Interhelical salt bridges, coiled-coil stability, and specificity of dimerization.

absorbed in the atmosphere, the NIR water vapor bands. Thus, we see no way of defending Stephen's statement that conclusions based on broad-band measurements (1) are inconsistent with other existing data sets where discrepancies were found to occur between measurements in spectral bands and theory. Instead, the broad-band solar absorption findings presented in our report (1) are entirely consistent with existing spectral data sets, as both show cloud absorption to be greater than the magnitude predicted by theory. Stephens makes another critical error when determining cloud absorption with the use of results shown in figure 6 of our report (1). Stephens determined cloud absorption, A, from the relationship A + R + T = 1, where R is reflectance and T is transmission, and he states that figure 6 in our report (1 ) indicates that a cloud with reflectance (albedo) 0.45 would absorb 0.40. His error is in assuming A occurs entirely in cloud. He does not account for the absorbing surface or for the lower 10 km of the atmosphere as well. An appropriate relationship between R and A is R + A = 1, where A now is the fractional absorption by the combined atmosphere-surface system (7). The analysis shown in figure 6 in our report is not by itself sufficient to determine cloud absorption; again we would refer the reader to figure 2 in our report (or Fig. 1 here). Consequently, contrary to Stephens erroneous conclusion, in our report (1) figure 1 is entirely consistent with figure 6 in our report, as those figures represent alternative ways of viewing