RELATIONSHIP BETWEEN SOYBEAN INDUSTRIAL-NUTRITIONAL QUALITY AND THE ASSIMILATE SOURCE UNDER HEAT AND WATER STRESS DURING SEED FILLING CARRERA, C.S. 1,2 ; CARRÃO PANIZZI, M.C. 3 ; GONTIJO MANDARINO, J.M. 4 ; LEITE, R.S. 4 ; ERGO, V.V 5 ; ANDRADE F. 2,6 ; PAROLA R. 1 ; LASCANO R.H. 2 ; VEGA C.R.C. 5 1 IFRGV-CIAP-INTA (X5020ICA) Córdoba, Argentina, carrera.constanza@inta.gob.ar; 2 CONICET; 3 Embrapa Trigo; 4 Embrapa Soja; 5 INTA Manfredi; 6 UNMdP-INTA Balcarce In crops, combination of heat stress (HS) with water stress (WS) modifies photosynthetic processes (RIZHYSKY et al., 2004) causing additive or multiplicative effects (PRASAD, 2004) that modify assimilate supply to seeds. The exposure of soybean to both abiotic stresses during the filling also affects final seed quality (DORNBOS & MULLEN, 1992; CARRERA et al., 2009). To our knowledge, quantitative relationships between seed chemical quality and photosynthetic markers describing the assimilate source are lacking for soybeans grown under heat and water stress during seed filling. An experiment was conducted in the EEA INTA Manfredi (31º 49'S, 63º 46' W) during the 2012-2014 crop seasons using two soybean cultivars (SPS4x4 and SPS4x99). The experimental design was a split-split plot with 2 replications, resulting in a three factorial arrangement: water level, genotype, and temperature level. Water levels were: i) non water stress (NWS), near field capacity, which was achieved by drip irrigation, and ii) water stress (WS), approximately 20% of available water content during 35 days from growth stage R5.5 (FEHR & CAVINESS, 1977). Temperature levels were: i) non heat stress (NHS), at environmental temperature (ET), and ii) heat stress (HS), comprising brief periods of exposure to temperature >32°C for 6 hours per day, during 21 days from R5.5. Field recorded variables were: quantum yield of photosystem II (ᶲPSII) and photochemical efficiency of PSII (Fv/Fm) measured with a modulated pulse meter (Hansatech, FMS2 model); canopy temperature (CT) at 12:30 and 14.30h measured with a hand held infra-red thermometer (Testo 845, Spain); leaf chlorophyll levels (estimated with a SPAD chlorophyll meter, Minolta SPAD-502); and leaf relative water content (RWC). The laboratory variables determined were: ferric reducing ability of plasma (FRAP, mmoles/m 2 ), malondialdehyde (MDA, moles/m 2 ), total chlorophyll (TChl, mg/m 2 ), leaf total soluble sugars (LSS, g/m 2 ) and leaf total starch (LSt, g/m 2 ), leaf total proteins (LPr, g/m 2 ) and total ureides (TU, μmoles/g dry leaf). Seed chemical determinations comprises: protein and oil percentage (PrP and OP respectively) by NIRS; oleic, linoleic and linolenic acid concentrations (Ol, Li, Ln, respectively) by gas chromatography and total isoflavones (TI, mg/100 g dry deffated flour) by liquid chromatography. At harvest yield (Y) (g m -2 ), seed weight (SW, g) and seed number (SN) were also determined. A multivariate analysis was carried out to explore correlations between soybean seed components and chemical-physiological variables characterizing the photosynthetic source of assimilates during the seed filling. The biplot obtained from the first two principal components (PC1 and PC2) explained 90.8% of total variability in the data (Fig. 1) and revealed that higher levels of protein content (PrC), oil content (OC), OP, Li and TI in seeds were positively correlated with LSt, SPAD, ᶲPSII, RWC, FRAP and Fv/Fm, and negatively correlated with PrP, Ol, LSS and CT. Figure 1 also shows that PrC, OC, OP, Li, TI, LSt, SPAD, ᶲPSII, RWC, FRAP and Fv/Fm trait vectors were orientated towards irrigated plots, whereas PrP, Ol, LSS and CT trait vectors pointed towards heat and water stressed plots. In the latter, photosynthesis was modified by the overlapping of low levels of primary metabolism parameters (LSt, ᶲPSII, Fv/Fm) with elevated levels of oxidative stress (low values of FRAP and high of MDA)