Seed germination trends of endemic vascular plants in the Valencian Community (Spain). M. C. ESCRIBÁ1, E. LAGUNA1 & M. GUARA2 1 Generalitat Valenciana, Conselleria de Medi Ambient, Servei de Conservació i Gestió de la Biodiversitat, C/. Arquitecto Alfaro, 39, E-46010 València, Spain (E-mail: email@example.com). 2 Departament de Botànica, Universitat de València, Facultat de Ciències Biològiques, Av. Dr. Moliner, 50, E-46100 Burjassot, València, Spain (E-mail: Miguel.Guara@uv.es). Summary Seed germination is a critical phase in the reproductive cycle of great importance for species fitness, and variation in germination percentage has been interpreted as an adaptation to ecological conditions. Understanding the germination process will help in the conservation strategies, especially ex situ strategies in order to obtain seedlings in greehouse. Germination test were carried out with seeds of 56 endemic, rare or threatened plants species collected between 1996 and 2002 in different populations localities from the Valencian Community. Correspondence analysis (CA) on the germination percentage matrix displays a gradient ordination onto axes I-II (43.37%). Cluster analysis on the three first axes point out three day-period germination groups, and four species groups. Trends of two Silene hifacensis populations considered in these analyses are different. Introduction Seed germination is a critical phase in the reproductive cycle of great importance for species fitness, and variation in germination percentage has been interpreted as an adaptation to ecological conditions (NAVARRO & GUITIÁN, 2003). Understanding the germination process will help in the conservation strategies, especially ex situ strategies in order to obtain seedlings in greehouse. The aim of the present communication is to show the germination capacity of the different endemic vascular plants, not previously studied, collected in the Valencian Community and, to analyze possible similarities among species and populations. Methods Germination test were carried out with seeds of 56 endemic, rare or threatened plants species collected between 1996 and 2002 in different localities from Valencian Community and bordering territories. Only, there were sampled two populations of Silene hifacensis (Figure 1). Seeds collected were cleanned and dryed progressevely up beetwen 30 and 45 days, then preserved in cold storage at 4º C. Previous shown, seeds were summerged during 10 minutes in NaClO. Incubated in a SANYO MLR-350 chamber at 18ºC-20ºC and 10ºC with 8 hours periods and total period of light of 10 hours, during 30 days calculating the germination percentage daily. Water was until the saturation point. With the percentage data of diary germination, a Correspondence Analysis (CA) was carried out to detect tendencies among species. A Cluster Analysis (Single, Complete and UPGMA methods) based on euclidean distance with the species-points and day-points coordenates of the three first factorial axes (FOUCART, 1982) to detect species-groups and days-groups was executed. The best dendrograms were selected for interpretation after Mantel test. Multivariate statistics package NTSYS-pc 2.1 (ROHLF, 2002) was used for ordination and cluster analysis. Results Germination test After the incubation period, only seeds of two species, Iberis carnosa subsp. hegelmaieri and Hippocrepis valentina, germinated completely and a few species passed the 95% (Centaurea saxícola, Saxifraga longifolia, Dianthus multiaffinis, Dipcadi serotinum, Aster aragonensis, Biscutella montana, Armeria alliacea subsp. alliacea). Between the 70% and the 95% of germination there are 19 species (Diplotaxis ibicensis, S. hifacensis –population 1–, Jasione mansanetiana, Vaccinium myrtillus, Gypsophila tomentosa, Seseli montanum subsp. granatense, G. struthium, S. diclinis, Sideritis spinosa subsp. subespinosa, Lycium intricatum, Teucrium lepicephalum, T. flavum, Senecio auricula, D. turolensis, Erodium aguilellae, Thymus willkomii, Vella lucentina, Muscari neglectum, Petrocoptis pardoi). S. hifacensis –population 2– reached the 63.5% of germination at the end. Twenty-eight species has passed the 70% at the end of 30 days period. Thirtytwo species reached or passed the 50% of seed germination at 15th day (those previously denoted, excepting E. aguilellae and T. flavum, and Biscutella lucentina, Dianthus hispanicus subsp. edetanus, Echium saetabense, Helianthemum caput-felis, Muscari atlanticum, Sideritis javalambrensis, Thymus granatensis subsp. micranthus, Verbascum fontqueri). And the same percentage at 5th day by only 11 species (A. alliacea subsp. alliacea, A. aragonensis, B. montana, D. multiaffinis, D. turolensis, D. ibicensis, G. tomentosa, G. struthium, I. carnosa subsp. hegelmaieri, S. diclinis, V. lucentina). Accumulated percentage of germination 100 90 80 70 60 50 40 30 20 10 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 A_alli A_font Ast B_mon B_luc M_atl M_neg Dipc C_sax C_lag Days D_mul D_tur D_edet Dipl Conv Chei Card Lina E_sang E_agui Eryn Gera G_tom G_stru H_cap H_squ Iber Jasi Leuc Lyci Echi Hipp Onon Peri Petr R_str R_bar Saxi Sene Sese S_gla S_spin S_ede S_jav Si_vir Si_hif1 Si_hif2 Si_dic T_flav T_lep T_wilk Th_mic T_pul Vacc V_luc V_sp V_font Figure 1. Accumulate percentage of seed germination of species considered. Comparison among species and germination days Ordination Correspondence analysis (CA) on the germination percentage matrix displays a gradient ordination onto axes I-II (43.37%) –arch effect– (Figure 2). The first eight germination days are located against the rest of days by the factorial axe I (28.61%), ordering species as a function of germination start time. Those were earlier germination with a pick over the 4th or 5th day (Armeria alliacea subsp. alliacea, Aster aragonensis, Biscutella montana, Dianthus multiaffinis, Diplotaxis ibicensis, Gypsophila struthium, G. tomentosa, Silene diclinis, Thymus granatensis subsp. micranthus, and Vella spinosa), were located at the right side. In front of these species, those with later germination with a pick over the 14th-15th or 19th days (Saxifraga longifolia, and Senecio auricula) were located at the left side. The factorial axe II (14.76%) located in the right extrem the first four days, in the middle the 5th to 13th germination days, and the last seventeen days in the left extrem. Species were ordered species with a very earlier germination pick, 2nd or 3th day, (Convolvulus valentinus, Dianthus turolensis and Helianthemum squamatum) against to species with middle later germination pick ,7th day, (Petrocoptis pardoi, Seseli montanum subsp. granatense, Thymus willkomii, and Vaccinium myrtillus). The factorial axe III (11.24%) –not represented– arranges the 12th to 13th germination days opposite to 30th day, ordering species with later germination pick (Centaurea lagascae, C. saxícola, Eryngium ilicifolium and Muscari atlanticum) front to species with germination delay and/or very later germination picks. 2.04 H squ d30 d2 d1 D tur Conv 1.31 d28 d26d27 d29 Chei V luc M neg d25 Peri d3 T flav d23 0.59 d19Gera Lina II Iber Onon d21 S gla V sp Th mic d24d18 d16 R bar d17 DiplB mon d14 Leuc d4 d15 d20 R str Ast G A Gtom alli stru d22 Sene Saxi Si hif2 Card Jasi D mul E agui H cap -0.13 d13 E sang d5 Eryn atl V font M C lag Si dic C sax Lyci D edet d12d11 A font Si vir d10 d9 Dipc T wilk d6 S spin Echi B luc S ede Vacc d8 S jav T pul Sese Petr d7 Hipp Si hif1 -0.86 T lep -1.72 -0.97 -0.23 0.52 1.27 I (28.61%) Figure 2. Correspondence analysis ordination on factorial axes I-II (43.37%). Clustering Cluster analysis on the three first axes point out three day-period germination groups (Figure 3). The earliest germination days (group T1: first three days), the middle germination days (group T2), with two subgroups (T2a: 4th to 10th days) and (T2b: 11th to 24th days), and the last germination days (group T3: 25th to 30th days). These three day-period germination are consistent with the CA axes interpretation below. Respect to species, cluster analysis shows four species groups (Figure 4). The first group (S1), with two subgroups, corresponding with earlier germination species with a pick over the 4th or 5th day and middle earlier germination species with a pick over the7th or 8th day. The second group (S2) formed by middle germination species with a pick on 12th to 14th days. The third group (S3) has a primary very low germination pick over the 14th or 15th days and secundary pick few days later (Muscari neglectum, Cheirolophus lagunae, Periploca laevigata subsp. angustifolia, Linaria cavanillesii, Geranium sanguineum, Teucrium flavum, Sideritis glauca). Finally, the fourth group (S4) has the earliest germination pick, 2nd or 3th day. As in the day-period germination groups, these species groups are consistent with the CA axes interpretation below. Discussion There are two populations of a unique species, Silene hifacensis, not coincident in ordination space or clustering. The population Si_hif1 has been included in group S1, while population Si_hif2 was in group S2. This circunstance seems more obvious among very different species taxonomically separated, and not so clear in close species (same genus, e.g. Dianthus, Muscari, Sideritis), but in multivariate analysis have not been considered other variables (habitat, phenology, plant age, climatic conditions, etc.). As it has been suggested (see references in Navarro & Guitián, op. cit.), variability in responses of seeds of different populations may be due to environment differences ultimately responsible for corresponding genotypic differences. Not all species have fully germinated during the tests, but a few germinated in a good percentage. This aspect is very important if there is necesity of obtaining seedlings, but the test was as simple as possible, based on International Seed Testing Association (ISTA) rules. Probably, comparing separated populations, differences would be detected, and testing different treatement conditions some germination percentage results were improved. References FOUCART, T.- 1982 - Analyse factorielle. Programmation sur micro-ordinateurs - Méthode + Programmes. Masson, Paris. LAGUNA, E.; M.B. CRESPO, G. MATEO, S. LÓPEZ-UDIAS, C. FABREGAT, LL. SERRA, J.J. HERRERO- BORGOÑÓN, J.L. CARRETERO, A. AGUILELLA & R. FIGUEROLA - 1998 - Flora endémica, rara o amenazada de la Comunidad Valenciana - Generalitat Valenciana, Conselleria de Medi Ambient. Valencia. NAVARRO, L. & J. GUITIÁN - 2003 - Seed germination and seedling survival of two threatened endemic species of the northwest Iberian Peninsula - Biological Conservation, 109: 313- 320. ROHLF, FJ. - 2002 - Ntsys. Numerical Taxonomy and Multivariate Analysis System. v. 2.1 - Exeter Sofware, Setauket, NY.