JARS v37n4 - Notes on the Pollen of Deciduous Azalea Cultivars

Notes on the Pollen of Deciduous Azalea Cultivars
Mark P. Widrlechner and Harold M. Pellett 1 , St. Paul, Minnesota

Scientific Journal Series Paper No. 13,552, Minnesota Agricultural Experiment Station.

Introduction
In the Summer 1982 issue of the Journal , we described a breeding method to take advantage of the occurrence of unreduced gametes in deciduous azalea hybrids (Widrlechner et al., 1982). We suggested that unreduced gametes might preserve desirable combinations of parental genes and that progeny from unreduced male gametes could be obtained by hybridizing the unreduced gamete source to species with twice as many somatic chromosomes. This method was proposed to preserve hybrid vigor and to bypass the failure of "normal" crosses due to pollen inviability.
The efficient use of unreduced gametes in a breeding program relies on the breeder choosing parents that produce unreduced gametes at moderate frequencies. Microscopic examination of pollen allows for rapid screening of potential parents. Unfortunately, a breeder may not have easy access to a microscope and stains.
When the breeding method was proposed, the frequency of unreduced gametes had only been studied in Rhododendron x kosterianum Schneider x R. prinophyllum (Small) Millais populations. Since that time, a review of the literature (Widrlechner et al., 1982a) has found circumstantial evidence for unreduced gamete production in R. occidentale (Torr. & Gray) A. Gray (Sax, 1930), R. alabamense Rehd. (Willingham, 1976), evergreen azaleas (Heursel and DeRo, 1981; Pryor and Frazier, 1970), and R. augustinii Hemsl. (Hardgrove, 1966). Few breeders of deciduous azaleas are working with R. x kosterianum x R. prinophyllum populations, but many breeders use named cultivars, especially Exbury and related groups, as sources of large flower size and bright flower colors.
The present study was undertaken to survey the pollen characteristic' of deciduous azalea cultivars of the Knaphill and Exbury groups. Most cultivars examined are available commercially and many are used widely in breeding. We hope that our results contribute to the success of future crosses, especially with regard to the success of hybrids between these cultivars and tetraploids, such as R. calendulaceum (Michx.) Torr.

Methods
During the Spring of 1983, anthers were collected from deciduous azalea plants, grown either at the Minnesota Landscape Arboretum, Chanhassen, MN or on the St. Paul campus of the University of Minnesota. An alphabetical list of the cultivars examined is given in Table 1. Pollen was stained with 3- (4,5-dimethyl-thiazolyl-2)-2, 5-diphenyl tetrazolium bromide, a vital stain shown to correlate significantly with in vivo germination (Widrlechner et al., 1983), and was examined microscopically using procedures described by Widrlechner et al., (1982).
The numbers of tetrads (normal pollen grains), dyads (possible unreduced gametes), and irregular sporads (monads, shrunken and undeveloped pollen grains) were noted from each cultivar. Counts were based on samples of 250 sporads, but larger numbers of pollen grains were examined to confirm that samples were representative of the plant.

Table 1 Cultivars Examined
Minnesota Landscape
Name Group Arboretum Accession Number
'Byron Mayo' Knaphill 790813
'Bullfinch' Knaphill 570019
'Cream Puff' Madison 780084
'Gibraltar' Exbury 620551 A
'High Sierras' Knaphill 790808
'Holiday' Unknown 800824
'Hotspur Orange' Exbury 790809
'Ilam Red Letter' Ilam 780099
'Ilam Whitecap' Ilam 780091
'June Bride' Madison 780083
K-10 Knaphill 780094
K-38 Knaphill 790805
K-47 Knaphill 790806
'Maori' Ilam 790807
'Marion Merriman' Knaphill 780096
'Orangeade' Knaphill 620550A
'Orient' Exbury 780098
'Oxydol' Exbury 780092
R. arborescens 790989
R. prinophyllum --
'Royal Lodge' Exbury 780100
'Suez' Knaphill 780097
Unknown Yellow Exbury Exbury --
Unnamed Mollis Hybrid Mollis --
Unnamed Orange Exbury Exbury 600267 H
'Whitethroat' Knaphill 780093

Results and Discussion
The pollen characteristics of the plants examined were quite variable. This might be expected because of the diverse parentage of the cultivars. Street (1959) reviewed the development of modern deciduous azalea hybrids and noted that at least seven species, including the tetraploid, R. calendulaceum , contributed to the parentage of modern hybrids. It is thus possible that the cultivars are not all of the same chromosome number.
We have found that the cultivars examined can be divided into four basic groups based on pollen characteristics (Table 2). Group 1 includes cultivars and species with essentially normal pollen, highly fertile, and over 90% tetrads. The unnamed mollis hybrid and the species, R. prinophyllum and R. arborescens (Pursh) Torr., are all in this group, but only 1/3 of the cultivars are included. 'Whitethroat' is included in Group 1 on the basis of pollen type, but the anthers of this plant are not well-formed and dehiscence is poor.
Group 2 can be distinguished from Group 1 by less regular pollen formation and lower pollen viability. The majority of the Exbury and Knaphill hybrids that we sampled produce between 60 and 90% tetrads and a significant proportion of dyads and other irregular sporads. For breeding purposes, we have divided Group 2 into two subgroups based on the proportion of dyads. Those cultivars producing dyads at frequencies greater than 10% (Group 2a) would be promising candidates for the use of unreduced gamete breeding methods (Widrlechner et al., 1982).
Group 3 includes only the cultivars 'Gibraltar' and 'High Sierras'. The pollen of these two plants is quite irregular and most likely is the result of major meiotic abnormalities. These two cultivars produce dyads at fairly high frequencies, thus making them good candidates for unreduced gamete breeding methods.
Group 4 is characterized by a total breakdown of normal pollen structure. Poorly developed sporads with varying numbers of microspores predominate. Minnesota Landscape Arboretum Accession Number (MLAA) 800096, a hybrid between R. calendulaceum and an Exbury hybrid, has already been described as having a Group 4 pollen type (Widrlechner et al., 1983a). The Knaphill selection, 'Byron Mayo', also fits this category. Crosses using Group 4 plants as pollen parents may give unpredictable results. We have found that progeny from MLAA 800096 are extremely variable in growth rate and it is possible that the sporads from Group 4 plants contain microspores with various chromosome numbers, thus producing aberrant progeny.

Table 2  Pollen Characteristics of Cultivars Examined
Group Cultivar Tetrads Dyads Irregular Sporads
1 R. prinophyllum 244 4 2
'Holiday' 243 0 7
Unnamed Mollis Hybrid 242 5 3
'Cream Puff' 241 4 5
'June Bride' 237 6 7
'Whitethroat' 237 10 3
R. arborescens 234 5 11
'Hotspur Orange' 234 5 11
Unknown Yellow Exbury 233 9 8
'Orient' 228 3 19
2 'Maori' 216 11 23
'Marion Merriman' 215 20 15
'Bullfinch' 213 18 19
K-47 207 22 21
'Suez' 204 18 26
2a K-38 206 26 18
'Royal Lodge' 202 33 15
'Ilam Red Letter' 201 33 16
'Orangeade' 200 41 9
K-10 190 29 31
'Oxydol' 187 53 10
'Ilam Whitecap' 184 32 34
Unnamed Orange Exbury 170 55 25
3 'Gibraltar' 104 81 65
'High Sierras' 1 155 94
4 'Byron Mayo' 38 49 163

Conclusion
A survey of the pollen characteristics of popular deciduous azalea cultivars has found more variability than would be expected for species in nature. This variability probably is a result of genetic and cytological instabilities in these complex, interspecific hybrids. We have found that many of the cultivars produce dyads at frequencies high enough to allow their use in crosses with tetraploids, such as R. calendulaceum . It may even be that some of the cultivars with flame azalea characteristics are tetraploid and have resulted from past matings involving unreduced gametes.
The common occurrence of pollen irregularities in this group of cultivars may help explain unusual results in breeding programs, such as success in crossing species of different chromosome numbers, poor seed set from particular pollen parents, and unusually variable progenies. We hope that our work will encourage breeders to take a closer look at their pollen parents and will help breeders more efficiently plan future crosses.

Acknowledgments
We sincerely wish to thank Drs. Peter Ascher, Florian Lauer, and James Luby for their helpful comments and advice in the preparation of this manuscript and Ms. Susan Moe for her help in collecting pollen samples.

Literature Cited
Hardgrove, D.L. 1966. Hybridizing experiences and recommendations. Q. Bull. Amer. Rhod. Soc. 20: 202-217.
Heursel, J. and R. De Roo. 1981. Polyploidy in evergreen azaleas. HortSci. 16:765-766.
Pryor, R.L. and LC. Frazier. 1970. Triploid azaleas of the Belgian-Indian series. HortSci. 5: 114-115.
Sax, K. 1930. Chromosome stability in the genus Rhododendron . Amer. J. Bot. 17: 247-251.
Street, F. 1959. Azaleas. London: Cassell & Co.
Widrlechner, M. P., H. M. Pellett, and P. D. Ascher. 1982. Unreduced gametes in azalea hybrids: a possible breeding method for using promising azaleas of low fertility. J. Amer. Rhod. Soc. 36: 98-100.
_____. 1982a. Progress in the study of the breeding and genetics of the genus Rhododendron L. In. Studies on the breeding potential and genetics of hybrid azalea. Rhododendron x kosterianum Schneider x Rhododendron prinophyllum

1 Former Research Assistant and Professor, Department of Horticultural Science & Landscape Architecture, University of Minnesota, St. Paul, MN.