428 ANNUAL REPORT SMITHSONIAN INSTITUTION, 19 64 



well within the limits imposed by statistical fluctuations and lack of 

 absolute knowledge concerning the neutrino spectrum. 



Evidence that the first pulse pair was due to a positron. — Here we 

 had two checks. We had dissolved a known positron-emitting radio- 

 active material (copper-64) in the water of a target tank and observed 

 the pulse amplitude spectrum obtained from it. The spectrum of 

 pulses in the first pair of reactor-produced events agreed with it 

 nicely. The second check consisted of placing thin sheets of lead as 

 an absorber between the water targets and the scintillation detector 

 tanks. By measuring the reduction in counting rate produced by 

 the lead, we could check the energy of the gamma rays in the first 

 pulse. They were found to be the two simultaneous gamma rays 

 produced when a positron-electron pair combines (or "annihilates," 

 in the vernacular) . 



Evidence that the second pulse pair was due to the capture of a 

 neutron by cadmium, and that the neutron had appeared in the 

 detector simultaneously with the positron. — Again, we had two 

 checks of this. We varied the amount of cadmium salt in the water 

 targets and observed the varying times for observation of the second 

 pulse following the first. These checked with the same data when a 

 known neutron source was placed near the detector and neutron 

 capture times measured. These capture time curves had already been 

 run on computers at Los Alamos for different cadmium concentra- 

 tions. These also agreed. The second check was the total pulse am- 

 plitude spectrum. This agreed with that obtained with known neu- 

 tron sources. The pulses were due to neutrons. The capture time 

 curves also demonstrated that the neutron had appeared with the 

 positron, for it was the interval between the two that was measured, 

 and this interval would not have checked had this not been so. Three 

 different runs were made with different cadmium concentrations. 



Dependence of the signal rate on the number of protons in the 

 target. — For this check, we reduced the amount of hydrogen in the 

 target to half, but did not reduce the amount of water. This was 

 done by replacing the ordinary water with a mixture of 50 percent 

 ordinary (light) water and 50 percent heavy water. Thus, 50 percent 

 of the hydrogen had been replaced by deuterium, which has a com- 

 paratively very low cross section for antineutrinos compared with 

 hydrogen. The signal rate fell when this was done as expected. 

 This checked another point at the same time. By putting deuterium 

 into the detector, we were sensitizing it to the effects of gamma rays 

 and neutrons. Such backgrounds can easily break up a deuteron 

 and mock up an antineutrino signal. Therefore, if the gamma ray 

 and neutron backgrounds were fooling us before, the signal rate 

 should have increased now rather than decreased as it was observed 

 to do. 



