Heterosis and Mixed Genetic Analysis of Flowering Traits in Cross Breeding of Day-Neutral Chrysanthemum (Asteraceae)

work


Introduction
The day-neutral multiflora chrysanthemum is an important variety group of chrysanthemums (Chrysanthemum × morifolium Ramat.) that can flower throughout the year without being influenced by daylength under appropriate temperature conditions [1,2]. Compared with short-day multiflora chrysanthemums, day-neutral multiflora chrysanthemums have a wider application range and lower production costs, although grower acceptance requires education regarding production changes. However, unsuitable flowering time can lead to dwarf plants, abnormal development of flower buds (termed "crown" Agronomy 2023, 13, 2107 2 of 16 or "abnormal buds"), and decreased petal layers of day-neutral chrysanthemums, seriously affecting their ornamental value [3]. To prevent early flowering, day-neutral stock plants are treated with ethylene to delay flower bud initiation [1][2][3]. A day-neutral crop ideotype was formulated [1] and tested [3]. Day-neutral chrysanthemums, by definition, produce equal numbers of leaves subtending the primary apical meristem in both shortand long-day photoperiods, and they flower regardless of the photoperiod [1,3]. Day neutrality is a recessive trait, although it is influenced by heat delay, as are many short-day chrysanthemums [1,3]. Previous studies have found that plant hormones [4][5][6], light [7,8], temperature [9], and nutrients [10] are critical factors affecting the plant type of chrysanthemums, while CYC and MADS-box genes are key factors affecting the flower type [11]. The budding date (flower bud initiation) and full flowering date (anthesis; flower bud development) are typical traits to describe the flowering stage [12] and they are influenced by the long-day leaf number [1,3], while the genetic mechanism of abnormal or crown buds in chrysanthemums has not yet been reported. Day-neutral chrysanthemums characteristically have equal leaf numbers and primary stem lengths when grown under both short-and long-day photoperiods [1,3]. In addition, studies have shown that the photoperiodic pathway, vernalization pathway, gibberellin pathway, and age pathway are all involved in regulating the flowering time of chrysanthemums [7,[13][14][15][16][17][18][19]. However, these studies have mainly focused on chrysanthemum cultivars that respond to short-day photoperiods to flower, and there have been few studies on the flowering mechanism of day-neutral chrysanthemums; furthermore, the related traits affecting their ornamental value are also lacking research. Therefore, this study uses phenotypic data obtained from cross breeding to analyze heterosis and inheritance patterns of these important traits of dayneutral chrysanthemums. This study has practical significance for breeding new cultivars of day-neutral chrysanthemums with commercially viable traits, and it can also provide a reference for the genetic mechanism of these traits.
Chrysanthemums have the characteristics of high ploidy levels (2n = 6x = 54) and high genomic heterozygosity [3]. F 1 traits show wide segregation, exhibiting the genetic characteristics of quantitative traits that cannot be analyzed according to classical genetic theory [20]. Studies have shown that the genes controlling quantitative trait inheritance have significant differences in their effects and may be controlled by major genes (qualitative), multiple genes (quantitative), or both, exhibiting major gene plus polygene mixed genetic models [21]. Zhang [22] summarized this inheritance pattern as the major gene plus polygene mixed genetic model. Currently, this model has been applied to quantitative trait research in various crops, such as the flowering time in Cicer arietinum [23] and Gossypium barbadense [24], plant height in Brassica napus L. [25], the grain-filling duration in Triticum aestivum [26], fruit-related traits in Chinese cherry (Cerasus pseudocerasus (Lindl.)) [27] and in Cucumis melo L. [28], cabbage black rot resistance in Brassica oleracea var. capitate [29], and melon aphid resistance in Cucumis sativus [30]. Studies have shown that many quantitative traits in chrysanthemums are consistent with a major gene plus polygene mixed genetic model. Gao [31] found that the leaf length/width of the small-flowered chrysanthemum was controlled by two pairs of additive-dominant major genes, and the lobe length/vein length was controlled by one pair of additive-dominant major genes; the heritabilities of the major genes were all greater than 30%. Song [32] found that the corolla tube merged degree (CTMD) and the relative number of ray florets (RNRF) in chrysanthemums were controlled by two pairs of additive-dominant major genes, and the heritabilities of the major genes were both >50%. Tang [33] found that the number of ray florets and the width of disc florets in anemone-typed chrysanthemums were controlled by two pairs of major genes, and the heritabilities of the major genes were 93% and 43%, respectively. Fu [34] found that the inheritance of aphid resistance in cut-flower chrysanthemums was controlled by two pairs of additive-dominant major genes, with a heritability of >90%. Zhao [35] found that the green-center (disc florets) relative area and the period of inflorescence during which the center stayed green of cut-flower chrysanthemums were controlled by two pairs of major genes, and the heritabilities of both major genes were >95%. Peng [36] found that the primary branch number and the branch height of cut-flower chrysanthemums were negatively and completely dominant, and the heritabilities of the major gene were 51% and 53%, respectively. Other researchers have also conducted genetic analyses on cold tolerance [37], flower type [38], and nutritional traits [39] in chrysanthemums. These studies provide important theoretical references for the utilization of heterosis and the breeding of new cultivars of chrysanthemums, but there is little research on the genetic analysis of flowering-related traits of day-neutral chrysanthemums. The genetic mechanism of flowering-related traits in day-neutral chrysanthemums is still unclear, including whether it is controlled by major genes, although day neutrality itself fits a single-gene recessive inheritance [3]. In this study, we use the major gene plus polygene mixed inheritance model to analyze the flowering traits of day-neutral chrysanthemums. This is the most common research method for the improvement of ornamental characteristics and genetic analysis of chrysanthemums to construct a hybrid population and to analyze the genetic model of main ornamental characteristics by using the sexual hybridization technique [32,40]. Long-term glasshouse and field cultivation have shown that early flowering in day-neutral chrysanthemums can lead to dwarf plants and uneven flower bud differentiation [1,3]. Based on this growth characteristic, this study constructed an F 1 population of the shorter plant type, day-neutral chrysanthemum '82-81-19,' and the taller plant type, late-flowering chrysanthemum '388Q-76.' Then, six traits were selected, including plant height, plant crown width, abnormal bud, budding date, full flowering date, and the number of petal layers, for major gene plus polygene mixed inheritance model analysis. This study aims to lay a foundation for further exploration of molecular markers associated with these important traits, while also providing a theoretical basis for the breeding of new cultivars.

Plant Materials
In the fall of 2017, the shorter plant type (low long-day leaf number), the day-neutral chrysanthemum '82-81-19,' was used as the female parent, and the taller plant type, the late-flowering short-day chrysanthemum '388Q-76' (higher long-day leaf number), was used as the male parent for hybridization (Table 1, Supplementary Figure S1). The dayneutral unnamed inbred-hybrid selection, '82-81-19,' was developed at the University of Minnesota, St. Paul, MN in 1982. It has yellow flowers and it is USDA Z4 winter hardy (Anderson, 1982, unpub. data). Pollinations to seed harvest were conducted at the Beijing Forestry University; a fully executed Material Transfer Agreement (MTA) between the University of Minnesota and Beijing Forestry University (MTA OTC Agreement 007 A20160059) permitted cross-hybridizations. When the ray florets just began to change color, the inner disc florets and the outer ray florets of the female buds were removed and bagged with sulfuric acid paper bags. Meanwhile, the first-opening capitula of the male parent were bagged, and pollen was collected on a sunny and windless morning after 9 o'clock. When the stigmas of the female extended and forked at acute angles (receptive), the female was repeatedly pollinated for three consecutive days using a brush. After pollination, the female plants were kept in a ventilated and dry place. The flower heads were harvested after about 70 days, and a total of 180 seeds were harvested. These seeds were sown in a solar greenhouse in January 2018, and the seedlings were transplanted into 81-hole plug trays when they grew to 3-4 true leaves. They were transplanted in 25 cm × 25 cm pots when they grew to 8-9 leaves on the primary meristem, and 135 hybrids were obtained. In March 2018, the 135 hybrids were numbered separately, and, together with their parents, they were cloned (rooted stem tip cuttings) into at least three plants/genotype. All F 1 generation individuals and parents were planted and managed under conventional field conditions at Beijing Forestry University plots for recording the traits.

Investigation of Field Traits
According to the Handbook of Distinctiveness, Uniformity, and Stability Test of New Plant Cultivars of the People's Republic of China (Chrysanthemum) and considering the growth characteristics of day-neutral chrysanthemums, six traits, such as abnormal (crown) buds, plant height and crown width, budding date at flower bud initiation (visible bud date, VBD), full flowering date (anthesis), and the number of petal layers, were selected as test indexes (Table 1). Data observations were carried out in the fall of 2018, and three clonal plants of each genotype were randomly selected for the measurement of six traits. The measurements were repeated three times, and the average value was taken. The measuring methods for the plant traits recorded for plant height, plant crown width, budding date (VBD), full flowering date (anthesis), and the number of petal layers are shown in Table 2. To ensure the accuracy and consistency of the results, all traits were measured by one person.

Plant Traits Measuring Methods
Plant height (cm) The distance from the base of the stem to the highest point of inflorescence Plant crown width (cm) The diameter of its development is measured from the top surface of the plant Budding date (VBD) The specific time when the bud size is about 5 mm Full flowering date (anthesis) The specific time when 50% of the buds reach full flowering (anthesis) No. of petal layers The total number of petal (ray floret) whorls in the capitulum

Definition and Measurement Method of Abnormal (Crown) Bud
The terms "abnormal" or "crown bud" refer to the reproductive phase change phenomenon of strap-or willow-shaped leaves at the top of the stem subtending the floral meristem [1,3]. Abnormal or crown buds occur when the external light and temperature cannot meet the requirements for flower bud development in chrysanthemums. Short-day chrysanthemums will develop crown buds under long-day photoperiods [1,3]. Once the abnormal or crown buds appear, the stems and leaves will naturally age, and the flower buds will develop malformed, resulting in unfilled and malformed flowers even if they bloom [41]. Crown bud or a lack thereof is one of the most important criteria for judging the quality of day-neutral chrysanthemums [1,3]. Currently, little research has focused on the formation and causes of abnormal buds, although phenotypic traits have been initially identified [1,3]. Therefore, in this study, additional criteria for testing the crown buds of day-neutral chrysanthemums were developed based on the number of crown buds and the influence degree of crown buds on the flowering of day-neutral chrysanthemums according to the data from field observations (Table 3), supplemented with previous research efforts [1,3].

Definition and Calculation of the Photoperiod
The photoperiod (light duration) of the VBD and full flowering (anthesis) date is another key criterion for judging the quality of day-neutral chrysanthemums [1,3,42]. According to Cockshull's research on the flowering period of the chrysanthemum, a daylength greater than or equal to 13.5 h is defined as long daylight [42]. Combined with the sunrise and sunset times in Beijing, the authors classified the flowering stages of chrysanthemums into five types: extremely early flowering type (flowering type with daylight longer than 13.5 h), early flowering type (flowering type with daylight of 12.0-13.5 h), medium flowering type (flowering type with daylight of 11.0-12.0 h), late flowering type (flowering type with daylight of 10.0-11.0 h), and extremely late flowering type (flowering type with daylight shorter than 10.0 h). Among them, the extremely early flowering type can normally produce buds and flowers under long-day photoperiods of greater than or equal to 13.5 h, and it belongs to the day-neutral chrysanthemum group. Therefore, this study recorded the budding date and full flowering date of 135 F 1 generation hybrids (Table 2) and converted them into the sunrise and sunset times for Beijing using records from the Beijing Astronomical Observatory (Supplementary Table S1). The light duration (photoperiod) was calculated using Equation (1), and the average value was taken. Based on the above criteria, the flowering types of F 1 generation chrysanthemums were categorized, and subsequent heterosis and mixed genetic analyses were carried out. At the same time, hybrid offspring with daylength greater than or equal to 13.5 h and normal flowering were selected as day-neutral chrysanthemum germplasm resources.

Heterosis Analysis and Significance Test
Heterosis is measured by mid-parent heterosis (MPV) and the mid-parent heterosis (Hm) rate (RHm) [43]. The MPV is the mean value of a certain trait in the parents. The Hm is defined as the difference between the average value of F 1 individuals (Fm) and the MPV, i.e., Hm = Fm − MPV, and the mid-parent heterosis rate (RHm/%) = [(Fm − MPV)/MPV] × 100%. Microsoft Excel 2007 and SPSS 18.0 software were used for the statistical analysis of basic descriptive data and t-tests for the sample means.

Mixed Genetic Analysis
The F 1 chrysanthemum population was considered a pseudo-F 2 population for genetic analysis by the double-pseudo-testcross strategy. A single-generation segregation analysis, as described by Gai [44], was used to analyze the mixed inheritance model of chrysanthemum flowering traits. A total of 11 kinds of genetic models, of which A represents a pair of major genes and B represents two pairs of major genes, were included, and the genetic parameters of each trait were calculated, including the maximum likelihood value (MLV) and Akaike's information criterion (AIC) [45]. The best-fitting genetic models were selected according to the principle of minimum AIC value and the results of the fitness tests, including the uniformity tests U 1 2 , U 2 2 , and U 3 2 , Smirnov's statistics (nW 2 ), and Kolmogorov's statistics (Dn). Using the selected optimal model, genetic parameters, such as Agronomy 2023, 13, 2107 6 of 16 effect values, variance, and heritability of the major gene were estimated by the least squares method. The major gene heritability was calculated using the equation h 2 mg = σ 2 mg /σ 2 p (h 2 mg means major gene heritability, σ 2 mg means major gene variance, and σ 2 p means phenotypic variance). The software for the mixed genetic model analysis was provided by the National Key Laboratory of Crop Genetics and Germplasm Enhancement, Soybean Research Institute, Nanjing Agricultural University [22].

Screening Methods for Hybrid Progeny
Individual plant selection is a simple and effective breeding method in flower breeding, and the percentile scoring method has been proven to be a fast and more economical approach in the early stage of many floral breeding programs [46]. In this study, the percentile scoring method was used for plant selection of chrysanthemum hybrid progeny. Based on the breeding objectives and previous observations, the traits of the hybrid progeny of chrysanthemum were divided into three major categories: plant type, flowering date, and petal type. A total of 30 points were assigned for the plant type (15 points each for plant height and plant crown width), 60 points were assigned for the flowering date (20 points each for abnormal bud, budding date, and full flowering date), and 10 points were assigned for the petal type (no. of petal layers). The specific scoring criteria are shown in Table 4. Table 4. Traits and scores for the day-neutral chrysanthemum individual plant selection.

Phenotypic Distribution and Heterosis of Flowering-Related Traits of the F 1 Population
The phenotypic data for the F 1 generation of the six flowering-related traits show that the coefficient of variation for the six traits observed ranged from 30% to 84%, with an average coefficient of variation of 54.2% (Table 5). Among them, the coefficient of variation for abnormal bud formation was 66%, the coefficients of variation for plant height and plant crown width were 34% and 30%, respectively, and the coefficient of variation for the daylength was 49% at the budding date and 65% at the full flowering date. The CVs of these six traits revealed great separation in the F 1 generation. From the skewness and kurtosis in Table 5 and the histogram of trait distribution in Figure 1, it can be seen that the distribution of these six traits was between the parents, and most of the traits showed a normal distribution trend, which was typical of traits controlled by multiple genes and was suitable for genetic analysis.
The mid-parent heteroses of the six traits in the hybrid combination, including abnormal (crown) bud, plant height, plant crown width, daylength at budding date, and daylength at full flowering date, were all positive, indicating positive genetic dominance ( Table 6). The distribution of the rate of mid-parent heterosis ranged from 2% to 45.5%. The mid-parent heterosis rate for petal layers was −0.3%, indicating that the heterosis for petal layers decreased in the F 1 generation. At the same time, there were individuals with positive or negative traits over their parents in the F 1 population, indicating the prevalence of super parental segregation in the hybrid progeny ( Figure 1). However, the average values of the six traits in the F 1 population were all between their parents, indicating that this super parental segregation phenomenon did not form a heterobeltiosis (when the hybrid's performance exceeds its best parent). showed a normal distribution trend, which was typical of traits controlled by multiple genes and was suitable for genetic analysis.  The mid-parent heteroses of the six traits in the hybrid combination, including abnormal (crown) bud, plant height, plant crown width, daylength at budding date, and daylength at full flowering date, were all positive, indicating positive genetic dominance ( Table 6). The distribution of the rate of mid-parent heterosis ranged from 2% to 45.5%. The mid-parent heterosis rate for petal layers was −0.3%, indicating that the heterosis for petal layers decreased in the F1 generation. At the same time, there were individuals with positive or negative traits over their parents in the F1 population, indicating the prevalence of super parental segregation in the hybrid progeny ( Figure 1). However, the average values of the six traits in the F1 population were all between their parents, indicating that this super parental segregation phenomenon did not form a heterobeltiosis (when the hybrid's performance exceeds its best parent).  Note: * and ** indicate significant differences at p < 0.05 and p < 0.01 levels, respectively. The difference between the F 1 and the mid-parent value was analyzed using a one-sample t-test.

Suitability Analysis of the Optimal Genetic Model for Flowering-Related Traits in Day-Neutral Chrysanthemums
The average values of phenotypic data for six traits of the cross-combination F 1 population in this study were analyzed with the major gene plus polygene mixed inheritance model for quantitative traits of the single-generation segregating method. The Akaike's information criterion (AIC) values were obtained by calculating the pairing of a total of 11 models with phenotypic data in both categories A and B (Table 7). According to the AIC minimum criterion, the models with the smallest AIC value and the models with an AIC value close to the minimum were selected as candidate models. Taking abnormal bud as an example, the calculation results of various genetic models were compared, and two models, B-3 and B-4, had the smallest AIC values of −3725.6512 and 401.9971, respectively; they could be used as candidate optimal models. The candidate models for each trait were selected according to the AIC minimum criteria ( Table 8). The fitness of the candidate models was tested using U 1 2 , U 2 2 , U 3 2 , nW 2 , and Dn, and the model with the least number of significant statistics was selected as the optimal model. Abnormal (crown) bud and daylength at budding date (VBD) both fit the B-3 model, which was controlled by two pairs of additive major genes. Plant height and plant crown width fit the A-0 model, which was not controlled by major genes and might be controlled by multiple genes with strong environmental influence. The number of petal layers fit the B-1 model, which was controlled by two pairs of additive major genes, and the major genes showed an additive-dominant-epistatic model. Daylength at full flowering date fit the A-1 model, which was controlled by one pair of major genes, with the major genes exhibiting additive and partially dominant or over-dominant effects.

Estimation of Genetic Parameters for Flowering-Related Traits in Day-Neutral Chrysanthemum
The genetic parameters of the optimal genetic model were estimated based on the maximum likelihood of the genetic model parameters for flowering-related traits ( Table 9). The inheritance of abnormal (crown) buds was controlled by two pairs of additive major genes (da), whose additive effects were da = 0.347 and da = 1.3349. The additive effect (db) of the second pair of genes was much larger than that of the first pair, indicating that the additive effect was dominated by the second pair of genes. The inheritance of daylength at budding (VBD) date was also controlled by two pairs of additive major genes, and the additive effects were da = 0.347 and db = 1.3349, respectively. The inheritance of daylength at full flowering (anthesis) date was controlled by one pair of additive-dominant major genes; the additive effect was da = 1.3595, and the dominant effect was ha = −1.3586. The inheritance of the number of petal layers was controlled by two pairs of additive-dominant major genes. Among them, the additive and dominant effects of the first pair of genes were 4.9622 and −7.4372, and the second pair of genes were db = 4.9527 and hb = −4.9179. The additive-additive, additive-dominant, dominant-additive, and dominant-dominant effects between the two pairs of major genes were i = 4.9516, jab = −4.9178, jba = −2.4682, and l = 7.3981, respectively. The effect of epistatic interaction between the two pairs of major genes was relatively large and mainly synergistic. The major gene heritabilities of these four traits were h 2 mg = 1.0, 0.9871, 0.7240, and 0.5612, all of which were highly heritable. Plant height and plant crown width were not controlled by major genes and were speculated to be controlled by micro-effective polygenes.

Correlation Analysis of Flowering Traits of Day-Neutral Chrysanthemum
The Pearson's correlation analysis (Table 10) showed that abnormal (crown) bud was significantly correlated with plant height, plant crown width, daylength at budding date, and daylength at full flowering date, among which abnormal bud and plant height showed a highly significant negative correlation (p < 0.01), indicating that the taller the plant, the less significant the abnormal bud. Plant height was negatively correlated with plant crown width, daylength at budding date, and daylength at full flowering date (p < 0.01), indicating that the earlier the flowering, the shorter the plant. Plant crown width, daylength at budding date, and daylength at full flowering date all showed highly significant positive correlations (p < 0.01), indicating that early flowering promoted branching. There was no significant correlation between the number of petal layers and the other five traits, i.e., abnormal (crown) bud, plant height, plant crown width, daylength at budding (VBD) date, and daylength at full flowering (anthesis) date (p > 0.05), indicating that petal layers may be an independent evolutionary trait. Note: * and ** indicate significant differences at p < 0.05 and p < 0.01 levels, respectively.

Screening of Selected Day-Neutral Chrysanthemum Genotypes
The traits and comprehensive evaluation scores of the selected day-neutral chrysanthemum genotypes are shown in Table 11. A total of eight genotypes were selected with comprehensive evaluation scores > 60. Among them, genotypes No.

Establishment of Evaluation Criteria for Flowering Time of Day-Neutral Multiflora Chrysanthemum
Flowering time is an important factor affecting the large-scale production and application of chrysanthemums. Several evaluation methods have been developed to define the flowering period of chrysanthemums [1][2][3]. For example, the guidelines for testing chrysanthemums implemented by the International Union for the Protection of New Varieties of Plants (UPOV) divided the flowering period of chrysanthemums into three classes of short-day response groups: early flowering, medium flowering, and late flowering [47]. Previous studies have recorded the flowering period by the number of days from planting to the first flower [42]. However, more recent studies creating a day-neutral ideotype have added other specific criteria [1,3]. Based on the statistics of the photoperiod length in Beijing combined with the flowering time of chrysanthemums, Zhang divided the flowering period of chrysanthemums into five major groups [40] based on the photoperiodic criteria [38], among which extremely early flowering types can flower normally under long-day conditions and belong to the day-neutral group. However, the above classification method cannot accurately reflect the flowering status of day-neutral chrysanthemums because of their long flowering duration and uneven opening. Therefore, this study quantified the main flowering traits of chrysanthemums, such as the budding (VBD) date, full flowering (anthesis) date, and the number of petal layers, using the flowering grading method and introduced the definition of "abnormal bud" into the evaluation criteria for flowering traits of day-neutral chrysanthemums, which reflected the opening status of chrysanthemums more objectively and comprehensively. However, this differed from previous research that has shown that the number of leaves subtending the primary apical meristem are equal in day-neutral genotypes, regardless of short-and long-day photoperiod treatments [1,3]. Based on this method, the flowering time of 135 F 1 generation chrysanthemums was converted into the daylength, and the degree of abnormal bud that affects their flowering was measured and described. This classification method can ensure the accuracy of the data and unify the genetic analysis of the flowering time of day-neutral chrysanthemums.

Heterosis of Flowering-Related Traits in Day-Neutral Chrysanthemums
Utilizing heterosis is one of the important ways to improve the flowering period of chrysanthemums. From a genetic basis, the probability of getting a super parental segregating population is higher when selecting parents with large differences in genetic loci [48]. Previously, heat-tolerant [3], day-neutral [1,3] and early-flowering [40] cultivars of chrysanthemums have been produced through cross-breeding. However, the high heterozygosity of the chrysanthemum genome leads to large differences in the heterosis of the same trait between different cross combinations and different traits between the same cross combinations, which increases the difficulty of obtaining hybrid progeny with comprehensive traits. Many researchers have obtained chrysanthemum genotypes or cultivars with improved traits through genetic modification [49][50][51]. However, it is still difficult to obtain and promote transgenic chrysanthemum cultivars due to the difficulty and high cost of transgenic technology and public acceptance of transgenic chrysanthemums. Sexual hybridization is still the most effective way to improve the flowering date of chrysanthemums.
In the early stage of this study, it was found that there were more super parental strains with early flowering and large plant types in the cross combination consisting of the dayneutral chrysanthemum Mn. Selection '82-81-19' and the late-flowering chrysanthemum '388Q-76.' Therefore, according to the chrysanthemum breeding objectives, we analyzed the heterosis of the 135 F 1 generation for six important flowering traits, including abnormal (crown) bud, plant height, plant crown width, daylength at budding (VBD) date, daylength at full flowering (anthesis) date, and the number of petal layers. The coefficients of variation for these six traits ranged from 30% to 84%, indicating a significantly widened genetic basis. Moreover, the heteroses of these six traits were all enhanced in the F 1 population, which indicated that this combination could be used for further breeding of day-neutral chrysanthemums. In future studies, increasing the number of F 1 generations can obtain hybrids with excellent comprehensive traits. The results of this study are inconsistent with the study by Jiang [52] but consistent with the results of Zhang [12], which may be related to the degree of heterogeneity of the parents and the number of hybrid progeny.

Genetic Effects of Flowering-Related Traits in Day-Neutral Chrysanthemums
The flowering period and floral traits of chrysanthemums are all quantitative traits with complex inheritance patterns. Statistical analysis based on the mean value can reflect the direction of trait variation, but it is difficult to essentially resolve the inheritance pattern of traits. In recent years, researchers have explored the function of key genes in the flowering of chrysanthemums based on the flowering pattern of the model plant Arabidopsis thaliana [7,16] and have made relatively good progress in research. It is currently unknown whether there are additional genes involved in the flowering of day-neutral chrysanthemums, which are not regulated by photoperiod. Previous studies have shown that the major gene plus polygene mixed inheritance model can help to resolve the inheritance pattern of complex ornamental traits to locate new genes in chrysanthemums [34,37]. Therefore, this study used this method to analyze six traits of F 1 generations, and the results showed that abnormal (crown) bud was controlled by a pair of major genes with a heritability of h 2 mg = 1.0; plant height and plant crown width were not controlled by major genes; budding date and the number of petal layers were controlled by two pairs of additive major genes with a heritability of ha = 0.98 and hb = 0.56, respectively; and full flowering date was controlled by a pair of major genes with a heritability of h 2 mg = 0.7240. The existence of these major genes provides an important research basis for finding molecular markers closely linked to these genes by the QTL gene mapping method and improving the breeding efficiency of chrysanthemums through molecular marker-assisted breeding methods. The heritabilities of the four traits of abnormal (crown) bud, budding (VBD) date, flowering (anthesis) date, and the number of petal layers were >50%, indicating high heritability and suitability for selection in early generations. In future studies, this genetic population can be used to construct a high-density genetic map to locate the key genes regulating these six traits.

Breeding Prospects of Day-Neutral Chrysanthemums
The selection and breeding of new cultivars of day-neutral chrysanthemums have not only enriched the chrysanthemum world but also greatly reduced production costs, which is of great significance to the year-round production of chrysanthemums. The first day-neutral and heat-delay-insensitive chrysanthemum was released to the market in 1988 [53]. This selection, '83-267-3,' flowered in 24 h light and was released for the potted chrysanthemum market. Commercial grower acceptance presented challenges because they had already invested in black cloths for the production of short-day chrysanthemums. Future research involving day-neutral cultivars will necessitate educating growers as well as consumers on their benefits.
In the process of breeding day-neutral chrysanthemum cultivars, we can identify the photoperiod sensitivity of early-flowering chrysanthemum cultivars, such as summerflowering chrysanthemum or four-season chrysanthemum, screen out chrysanthemum genotypes that are not sensitive to photoperiod in flower bud differentiation and developmental stages, and use them as parents for hybrid combinations.
Studying the genetic patterns and molecular mechanisms of important traits in dayneutral chrysanthemums is beneficial to hybrid combination selection and transgenic breeding. Based on the molecular regulatory mechanism of the model plant, A. thaliana, and light-insensitive crops, genetic linkage maps and physical maps can be constructed for the genetic characteristics of day-neutral chrysanthemums, and the day-neutral loci can be finely located to isolate and identify the genes responsible for causing the chrysanthemum's photoperiod insensitivity. At the same time, the separation and identification of homologous genes, such as PhyB, PRR, CO, FT, etc., in chrysanthemums can be carried out to study the gene expression differences among different photoperiodic chrysanthemum cultivars and analyze the causes of the differences to analyze the molecular mechanism and guide the breeding/selection of day-neutral chrysanthemums. In addition to internal genetic factors, external environmental factors can also significantly affect plant growth and development. Low temperature is an important factor affecting the flowering of chrysanthemums. It is possible to research the low-temperature requirements of existing day-neutral chrysanthemum cultivars to understand the optimal temperature range for the induction of flower bud differentiation and induction in order to guide the large-scale production of day-neutral chrysanthemums.

Conclusions
The heterosis and genetic effects of six important flowering-related traits in day-neutral chrysanthemums were estimated by phenotypic data observations of two chrysanthemum genotypic selections with significant differences in flowering time and plant type as parents. The results showed that heterosis and super parental segregation were generally present in these six traits in the hybrid progeny. Using the major gene plus polygene mixed inheritance model in the quantitative traits, the genetic analysis of these traits was performed in a single segregating generation. Plant height and plant crown width were mainly controlled by micro-effective polygenes, while abnormal (crown) bud and budding date were controlled by two pairs of major genes. The full flowering date was controlled by a pair of major genes, and the number of petal layers was controlled by two pairs of additive genes. Moreover, the major genes of these four traits mainly exhibited additive effects. Using a percentile scoring method, eight day-neutral chrysanthemum genotypes were selected from the hybrid progeny, which can be used for breeding new day-neutral chrysanthemum cultivars. Given the importance of day-neutral chrysanthemums in the year-round production of chrysanthemums, materials with photoperiod insensitivity and excellent comprehensive traits can be selected as parents for cross-breeding. Furthermore, the existence of these major genes provides an important theoretical basis for gene localization and molecular markerassisted breeding research of flowering-related traits in day-neutral chrysanthemums.