fied at present. It may be due to an unreported band of water vapor or NO. It does not, how- ever, have the appearance of a typical NO, section of spectrum.| With regard to atomic spectra present in Mike and King, no atomic lines are prominent below 4800 A. However, since every single line in this region has not yet been identified, some might be present. The three atomic oxygen lines at 7771, 7774, and 7775 A, observed in the total King and in the King first-maximum spectra, are very broad. A further study of their appearance is suggested. This region in Mike was blackened by accident, and measurements were impossible. Sodium lines at 5890 and 5896 A are observed in Mike and King but are of no special interest here, 4.5 ROTATIONAL AND VIBRATIONAL TEMPERATURES Information on temperature is rather meager, especially for the Ivy shots. It was not possible to obtain any value from the Mike spectra and only one value for King. Rotational temperatures were calculated by both the maximum method and the slope method using the measured intensities and J values for the R-branch of the S-R 0-13 and S-R 0-15 Table 4.7-—-APPROXIMATE ROTATIONAL TEMPERATURES Yield, Exposure S-R O, Temp. by max. Temp. by slope Shot kt time band Tmax. method, K method, °K Buster Charlie Tumbler-Snapper 3 14.0 30.7 0~20 msec =Total 0-13 0-13 1l 37 700 8,000 600 2,500 Tumbler-Snapper 3 Tumbler-Snapper 7 30.7 14.6 Total 0~100 usec 0-15 0-15 40 11 10,000 700 5,000 0-15 40 10,000 10,000 Buster Charlie King 14.0 540.0 0-20 msec 0-15 1.0—1.1 sec 9 400 bands as given in Tabies 4.2 and 4.3. These temperatures are listed in Table 4.7. The graphs of log, [1/(2J + 1)] vs E from which the “slope” temperatures were determined are shown in Figs. 4.6 and 4.7. In these graphs the rotational energy is that of the lower state and is plotted in wave numbers (cm™'). (See Sec. 1.3 for comment on the maximum method.) Theoretically, the same temperature should be arrived at by both methods for a certain shot; but, as seen in Table 4.7, the 0-13 S-R band of Buster Charlie gave about 600°K by the slope method as compared with 700°K by the maximum method. Similar variations were obtained for the other shots, except in the case of King where both methods gave roughly 10,000°K. The slope method is considered more reliable for obtaining rotational temperatures since more points were used in the determination. Both methods are affected by large error due to overlapping of branches and rotational components themselves. The vibrational temperature of the early-time spectra is very high in comparison with the rotational temperature, as is evidenced by the existence of transitions from a very high vibrational level (v ranging in O, from 21 or so, upward). Calculation of this type of temperature is not possible at present because of lack of good data on transition probabilities, and, also, the experimental data which are measurable do not cover a wide enough energy range of the initial quantum state. Although it is true that thermal excitation does not produce these high vibrational levels at the early stage, it does seem that the levels could be thermally excited in the later stages. Before this can be verified, reliable transition probabilities over a wide range of observed frequencies should be determined. An electronic temperature for O02, for example, cannot be calculated because of lack of observation of different electronic levels from which transitions take place. 30 ‘