Bimodal

Bimodal describes a distribution that seems to be a mixture of two basic distributions

Examples of bimodal distributions often use two gaussian-ish humps, in part to point out these are not entirely separable, while still being pretty useful to us.

Pointing out that something is bimodal often comes up in cases where it's convenient to simplify into something with two distinct, clearly separated cases, and more so in cases where that is a simplification with problems.

...or, more generally, a multimodal distribution that is made of two or more.

See also:

• https://en.wikipedia.org/wiki/Multimodal_distribution