This article uses a novel database of Ottoman probates and examines some of the methodological difficulties that arise in very long-term analysis. Wealth statistics, spanning from 1460 to 1920 in the longest subsample, indicate approximately an inverted U-shaped pattern that may signal the limits of extensive growth. While plausible, severity of the drop on the right side of the wealth curve does not entirely match recent scholarship on the Ottoman Empire. Examining the effect of biases and changes in probate demography on wealth, we explore how real the observed wealth pattern is. We employ descriptive statistics, linear regression and Oaxaca-Blinder decomposition, and find that demographic composition matters but does not alter the shape of the wealth curve. Explanation for the gap between probate findings and current historiography, therefore, must lie elsewhere.