Introduction
The African marigold (Tagetes erecta L.), a member of the Asteraceae family, is native to Mexico and Central America and widely cultivated for its vibrant, hybrid flowers. In India, it is extensively grown for ornamental purposes such as garlands, landscaping and festival decorations. Beyond aesthetics, marigold plays a functional role in integrated pest management as an effective trap crop and a natural source of nematicidal compounds that protect vegetables and cereals (Hooks et al., Reference Hooks, Wang, Ploeg and McSorley2010). It also holds industrial significance in the extraction of essential oils and carotenoids (Shaikh et al., Reference Shaikh, Ray and Singhal2023).
As of 2023–24, marigold cultivation in India spans approximately 86.05 thousand hectares, producing 954.7 thousand metric tonnes of flowers annually (Anonymous, 2024). The crop exhibits a high degree of cross-pollination and has a composite inflorescence (capitulum) composed of ray and disc florets. Male-sterile and male-fertile lines, each with distinct floral morphologies, have enabled successful F₁ hybrid seed production. ICAR-Indian Institute of Horticultural Research, Bengaluru, India has developed and released four public-sector marigold hybrids. However, the rising demand for hybrid seeds has outpaced supply, underscoring the need for efficient breeding strategies and deeper understanding of reproductive biology, particularly pollen performance.
Pollen biology including morphology, viability and in vitro germination is central to reproductive success and hybrid seed production. These parameters are critical not only for optimizing artificial pollination but also for understanding taxonomic relationships, guiding selection in breeding programmes and preserving germplasm. Pollen morphological traits, such as size, shape, number of apertures and exine ornamentation, aid in cultivar identification and phylogenetic studies (Souza et al., Reference Souza, Oliveira, Aona, Souza, Soares, Rossi and Souza2021; Dos Santos Mota et al., Reference Dos Santos Mota, de Faro, Soares, de Almeida, Souza, Aona and de Souza2024). However, pollen viability tends to decline rapidly after dehiscence due to environmental stress, directly affecting pollination efficiency and seed set (Mog et al., Reference Mog, Veena, Adiga, Hebbar, Shamsudheen, Manjesh and Yadav2023).
Although previous studies have described marigold pollen as medium-sized and prolate spheroidal with a polar/equatorial (P/E) ratio of 1.02 (Egbe et al., Reference Egbe, Iso and Andrew2023; Hayat et al., Reference Hayat, Khan, Khan and Shah2023), there remains a lack of detailed comparative analysis of pollen traits across genotypes, especially using advanced techniques like scanning electron microscopy (SEM). Furthermore, while in vitro germination has been applied in marigold (Thaneshwari et al., Reference Thaneshwari, Prakash, Kumari, Thakur and Sharma2019) and other ornamentals such as chrysanthemum (Miler and Wozny, Reference Miler and Wozny2021; Sunny et al., Reference Sunny, Kumar, Rajasekharan and Dhananjaya2024), standardization of culture conditions including the role of osmotic agents like polyethylene glycol (PEG) and sugars like sucrose remains underexplored for marigold.
This study addresses these gaps by:
1. Characterizing pollen morphology of selected marigold genotypes using SEM.
2. Assessing pollen viability using two staining methods Alexander’s stain and acetocarmine.
3. Standardizing in vitro pollen germination media with varying concentrations of sucrose and PEG.
The findings will not only contribute to optimizing hybrid seed production in T. erecta but also support pollen conservation strategies for both short- and long-term breeding applications.
Materials and methods
Plant materials
Four marigold (T. erecta) inbred lines (Fig. 1), IIHRMO 12, IIHRMO 53, IIHRMY 2-1 and IIHRMY 1-4, that are pollen parents of four F1 hybrids, namely Arka Shubha (IIHRMOS-1 × IIHRMO 12), Arka Vibha (IIHRMOS-1 × IIHRMO 53), Arka Abhi (IIHRMOS-1 × IIHRMY 2-1) and Arka Bhanu (IIHRMOS-1 × IIHRMY 1-4), respectively, and pollen grains were used for pollen morphology, viability and in vitro germination studies. The plants were cultivated under polyhouse conditions at the ICAR-Indian Institute of Horticultural Research, Bengaluru, Karnataka, India (13.13°N, 77.49°E, elevation: 890 m above sea level) during 2022 and 2023. Pollen grains were collected at peak anthesis between 9:30 AM and 11:00 AM. Pollen extraction was performed by gently tapping the flowers with forceps, paint-brushes or needles. In some cases, handheld vacuum cleaners were used for more efficient pollen collection.
Pollen Parents of different genotypes of marigold A) IIHRM0 12, B) IIHRMO 53, C) IIHRMY 2-1 and D) IIHRMY 1-4.
Pollen morphology
Fresh pollen grains were collected from fully open flowers of marigold genotypes by gently tapping them with a soft paintbrush. The pollen was transferred to butter paper envelopes and desiccated using silica granules for one hour to eliminate surface moisture. For SEM analysis, the desiccated pollen grains were mounted on double-sided copper tape affixed to aluminium stubs. The samples were examined under a Hitachi S3000N field emission SEM, and high-resolution images were captured for morphological evaluation. Morphometric parameters such as pollen size and shape were quantified using Fiji and ImageJ software (Schneider et al., Reference Schneider, Rasband and Eliceiri2012).
Pollen viability
To assess pollen viability, fresh pollen grains from each genotype were collected and transferred to individual glass petri dishes. A portion of the pollen was mounted on clean glass slides and stained using two contrasting staining techniques, Alexander’s stain and 1% acetocarmine solution. Each slide was covered with a coverslip and observed under a ZEISS Axio Cam ICc 5 light microscope at 20 × magnification. The experimental design included four marigold genotypes (IIHRMO 12, IIHRMO 53, IIHRMY 2-1 and IIHRMY 1-4) as Factor A and the two staining methods as Factor B. Each genotype-stain combination was replicated five times, and for each replication, 10 microscopic fields were analysed. Viable pollen grains were identified based on colour: blue for Alexander’s stain (non-viable grains appeared green) and red for acetocarmine (non-viable grains remained colourless). Pollen viability was expressed as a percentage, calculated using the formula:
\begin{align*}
\text{Pollen viability}\ (\%) & = [\text{Number of viable pollen grains/(Number}\\ & \ \quad \text{of viable {+} non-viable pollen grains})]\,{\times}\,100
\end{align*}In vitro pollen germination
To standardize in vitro pollen germination conditions, different combinations of sucrose and PEG concentrations were tested in Brewbaker and Kwack medium (Brewbaker and Kwack, 1963), which included boric acid, calcium nitrate, magnesium sulfate and potassium nitrate. In the initial phase, both desiccated and fresh pollen were cultured in media containing equal concentrations of sucrose and PEG, ranging from 5% to 10% (i.e., 5% sucrose + 5% PEG, 6% sucrose + 6% PEG and up to 10% sucrose + 10% PEG).
Subsequently, the role of PEG was examined by holding the sucrose concentration constant at 10% and varying PEG levels from 10% to 30% (i.e., 10% sucrose + 10% PEG, 10% sucrose + 12% PEG and up to 10% sucrose + 30% PEG). A third set of treatments involved fixing PEG concentration at 15% while varying sucrose from 10% to 30% (i.e., 10% sucrose + 15% PEG, 15% sucrose + 15% PEG and up to 30% sucrose + 15% PEG). These combinations were evaluated to identify the optimal medium that supports maximum pollen germination and pollen tube elongation under in vitro conditions.
Statistical analyses
The experiments (pollen morphology and pollen germination) were conducted using a completely randomized design (CRD) with five replicates per treatment. For pollen viability, experiments were laid out in CRD (factorial) with the eight treatments and five replication, and the obtained results were analysed by two-way analysis of variance. Data from 10 microscopic fields per replicate were analysed. Mean differences were considered statistically significant at p ≤ 0.05. Variations among treatments were assessed using the least significant difference test and Pearson correlation was done using R software. All statistical analyses were performed using SAS software (version 9.3).
Results
Pollen morphology
Significant differences were observed among the genotypes for various pollen traits (Table 1, Fig. 2). The polar and equatorial axes, which determine pollen grain size, indicated that all marigold genotypes belonged to the medium-sized category (25–50 µm). Genotype IIHRMY 2-1 exhibited the largest pollen grains (polar diameter = 31.40 ± 0.98 µm; equatorial diameter = 36.32 ± 0.65 µm), followed by IIHRMO 53 (polar = 29.81 ± 0.66 µm; equatorial = 33.54 ± 0.65 µm) and IIHRMY 1-4 (polar = 29.09 ± 0.60 µm; equatorial = 33.28 ± 1.08 µm). The smallest pollen size was observed in IIHRMO 12 (polar = 27.60 ± 0.84 µm; equatorial = 31.85 ± 0.81 µm). Based on the P/E ratio, IIHRMY 2-1 (0.87 ± 0.04) and IIHRMO 12 (0.84 ± 0.05) were categorized as suboblate, while IIHRMO 53 (0.88 ± 0.04) and IIHRMY 1-4 (0.88 ± 0.04) were classified as oblate-spheroidal. The highest pollen perimeter was recorded in IIHRMY 2-1 (108.32 ± 1.79 µm), followed by IIHRMO 53 (103.77 ± 2.39 µm), with the lowest in IIHRMO 12 (98.76 ± 1.35 µm). All genotypes exhibited tricolporate pollen grains, characterized by three colpi and three endoaperture. Aperture lengths ranged from 3.88 to 4.90 µm, with IIHRMY 2-1 showing the longest apertures (4.90 ± 0.18 µm) and IIHRMO 12 the shortest (4.06 ± 0.24 µm). IIHRMY 2-1 also recorded the maximum colpi length (31.58 ± 1.21 µm) and width (6.72 ± 0.26 µm). Spine morphology revealed pyramid-shaped spines with pointed or blunt apices. IIHRMY 1-4 had the longest spines (4.77 ± 0.34 µm), while IIHRMO 12 had the shortest (2.43 ± 0.15 µm).
Pollen structure images of different genotypes A) IIHRMO 12, B) IIHRMO 53, C) IIHRMY 2-1 and D) IIHRMY 1-4 by using SEM.
Pollen morphological characteristics of four marigold genotypes observed under SEM
Note: Data represent the mean ± standard error of the mean (SEM) for each genotype. Parameters measured include polar axis length, equatorial diameter, P/E ratio, perimeter, aperture size, spine length and dimensions of colpi (length and width). Superscript letters such as a, b and c indicate statistically significant differences among genotypes (p ≤ 0.05) based on CD (critical difference) at 5% level.
Pollen viability
Pollen viability was assessed using Alexander’s stain and acetocarmine (Table 2, Fig. 3). Among the genotypes, IIHRMY 1-4 recorded the highest viability (96.54%), followed closely by IIHRMY 2-1 (96.06%). The lowest viability was observed in IIHRMO 12 (91.04%). Between the two staining methods, acetocarmine produced a slightly higher average viability (94.44%) compared to Alexander’s stain (93.52%). The interaction between genotype and staining method had a significant impact: the highest viability was seen in IIHRMY 1-4 with acetocarmine (96.66%), while the lowest was in IIHRMO 12 with Alexander’s stain (90.02%).
Pollen Viability of different genotypes A & E) IIHRMO 12, B & F) IIHRMO 53, C&G) IIHRMY 2-1 and D& H) IIHRMY 1-4 by using Aceto carmine and Alexander stain. Where, red arrow indicates viable pollen grains and green arrow indicating non-viable pollen grains.
Pollen viability (%) of four marigold genotypes assessed using two staining methods (Alexander’s stain and acetocarmine)
Note: Factor A represents the different genotypes, while Factor B represents the staining solutions used to assess viability. Values are expressed as percentages of viable pollen observed. The mean values followed by different letters a, b and c indicate significant differences at p ≤ 0.05 as determined by the CD test. Statistical comparisons were made for genotype effect (A), staining method effect (B), and their interaction (A × B).
Pollen germination media standardization
Desiccated and fresh pollen with sucrose and PEG (5–10%)
Pollen germination was evaluated using fresh and desiccated pollen under different combinations of sucrose (5–10%) and PEG (5–10%). The highest germination rate (22.65 ± 0.74%) was achieved with 10% sucrose + 10% PEG using desiccated pollen. The lowest germination (9.14 ± 0.34%) was observed with 5% sucrose + 5% PEG using fresh pollen (Fig. 4A).
In vitro pollen germination media standardization using different concentrations of Sucrose and Polyethylene Glycol (PEG) in marigold pollen grains. Where, A) Pollen grains with sucrose (5 to 10 %) and PEG (5 to 10 %), B) Sucrose (10%) with varying PEG concentrations ranging 10 to 15 % and C) PEG (15 %) with varying sucrose concentrations ranging 10 to 30 %.
Sucrose standardization with varying PEG (10–30%)
Using 10% sucrose with varying PEG concentrations, the combination of 10% sucrose + 15% PEG resulted in the highest germination (38.56 ± 0.77%), followed by 10% sucrose + 20% PEG (32.94 ± 0.91%). The lowest germination was recorded with 10% sucrose + 10% PEG (20.05 ± 0.95%) (Fig. 4B).
PEG standardization with varying sucrose (10–30%)
Using a fixed 15% PEG with varying sucrose concentrations, the highest germination was recorded with 15% sucrose + 15% PEG (37.12 ± 0.51%), followed by 10% sucrose + 15% PEG (32.82 ± 1.55%). The lowest germination occurred with 25% sucrose + 15% PEG (24.00 ± 0.51%). The combination 15% sucrose + 15% PEG was identified as the optimized germination medium and used for subsequent experiments (Fig. 4C).
In vitro pollen germination
Significant differences in pollen germination rates were observed among the genotypes. The highest germination was recorded in IIHRMY 1-4 (54.09 ± 1.15%), followed by IIHRMY 2-1 (47.05 ± 1.27%). The lowest rates were observed in IIHRMO 12 (32.24 ± 0.35%) and IIHRMO 53 (Table 3, Fig. 5).
In vitro pollen germination and pollen tube growth of marigold pollen genotypes A) IIHRMO12, B) IIHRMO 53, C) IIHRMY 2-1 and D) IIHRMY 1-4. Where, Pollen germination was carried out in standardized germination medium containing 15 % sucrose and 15 % polyethylene glycol (PEG).
Pollen germination percentage and pollen tube length (µm) of four marigold genotypes in standardized germination medium containing 15% sucrose and 15% PEG
Note: Values are presented as mean ± SEM. Superscript letters a, b and c denote statistically significant differences among genotypes at p ≤ 0.05 based on the CD test. The standardized medium was optimized for assessing pollen performance under in vitro conditions.
Pollen tube length (µm)
Pollen tube length also varied significantly among genotypes. IIHRMY 1-4 had the longest tubes (55.74 ± 3.28 µm), followed by IIHRMY 2-1 (53.74 ± 3.83 µm). The shortest tubes were recorded in IIHRMO 53 (38.21 ± 2.48 µm) (Table 3).
Pearson correlation among pollen traits
Pearson correlation coefficients (r) were calculated to assess the relationships between pollen morphological traits, viability and germination (Supplementary Fig. 1). Pollen germination showed a highly significant positive correlation with pollen viability (r = 0.978), indicating that higher viability corresponds with better germination. Additionally, pollen germination was positively correlated with aperture size (r = 0.911), colpi length (r = 0.938) and colpi width (r = 0.833). Pollen viability was also significantly correlated with colpi length (r = 0.989), colpi width (r = 0.913) and aperture size (r = 0.943). Furthermore, aperture size was positively correlated with both colpi length (r = 0.928) and colpi width (r = 0.972).
Discussion
This study provides a comprehensive assessment of pollen morphology, viability and in vitro germination across four genotypes of T. erecta (African marigold), integrating SEM-based morphometric characterization with physiological evaluations using optimized culture media. The findings reveal significant genotypic variation in pollen traits, establishing a clear link between pollen structure and functionality. Notably, this is among the few studies to combine detailed SEM imaging with a dual-stain viability approach and PEG sucrose-based germination media standardization in marigold filling a critical knowledge gap in the reproductive biology of this economically important ornamental crop.
The pollen grains of all four genotypes were found to be medium-sized (25–50 µm), suboblate to oblate-spheroidal in shape, and tricolporate, bearing three colpi and endoaperture features as characteristics of the Asteraceae family (Khan et al., Reference Khan, Ahmad, Zafar and Akhter2012; Salamah et al., Reference Salamah, Luthfikasari and Dwiranti2018; Sánchez-Chávez et al., Reference Sánchez-Chávez, Vovides and Sosa2022). While such features are consistent with earlier reports in related genera such as Dahlia and Hidalgoa (Sánchez-Chávez et al., Reference Sánchez-Chávez, Vovides and Sosa2022), the present study provides finer resolution and morphometric precision via SEM, highlighting subtle genotypic differences, especially in spine morphology and aperture dimensions. These traits, though often overlooked, are critical for taxonomic resolution and may influence pollen–stigma compatibility and adhesion efficiency. The genotype IIHRMY 2-1 exhibited the largest polar and equatorial dimensions, suggesting enhanced pollen dispersal and potentially better interaction with stigmatic surfaces traits desirable in a pollen donor for hybrid seed production.
Pollen viability and in vitro germination tests revealed variability among genotypes. Pollen viability is a crucial parameter for successful pollination and seed set, as reported by Koshy and Jee (Reference Koshy and Jee2001) in Bambusa vulgaris. High pollen viability is desirable in male parents, as it correlates with successful fertilization and seed formation in the female parents (Meral et al., Reference Meral, Kırbay, Haspolat and Kazaz2024). Pollen viability varied significantly among the genotypes, ranging from 90.02% to 96.66%, depending on the staining method used. IIHRMY 1-4 exhibited the highest viability using both acetocarmine (96.66%) and Alexander’s stain (96.42%). This high viability is a strong indicator of successful pollen performance, essential for efficient fertilization and seed set, especially under field conditions where environmental stresses often compromise reproductive success. The dual-stain comparison adds methodological robustness, as it confirms consistent results across histochemical approaches. Interestingly, the yellow-pollen genotypes (IIHRMY 1-4 and IIHRMY 2-1) performed better than the orange-pollen lines (IIHRMO 12 and IIHRMO 53), suggesting a possible link between pollen pigmentation and physiological viability, warranting further biochemical investigation.
Earlier studies, such as Thaneshwari et al. (Reference Thaneshwari, Prakash, Kumari, Thakur and Sharma2019), reported viability rates nearing 99% in the cultivar Pusa Narangi Gainda, while Sunny et al. (Reference Sunny, Kumar, Rajasekharan and Dhananjaya2024) documented broader variability (49.55–95.02%) in chrysanthemum. The current findings align well with these results but are distinctive in offering genotype-specific insights that are directly applicable to ongoing hybridization efforts in Indian public-sector breeding programmes. Moreover, while previous studies often relied on a single stain, the current dual-method approach improves reliability and reproducibility.
Standardization of in vitro pollen germination media represents another major contribution of this work. By testing a gradient of sucrose and PEG combinations, the study identified 15% sucrose + 15% PEG as the optimal medium for germination, a finding that is consistent with previous reports in marigold (Yong-ping et al., Reference Yong-ping, Ya, Xiao-ling, Yong-kang, Xiu-hua and Mei-xiu2010) and Heptacodium miconioides (Qi et al., Reference Qi, Fang, Peng and Gen-di2006). The addition of PEG plays a key role in modulating osmotic potential, thereby mimicking the stigmatic environment and enabling accurate prediction of pollen performance. Importantly, higher sucrose concentrations were found to inhibit germination, likely due to osmotic stress, while PEG helped to stabilize the medium by reducing water availability to non-viable pollen thereby increasing selectivity for functionally competent pollen grains (Sunny et al., Reference Sunny, Kumar, Rajasekharan and Dhananjaya2024).
The germination percentages ranged from 32.24% (IIHRMO 12) to 54.09% (IIHRMY 1-4), with pollen tube lengths closely paralleling these trends. The highest pollen tube elongation was also observed in IIHRMY 1-4 (55.74 µm), followed by IIHRMY 2-1 (53.74 µm), reinforcing the idea that tube length is a reliable proxy for pollen vitality and competitive fertilization potential. Jayaprakash et al. (Reference Jayaprakash, Sheeba, Vikas, Sivasamy and Sabesan2018) observed similar positive correlations between pollen viability, germination and tube growth in eggplant. These results underscore the importance of selecting male parents with consistently high pollen tube growth rates to ensure timely and effective fertilization in hybrid seed production programmes.
Compared to previous studies that primarily focused on single aspects, either morphology, viability or germination, this study is novel in its integrated approach. It offers a full-spectrum characterization of pollen traits in T. erecta, facilitating both basic understanding and applied breeding. The clear identification of superior pollen donors (notably IIHRMY 1-4 and IIHRMY 2-1) provides actionable information for breeders aiming to improve F₁ hybrid seed production efficiency.
Future directions
The insights gained from this study open several avenues for future research. Biochemical profiling of pollen, especially in relation to pigment composition and antioxidant activity, may reveal deeper links between structure and function. Molecular characterization of pollen-specific gene expression could further enhance our understanding of the determinants of pollen viability and tube growth. Additionally, integrating these findings into cryopreservation or long-term pollen storage strategies could significantly enhance year-round hybrid seed production. Ultimately, combining morphological and physiological screening with molecular markers will offer a robust pipeline for selecting elite pollen parents in marigold and other cross-pollinated ornamentals.
Conclusion
This study provides a comprehensive characterization of pollen traits in four genotypes of T. erecta, integrating morphological, viability and germination assessments to inform hybrid breeding strategies. Through detailed SEM analysis, we confirmed that all genotypes produce medium-sized, tricolporate pollen with suboblate to oblate-spheroidal shapes and distinctive exine ornamentation – traits that reinforce their taxonomic placement within Asteraceae and enable precise genotype differentiation. Importantly, the genotypes IIHRMY 1-4 and IIHRMY 2-1 consistently outperformed others in terms of pollen viability and in vitro germination, with IIHRMY 1-4 achieving the highest viability (96.66%) and germination rate (54.09%) in a medium optimized with 15% sucrose and 15% PEG. This media standardization is a significant advancement, offering a reliable protocol for assessing functional pollen performance in marigold and potentially other ornamental species. The identification of high-performing pollen donors provides actionable information for enhancing the efficiency of hybrid seed production. Moreover, the dual-staining approach validated here strengthens current practices in viability testing by offering reproducible and comparative insights. Together, these findings bridge critical gaps in reproductive biology and offer a robust framework for future marigold breeding programmes, germplasm conservation and precision pollination management.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1479262125100233.
Acknowledgements
The authors are thankful to the ICAR-Indian Agricultural Research Institute, New Delhi and ICAR-Indian Institute of Horticultural Research, Bengaluru, India, for providing the necessary research facilities.
Author contributions
P.V.: Investigation, Formal Analysis, Data Curation, Writing Original Draft and Editing; L.D: Editing; P.E.R.: Supervision, Resources, Reviewing and Editing; D.S.: Resources and Methodology; T.P.: Conceptualization, Supervision, Visualization, Reviewing and Editing.
Data availability statement
All the data generated and analysed during this study are included in this manuscript.
Competing interests
The authors declare that they have no conflict of interest.
