Blue-and green-light-emitting diodes (LEDs) based on InGaN/ GaN films that are typically grown along the polar c-axis suffer from strong polarization-related internal electric fields that limit the radiative recombination efficiency due to spatial separation of holes and electrons. LEDs based on nonpolar GaN orientations, including the (1100) m-plane and (1120) a-plane, are free from polarization-related internal electric fields, and thus have the potential for improved efficiencies. Accordingly, research efforts in developing nonpolar InGaN/GaN LEDs have dramatically increased recently. However, a major problem for the development and adoption of efficient nonpolar LEDs is the limited availability of high-quality substrates. Growth of nonpolar a-plane GaN epilayers using nucleation-layer techniques results in films with high threading dislocation densities (TDDs) in the mid-high 10 10 cm ร2 range, and basal-plane stacking-fault densities in the mid-high 10 5 cm ร1 range. The presence of high levels of extended defects, particularly threading dislocations, leads to lower device lifetimes and reduced efficiencies and output powers. Recently, dramatic improvements in a-plane and m-plane LED performance have been reported for bulk a-plane and m-plane GaN substrates with low TDD in the 10 6 cm ร2 range. These results confirm that a low-dislocation-density material is necessary for high efficiency nonpolar nitride LEDs. Unfortunately, the size, availability, and cost of nonpolar freestanding bulk GaN substrates are serious barriers to the development and adoption of inexpensive nonpolar LEDs. While the size, cost, and availability of bulk freestanding GaN substrates are likely to continue to improve in the future, the cost is unlikely to come close to that of sapphire substrates for the foreseeable future, rendering them impractical for inexpensive devices such as LEDs. Epitaxial lateral overgrowth (ELO) techniques and variations of ELO have recently been employed to produce nonpolar a-plane and m-plane GaN films with TDD in the 10 6 -10 8 cm ร2 range in the wing regions. However, given the process complexity, and hence high cost, of ELO-based buffers,