Color any flat map so that no two neighboring regions share a color and you will never need more than four crayons. That simple observation, made by Francis Guthrie while coloring the counties of England in 1852, stumped mathematicians for 124 years. This deep dive follows the puzzle from Augustus De Morgan’s inability to explain it, through Alfred Kempe’s celebrated 1879 proof and Peter Guthrie Tait’s 1880 follow-up, both of which stood unchallenged for eleven years until Percy Heawood and Julius Petersen found fatal flaws and salvaged only a five color theorem.

We explain how maps become planar graphs of vertices and edges, why the pie chart and empire loopholes must be ruled out, why three-coloring is NP-complete, and how odd and even neighbor counts around Missouri and Nevada force a fourth color. Then we cover the 1976 breakthrough by Kenneth Appel and Wolfgang Haken at the University of Illinois, who used discharging to reduce infinity to an unavoidable set of configurations and over 1,000 hours of computer time to check them, sparking a philosophical fight over whether an unreadable machine proof counts as mathematics. We close with Mobius strips needing six colors, a torus needing seven, infinite colors in three dimensions, and the 2005 formal verification that settled the matter.

• Francis Guthrie’s 1852 observation and the false proofs that fooled mathematicians for a decade
• Kempe chains and the tangled configuration that broke them
• Turning geography into graph theory: vertices, edges and the rules for what counts as a border
• Discharging, the unavoidable set and the 1976 computer-assisted proof
• Why donuts need seven colors and 3D solids need infinitely many