N UCLEAR magnetic resonances arising from the isotopes of thallium (Tl 203 and Tl 205 ) have been observed. The ratio of the resonant frequencies of Tl 205 and H 1 was measured.Thallium in solution was used for a sample in order to obtain the narrow resonance lines characteristic of liquids. Two r-f coils, J inch long and •£$ inch in diameter, were wound on a" glass coil form containing a saturated solution of thallium acetate, the most soluble of the thallium compounds available. Each coil was part of a circuit similar to one which has been described elsewhere. 1 A |-inch separation between coils reduced the coupling between them sufficiently so that adjustment of one circuit did not appreciably disturb the other. The M.I.T, cyclotron magnet provided the magnetic field.Proton resonances were observed in one circuit at about 30.5 Mc/sec. In the other circuit, operated at a frequency calculated from spectroscopic values 2 [spin: \; magnetic moment: 1.45 \x N (nuclear magnetons); M205/M203 = 1.00966] to give thallium resonances for the same magnetic field, two resonances separated by a one percent difference in field were observed for a field about 10 percent less than expected. The stronger resonance occurred at the lower magnetic field and was identified as caused by Tl 205 , since the ratio of its abundance 3 to Tl 203 is 2.5 and their spins are the same. Also, from the resonance condition, hv = gwH> it is seen that at a given frequency, resonance will occur for the larger moment in a smaller field. The spectroscopic value of the moments is thus shown to be about 10 percent too low, while the value of their ratio is roughly confirmed.With the thallium circuit operating in the vicinity of 17.5 Mc/sec. both proton and thallium resonances could be observed simultaneously, using an electronic switch to display the output of the receiver in each circuit on the same oscilloscope. The Tl 205 and proton resonances were aligned, and measurements were made of the ratio of resonant frequencies for slightly differing oscillator frequencies. Another run was made on a subsequent occasion and gave results in close agreement with those of the first run. The signal-to-noise ratio for the weak Tl 203 resonance was unfortunately too low at this time to permit an accurate check of the ratio of the moments of the two isotopes.Frequencies were measured with a General Radio 620A frequency meter. The crystal oscillator in the meter was check against that in a Signal Corps BC-221 frequency meter. The agreement was exact as far as could be measured. Systematic errors that might arise from inhomogeneities in the steady magnetic field or modulating field over the length of the coil form, or from differences in phase shift in the audio stages of the receivers used were studied, the indications being that errors from these effects did not exceed those in the frequency measurements.The result of the measurements gives for the ratio of resonant frequencies KTl 205 )A(/f 1 ) = 0.577135=fc.00005.

No abstract