【正文】 s work were to estimate heritabilities for isoflavone contents in soybean seeds and the correlation between isoflavone and protein contents. Commercial variety IAC100 (high isoflavone and normal protein contents) and the line BARC8 (low isoflavone and high protein contents) were crossed, and one single F1 plant derived 97 F2 seeds, which were used to obtain F3 seeds. A sample of F3 seeds from each F2 plant was used for isoflavone determination by HPLC and protein by the Kjeldahl method. Six isoflavone forms were detected: daidzin, genistin, glycitin, malonyldaidzin, malonylgenistin and malonylglycitin. Total isoflavone contents ranged from to μg per gram of dry seed and the protein content ranged from to % in BARC8 and IAC100, respectively. Our results indicate that it is possible to select for high isoflavone content in early breeding generations because the broad sense heritabilities for the contents of the various isoflavone forms were higher than 90%. In addition, high correlation values among the contents of the individual isoflavone forms were observed (between and ). However, negative correlation values were obtained between isoflavone and protein contents, ranging from – to – for the different isoflavone forms. The correlation value of – between total isoflavone and protein contents confirmed the negative correlation between these two parameters, as reported by other authors.——▼. Maughan1。_, . Saghai Maroof amp。 . Buss，Identification of quantitative trait loci controlling sucrose content in soybean (Glycine max)，Molecular Breeding 6: 105–111, 2000Sucrose is a primary constituent of soybean (Glycine max) seed。 however, little information concerning the inheritance of seed sucrose in soybean is available. The objective of this research was to use molecular markers to identify genomic regions significantly associated with quantitative trait loci (QTL) controlling sucrose content in a segregating F2 population. DNA samples from 149 F2 individuals were analyzed with 178 polymorphic genetic markers, including RFLPs, SSRs, and RAPDs. Sucrose content was measured on seed harvested from each of 149 F2V3 lines from replicated field experiments in 1993 and 1995. Seventeen marker loci, mapping to seven different genomic regions, were significantly associated with sucrose variation at P 0:01. Individually, these markers explained from % to % of the total phenotypic variation for sucrose content in this population. In a bined analysis these genomic regions。 explained 53% of total variation for sucrose content. No significant evidence of epistasis(上位性) among QTLs was observed. Comparison of our QTL mapping results for sucrose content and those previously reported for protein and oil content (the other major seed constituents in soybean), suggests that seed quality traits are inherited as clusters of linked loci or that ‘major’ QTLs with pleiotropic effects (多效性效應(yīng)) may control all three traits. Of the seven genomic regions having significant effects on sucrose content, three were associated with significant variation for protein content and three were significantly associated with oil content.——▼Hany A. ElShemy1,2, Masayoshi Teraishi1, Mutasim M. Khalafalla et al. Isolation of soybean plants with stable transgene expression by visual selection based on green fluorescent protein, Molecular Breeding 14: 227–238, 2004.Particle bombardment is a mon platform for soybean transformation but tends to cause transgene silencing due to the integration of rearranged or multiple copies of transgenes. We now describe the isolation of a total of 44 independent transgenic soybean plants after transformation by particle bombardment with one of two gene constructs, pHV and pHVS. Both constructs contain the hygromycin phosphotransferase gene （hpt）as a selectable marker and a modified glycinin gene (V31) for evaluation of homologydependent silencing of endogenous glycinin genes。 pHVS also contains sGFP(S65T), which encodes a modified form of green fluorescent protein (GFP), as a reporter gene in the flanking region of V31. Fluorescence microscopy revealed that the leaves of 8 of the 25 independent transgenic plants obtained with pHVS expressed GFP。 most of these GFPpositive plants also contained V31 mRNA and an increased glycinin content in their seeds, and they exhibited simple banding patterns on Southern blots that were indicative of a low copy number of each of the three transgenes. In contrast, most of the transgenic plants obtained with pHVS that did not express GFP, as well as most of those obtained with pHV, lacked endogenous glycinin in their seeds and exhibited more plex patterns of transgene integration. The use of a reporter gene such as sGFP(S65T) in addition to an antibiotic resistance gene may thus help to reduce the problem of gene silencing associated with direct DNA transformation systems and facilitate the recovery of transgenic plants that stably express the gene of interest.