Abstract The evolution of wings and asymmetrical flight feathers in non‑flying theropod dinosaurs remains a contested problem. Existing hypotheses emphasize lift, thrust, or wing‑assisted incline running, but none account for a fundamental property of vertebrate movement: angular momentum is the universal currency of all locomotor and non‑locomotor behaviors. Any change in body orientation, direction of travel, or segmental position arises from the generation, transfer, and redistribution of angular momentum within the biomechanical unit. I propose that the primary selective advantage of early wings (Level 4 on the four‑stage feather scale described by Potts (2026a), was not flight, but the ability to generate highly reliable angular momentum by pushing against atmospheric gases—a fluid medium with consistent, predictable viscosity and density. We introduce the Coefficient of Reliability (CR) = Var(angular impulse) / (mean angular impulse)² for a fixed motor command. For wings in air, CR → 0 (perfect reliability); for limbs pushing against natural substrates (sand, mud, gravel, vegetation), CR is very large or infinite. Because the vertebrate nervous system and body plan evolved in water—another high‑reliability fluid—terrestrial substrates impose an ancestral mismatch. Wings therefore represent a return to a fluid regime, restoring the predictability of force, torque, angular momentum, and their cascading consequences. Continuous sensory feedback from earlier feather stages (Levels 1‑3) provides real‑time information on atmospheric conditions, enabling predictive compensation and extending reliability to gusty, turbulent environments. Low CR at the input guarantees low variance throughout the kinetic chain, allowing animals to redirect angular momentum into infinitely many locomotor and non‑locomotor trajectories with confidence. This framework yields testable predictions regarding avian kinematics, avian biomechanics, avian evolutionary biomechanics, fossil joint morphology, robotic CR measurements, and neural homologies between fish and birds.
Charles Darryl Potts (Fri,) studied this question.