Discussion: Panel: (G) Sampling Principles and Rockets J. R. Banister, moderator; M. Markels, P. F. Gustafson, H. R. Vaughn, Dr. Shreve: I would like to ask each panel member to come forward and spend a few minutes giving his impressions of the meeting. I'll call first, for impromptu remarks, on the panel moderator, John Banister, to be followed in turn by the other panel members. Dr. Banister: I haven't got my thoughts too well in order yet, since I found out about this only about three minutes ago. The thing I think that has been left out of all of our discussions is that the sampling methodis intimately related with the radiochemical analysis of the samples. I think that we are going to have to go ahead and try systems, having faith that there is enough stuff aloft to get an appreciable sample and that we will be able then to carry out an analysis. Ithink this will be the first stage of this program if we go into it. Ithink, then, that we have a second stage of refinement of technique, to develop something that is economically feasible after we gain more understanding of what conditions are that we are sampling under. particle size is at this altitude. I think another unknown is what the actual We should try to design systems which will sample over a wide particle range. On the sampling approach that Atlantic Research proposed, I would suggest that their system is somewhat more effective than they have indicated in their talk. I believe diffusion is actually a rather important phenomenon in the smaller particle range, and I think their analysis could be strengthened and improved by going to diffusion, Those are about all the comments that occur to me. Dr. Markels: Two things have come to myattention in the course of the conference which I would like to mention. The first is that in this upper-atmosphere sampling problem, we are in a regime in which engineering possibilities for collecting the samples which were just out of range are now becoming possible. I presented an approach; Sandia has presented another one; we have athirdin the cryogenic sampler. This indicates to me that there is a lot of room for creative thinking in solving the particular problems we face. As we think more and more about this and get more and more information about the physics and engineering required, the possibilities for inexpensive, effective sampling will actually improve rather than go the other way. started to think about this business, it looked hopeful but very difficult. I know that when we first We now seem to find two, or maybe three, systems that look quite feasible and require only the necessary funding and support to produce effective sampling. This brings me, then, to the next thing I would like to share with you: namely, that one of the big gaps that is holding back our imagination and holding back our working on some of these newer ideas is our lack of knowledge about a lot of the basic science of what may be going on up there ~ the particles and how theyinteract with filters, how they interact with devices such as impactors, what the aerodynamics might be, what the 159