Imagine a mountaineering enthusiast who decides to write a comprehensive guide for climbers. This would-be author spends many years climbing every peak throughout the world, gathering material for a book, and then starts to write. In the end, our enthusiast produces a guidebook, listing each peak, its longitude and latitude, and the number of climbers the author encountered at the summit.

Clearly, this guidebook is unlikely to make the best-seller list or even to become a citation classic. Readers of this guide would undoubtedly have many additional questions: What is the height of each peak? How steep is the approach to the summit from different directions? Are there ridges leading to the peak? Are there other peaks nearby? When asked about these questions, the author replies in defense, “Climbers climb peaks, so I wrote about peaks.”

In our view, many (perhaps most) studies of optimality in biology seem similar to this guidebook. The primary questions of interest are “Where is the fitness peak?” and “Is the population or species of interest currently at or near the peak?” The rationale for this emphasis is that, because evolution by natural selection is expected (under certain conditions) to increase the mean fitness of a population, one might predict that populations would tend to reside at the peaks in a fitness landscape.

Although fitness peaks are interesting, we fear that an exclusive emphasis on them may generate an understanding of phenotypic selection and evolution that is as unsatisfying as our imaginary mountaineering guidebook. We suggest that a broader view that explores the topography of fitness landscapes may be useful.