In this article, we will defend the epistemological claim that observation is mathematically- laden. A well-known thesis in the philosophy of science is that observation is theory-laden. The claim that observation is mathematically laden can be similarly justified. The first part of the paper focuses on the definition of mathematically-ladeness and its relations to the best-known problem of theory-ladeness. The second part of the paper presents some explicit examples and outlines the consequences and the difficulties of this epistemological limit. Finally, a specific context for this problem is discussed in detail: we will analyze the question of deterministic chaos as a paradigmatic example of mathematically-ladeness and show that the deterministic or indeterministic nature of chaos is strongly linked to the choice of a particular mathematical description.