[1] We present the first microphysical calculations confirming that photolysis of sulfuric acid vapor by visible light is responsible for the formation of the springtime ''CN layer'' observed in the polar stratosphere. Our calculations show that the recently proposed photolysis mechanism is also sufficient to explain observations of vertically increasing SO 2 mixing ratios in the upper stratosphere. Such photolysis, however, does not sufficiently explain the limited observations above 40 km of vertically decreasing H 2 SO 4 and SO 3 vapor, suggesting an additional loss mechanism there. We rule out previous speculation regarding reaction of H 2 SO 4 with O( 1 D) or OH as inconsistent with observations of SO 2 . Rather, the loss is consistent with a permanent sink for sulfur, such as H 2 SO 4 neutralization by metals on meteoritic dust. In light of such removal, H 2 SO 4 photolysis to SO 2 gains added importance in preserving gaseous sulfur in the upper stratosphere. We also present new cross sections derived from ab initio calculations for H 2 SO 4 absorption at visible and Lyman-a wavelengths and evaluate their atmospheric implications. Our atmospheric model reveals that photolysis of H 2 SO 4 by Lyman-a radiation is responsible for up to one third of the SO 2 in the mesosphere and 10% of the particles in the polar stratospheric CN layer.