Stratospheric aerosols were collected on special surfaces to determine both the physical state of the particles and the nature of included undissolved granules. Physically, each aerosol particle is composed of a somewhat volatile slurry mixture of crystalline-like material in a liquid matrix. Variations in the relative amounts of liquid at different times cause changes in particle fluidity. Although all particles contain a significant portion of the crystalline material, they are fluid in the atmosphere in the altitude range from 12 to 32 km and are physically similar at all geographic locations from 5øN to 89øN latitude and from 55øW to 175øW longitude. No evidence of frozen particles has been found even at stratospheric temperatures of -65øC. Undissolved granules are found in only one third of the aerosol particles, and sometimes the granules are found alone outside the slurry matrix. Thus we conclude that in general, these undissolved granules are not active nucleating agents for stratospheric aerosols. Furthermore, these granules now contain mostly sulfur, sodium, and possibly lighter elements, whereas before they contained silicon and other heavy elements attributed to volcanic eruptions. We conclude that other important sources besides volcanos dominate the granular component after long quiet periods without significant volcanic activity.

We have obtained stratospheric aerosols from tropical to northern latitudes using special collectors on U-2 aircraft during 1976 and 1977. Aerosols characterized by large numbers of small particles are found in the tropical zone suggesting this is a region of particle growth; whereas aerosols containing mostly larger particles are distributed throughout the Northern Hemisphere indicating a well-mixed, mature population. We find the aerosol layer extends from higher altitudes near the equator to lower ones toward the pole. Although this gradient suggests mature aerosols may leave the stratosphere at high latitudes, the data are, as yet, inconclusive. Comparisons of our data with those of other investigators using different instruments are generally encouraging, suggesting that if similar populations were sampled, the results would be similar. When our calculated sulfate mass mixing ratios are compared with those measured directly by others, we find better agreement is achieved if we assume more dilute sulfate and water mixtures than previously proposed.

Two forms of nitrosyl sulfuric acid (NOHSO4 and NOHS:O7) have been tentatively identified in stratospheric aerosols. The first of these can be formed either directly from gas reactions of N O: with or by gas-particle interactions between NO: and H:SO4. The second product may form when SOs is involved. Estimates based on these reactions suggest that the maximum quantity of NO that might be absorbed in stratospheric aerosols could vary from one-third to twice the amount of NO in the surrounding air. If these reactions occur in the stratosphere, then a mechanism exists for removing nitrogen oxides from that region by aerosol particle fallout. This process may typify another natural means that helps cleanse the lower stratosphere of excessive pollutants.

In this paper we describe a sampling and analysis technique that uses the binomial distribution to characterize stratospheric aerosol populations at the 95% level of confidence. Particle samples obtained over Alaska during July 15‐19, 1979, are used; the results show the presence of more small particles at lower altitude than at high altitudes. We also give calculations of the surface area and volume distributions for all aerosol samples collected. Evidence from these data suggests either that Aitken nuclei are injected or diffused across the tropopause and rise into the stratosphere, where they mature into larger particles, or nuclei form in the lower stratosphere and become mature aerosols at high altitude. Samples obtained at another site give the same results, supporting the view that the process of injection or nucleation and maturing of aerosols with altitude may be global and need not occur only in locations exhibiting unique meteorologic features.

Stratospheric aerosols were collected at Poker Flat, Alaska, in July, 1979, to determine particle properties, confirm coincident satellite SAGE measurements, and compare similar results obtained with different airborne samplers. Because of the steep slopes in size‐distribution curves for larger particles, we found properties such as concentration, aerosol mass, and optical extinction are very sensitive to small errors in radii. We calculate that our concentration measurements agree with photoelectric particle counter results when a 16% radius change is introduced. An 8% radius change matches our calculated sulfate mass with filter mass measurements. And a 13% radius change results in agreement between our calculated optical extinction and coincident SAGE satellite results. Recognizing that different instruments can produce 10‐20% differences in measured sizes, we believe the results of these comparative measurements of SAGE and in situ instruments are essentially in agreement.