RNA has been called a "prebiotic chemist's nightmare" because of its combination of large size, carbohydrate building blocks, bonds that are thermodynamically unstable in water, and overall intrinsic instability. However, a discontinuous synthesis model is well-supported by experimental work that might produce RNA from atmospheric CO(2), H(2)O, and N(2). For example, electrical discharge in such atmospheres gives formaldehyde (HCHO) in large amounts and glycolaldehyde (HOCH(2)CHO) in small amounts. When rained into alkaline aquifers generated by serpentinizing rocks, these substances were undoubtedly converted to carbohydrates including ribose. Likewise, atmospherically generated HCN was undoubtedly converted in these aquifers to formamide and ammonium formate, precursors for RNA nucleobases. Finally, high reduction potentials maintained by mantle-derived rocks and minerals would allow phosphite to be present in equilibrium with phosphate, mobilizing otherwise insoluble phosphorus for the prebiotic synthesis of phosphite and phosphate esters after oxidation. So why does the community not view this discontinuous synthesis model as compelling evidence for the RNA-first hypothesis for the origin of life? In part, the model is deficient because no experiments have joined together those steps without human intervention. Further, many steps in the model have problems. Some are successful only if reactive compounds are presented in a specific order in large amounts. Failing controlled addition, the result produces complex mixtures that are inauspicious precursors for biology, a situation described as the "asphalt problem". Many bonds in RNA are thermodynamically unstable with respect to hydrolysis in water, creating a "water problem". Finally, some bonds in RNA appear to be "impossible" to form under any conditions considered plausible for early Earth. To get a community-acceptable "RNA first" model for the origin of life, the discontinuous synthesis model must be developed. In particular, the model must be refined so that it yields oligomeric RNA from CO(2), H(2)O, and N(2) without human intervention. This Account describes our efforts in this direction. Our hypothesis centers on a geological model that synthesizes RNA in a prebiotic intermountain dry valley (not in a marine environment). This valley receives high pH run-off from a watershed rich in serpentinizing olivines and eroding borate minerals. The runoff contains borate-stabilized carbohydrates, formamide, and ammonium formate. As atmospheric CO(2) dissolves in the subaerial aquifer, the pH of the aquifer is lowered. In the desert valley, evaporation of water, a solvent with a nucleophilic "background reactivity", leaves behind formamide, a solvent with an electrophilic "background reactivity". As a result, nucleobases, formylated nucleobases, and formylated carbohydrates, including formylated ribose, can form. Well-known chemistry transforms these structures into nucleosides, nucleotides, and partially formylated oligomeric RNA.