m/d shows month/date, d means days. \* Values followed by different letters within the same column in the same year are significantly different at probability levels (*p* < 0.05) according to the Least Significant Difference (LSD) test.

None of the specific cotton growth periods were affected by the K fertilizer amounts within the same year. However, the cotton growth period in 2016 was 18 d longer than that in 2017, due to 15 d longer in seedling and 11 d longer in boll setting, but 8 d shorter in squaring.

### *3.2. Cotton Plant Biomass Accumulation*

Cotton plant biomass (CPB) was significantly increased with increased K amounts in both years (Table 2). The same trends were observed in root and stem biomass. Compared with K1, K<sup>2</sup> increased root and stem 11.55% and 2.11% in 2016, respectively. However, the root biomass in K<sup>2</sup> was lower than that in K<sup>1</sup> in 2017 with no significant difference and stem biomass in K<sup>2</sup> was 13.87% higher than that in K1. Cotton plants in K<sup>3</sup> produced 24.71% (2016) and 0.65% (2017) more root biomass and 27.36% (2016) and 26.40% (2017) more stem biomass compared with K1. Leaves and reproductive parts accumulated higher in K<sup>2</sup> and K3, and significantly lower in K1. The ratios of RPB to CPB had no significant difference among the three K amounts in 2016, but that is significantly higher in K<sup>2</sup> and K<sup>3</sup> than K1. There were no significant differences between K<sup>2</sup> and K<sup>3</sup> for Leaves and RPB and the ratios of RPB to CPB. Furthermore, the ratios of RPB/CPB in K<sup>2</sup> were higher than other treatments.


**Table 2.** Cotton and each part biomass accumulation influenced by K fertilizer amounts. *Agronomy* **2020**, *10*, x FOR PEER REVIEW 5 of 10

\* Values followed by different letters within the same column in the same year are significantly different at probability level (*p* < 0.05) according to Least Significant Difference (LSD) test. Average 1149.6 2692.7 472.6 3723.6 8038.5 45.60 \* Values followed by different letters within the same column in the same year are significantly

different at probability level (*p* < 0.05) according to Least Significant Difference (LSD) test.

The growth curves of CPB, VPB, and RPB increased along with the cotton growth stage following a sigmoid curve with different slopes from K fertilizer amounts (Figure 1). The growth curves of CPB, VPB, and RPB increased along with the cotton growth stage following a sigmoid curve with different slopes from K fertilizer amounts (Figure 1).

**Figure 1.** Cotton plant, vegetative parts, reproductive parts, and reproductive related parts biomass of field-grown cotton influenced by K fertilizer amounts at different growth stages in 2016. SQ, FB, **Figure 1.** Cotton plant, vegetative parts, reproductive parts, and reproductive related parts biomass of field-grown cotton influenced by K fertilizer amounts at different growth stages in 2016. SQ, FB, PB, BO, and PS indicate squaring (51 days after emergence (DAE)), first bloom (66 DAE), peak bloom (81 DAE), boll opening (128 DAE), and plant senescence (168 DAE) stage, respectively. Error bar plus shows SEMs.

The growth curve slopes of CPB, VPB, and RPB were gradually increased until the boll opening stage and then decreased. Compared with RPB, the slopes of VPB curves were higher before peak bloom stage, and 41.04%–44.47% and 35.05%–40.90% VPB was produced in 2016 and 2017, respectively. However, RPB grew faster after peak bloom stage, and 81.75%–82.40% and 82.09%–84.82% RPB were produced in 2016 and 2017, respectively. The growth curves of CPB, VPB, and RPB in different K amounts gradually diverged from peak bloom. At the plant senescence stage, K3 plants produced 4.84% CPB, 5.90% VPB, and 3.76% RPB more than K2 plants and 7.31% CPB, 8.30% VPB, and 6.30% RRB more than K1 plants, respectively, in 2016. K3 plants produced 9.10% CPB, 9.15% VPB, and 9.04% RPB more than K2 plants and 18.64% CPB, 18.88% VPB, and 18.36% RRB more than K1 plants, respectively, in 2017.

### *3.3. Simulation of Biomass Accumulation*

The biomass of cotton plants and each plant part accumulated following the logistic regression Equation (1) (*p* < 0.01) and showed different accumulation characteristics during FAP in both years (Table 3). Eigenvalues of cotton plant biomass accumulation were calculated by Equations (2)–(6) using the coefficient of Equation (1).


**Table 3.** Regression equation and Eigenvalues of cotton plant biomass accumulation of field-grown cotton influenced by K fertilizer amount in 2016 and 2017.

Where *t*<sup>1</sup> and *t*<sup>2</sup> (DAE) mean the initiation and termination, respectively, of the fast accumulation period (FAP); ∆*t* (d) means the duration of FAP; *V<sup>T</sup>* and *V<sup>M</sup>* (g d−<sup>1</sup> ) mean the average, and the highest biomass accumulation rate, respectively, during FAP.

The *W<sup>M</sup>* values of the logistic regression equation in CPB, VPB, and RPB were higher with increased K amounts (Table 2). The average and the highest biomass accumulation rates of CPB and each part biomass during FAP showed higher values in higher K amounts in both years.

Compared with RPB, VPB initiated FAP 25 d earlier in 2016 and 18 d in 2017 and terminated FAP 10 d earlier in 2016 and 24 d in 2017 with 15.5 d longer duration in 2016 and 6 d shorter in 2017. CPB initiated FAP 13 d (2016) and 12 d (2017) earlier than RPB and terminated 4 d (2016) and 13 d (2017) earlier with 9 d longer duration in 2016 and 1 d shorter in 2017. Compared with the average accumulation rates of FAP in RPB, the rates in CPB was 34.3 kg ha−<sup>1</sup> d −1 (2016) and 71.2 kg ha−<sup>1</sup> d −1 (2017) faster, and the rates in VPB was 14.6 kg ha−<sup>1</sup> d −1 slower in 2016 and 23.8 kg ha−<sup>1</sup> d −1 faster in 2017.

CPB initiated FAP in flowering and boll setting period (74 DAE in 2016 and 68 DAE in 2017) and terminated at 120 DAE (2016) and 104 DAE (2017) with the duration of 46 d and 36 d averaged across three K fertilizer amount in 2016 and 2017, respectively. The FAP initiations in K<sup>2</sup> and K<sup>3</sup> were similar but later than that in K<sup>1</sup> in both years. The FAP termination was earlier with increased K amounts in 2016, but later in 2017. The FAP duration was decreased with increased K amounts in 2016, but no similar result was observed in 2017.

VPB initiated FAP at 62 DAE in both years and terminated in flowering and boll setting period at 114 DAE (2016) and 93 DAE (2017), respectively. Many differences existed in FAP durations and biomass accumulation rates of VPB between both years which reflected the VPB were accumulated more slowly in 2016 than in 2017. With the increase in K amount, the duration of FAP become shorter and the accumulation rates were higher in 2016, but the shortest duration and highest accumulation of FAP in VPB were observed in K<sup>2</sup> in 2017.

RPB initiated and terminated FAP in the flowering and boll setting period with 37 d FAP duration in both years. The accumulation rates were higher in 2017 than in 2016. With increased K amounts, FAP durations decreased and the accumulation rates increased.

### *3.4. Fiber Quality*

K amount significantly affected the fiber length, micronaire, and fiber strength. With increased K amount, the fiber length and the fiber strength increased significantly but there was no significant difference between that in K<sup>2</sup> and K3. The micronaire values in different K amounts were no significant difference in 2016, but significantly lower in K<sup>1</sup> in 2017 with no significant difference between that in K<sup>2</sup> and K3.

### **4. Discussion**

K fertilizer is one of the main cotton fertilizers and has great correlations with cotton growth and the economic benefits of cotton production.

Many studies also reported that K deficiency could accelerate the growth process of cotton and result in premature senescence [28–31]. However, another study showed that K deficiency elicited similar effects on cotton earliness with late sowing which delayed flowering and boll development [32]. In the present study, the growth period was not affected by K amount, although apparent differences existed between the two growing seasons (Table 1). This might be due to the closeness of the K amount range in this study which was not sufficient to bring significant differences in growth period among different K amounts in the same year. Furthermore, the K amounts of three treatments were within the appropriate range for cotton growth under medium fertility, ensuring no K-deficiency in this study. The big differences between the two growing seasons were possibly due to a large amount of precipitation during the early cotton growth period but draught occurred in the flowering and boll setting periods in the 2016 growing season.

The biological yield was the basis of economic yield. Biomass accumulation could be explained in the context of plant photosynthesis, photo-assimilate translocation from vegetative to reproductive parts. K fertilizer affected the photo-assimilate export from leaves to sink parts and regulated the sugar signaling in reproductive parts [24]. Potassium deficiency led to a reduction of main stem length, nodes and bolls, and also leaf photosynthesis and stomatal conductance [18,23,33,34], resulting in less carbohydrate production, a small sink and an in-balanced source-sink ratio. A previous study also revealed that a linear effect between K amounts and the growth efficiency of the reproductive part [35]. In this study, the RPB and the ratios of RPB to CPB were significantly higher in K<sup>2</sup> and K<sup>3</sup> with no significant difference between the two treatments (Table 2). This indicated that K<sup>2</sup> can benefit the carbohydrate production transportation from vegetative parts to reproductive parts and get the approximate RPB with K3. Higher carbohydrate production in the reproductive parts can result in higher yield [36]. In the present study, the biomass of each cotton part was increased along with the increase in K amount (Table 2 and Figure 1). Similar studies revealed that increasing the K amount can increase the biomass of total plant and cotton bolls [14,18,37]. Furthermore, in this study, the FAP of CPB and RPB were initiated in the flowering and boll setting period and the FAP durations of RPB in both years were the same but the FAP accumulation rates were higher in 2017 with higher RPB (Table 3). With the increased K amount, the FAP accumulation rates were higher with higher biomass accumulation, but the duration of FAP shortened (Table 3). Similar results were also observed in Khan et al. [6] and Tung et al. [38]. This indicated that higher K amounts increased the cotton plant biomass mainly by higher accumulation rate during flowering and boll setting period.

In this study, fiber length, fiber strength, and micronaire were significantly affected by K amounts and better fiber quality traits were observed in K<sup>2</sup> and K<sup>3</sup> with no significant difference between K<sup>2</sup> and K<sup>3</sup> (Table 4). That indicated the fiber quality in K<sup>2</sup> was as well as that in K<sup>3</sup> and significantly better than that in K1.


**Table 4.** Cotton fiber quality influenced by K fertilizer amounts.

\* Values followed by different letters within the same column in the same year are significantly different at probability level (*p* < 0.05) according to Least Significant Difference (LSD) test.

### **5. Conclusions**

K amounts ranging from 168–252 kg K2O ha−<sup>1</sup> have not altered the cotton growth period. Higher K increased cotton biomass due to a higher accumulation rate during FAP. Nevertheless, K<sup>2</sup> had similar fiber qualities, biomass accumulation, and partitioning as K3.

The results suggest that an equal K amount to lowed N of 210 kg ha−<sup>1</sup> should be the optimal strategy under this new planting model in Yangtze River Valley, China, and similar regions.

**Author Contributions:** Conceptualization, X.M. and G.Y.; methodology, X.M.; data curation, X.M., S.A., A.H., A.L., J.L., Z.Z., D.L., A.N.S. and G.Y.; writing—original draft preparation, X.M.; writing—review and editing, X.M. and G.Y.; funding acquisition, G.Y. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the National Natural Science Foundation of China (31271665, 31771708).

**Conflicts of Interest:** The authors declare no conflict of interest.

### **References**

1. Reddy, N.; Yang, Y. Properties and potential applications of natural cellulose fibers from the bark of cotton stalks. *Bioresour. Technol.* **2009**, *100*, 3563–3569. [CrossRef]


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*
