INTRODUCTION Seabuckthorn (Genus Hippophae) is a berry-bearing, hardy shrub of the family Elaeagnaceae, naturally distributed in Asia and Europe and also introduced in North and South America. It includes 4 species (Hippophae rhamnoides, Hippophae salicifolia, Hippophae tibetana and Hippophae neurocarpa) and further 9 subspecies of Hippophae rhamnoides are reported so far from many parts of world out of which three species of seabuckthorn (Hippophae rhamnoides L. subsp. turkestanica Rousi, Hippophae salicifolia D. Don and Hippophae tibetana Sch.) are found in Spiti, Himachal Pradesh (Singh et al., 2012). It is a unique and valuable genotype resource currently cultivated in various parts of the world. The natural habitat of Seabuckthorn extends widely in China, Mongolia, Russia, and most parts of North Europe. It can withstand extreme temperatures from -43ºC to 40ºC (Lu 1992) and is considered to be drought resistant. The cold deserts in Himachal Pradesh are found in the districts of Lahaul and Spiti, parts of Kinnaur and Pir Panjal region of Chamba. These areas are characterized by high ridges, difficult terrains with ice field, perpetual snow covered peaks and hostile climate. Among various indigenous and under exploited genotype resources of high mountain area, Seabuckthorn (Hippophae salicifolia D. Don) is one of the best solution and can certainly metamorphose the ecology of cold desert by reclaiming these bare fragile mountains. Willow leaved Seabuckthorn and indigenous source locally Sutz/Sarla offers an opportunity to maintain more sustainable livelihood qualities as well as unique option for the simultaneous management of several problems such as capability to grow and survive under adverse climatic conditions, extensive root system with soil binding ability/ soil stabilization/control of river bank/ water retention, This Frankia association accounts for atmospheric nitrogen fixation, hence adding to the soil-fertility (Lu 1992; Singh, 2001), nitrogen fixing upto 60-180kg/ha/year (Mathew et al., 2007), higher vitamin-C content and economic value of fruit and seed oil, excellent fodder and fuel wood qualities (Roomi et al., 2015), wider application in food, cosmetics, therapeutic, pharmaceutical, and cosmetic properties and other pharmaceutical products (Kaushal et al., 2013), excellent fencing hedge and social fencing. Though seabuckthorn is widely found under agroforestry system as well as hazard zones, yet no any systematic study has been carried out so far to understand its potential under agroforestry/forestry perspectives. So it help to be a valuable tool for land restoration and conservation in the cold desert of the Lahaul valley (Sankhyan et al., 2018). The first pre-requisite step to undertake breeding programme and to obtain improved genetic gain is selection of best population and best individuals within the population. Knowledge of the distribution of genetic diversity provides a guide to the proper management of the genetic resources of species in effective genetic conservation programs (Barrett and Kohn, 1991). Hence present study was undertaken to study variation in quantitative characteristics among and between different populations to select plus trees of Hippophae salicifolia D. Don. MATERIAL AND METHODS The present study was carried out in the fields of Baspa valley of District Kinnaur and Spiti valley of District Lahaul and Spiti as well as in the laboratories of the Department of Tree Improvement and Genetic Resources of Dr. Y. S. Parmar University of Horticulture and Forestry, Nauni, Solan, Himachal Pradesh, India during the period 2018- 2020.After surveyed and proper identification of genotypes and population with selecting three natural populations in each valley and from each population five female genotypes were selected at the time of fruit set, i.e., during August-September, marked and taken for further investigation and recording quantitative characteristics. Leaves and berries were taken and carried from experimental sites to laboratory for further study. Altitude of populations ranged from 2590m amsl to 3538m amsl in the study area at different sites of Himachal Pradesh. Observations on the following morphological characteristics were recorded. Genotype height (m), Branch length (cm), Leaf length (cm), Leaf width (cm), Leaf area (cm²), Leaf density, Number of thorns (average per branch), Fresh fruits weight (g -100 Fruits), Fruit length (mm), Fruit diameter (mm), Seeds weight (g -1000 Seeds), Seed length (mm), Seed width (mm). ANOVA (Analysis of Variance) was carried out for quantitative characteristics as described by Panse and Sukhatme (1967); Chandel (1984). Design RBD(Factorial) with 6 treatments, 5 genotypes, 3 replication was used. Genotypic coefficient of variation (GCV) and phenotypic coefficient of variation (PCV) were estimated as per suggested by Burton and Devane(1953), heritability was estimated as per suggested by Burton and Devane (1953); Johnson et al. (1955), genetic advance as per procedure of Allard (1960) and genetic gain as per procedure of Johnson et al. (1955). RESULT AND DISCUSSION A. Quantitative Characters Genotype height (m). It is evident from data presented in Table 1 that maximum genotype height was recorded in Chitkul (3.34 m) followed by Kupa (2.74 m), while minimum genotype height was observed in Shiego (1.20 m) followed by Giu (1.60 m). Between the genotype of different sites the maximum (4.3 m) genotype height was obtained from genotype number 5 of Chitkul (CG5), whereas the minimum genotype height (0.9 m) was noticed in Sheigo genotype 1 (SG1), respectively. The interaction between the population and among population was statistically significant. Similar genotype height up to 6 m was recorded by Yadav et al. (2006). Branch length (cm). The maximum branch length was recorded in Kupa (122.67 cm) followed by Chitkul (111 cm) whereas minimum was observed in Shiego (58.27 cm) (Table 2). The highest value (180.00 cm) for branch length was recorded in genotype number 4 in Kupa (KG4), whereas minimum (46.67 cm) was observed in genotype 3 in Giu (GG3). The interaction among and between population was statistically significant. Leaf length (cm). The mean leaf length ranged from 7.07 to 3.92 cm presented in Table 3. Maximum value was recorded for Kupa (7.07 cm) and the minimum value was noticed in Shiego (3.92 cm). From the studied genotypes, the maximum leaf length (8.10 cm) was observed in genotype 4 for Kupa (KG4), while the minimum (3.34 cm) was recorded in Sheigo genotype 3 (SG3). The interaction among and between population was observed to be significant whereas it was non-significant between population .On the other hand, the variation among population was also shown to be significant. Similar leaf length was recorded by Yadav et al. (2016) which was ranged from 8.46-4.83 cm. Leaf width (cm). Content of the data presented in Table 4 showed that the variation among the population as well as the interaction between among and between population showed significant variation on leaf width. Maximum leaf width was recorded in Kupa (1.01 cm) followed by Chitkul (0.88 cm) and minimum was recorded in Shiego (0.48 cm) followed by Giu (0.55 cm). However, among population, the maximum leaf width (1.34 cm) was recorded in genotype 3 for Kupa (KG3) whereas, minimum was recorded in genotype 3 (0.37 cm) for Sheigo (SG3). The variation among population as well as between population was also found to be significant. Similar leaf width was recorded by Yadav et al. (2016) which was ranged from 1.4-037 cm. Leaf area (cm²). The mean data in Table 5 revealed that the maximum leaf area was recorded in Chitkul (4.00 cm²) followed by Kupa (3.83 cm²) and minimum was recorded in Shiego (2.21 cm²) followed by Giu (2.32 cm²). A significant variation was found between population and the interaction between population and among population was also found to be significant. Between population, the maximum (7.88 cm2) leaf area was recorded in Chitkul genotype 1 (CG1) where minimum (1.79 cm²) in genotype 3 for Sheigo (SG3). Yadav et al. (2006) also reported similar leaf area ranged from 2.28 to 9.35 cm2. Leaf density (10 cm of branch). The data presented in Table 6 revealed that the maximum leaf density was recorded in Chitkul (24.20 per 10 cm) and minimum was observed in Shiego (14.67 per 10 cm). Between population, the highest (27.00 per 10 cm) leaf density per 10 cm was recorded in Chitkul genotype 5 (CG5), whereas the minimum was in Kupa genotype 5 (13.00 per 10 cm). The interaction among population and between population was found to be significant but it was non-significant between the population. Similar leaf density was recorded by Yadav et al. (2006) that ranged from. 29.83 leaves per 10 cm. Number of thorn (per 10 cm of shoot). The data on number of thorns is detailed in Table 7. The maximum number of thorn was recorded in (4.53 per 10 cm) and the minimum was recorded in Kupa (1.47 per 10 cm), respectively. Between the population of all the sites, the maximum number of thorns per 10 cm was recorded in genotype 5 (5.33 per 10 cm) of Giu (GG5) and minimum number was recorded in Mane genotype 5(MG5) (0.67 per 10 cm). The interaction was also significant between population and among populations. Similar number of thorn was recorded by Singh V (2006) that ranged from 4.3-6.9 per 10 cm. Fresh fruit weight (g -100 fruits). The fruit weight showed significant variation among population, which varied from 17.09 to 11.14 g in Table 8. The maximum fruit weight was noticed in Sheigo (17.09 g) followed by Mane (16.45 g) and minimum was recorded in Badseri (11.14 g) which was differed statistically. Between the population, the maximum fruit weight (20.11 g) was recorded in Mane genotype 4 (MG4) and minimum was in (MG1) Mane genotype 1 (8.84 g). The variation between population was noticed to be significant and the interaction was also statistically significant for among population and between population. Nawaz et al. (2018) also found the maximum and minimum fruit weight value of 20 berries ranged from 6.28 to 1.08 g. Fruit length (mm). The data pertaining to the fruit length is presented in Table 9, which showed significant variations among studied population. Maximum fruit length (7.00 mm) was observed in Kupa which was statistically different from other population and minimum was recorded in Giu (5.73 mm) which was at par with Sheigo (5.88 mm) respectively. Between population, the maximum fruit length (7.23 mm) was also noticed in Kupa genotype 2 (KG2) which was followed by (CP2), (CP5), (KP3), (CP3) whereas, minimum was recorded in Giu Genotype 5 (GG5) (5.17 mm). The interaction among population and between population was found significant but it was non-significant between population. Yadav et al. (2006) reported almost similar maximum and minimum range of fruit length varied from 5.78 to 7.92 mm. Fruit diameter (mm). The data pertaining to fruit diameter are presented in Table 10 which revealed a significant variation among all the population. Mane showed the highest (7.17 mm) fruit diameter which was statistically different from all other population whereas, Badseri reported the minimum (5.88 mm) fruit diameter which was at par with Shiego (5.96) and Chitkul (5.98), respectively. Between population, the maximum (7.63 mm) fruit diameter was noticed in Mane genotype 4 (MG4) and minimum (5.29 mm) was recorded in Badseri genotype 3 (BG3). All the population showed a significant variation between each other and the interaction among population and between population also found to be significant. Yadav et al. (2006) reported almost similar fruit diameter ranged from 5.51 to 7.24 mm. Seed weight (g -1000 seeds). The population taken for study showed considerable variation among themselves (Table 11). The highest value (21.30 g) for seed weight was recorded in Sheigo was statistically different from other whereas, the minimum seed weight was recorded in Badseri (11.30 g). Between all population, the maximum (24.00 g) seed weight was observed in Sheigo genotype1 and minimum (10.00 g) was recorded in Badseri genotype5 (BG5). The interaction between population and among population showed significant variation and the variation was also statistically significant for among population and between population. Similar seed weight was recorded by Tomar and Rattan (2012) which was ranged between 2.83-1.30 g. Seed length (mm). The genotypes taken for study showed considerable variation among themselves (Table 12). The highest value (5.83 mm) for seed length was recorded in Chitkul which was significantly different from other population, whereas the minimum seed length was recorded in Mane (4.17 mm). Between population, the maximum seed length (6.50 mm) was noticed in Chitkul genotype 2 whereas, minimum (3.50 mm) was recorded in Mane genotype1 (MG1). The interaction between population and among population as well as between population showed statistically significant variation. The present investigations are in line with the findings of Mir et al., (2018); for seed length ranged from 4-7 mm. Seed width (mm). The data pertaining to the seed width is presented in Table 13, which showed significant variations among studied population. Badseri showed the highest (3.16 mm) seed width which was statistically different from other population and Kupa reported the minimum (2.18 mm) seed width. Between population, the maximum seed width was noticed in Badseri genotype 4 (3.50 mm) and minimum (1.80 mm) was recorded in Kupa genotype 3. Between population showed non-significant variation among each other and the interaction between population and among population showed statistically significant. The present investigations are in line with the findings of Kaushal and Sharma (2012) for seed width varied from 2.5-3.5 mm. Variability Estimate and genetic parameters for quantitative characters. Noteworthy distinction was observed during the investigation of coefficient of variation, heritability, genetic advance and genetic gain for various parameters of between different population of Himachal Pradesh. The characters evaluated for variability and genetic parameters included quantitative (plant height, branch length, leaf characters, fruit characters and seed characters). The consequences found for variability and genetic parameters are accessible in the certifying tables and unfolding as follows: The total variance panel into different mechanisms as described in Table 14 showed that the highest genotypic and phenotypic coefficient of variability among quantitative traits was observed in number of thorns (45.3; 53.36) followed by leaf area (43.87; 46.21) and minimum value in fruit diameter (8.44; 9.69). Maximum heritability was shown by 100 seeds weight (99.98%) followed by fruit weight (99.97%), plant height (99.54%), leaf length (91.25%), leaf area (90.16%) and minimum value was shown by seed width (47.11 %). Genetic advance was maximum in branch length (38.70) and minimum in leaf width (0.47). Genetic gain was maximum for leaf area (85.82 %) and minimum was for fruit diameter (15.15 %). These results support the findings of Mohapatra (1996) in Acacia catechu, Manga and Sen (1998) in Prosopis cineraria, Gera et al. (2002) in Dalbergia sissoo, Khosla et al. (1982) in Santalum album, (Kaushal 1978; Khosla et al., 1980; Gupta 1993; Sehgal and Jaswal 1996; Rathore 1997; Bhat 2010) in Grewia optiva. This suggests the selection of the required traits, so as to exploit the potentialities and huge variation existing in the valuable species, which is of great agroforestry importance in cold desert of Spiti Valley and Baspa Valley of Kinnaur.