Among amniotes, mammals are distinguished by a suite of unique features in the pharynx. These features are thought to facilitate several novel feeding behaviors, including a new way of swallowing food and liquid bolus by explosively propelling them out of the pharynx and into the esophagus. The evolutionary origin and transformation pattern of the mammalian pharynx remain obscure, however. The hyoid is an important structural component of the pharynx and larynx, and plays an integral role in swallowing, respiration and vocalization. Morphological transformations in the hyoid can thus yield novel insights into the timing and pattern of oral-pharyngeal reorganization during mammalian evolution, and how unique mammalian feeding behaviors originated. First evolved in Jurassic mammaliaforms, the unique morphology and posture of the mammalian hyoid are thought to coincide with a muscularized pharynx for powered swallowing of liquid and masticated food. However, the relationship between a novel mammalian hyoid morphology and the mammalian powered swallowing mechanics is poorly understood. Moreover, hyoid form continues to diversify within crown Mammalia, yet we know little about the mechanical consequence of this morphological diversity for swallowing performance. Here I leverage the diversity of hyoid morphology among living mammals to test hypotheses on the evolution and form-function relationships of the mammalian hyoid apparatus. Using euarchontan mammals (primate and kin) as a model system, I demonstrated that multiple distinct evolutionary trajectories are viable for the origin of novel hyoid phenotypes in mammals (Chapter I), although specialized hyoid morphology did not necessarily evolve in tandem with novel vocalization behaviors. In Chapter II, I demonstrated that head-neck posture variation could confound inference on hyoid resting position based on hyoid morphology alone, and instantaneous head posture may impact hyoid muscle function during in vivo hyoid movement. In Chapter III, I examined the mechanical consequence of interspecific variation in hyoid form and resting position on hyoid mobility during swallowing. Contrary to previous hypotheses, hyoid arch ossification does not seem to restrict hyoid range of motion, which is more heavily mediated by hyoid-skull geometry. Finally in Chapter IV, I demonstrated that similar hyoid morphology and/or hyoid resting position does not necessarily indicate similar biomechanical function for bolus propulsion strategy during swallowing: while the macaque and dog hyoid can act like a piston to drive tongue base retraction for bolus propulsion, tongue base retraction may be mechanically decoupled from hyoid movement in opossums. Based on evidence from a phylogenetic, experimental and modeling perspective, I argue that hyoid morphology alone is insufficient to derive information about hyoid resting position (Chapter II), hyoid in vivo mobility (Chapter III), and hyoid biomechanical function during swallowing (Chapter IV) and vocalization (Chapter I). The lack of a strong form-function relationship of the mammalian hyoid apparatus suggests evolution of novel hyolingual morphology may occur without significant performance trade-off between different behaviors. It also highlights the decoupling between morphological and functional diversity during the evolution of complex biomechanical systems.