Caryologia. International Journal of Cytology, Cytosystematics and Cytogenetics 72(3): 23-34, 2019 Firenze University Press www.fupress.com/caryologiaCaryologia International Journal of Cytology, Cytosystematics and Cytogenetics ISSN 0008-7114 (print) | ISSN 2165-5391 (online) | DOI: 10.13128/caryologia-755 Citation: K. Gautam, R. Raina (2019) Floral architecture, breeding system, seed biology and chromosomal stud- ies in endangered Himalayan Angelica glauca Edgew. (Apiaceae). Caryologia 72(3): 23-34. doi: 10.13128/caryolo- gia-755 Published: December 13, 2019 Copyright: © 2019 K. Gautam, R. Raina. This is an open access, peer- reviewed article published by Firenze University Press (http://www.fupress. com/caryologia) and 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 Statement: All rel- evant data are within the paper and its Supporting Information files. Competing Interests: The Author(s) declare(s) no conflict of interest. Floral architecture, breeding system, seed biology and chromosomal studies in endangered Himalayan Angelica glauca Edgew. (Apiaceae) Kamini Gautam1,2,*, Ravinder Raina1,3 1 Dr. YSP University of Horticulture and Forestry, Solan, Himachal Pradesh, India 2 Grassland and Silvipasture Management Division, ICAR-Indian Grassland and Fodder Research Institute, Jhansi, Uttar Pradesh, India 3 Amity Food and Agriculture Foundation, Amity University, Noida, Uttar Pradesh, India *Correspondence author: kaminigautam1989@gmail.com Abstract. Endangered Angelica glauca an important medicinal plant of temperate Himalaya is valued for its roots which are used to treat several diseases besides food flavouring. Reproductive biology studies conducted in this species for the first time have revealed i). presence of umbels of different orders with only bisexual flowers ii). occurrence of sterile seeds (without embryo) apart from fertile ones iii). seed set in only early blooming umbels (primary and lateral-I) iv). 2n=22 chromosomes besides presence of chromosomes in a group at metaphase and anaphase-I and cytomixis in some pollen mother cells and v). extreme protoandry and cross pollination behavior (upto 95%) of the species. These observations have implications for developing any conservation plan for the species. Keywords. Endangered, umbel order, Apiaceae, cross pollination, low seed set, embryo less seeds. INTRODUCTION Apiaceae an angiosperm family consists of 300-455 genera and 3000- 3750 species worldwide (Pimenov and Leonov 2004) and many of these spe- cies are highly valued for being economically and medicinally important (Butola and Badola 2006; Sher et al. 2011). The genus Angelica is one of the very important genera belonging to family Apiaceae and is represented by about 110-115 species worldwide and almost 87 species in Asia (Pimenov and Leonov 2004). This genus is represented by mainly three species viz. A. glau- ca Edgew., A. archangelica L. and A. nubigena Clarke in Himalayan region. A. nubigena is poorly known species found in Sikkim (Pimenov and Kljuyk- ov 2003) and other species i.e. A. cyclocarpa (C.Norman) M. Hiroe and A. oreadum Diels have also been reported from Indian Himalaya, Pakistan and Afghanistan (Pimenov and Kljuykov 2003). 24 Kamini Gautam, Ravinder Raina Angelica glauca Edgew. (Family: Apiaceae; English name: Himalayan Angelica; Local name: chora, chokha- ra & gandrayan) is an endangered perennial temper- ate medicinal and aromatic herb distributed in moist and shady regions of Himalaya at an altitude of 2000- 3800m amsl (IUCN 1993; Samant et al. 1998; Chauhan 1999; Butola and Budola 2004; Samant et al.2009) in Afganistan, Pakistan and India (Jammu and Kashmir, Himachal Pradesh and Uttrakhand) (Bisht et al. 2003; Butola and Budola 2004; Saeed and Sabir 2008; Butola and Vashistha 2013).Valued for roots which are used to treat dismenhorrea, metorrhagia, amenhorrea, polycys- tic ovary syndrome, rheumatism, infantile atrophe (Bisht et al. 2003; Butola and Samant 2006; Butola and Budola 2008; Butola and Vashistha 2013; Goswami et al. 2012), also acts as stimulant, cholagogue, cardio-active, car- minative, sudorfic and expectorant (CSIR 1985; Singh and Rawat 2000; Bisht et al. 2003; Butola and Samant 2006). Besides this, roots also yield essential oil used to flavour liquor and food items (Nautiyal and Nautiyal 2004; Butola and Samant 2006). Due to remote distribu- tion of A. glauca in inaccessible areas of Himalaya cou- pled with small size of flower because of being an Api- aceae member, very less work has been carried out on its reproductive biology. However, breeding behavior and reproductive biology studies are crucial for understand- ing plant pollinator interaction, reproductive bottlenecks as well as for developing conservation plan. Therefore reproductive biology has been studied in details for the first time in this species under the present investigation. Material and methods Studies were carried out at Shillaru (2130m amsl, 30°45’00.48’’N, 76º59’12.22’’E; District Shimla, Himachal Pradesh, India); at Shilly (1550m amsl; 30°54’30’’N, 77° 07’30’’E; District Solan, Himachal Pradesh, India) and Medicinal plants laboratory of department of Forest Products, College of Forestry, Dr. Y. S. Parmar Univer- sity of Horticulture and Forestry, Nauni, Solan-173230 (Himachal Pradesh, India) during years 2013-2016. Pop- ulation Shillaru had almost 100 plants whereas in Shilly population 50 plants were present. The vegetative and floral studies were conducted as per standard literature (Lawrence 1951; Weberling 1989; Kaufman et al. 1989). Pollen-ovule ratio was studied as per Cruden (1977) and pollen viability was calculated on the basis of one percent acetocarmine staining test. Young floral buds, fixed in absolute alcohol, glacial acetic acid and chloroform in the ratio of 1:1:1(v:v:v) for 24 hours, washed and then stored in 70% alcohol at low temperature, were used for meiotic studies. One percent acetocarmine stain was used for chromosomal staining by usual squash method. For open pollination plants with unopened healthy umbels were tagged and left as such, whereas for assessing autogamy umbels were enclosed (Figure 2a) at pre flowering stage. Increase in ovary size and its transformation into fruit was taken as the basis of fruit set. Seed viability was tested by Topo- graphical Tetrazolium Test (TTZ) test and germination by petri plate method. Petri-plate germination test was performed at 23 ± 2°C in growth chamber and radicle protuberances were taken as sign of germination. Topographical Tetrazo- lium Test (TTZ) (0.1% pH 6.0) for 48 hours after extract- ing seeds was conducted by soaking seeds in water, then excised to expose embryo followed by immersing in TTZ solution (0.1% pH 6.0) under dark conditions for 48 hours. Darkly red stained embryos were taken as viable. Statistical analysis was made as per CRD factorial (seed germination and viability testing under laboratory conditions), RBD factorial (seed set % in field) as well as T-test (pollination studies). Statistical analysis was con- ducted as per Gomez and Gomez (1984). Ocular and stage micrometer (ERMA, Tokyo, Japan) were used for micro-measurements and microscopic examination was made using Olympus trinocular research microscope (Model - CH20iBIMF, New Delhi, India). RESULTS Morphology and floral architecture The qualitative and quantitative features are tabu- lated (Table 1). Plants of this species are erect perennial herbs, inflorescence is compound umbel with umbels of different order i.e. primary, lateral-I, lateral-II and lateral -III umbels based on whether they are borne on main stem or lateral branches (Figure1a, b). Percentage of plants with different umbel order varied in two stud- ied populations (Table 2) and quantitative features of umbels are presented in Table 3. All the morphological features of stem, roots, leaves, inflorescence, flower, fruit and seeds were similar to earlier reports except for the presence of variation in seed size and presence of seeds without embryo which are is being reported for the first time in A. glauca (Figure 2b, c). Phenology Sprouting starts in spring season (last week of April month onwards) and continues up to June month (first week). Floral buds start appearing during July first week 25Floral architecture, breeding system, seed biology and chromosomal studies in Angelica glauca to mid of August month. Primary umbel floral buds appear first followed by buds of lateral-I, lateral-II and lateral-III umbels. Peak flowering occurs during August and is asynchronous even among plants occurring at same niche. Within a plant also, phenological events are asynchronous among primary, lateral-I and lateral-II umbels. Fruit formation commence during the last week of August completing (full maturity) by September last week. Fruit shedding occurs with beginning of October month onwards and physical as well as physiological changes leading to the prennation commence with the beginning of autumn season. This inactive phase lasts upto next spring season. Breeding system studies Floral biology Anther dehiscence (Figure 1c) asynchronously starts with the opening of floral buds through longitudinal slits and continues for 2-3 days. Stigma become receptive (observed by in vivo artificial pollination and resultant pollen germination) when all the anthers of a flower are shed and style reached its maximum length after pro- truding out of stylopodium (Figure 1d). At stigma recep- tive stage, stylopodium is with shiny surface. Period of stigma receptivity depended upon umbel order. Elongated, trinucleate (at shedding stage) and bicol- pate pollen show 76% to 100% (average 92.64 ± 1.71%) stainability and their number/flower vary from 17400 to 40700 (average 26128.75± 1323.41). With two ovules/ flower, pollen-ovule ratio ranged from 8700 to 20350 (average 13064.37± 661.71). Number of seeds produced ranged from 334 to 1024 (average 670.27 ± 65.24) in pri- mary umbel and 112 to 344 (average 216.64 ± 18.88) in lateral-I umbel whereas, no seed set in lateral-II and lat- eral-III umbels was observed as these umbels dry before fruit formation. A D CPrimary L - I L - II L - I L - II L - II L - III L - III B A HG FED CB KJ I L Fig. 2. A. glauca a. Bagging for autogamy; b & c. Seed without and with embryo respectively after TTZ staining; d. Pollen germi- nation on receptive stigma; e. Metaphase-I (n=11, 2n=22); f. Ring formation at poles at anaphase-I; g. Clumping of chromosomes at Metaphase-I; h. Pollen mother cells showing cytomixis i. Decussate and tetrahedral tetrads; j. Isobilateral tetrads; k. Trinucleate pollen grains; l. Seed viability through TTZ test. Fig. 1. A. glauca a. Arrangement of umbel orders primary, lateral-I (L-I), lateral –II (L-II), lateral –III (L-III); b. Schematic representation of a; c. Anther dehiscence stage; d. Protruding style at receptive stage. 26 Kamini Gautam, Ravinder Raina Chromosomal studies In most of the pollen mother cells, the bivalent upto metaphase stage appeared to be clumped together with- out clear separation (Figure 2g), however, some cells with separate 11 bivalents were also observed (Figure 2e). Anaphase-I was interesting as chromosomes at each pole were present in groups forming ring like structure and the number of chromosome in each group at each pole was 11 (Figure 2f ). In 7-8% pollen mother cells, cytomixis (Figure 2h) was observed, however other abnormalities like laggards, bridges, etc. were absent. Pollen grains were trinucleate at pollen shedding stage (Figure 2k). Breeding system Floral visitors like bees, flies, beetles, butterflies and ants were observed visiting its flowers. Open pollina- tion resulted in 670.27 ± 65.24 seeds/primary umbel and under autogamous conditions, only 21.82 ± 15.36 seeds/ primary umbel were set (Table 4). Based on the aver- age number of flowers per primary umbels (591 ± 38.96) with two ovules per flower, 56.71 ± 5.52% seed set under open pollination conditions and 2.63 ± 1.82% under autogamous conditions was observed. Out of is 56.71 ± 5.52% seed set under open conditions, after micro- scopic examinations, 27.49 ± 2.67% of such seeds was with embryo with the rest (29.22 ± 2.84%) being with- Table 1. Qualitative and quantitative morphological features of A. glauca. Plant part Qualitative Quantitative Habit and habitat Erect perennial temperate and alpine herb. Stem Erect, cylindrical, hollow inside, smooth and swollen at nodes. Covered with white powder and variously colored (entire shoot purple; purple upto middle and green at top or opposite; to green with purple patches). L: 172.67 ± 6.36 cm Roots Perennial consisting of tuberous roots, pale yellow to yellowish brown, surface smooth or wrinkled and occasionally tap root splits into two near collar region. Tap L: 18.90 ± 1.66 cm B: 13.56 ± 0.97 mm Secondary L:12.53 ± 1.23 cm B: 1.40 ± 0.36 mm Leaves Large, petiolated, tripinnate, alternate, with very long rachis. Petiole base sheathing. Leaflets: lance-ovate to ovate, tip narrowly-acute to acute, base cuneate, margin irregularly toothed and reticulate venation. Adaxial surface of leaflets dark green and smooth. Abaxial surface grayish white and smooth. Cauline leaves/ plant: 5-12 Inflorescence Compound umbel with umbels of different orders. Umbels/plant: 2 –9 Involucre bracts 6-10 in number, linear and green colored. L: 2.73 ± 0.27 cm Bracts 4-11 in number, linear and green colored. L: 1.72 ± 0.17 cm Flower Bisexual, pedicillate, epigynous, actinomorphic and pentamerous. Spread: 3.39 ± 0.08 mm Pedicel Green colored, length decrease from peripheral towards the centre. Calyx Absent or obsolete. Corolla Petals five, free, valvate, obovate with inward curved tip, green in bud stage and whiter on maturity. L: 2.08 ± 0.03 mm B: 1.52 ± 0.07 mm Androecium Stamens five, green colored, bilobed, dorsifixed, exerted, alternate to petals, dehisce by longitudinal slits and filaments green colored. Anthers remain bend inwards in bud stage and spread outwards at maturity. Filament L: 3.15 ± 0.09 mm Anther lobe L: 0.97 ± 0.02 mm Anther lobe B: 0.77 ± 0.01 mm Gynoecium Ovary inferior, bicarpillary syncarpous, bilocular bearing single solitary ovule in each locule and placentation apical. Style bifid, erect, white coloured and attain full development after anther dehiscence. Stylar base swollen to form stylopodium. Ovary L: 1.30 ± 0.07 mm B: 1.72 ± 0.06 mm Style L:1.67 ± 0.06 mm Ovule size: 0.65 ± 0.04 × 0.29 ± 0.01 mm Fruit Fruit mericarp, green colored, oblong, smooth, flat, pale white to brown, on maturity divides longitudinally into two halves joined with the help of carpophores bearing a single seed in each half. L: 1.66 ± 0.05 cm Seed Flat, pale whitish to brown, with five ridges, two lateral ridges form oblong membranous wings that surrounds the seed, wing color pale white or brown. Small seed L: 0.59 ± 0.03 cm Medium seed L: 0.94 ± 0.02 cm Large seed L: 1.34 ± 0.04 cm Floral formula ⚥ , ⨁, K 0 or obsolete, C5, A5, G (2) L: length; B: breadth. 27Floral architecture, breeding system, seed biology and chromosomal studies in Angelica glauca Table 2. Percentage of plants with different umbel order in A. glauca: Population Primary Primary + Lateral- I Primary + Lateral -I + Lateral -II Primary + Lateral -I + Lateral -II+ Lateral- III Shillaru 100% 24% 72% 4% Shilly 100% 60% 40% 0 % Table 3. Quantitative features of umbels of different order in A. glauca: Umbel order Characters Primary Lateral-I Lateral-II Lateral-III Number per plant 1 2-4 0-5 0-1 Diameter 15.56 ± 0.52 cm × 15.6 ± 0.53 cm 11.59 ± 0.80 × 11.53 ± 0.81 cm 3.93 ± 0.21 cm × 3.93 ± 0.21 cm It was observed to be simple umbel, with upto 10 flowers, very weak and dried later on before blooming. Umbelet number 18.50 ± 0.86 18.05 ± 0.85 15.08 ± 1.22 Number of flowers 591 ± 38.96 539.35 ± 16.92 247.47 ± 31.09 Diameter of peripheral umbelet 3.05 ± 0.07 cm × 3.05 ± 0.07 cm 2.32 cm ± 0.13 × 2.28 ± 0.14 cm 0.75 ± 0.08 cm × 0.75 ± 0.08 cm Diameter of central umbelet 2.43 ± 0.06 cm × 2.42 ± 0.06 cm 1.61 ± 0.16 cm × 1.61 ± 0.16 cm 0.43 ± 0.03 cm × 0.43 ± 0.03 cm Number of flowers in peripheral umbelets 33.65 ± 0.91 30.95 ± 1.01 16.05 ± 1.15 Number of flowers in central umbelets 24.90 ± 0.91 20.8 ±0.89 10.9 ± 1.03 cm Length of peripheral rays 8.2 ± 0.45 cm 5 ± 0.29 cm 1.17 ± 0.80 cm Length of central rays 5.2 ± 0.35 cm 3.01 ± 0.26 cm 0.64 ± 0.05 cm Length of flower stalk in peripheral flowers of peripheral umbelets 1.09 ± 0.07 cm 0.85 ± 0.04 cm 0.25 ± 0.02 cm Length of flower stalk in central flowers of peripheral umbelets 0.48 ± 0.04 cm 0.32 ± 0.03 cm 0.16 ± 0.02 cm Length of flower stalk in peripheral flowers of central umbelets 0.73 ± 0.04 cm 0.54 ± 0.03 cm 0.1 cm Length of flower stalk in central flowers of central umbelets 0.36± 0.03 cm 0.19± 0.02 cm Central flowers were underdeveloped Table 4. Impact of different pollination methods on seed set and viability in primary umbel of A. glauca: Pollination Conditions Observations – Primary umbel Average number of seeds** per umbel* Total seed** set* % Seed set % (with embryo)* Seed set % (without embryo)* Seed viability %*** 100 Seed# weight grams*With embryo Without embryo Open Pollination 670.27 ± 65.24 56.71 ± 5.52 27.49 ± 2.67 29.22 ± 2.84 100 0.00 1.20 ± 0.44 g Self Pollination 21.82 ± 15.36 2.63 ± 1.82 1.27 ± 0.88 1.36 ± 0.93 100 0.00 0.85 ± 0.06 g T calculated value 9.22 8.87 8.87 8.87 4.19 * Statistically significant. **On the basis of number of ovules involved in study. Refers to all these structures that appeared to be like seed (with or without embryo). *** Refers to seeds with embryo only. Seeds without embryo did not show any positive viability due to absence of embryo. # Refers to all these structures that appeared to be like seed (with or without embryo). 28 Kamini Gautam, Ravinder Raina out embryo. Similarly out of the 2.63 ± 1.82% seed set under autogamous conditions, 1.27 ± 0.88% seed was with embryo with the rest 1.36 ± 0.93% without embryo respectively. 100 seed test weight under open pollination (1.20 ± 0.44 g) was statistically higher to 0.85 ± 0.06 g under autogamous pollination. TTZ test revealed 100% seed viability (Figure 2l) in seeds with embryo in both open as well as autogamous conditions and on the con- trary none of the seed without embryo was found to be viable. Seed biology Seed size Differences in seed size were noticed and were cat- egorized into i). small, ii). medium and iii). large seeds (Table 5). Seed set percentage in different umbel orders of A. glauca 670.27 ± 65.24 seeds were obtained in primary umbel which was statistically higher than 216.64 ± 18.88 obtained in lateral-I umbels thus, seed set percentage was 56.71 ± 5.52% in primary and 20.08 ± 1.75% in lat- eral-I umbel. Out of these only 27.49 ± 2.67% and 6.53 ± 0.57% with seeds with embryo were present in pri- mary and lateral-I umbel respectively which was statisti- cally significant. 100% viability in seed with embryo was observed irrespective of umbel order by TTZ test. Seed set by primary umbel had statistically significant 100 test seed weight (1.20± 0.44 g) as compared to 0.94 ± 0.28 g in lateral-I umbel (Table 6). Seeds (with embryo) set percentage as influenced by location, umbel order and position within umbel In Shillaru population (2130 m amsl, district Shim- la, HP, India), statistically non-significant difference in percentage of seed with embryo among primary and lateral-I umbels as well as among peripheral and central regions of these umbel orders was observed (Table 7). 53.80% (maximum) seeds with embryo were observed in peripheral regions of primary umbels and 32.38% (mini- mum) in central region of lateral-I umbels which was however statistically non significant. On overall basis, 48.47% seeds with embryo were obtained in primary and 32.52% in lateral-I umbel (Table 7). In Shilly population (1550 m amsl, district Solan, HP, India), maximum (66.18%) seeds with embryo were obtained in central region of primary umbel and mini- mum (36.68%) in central region of lateral-I umbel which was however statistically non-significant (Table 7). On overall basis maximum (56.56%) seeds with embryo were obtained in primary umbel which was statisti- cally higher to minimum (40.11%) obtained in lateral-I umbels. Maximum (51.43%) seeds with embryo were obtained in central region of umbels and minimum (45.24%) in peripheral regions of umbels which was, however statistically non-significant (Table 7).Amongst the two populations, on overall basis 48.34% (Shilly) and 40.49% (Shillaru) seeds with embryo were obtained which was statistically non-significant (Table 8). Table 5. Different seed size classes in A. glauca. Size class Size of seeds (cm) 100 seed weight in grams (g) Small 0.59 ± 0.03 (0.4-0.7) 0.72 ± 0.03 g Medium 0.94 ± 0.02 (0.8-1.0) 1.02 ± 0.04 g Large 1.34 ± 0.04 (1.1-1.6) 1.18 ± 0.07 g Table 6. Seed set percentage in different umbels of A. glauca: Umbels Observations Average number of seed per umbel*# Total seed set*# % Seed set % (with embryo)* Seed set % (without embryo)* Seed viability %** 100 Seed weight*** grams*With embryo Without embryo Primary Umbel 670.27 ± 65.24 56.71 ± 5.52 27.49 ± 2.67 29.22 ± 2.84 100 0.00 1.20 ± 0.44 g Lateral-I Umbel 216.64 ± 18.88 20.08 ± 1.75 6.53 ± 0.57 13.55 ± 1.18 100 0.00 0.94 ± 0.28 g T calculated value 6.37 6.03 7.30 4.85 4.37 * Statistically significant # On the basis of number of ovules involved in study. Refers to all these structures that appeared to be like seed (with or without embryo). **Refers to seeds with embryo only. Seeds without embryo did not show any positive viability due to absence of embryo. *** Refers to all these structures that appeared to be like seed (with or without embryo). 29Floral architecture, breeding system, seed biology and chromosomal studies in Angelica glauca Seed size vis-à-vis percentage of seeds with embryo The large sized seed consisted of 52.12% seeds with embryo at Shillaru as against 58.12% obtained at Shilly, (statistically non significant). Amongst the small seeds, only 37.67% (Shillaru) and 39.15% (Shilly) seeds were with embryo (Table 9). Amongst medium sized seeds 28.66% (Shillaru) and 41.66% (Shilly) were with embryo (table 9). There was observed no statistically significant difference amongst the two locations i.e. Shilly and Shillaru but significant difference in percentage of seed with embryo amongst seeds of different size class was observed at both locations (Table 9) with large seeds having higher proportion of seeds with embryo (55.12%). Seed germination Seed size class wise, inter and intra population seed germination was conducted and Shillaru (2130 m amsl, district Shimla, HP, India) population gave maximum (31.00%) germination which was statistically higher (Table 10). Seeds of Kilba (3200 m amsl, 31°31’ 18.17’’ N; 78°11’ 49.30’’E district Kinnaur, HP, India) population did not germinate at all and in case of Khan Jungle (2300 m amsl, 30°49’ 13.40’’ N; 77°27’ 47.82’’ E, district Sirmour, HP, India) population, large sized seeds gave maximum germination (24.00%) and medium seeds gave minimum germination (6.00%) (Table 10). In case of Jagatsukh (1982 m amsl, 32° 11’ 43.20’’ N; 77° 12’ 31.82’’ E, district Kullu, HP, India) population large seeds gave maximum germination (8.00%) and small seeds did not germinate at all (Table 10). In case of Thandi Dhar (2240 m amsl, 30° 54’ 51.42’’N; 77° 24’ 44.45’’E, district Sirmour, HP, India) population, medium seeds gave maximum germi- nation (26.67%) and large seeds gave minimum germina- tion (8.33%) (Table 10). In case of seeds from Rohru For- est Division (2700 m amsl, 31°07’09.49’’N; 77°37’35.45’’E, district Shimla, HP, India), small seeds gave maximum germination (34.00%) and large seeds minimum (13.33%) (Table 10). In case of Shillaru population, medium seeds gave maximum germination (48.00%) and large seeds gave minimum germination (10.00%) (Table 10). On over- all basis, non significant impact of seed size on seed ger- mination was observed (Table 11). Table 7. Percentage of seeds with embryo among different umbel order vis-à-vis umbel part in population in A. glauca. Umbel order Umbel part Shillaru population Seed set % Shilly population Seed set % Peripheral Central Mean Peripheral Central Mean Primary 53.80 (47.13) 43.13 (39.15) 48.47 (43.14) 46.94 (46.21) 66.18 (58.12) 56.56 (52.16) Lateral-I 32.65 (34.36) 32.38 (32.37) 32.52 (33.36) 43.55 (41.17) 36.68 (37.12) 40.11 (39.15) Mean 42.23 (40.74) 37.76 (35.76) 45.24 (43.69) 51.43 (47.62) Cd0.05 Umbel order Within umbel Umbel order X within umbel NS* NS NS 12.13 NS NS Values in parentheses are Arc Sine transformed values. * Non significant. Table 8. Overall percentage of seeds (with embryo) comparison between two population of A. glauca. Populations Seed with embryo (%) Shillaru 40.49% Shilly 48.34% T calculated value 1.37* * Non significant. Table 9. Percentage of seed with embryo amongst different seed size classes in A. glauca. Population Seed size Small Medium Large Mean Shillaru 37.67 (34.96) 28.66(31.64) 52.12(46.34) 39.49(37.65) Shilly 39.15(38.42) 41.66(39.90) 58.12(54.95) 46.31(44.42) Mean 38.41 (36.69) 35.16(35.77) 55.12 (50.65) Cd0.05 Sites Seed size Site X size NS* 9.69 NS * Non significant. Values in parentheses are Arc Sine transformed values. 30 Kamini Gautam, Ravinder Raina DISCUSSION Morphology and floral architecture The traded roots of A. glauca are sometimes adulter- ated by roots of Pleurospermum angelicoides (Wall. ex DC) Benth. ex C. B. Clarke and Angelica archangelica L., thereby making morphological studies crucial to check the genuiness of the species. Although the studied popu- lations were of genuine A. glauca being similar in mor- phological features reported earlier (Clarke 1885; Kirti- kar and Basu 1984; Bisht et al. 2003; Nautiyal and Nau- tiyal 2004; Vashistha et al. 2006), yet with regard to the sex type present observations have established beyond doubt presence of only bisexual flowers that has been reported earlier by Butola et al. (2010) also however, Bisht et al. (2008) have reported A. glauca as andromo- noecious (both bisexual and staminate flowers on same individual). Apiaceae members exhibit diverse sexual expression with most of the species being andromonoe- cious, few bisexual (wild Foeniculum vulgare Mill.) and rest either dioecious (Aciphylla or Anisotome) or gynodi- oecious (Gingidia, Scandia and Lignocarpa etc.) (Koul et al. 1993; Reuther 2013). Seed size variation corresponding to test weight is being reported for the first time in this species and such variation was observed irrespective of the umbel order. However, the primary umbel followed by lateral-I umbel only set seed with almost nil seed set by lateral-II and lateral-III umbels. This suggests that only two former types of umbels should be targeted for seed harvest. Seed size variation is also known in the Apiaceae spe- cies like Anethum graveolens L. and Pastinaca sativa L. wherein such variations is correlated with umbel order as well as the portion of flowers within an umbel (Hen- drix 1984; Hołubowicz and Morozowska 2011). Another important seed feature having implica- tions for its reproductive fitness that has been observed is presence of seeds without embryo (thereby sterile) in the species. Low seed germination in A. glauca is already known and seeds without embryo probably are the rea- son. Low seed fertility due to the embryo less seeds may be the reason of its sporadic populations thereby leading to its rarity in nature. This is an important finding and any strategy towards sustainable utilization shall have to factor in this feature. This feature was irrespective of seed produced by different umbel order as well as polli- nation systems indicating physiological causes. Reduced fertile seed output may have some advantages like allow- ing enough space for progeny to grow but limit their number. In self incompatible Stevia rebaudiana Bertoni, two types of sterile and fertile seeds are produced, how- ever that is due to genetic reasons (Raina et al. 2013). Table 10. Site wise open pollination set seed germination response vis-à-vis seed size in A. glauca. Sites Category Small Medium Large Mean Khan Jungle 20.00 (26.54) 6.00 (14.12) 24.00 (29.30) 16.67 (23.32) Kilba - 0.00 (0.00) - 0.00 (0.00) Jagatsukh 0.00 - 8.00 (16.37) 4.00 (8.18) Thandidhar 23.33 (28.83) 26.67 (31.06) 8.33(16.72) 19.45 (25.54) Rohru forest division 34.00(35.64) - 13.33 (21.37) 23.67 (28.50) Shillaru 35.00 (36.25) 48.00 (43.83) 10.00(18.27) 31.00 (32.79) Cd0.05 1. Size categories with in populations 2. Between population with number of size categories 1 and 2 1 and 3 2 and 3 2 and 2 3 and 3 2.12 1.84 1.73 1.37 1.50 1.22 Values in parentheses are Arc Sine transformed values. Table 11. Impact of seed size on germination percentage. Seed size Germination% Small 22.47 (25.48) Medium 20.17 (23.16) Large 12.73 (20.48) CD0.05 NS* * Non significant. 31Floral architecture, breeding system, seed biology and chromosomal studies in Angelica glauca Phenology Phenology of A. glauca follows the general pattern of temperate perennial herbs that undergo perennation dur- ing winter period only to sprout back after snow melt- ing. Flowering commences with the summer season with primary umbels blooming first followed by lateral-I, lat- eral-II and lateral-III umbels with peak flowering during August month. Seed maturation and shedding commenc- es from last week of August month till October. Flower- ing is asynchronous among plants occurring in same niche and within plant too, different phenological events were asynchronous even among primary, lateral-I and lateral-II umbels which appears to be an adaptation to limited pollinator services especially insects. Vashistha et al. (2010) have also reported similar phenological events. Floral biology Flower of A. glauca have been observed to be pro- tandrous with anther dehiscence beginning with the opening of floral buds that continues for 2-3 days. Stig- ma become receptive after complete anther dehiscence that is also characterized by style extending full beyond stylopodium indicating complete intra floral dichogamy. Shiny stylopodium is also indicator of stigma receptiv- ity. Late maturation of stigma coupled with elongation of style after anther dehiscence facilitate dichogamy in A. glauca and appears as an adaptation to avoid autogamy as well as inbreeding depression. Protandry in A. glauca has also been reported by Bisht et al. (2008). In Chaero- phyllum bulbosum L.(Apiaceae), also styles elongates only after pollen is shed and sexual phases are clearly distin- guishable indicating extreme ‘protandry’ (Reuther and Claßen-Bockhoff 2013). Foeniculum vulgare Mill. other member of Apiaceae, has also been found to be strongly ‘protandrous’ as pollen are released much before stigma receptivity (Koul et al. 1996). As A. glauca thrives in hos- tile climatic condition, production of trinuclear pollen grains seems to be an adaptive feature for faster germina- tion on stigma leading to reproductive assurance. Pollen ovule ratio of 13064.37± 661.71, studied for the first time in present studies indicates the species to be an obligate outcrosser as per Cruden (1977). Foenicu- lum vulgare, another member of Apiaceae, is also char- acterized by high pollen ovule ratio of 12005-14635 (Koul et al. 1996). Chromosomal studies The present gametic chromosome count of n=11 is in conformity with the previous diploid count of 2n=22 (Kumar and Singhal 2011) from northwest Himalaya. However, grouping of bivalents at diakinensis and met- aphase-I stage into a ring structure has been observed for the first time in this species. In, only few cells could clear 11 bivalents be observed at these stages. Anaphase-I too was characterized by the presence of two rings of 11 chromosomes at each poles. Presence of ring of 11 chro- mosomes at anaphase-I in A. glauca poles appears to be similar to the ‘renner’ complexes (entire haploid genom- es which are inherited as single units) present in genus Oenothera, wherein due to reciprocal translocations of chromosome arms, all the 14 chromosomes form two rings of seven chromosome each (Greiner 2008). As has been discussed earlier, significant proportion of seeds of A. glauca were without embryo leading to reduced germination, which may be due to this meiotic anomaly. Further studies on female meiosis, embryo development as well as more extensive studies on male meiosis are required to establish the consequences of meiotic anom- aly in A. glauca. Although meiotic abnormalities like lag- gards, bridges, etc. were not observed but in some pollen mother cells, cytomixis was observed in A. glauca. Cyto- mixis often leads to abnormal meiotic behavior, variation in pollen grains size and low pollen viability or sterility e.g. in Alopecurus arundinaceus Poir. (Koul 1990), Polygo- num tomentosum Willd. (Haroun 1995), Hordeum vul- gare L.(Haroun 1996), Brassica napus var. Oleifera Delile. and B. campestris var. oleifera DC (Souza and Pagliarini 1997), Vicia faba L. (Haroun et al. 2004), and Meconopsis aculeate Royle (Singhal et al. 2008). Despite chromosome arranged in rings as well as cytomixis, pollen stainability did not seem to be impacted much as it ranged from 76% to 100% in different plants of A. glauca studied. Breeding system studies Significant seed set under open pollination condi- tions in comparison to autogamous conditions estab- lished the species strongly favouring (about 95%) cross pollination. This aspect is being reported for the first time in this species. Interestingly open pollination also resulted in much higher seed (with embryo) set (27.49 ± 2.67%) as compared to autogamous seed (1.27 ± 0.88%), again indicating strong presence of cross pollination. As it is generally presumed that selfing rates increase with increasing altitudes (Schroter 1926; Bliss 1962; Gar- cia-Camacho and Totland 2009; Korner and Paulsen 2009) as pollinator abundance and activity become lim- iting factors due to hostile climatic conditions at higher altitudes (Arroyo et al.1982, 2006; Bingham and Orthner 1998; Medan et al. 2002). Contrary to this view, the species under investigation (A. glauca) favours cross pollination 32 Kamini Gautam, Ravinder Raina and dichogamy seems to play a key role in its cross fertili- zation. Eritricium nanum (Vill.) Schrad.ex Gaudin (Borag- inaceae), Chaetanthera renifolia  (J. Remy) Cabrera (Aster- aceae) and Nardostachys grandiflora DC (Valerianaceae) other temperate plant species, are also known for higher cross pollination rates at high altitudes (Writh 2010; Diaz et al. 2011; Gautam and Raina 2016). Inflorescence attrib- utes, high pollen ovule ratio and asynchronous opening of flowers in A. glauca are also evidence for its cross pollinat- ing nature. However, low seed set in A. glauca limits, natu- ral variation essential for genetic improvement. Seed biology As has been already discussed only primary and lateral-I umbels set seeds. Among these too primary umbel set significantly more (56.71 ± 5.52%) seed than lateral-I umbel (20.08 ± 1.75%). Higher seed set in pri- mary umbel in Eryngium alpinum L. and Carum carvi L. (family Apiaceae) is already known (Bouwmeester and Smid 1995; Gaudeul and Bottraud 2004). Interestingly, despite blooming, lateral-II as well as lateral-III umbels do not set any seed which seems due to their late development or restricted resource alloca- tion and they seem to only for enhancing floral visibil- ity of its plants for pollinator attraction. Of the 56.71± 5.52% and 20.08 ± 1.75% seed set by primary and later- al-I umbels only27.49 ± 2.67% and 6.53 ± 0.57% seed is with embryo respectively indicating higher fertile seed production by primary umbels. This low fertile seed pro- duction in A. glauca may be the reason for generally low germination rates in this species. Fruit and seed set percentage is generally low in late blooming flowers than early blooming ones (Zimmer- man and Aide 1989), and several reasons like resource competition among the ovaries of an inflorescence (Lee 1988); reduced pollen receipt by later blooming inflo- rescence (Lee 1988); intrinsic features (Berry and Calvo 1991; Diggle 1995) may be the reasons. Non significant variations in fertile seed production between peripheral and central flowers of a umbel and also among seed from two ecologically different popula- tions (Shilly and Shillaru) indicates that A. glauca as a strategy, limits fertile seed production either for nutrient resource conservation or to ensure better quality fertile seed that can produce a healthy progeny. Seed germination Seed germination behavior of any species impacts the genetic variability in any species and lower germina- tion rates deprive such variation. Inter population ger- mination variation was observed amongst the six popu- lation viz. Khan Jungle, Kilba, Jagatsukh, Thandidhar, Rohru forest division and Shillaru with Shillaru popula- tion excelling others (31.00% seed germination) with no germination in seeds of Kilba. Apiaceae family members are known for generally low germination rates (Koul et al. 1993) and in A. glauca poor and erratic germination with maximum of 8% ger- mination is already reported (Nautiyal et al. 2002; Butola and Budola, 2004; Butola and Samant 2006). Present stud- ies have revealed that low germination in A. glauca is not due to dormancy but production of seeds without embryo. CONCLUSION Only primary and lateral-I umbels set seeds as other lateral-II and lateral-III only attract pollinators without setting any seed. Production of embryo less seeds (ster- ile seed) is a major reproductive bottleneck in this spe- cies. Seed size variation occurs within same plant as well as within same umbel with large seeds having higher proportion of fertile seeds (with embryo). The species is strongly cross pollinating. ACKNOWLEDGEMENT This study was funded by Department of Biotech- nology (DBT), Government of India. The first author would like to acknowledge Department of Science and Technology (DST), Government of India for INSPIRE (Innovation in Science Pursuit for Inspired Research) Fellowship.Authors would also like to acknowledge Director, HFRI (Himalayan Forest Research Insti- tute, Conifer Campus, Panthaghati, Shimla−171009, Himachal Pradesh, India) and Dr. Sandeep Sharma (Scientist−E, HFRI) for providing necessary facilities for conducting research experiments. REFERENCES Arroyo MTK, Munoz MS, Henriquez C, Bottraud TI, Perez F. 2006.Erratic pollination, high selfing levels and their correlates and consequences in an altitudi- nally widespread above tree-line species in the high Andes of Chile. Acta Oeco. 30(2): 248–257. Arroyo MTK, Primack R, Armesto J. 1982.Community- studies in pollination ecology in the high temperate Andes of central Chile: pollination mechanisms and altitudinal variation. Am J Bot. 69(1): 82–97. 33Floral architecture, breeding system, seed biology and chromosomal studies in Angelica glauca Berry PE, Calvo RN. 1991. Pollinator limitation and posi- tion dependent fruit set in the high Andean orchid Myrosmodes cochleare (Orchidaceae). Plant Syst Evol. 174(1): 93–101. Bingham RA, Orthner AR. 1998.Efficient pollination of alpine plants. Nature. 391(6664): 238–239. Bisht AK, Bhatt A, Rawal RS, Dhar U. 2008.Assessment of reproductive potential of different populations of Angelica glauca Edgew: a critically endangered Hima- layan medicinal herb. J Mt Sci. 5(1): 84–90. Bisht AK, Manjkhola S, Joshi M. 2003. Comparative account of two high value species of Himalayas: Angelica glauca Edgew. and Angelica archangelica L. Indian For.129 (10): 1241–1248. Bliss LC.1962.Adaptations of arctic and alpine plants to environmental conditions. Arctic. 15(2): 117–144. Bouwmeester HJ,  Smid HG. 1995. Seed yield in caraway (Carum carvi): role of pollination. J of Agric Sci. 124(2): 235–244. Butola JS and Badola HK. 2006. Chemical treatments to improve seedling emergence, vigour and survival in Heracleum candicans Wall. (Apiaceae): a high value threatened medicinal and edible herb of Himalaya. J. Plant Biol. 33(3): 215–220. Butola JS, Badola HK. 2004. Effect of pre-sowing treat- ment on seed germination and seedling vigor in Angelica glauca: a threatened medicinal herb. Curr Sci. 87(6): 796–799. Butola JS, Badola HK. 2008. Propagation conditions for mass multiplication of three threatened Himalayan high value medicinal herbs. P G R Newsletter. 153: 143–147. Butola JS, Samant SS. 2006. Physiological studies on seed germination of Angelica glauca. J Trop Med Plants. 7(2): 205–212. Butola JS, Vashistha RK, Samant SS, Malik AR. 2010. Technology for propagation and cultivation of Angel- ica glauca Edgew.: a threatened high value Himalayan medicinal cum edible herb. Emer Med Plants. 2(1): 67–72. Butola JS, Vashistha RK. 2013. An overview on conserva- tion and utilization of Angelica glauca Edgew. in three Himalayan states of India. Med Plants. 5(3):171–178 Chauhan NS. 1999. Medicinal and aromatic plants of Himachal Pradesh. New Delhi (India): Indus Pub- lishing Company. Clarke CB. 1885. The flora of British India. VolI. L. Lon- don: Reene and Co. Corner EJH. 1976. Seeds of dicotyledons. Cambridge: Cambridge University Press. Cruden RW. 1977. Pollen ovule ratio: a conservative indi- cator of breeding system in flowering plants. Evolu- tion. 31(1):32–46. Diaz CT, Gonzalez SG, Stotz GC, Morales PT, Paredes B, Millaqueo MP, Gianoli E.2011. Extremely long-lived stigmas allow extended cross-pollination opportuni- ties in high Andean plant. PLoSONE. 6(5): e19497. doi:10.1371/journal.pone.0019497. Diggle PK. 1995. Labile sex expression of andromonoe- cious Solanum hirtum: pattern of variation in floral structure. Can J Bot. 69(9): 2033–2043. Garcia-Camacho R, Totland O. 2009. Pollen limita- tion in the alpine: a meta-analysis. Arct Antarct Alp Res. 41(1):103–111. Gaudeul M, Bottraud IT. 2004. Reproductive ecology of the endangered alpine species Eryngium alpinum L. (Apiaceae): phenology, gene dispersal and reproduc- tive success. Ann Bot. 93(6):711–721. Gautam K, Raina R. 2016.New insights into the phenol- ogy, genetics and breeding system of critically endan- gered Nardostachys grandiflora DC. Caryologia. 69 (1): 91–101. Gomez KA, Gomez AA. 1984. Statistical procedures for agricultural research. 2nd edition. New York (John Wiley). Goswami PK, Khale A, Ogale S. 2012.Natural remedies for polycystic ovarian syndrome (PCOS): a review. Inter J Pharmaceut and Phytopharmaco Res. 1(6): 396–402. Greiner S. 2008. Oenothera, a unique model to study the role of plastids in speciation, Dissertation der Fakultätfür Biologie der Ludwig-Maximilians-Univer- sität München (Dissertation at the Faculty of Biology, at the Ludwig Maximilian University of Munich). Haroun SA, Al Shehri AM, Al Wadie HM. 2004. Cyto- mixis in the microsporogenesis of Vicia faba L. (Fabaceae).Cytologia. 69(1): 7–11. Haroun SA. 1995. Cytomixis in the pollen mother cells of Polygonatum tomentosum Schrank. Cytologia. 60(3): 257–260. Haroun SA. 1996. Induced cytomixis and male sterility in pollen mother cells of Hordeum vulgare L. Delta J Sci. 20(1): 172–183. Hendrix SD. 1984. Variation in seed weight and its effects on germination in Pastinaca sativa L. (umbelliferae). Am J Bot. 71(6): 795–802 Hołubowicz R, Morozowska M. 2011.Effect of umbel position on dill (Anethum graveolens L.) plants grow- ing in field stands on selected seed stalk features. Folia Hort. 23(2): 157–163. IUCN.1993. Draft IUCN red list categories. Gland, Swit- zerland. Kirtikar KR, Basu BD. 1984.Indian Medicinal Plants.2nd edition. Dehradun (India).Bishen Singh Mahendra Pal Singh, 1592 p. 34 Kamini Gautam, Ravinder Raina Korner C, Paulsen J. 2009. Exploring and explaining mountain biodiversity. In: Spehn M and Korner C, Editors. Data mining for global trends in mountain biodiversity. Florida (USA): CRC Press, 10 p. Koul K K. 1990. Cytomixis in pollen mother cells of Alo- pecurus arundinaceus Poir. Cytologia. 55(1): 169–173. Koul P, Sharma N, Koul AK. 1993. Pollination biology of Apiaceae. Curr. Sci. 65(3): 219–222. Koul P, Sharma N, Koul AK. 1996. Reproductive biol- ogy of wild and cultivated fennel (Foeniculum vulgare Mill.). Proc Indian Natn Sci Acad. B62 (2): 125–134. Kuafman PB, Carison TF, Dayanadan P, Evans ML, Fisher JB, Parks O, Wells JR. 1989.Plants, their biology and importance. New York (USA): Hopper and Raw Pub- lishers. Kumar P, Singhal VK. 2011. Chromosome number, male meiosis and pollen fertility in selected angiosperms of the cold deserts of Lahaul-Spiti and adjoining areas (Himachal Pradesh, India).Plant Syst and Evol. 297(3-4): 271–297. Lawrence GHM. 1951.Taxonomy of vascular plants. New York: McMillan. Lee TD. 1988. Patterns of fruit and seed production. In: Doust J L and Doust L L, Editors, Plant reproductive ecology: patterns and strategies. New York (USA): Oxford University Press, 179–202 p. Medan D, Montaldo NH, Devoto M, Mantese A, Vasel- latiV, Bartoloni NH. 2002. Plant–pollinator rela- tionships at two altitudes in the Andes of Mendoza, Argentina. Arct Antarct Alp Res. 34(3): 233–241. Nautiyal MC, Prakash V, Nautiyal BP .2002. Cultivation techniques of some high altitude Medicinal herbs. Ann For. 10(1): 62–67. Nautiyal MC, Nautiyal BP. 2004.Agrotechniques for high altitude medicinal and aromatic plants. High Altitude Plant Physiology Research Centre, Dehradun, India, pp. 99–133. Pimenov MG, Leonov MV. 2004. The Asian Umbelliferae biodiversity database (ASIUM) with particular reference to South-West Asian taxa. Turk J Bot. 28(1–2): 139–145. Pimenov, MG, Kljuykov EV. 2003. Notes on some Sino- Himalayan species of Angelica and Ostericum (Umbelliferae). Willdenowia 33: 121–137. Raina R, Bhandari SK, Chand R, Sharma YP. 2013. Strat- egies to improve poor seed germination in Stevia rebaudiana: a low calorie sweetener. J of Med Plants Res.7(24): 1793–1799. Reuther K, Claßen-Bockhoff R. 2010.Diversity behind uniformity: inflorescence architecture and flowering sequence in Apiaceae-Apioideae. Plant Div and Evol. 128(1–2): 181–220. Reuther KU. 2013. Spatial and temporal flower presentation in Apiaceae-apioideae. Dissertation, Doctor of Science, Department of Biology at the University of Mainz. Saeed MA, Sabir AW. 2008. Irritant and cytotoxic cou- marins from Angelica glauca Edgew Roots. J Asian Nat Prod Res. 10(1): 49–58. Samant SS, Dhar U, Palni LMS. 1998. Medicinal plants of Himalaya: diversity distribution and potential values. Nainital, India, Himvikas, GyanodayaPrakashan. Samant SS, Nandi SK,  Butola JS.2009.Conservation sta- tus and cultivation of selected medicinal plants in the Indian Himalayan Region. In: Chaurasia SB, Yadav OP, Rimando, AM, Terrill, TH. Editors. Advances in Agriculture, Environment and Health: Fruits, Vegeta- bles, Animals and Biomedical Sciences. Delhi (India): Satish Serial Publishing House. pp 185–214. Schroter C. 1926. Das Pflanzenleben der Alpen [The plant life in the Alps]. Albert Raustein, Zurich, Switzer- land. Sher H, Elyemeni M, Kamran H and Sher H. 2011. Ethn- obotanical and economic observations of some plant resources from the northern parts of Pakistan, Eth- nobot Res Appl. 9: 027–041. Singh SK, Rawat GS. 2000. Flora of great Himalayan national park, Himachal Pradesh, Dehradun (India) Bishen Pal Singh and Mahendra Pal Singh. Singhal VK, Kumar P. 2008.Impact of cytomixis on mei- osis, pollen viability and pollen size in wild popu- lations of Himalayan poppy (Meconopsis aculeata Royle).J Biosci. 33(3): 371–380. Souza AM, Pagliarini MS. 1997.Cytomixis in Brassica napus var. oleifera and Brassica campestris var. Oleif- era (Brassicaceae).Cytologia. 62(1): 25–29 The wealth of India: a dictionary of Indian raw materials and industrials products. 1985. Council for Scientific and Industrial Research, I. A. Public and Information Directorate, New Delhi, India, 153 p. Vashistha R, Nautiyal BP, Nautiyal MC. 2006.Conserva- tion status and morphological variations between populations of Angelica glauca Edgew. and Angelica archangelica Linn. in Garhwal Himalaya. Curr Sci. 91(11): 1537–1542. Vashistha RK, Butola JS, Nautiyal BP, Nautiyal MC. 2010. Phenological attributes of Angelica glauca and A. archangelica expressed at two different climatic zones in Western Himalaya. Open Access J Med Arom Plants. 1(1): 7–12. Weberling F.1989. Morphology of flower and inflores- cence. New York (USA) Cambridge University Press. Wirth LR, Graf R, Gugerli F, Landergott U, Holderegger R. 2010. Lower selfing rate at higher altitudes in the alpine plant Eritrichium nanum (Boraginaceae). Am J Bot. 97: 899–901. Zimmerman J K and Aide T M.1989. Patterns of fruit production in a neotropical orchid: pollinator vs. resource limitation. Am. J Bot. 76(1):67–73. Substantia An International Journal of the History of Chemistry Vol. 2, n. 1 - March 2018 Firenze University Press