A cytological characterization of genomes of Alstroemeria, the production of interspecific hybrids, and their performance during micropropagation.

The vegetatively propagated ornamental Alstroemeria has become highly popular in a relatively short period. During the last 40-50 years botanists and breeders realized that the genus Alstroemeria has tremendous potential as a cut flower crop, a bedding plant and a potted plant. They also became aware of the limited knowledge regarding the species and the interspecific hybrids that in several cases occurred spontaneously in their nurseries. There was a growing need for species descriptions and background information on their natural habitats. During the last 25 years, and especially during the last ten years, much information became available on the cultivation methods, tissue culture techniques and on the application of mutation breeding. Biosystematic studies and cytological investigations on chromosomes were, however, scarce in comparison to other major cut flowers.In the biosystematic work on Alstroemeria frequent use was made of characteristics of the flowers. The underground organs, i.e. the rhizome and the root system, have not been described extensively. With the assumption that the morphological features of the rhizomes may affect micropropagation, the underground organs of several species and hybrids were studied. This investigation resulted in descriptions and illustrations of the general morphological structure of the rhizome, and of the underground organs of nine species and three hybrids in particular. The rhizomes varied considerably in length, diameter, internodal length, and the number of lateral rhizomes and visible axillary buds, giving more substance to the idea that the different species and hybrids respond differently when propagated in vitro .A large part of the research that is described in this thesis concerned the chromosomes and genomes of Alstroemeria species and hybrids. The chromosomes of eight species were analysed after staining with Feulgen's reagent. The morphology of the chromosomes, i.e. the length, arm length ratio and the secondary constrictions, was clearly visible and chromosome arms could be measured. The application of Giemsa C-banding displayed unique banding patterns on the chromosomes of each of the eight species that were investigated. These banding patterns have cytotaxonomic value, and, once a larger number of species and species accessions has been studied, can throw light on the evolutionary events that occurred within the genus. The length of the chromosome complement already indicated that the species of Alstroemeria possess large genomes.This finding was confirmed through flow cytometric analysis of nuclear DNA amounts. The amount of nuclear DNA in the diploid cells (2C-values), as determined through flow cytometric measurement of the relative fluorescence of isolated nuclei that were stained with propidium iodide (PI), ranged from 36.5 to 78.9 pg among the accessions of 12 species. When nuclei were stained with DAPI, a fluorescent dye with different staining properties, different values were obtained. The ratio of PI and DAPI fluorescence varied from 1.60 to 1.88. The Brazilian species had lower PI/DAPI ratios (1.60-1.67) than the Chilean species (1.68-1.88). The 2C-values together with the PI/DAPI ratio of the twelve species enabled the separation of species into four groups. These groups were (1) A. magnifica ssp. magnifica , A. pelegrina , A. philippii and A. pulchra ssp. pulchra , (2) A. angustifolia ssp. angustifolia , A. aurea and A. hookeri ssp. hookeri , (3) A. ligtu ssp. ligtu and A. ligtu ssp. simsii , and (4) A. brasiliensis , A. caryophyllaea , A. inodora and A. psittacina .There were remarkable karyotype similarities within each of these groups, with regard to total chromosome length and C-banding pattern. A survey of the extent of intraspecific variation in genome size of three Chilean species revealed up to 1.21 fold variation in A. ligtu . The variation in genome size could be attributed to C-band and chromosome length polymorphism. The intraspecific variation in A. magnifica ssp. magnifica was discontinuous, and might be due to a broad taxonomic species concept of this particular taxon [as, according to Bayer (1987), the taxa A. gayana and A. sierrae are conspecific with A. magnifica ssp. magnifica ]. Giemsa C-banding and flow cytometry could both be used to check the hybrid nature of the plants that were produced in the interspecific hybridization programmes. Individual chromosomes of the parental species could be recognized in the C-banded karyotypes of most hybrids, and 2C-values as determined through flow cytometry were intermediate between those of the two parents. With the technique of flow cytometry it was possible to distinguish between aneuploid (2n=4x+1=33) and euploid (2n=4x=32) plants.In order to obtain well defined plant material, species were collected from breeders, botanical gardens and research institutes, and interspecific crosses were made between five species, that are often used in the development of cultivars. These were the Chilean species A. aurea , A. pelegrina and A. magnifica ssp. magnifica and the Brazilian species A. inodora and A. psittacina . Seed set in interspecific crosses, however, was poor. Histological observations of the fertilized ovules revealed a poor development of the endosperm, and a reasonable to good development of the embryo during the first two to three weeks after pollination. Through in vitro culture of dissected ovules, it was possible to produce more than 250 hybrids.Conventionally, Alstroemeria plants are vegetatively propagated by division of greenhouse grown rhizomes, usually once a year. The multiplication rate with this practice of propagation is rather low (about two to six new plants), because it is restricted by the number of lateral rhizomes that are present on the mother plant. The development of micropropagation methods, using in vitro grown rhizomes, has enabled the mass propagation of plant material. However, the rhizome multiplication rate of hybrids is mostly unpredictable, and for groups of cultivars it is extremely low. An analysis of the rhizome multiplication rate of defined plant material, consisting of species and their interspecific hybrids, demonstrated that there is a genetic base for the performance during micropropagation. Thus, it is worthwhile to incorporate rhizome multiplication rate as a selection criterion in Alstroemeria breeding programmes. The choice of parental species and genotypes, to be used in interspecific crosses, is crucial. Because of the high degree of heterozygosity of the parental genotypes, selection among full sibs can be profitable. Morphological features of the rhizomes, such as the number of rhizome nodes, the degree of branching, and internodal length, may serve as criteria for indirect selection. It is suspected that the complementation of these morphological traits in the parental genotypes might lead to a superior micropropagation performance of the hybrid.The research that is described in this thesis has contributed by providing defined plant material and basic information on species and hybrids of Alstroemeria . The results are beneficial for the breeding of new cultivars, and for further applied and fundamental research in the genus Alstroemeria