We evaluated spatial heterogeneity for pH and a comprehensive set of nutrient and trace elements in surface (0–0.1 m depth) and subsurface (0.3–0.4 m depth) soils across 26.6 ha of old-growth, lowland, tropical moist forest, established on a highly weathered soil in Panama. Little is known about spatial heterogeneity patterns of soil properties in tropical forest soils. Soil was moderately acidic (pH 5.28) with low concentrations of exchangeable base cations (13.4 cmol_c/kg), Bray-extractable PO₄ (2.2 mg/kg), KCl-extractable NO₃ (5.0 mg/kg), and KCl-extractable NH₄ (15.5 mg/kg). The coefficient of variation for soil properties ranged from 24% to >200%, with a median value of 84%. Geostatistical analysis revealed spatial dependence at a scale of 10–100 m for most of the soil properties; however, pH, NH₄, Al, and B had spatial dependence at a scale up to 350 m. Best-fit models to individual variograms included random, exponential, spherical, Gaussian, linear, and power functions, indicating many different spatial patterns among the set of soil properties. Correlation among individual elements was poor, indicating independent patterns. Our results show complex spatial patterns in soil chemical properties and provide a basis for future investigations on soil–plant relationships and soil nutrient niche differentiation.