1. F. Amman et al.:Proceedings of the International Conference on High-Energy Accelerators (Dubna, 1963), p. 249;F. Amman et al.: Proceedings of the V International Conference on High-Energy Accelerators (Frascati, 1965), p. 703;F. Amman et al.: Proceedings of the International Symposium sur les Anneaux de Collisions (Saclay, 1966), III-2-1;F. Amman, R. Andreani, M. Bassetti, M. Bernardini, A. Cattoni, V. Chimenti, G. F. Corazza, D. Fabiani, E. Ferlenghi, A. Massarotti, C. Pellegrini, M. Placidi, M. Puglisi, F. Soso, S. Tazzari, F. Tazzioli andG. Vignola:Lett. Nuovo Cimento,1. 729 (1969).
2. B. Bartoli, B. Coluzzi, F. Felicetti, V. Silvestrini, G. Goggi, D. Scannicchio, G. Marini, F. Massa andF. Vanoli:Nuovo Cimento,70 A, 603 (1970).
3. The absorbers in the telescopesT i and in front of the CR counters set the following limits on the kinetic energyT of the observed particles: ∼80 MeV ≲ ≲T≲350 MeV. The evaluation of the corresponding corrections (due to their smallness) does not critically depend on the momentum distribution of the emitted particle; we have used the experimental momentum distribution of pions from $$p\bar p \to 2\pi ^ + + 2\pi ^ - $$ annihilations (see,e.g.,C. Baltay,P. Franzini,G. Lütjens,J. C. Severiens,D. Tycko andD. Zanello:Phys. Rev.,145, 1103 (1966)).
4. We recall that the luminosity ℒ (essentially the product of the e+ and e− beam intensities divided by their effective section) is a measurement of the machine intensity. The number of events produced is directly given by the product ℒ·σ, where σ is the cross-section of the process.
5. Our «source» has a finite longitudinal dimension and is expected to have a Gaussian distributionN(l)=N 0 exp $$\left[ { - l^2 /2\bar l^2 } \right]$$ . Theoretically $$\bar l$$ is expected to be $$\bar l\left( {cm} \right) = 20E_ \pm ^{\frac{3}{2}} $$ and experimentally has been measured to be $$\bar l\left( {cm} \right) = \left( {22 \pm 2} \right)E_ \pm ^{\frac{3}{2}} $$ , whereE ± is in GeV. (Private communications from the Adone machine staff.) As far as the ∈(n)’s are concerned, their dependence on $$\bar l$$ is $$\varepsilon \left( n \right) \propto l/\bar l$$ , but since our conclusions involve only ratios of ∈(n)’s they are independent of the actual value used for $$\bar l$$ .