Determination of Pollen Viability and in Vitro Pollen Germination of Rosa Dumalis and Rosa Villosa

Pollen quantity, viability and germination of four genotypes of Rosa dumalis and R. villosa were investigated. The number of anthers per flower were 90.3 in genotype 4 (Rosa villosa) and 116.4 in genotype 1 (Rosa dumalis). Genotype 1 showed the highest pollen viability on the basis of both TTC and IKI tests. Most of the rose species are diploid but there are several species with higher ploidy levels as well, especially tetraploids and also pentaploids, in section Caninae. Diploid species are usually self-sterile whereas the polyploids are self-compatible. The deviating section of Caninae contains mostly pentaploid species and is characterized by the Caninae meiosis with an uneven chromosomal contribution from the two parents. In addition, these species are self-fertile, sometimes apomictic, and have reduced pollen viability but a high seed set (Nybom et al. 2005). In general, there is a linear relation between pollen viability and germination capability in many fruit species (Grigs et al. 1971). Germination capability of pollen depends on various factors, namely nutrition conditions of species and varieties used and environmental factors (Eti and Stosser 1988). To investigate pollination potential, estimates should be made of pollen quantity and viability, as well as of pollen germination capability. This study was designed to determine the qualitative and quantitative characteristics of the pollen and to assess in vitro pollen germination of some rose genotypes that are presently being domesticated in Turkey. In the present study, four genotypes from Rosa section Caninae were used belonging to Rosa dumalis (genotypes 1 and 2) and Rosa villosa (genotypes 3 and 4). The number of anthers per flower were determined by counting anthers of 40 flowers from each genotype and the number of pollen per anther was determined by hemacytometer with six replications (Eti 1990). Pollen viability was tested according to Eti (1991) in 1% TTC (2,3,5-triphenyl tetrazolium chloride) and IKI (iodine+potassium iodide). The morphological homogeneity level of pollen was also investigated according to Eti and Stosser (1988). For in vitro pollen germination, 5, 10, 15, 20, 25, 30, 35 and 40% sucrose and 0.03, 0.01 and 0.1% boric acid were used in hanging drop method. In addition 1% agar + 15% sucrose combinations in Petri dishes were also used. The experiment was set according to the randomized block design with four replications (each replication includes 50 pollen) and the values were evaluated by the DMRT. Data were arc-sine transformed for analysis but non-transformed means are presented. There were statistical differences (p < 0.01) among genotypes in terms of the number of anther per flower, the number of pollen per flower, the number of pollen per anther, pollen homogeneity and pollen viability. Average anther number in a flower and pollen numbers per anther was highest in genotype 1 as 116.4 and 1617, respectively (Table 1). The highest pollen number per flower was 188.219 in genotype 1 and followed by the genotype 2 (R. dumalis) as 154.157. The genotype 3 and 4 (R. villosa) had the lowest pollen numbers with 127.075 and 102.400, respectively (Table 1). Gunes et al. (2005) reported that the number of anther per flower of rose species were between 81.4 (R. villosa) and 148.1 (R. elliptica) similar to present result. E-mail: [email protected]