A preliminary temperature-programmed desorption (TPD) study of hydrogen desorbing from diamond (100) concludes that the previously-reported non-saturation behavior of the surface is an experimental artifact resulting from desorption from the Ta heater. Critical conditions for conducting an accurate TPD measurement are identified. In its first application to the diamond surface study in which lateral periodic boundary conditions are incorporated, the molecular mechanics method (MM3) has been shown to be a useful tool in determining surface structures and energetics, employing modest-sized clusters. Atomic structures and energetics of the diamond (100)-(2$\times1),(100)−(2\times1):H,(100)−(1\times1):2H,and(100)−(3\times1):1.33Hsurfaceshavebeencalculated.Pairsofsurfacecarbonatomsformsymmetricdimersonthereconstructeddiamond(100)−(2\times1),(100)−(2\times1):H,and(100)−(3\times1):1.33Hsurfaces,withdimerbondlengthsof1.46A,1.63A,and1.59A,respectively,correspondingtostraineddoubleorsinglebonds.Thefull(100)−(1\times1):2HdihydrideishighlystrainedduetoH−Hrepulsions,causingareductionoftheH−−C−−Hbondangleandtwistingaboutthesurfacenormal,andispredictedtobethermodynamicallyunstablewithrespecttodehydrogenationtothemonohydride.Someimportantgas−surfacereactionsinvolvinghydrogenandthediamond(100)and(100)−(2\times$1):H are discussed in light of the derived energetics. We have noted that preferential pairing of chemisorbed hydrogen on Si(100) is a direct result of the partial π-bond existing on the surface dimers. A lattice-gas model has been developed based on this concept, and predicts that, with a modest pairing energy, hydrogen desorption adopts near-first-order kinetics at high coverages but deviates from first-order kinetics at low coverages. We calculated the pairing energy of adsorbed H to be about 7.5 kcal/mol, based on a comparison of the predictions of the model with experiment. We conclude that preferential pairing on dimers is a general feature of hydrogen adsorption on the (100) surfaces of group IV semiconductors.