Pollen fertility in Musa: Viability in cultivars grown in Southern Australia

Abstract

Pollen viability was examined by recording the number of viable as well as non-viable microspores in anthers just before anthesis, as it could be an efficient and rapid indicator of female fertility. It was thought that competency in meiotic restitution in microspores could indicate similar competency in megaspores. Pollen fertility was compared among seeded diploids and edible triploids of Musa and correlated with ovule fertility. Viability was examined using the Alexander’s pollen stain procedure. The seeded diploid species M. acuminata, M. balbisiana, and M. ornata had 3 times more viable pollen than the edible tetraploids (AAAB). M. balbisiana and M. ornata had significantly more viable pollen than M. acuminata. The tetraploids contained 3 times more viable pollen than the edible triploids AAA and ABB and 4 times more than the AAB cultivars. The genome A or B did not affect pollen viability within the triploid cultivars examined. The AAA triploid Gros Michel had the highest percentage of viable pollen at 13% and Green/Red the lowest at 3%. Pollen viability was influenced by meiotic disturbances and an association was made between viability of pollen and viability of embryo sacs. Pollen viability from anthers of Australian grown species and cultivars of Musa a. ssp. (undescribed subspecies) was compared with published reports of pollen viability from Indian grown species and cultivars. The pollen viability between the Indian cultivated and wild diploids of Musa a. ssp. was similar at 50–66%, but less than the pollen viability of Australian diploid Musa a. ssp of 84%. The Indian-grown triploid cultivars had 21–29% viable pollen and the tetraploids had 28% viable pollen, whereas the Australian grown triploids had 6–10% viable pollen and the tetraploids 29% viable pollen. There was a difference in pollen viability between genome groups and within genome groups. Different species and cultivars of Musa possess different levels of competency in the production of microspores, which correlated positively with levels of megaspore fertility in the same species and cultivars.

Male fertility and occurrence of 2n gametes in East African Highland bananas (Musa spp.)

“…This result is similar to that reported by Fortescue and Turner (2004) who found a large variability in pollen viability among diploid, triploid and tetraploid banana accessions. Pollen viability differed between and within genome groups.…”

“…Adeleke et al (2004) showed that pollen viability ranged from 30.3 to 48.8% in some AAA varieties. As in previous studies (Dumpe and Ortiz 1996;Tenkouano et al 1998;Adeleke et al 2004;Fortescue and Turner 2004) this study also showed that pollen stainability is higher in diploids than in triploids (Table 3, 4). Low pollen stainability of the triploid (2n = 3x = 33) EAHB varieties may be due to the large number of inviable and non-fertile pollen due to meiotic irregularities in triploids (Shepherd 1960).…”

“…The mean pollen output has been found to vary significantly between and within different genome groups in banana (Sathiamoorthy and Balamohan 1993;Fortescue and Turner 2004). It was unusual to find that very little variation in pollen output existed among the EAHB varieties in this study.…”

“…Pollen viability differed between and within genome groups. In their study, Fortescue and Turner (2004) reported that pollen viability ranged from 5.6 to 18.0% within the AAA dessert bananas and from 70.8 to 97.9% among diploids. In this study, pollen Table 6 Proportions of x, 2x and 3x pollen in East African Highland and diploid bananas…”

“…Very little experimental information is available about the actual male fertility of the EAHB. Other studies showed that pollen output and viability varied among banana cultivars and also along the rachis (Sathiamoorthy 1973;Sathiamoorthy et al 1984;Krishnakumar et al 1992;Fortescue and Turner 2004).…”