Cytological Conditions and Evidences for Hybridity in North American Wild Roses
1929; University of Chicago Press; Volume: 87; Issue: 4 Linguagem: Inglês
10.1086/333956
ISSN1940-1205
Autores Tópico(s)Plant Pathogens and Resistance
Resumo1. In this study 107 wild roses from known stations in North America were examined cytologically. The majority of the specimens fall into the three classes, diploid, tetraploid, and hexaploid, seven being the basic number. One specimen was octoploid, two were triploid, and three aneuploid with the somatic number not a multiple of seven. 2. Measurements show that the cells are larger in the polyploid forms. 3. The diploid American roses are robust plants, larger as a rule than the related polyploid species. Many of them exhibit irregularities during meiosis. In some individuals incomplete pairing at diakinesis, lagging chromosomes, and polyspory are frequent; these are believed to be spontaneous hybrids produced by the crossing of different species, or the descendants of such hybrids. 4. It is concluded that crossing occurs in the field between different diploid species in Rosa, giving rise to at least partially fertile diploid offspring. 5. Three fertile aneuploid plants, two with sixteen somatic chromosomes and one with fifteen, have been found among cultures of diploid plants. One such culture exhibits a great deal of variation and appears to be a later generation of a cross between the diploid R. blanda and the hexaploid R. acicularis of Michigan. Such a cross would give, in the first generation, tetraploid hybrids with seven pairs and fourteen unpaired chromosomes. During meiosis in these hybrids the unpaired chromosomes probably lag on the spindle, giving some fertile gametes with only seven or eight chromosomes, which by self-fertilization give rise to fertile diploid or aneuploid plants. 6. Two triploid plants have been examined cytologically. Both have seven paired and seven unpaired chromosomes at first reduction division. One exhibits the "Drosera-type" of meiosis and is sterile. The other has the "Rosa-type" of meiosis, in which the seven univalent chromosomes divide twice, making it possible for some spores to receive fourteen chromosomes, although this is prevented in many cells by lagging on the spindle. This latter plant is not apomictical yet it sets good fruit. Reduction in the megasporocyte of the fertile triploid has not been observed. The plant may be a starting point for a race of triploid roses analogous cytologically to the pentaploid Caninae, or it is able to give either tetraploid or diploid offspring by self-fertilization. A third plant with over 90 per cent of the pollen sterile is suspected of being a triploid hybrid from R. palustris x R. virginiana. 7. R. californica and related species were found all to be tetraploid with fourteen pairs at diakinesis. 8. The tetraploid species examined cytologically nearly all exhibit irregularities during meiosis and also polyspory. They thus resemble hybrid forms in their cytological behavior. 9. Both tetraploid and diploid races morphologically similar were found in R. subserrulata Rydb. 10. The tetraploid forms often exhibit some rings of four chromosomes at diakinesis, but none have been found that are completely quadrivalent. It is concluded that they originated by duplication of diploid hybrids and have hybridized with each other. 11. The tetraploids are characterized by being able to flower on the season's turions and by a long flowering period. Many of them are semi-herbaceous or dwarf, and are believed to be recent or derived forms adapted to prairie and dry upland habitats. 12. R. nutkana and related species, and all but one of the specimens of R. acicularis and its varieties, were found to be hexaploid, with twenty-one pairs at diakinesis. 13. Hexaploids exhibit few meiotic irregularities. 14. All hexaploids in America apparently occupy one continuous range, and it is thought that they have not originated on this continent from a sterile triploid hybrid by duplication, but that they migrated from Asia as hexaploids spreading southward and eastward. 15. There are two parallel series of American Cinnamomeae, each consisting of hexaploid, diploid, and tetraploid types: (1) on the Pacific Coast R. nutkana, R. pisocarpa, and R. californica; (2) in the Mississippi Valley R. acicularis, R. blanda, and R. suffulta. 16. A specimen of R. acicularis from Alaska was found to be octoploid with twenty-eight pairs at diakinesis. The plant does not differ markedly from the hexaploid variety lacorum, but is distinguished by coming into flower earlier. It has pure pollen. 17. Examination showed that the hexaploids have a small proportion (under 10 per cent) of sterile pollen. Some diploid specimens have almost entirely good pollen while others have 10-15 per cent sterile grains. Most of the tetraploids showed 15-30 per cent sterile pollen, although individuals of R. carolina and of R. virginiana were found with 90 per cent of their pollen good. One tetraploid had 56-77 per cent of the pollen bad. 18. Good fruit was obtained after castration in one case only, that of a specimen of R. blanda (diploid) from Illinois which had 52 per cent bad pollen. 19. The flowering times of each of the large species groups are specific and characteristic when taken in relation to each other. The phenology of the species of American Cinnamomeae seems to coincide with phylogenetic relationship, those coming into flower first being the most primitive. The flowering order is: octoploids, hexaploids, diploids, tetraploids. 20. Cytological and morphological evidence support the theory of Harrison and Blackburn of the hybrid origin of orthoploid series in Rosa. A possible origin of the North American polyploid series is worked out from this premise. 21. The presence of octoploid and hexaploid races of R. acicularis in America, and also of diploid and tetraploid races of R. subserrulata, is suggested as evidence in support of Hurst's theory. 22. Cytological observations on American roses do not support Hurst's theory of differential septets in Rosa which would require that sets of seven chromosomes separate together during meiosis. Extrusion is thought to be due partly to imperfect fixation and partly to disturbances consequent upon hybridization.
Referência(s)