The Improvement Effect Of Different Recurrent Selection Methods On Narrow Base Maize Populations

China is the second largest country in maize production all over the world. The maize breeding was relatively advanced, but also the germplasm resources were relatively poor in china. The lack of maize germplasm result in the narrow basis, it was the main factor that block the development of maize breeding. This fact requires us to focus on the expansion and improvement and innovation of germplasm. The recurrent selection is the basic method for the improvement and expansion of maize germplasm. It could break the chain linkage between genes effectively and increase the superior gene frequency continuously. These populations with the performance improved then can serve as a potential source of superior inbred lines and can inhabit development of a possible genetic ceiling for further hybrid improvement. In this study, two synthetic narrow-base maize populations P3 and P4 were improved by 5 cycles of biparental mass selection method to obtain 12 populations. In different ecological environment, the phenotypes of the populations per se were analysis and the combing ability was tested according to an incomplete diallel model to study the effect of MS on different populations. The narrow base population P4 was improved by different recurrent selection method to obtain 11 populations. In different ecological environment, the phenotypes of the populations per se were analysis and the combing ability was tested according to an incomplete diallel model to study the effect of different recurrent selection method on a same population, also the effect of different recurrent selection methods on genetic diversity of populations were analysis by SSR markers with the objective to make suggestions on how to raise the efficiency of improvement and utilization of synthetic populations. Three inbred lines derived from each of the 6 populations, including P4C0, P4MS2, P4HSC1, P5C0, P5MSC2 and P5HSC1, to make a total of 18 lines. The phenotypes of the lines per se were analysis, and the combing ability was tested according to an incomplete diallel model, the genetic diversity of the lines were analysis by SSR markers with the objective to make suggestions on the usage of these lines and generating strategies for deriving lines from a improved population. The main results were summarized as follows: 1. After 5 cycles of biparental mass selection (MS), for both of the base population P3C0 and P4C0, most traits of the populations per se and the GCA were improved significantly, whereas the SCA of most traits showed no significant increase. The mean of plant height and ear height per se and their GCA increased with the advance of selection for the both base populations. For P3C0 and its improved descendants, the mean of kernel depth and rows per ear were smallest at P3C0 and largest at P3MSC5. The GCA effects of them were smallest at P3C0, and largest at P3MSC2 and P3MSC5, respectively. The yield per plant of populations per se and its GCA effects were both smallest at P3C0, and largest at P3MSC4 and P3MSC2, respectively. For P4C0 and its improved descendants, the mean of kernel depth and rows per ear were smallest at P4C0, and largest at P4MSC5 and P4MSC4, respectively. The GCA effects of them were both smallest at P4C0, and largest at P4MSC4. The yield per plant of the populations per se and its GCA effects were both smallest at P4C0, and largest at P4MSC2 and P4MSC4, respectively. For most of the other traits per se and their GCA effects, the improved descendants were larger than their respective CO. There were different effects of improvement on different populations and traits at different cycles. The general trend showed that after cycles of biparental mass selection, some traits of populations per se and their GCA were improved simultaneously, while the improvement of population per se and the GCA did not display a synchronicity for some others. After the largest selection response to biparental mass selection was obtained, a sustained biparental mass selection would lead to a decrease in genetic gain or even a negative genetic gain.2. The effect of different recurrent selection method on the main traits and their GCA were different. After 5 cycles of MS, the mean of plant height and ear height per se and their GCA increased with the advance of selection for P4C0,most traits of the populations per se and the GCA were improved significantly, and the genetic gain were obtain in first 2 cycles, the genetic gain decrease from the 3rd cycle. After 1 cycles of HS-S2:3 selection, most traits of the populations per se and the GCA were improved significantly. After 3 cycles of MS, most traits of the population P4HSC1 per se and the GCA were improved further. After add the inbred lines to the population P4HSC1, most traits per se and their GCA display a significant increase over P4HSC1.In time, the effect on yield and the yield component and their GCA were better than HS-S2:3 selection, but the HS-S2:3 selection was more beneficial to maintain the plant and ear height of population. In turn, adding lines to the population were more effective on the improvement of yield and the yield component and their GCA than MS, but also the plant and ear height increase more.3. After five cycles of MS, the excellent individual frequency increase with the advance of selection with a wave phenomenon for both P3C0 and P4C0.After 1 cycle of HS-S2:3 selection, the excellent individual frequency increase over P4C0. After add the inbred lines to the population P4HSC1, the he excellent individual frequency increase. After 3 cycles of bi-parents mass selection, the excellent individual frequency increase with the advance of selection with a wave phenomenon.4. The effects of different recurrent selection methods on the genetic diversity were different. After 5 cycles of MS on P4C0, the number and the percentage of polymorphic loci displayed a descending trend with the advance of selection, the mean genetic distance represent slightly fluctuate declining. The expected heterozygosity (He) and Shannon genetic diversity index (I) increased in the first two cycles while decreased from the 3rd cycle with the advance of selection. The number of genotype presented the trend of declining with the advance of selection and showed an alternative growth and descends phenomenon related to the location where the populations were developed. The 6 index of genetic diversity decreased after HS-S3, after 3 cycles MS were carried out on P4HSC1 or adding 2 inbred lines into P4HSC1, the genetic diversity of population increase in varying degrees.5. In our study, the number of genotype presented the trend of declined with the selection process. In additional, the number of genotype appears to be related to the location where the populations were developed. When the population was developed in Yunnan province, the number of genotype increased while decreased when the population was developed in Sichuan province.6. The 11 populations devlopeped from P4 by different recurrent selection method were classified into 5 types based on the method the population were improved. The 1st type included P4C0, the 2nd type included P4MSC1, P4MSC2, P4MSC3, P4MSC4 and P4MSC5, the 3rd type included P4HSC1, the 4th type included P4HSC1-AP, the 5th type included P4HSC1-MSC1, P4HSC1-MSC2 and P4HSC1-MSC3.The AMOVA showed that the genetic various among regions, among populations and within populations were 5%, 6% and 89%, respectively. The genetic diversity changed with the advance of selection, while the genetic diversity within populations was much more than that of among populations. The result of principle coordinates analysis showed that the individual distribute in a larger range in the P4C0 and early cycles of MS while the distribution range become smaller in the advance cycles of MS and the P4HSC1 and the populations developed from P4HSC1 by MS and. The directional changes occurred in the distribution of the individual with the advance of selection, but degrees and direction vary from method to method.7. The inbred lines derived from different population, and the inbred lines derived from different basic individual in a same population vary in the phenotype and combing ability.8. The inbred lines P4MSC2-1, P4MSC2-2, P5MSC2-2 and P5MSC2-3 derived from improved population were superior in yield and yield component and their GCA, the hybrid between these lines and the testers were high in yield. In summary, most of the inbred lines derived from the improved population were better than the inbred lines derived from the base population in yield and yield component and their GCA. However the inbred lines derived from the improved population were taller in the plant height and ear height.9. The number and percentage of homozygote of the lines derived from the improved populations were roughly equivalent to that of the lines derived from the basic populations with one inbred generation less. The genetic diversity of the lines derived from the improved population showed a decrease trend, but the genetic diversity was still rich relatively.