Genetic Variation, Clonal Diversity And Evolution Of Tristylous Breeding System In Eichhornia Crassipes, An Invasive Clonal Plant In China

Understanding the breeding system and population genetic structure of invasive weed species iscrucial for invasion management, and contributes to our understanding of the evolutionary processes ofplant breeding system. In this paper, we compared the genetic variation and clonal diversity & structureof Eichhornia crassipes (Mart.) Solms populations throughout China by randomly amplifiedpolymorphic DNA (RAPD) fingerprinting to detect the relevance of sexual and clonal reproduction inthis species and its contributions for the plant's success as an invader. One population from Kenya(Aferica) was also studied on its genetic variation and the possible genetic relationship betweenChinese populations. The between-genotype variation in clonal growth was also experimental studiedand its influence on the plant's invasion mechanism was disccussed.The breeding system of E. crassipes is tristyly in the native region, while in the invaded places, allthe populations contain only one or two styly forms. In china, most of the populations throughtoutChina were monomorphic for M morph, only the populations in Chongqing city (in West China, on theYangtze River) and Nanning city (In southwest China, near tropical zone) contained L morph, with Mmorph still predominating the populations. Some populations were found with flowers withsemi-homostyly (modified M flower, with one anthers of upper set stamens adjacenting to stigma).Semi-homostyly was thought to be a derived form of M morph, and was selected in the invaded regionssince it could increase fecundity through autogamy duringthe periods of unreliable and ineffientpollination service.34 populations were studied to detect genetic variation by using RAPD markers. A total of 1 009individuals were analyzed, for which 12 RAPD primers amplified 69 reproducible bands with 22polymorphic (31.88%). The percentage of polymorphic bands (PPB ) within a population ranged from4.35% to 17.39%, and mean Nei's gene diversity (He) was 0.0461±0.0145, indicating a low geneticdiversity of E. crassipes in China. Each population contained more than 4 RAPD phenotypes(genotypes), and was dominated by the same genotype. Analysis of molecular variance (AMOVA)revealed a large proportion of genetic variation (83.93%) resided within the populations (only 16.07%among populations), and a greater majority of genetic variation (88.11%) resided within localities, onlya small amount of variation (11.89%) presented differences among localities. These results suggestedlittle genetic divergence among populations, both locally and regionally, in China. Human-mediateddispersal, vivid clonal growth and a generally low level of sexual reproduction in E. crassipespopulations were thought responsible for such a pattern of genetic structure. The population fromKenya (Africa) showed a relatively higher genetic variation (He = 0.0735±0.1590,PPB = 18.84%),and a much larger genetic divergence from China's populations.UPGMA dendrogram was produced to illuminate the genetic relation of populations studied. Thereis no significant relation between geographic distance and genetic distance (Mantel test, p = 0.3390),and genetic variation (He) does not differ significantly (Mann-Whitney test, p>0.3) between still waterhabitat (lakes and ponds) and flowing-water habitat (rivers and streams). Human–mediated dispersalover a long distance were believed to be the main causes for this phenomenon.Clonal diversity was low in this species with only 5 genets were detected from 209 individualsusing 5 RAPD primers. The five populations studied were all composed of multiple, highly intermixedclones, suggesting this plant has a guerilla-type clonal growth.Results based on a primary experiment showed that individuals with M floral morph (M genotype)had a relatively quicker clonal growth rate and a greater clonal potential (more ramets produced by Mindividuals) than that of L genotype. These between-genotype differences in clonal growth werethought to be a causes for the M predomination in E. crassipes populations all over the world, andconsequently possess a effect on the invasion of this noxious clonal plant.