Pkt t oo “. a wo age + ee a als Bd Ts at wee ie UY de orks uk. oa ate al weSR _ . . hie Cecured Allie dl ee el od Elia lee ate te el le et ot 179 than in developing a source-oriented model. In a source-oriented modcl, it is necessary to calibrate the urban terrain by determining the variation of ¢, and o, with distance for a number of direction bands. Surface roughness «affects these relationships. In addition, it is essential to make a detailed source inventory. With a receptor-oriented model, based on a Tabulation Prediction Scheme, a detailed source inventory is not necessary. However, one must know whether or not changes are occurring. It appears possible to develop a Tabulation Pre- diction Scheme which would have the advantages of both the source-oriented and receptor-oriented models. By maintaining surveillance over the addition or subtraction of sources within the urban area, a modification to the tabulation may be made just as the new sources come on line. In addition, by applying significance tests to these tabulations on a monthly basis, it appears likely that the addition or subtraction of new sources in the area could be detected and the tables updated accordingly. In fact by triangu- lation techniques (using two or more stations) the tabulation technique can be used to delineate the area which shows a change in source strength. Thus, a skillful use of the Tabulation Prediction Technique maybe used to monitor the distribution of sources. The Tabulation Prediction Scheme is a technique which cannot be constructed blindly; it must be a dy- namic system in which changes are continuously made. Similarly, a source-oriented model must be a dynamic system sensitive to the addition or subtrac- tion of sources. By applying the diffusion algorithms of the source-oriented type, the receptor-oriented tabulation prediction may be updated frequently. Of course, each station of the network would have to be updated. Such a hybrid technique may provide a superior system for forecasting air pollution levels based on meteorological variables. REFERENCE 1, Roberts, J. J., Reactor Engineering Division, Argonne National Laboratory. Private communication. RADIATION PHYSICS FURTHER STUDIES ON FLUORESCENCE POLARIZATION W. R. Anderson and I. B. Berlman Polarization studies of several compounds in solution have been made and the results are reported. For 1,3,6,8-tetraphenyl pyrene, the polarization values start to become negative at about 3100 A in agreement with other findings that the transition moment for absorption in this region is in a ‘different direction from that of fluorescence. In the case of substituted binaphthyls, an absorption band associated with a weak transition can be located at 3300 A. INTRODUCTION In a previous report) the reasons were given for interest in these studies, the polarimeter used was de- The principle involved in the first of these is as follows: In general, the absorption band associated with a given electronic transition overlaps that of others. Because of this, a weak and, therefore, seemingly narrow band within the span of a strong band may be obscured in an ordinary absorbance scan, but maybe disclosed if changes in the degree of polarization of the fluorescent emission are noted during the scan, provided a 90° angular shift between absorbing and emitting oscillators exists for one, but not both, of the transitions. Suppose that for a particular mole- scribed, the theory involved was outlined, technical cule the electronic transition representing the change to the first excited level happens to be 1A — 1Z, “) taking the studies are, in summary, the following: The states 1Z, and 'L, represent oscillators at 90° to each other. Fluorescence takes place from the lower of the two and is represented by 1L,_ — 1A; therefore, the molecules originally excited to the 1ZLy state undergo a 90° oscillator shift in dropping to the difficulties were discussed, and information obtained to that point was tabulated. The reasons for under1. To obtain information on the electronic transi- tions; more specifically, to look for hidden transitions; and 2. To study the effect of intrinsic viscosity on the degree of polarization. and that to the second excited level 1A — 1L,. 17,, state through internal conversion before emission.