INTRODUCTION Tuberose (Agave amica Medik.) (Syn. Polianthes tuberosa L.) is one of the commercially important bulbous flower crop which belongs to the family Asparagaceae (Chase et al., 2016). It is native to Mexico (Bailey, 1919) and is grown wildly in the tropical and subtropical regions of India (Benschop, 1993). It is commonly called as Rajanigandha or Nishigandha because of its night blooming habitat. It is cultivated in major states of India like Tamil Nadu, Karnataka, West Bengal, Maharashtra, Haryana, Punjab, Andhra Pradesh, Telangana, Uttar Pradesh and Delhi In India, total production of tuberose is 1,97,210 tonnes (2015-2016) (NHB). In Tamil Nadu, the total area under tuberose cultivation during 2019-20 is 7,654 ha and its production is 66,355 tonnes (Salient Statistics on Agriculture, Agriculture Department, Govt. of Tamil Nadu, 2021). Tuberose is mainly valued for its ornamental beauty, sweet fragrant spikes with excellent keeping quality (Sadhu and Bose, 1973). Flowers are used in bouquets, garland, floral arrangements, floral ornaments and buttonholes. Tuberose is classified in to three types based on the number of tepal whorls viz single (one whorl), semi double (2-3 whorl) and double (more than 3 whorl). Variegated type also available which is used in landscape gardening. Essential oil of tuberose occupies a special place in the international market. Single type is preferred for the concrete extraction over double type because of its high recovery percentage of 0.08- 0.11% by solvent extraction (Singh, 1995) and double type is recommended for cut flower purposes. There is limited variety available in tuberose due to the constraints like self incompatibility, limited genetic resources, dichogamy and poor seed setting (Hemanta et al., 2016). In tuberose, developing new variety with improved quantitative and qualitative character is essential. Before conducting any breeding work thorough knowledge on the plant characteristics is a prerequisite. In tuberose seed setting is quiet erratic in single type, not observed in double type and high percentage of seed setting is found in variegated types (Datta, 2017). Joshi and Pantulu (1940) reported that the reason for sterility in tuberose is not known but it is not due to any defects or deformation in pollen grains and embryo sac development. Hence, it is possible to create genetic variability in tuberose by allowing genes to express in different combinations. This can be achieved by selecting seedlings from the open pollinated or cross pollinated types (Hemanta et al., 2016). With these basic strategies the open pollination study was conducted in selected tuberose genotypes. MATERIALS AND METHODS This study was conducted in the Department of Floriculture and Landscape Architecture, Tamil Nadu agricultural university, Coimbatore during the year 2019-2020. Seven single type tuberose genotypes were selected for this study viz, Arka Nirenthara, Variegated Single, GK-T-C-4, Bidhan Rajani-1 (Shnigdha), Bidhan Rajani-2 (Ujwal), Bidhan Rajani-3 (Joythi) and Arka Prajwal (Table 1). Fully developed flower buds in the selected genotypes were allowed for open pollination. Only one spike per plant was used, remaining spikes were bagged to avoid contamination. This experiment was laid out in randomized block design (Panse and Sukhatme 1961). After two weeks, observations like days taken for maturity, number of pods per spike, fruit set percentage and fruit number of seeds per capsule (from fully matured) were recorded. From the matured pods, the seeds were collected and seed germination study was carried out. Observation viz, germination percentage, days taken for seed germination (germination complete once the radicle protruded about 1cm in length), seedling height, vigour index (Germination percentage × (root + shoot length) by Abdul Baki and Anderson, 1973), root and shoot growth was recorded. This experiment was laid out in completely randomized design (Panse and Sukhatme, 1961). Statistical analysis. Observations recorded were subjected to analysis of variances (ANOVA) using SPSS 20.2 statistical software. The significant difference between the genotypes was determined by least significant difference (LSD) at 5% probability. RESULT AND DISCUSSION Open pollination in different genotypes. Initial number of florets per spike of each genotype was recorded before it was allowed for open pollination. It is observed from the Table 2 that maximum number of florets per spike was recorded in BR-I (32.67) followed by Arka Prajwal (31.83) and minimum number of florets was observed in GKTC-4 (27.53).After two weeks of pollination, the fruits starts to develop. Fruit formation was recorded in genotypes such as Arka Nirantara, Variegated Single, GK-T-C-4, Bidhan Rajani-2, Bidhan Rajani-3 and not in Arka Prajwal and Bidhan Rajani-1. Fruit set was not observed in all the genotypes. In some genotypes there was no fruit development. Poor fruit set and failure of fruit formation maybe due to incompatibility, lack of pollen germination and pollen tube growth. Hemanta et al. (2016) reported that in open pollination, the low frequency of fruit set and capsule formation may be attributed to the less presence of insect vectors and pollinators. Whereas Bharathi et al. (2021) also observed degeneration of embryo sac after zygote formation in genotype Arka Prajwal due to incompatibility. Days taken for maturity was calculated from the day of pollination to the fruit development stage (dried pods). Days taken for fruit maturity were significantly different between the genotypes. Minimum number of days taken for maturity was observed in Bidhan Rajani-2 (58.66) followed by Arka Nirantara (59.34) and maximum days was observed in Variegated Single (64.87). Bidhan Rajani-2 had higher number of fruit per spike of 7.37 and lower in Bidhan Rajani-3 of 2.12. Percentage fruit set was observed higher in Bidhan Rajani-2 (24.39%) followed by Arka Nirantara (20.93%) and Bidhan Rajani- 3 with a minimum of 6.66% of fruit set. Number of seeds per capsule significantly varies between the genotypes. Maximum number of seeds per capsule was recorded in Bidhan Rajani-2 (26.77) followed by Arka Nirantara (20.43) and minimum seed number was in Bidhan Rajani-3 (15.82). Variation in seed set by different genotype might be due to genetic and environmental factors. Ranchana et al. (2014) observed that under open pollination, the highest percentage of fruit set was reported in Variegated Single (89.00%) and similar result was also obtained by Seetharamu et al. (2000). In Arka Prajwal the pod abscission observed after crossing was due to post zygotic barrier (Bharathi et al., 2021). Similarly, Hemanta et al. (2016) also studied open pollination in tuberose and recoded maximum mean number of fruit set (7.33) and percentage of fruit set in a spike (21.77%) in Arka Nirantara. There was no fruit set in Arka Prajwal, this was also observed by Ranchana and Kannan (2016). Germination and seedling growth of open pollinated seeds. Matured pods from the genotypes were collected then seeds germination and seedling development were observed (Table 3). There was significant difference among the genotypes in the seed germination characteristics. Minimum number of days taken for seed germination was recorded in Bidhan Rajani-2 (22.13) which is on par with Arka Nirantara (22.56) and maximum number of days was in Bidhan Rajani- 3 (32.48). Germination percentage was observed higher in Bidhan Rajani-2 (76.89%) followed by Arka Nirantara (75.45%) and lower germination percentage in Bidhan Rajani- 3 (20.11). Variation in the days taken for germination and germination percentage may be due to the genetic makeup of the genotype and environmental conditions. Similarly, Raja et al. (2003) also observed a maximum of 66% germination in tuberose seeds when fresh seeds were sown and Hemanta (2015) recorded 26% germination. Seedling growth parameters were recorded after one month of sowing the seeds. Number of roots per seedling was on par between the genotypes. Maximum number was observed in Arka Nirantara (2.13) and minimum number in Bidhan Rajani- 3 (1.07). Number of leaves per seedling was also recorded and mostly in all the genotypes two leaves develops. Higher root and shoot length was recorded in Arka Nirantara (11.56, 15.34 cm respectively), lower root length in Variegated Single (9.44 cm) and lower shoot length in GKTC-4 (10.31). Ranchana et al (2014) also studied the germination of open pollinated tuberose seeds with different seed treatments and obtained root length in the range of 7.46-12.80 cm with shoot length range of 7.21-14.40 cm. Based on the germination percentage and seedling growth the vigour index of the genotype was calculated. Maximum vigour index was obtained in Arka Nirantara (1426.01) followed by Bidhan Rajani- 2 (1380.18) and minimum number in Bidhan Rajani- 3 (256.40) (Fig. 1). Similarly, variation in vigour index of China aster genotype was observed by Pandey et al. (2017).