This paper presents an experimental realization of the Einstein-Podolsky-Rosen (EPR) paradox for continuous variables. The EPR paradox, originally proposed by Einstein, Podolsky, and Rosen, questions the completeness of quantum mechanics by highlighting the apparent nonlocality of quantum states. The experiment demonstrates that the correlations between two spatially separated particles can be used to infer the properties of one particle from the other, even when the particles are not directly measured. The experiment uses nondegenerate optical parametric amplification to generate correlated amplitudes for signal and idler beams of light. The amplitudes of the signal beam are inferred from those of the idler beam. The uncertainty product for the variances of these inferences is observed to be 0.70 ± 0.01, which is below the limit of unity required for the demonstration of the paradox. The experiment involves measuring the quadrature-phase amplitudes of the fields emerging from the optical parametric oscillator. The results show that the product of the variances of the inferred amplitudes is less than one, demonstrating the EPR paradox. The experiment also shows that the correlations between the signal and idler beams are strongly nonclassical, with the quadrature amplitudes of the signal and idler beams becoming "quantum copies" of one another over a bandwidth set by the OPO linewidth. The results of the experiment are consistent with the theoretical predictions and demonstrate the nonclassical nature of the correlations. The experiment also has potential applications in precision measurement and quantum communication, as the correlations can be used for noise suppression below the vacuum-state limit in various dual-beam arrangements. The experiment highlights the importance of the EPR paradox in understanding the nature of quantum mechanics and the challenges in extending the Bell inequalities to continuous variables.This paper presents an experimental realization of the Einstein-Podolsky-Rosen (EPR) paradox for continuous variables. The EPR paradox, originally proposed by Einstein, Podolsky, and Rosen, questions the completeness of quantum mechanics by highlighting the apparent nonlocality of quantum states. The experiment demonstrates that the correlations between two spatially separated particles can be used to infer the properties of one particle from the other, even when the particles are not directly measured. The experiment uses nondegenerate optical parametric amplification to generate correlated amplitudes for signal and idler beams of light. The amplitudes of the signal beam are inferred from those of the idler beam. The uncertainty product for the variances of these inferences is observed to be 0.70 ± 0.01, which is below the limit of unity required for the demonstration of the paradox. The experiment involves measuring the quadrature-phase amplitudes of the fields emerging from the optical parametric oscillator. The results show that the product of the variances of the inferred amplitudes is less than one, demonstrating the EPR paradox. The experiment also shows that the correlations between the signal and idler beams are strongly nonclassical, with the quadrature amplitudes of the signal and idler beams becoming "quantum copies" of one another over a bandwidth set by the OPO linewidth. The results of the experiment are consistent with the theoretical predictions and demonstrate the nonclassical nature of the correlations. The experiment also has potential applications in precision measurement and quantum communication, as the correlations can be used for noise suppression below the vacuum-state limit in various dual-beam arrangements. The experiment highlights the importance of the EPR paradox in understanding the nature of quantum mechanics and the challenges in extending the Bell inequalities to continuous variables.