https://orcid.org/0000-0001-5591-8212
Figures
Abstract
This study aimed to characterize the diversity and variation of fruit phenotypic traits in natural provenances of Cinnamomum camphora to facilitate the selection of superior germplasm. According to a provenance-based sampling design, 10 fruit phenotypes were investigated across 170 individuals from 11 provenances from Jiangxi province, China. The results showed that the average phenotypic differentiation coefficients among and within provenances were 73.82% and 26.18%, respectively, indicating that variation among provenances was the main source of phenotypic diversity in C. camphora in Jiangxi province, China. The mean Shannon-Wiener index (H) for the fruit phenotypic traits of C. camphora was 2.62, with the high diversity observed in fruit horizontal diameter/fruit vertical diameter (FHD/FVD, 2.848), fruit thousand-grain weight/fruit volume (FTW/FV, 2.835) and the fruit side diameter/fruit horizontal diameter (FSD/FHD, 2.817). The coefficients of variation of fruit phenotypic traits among different provenances ranged from 2.64% to 25.45%, with the largest value for peel thickness (PT). Furthermore, significant or highly significant correlations existed among fruit phenotypic traits. There was a significant correlation among fruit phenotypic traits of C. camphora and latitude, longitude, climate factors in different provenances of Jiangxi province. Based on the results of principal component analysis and cluster analysis, the cumulative contribution rate reached 78.78%, and the fruit side diameter (FSD), fruit side diameter/fruit vertical diameter (FSD/FVD), fruit side diameter/fruit horizontal diameter (FSD/FHD) could comprehensively reflect the information and ranking of the 10 traits. The fruit phenotypic traits from subgroup B2 were bigger. These selected individuals can be used as basic materials for variety improvement and conservation. The high phenotypic diversity indicates a rich genetic base, suggesting that future studies should employ molecular markers to assess genetic diversity and identify associated genes.
Citation: Wan S, Zhao J, Ma J, Zhang J, Liu Q, Xiao C, et al. (2026) Variation of fruit phenotypic traits of Cinnamomum camphora from 11 provenances in China. PLoS One 21(5): e0319877. https://doi.org/10.1371/journal.pone.0319877
Editor: Zhihong (Arry) Yao, Southwest Jiaotong University, CHINA
Received: February 9, 2025; Accepted: April 9, 2026; Published: May 20, 2026
Copyright: © 2026 Wan et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: This work was supported by Jiangxi Provincial Department of Science and Technology (Grant No. 20203ABC28W016, awarded to Z.J.) and Education Department of Jiangxi Province (Grant No. GJJ2201515, awarded to Q.L.).
Competing interests: The authors have declared that no competing interests exist.
Introduction
Cinnamomum camphora (Lauraceae) is a subtropical evergreen tree species of significant economic and ecological value, integrating timber production, landscaping, medicinal applications, and biochemical functions into its utility [1–4]. Jiangxi renowned as “the hometown of camphor wood” in China, represents a modern distribution center for this species and conserves abundant natural resources resulting from millennia of cultivation history [5]. Despite this rich germplasm foundation, strategic conservation and genetic improvement programs in Jiangxi remain markedly underdeveloped.
The comprehensive collection and preservation of germplasm resources demand substantial investment of human, financial, and material resources [6–7], underscoring the critical need for efficient screening strategies to optimize conservation and utilization efficiency [8]. Plant phenotypic traits, which arise from the interplay of genetic and environmental factors, serve as primary indicators for assessing diversity and preliminary selection [9–11]. Considerable studies on diverse tree species, including Idesia polycarpa Maxim. (Salicaceae) [12], Elaeagnus mollis Diels. (Elaeagnaceae) [13], Canarium album L. (Burseraceae) [14] and Melia azedarach (Meliaceae) [15] have demonstrated substantial provenance-based variation in fruit traits, providing a critical basis for elite germplasm selection. These phenotypic differences were frequently correlated with geographic and climatic factors, such as longitude, latitude and temperature [16–17].
Despite the ecological and economic importance of C. camphora in Jiangxi, a significant knowledge gap exists regarding the patterns of fruit phenotypic diversity and its relationship with geographical gradients in this region [18]. Furthermore, it remains unknown which specific fruit traits are most influential for discriminating among provenances and selecting superior germplasm, and is geographically limited to regions such as Sichuan and Zhejiang provinces in China. The patterns of fruit phenotypic variation in Jiangxi which is a region critical to the species distribution remain completely unexplored.
While assessing genetic diversity using molecular markers provides the most direct measure of heritable variation [19], these approaches can be cost-prohibitive and technically demanding for initial screening of large germplasm collections. Therefore, the evaluation of phenotypic diversity remains a fundamental and cost effective first step for identifying promising germplasm and capturing potential genetic variation, upon which subsequent molecular analyses can be focused.
Therefore, to address this knowledge gap, this study investigated the fruit phenotypic diversity of 170 C. camphora individuals sampled from 11 natural provenances across Jiangxi province. We aimed to quantify the diversity and differentiation of fruit traits, elucidate their relationships with geographical and climatic factors, and identify superior germplasm resources. Our findings would provide a scientific basis for breeding selection and the conservation of C. camphora germplasm resources.
Materials and methods
Study area and plant materials
A total of 170 healthy and vigorous plants showing no visible signs of disease or pest infestation and with intact canopy architecture were selected. The fruit collection was conducted in December 2018 across 45 counties within 11 cities of Jiangxi province, China, with official permission from Jiangxi provincial Department of Forestry. The geographic location information and meteorological data from the monitoring information of meteorological stations of 11 provenances were shown in Table 1. Jiangxi province has a subtropical monsoon climate. The annual average sunshine duration is 1447.98–1805.13 h, and the mean altitude is 17.16–388 m. The annual mean temperature is 17.65–18.98 ℃. The annual mean precipitation is 1461.50–1898.14 mm, and the soil is dominated by red soil. Several normal-growing C. camphora fruits with a diameter at breast height greater than 60 cm were collected using a systematic sampling method at 2 km intervals to ensure spatial independence of samples and effectively capture intra provenance variation [20]. The organs of the plant were presented in Fig 1.
(A) Overall plant morphology; (B) The front (adaxial) side of a leaf; (C) The back (abaxial) side of a leaf; (D) Fruit.
Data collection
In this study, approximately 2 kg of mature fruits were collected from the four cardinal directions (east, south, west and north) of each tree. From this bulk sample, one hundred fruits per tree (n = 100) were randomly selected for the measurement of the fruit thousand-grain weight (FTW, g) and fruit volume (FV, cm³). The fruit thousand-grain weight (FTW, g) was measured using an electronic balance with an accuracy of 0.1 g, while the fruit volume (FV, cm³) was determined by the water displacement method. First, we added an appropriate amount of water to a graduated cylinder and recorded the initial volume (V1, mL), then slowly and completely immersed the fruit in the water, ensuring no air bubbles adhere to it. After the liquid stabilizes, the final volume (V2, mL) was recorded. The fruit volume was calculated as the difference between the two readings (V = V2-V1, mL). If the fruit floated, it was submerged fully before taking the reading.
Additionally, thirty fruits per tree (n = 30) were randomly chosen to measure the following morphological traits of fruit vertical diameter (FVD, mm), fruit horizontal diameter (FHD, mm), fruit side diameter (FSD, mm) and peel thickness (PT, mm), using an electronic vernier caliper with an accuracy of 0.01 mm. The fruit vertical diameter (FVD, mm) was defined as the maximum distance from the top to the base of the longitudinal section, the fruit horizontal diameter (FHD, mm) as the maximum distance from the ventral suture line to the opposite side in the transverse section, and the fruit side diameter (FSD, mm) as the maximum distance across both sides of the ventral suture line [21]. Three replicates were performed for each trait.
Finally, the following of fruit side diameter/fruit horizontal diameter (FSD/FHD), fruit side diameter/fruit vertical diameter (FSD/FVD), fruit horizontal diameter/fruit vertical diameter (FHD/FVD) and fruit thousand-grain weight/fruit volume (FTW/FV, g·cm-3) were calculated based on the above measurements.
Data and statistics
Microsoft Excel 2023 was used to examine the variation of fruit phenotypic traits, including mean value, standard error and coefficient of variation. SPSS27.0 software, minitab software and PAST software were used to analyze the diversity of each phenotype, including multiple comparisons, analyzed using one-way analysis of variance (ANOVA) and regression analysis. Means comparison method were conducted both among and within provenances, and one way ANOVA was performed to assess the differences among provenances. When ANOVA revealed significant effects, means were separated by Duncan’s new multiple range test at a significance level of p < 0.05. And Origin 2024 software was further used for principal component analysis and cluster analysis of fruit phenotypes in 11 provenances.
The formula CV= ×100% was used to calculated the coefficient of variation, where CV was the coefficient of variation,
was the average value of phenotypic trait, s was the trait standard deviation [22].
Shannon-Wiener index was a quantitative representation of the variation of trait diversity. The calculation formula was H = . In the formula, H was the diversity index, and pi was the effective percentage of the distribution frequency within the material at level i of a trait [23].
The formula VST= [δ2t/S/(δ2t/S + δ2S)] × 100% was used to calculated the variation of phenotypic differentiation coefficient among provenances. VST was the phenotypic differentiation coefficient, representing the percentage of among provenances of total genetic variation. δ2t/S was variance component among provenances, and δ2S was variance component within provenances. And the sum of variation of phenotypic differentiation coefficient among and within provenances was one [24].
Results
Variation of fruit phenotypic traits of C. camphora in different provenances
The variation of fruit phenotypic traits was shown in Table 2. The differences among provenances of FTW, FV, FHD, PT, FSD/FHD and FSD/FVD reached a extremely significant level (P < 0.01). There were significant differences of FTW, FSD/FHD, FSD/FVD within provenances (P < 0.05). The mean variance component among provenances (23.87%) was greater than that within provenances (10.44%). The phenotypic differentiation coefficient among provenances ranged from 37.04% to 94.45% across all traits, with all values exceeding 30%. The highest value was observed for FVD (94.45%), while the coefficients for FV and FSD were all above 80%. In contrast, the phenotypic differentiation coefficient within provenance ranged from 5.55% to 62.96%. It was indicated that variation among provenances was the main source in C. camphora in Jiangxi province, China.
Significant differences were observed in fruit phenotypes of C. camphora among 11 provenances in Jiangxi, China (Table 3). The traits exhibited considerable variation, with ranges of 454.9–608.74 g for FTW and 438.19–586.43 cm³ for FV. The JJ provenance recorded the lowest values for both these traits.
Similarly, the ranges for FSD, FHD, FVD, and PT were 8.91–9.77 mm, 8.09–9.08 mm, 8.45–9.36 mm and 0.89–1.28 mm, respectively. The YT provenance showed the lowest values for FSD, FHD, and FVD. In contrast, the JA provenance consistently exhibited the largest values for FTW, FV, FSD, FHD, FVD and PT among all provenances.
The ranges of FSD/FHD, FSD/FVD, FHD/FVD and FTW/FV were 1.05 to 1.11, 1.02 to 1.09, 0.94 to 0.99 and 1.03 to 1.06 g·cm-3, respectively. The largest of FSD/FHD was found in the NC and YT provenances, while the smallest was in XY. The YC provenance had the largest FSD/FVD, and the JJ provenance had the smallest. The XY provenance showed the largest FHD/FVD, and the NC provenance had the smallest. No significant differences were detected among provenances for the FTW/FV.
The Shannon-Wiener index for ten traits ranged from 2.40 to 2.85. The highest diversity was observed in FHD/FVD, while the lowest was in FTW and FV. These results collectively indicated that C. camphora fruits from different provenances possessed rich genetic diversity.
Variation coefficient of phenotypic traits in different provenances was shown in Table 4. Among 10 fruit phenotypic traits in different provenances, the largest coefficient of variation was PT (25.45%) followed by FV as the second highest (17.83%), and the smallest coefficient of variation was FSD/FHD (2.64%). According to the average coefficient variation of phenotypic traits in 11 provenances, 2 provenances were more than 10%, among which PX provenance was the largest with the value for 11.14%, followed by NC provenance (10.68%), and YT provenance was the smallest (5.15%). The mean value of the coefficient of variation among provenances (9.65%) was greater than the variation within provenances (8.52%) for fruit phenotypic traits of C. camphora.
Correlation analysis on fruit phenotypic traits of C. camphora in Jiangxi province
There were certain correlations among fruit phenotypic traits of C. camphora from different provenances in Jiangxi province, China (Fig 2). There were significant positive correlations between FTW and four traits of FV, FSD, FHD and FVD (P < 0.05). FV showed a highly significant correlation with FSD, FHD and FVD. FSD showed significantly positively correlated with FHD and FV. FHD was significantly positively correlated with FV, while negatively correlated with FSD/FHD and FTW/FV. FVD showed significantly positively correlated with PT, while there were significant negative correlations with FSD/FVD, FHD/FVD and FTW/FV. In summary, most of the relevant traits of C. camphora fruits were closely related, and when one trait was changed, it might cause changes in other indexes.
FTW: fruit thousand-grain weight; FV: fruit volume; FSD: fruit side diameter; FHD: fruit horizontal diameter; FVD: fruit vertical diameter; PT: peel thickness; FSD/FHD: ratio of fruit side diameter to horizontal diameter; FSD/FVD: ratio of fruit side diameter to vertical diameter; FHD/FVD: ratio of fruit horizontal diameter to vertical diameter; FTW/FV: ratio of fruit thousand-grain weight to volume. Values are Pearson correlation coefficients ranging from 0 to 1.The color gradient bar on the right (from blue to red) indicates the strength of correlation, with darker red representing stronger positive correlations. *p < 0.05.
Response of fruit phenotypic traits to latitude, longitude and climatic factors from different provenances
Regression analysis revealed significant geographic patterns in the phenotypic traits of C. camphora different provenances (Fig 3 and Fig 4). Both FTW and FV showed a highly significant negative correlation with latitude (P < 0.01). Similarly, FSD and FHD were significantly negatively correlated with latitude (P < 0.05). Regarding longitude, FSD, FHD, FSD/FVD and FHD/FVD all exhibited a significant negative correlation (P < 0.05), indicating pronounced geographic variation in the phenotypic traits of C. camphora in Jiangxi province.
Each panel shows the relationship between one fruit phenotypic trait (y-axis) and latitude (x-axis, °N). Fitted linear regression lines are presented with corresponding regression equations, correlation coefficients (R2), and significance levels (**P < 0.01, *P < 0.05).
Multiple linear regression lines are shown, with the corresponding regression equations labeled within the figure. The X-axis represents longitude (°); the Y-axis represents fruit phenotypic traits (see labels within the figure or the main text for specific trait names). Each line was fitted based on the original data points; in the equations, x denotes longitude, and y denotes the corresponding fruit phenotypic trait value. Raw data points are not shown in the figure; please refer to the main text or supplementary materials for details.
The correlation between fruit phenotypic traits and geo-climatic factors was detailed in Table 5. FTW and FV were positively correlated with the mean temperature in January, mean altitude (P < 0.01) and annual mean temperature (P < 0.05), but negatively correlated with the annual average sunshine duration (P < 0.01) and mean temperature in July (P < 0.05). FSD showed a significant negative correlation with annual sunshine duration (P < 0.01). FHD was negatively correlated with the mean temperature in July and annual sunshine duration (P < 0.05). Furthermore, FVD was positively correlated with mean altitude (P < 0.05), while PT was positively correlated with both mean altitude and annual mean precipitation (P < 0.05). Among the ratio traits, FSD/FHD was positively correlated with annual precipitation (P < 0.05), whereas FSD/FVD was negatively correlated with annual sunshine duration (P < 0.05).
Principal component analysis of fruit phenotypic traits of different provenances
Principal component analysis was conducted in fruit phenotypic traits from different provenances in Jiangxi province of China, in which three principal components with eigenvalues greater than 1.00 were yielded with a cumulative contribution rate of 78.78%. It was indicated that the three principal components represented most of the information on phenotypic traits of C. camphora (Fig 5). Among them, the 1st principal component had the highest contribution rate of 42.6%, with an eigenvalue for 4.26. And eigenvector absolute value of FSD (0.96), FHD (0.91), FV (0.91) and FTW (0.89) were larger, which indicated that the 1st principal component was mainly associated with the traits related to fruit size. The contribution rate of the 2nd principal component was 22.5% with the eigenvalue for 2.255, and the absolute values of eigenvectors in 3 phenotypic traits of FSD/FVD (0.92), FHD/FVD (0.78) and FVD (−0.67) were larger, indicating that the 2nd principal component was related to fruit shape. The contribution rate of the 3rd principal component was 13.6%, and the eigenvalue was 1.361, in which FSD/FHD (0.76) and FHD/FVD (−0.54) had the largest absolute value, indicating that the 3rd principal component was also a trait mainly reflecting fruit shape. In summary, FSD, FSD/FVD and FSD/FHD could comprehensively reflect the information of 10 traits.
The figure shows a 3D PCA plot. Points represent samples from different provenances; vectors represent individual fruit phenotypic traits. The X-axis (PC1), Y-axis (PC2), and Z-axis (PC3) are the first, second, and third principal components, respectively. For definitions of FTW, FVD, FSD, FHD, and their ratios, please refer to Fig 2.
According to fruit phenotypic trait load coefficient of C. camphora on the three principal components and the contribution rate of the three principal components, the comprehensive score of each C. camphora tree was calculated, so as to obtain the top 20 plants of C. camphora in the comprehensive score, with the sample number for 79, 67, 112, 166, 72, 137, 98, 165, 142, 120, 66, 159, 80, 97, 87, 11, 59, 135, 109 and 157. And the corresponding scores were 3.08, 2.12, 1.82, 1.81, 1.77, 1.76, 1.60, 1.59, 1.56, 1.53, 1.52, 1.483, 1.47, 1.29, 1.25, 1.24, 1.19, 1.18, 1.15 and 1.12, respectively. Among them, there were 6 plants from YC provenance, 5 plants from GZ provenance, 4 plants from JA provenance, 2 plants from FZ provenance, 2 plants from PX provenance and 1 plant from JDZ provenance (Table 7).
Cluster analysis based on fruit phenotypic traits of C. camphora from different provenances
Hierarchical cluster analysis was performed based on the Pearson correlation coefficient using the between-groups linkage method to analyze the 10 fruit phenotypic traits of C. camphora from 11 provenances in Jiangxi province, China (Fig 6). The mean values of fruit phenotypic traits for each resultant group were calculated (Table 8).
The figure shows a dendrogram based on euclidean distance. The Y-axis represents provenance abbreviations; full names are provided in Table 1. The X-axis represents the euclidean coefficient, indicating the dissimilarity of fruit traits among provenances. Provenances with smaller distances exhibit more similar fruit traits.
The dendrogram showed that C. camphora from different provenances could be divided into 2 major groups at a distance coefficient of 1.2. These were further subdivided at a distance coefficient of 0.9, where group A emerged as a distinct cluster, and group B was partitioned into subgroups B1 and B2.
Group A included four provenances (NC, YT, JJ, FZ). Subgroup B1 comprised three provenances (SR, JDZ, XY) and subgroup B2 contained the remaining four (JA, YC, GZ, PX). The fruit phenotypic traits of group B provenances were significantly greater than those of group A, as evidenced by thicker peel, larger fruit size and heavier fruit weight. Furthermore, within group B, FTW, FV and FSD in the B2 provenances were significantly higher than in B1, while PT in B2 was significantly lower than in subgroup B1.
Discussion
Diversity and variation of fruit phenotypic traits of C. camphora
Our findings revealed exceptionally high phenotypic differentiation among provenances, significantly exceeding that reported for Xanthoceras sorbifolium (Sapindaceae) and Quercus fabri (Fagaceae) [25–31]. This suggested that local adaptation, rather than gene flow or genetic drift, was the predominant force driving fruit trait diversification in C. camphora across Jiangxi’s complex landscape. The substantial geographic isolation and habitat heterogeneity in this region likely imposed strong selective pressures, leading to pronounced genetic divergence among provenances. The moderate to high Shannon-Wiener indices (2.395–2.835) further confirmed that among provenance diversity was preserved despite this strong differentiation, a pattern consistent with that observed in Zanthoxylum armatum (Rutaceae) [32–33]. The notably lower diversity in species like Ziziphus jujuba Mill. (Rhamnaceae) may reflect their more restricted genetic base or different breeding systems [34–35].
The coefficient of variation provides insights into the stability and evolvability of traits. The highest coefficient of variation observed in PT (25.45%) indicated that this trait was highly responsive to environmental cues or under weaker genetic constraint, making it a potential target for selective breeding for specific fruit characteristics. This aligned with findings by Li et al. [36] and Gao et al. [37]. In contrast, the lower coefficient of variation in dimensional traits like FHD and FVD suggested they were under stronger stabilizing selection or were more canalized developmentally, which was crucial for maintaining basic fruit architecture. The high stability of FSD, FHD and FVD might be a unique feature of our sampled provenances [38–39].
Geographical variation and clustering results of fruit phenotypic traits in C. camphora
The correlation patterns unraveled the complex interplay between genotype and environment. The unexpected positive correlation of FTW, FV and PT with altitude contradicted the common trend of smaller fruits at higher elevations due to resource limitation [40–41]. This anomaly could be explained by our specific altitudinal range (32–388 m). Within this range, increased altitude likely mitigated summer heat stress and amplified diurnal temperature fluctuations, both of which could enhance photosynthetic efficiency and nutrient accumulation, ultimately facilitating larger fruit development [42]. Conversely, the negative correlation with summer temperature and sunshine duration underscored heat stress as a major limiting factor. Excessive sun and heat during Jiangxi’s summer might inhibit metabolic processes crucial for fruit growth [43]. The latitudinal and longitudinal clined further support that gradients in temperature and precipitation, tied to geographic position, were key drivers of phenotypic variation [44].
Principal component analysis efficiently reduced trait dimensionality, identifying FSD, FSD/FVD and FSD/FHD as key integrators of fruit morphology. This suggested that selection on these easily measurable traits could effectively improve multiple fruit characteristics simultaneously, a strategy more efficient than independent culling [45]. Cluster analysis grouped provenances into three distinct clusters, not strictly by geography, implying that microhabitat factors like soil composition and topography were important alongside climate [46]. The B2 cluster, characterized by larger and heavier fruits, represented a prime candidate pool for germplasm conservation and breeding programs aimed at enhancing yield [47]. Building on this, future research should directly link these advantageous phenotypic traits to economically important qualities, such as fruit oil content and nutritional value, as seen in Xanthoceras sorbifolium Bunge (Sapindacea) [48] and Fortunella crassifolia (Rutaceae) [49]. Furthermore, exploring the potential relationship between leaf essential oil content and fruit phenotypes could enable the non-destructive selection of high oil yielding varieties based on fruit traits, revolutionizing breeding strategies for C. camphora.
Conclusions
In conclusion, our study comprehensively demonstrated the rich diversity in fruit phenotypic traits among natural provenances of C. camphora in Jiangxi Province, which primarily originated from differences among provenances and reflected strong local adaptation to heterogeneous environments, as evidenced by significant correlations between traits and geographical-climatic factors. Furthermore, FSD, FSD/FVD, and FSD/FHD were identified as key integrative indicators for efficiently evaluating fruit traits. The superior germplasm resources identified, particularly those from the B2 subgroup characterized by larger fruit size and weight, provide a valuable material base for conservation and breeding. Based on these findings, future research should employ molecular markers to quantify the underlying genetic diversity and provenance structure, thereby establishing a foundation for the screening of germplasm resources.
Supporting information
S1 File. The values behind the means, standard deviations and other measures reported.
https://doi.org/10.1371/journal.pone.0319877.s001
(XLSX)
S3 File. The points extracted from images for analysis.
https://doi.org/10.1371/journal.pone.0319877.s003
(XLSX)
References
- 1. Lv DD. Study on the characteristics of old and notable trees in Jiangxi province. Acta Agric Univ Jiangxiensis. 2023.
- 2. Xiao ZF, Ai Q, Jin ZN, Zhang BH, Wang YB, Zhu YB. A study on the dynamic changes in growth rhythm and essential oil in linalool type Cinnamomum camphora dwarf forest. Acta Agric Univ Jiangxiensis. 2021;43(4):834–41.
- 3. Hu M, Wang YH, Yang MX, Zhong YD, Yu FX. Research advances on ornamental Cinnamomum camphora. Jiangxi Sci. 2020;38(6):846–50.
- 4. Yin M, Song X, He C, Li X, Li M, Li J, et al. Publisher Correction: The haplotype-resolved genome assembly of an ancient citrus variety provides insights into the domestication history and fruit trait formation of loose-skin mandarins. Genome Biol. 2025;26(1):78. pmid:40156054
- 5. Zhang B. Cultivation technology of Cinnamomum camphora and its role in garden landscape. Agric Technol Equip. 2021;4:144–5.
- 6. Wu W. Regional experiment and evaluation of growth and leaf essential oil characteristics of Cinnamomum camphora var. linaloolifera clones. J Fujian For Sci Technol. 2025;52(1):80–8.
- 7. Zhong YD, Liu LP, Li YQ, Wu ZX, Yang AH, Liu SJ. Sampling strategy of primary core germplasm of Cinnamomum camphora in China. J Southwest For Univ. 2020;40(4):1–13.
- 8. Eltaher S, Li J, Freeman B, Singh S, Ali GS. A genome-wide association study identified SNP markers and candidate genes associated with morphometric fruit quality traits in mangoes. BMC Genomics. 2025;26(1):120. pmid:39920570
- 9. Zhang Q, Luan R, Wang M, Zhang J, Yu F, Ping Y, et al. Research Progress of Spectral Imaging Techniques in Plant Phenotype Studies. Plants (Basel). 2024;13(21):3088. pmid:39520006
- 10. Sun Y, Ku TT, Xiong W, Liu QT, Dang TY, Zhang C, et al. Analysis on evolution rule of main traits of Zea mays L. varieties approved in Shandong Province in recent 20 years. Shandong Agric Sci. 2025;57(3):37–46.
- 11. Huang Y, Ren Z, Li D, Liu X. Phenotypic techniques and applications in fruit trees: a review. Plant Methods. 2020;16:107. pmid:32782454
- 12. Su S, Li ZJ, Ni JW, Geng YH, Xu XQ. Diversity analysis of fruit spikes and fruit traits in Picea abies. J Plant Res Environ. 2021;30(1):35–44.
- 13. Xu JM, Yin JJ, Cai GJ. Phenotypic diversity of fruits of artificially cultivated Elaeagnus mollis Diels. For Sci Technol. 2021;7(7):48–51.
- 14. Xie Q, Zhao QQ, Jiang L, Chen QX. Classification of phenotypic traits, identification of volatile compounds and characteristic aroma analysis of Canarium album fruit. Food Sci. 2024;45(1):175–85.
- 15. Peng HL. Study on the growth traits of seed and seedling of Melia azedarach from different provenances. Acta Agric Univ Jiangxiensis. 2022.
- 16. Wang Y, Wu SQ, Zhang JK, Wang YL, Feng Z, Zang DK, et al. Analysis on diversity of fruit phenotypic traits of Rhus chinensis from different provenances in Shandong Province. J Plant Res Environ. 2020;29(1):18–25.
- 17. Li DB, Wu M, Yu R, He TY, Rong JD, Zhen YS. Phenotypic diversity and its correlation with environmental factors in different provenances of Dendrocalamus latiflorus Munro. J Plant Res Environ. 2023;32(5):39–50.
- 18. Li CJ, Zhou L, Lu B, Dong KX, Liu CY. Study on phenotypic diversity of Cerasus tianschanica pojark germplasm resources. Acta Agric Boreal-Occident Sin. 2018;27(1):91–7.
- 19. Yao LM, Yuan J, Ma XH, Wang B. Genetic diversity analysis of Lablab purpureus germplasm resources based on morphological trait and SSR markers. Crops. 2025;41(1):35–45.
- 20. Sun H, Lin H, Shi JN, Mo DK, Zhu GY, Zang Z. Comparative study of remote sensing sampling technique in national forest inventory of Hunan province. J Cent South Univ For Technol. 2010;30(11):26–31.
- 21. Wu YD, Liu HF, Guan FC, Zong XC, Lu X, Cai CP, et al. Study on variation of fruit phenotypic traits of Tibetan apricot. J Southwest Minzu Univ (Natural Science Edition). 2019;45(3):254–8.
- 22.
Sokal RR, Rohlf FJ. Biometry: the principles and practice of statistics in biological research. WH Freeman; 1995.
- 23. Shannon CE. A mathematical theory of communication. Bell Syst Tech J. 1948;27(3):379–423.
- 24. Chen HW, Ji JY, Ye LY, Shen B, Cai C, Luo ZS, et al. Analysis of growth and leaf morphological variation of Ormosia hosiei from different origins and selection of excellent strains. J Zhejiang For Sci Technol. 2025;45(1):73–81.
- 25. Chen H, Chen J, Zhai R, Lavelle D, Jia Y, Tang Q, et al. Dissecting the genetic architecture of key agronomic traits in lettuce using a MAGIC population. Genome Biol. 2025;26(1):67. pmid:40122830
- 26. Doan PPT, Vuong HH, Kim J. Genetic foundation of leaf senescence: insights from natural and cultivated plant diversity. Plants (Basel). 2024;13(23):3405. pmid:39683197
- 27. Ren TT, E ST, Chen JH, Pan JJ, Dong SJ. A taxonomic study on phenotypic traits of Acer mono and Acer truncatum. Non-wood For Res. 2023;41(3):24–56.
- 28. Zhao J. Analysis on variation of leaf phenotype of Cinnamomum camphora (L.) Presl from different provenance in Jiangxi province. Chin Wild Plant Res. 2021;40(7):7–14.
- 29. Wang YL, Gu ZJ, Jia QX. Genetic variation analysis of phenotypic traits of 37 Xanthoceras sorbifolium elite germplasm. For Res. 2022;35(5):52–62.
- 30. Xiong SF, Wu LW, Chen YC, Gao M, Zhou XH, Wang YD. Variation in morphological characters and nutrient contents of Quercus fabri fruits from different provenances. For Res. 2020;33(2):93–102.
- 31. Dong SJ, Wang RX, Zhang HK, Chen JH, Liu LX, Yu QF. Analysis on diversity of fruit phenotypic characters of Armeniaca mandshurica from different provenances. J Plant Res Environ. 2020;29(6):42–50.
- 32. Wang HF, Wen K, Chen Z, Zang P, Gong X, Tang Y. Analysis of fruit phenotypic characters and screening of excellent germplasm of Zanthoxylum armatum DC. in Sichuan. Non-wood For Res. 2023;41(3):224–34.
- 33. Wang YH, Wu HQ, Yang Y, Wang J, Liu XQ, Wang LY, et al. Analysis of genetic diversity in phenotypic traits of 215 Capsicum annuum L. resources. J Shanxi Agric Univ. 2025;45(1):29–40.
- 34. Wu H, Su WL, Shi MJ, Xue XF, Ren HY, Wang YK, et al. Diversity analysis and comprehensive evaluation of Ziziphus jujuba Mill. fruit traits. J Plant Genet Resour. 2022;23(8):1613–25.
- 35. Zhai DC, Wu JF, Bai XH, Ye SF, Jiang YL, Yao Q. Preliminary evaluation of fruit characteristics and nut quality of Juglans cathayensis Dode var. formosana (Hayata) from Anhui province. For Sci Res. 2019;32(4):129–36.
- 36. Li HL, Hu WB, Hong QM, Pu WH, He Y, Zhang DX. Genetic diversity analysis of fruit traits of Hylocereus undatus germplasm resources. J Trop Subtrop Bot. 2019;27(4):432–8.
- 37. Cao Y, Tian D, Tang Z, Liu X, Hu W, Zhang Z, et al. OPIA: an open archive of plant images and related phenotypic traits. Nucleic Acids Res. 2024;52(D1):D1530–7. pmid:37930849
- 38. Li KX, Liang XJ, Zhu CS, An JC, Zeng XY, Yang ZY. Comparative analysis of phenotypic traits of fruits and seeds of Cinnamomum camphora from different provenances. Guangxi For Sci. 2020;49(1):1–6.
- 39. Gao YH, Wei M, Li KD, Mou FJ. Phenotypic variation of Citrus hongheensis fruit from different provenances. Non-wood For Res. 2021;39(4):88–96.
- 40. He MY, Zhang BH, Xiao ZF, Cheng HJ, Luo MT, Xiao CL, et al. Study on phenotypic diversity of Cinnamomum camphora in Hu’nan province. Tillage Cultiv. 2022;42(5):15–9.
- 41. Xing KF, Xie HX, Zhang LD, Zhou J, Feng LY, Zhang HX, et al. Diversity analysis on fruit phenotype of wild Camellia oleifera from different provenances. J Plant Genet Resour. 2024;25(5):1106–17.
- 42. Guo CC, Liao K, Shi D, Jiang NL, Tang YY, Diao YQ. Relationship between fruit and seed traits and altitude in wild Prunus armeniaca. J Fruit Sci. 2023;40(2):316–26.
- 43. Corbin JPM, Best RJ, Garthwaite IJ, Cooper HF, Doughty CE, Gehring CA, et al. Hyperspectral leaf reflectance detects interactive genetic and environmental effects on tree phenotypes, enabling large-scale monitoring and restoration planning under climate change. Plant Cell Environ. 2025;48(3):1842–57. pmid:39497286
- 44. Yuan CJ, Li H, Chen R, Yu DH. Fruit trait characteristics of Magnolia wilsonii (Finet et Gagn) Rehd. and its phenotypic diversity. Jiangsu Agric Sci. 2020;48(8):142–8.
- 45. Gong LD, Ma J, Li ZQ, Wu C. Analysis and evaluation of fruit characters in 101 Macadamia germplasm resources. Trop Agric Sci Technol. 2025;48(2):13–8.
- 46. Tao SY, Wang P, Pang HY, Li HL. Comprehensive evaluation of phenotypic traits and quality of Actinidia arguta fruits from different provenances. Jiangsu Agric Sci. 2024;52(22):168–73.
- 47. Tang Y, Li L, Wang HF, Gong X, Wen K. The comprehensive evaluation of phenotypic traits of thornless (less thorny) Zanthoxyhum bungeanum types in ganzi prefecture. Non-wood For Res. 2025;43(1):40–8.
- 48. Li YY, Ao Y, Zhao LL, Xu XY, Chen YH, Na RS. Regularities of fruit growing and development and its inclusion of Xanthoceras sorbifolium. J Nanjing For Univ. 2025;49(1):31–7.
- 49. Liu XF, Zhang J, Zhu SP, Jiang D, Zhao XC. Relationships between essential oil content, oil gland density and fruit morphology in Citrus japonica Thunb. S China Fruits. 2013;42(1):1–4.