Impaginato 45 1. Introduction The genus Hydrangea L. includes 23 species, mainly distributed in the American and Asiatic conti- nents (McClintock, 1957). It is a very popular orna- mental plant for both garden and interior design and has recently been commercialized as a high value fresh or dried cut flower. Interest in Hydrangeas is mainly due to the strik- ing coloration of its inflorescences (corymbs or pani- cles), that range from pink, blue, white, to light pur- ple or dark purple. Flowers are produced from early spring to late autumn and have two inflorescence morphologies: ‘mophead’ - with large flowers form- ing spherical flower heads; and ‘lacecap’ that resem- ble round, flat flower heads with a center core of subdued, fertile flowers surrounded by outer rings of showy, sterile flowers. To further increase their popularity, new hybrids and cultivars should be developed. In flowering plants, the main objective of selective breeding is to increase genetic variability in ornamental traits such as flower color, flower shape and plant form. To achieve this objective, intra- and inter-specific hybridizations have been widely used in breeding programs. Hybrids between H. macrophylla (Thunb.) Ser. and H. paniculata Sieb. were produced using embryo rescue, but the resulting plants were sterile and lacked vigor (Reed et al., 2001; Reed, 2004). In vitro embryo rescue procedures have been used to facilitate the recovery of interspecific hybrids of many genera (Bridgen, 1994; Sharma et al., 1996), and have recently been used to recover a putative H. macrophylla (Thunb.) Ser. x H. arborescens L. hybrid (Kudo and Niimi, 1999 a, b). Hybrid embryos often resume growth and develop into normal plants when removed from ovules and placed on an aseptic nutri- ent media. This procedure is known as in ovule Adv. Hort. Sci., 2017 31(1): 45-51 DOI: 10.13128/ahs-20725 Development of pollination and in vitro germination techniques to improve the hybridization in Hydrangea spp. G.A. Venturieri 1, B. Nesi 2, S. Lazzereschi 2, S. Pecchioli 2, G. Burchi 2 1 UFSC/CCA-FIT, Federal University of Santa Catarina, Campus Itacorubi, Rod. Admar Gonzaga, 1346 Itacorubi, CEP 88034-000, Florianopolis (SC), Brazil. 2 Consiglio per la Ricerca in Agricoltura e l’Analisi dell’Economia Agraria, Unità di Ricerca per il Vivaismo e la Gestione del Verde Ambientale ed Ornamentale (CREA-VIV), Via dei Fiori, 8, 51017 Pescia (PT), Italy. Key words: breeding, fertilization barriers, fruit cut system, hortense, seed disinfection, sowing system. Abstract: Hydrangea is a genus of ornamental plants which is gaining new markets mainly as a fresh or dried cut flower, but it is also important as a pot plant and for landscaping. To expand its market, new hybrids should be developed. To increase the hybridization efficiency, some techniques were developed and tested: i) evaluation of two pollination sys- tems; ii) comparison among fruit-cut systems before in vitro cultivation to develop embryos and to allow the growth of new genotypes; iii) evaluation of seed disinfection systems for in vitro germination; iv) sowing systems using seeds and fruits from stocks cultivated in two environments. To increase inter- and intra-specific hybridization, pollination by dis- persion of previously collected pollen on the top of a corymb by a brush was more effective than pollination using the corymb itself as a brush. A longitudinal cut system can be considered the best treatment to be applied on fruits before in vitro cultivation to allow growth of seedlings. Sterilization of seeds can be done by immersion in a solution of commer- cial bleach for 5 minutes on MS culture medium with PPM®. When stocks are cultivated in greenhouses, in vitro contami- nation is lower and seeds have a better rate of germination. The results of these experiments were applied in a breeding program on Hydrangea using sexual crosses. (*) Corresponding author: giorgini.venturieri@ufsc.br Received for publication 5 October 2016 Accepted for publication 15 February 2017 Copyright: © 2017 Author(s). This is an open access article 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. http://creativecommons.org/licenses/by/4.0/ http://creativecommons.org/licenses/by/4.0/ Adv. Hort. Sci., 2017 31(1): 45-51 46 embryo culture (Reed, 2000). In vitro germination of seeds is a common tech- nique used to overcome incompatibility barriers in ornamental plant hybrids (Eeckhaut et al., 2006 a, b; Lazzereschi et al., 2012; Nesi et al., 2012). Since most Hydrangea seeds are so small (diameter of 0.5 mm) and hybrid seeds production is usually low (0-5 seeds/fruit), so, the development of a successful in vitro method for Hydrangea seed germination would be an important tool for breeding programs. To increase seed germination, Greer and Rinehart (2010) have developed an in vitro method for cultiva- tion and assay of H. macrophylla (Thunb.) Ser. and H. paniculata Sieb. seeds, through germination on solid media in conjunction with Plant Preservative Mixture (PPM®, Plant Cell Technology, Inc., Washington, DC, USA), and by sterilizing seed with trichloro-s-tri- azinetrione (Trichlor). The objective of the present study was to develop and test several techniques for the generation and recovery of hybrids in Hydrangea. These included: two pollination systems (using a brush or an inflores- cence on the top of the corymb); different ovary cut systems, aimed at embryo and seed rescue; different seed disinfection systems for in vitro germination (evaluating the time of immersion in a bleach solu- tion and the addition of Plant Preservative Mixture - PPM® in media); comparison among three sowing systems using seeds and immature fruits in aseptic conditions, and in compost in climatized beds, from stocks cultivated in two environments (in a green- house and under shading net). The outcomes of these experiments were used to determine best practices for the hybridization of Hydrangea species using sexual propagation. 2. Materials and Methods The experiments were carried out at the Council for Agricultural Research and Economics - Landscaping Plants and Nursery Research Unit (CREA- VIV) in Pescia (PT) (43° 49’ 00” N; 10° 48’ 00” E), Italy. At CREA-VIV, a germplasm collection of Hydrangea, composed of 66 genotypes belonging to H. macro- phylla ssp. macrophylla (Hortensia and Lacecap Group), H. paniculata, H. serrata, H. villosa, H. querci- folia, H. anomala ssp. petiolaris, H. arborescens, H. heteromalla, H. involucrata, H. aspera and three genotypes of Schizophragma hydrangeoides, was maintained. Some trials were also arranged in Sanremo (IM) (43°49′ N; 7°47′ E) with some geno- types replicated from the CREA-VIV collection. Selected cultivars of this collection were used for hybridizations. During the summer of 2014, crosses among different genotypes, belonging to H. macro- phylla ssp. macrophylla (38 cultivars), H. macrophylla ssp. serrata and H. paniculata (5 cultivars each); H. arborescens, H. aspera, H. quercifolia and H. involu- crata (one cultivar for each species) were made (Table 1). Then, the capsules obtained from con- trolled pollinations were collected and used as start- ing material in the experiments, as described below. Pollination systems Twelve crosses, involving H. macrophylla, H. arborescens and H. quercifolia, randomly distributed in a germplasm collection, were subjected to two pollination systems (treatments). Before pollination, sterile flowers and all extremely immature fertile were removed from inflorescences to be used as females prior to opening of the fertile flowers. After, the petals and anthers of all remaining fertile flowers were removed and the inflorescence was covered with a breathable plastic bag. Inflorescences to be used as males were also bagged prior to flower open- ing. Pollination experiments were performed 1 to 4 days following emasculation. The two pollination treatments included: a) polli- nation by the dispersion of previously collected pollen on the top of a corymb, aided by a brush; b) a simple dispersion of pollen using the corymb as a brush, where anthers presenting freshly-dehisced pollen were touched with the exposed stigmas of the Table 1 - Plant material used for intra- and inter-specific crosses between different genotypes of Hydrangea spp. Species Cultivars H. macrophylla ssp. macrophylla ‘Alberta’, ‘Alpen Gluhen’, ‘Ayesha’, ‘Benxi’, ‘Bianca Ceriana’, ‘Dienemann’, ‘Elbatal’, ‘Endless Summer’, ‘Etoille Violette’, ‘Europa’, ‘First Red’, ‘Grattino’, ‘Green Shadow’, ‘Hanaby’, ‘Harlequin’, ‘Intermezzo’, ‘Lake San Markos’, ‘Lanarth White’, ‘Lemon Wave’, ‘Libelle’, ‘Magical Coral’, ‘Magical Garnet’, ‘Magical Jade’, ‘Magical Noblesse’, ‘Masja’, ‘Myharayama Yae’, ‘Nymphe’, ‘Paris’, ‘Red Beauty’, ‘Renate Wate’, ‘San Baronto’, ‘Schnball’, ‘Seour Therese’, ‘Sibilla’, ‘Tricolor’, ‘White First’, ‘Zorro’ and ‘Kardinal’ H. macrophylla ssp. serrata ‘Acuminata’, ‘Blue Bird’, ‘Miranda’, ‘Preziosa’ and ‘Yae-no-amacha’ H. paniculata Limelight’, ‘Phanthom’, ‘Pink Diamond’, ‘Unique’, and ‘Vanilla Fraise’ H. arborescens ‘Annabelle’ H. aspera ‘Rowellane’ H. involucrata ‘Yorakutama’ H. quercifolia ‘Snow Queen’ Venturieri et al. - Pollination and in vitro germination techniques applied to Hydangea hybridizations 47 emasculated flowers. After pollination, inflores- cences were covered again with the bags which remained on the plants until fruit collection. Effect of treatment was evaluated by the number of develop- ing fruits on bagged female flowers. A t test was used to compare averages between treatments (Sokal and Rohlf, 1981). In total, one hundred and thirty-five hand pollina- tions were carried out in Pescia, and 33 in Sanremo, using a different subset of Hydrangea species. Each species cross was repeated 3 times. Embryo and seed rescue from immature fruits Approximately 90 days after pollination (DAP), well-developed fruits (n = 444), obtained from the breeding program were used to investigate rates of embryo and seed rescue from immature fruits. Fruits were sterilized in a solution of commercial bleach (5% of active chlorine), in distilled water (1:2, v/v) plus 2 μl/100 ml Tween20® (Sigma, St. Louise, MO, USA) for 10 minutes. Subsequently four ovary cut systems (treatments) were applied: a) stigma off - stigmatic branches were removed; b) longitudinal cut - stigmatic branches were removed and a longitudi- nal cut, from the top of the fruit up to approximately to peduncle insertion, was made; c) equatorial cut - stigmatic branches were removed and a transversal cut at the larger diameter of the fruit was made (both sections were cultivated); d) top cut - 1/3 of the distal part of the fruit was removed (Fig. 1 a-d). Fruits were cultured in a one-half strength Murashige and Skoog (MS) medium (Murashige and Skoog, 1962) supplemented with 1 mg/L of naphtha- lene acetic acid (NAA), 1 ml/L of PPM®, 30% sucrose, 6 g/L of agar and pH was adjusted to 5.8. Fruits were individually placed in test tubes. Six replication were used by cross. All vials were cultivated in a chamber maintained at 23±1°C under 16 h/day photoperiod provided by fluorescent tubes at 35 µmol/m∙s. Ovaries that were considered as contaminated or dead were counted 7 days after, and their proportion (p) was a square root of transformed arcsine (Ayres et al., 2007) and evaluated using Analysis of Variance. Averages were compared by Fisher LSD test of significance for α= 0.05 (Ayres et al., 2007). Values were expressed as a percentage. Physical attributes of germinated seedlings were described for each treatment. Disinfection of seeds for in vitro germination Fruits considered as mature (130±10 DAP) were collected and left to dry on a laboratory bench, and then grinded carefully using a mortar and pestle to liberate seeds from capsules. The obtained mass was passed through 0.71 mm mesh of a soil gradation sieve (Grade 25, Giuliani Tecnologie, Scientific Instrument, Torino, IT) to remove large debris, and used in two disinfection experiments as described below. Sterilization as a function of time immersion in bleach solution. About 0.025 g of the mass of seeds and debris was placed in a piece of TNT envelope and submerged for the disinfection process (Fig. 2 a). The envelopes were dipped in 70% ethanol for 30 sec, fol- lowed by sterilization in a commercial bleach solu- tion, as already as described above, for 5, 10, 20, and 30 minutes and rinsed in sterilized distilled water twice (Fig. 2 b). Then the envelopes were opened (Fig. 2 c) and their contents were laid on a one-half MS culture medium supplemented with 1 mg/L of NAA, 1 ml/L of PPM®, and 30% sucrose, and adjusting pH to 5.8. TNT was used to cover petri dishes (Fig. 2 d), and then cultures were placed in the chamber as described above. Germinated seeds and contamina- tion of petri dishes were evaluated after 15 days. Frequency of contaminated and uncontaminated petri dishes that contained germinated seeds was used to evaluate the effect of contamination on ger- Fig. 1 - Graphic representation of the different cuts applied to the immature fruits of Hydrangea: a) stigma off; b) longitudinal cut; c) equatorial cut; d) top cut (Fig. 1 a-d); e) development of new plantlet from immature fruits in vitro cultured. Adv. Hort. Sci., 2017 31(1): 45-51 48 mination by Fisher’s test. Spearman rank order corre- lation was calculated between number of germinated seeds and disinfection time (Ayres et al., 2007). Sterilization as a function of duration in bleach solution and PPM® concentration in the media. In this experiment, culture medium and in vitro sowing sys- tem were the same as described above, but using a factorial design of two variables: PPM® in the medi- um at 3 different concentrations (2, 4 and 6 ml/L) and four disinfection times (5, 10, 20 and 30 min- utes). About 0.025 g of the mass of seeds was utilized to the disinfection process. Each treatment was repeated 3 times. The number of germinated seeds per petri dish and the proportion of contaminated petri dishes were used as parameters. A factorial analysis of variance was performed and averages were compared by Fisher LSD test of significance for α=0.05. Sowing systems using seeds and fruits from stocks cultivated in two environments Seeds were sown in Pescia, using material from two environments (Pescia and Sanremo), following three different systems. From 130 different crosses made in Pescia, seeds from 33 crosses were sown in vitro, well developed but still immature fruits, from 22 crosses were cultivated in vitro and seeds from 130 crosses were sown in organic compost (peat and perlite, 1:2, v/v) on artificially climatized beds, inside a greenhouse. From 34 different crosses made in Sanremo, seeds from 19 crosses were sown in vitro, well developed but still immature fruits from 24 crosses were cultivated in vitro and seeds from 34 crosses were sown in the same compost and condi- tions as described above. About 0.025 g of the mass of seeds was utilized to in vitro sowing for each com- bination of cross. Three ovaries of each cross were in vitro cultivated, and of the total mass of seeds were sown in organic compost. Each sowing system was repeated 3 times. All in vitro germinations were carried out in fall- winter of 2013 and compost germination in summer of 2014. Due to the difficulty in counting the number of seeds sown inside a fruit or inside a mass of sieved seeds and debris, the comparison was based in num- ber of germination events that happened per treat- ment. A nonparametric χ2 statistical analysis was applied, and where significant intergroup differences were found, multiple comparisons were conducted using the partitioning χ2 test to differentiate between treatments, both for α=0.05 (Ayres et al., 2007). 3. Results Pollination systems According to the t-test, pollination using a brush with pollen collected from the corymb (average of 7.6 fruits per inflorescence cross), showed a mean significantly higher than using a corymb as a brush (average of 4.8 fruits per inflorescence cross) (p= 0.001). Although, the time needed for pollination using a brush is considerably greater than that using only a corymb. Embryo and seed rescue from immature fruits No statistical significance (p=0.07) was observed in the rate of contamination between different cut treatments. A lower percentage of contamination was observed in the treatment “stigmas off” (Fig. 1 a), probably due to less damage in the fruit tissues compared to the other cutting systems (Table 2). The number of dead fruits was not affected by the differ- ent fruit cut systems applied (p=0.50). Cut system were further differentiated based on variation in ger- mination behavior. In the treatments “stigmas off” and “top cut”, fruits swelled but seedlings did not emerge from fruit, suggesting poor germination. Furthermore, some seedlings were confined inside the fruit and did not develop and grow outside of the fruit. In the “equatorial cut” treatment, two portions of the fruit were grown: seeds were able to germi- Fig. 2 - Disinfection process of seeds obtained from mature capsules harvested: (a) TNT envelops with seeds plus debris inside; (b) Envelops dipped in the disinfection solution; (c) Staples were taken off and envelop opened; (d) Seeds plus debris were layered on medium and TNT was left to cover seeds. Venturieri et al. - Pollination and in vitro germination techniques applied to Hydangea hybridizations 49 nate and develop new seedlings from the lower por- tion, but not from the upper portion. This suggests a potential loss of seedlings. Conversely, the treatment “longitudinal cut” (Fig. 1 b) allowed the seeds to ger- minate and seedlings promptly grew and developed upright (Fig. 1 e). Disinfection of seeds for in vitro germination Sterilization as a function of time immersion in bleach solution. No statistical differences were observed in the number of contaminated petri dishes between genotypes (p=0.37) or duration of disinfec- tion period (p=0.22). Seed germination was lower in contaminated petri dishes only, where sterilization time was positively correlated with the number of germinating seeds (r s of Spearman=0.51; p=0.01) (Table 3). Sterilization as a function of duration in bleach solution and PPM® concentration in the media. Seed immersion duration in bleach solution did not signifi- cantly affect the rate of contamination (p=1.00); however, a significant difference was observed for PPM® concentration (p≈0.00). No interaction between these two variables was observed (p=1.00). Therefore, seeds surface sterilization with bleach solution was not enough to prevent contamination, but sterilization was achieved only with the addition of PPM®. Contamination occurred only when 2 ml/L of PPM® were added to the culture medium; when a higher concentration of PPM® was used, no contami- nation was observed at any time of immersion in the solution of bleach (Table 4). So, to sterilize seeds for in vitro germination, the use of a culture media with 4 ml/L PPM® and a steril- ization of 5 minutes is recommended. Sowing systems using seeds and fruits from stocks cultivated in two environments The average percentage of germination in Pescia was 30.5%; and in Sanremo, 55.8%. Inside the envi- ronment “Pescia”, no statistical differences were observed between sowing systems (p=0.8, d.f.=2), but in “Sanremo” there were differences (p≈0.00 d.f.=2). Partitioning χ2 tests revealed that treatments in the Sanremo environment were all statistically dif- ferent from each other (p<0.007, d.f.= 1), with imma- ture fruits cultivated in vitro demonstrating the high- est percentage of germination among treatments (Fig. 3). 4. Discussion and Conclusion In the present paper, we use several experiments to define work strategies and priorities to hybridize Table 2 - Proportions of contaminated and dead fruits, by ovary cut system, and behaviour of eventual germinated seeds from survived fruits (z) Values followed by the same letter in each column do not dif- fer statistically for α=0.05. Table 3 - Effect of application of bleach solution in function of duration of disinfection by different Hydrangea geno- types (z) Symbols: ○ uncontaminated petri dishes; ● contaminated petri dishes; ♣ number of germinated seed. Table 4 - Evaluation of PPM® concentration on the number of Hydrangea spp. germinated seeds/per petri dishes, and proportion of contaminated petri dishes (z) Averages and percentage (by column) followed by the same letter do not differ at α>0.001. Cutting systems Sample size (N) Contamination (%) Death (%) Behaviors of germinated seedlings Stigma off 150 31 a z 31.3 a Fruits swelled but seedlings could not emerge from fruit and develop Longitudinal cut 150 34 a 38.0 a Seeds germinated prompt- ly, grew up and develop upright E q u a t o r i a l cut 72 38 a 29.2 a Germinated seed devel- oped but none from the top slices, suggesting potential seedlings loses Top cut 72 49 a 31.9 a Some seedlings were con- fined inside the fruit and did not emerged Genotype Duration of disinfection (minutes) Contami- nated Petri dishes/ genotype5 10 20 30 H. macrophylla spp. macro- phylla Proc. Izu Ohoshima ○z ○ ○ ○ 0 H. macrophylla spp. macro- phylla Proc.Takeoka Chiba ○ ○ ○ ○ 0 H. involucrata Proc. Yamamae Yoko Tama ○ ○ ○ ○ 0 H. macrophylla ‘Libelle’ x H. paniculata ‘Limelight’ ● ● ●(♣) ●(♣♣♣♣♣) 4 H. involucrata ‘Myharayama kokonoe Tama’ x H. macro- phylla ‘Alberta’ ● ● ● ●(♣♣♣) 4 H. macrophylla ‘Libelle’ x H. macrophylla ‘Europa’ ●(♣) ●(♣) ● ● 4 Total of contaminated Petri dishes/duration of disinfection 3 3 3 3 PPM® concentration (ml/L) Germinated seeds/petri dishes (N) Contaminated petri dishes (%) 2 5.5 a z 66.7 a 4 3.8 a 0 b 6 4.0 a 0 b Adv. Hort. Sci., 2017 31(1): 45-51 50 Hydrangea species using sexual propagation. The aim of the project was to hybridize all species involved in a breeding program, therefore pollination and in vitro germination techniques had to be developed to maximize efficiency. To efficiently produce fruits of inter- and intra- specific hybrids, pollination by the dispersion of pre- viously collected pollen on the top of a corymb aided by a brush resulted in better pollination than using the corymb itself as a brush. Nevertheless, pollina- tion using only the corymb was used for the purpose of the present project because it was faster and more convenient. However, the hybrid origin should be certified using molecular markers when seedlings are established or at flowering, based on morphological characters. Based on the germination behaviour, the “longitu- dinal cut” system applied to fruits prior to their trans- fer to the culture medium is recommended to pro- mote vigorous growth of seedlings. It is considered as the best system because seeds readily germinated and developing seedlings promptly grew upright. Similar responses were also observed in the produc- tion of interspecific hybrids of Lilium longiflorum Thunb. and L. × elegans (Roh et al., 1996). Seed surface sterilization for in vitro germination with bleach solution was not enough to prevent cont- amination, but sterilization was achieved with the addition of PPM®. In some plant species, germination is favoured by the presence of microorganisms, usu- ally attributed to Rhizobacteria (Saharan and Nehra, 2011), but seldom adapted to in vitro germination of ornamental plants except in orchids (Tsavkelova et al., 2007); this probably could be an explanation of what happened in the case of Hydrangea seeds. For the objectives of the present study, however, it would not be an advantage because almost all cul- ture media were completely covered by fungal mycelia causing seedlings to collapse within the first week following germination. Germination in fruits showed a higher number of successful germinations than the other two systems. The highest proportion of germination success, observed for “in vitro inside fruit with a longitudinal cut for embryo and seed rescue” system, could be due to supplementary nutrients, given by the medi- um, that ensured development of hybrid seeds with- out endosperm (Eeckhaut et al., 2006 a and b). Nevertheless, specific studies of endosperm lacking in seeds from Hydrangea hybrid crosses are unknown. The number of germinated seeds in Pescia was lower than in Sanremo: in Pescia the stock plants were cultivated under shading net while in Sanremo in greenhouse conditions. In fact, the contamination of explants cultivated in vitro can be associated to management of stock plants. Pollination system and in vitro germination tech- niques have allowed us to obtain a number of new individuals with potential ornamental traits. Currently this material is under selection, providing the basis for the development of a Hydrangea breed- ing program in course at CREA-VIV at Pescia - Italy. Acknowledgements The authors thank to Drs. Francesco Ferrini and Laura Mugnai, throughout the international coopera- tion agreement between the Federal University of Santa Catarina - Brasil and the University of Florence - Italy; to the Council for Agricultural Research and Agricultural Economics Analysis/ Landscaping and Nursery Plants Research Unit, at Pescia - Italy, for the laboratories and support. We also thanks for the Brazilian Government, by the Program Science Without Border from CAPES/CNPq, for the grant given for the first author. And we also acknowledge Dr. Alistair Campbell for linguistic advice. References AYRES M., AYRES J.M., AYRES D.L., SANTOS A.A.S., 2007 - BioEstat 5.0.: aplicações estatísticas nas áreas das Fig. 3 - Events with successful germination of potential Hydrangea spp. hybrids by different sowing systems, in two environments. Columns, by environment, followed by the same letter do not differ statistically for α=0.05. Venturieri et al. - Pollination and in vitro germination techniques applied to Hydangea hybridizations 51 Ciências Biomédicas. - Sociedade Civil Mamirauá, Belém, Pará-Brasil, pp. 324. BRIDGEN M.P., 1994 - A review of plant embryo culture. - HortScience, 29: 1243-1245. EECKHAUT T., VAN HUYLENBROECK J., VAN LAERE K., VAN BOCKSTAELE E., 2006 b - Interspecific hybridization in woody ornamentals: how to deal with barriers? - Acta Horticulturae, 725: 117-126. EECKHAUT T., VAN LAERE K., DE RIEK J., VAN HUYLEN- BROECK J., 2006 a - Overcoming interspecific barriers in ornamental plant breeding, pp. 540-551. - In: DA SILVA J.A.T. (ed.) Floriculture, ornamental and plant biotech- nology. Advances and Tropical Issues. Vol. I. Global Science Books, UK, pp. 646. GREER S.P., RINEHART T.A., 2010 - Dormancy and germina- tion in vitro Response of Hydrangea macrophylla and Hydrangea paniculata seed to light, cold-treatment and gibberellic acid. - JEH, 28(1): 41-47. KUDO N., NIIMI Y., 1999 a - Production of interspecific hybrids between Hydrangea macrophylla f. hortensia (Lam.) Rehd. and H. arborescens L. - J. Jpn. Soc. Hort. Sci., 68(2): 428-439. KUDO N., NIIMI Y., 1999 b - Production of interspecific hybrid plants through cotyledonary segment culture of embryos derived from crosses between Hydrangea macrophylla f. hortensia (Lam.) Rehd. and H . arborescens L. - J. Jpn. Soc. Hort. Sci., 68(4): 803-809. LAZZERESCHI S., NESI B., PECCHIOLI S., GRASSOTTI A., 2012 - Costituzione, selezione e propagazione di ibridi inters- pecifici di Lilium. - Acta Italus Hortus, 6: 188. McCLINTOCK E., 1957 - A monograph of the genus Hydrangea. - Proc. Calif. Acad. Sci., 29: 147-255. MURASHIGE T., SKOOG F., 1962 - A revised medium for rapid growth and bioassays with tobacco tissue culture. - Physiologia Plantarum, 15(3): 473-497. NESI B., LAZZERESCHI S., PECCHIOLI S., CARDONA-SUAREZ C.M., COLLA G., CARDARELLI M.T., GRASSOTTI A., 2012 - Costituzione, selezione e propagazione di ibridi inter- specifici di Lilium. - Acta Italus Hortus 4: 85-94. REED S.M., 2000 - Development of an in ovule embryo cul- ture procedure for Hydrangea. - J. Environ. Hort., 18(1): 34-39. REED S.M., 2004 - Floral characteristics of a Hydrangea macrophylla × H. paniculata hybrid. - Proc. Southern Nursery Assn. Res. Conf., 49: 580-582. REED S.M., RIEDEL G.L., POOLER M.R., 2001 - Verification and establishment of Hydrangea macrophylla ‘Kardinal’ × H. paniculata ‘Brussels Lace’ interspecific hybrids. - J. Environ. Hort., 19(2): 85-88. ROH M.S., GRIESBACH R.J., GROSS K.C., LAWSON R.H., 1996 - Production and identification of interspecific hybrids of Lilium longiflorum and L. × elegans. - Acta Horticulturae, 397: 93-100. SAHARAN B.S., NEHRA V., 2011 - Plant growth promoting rhizobacteria: A critical review. - Life Sci. Medicine Res. LSMR., 21: 1-30. SHARMA D.R., KAUR R., KUMAR K., 1996 - Embryo rescue in plants - a review. - Euphytica, 89: 325-337. SOKAL R.R., ROHLF F.J., 1981 - The principles and practice of statistics in biological research. - 2nd ed. State University of New York at Stony Brook, San Francisco, CA, USA, pp. 776. TSAVKELOVA E.A., CHERDYNTSEVA T.A., KLIMOVA S.Y., SHESTAKOV A.I., BOTINA S.G., NETRUSOV A.I., 2007 - Orchid-associated bacteria produce indole-3-acetic acid, promote seed germination, and increase their microbial yield in response to exogenous auxin. - Arch. Microbiol., 188(6): 655-664.