No abstract

The directional correlation between the 1930-keV 0 group and the 551-keV y in the decay of 5-day 148 Pm has been measured. The anisotropy is found to be only a few percent over the fi energy range from 1050 to 1800 keV. Directional correlations for the following 7-7 cascades in the decay of 41-day 148 Pm have been measured: 414-630 keV, A 2 = -0.07±0.03; 414-551 keV, A*=* ~~0.04db0.03; 288-630 keV, ^2=0.17±0.02; 288-551 keV, J2=0,19±0.02. A note on simple properties of directional correlations involving unobserved y rays which hold in special cases is appended.

No abstract

LETTERS TO THE EDITORcomes nearer to c\ as the temperature is decreased (i.e., as the mean free path becomes longer) and as the sensitivity of the receiver is increased. In this picture, the sharp rise in velocity at intermediate temperatures is caused by the disappearance of the high temperature excitations (bosons or rotons). As soon as these have disappeared, the velocity of heat pulses should become ci/y/3, and remain there until the mean free path for phonon-phonon collision becomes long, at which temperature it should rise to near c\. The pressure dependence should be in the direction found experimentally, since (as mentioned by Maurer and Herlin) the number of phonons at a given temperature decreases with pressure, and hence any long mean-free-path effect should increase with pressure.The shape of the received pulse at low temperatures and at all pressures observed shows the behavior reported by Pellam and Scott 9 and by Atkins and Osborne. At the lowest temperatures, where the velocity is high and does not vary rapidly with temperature, the pulse shape is essentially that reproduced in the letter of Atkins and Osborne. The rise is fairly sharp, with a well-defined foot, but the tail is extremely long. At the intermediate temperatures, coincident with the rapid drop in velocity with temperature, the tail becomes shorter, the rise longer, and the foot less well defined, so that the pulse becomes symmetrical and assumes a roughly Gaussian shape. At the higher temperatures the entire pulse becomes well defined. The extremely large dispersion in the intermediate-temperature region may possibly be due to the interaction of the phonons and the higher temperature excitations, since in this region they are competing for dominance.

1259 the yield of Er iw is high by a factor of 6 from that obtained from the Gaussian curve. We feel that most of this discrepancy is due to a combination of Er, Tm, and Yb target impurities. Part of it could also be due to radioactive impurities in the low-counting Er samples.These two sources of error operate in opposite directions, one tending to lower an observed yield, the other tending to raise it. An error in one measurement changes the Gaussian curve that we have fitted, using that value, thus obscuring errors in other measurements. Fortunately, the two sources of error are probably not important for any one product. The combined effect is then a general increase in scatter from the smooth Gaussian curve. As noted before, the average deviation from the curve is about 15%, which is not much greater than the average accuracy of the individual results. SUMMARYWe have measured the fission yields of a number of products formed in low yield from 14.8-MeV neutron fission of Th 232 , U 235 , and U 238 . The results show that, within experimental uncertainty, the wings of the mass-