The R programming language is widely used for statistical computing. To enable interactive data exploration and rapid prototyping, R encourages a dynamic programming style. This programming style is support by a number of features including first-class environments. R is the only language, with millions of users, that provides a full reflective interface for manipulating environments. With the great flexibility afforded by first-class environments, comes challenges for reasoning about code and significant restrictions about what transformations a compiler is allowed to apply to programs. This paper is an overview of the environment interface refined over two decades. We explain the rationale behind the design and document how environments are used in the wild by the means of a corpus analysis.