High-temperature origin-of-life theories require that the components of the first genetic material are stable. We therefore have measured the half-lives for the decomposition of the nucleobases. They have been found to be short on the geologic time scale. At 100 degreesC, the growth temperatures of the hyperthermophiles, the half-lives are too short to allow for the adequate accumulation of these compounds (t1/2 for A and G approximately 1 yr; U = 12 yr; C = 19 days). Therefore, unless the origin of life took place extremely rapidly (<100 yr), we conclude that a high-temperature origin of life may be possible, but it cannot involve adenine, uracil, guanine, or cytosine. The rates of hydrolysis at 100 degreesC also suggest that an ocean-boiling asteroid impact would reset the prebiotic clock, requiring prebiotic synthetic processes to begin again. At 0 degreesC, A, U, G, and T appear to be sufficiently stable (t1/2 >/= 10(6) yr) to be involved in a low-temperature origin of life. However, the lack of stability of cytosine at 0 degreesC (t1/2 = 17, 000 yr) raises the possibility that the GC base pair may not have been used in the first genetic material unless life arose quickly (<10(6) yr) after a sterilization event. A two-letter code or an alternative base pair may have been used instead.