LETTERSTO parities as explained below. The arguments are based on two assumptions: (1) Gamow-Teller selection rules govern the beta-decay process, and (2) the nuclear shell model 1 is a reliable guide in the region Z^-40 and N~50.The nucleus 4 oZr 90 contains closed shells of 40 protons and 50 neutrons. Thus, 1 = 0 and the parity is even in the ground state.39Y 90 is an odd-odd nucleus. It is characterized by a (3p)~1(4d) 1 configuration in the notation of Feenberg and Hammack. Thus, the ground state parity is odd. 3 3Sr 90 is an even-even nucleus. It is characterized by a (3p)~2(4:d) 2 configuration. In the ground state the parity is even and, most probably, 7 = 0.Both beta-transitions are associated with change of parity. Thus by G-T rules, both are first-forbidden or both are thirdforbidden. The ft product for the first transition is ~10 9 and for the second ^40 8 . These values are appropriate for firstforbidden transitions with AJ= ±2 (1_2) , but are too small for third-forbidden. Consequently, 1 = 2 in the ground state of Y 90 .Thus the two beta-transitions of Fig. 1 should produce peculiar energy distributions similar to those found in Y 91 , Cs 137 , K 42 , and Rb 86 by various investigators. 2-7 In such cases the energy distribution differs 8 from the allowed form by a factor G~(W Q -W) 2 + (W 2 -l). A carrier-free sample of Sr 90 was obtained from Oak Ridge. This sample had been aged to permit an associated 55-day activity of Sr 89 to die out. Three spectra were measured:(1) Sr 90 and Y 90 in equilibrium together, (2) Sr 90 alone, and (3) Y 90 alone. For the latter two runs, the isotopes were chemically separated by a method due to Kurbatov and Kurbatov. 9 Samples were prepared for the spectrometer by evaporation from solution on thin Zapon films, following the insulin technique of Langer. 10 The spectra were measured in the double-focusing spectrometer described by Kurie, Osoba, and Slack. 11 In the interests of higher counting rates, a wide counter window (0.25 in.) was used, which dropped the resolving power to 1 percent. The window was Zapon film of 3-to 4-kev stopping power, and was supported lengthwise by a single 5-mil wire. A pressure regulator 12 was used to avoid loss of counter pressure caused by window leakage.The results are shown as FK (Fermi-Kurie) plots in Fig. 2 for Sr 90 and in Fig. 3 for Y 90 . As indicated in Fig. 2, the Sr 90 data was obtained by subtracting the Y 90 distribution curve from that for an equilibrium mixture of both Sr 90 and Y 90 . * Assisted by the joint program of the ONR and the AEC.