Impact of self-pollination on the genetic diversity of inbred families of Psidium guajava']java L

In guava (Psidium guajava), the impact of self-pollination on the quantitative traits of the fruits is not fully understood, necessitating further investigation. This study aimed to estimate the effects of selfing on fruit traits in S-1 and S-2 inbred families of guava and to explore potential impacts on genetic diversity. Eighteen S-1 families were generated through selfing of progenies from biparental crosses, and ten S-2 families were produced by selfing superior genotypes from S-1 families. The experiment was conducted at the Experimental Station of Ilha Barra do Pomba, in the municipality of Itaocara-Rio de Janeiro, Brazil. It utilized a randomized complete block design, with three replications and ten plants per plot. Evaluated traits included fruit weight, length, diameter, length-diameter ratio, endocarp thickness, mesocarp thickness, pulp weight, and soluble solids content. The data underwent individual analysis of variance, yielding predicted mean trait values for S-1 and S-2 generations, alongside correlation and homozygosity estimates. Genetic diversity was assessed using Mahalanobis distance and UPGMA cluster analysis, and comparative box plots between inbred populations were created for the evaluated traits. Box plot analysis revealed symmetry in most evaluated traits, suggesting uniformity in the data due to the selfing strategy. Analysis of variance indicated statistically significant differences in all traits, highlighting variability between populations S-1 and S-2. Fruit and pulp weights exhibited high homozygosity levels, with values of 90.86 and 102.59 respectively, linked to increased fruit traits in the S-2 population, indicating their importance in the fixation of favorable alleles. Fruit weight, length, and diameter, endocarp thickness, and mesocarp thickness showed strong correlations, exceeding 0.70. Genetic diversity assessment via Mahalanobis distance indicated a decrease in genetic variability, evidenced by fewer groups in S-2 compared to the S-1 population. However, this reduction did not noticeably affect the average performance of the S-2 population. The results indicate that the two generations of self-pollination did not negatively affect the phenotypic values of the evaluated traits.