ACTA BIOLOGICA CRACOVIENSIA Series Botanica 48/2: 49–58, 2006 SEEDLING EMERGENCE IN THE ENDANGERED JUNIPERUS OXYCEDRUS SUBSP. MACROCARPA (SM.)BALL IN SOUTHWEST SPAIN ROCÍO JUAN, JULIO PASTOR, INMACULADA FERNÁNDEZ*, AND JUAN CARLOS DIOSDADO Department of Plant Biology and Ecology, University of Sevilla, Box 1095, 41080 Sevilla, Spain Received April 10, 2006; revision accepted July 20, 2006 Juniperus oxycedrus subsp. macrocarpa is an endangered species in southwest Spain, with seed dormancy as found in other species of the same genus. This study employed different experiments to determine a method to improve the seedling emergence in this species. Three types of seedling emergence trials were performed: (a) untreated seeds under greenhouse conditions, (b) untreated seeds under natural conditions, and (c) treated seeds under greenhouse conditions, with different acids (sulphuric, hydrochloric and nitric) for 10 and 30 min, followed or not by cold stratification for 3 months. In all trials, seeds derived from both mature and immature cones were used to verify which one produced higher seedling emergence. Previously, seed viability was verified and a proper substrate for greenhouse sowing was selected. The best percentage of seedling emergence was obtained in the "a" and "b" trials. In trial "a", seeds derived from immature cones germinated significantly better than mature ones. Chemical scarification of seeds with or without cold stratification yielded less seedling emer- gence than the other trials. Key words: Cupressaceae, Juniperus oxycedrus subsp. macrocarpa, dormancy, seedling emer- gence, viability. INTRODUCTION Lee et al., 1995; Owens and Schliesing, 1995; Jordan de Urríes, 1997; Cantos et al., 1998; Ortiz et Juniperus oxycedrus subsp. macrocarpa (Sm.) Ball al., 1998; García, 2001; Juan et al., 2003). is a shrub or small tree typical of coastal environ- Seed germination difficulties in this genus are ments in the Mediterranean region (Jalas and high because of dormancy directly associated with Suominen, 1973; Castroviejo et al., 1986). Although the embryo, seed coat impermeability to water, this taxon, common to both dunes and cliffs, used to and/or the presence of germination inhibitors be spread along the seacoast, it has suffered severe (Hajar, 1991; Cregg et al., 1994; Lee et al., 1995; regression due to human pressure, leaving reduced Ueckert, 1997; Cantos et al., 1998). Some trials sug- populations in small isolated patches, as for other gest that it is due to physical dormancy directly species of the genus (Hajar, 1991; Clifton et al., related to the seed coat, because germination 1997; Sharew et al., 1997). In southwest Spain in increases when the seed coat is removed (Cantos et particular, it is common to find low-density popula- al., 1998). However, Baskin and Baskin (1998) tions of this taxon because of severe urban pressure point out that seeds from Juniperus plants exhibit along coastal areas, resulting in a discontinuous dis- physiological dormancy in which the embryo is tribution (Pastor and Juan, 1999). The progressive unable to develop a radicle due to an inhibition destruction of its natural habitat in southwest Spain mechanism. Pardos and Lazaro (1983) suggested has caused coastal juniper to be classified as in dan- that seeds of J. oxycedrus have a double dormancy ger of extinction (Hernández and Clemente, 1994). feature involving both endogenous and exogenous Previous studies have shown that natural regen- factors. Other authors have also acknowledged that eration of J. oxycedrus subsp. macrocarpa and Juniperus seeds show both seed coat and embryo other taxa in this genus is not easy due to several dormancy (Herrero, 1959; Cregg et al., 1994; factors, including slow growth (Ceballo and Ruíz de Johnson, 1995; Lee et al., 1995; Barbour, 1999). la Torre, 1979), low seed viability and/or difficulty in A phenomenon which may occur in other germinating (Pardo and Lazaro, 1983; Hajar, 1991; Juniperus species is one first suggested by Herrero *e-mail: [email protected] PL ISSN 0001-5296 © Polish Academy of Sciences, Cracow 2006 50 Juan et al. (1959) in J. thurifera and confirmed later by other years to develop mature cones. During the first year authors (Balboa-Zavala and Dennis, 1977; Pinfield the cones are small, hard and green-glaucous in et al., 1990; Pinfield and Gwarazimba, 1992; Jensen color. During the second summer, these cones reach and Eriksen, 2001) for some angiosperm species their final size; they are still hard but turn green-yel- belonging to Acer, Prunus or Malus genera. By this lowish (immature cones). Finally, during the second phenomenon, seeds become dormant as they autumn the cones are fleshy and show a reddish color mature, so that when they are fully mature the seeds (mature cones). It is possible to find cones at different lose their germination ability, while immature seeds stages of maturity on the same plant simultaneously. are able to germinate. This feature could help Seed for these trials were harvested from late explain the low percentage of germination found in summer to autumn 1999. Mature and immature some species within Juniperus, in studies always cones were collected from four populations in south- using seeds derived from mature cones in the exper- west Spain whose cone characteristics had been pre- iments (Pardos and Lazaro, 1983; Rietveld, 1989; viously studied (Juan et al., 2003): Doñana (DO), Cregg et al., 1994). Punta Umbría (PU), Caños de Meca (CM) and Punta Seed dormancy can be overcome through dif- Paloma (PP). The first three are protected areas ferent methods. Cold or hot stratification, or both under different categories (national park, natural methods applied in turn, are among the treatments site and natural park, respectively), while the last is most used to break physiological seed dormancy in an area not under protection. and therefore increase germination frequency in many conifers (Herrero, 1959; Johnsen and VIABILITY Alexander, 1947; Catalan, 1978; Gosling, 1988; Poulsen, 1996; Downie et al., 1998; Offord and All trials were carried out with sinking seeds, Meagher, 2001). On the other hand, in Juniperus because floating ones are empty (Juan et al., 2003). species the method most usually employed to over- Before starting the germination trials, the via- come physical seed dormancy is pretreatment with bility of the seeds to be used in this work was different acids such as commercially pure (96–98%) checked. From 800 to 1200 randomly selected seeds sulphuric acid (Herrero, 1959; Pardo and Lazaro, from immature cones (~3 months from full maturity) 1983; Hajar, 1991). were tested for each population, as seed viability Juniperus oxycedrus is of ecological interest derived from mature cones had been previously pub- because it is a pioneer of inhospitable habitats in lished (Juan et al., 2003). Seed viability of both nutrient-poor and unstable soil with strong sea mature and immature seeds was tested by both the spray, habitats which later can be colonized by other cutting method and the tetrazolium chemical test less tolerant species. Recently, Joy and Young (ISTA, 1976; 1999). In the first method, each seed (2002) suggested that J. virgiana, another pioneer was opened to determine the presence or absence of species in coastal environments, can facilitate and an embryo. In the second method, the seeds were accelerate primary succession, due to reduction of immersed in 2,3,5-triphenyltetrazolium chloride solar radiation and temperature fluctuations solution to verify the condition of the embryo. beneath its canopy, which help to retain soil mois- ture and organic matter levels. Previously, Cregg et SUBSTRATE SELECTION al. (1994) and Lee et al. (1995) pointed out the high ecological value of some species of Juniperus, which For selection of the substrate used in all trials, act as natural windbreaks or shelterbelts. seedling emergence trials were carried out with The aim of the present study was to determine mature seeds from six individuals randomly chosen the effects of different variables (substrate, seed from Punta Umbría (PU) and Caños de Meca (CM), ripeness, cold stratification, chemical scarification, two populations with high cone production. From etc.) on seedling emergence in J. oxycedrus subsp. each tree, 200 seeds were extracted and sown in macrocarpa. Determining the most appropriate three different substrates: (a) pine humus (PH), method to improve the germination rate should con- material derived from pine bark and needle com- tribute to the conservation and recovery of an post, (b) a mixture of blonde and black peat in a endangered species. 20:80 ratio (Blo/Bla 20:80), and (c) the same mix- ture but in an 80:20 ratio (Blo/Bla 80:20). MATERIALS AND METHODS SEEDLING EMERGENCE SPECIES AND STUDY AREA Untreated seeds in greenhouse trials Juniperus oxycedrus subsp. macrocarpa is a dioe- This trial started in November 1999, using 500 cious species whose female individuals require two seeds (both mature and immature) per individual Seedling emergence of Juniperus oxycedrus 51 and population. For each individual, seeds were TABLE 1. Immature seed viability by cutting and tetra- allocated to four replicates of 125 seeds each. Each zolium methods was placed in a tray (40 × 25 × 10 cm) maintained under greenhouse conditions (25°C day/17–18°C night). The number of individuals used from each population was 22 in PU, 20 in CM, and 15 in DO and PP. To monitor the timing of seedling emergence, during the first month the trays were observed daily, and thereafter weekly until September 2000. Untreated seeds in natural habitat trials This trial started in March 2000, using mature and immature seeds derived from 10 randomly selected Data in parentheses correspond to mature seed viability previ- individuals of each studied population. From each ously published (Juan et al., 2003) individual, 200 mature seeds were extracted and distributed as follows: 50 seeds exposed to sunlight without watering (full sun/not watered), 50 seeds STATISTICAL ANALYSIS exposed to sunlight with watering (full sun/watered), The variation of each response variable was ana- 50 seeds sunlight-protected without watering (shad- lyzed by one-way or two-way ANOVA.