fruiting branches and different fruiting positions is not comparable (Errington, 2013), which might result in different responses to waterlogging. Significant interactions revealed that the effects of waterlogging days on the vegetative and reproductive biomass accumulation differed by years and fruiting branches. As the study showed, the vegetative and reproductive biomass of FB1–8 was significantly reduced under waterlogged conditions, as compared with WL0 (Table 1). The decreased vegetative growth possibly resulted from waterlogging restricting the supply of nitrogen to the shoot, which was observed in an earlier experiment by Trought and Drew (1980), while the reduction in the reproductive biomass was mainly due to an increasing shedding rate caused by the waterlogging (Longnecker and Erie, 1968). Moreover, limited carbohydrates resulted from the significant reduction in Pn also led to the reduced vegetative and reproductive biomass in our research (Kuai et al., 2014). In contrast, the biomass of the vegetative and reproductive organs of FB9–16 was significantly increased under waterlogged conditions compared with WL0 ; the highest biomass value was observed in WL6 (Table 1). This was consistent with previous research that showed an increase in the vegetative growth after the loss of fruiting forms from the earliest positions (Eaton, 1955; Kennedy et al., 1986). Waterlogging caused a reduction in plant growth measured in terms of the plant height, total number of fruiting branches, total number of fruiting positions, number of bolls and increased shedding rate (Fig. 3). The shedding of flowers and the loss of fruit

set due to waterlogging has also been observed in other studies that measured the entire cotton plant (Ahmad et al., 2002, 2003; Lakitan et al., 1992; Umaharan et al., 1997). The changes in the pattern of growth during the cotton’s ontogeny after waterlogging, determine an imbalance in the vegetative and reproductive growth. This might be the result of complex nutritional and hormonal influences that may be altered by fruit shedding. Materials that would have been partitioned to the damaged organs are partitioned to undamaged ones. Loss of reproductive organs, prolonging flower bud production and allowing some of the additional buds to set fruit increased the rate of late flowering and the number of fruiting positions (Sadras, 1995). Consequently, the lower shedding rate under waterlogged conditions in 2012 may be due to an increased number of bolls forming on FB13–16 as a result of compensation, which is limited to long-season environments (Tables2 and 4b). The lint yield of the bolls at position 1–2 on FB1-12 was higher in 2012 than in 2011, while the lint yield for the bolls at position 3 on FB5–16 was lower in 2012 than in 2011 (Table 4). This was likely due to the improved growth resulting from a sufficient number of day degrees and solar daily radiation in 2012 for the bolls at position 1–2 before waterlogging, whereas the bolls at position 3 or higher that had greater growth in 2011 after waterlogging could be attributed to improved compensation in response to waterlogging under lower temperatures or radiation (Fig. 1). Among the yield components, the CVs indicated that boll number was the most sensitive to waterlogging, whereas the boll biomass was the least sensitive to waterlogging. Consequently, waterlogging reduced the lint yield of FB1–8 primarily by reducing the boll number per plant (Table 2). Similar results have been reported in Bange et al. (2004). Bottom bolls occur earlier than middle and top bolls and have less competition (Hearn, 1976). Other studies also revealed that the first fruiting branches are usually less vigorous than subsequent ones (Mauney, 1968), and the first fruiting position on the middle fruiting branch normally contributes most to cotton yield (Constable, 1991; Heitholt, 1993). In the present study, waterlogging caused a significant reduction in the number of bolls at position 1–2, especially on FB1–8 , while an increase in the number of bolls at position 3 or higher was observed on FB9–16 (Table 4). The contribution of the bolls at position 3 or higher to the yield increased as the duration of the waterlogging increased. Previous studies indicated that the loss of bolls from the earliest (lower) positions resulted in more productive fruiting from later-developed positions that partially compensated for early losses (Dale, 1959; Kletter and Wallach, 1982; Ungar et al., 1987), as an increase in the sink strength was

68

Table 3 The statistical significance and interactions of the effects of waterlogging (WL), years (Y) and fruiting branches (FB) on lint yield and yield components. Position 1–2 bolls

Average across fruiting branches

Significance Year (Y) Fruiting branches (FB) Waterlogging (WL) Y × FB Y × WL FB × WL Y × FB × WL a b c d * **

Boll number

Boll biomass (g)

Lint percentage (%)

Lint yield (gFP− 1a )

Contribution rate to yield (%)

Boll number

Boll biomass (g)

Lint percentage (%)

Lint yield (gFP−1 )

Contribution rate to yield (%)

2011 2012

11.7bb 12.6a

4.3b 5.1a

38.6b 40.4a

21.2b 28.2a

60.3 71.5

7.7a 6.6b

4.5a 4.1b

39.0a 37.0b

13.5a 10.2b

39.7 28.5

WL0 c WL3 WL6 WL9 WL12

16.2a 14.4b 13.9b 11.7c 9.6d

5.1a 4.9ab 4.7abc 4.6bc 4.4c

42.7a 40.8ab 39.7bc 38.0 cd 36.5d

34.8a 28.7b 25.9b 17.8c 15.3d

68.3 67.1 67.5 66.4 60.1

8.1a 7.7ab 7.5ab 6.5b 6.5b

4.6a 4.4ab 4.2b 4.2b 4.2b

40.2a 39.5a 38.1b 36.8c 35.4d

15.1a 13.3ab 12.0bc 14.2c 9.6c

31.7 32.9 32.5 33.6 39.9

FB1–4 FB5–8 FB9–12 FB13–16

3.8a 3.5a 3.3ab 2.5b

4.6b 4.7ab 4.9a 4.7b

39.4a 39.0a 39.8a 39.8a

7.1a 6.6a 6.4a 4.5b

18.6 17.5 17.2 12.7

1.9ab 1.5b 2.2a 1.6ab

4.5a 4.5a 4.4a 3.9b

39.6a 37.3b 38.4b 36.6c

3.4ab 2.5ab 3.8a 2.3b

8.7 7.0 11.3 7.1

**

**

**

**

**

**

**

**

**

**

**

**

**

**

**

**

**

**

**

**

**

**

**

**

**

**

**

**

**

**

**

**

NSd

NS

NS

**

NS

**

*

**

**

**

**

**

**

**

**

**

NS

NS

**

**

**

**

**

**

FP, fruiting position. Values followed by different letters within the same column are significantly different at P = 0.05 probability level Each data represents the mean of three replications (eight plants in each replication). WL0 , WL3 , WL6 , WL9 and WL12 stand for waterlogging days of 0, 3, 6, 9 and 12 d, respectively. NS, not significant. Significant differences at 0.05 probability levels. Significant differences at 0.01 probability levels.

J. Kuai et al. / Europ. J. Agronomy 67 (2015) 61–74

Average across years Average across waterlogging treatments

Position 3 or higher bolls

J. Kuai et al. / Europ. J. Agronomy 67 (2015) 61–74

69

Table 4 (a) Effects of short-term waterlogging on lint yield, yield components and contribution rate to yield for bolls on different fruiting branches and positions in 2011. (b) Effects of short-term waterlogging on lint yield, yield components and contribution rate to yield for bolls on different fruiting branches and positions in 2012. (a) Fruiting branches

Days of waterlogging

Position 1–2 bolls

Position 3 or higher bolls

Boll number

Boll biomass (g)

Lint percentage (%)

Lint yield (g FB−1 d )

Contribution rate to yield (%)

Boll number

Boll biomass (g)

Lint percentage (%)

Lint yield (gFB−1 )

Contribution rate to yield (%)

FB1–4

WL0 a WL3 WL6 WL9 WL12 CV%c

4.8ab 3.5b 3.3bc 2.5 cd 2.0d 32.87

4.6a 4.3b 4.1c 3.9d 3.7e 8.69

43.2a 41.1b 40.2b 38.2c 36.7d 6.39

9.4a 6.2b 5.3b 3.7c 2.7c 47.34

19.6 16.0 13.9 12.8 12.0 20.52

2.8a 1.8b 1.5b 1.3b 1.0b 40.94

4.5a 4.5a 4.4b 4.4b 4.4b 1.64

43.0a 42.0ab 41.5b 40.6b 38.2c 4.42

5.4a 3.3b 2.7b 2.2b 1.7b 46.73

11.2 8.5 7.1 7.5 7.5 19.79

FB5–8

WL0 WL3 WL6 WL9 WL12 CV%

4.5a 3.3b 2.8bc 2.3 cd 1.8d 42.55

5.0a 4.7b 4.5c 4.2d 4.0e 10.42

42.8a 40.3b 37.3c 36.4c 35.0d 9.02

9.7a 6.2b 4.6bc 3.4 cd 2.4d 54.03

20.1 15.9 12.0 11.7 10.8 27.68

2.0a 1.8a 1.5a 1.3a 1.3a 23.57

4.4b 4.3b 4.4b 4.4ab 4.6a 2.02

41.5a 40.5b 39.4c 37.8d 36.4e 5.72

3.7a 3.1ab 2.6b 2.1b 2.1b 24

7.6 7.9 6.8 7.2 9.6 13.73

FB9–12

WL0 WL3 WL6 WL9 WL12 CV%

4.3a 3.5b 3.8ab 2.8c 2.0d 27.20

4.5c 4.8b 4.9a 4.5c 4.4d 3.64

39.7a 38.6b 38.3b 37.1c 34.3d 5.47

7.6a 6.5a 7.1a 4.6b 3.0c 33.16

15.8 16.7 18.5 15.9 13.3 11.57

2.3a 2.5a 3.3a 3.0a 2.5a 15.21

4.5b 4.8a 4.8a 4.8a 4.8a 2.80

41.1a 41.0a 38.9b 36.5c 35.2c 6.81

4.2b 4.9b 6.0a 5.2ab 4.3b 15.59

8.6 12.6 15.7 17.8 19.0 28.42

FB13–16

WL0 WL3 WL6 WL9 WL12 CV%

3.0ab 3.3ab 3.8a 3.3ab 2.8b 11.59

4.5a 4.5a 4.2ab 4.1bc 3.8c 6.87

41.9a 40.1ab 38.3b 37.8bc 35.63c 6.17

5.7b 5.8ab 6.4a 5.1b 3.8c 18.44

11.9 14.8 16.6 17.3 16.9 14.44

1.5a 1.8a 2.3a 1.8a 1.5a 17.50

4.1c 4.3b 4.5a 4.6a 4.7a 5.29

41.2a 38.8b 36.0c 35.6c 35.1c 7.03

2.5b 2.9ab 3.6a 2.8ab 2.5b 16.38

5.2 7.6 9.5 9.7 11.0 26.05

(b) Fruiting branches

Days of waterlogging

Position 1–2 bolls

Position 3 or higher bolls

Boll number

Boll biomass (g)

Lint percentage (%)

Lint yield (gFB−1 d )

Contribution rate to yield (%)

Boll number

Boll biomass (g)

Lint percentage (%)

Lint yield (gFB−1 )

Contribution rate to yield (%)

FB1–4

WL0 a WL3 WL6 WL9 WL12 CV%c

5.5ab 5.0ab 4.5bc 3.8cd 3.0d 22.84

5.5a 5.2ab 5.1ab 5.0bc 4.7c 5.72

45.6a 41.3b 39.0c 35.6d 33.3e 12.37

13.7a 10.8ab 9.0bc 6.7 cd 4.7d 39.02

26.0 23.6 23.7 21.6 17.2 14.89

3.8a 2.8b 1.5c 1.5c 1.0c 53.64

5.2a 4.8b 4.4c 4.1d 3.9e 12.08

39.8a 39.5a 38.7a 37.3b 35.4c 4.75

7.7a 5.3b 2.6c 2.3c 1.4c 67.17

14.6 11.6 6.8 7.4 5.1 42.73

FB5–8

WL0 WL3 WL6 WL9 WL12 CV%

5.5a 4.8b 4.0c 3.3d 2.8d 27.40

5.5a 5.3a 4.9b 4.8bc 4.6c 6.90

46.1a 42.2b 40.8c 36.0d 33.4e 12.71

13.8a 10.6b 8.0c 5.7d 4.2d 45.42

26.2 23.2 21.1 18.4 15.3 20.23

1.8a 1.8a 1.3ab 1.0b 1.0b 28.08

5.1a 4.6b 4.4b 4.5b 4.1c 8.56

39.2a 36.8b 34.9c 33.6d 33.1d 7.11

3.5a 3.0ab 1.9c 1.5c 1.4c 41.58

6.7 6.6 5.0 4.8 5.1 15.92

FB9–12

WL0 WL3 WL6 WL9 WL12 CV%

3.5ab 4.0a 3.5ab 3.3b 2.5c 16.35

5.3a 5.3a 5.3a 5.4a 5.1b 1.48

40.3b 41.3ab 42.7ab 42.6ab 43.4a 2.96

7.4a 8.8a 7.9a 7.4a 5.5b 16.30

14.1 19.3 20.8 23.9 20.1 18.13

1.0b 1.8b 1.5b 1.3b 3.0a 45.80

5.1a 4.2b 3.7c 3.6c 4.0b 14.58

39.2a 40.0a 39.6a 37.5b 35.1c 5.23

2.0b 2.9b 2.2b 1.7b 4.2a 38.37

3.8 6.3 5.8 5.5 15.3 62.04

FB13–16

WL0 WL3 WL6 WL9 WL12 CV%

1.3a 1.3a 2.0a 2.0a 2.3a 24.91

5.4a 5.4a 5.2ab 5.0b 4.6c 5.99

42.0a 41.3ab 40.6ab 40.0b 39.9b 2.17

3.0b 3.0b 4.2a 4.0a 4.3a 17.52

5.7 6.6 11.1 12.9 15.7 40.71

1.0b 1.0b 2.0a 1.7ab 1.7ab 30.52

4.2a 3.5b 3.1c 2.9d 2.9d 16.15

36.7ab 37.4a 35.4b 35.1b 34.9b 3.08

1.5ab 1.3b 2.2a 1.7ab 1.7ab 19.92

2.9 2.8 5.8 5.5 6.2 35.62

a

WL0 , WL3 , WL6 , WL9 and WL12 stand for waterlogging days of 0, 3, 6, 9 and 12 d, respectively. Values followed by different letters within the same column are significantly different at P = 0.05 probability level Each data represents the mean of three replications (eight plants in each replication). c CV%, cofficient of variation. d FB, fruiting branch. b

70

Table 5 Biomass accumulations for the boll wall, fiber and seed in a cotton boll on different fruiting branches and fruiting positions after short-term waterlogging in 2011 and 2012. Fruiting branches

Days of waterlogging

Position 1–2 bolls

Position 3 or higher bolls

2011

2012

2011

2012 Seed (g)

Fiber (g)

WL0 a WL3 WL6 WL9 WL12 CV%c

1.17ab 1.10b 1.07bc 1.04c 1.02c 5.40

2.61a 2.53ab 2.45bc 2.42bc 2.32c 4.50

1.98a 1.77b 1.64c 1.50d 1.34e 15.02

1.46a 1.28b 1.21c 1.14d 1.07e 12.04

2.97a 3.07a 3.12a 3.21a 3.11a 2.83

2.49a 2.14b 2.00b 1.77c 1.55d 17.95

1.10d 1.19c 1.28b 1.41a 1.33ab 9.39

2.58b 2.61b 2.58b 2.58b 2.72a 2.34

1.94a 1.89a 1.83b 1.77c 1.68d 5.66

1.23c 1.34b 1.37b 1.46a 1.22c 7.63

3.13a 2.92b 2.69c 2.59cd 2.49d 9.39

2.07a 1.90b 1.70c 1.54d 1.36e 16.38

FB6–7

WL0 WL3 WL6 WL9 WL12 CV%

1.26a 1.19ab 1.11bc 1.07c 1.05c 7.76

2.87a 2.81a 2.79a 2.66b 2.57b 4.45

2.15a 1.89b 1.66c 1.52d 1.38e 17.63

1.49a 1.30b 1.19c 1.12cd 1.05d 13.93

2.94a 3.07a 2.92a 3.09a 3.07a 2.73

2.51a 2.24b 2.01c 1.74d 1.54e 19.28

1.20c 1.20c 1.23c 1.26b 1.34a 4.61

2.57c 2.58c 2.66bc 2.76b 2.91a 5.28

1.82a 1.76b 1.72bc 1.67c 1.66c 3.78

1.27d 1.34bc 1.40ab 1.42a 1.29 cd 4.78

3.13a 2.90bc 2.87bc 2.95ab 2.73c 4.97

2.02a 1.69b 1.53c 1.49c 1.35d 15.76

FB10–11

WL0 WL3 WL6 WL9 WL12 CV%

1.23c 1.23bc 1.27abc 1.33ab 1.36a 4.63

2.79a 2.76a 2.74a 2.75a 2.75a 0.70

1.83a 1.73b 1.70bc 1.62c 1.44d 8.95

1.22a 1.34a 1.35a 1.32a 1.29a 3.96

3.14a 3.19a 3.04a 3.02a 2.98a 2.84

2.41a 2.35a 2.26b 2.25b 2.21b 3.61

1.14e 1.23d 1.29c 1.34b 1.40a 8.06

2.65d 2.83c 2.92bc 3.03ab 3.12a 6.23

1.84b 1.96a 1.86b 1.74c 1.70c 5.79

1.21d 1.26c 1.32b 1.45a 1.41a 7.57

3.08a 2.52b 2.23c 2.23c 2.58b 13.83

1.99a 1.68b 1.46c 1.33d 1.39 cd 16.98

FB14–15

WL0 WL3 WL6 WL9 WL12 CV%

1.28a 1.42a 1.41a 1.49a 1.46a 5.64

2.64ab 2.67ab 2.73a 2.56ab 2.47b 3.88

1.91a 1.78ab 1.69bc 1.55c 1.37d 12.57

1.42c 1.49ab 1.52ab 1.54a 1.47bc 3.25

3.12ab 3.14a 3.08ab 3.02b 2.78c 12.14

2.26a 2.21ab 2.11bc 2.02c 1.85d 7.90

1.27e 1.32d 1.50b 1.47c 1.57a 8.75

2.40d 2.63c 2.87b 2.95ab 3.03ab 9.31

1.69a 1.67a 1.61a 1.63a 1.64a 1.77

1.23d 1.31c 1.39b 1.43a 1.38b 5.83

2.63a 2.19b 1.99c 1.89d 1.88d 14.77

1.52a 1.31b 1.09c 1.02c 1.01c 18.64

a

c

Seed (g)

Fiber(g)

Boll wall (g)

Seed (g)

Fiber (g)

Boll wall (g)

Seed (g)

Fiber (g)

WL0 , WL3 , WL6 , WL9 and WL12 stands for waterlogging days of 0, 3, 6, 9 and 12 d, respectively. Values followed by different letters within the same column are significantly different at P = 0.05 probability level Each data represents the mean of three replications (eight plants in each replication). CV%, cofficient of variation.

J. Kuai et al. / Europ. J. Agronomy 67 (2015) 61–74

Boll wall (g)

FB2–3

b

Boll wall (g)

J. Kuai et al. / Europ. J. Agronomy 67 (2015) 61–74

71

Fig. 4. Biomass proportion of the boll wall, the seed and the fiber for cotton bolls on different fruiting branches and positions after short-term waterlogging (a) position 1–2 bolls in 2011 (b) position 3 or greater bolls in 2011 (c) position 1–2 bolls in 2012 (d) position 3 or higher bolls in 2012. Data (n = 3) are mean values ± SD, calculated from three replications (eight plants in each replication). WL0 , WL3 , WL6 , WL9 and WL12 stand for waterlogging days of 0, 3, 6, 9 and 12 d respectively.

observed when the bolls from position 1 of the same sympodial branch were absent (Peoples and Matthews, 1981; Sadras, 1995). Preliminary experiments showed that enhanced photosynthesis

may be an important component of compensatory growth in that the radiation-use efficiency of cotton increased (Sadras, 1995). In contrast, Milroy and Bange (2013) reported that a single transient

72

J. Kuai et al. / Europ. J. Agronomy 67 (2015) 61–74

Table 6 Correlation between “source (LSCB)–sink (boll)” after short-term waterlogging in 2011 and 2012. Correlation with

2011

Boll number Boll biomass

2012

Pna

Vegetative organs biomass

Reproductive organs biomass

Pn

Vegetative organs biomass

Reproductive organs biomass

0.827**, b 0.781**

0.889** 0.678**

0.527* 0.332

0.378 0.745**

0.932** 0.298

0.544* 0.368

Biomass

Boll wall Fiber Seed

0.362 0.805** 0.423

0.418 0.709** 0.580**

0.118 0.485* 0.097

0.520* 0.800** 0.081

−0.106 0.315 0.106

0.427 0.434 −0.277

Biomass proportion

Boll wall/boll Seed/boll Fiber/boll Fiber/seed

−0.081 −0.736** 0.570** 0.755**

−0.047 −0.584** 0.414 0.566**

−0.116 −0.506* 0.469* 0.540*

0.069 −0.591** 0.701** 0.747**

−0.318 −0.108 0.343 0.35

0.289 −0.498* 0.447* 0.517*

a b * **

Net photosynthetic rate (Pn) data were cited in Kuai et al. (2014). (n = 20, R0.05 = 0.444, R0.01 = 0.562). Significant differences at 0.05 probability levels. Significant differences at 0.01 probability levels.

waterlogging event had a long-term impact on the performance of a cotton crop: RUE did not recover from a single large waterlogging event early in crop development and remained low for the rest of the season. In the present study, the increased lint yield of the bolls at position 3 or higher on FB9 –12 and the bolls on FB13 –16 after waterlogging in 2011 was due to an increase in boll number and boll biomass resulted from the better-developed leaf subtending a given boll [i.e., higher Pn; data presented in Kuai et al. (2014)] in the waterlogged treatments compared with WL0 (Table 4a). Different cultivars used in these experiments might result in different responses of Pn to waterlogging. An increase in the boll number led to increased lint yield in 2012 at position 3 or higher on FB9 –12 and the bolls on FB13 –16 after waterlogging (Table 4b). Similar results were observed in mungbeans [Vigna radiata (L.) Wilczek], where a higher yield in tolerant cultivars was due to an increase in the number of pods, a higher rate of photosynthesis and increased availability of plant nitrogen under waterlogged conditions (Kumar et al., 2012). Competition for carbohydrates existed between the boll biomass and production of flower buds. Insufficient assimilates were applied for boll biomass accumulation after waterlogging, resulted in the reduced boll biomass for the position 3 bolls on FB13 –16 of the waterlogged plants in 2012, compared with the boll biomass of WL0 , while the boll number was able to be restored to the level of WL0 . As seen, actual cotton yield response will depend on the combined effects of fruit loss on yield potential (as affected by changes in acquisition and partitioning of carbon and nitrogen) and on the growing conditions (as affected by changes in the spatial pattern on the plant and seasonal timing of fruiting) (Sadras, 1995). Significant Y × WL interactions for the boll biomass and lint percentage and significant Y × WL × FB interactions for the boll number and boll biomass were observed at position 3 or higher (Table 3). These indicated that the effect of waterlogging on the bolls in these positions were more easily influenced than the bolls at position 1–2 when the weather conditions changed during waterlogging event and the boll development. Within the cotton boll, a reduction in the fiber biomass was observed for the bolls on all fruiting branches and positions. A higher boll wall biomass was observed in the bolls at position 3 or higher on FB2 –3 and FB6-7 and for all bolls on FB10 –11 and FB14 –15 under waterlogged conditions (Table 5). The boll wall serves as both “source” and “sink” during boll development. Increased assimilates stored in the boll wall would help afford nutrients for seed and fiber development after waterlogging. Correlation analyses revealed that the boll number had significant positive correlations with vegetative biomass and reproductive biomass, while Pn had significant positive correlations with fiber and boll biomass (Table 6). These results indicated that the loss of bolls after waterlogging occurred

mainly because of the impaired growth of cotton, while the inhibition of Pn resulted in a reduction in the fiber biomass and boll biomass, as previously reported by Kuai et al. (2014). Similar results were found in Liu’s study (Liu et al., 2013). Storage materials may account for up to 15% of the biomass in mature cotton fruits (Constable and Rawson, 1980). It has been proposed that the contribution of carbon stores to the yield increases with the ratio between sink-demand and current carbon assimilation (Sadras et al., 1993). This ratio could be high in the top of the plant if carbon assimilation is limited. The ratio may also increase, and a greater percentage of yield arises from stores when crop photosynthesis during fruit growth is reduced by stress induced by abiotic (e.g., water or nitrogen deficit) or biotic factors (e.g., leaf damage by pests and diseases) (Sadras, 1995). This might be attributed to the fact that Pn was correlated with just boll biomass and not total vegetative and reproductive biomass, while total reproductive biomass was not correlated with boll biomass. Waterlogging altered the biomass of the boll wall, seed, and fiber, as well as the biomass distribution within a cotton boll through a reduction in the fiber/boll and fiber/seed ratio and an increase in the proportion of the boll wall biomass and seed biomass (Fig. 4). In the present study, low Pn corresponding to high seed biomass per boll reduced carbohydrate availability for lint production (Table 6), as fiber development needed a high level of carbohydrates and energy. Furthermore, the Pn and the reproductive biomass were negatively correlated with the seed biomass proportion and were positively correlated with the fiber biomass proportion within a cotton boll (Table 6). These correlations suggested that higher Pn and better reproductive growth were favorable for the fiber biomass accumulation, while lower Pn accompanied by poor reproductive growth inhibited the transfer of assimilates from the seeds to the fiber. These results suggested that more assimilates were retained by the boll wall and seed after the waterlogging. The increased proportions of the boll wall biomass and seed biomass appear to be an ecological adaptation to the waterlogging. A modification in the distribution of the boll wall biomass will protect the seed and fiber from other stresses, while a modified seed biomass distribution will be beneficial to breeding (Bradow and Bauer, 2010; Lv et al., 2013). Despite the damaged plants having a similar number of fruit compared to the control plants or even more as compensation occurred on upper fruiting branches and higher fruiting positions, delayed fruit setting and boll opening on FB9–16 were reported by Kuai et al. (2014). In many cases, delayed fruit setting can be expected to shift fruit growth toward relatively unfavorable conditions, restricting the rate and amount of cellulose synthesis, which in turn results in underdeveloped fiber with poor physical and

J. Kuai et al. / Europ. J. Agronomy 67 (2015) 61–74

chemical properties (Roberts et al., 1992). Therefore, fruit shedding caused by waterlogging could affect not only the yield but also the earliness and quality of cotton. Can compensatory bolls on upper fruiting branches and higher boll positions maintain fiber quality? Further work is urgently needed to investigate the effect of waterlogging on fiber quality × fruiting branch and boll position. 5. Conclusions In the present study, the long-term growth of cotton was severely affected by short-term (3–12 d) exposure to waterlogging. Waterlogging of 3, 6, 9 and 12 d inhibited the bolls development on FB2 –3 and FB6 –7 . A marked compensation driven by altered fruiting dynamics as a result of previous waterlogging was observed for bolls on FB10 –11 and FB14 –15 in terms of greater biomass accumulation and higher lint yield, indicating this compensation was limited to long-season environments. The boll losses on FB1 –8 and the bolls at position 1–2 on FB9 –12 driven by waterlogging altered fruiting dynamics and promoted the formation of position 3 or higher bolls on FB9 –16 , resulting in an increase in the contribution of these bolls to the cotton yield. Waterlogging increased the proportion of the boll wall biomass and seed biomass within a cotton boll, while the proportion of the fiber biomass decreased. Insufficient assimilates were preferred compensation in boll number to compensation in boll biomass after waterlogging. These results highlight the importance of retaining the bolls at position 3 or higher on the upper fruiting branches to compensate for the loss in lint yield caused by waterlogging and provide an understanding of how waterlogging affects the growth, development and yield of cotton with respect to boll position. In addition, this work has important implications for establishing guidelines for decision making for disaster mitigation in situations where cotton has been subjected to short-term waterlogging events. Acknowledgment We are grateful for the financial support from the National Natural Science Foundation of China (31171487, 31371583). References Ahmad, R., Ikraam, M., Ullah, E., Mahmood, A., 2003. Influence of different fertilizer levels on the growth and productivity of three mungbean cultivars. Int. J. Agric. Biol. 5, 335–338. Ahsan, N.L.D., Lee, S.H., Kanga, K.Y., Bahka, J.D., Choi, M.S., Lee, I.J., Renaut, J.L.B., 2007. A comparative proteomic analysis of tomato leaves in response to waterlogging stress. Physiol. Plant 131, 555–570. Bange, M., Milroy, S., Thongbai, P., 2004. Growth and yield of cotton in response to waterlogging. Field Crops Res. 88, 129–142. Bondada, B.R., Oosterhuis, D.M., 2001. Canopy photosynthesis, specific leaf weight, and yield components of cotton under varying nitrogen supply. J. Plant Nutr. 24, 469–477. Boquet, D.J., Moser, E.B., 2003. Boll retention and boll size among intrasympodial fruiting sites in cotton. Crop Sci. 43, 195–201. Bradow, J.M., Bauer, P.J., 2010. Germination and seedling development. In: Stewart, J.M., Oosterhuis, D.M., Heitholt, J.J., Mauney, J.R. (Eds.), Physiology of Cotton. Springer, Netherlands, pp. 48–56. Brook, K.D., Hearn, A.B., Kelly, C.F., 1992a. Response of cotton to damage by insect pests in Australia: pest management trials. J. Econ. Entomol. 85, 1356–1367. Brook, K.D., Hearn, A.B., Kelly, C.F., 1992b. Response of cotton (Gossypium hirsutum L.) to damage by insect pests in Australia: manual simulation of damage. J. Econ. Entomol. 85, 1368–1377. Brook, K.D., Hearn, A.B., Kelly, C.F., 1992c. Response of cotton to damage by insect pests in Australia: compensation for early season fruit damage. J. Econ. Entomol. 85, 1378–1386. Christianson, J.A., Danny, J.L., Elizabeth, S.D., Wilson, I.W., 2010. Global gene expression responses to waterlogging in roots and leaves of cotton (Gossypium hirsutum L.). Plant Cell Physiol. 51, 21–37. Conaty, W.C., Tan, D.K.Y., Constable, G.A., Sutton, B.G., Field, D.J., Mamum, E.A., 2008. Genetic variation for waterlogging tolerance in cotton. J. Cotton Sci. 12, 53–61. Constable, G.A., Rawson, H.M., 1980. Photosynthesis, respiration and transpiration of cotton fruit. Photosynthetica 14, 557–563.

73

Constable, G.A., 1991. Mapping the production and survival of fruit on field-grown cotton. Agron. J. 83, 374–378. Dale, J.E., 1959. Some effects of continuous removal of floral buds on the growth of the cotton plant. Ann. Bot. 23, 636–649. Das, K.K., Panda, D., Sarkar, R.K., Reddy, J.N., Ismail, A.M., 2009. Submergence tolerance in relation to variable floodwater conditions in rice. Environ. Exp. Bot. 66, 425–434. Eaton, F.M., 1955. Physiology of the cotton plant. Annu. Rev. Plant Physiol. 6, 299–328. Gillham, F.E.M., Bell, T.M., Arin, T., Matthews, G.A., Le Rumeur, C., Hearn, A.B., 1995. Cotton production prospects for the next decade. World Bank Technical Paper, 287. The World Bank, Washington, DC, pp. 277. Grassini, P., Indaco, G.V., Pereira, M., Hall, A.J., Tra´ıpani, N., 2007. Responses to short-term waterlogging during grain filling in sunflower. Field Crops Res. 101, 352–363. Gu, L.L., Wang, X.S., Zhou, Z.G., Chen, D.H., Xu, L.H., 2010. Researches of high yield cotton cultivations in Jiangsu province. China Cotton 4, 14–16 (in Chinese). Guang, C., Wang, X.G., Liu, Y., Luo, W.B., 2012. Effect of water logging stress on cotton leaf area index and yield. Procedia Eng. 28, 202–209. Hearn, A.B., 1976. Crop physiology. In: Arnold, M.H. (Ed.), Agriculture Research for Development. Cambridge University Press, London. Heitholt, J.J., 1993. Cotton boll retention and its relationship to lint yield. Crop Sci. 33, 486–490. Hodgson, A.S., Chan, K.Y., 1982. The effect of short-term waterlogging during furrow irrigation of cotton in a cracking grey clay. J. Agric. Res. 33, 109–116. Hodgson, A.S., 1982. The effects of duration, timing and chemical amelioration of short-term waterlogging during furrow irrigation of cotton in a cracking grey clay. J. Agric. Res. 33, 1019–1028. Huang, B., Liu, X., Fry, J.D., 1998a. Shoot physiological responses of two bentgrass cultivars to high temperature and poor soil aeration. Crop Sci. 38, 1219–1224. Huang, B., Liu, X., Fry, J.D., 1998b. Effects of high temperature and poor soil aeration on growth and viability of creeping bentgrass. Crop Sci. 38, 1618–1622. Jackson, M., Colmer, T., 2005. Response and adaptation by plants to flooding stress. Ann. Bot. 96, 501–505. Jenkins, J.N., McCarty, J., Parrott, W., 1990. Effectiveness of fruiting sites in cotton. Yield Crop Sci. 30, 365–369. Kennedy, C.W., Smith, J.W.C., Jones, J.E., 1986. Effect of early season square removal on three leaf types of cotton. Crop Sci. 26, 139–145. Kletter, E., Wallach, D., 1982. Effects of fruiting form removal on cotton reproductive development. Field Crops Res. 5, 69–84. Kuai, J., Liu, Z., Wang, Y., Meng, Y., Chen, B., Zhao, W., Zhou, Z., Oosterhuis, D.M., 2014. Waterlogging during flowering and boll forming stages affects sucrose metabolism in the leaves subtending the cotton boll and its relationship with boll weight. Plant Sci. 223, 79–98. Kumar, P., Pal, M., Joshi, R., Sairam, R., 2012. Yield: growth and physiological responses of mung bean [Vigna radiata (L.) Wilczek] genotypes to waterlogging at vegetative stage. Physiol. Mol. Biol. Plants 19, 1–12. Lakitan, B., Wolfe, D.B., Zobel, R.W., 1992. Flooding affects snap bean yild and genotypic variation in leaf gas exchange and root growth response. J. Am. Soc. Hortic. Sci. 117, 711–716. Lv, F., Liu, J.R., Ma, Y.N., Chen, J., et al., 2013. Effect of shading on cotton yield and quality on different fruiting branches. Crop Sci. 53, 2670–2678. Mauney, J.R., 1968. Morphology of the cotton plant. In: Elliot, F.C., Hoover, M., PorterJr, W.K. (Eds.), Advances in Production and Utilization of Quality Cotton: Principles and Practices. The Iowa State University Press, Amew, IA, pp. 24–40. Meyer, W., Reicosky, D., Barrs, H., Smith, R., 1987. Physiological responses of cotton to a single waterlogging at high and low N-levels. Plant soil 102, 161–170. Milroy, S.P., Bange, M.P., 2013. Reduction in radiation use efficiency of cotton (Gossypium hirsutum L.) under repeated transient waterlogging in the field. Field Crops Res. 140, 51–58. Peoples, T.R., Matthews, M.A., 1981. Influence of boll removal on assimilate partitioning of cotton. Crop Sci. 21, 283–286. Pettigrew, W., 2004. Moisture deficit effects on cotton lint yield yield components and boll distribution. Agron. J. 96, 377–383. Roberts, E.M., Rao, N.R., Huang, J.Y., Trolinder, N.L., Haigler, C.H., 1992. Effects of temperature on fiber metabolism in cultured cotton ovules. Plant Physiol. 100, 979–986. Sadras, V.O., Connor, D.J., Whitfield, D.M., 1993. Yield, yield components and source–sink relationships in water-stressed sunflower. Field Crops Res. 31, 27–39. Sadras, V.O., 1995. Compensatory growth in cotton after loss of reproductive organs. Field Crops Res. 40, 1–18. Sairam, R.K., Dharmar, K., Chinnusamy, V., Meena, R.C., 2009. Waterlogging-induced increase in sugar mobilization, fermentation, and related gene expression in the roots of mung bean (Vigna radiata). J. Plant Physiol. 166, 602–616. Trought, M.C.T., Drew, M.C., 1980. The development of waterlogging damage in wheat seedlings (Triticum aestivum L.). Plant Soil 54, 77–94. Umaharan, P., Ariyanayagam, R.P., Haque, S.Q., 1997. Effect of short-term waterlogging applied at various growth phases on growth, development and yield in Vigna unguiculata. J. Agric. Sci. 128, 189–198.

74

J. Kuai et al. / Europ. J. Agronomy 67 (2015) 61–74

Ungar, E.D., Wallach, D., Kletter, E., 1987. Cotton response to bud and boll removal. Agron. J. 79, 491–497. Wang, K., Bian, S., Jiang, Y., 2009. Anaerobic metabolism in roots of Kentucky bluegrass in response to short-term waterlogging alone and in combination with high temperatures. Plant soil 314, 221–229.

Weatherley, P.E., 1950. Studies in water relations of cotton plants 1. The field measurement of water deficit in leaves. New Phytol. 49, 81–87. Yang, W., Zhu, J.Q., Liu, W.H., 2012. Impact of waterlogging coupling with high temperature during cotton in flowering and boll-bearing on its photosynthetic physiology and yield. Adv. J. Food Sci. Technol. 4, 344–347.