particles are themselves, what the atmosphere is doing, and so on. I suggest that there may be a very fertile field here for rather basic research on individual component areas of this whole problem with which we are dealing, and that such research, in support of these individual areas which make up the whole technological field which we are attaching, might be very worth while. Dr. Gustafson: During the course of this meeting, I have been very impressed with the amount of work that has gone into particle collection devices, filter devices, things which are very sophisticated in comparison with the filter systems which we have used in the limited amount of work we have done inthis area. The comments I would like to make are in regard, not to these devices, but rather to another area which I don't believe would substitute for this type of sampling, but which might lead to the collection of a considerable amount of auxiliary data. This is gamma-ray spectromeiry, particularly in situ measurements of the atmosphere as far out as one can get, whether this device is mounted on balloons or in rockets. has not been done with any great degree of fervor. This is, to my knowledge, something that Measurements have been made in balloons, but the things that have been looked at are not in the fission-product energy region nor in the region where natural activity such as beryllium emits. lection. At first glance, the problem is a difficult one, but I think so is that of particle col- It is a matter of how much time and effort one wants to devote to this to effect a reasonable degree of success. Certainly the first looks we had at the composite gamma spectrum of soil were anything but en- couraging. Therefore, I don't know that the great secondary showers of particles that are going to becloud the characteristic line of the various fission products need discourage us to the point where we don't at least try it. I would like to make a plea that this be done, and, in the limited way at Argonne that we are able to, we are going to try to do our part in this portion of the program. Mr. Vaughn: Iam a member of the Aerodynamics Department here, and, as such, my primary interest is in helping to furnish the vehicle to do the job. In order to do this properly we should have a very good idea of the very best means of sampling as well as the sample size. My impressions from the talks have been that, from an AEC, or particulate sampling standpoint, probably the best means would be an impaction - diffusion device, and the rotating-type vane scheme seems to be one of the best. I have also gotten the impression that cryogenic-type devices will be very important for whole-air sampling. It is not clear to me whether the AEC should undertake this sort of thing or not, but I can certainly see a lot of value in it. I have the impression that the sample size of 1000 cubic feet of air would be satisfactory, but there seems to be quite a bit of doubt along this line. have been trying to get something on the order of 10, 000 standard cubic feet of air. would need. We This may be more than we However, I do want to make one point that wasn't brought out in our discussion: the rockets that we would use in the development of this device are surplus types that cost $2000 apiece or less. So, from a development standpoint, the size of the sample is not too important. However, in a final device, the size would be important as well as the number of shots that might be required. It may be too far in the future to try to de- termine how many shots would be necessary, but this might give us an insight into the design needs of the final device. One thing I feel very strongly about: there have been a number of papers indicating very strongly that at least there is something up there, so I feel somewhat more confident about being able to get a good sample back. 160