Insulin resistance (IR) is a characteristic feature of non-insulin-dependent diabetes mellitus (NIDDM) as well as obesity, and a majority of NIDDM patients are obese. To assess the effect of obesity independent of NIDDM on IR, we studied the relationship between IR and obesity in 65 normal and 58 NIDDM subjects; we used body mass index (BMI) as a measure of obesity and glucose infusion rate (GINF) during a euglycemic hyperinsulinemic (120 mU · m⁻² · min⁻¹) glucose clamp as a measure of IR. In lean normal subjects, GINF was 57.7 ± 2.2 μmol · kg⁻¹ · min⁻¹ (10.4 ± 0.4 mg · kg⁻¹ · min⁻¹) and the lean NIDDM subjects were markedly insulin-resistant, with a GINF of 34.4 ± 2.8 μmol · kg⁻¹ · min⁻¹ (6.2 ± 0.5 mg · kg⁻¹ · min⁻¹). Obese normal subjects were also insulin-resistant compared with lean normal subjects, with a GINF of 36.1 ± 2.2 μmol · kg⁻¹ · min⁻¹ (6.5 ± 0.4 mg · kg⁻¹ · min⁻¹), and obesity caused an increase in IR in NIDDM, with a GINF of 21.1 ± 1.4 μmol · kg⁻¹ · min⁻¹ (3.8 ± 0.25 mg · kg⁻¹ · min⁻¹) in the obese NIDDM subjects. Therefore, ∼61% of the IR in obese NIDDM subjects is due to NIDDM, with 39% due to obesity, demonstrating a greater impact of NIDDM than of obesity in causing IR. The correlation between GINF and BMI was much better in normal subjects (r = −0.75) than in NIDDM subjects (r = −0.50) as was the relationship between fasting insulin level and BMI (r = −0.59 in normal subjects, r = −0.48 in NIDDM subjects). As expected, the fasting insulin level was also strongly correlated to GINF in normal subjects (r = −0.61); however, this relationship was weaker in NIDDM subjects (r = −0.46). In conclusion, 1) obesity has a major impact to cause insulin resistance in nondiabetic subjects, but the effect of obesity on IR in NIDDM is less; 2) NIDDM per se is the major contributor to IR in NIDDM; and 3) the fasting insulin level is a better surrogate marker of IR in nondiabetic subjects than in NIDDM patients.