sections are found for the two target nuclei with respect to both two-and four-nucleon ("d" and "a") removals, a large difference is observed in the removal of both one and three nucleons. In 27 A1 we observe the removal of a proton and also of a neutron. In 28 Si, on the other hand, the removal of a proton is clearly observed and with large cross section, while the neutron removal is very weak, if at all present (see Table I). Also, in 27 A1 a "t" removal is observed with a large cross section, but a " 3 He" removal is absent, while the opposite is observed in 28 Si.(3) The fact that in 28 Si the proton removal is much stronger than the neutron removal is surprising in view of the fact that the levels populated by removal of a neutron or a proton (in 27 Si and 27 A1, respectively) are mirror states. From symmetry it is expected that the cross sections for (n",ir~p) and (ir + ,ir + n) to the mirror states should be equal-and this is not the case (Coulomb effects can hardly explain this difference). It should be noted that neutron and proton pickup reactions on 28 Si yield similar spectroscopic factors for the mirror states. 7 * 8 A similar difference is observed in the removal of other clusters such as 2p-2n, "f"-" 3 He", and " 5 He"-" 5 Li."(4) Finally we consider the ratio of cross sections for the knockout of a proton by positive and negative pions leading to a single final state. An equivalent ratio for neutron knockout leading to many final states has recently been the subject of theoretical and experimental studies. 9 It was found that over a wide energy range the experimental ratio was smaller (by a factor of ~3) than the ratio predicted from the free pion-nucleon scattering in a simple impulse approximation. In the present work we find that at 70 MeV the value of a(7r + ,7r + /))/[a(7T",7r"/)) + a(7r", ir°n)] is 0.52 A recent paper 1 reported angular distributions for the process 32 S( le O, 12 C) 36 Ar leading to the ground and first excited states of the residual ±0.16 for 27 A1 and 0.28 ± 0.13 for 28 Si. These numbers are again smaller than the value of about 1.3 expected from the simple impulse approximation. 9The differential cross section for 32 S( ie O, 12 C) 36 Ar has been analyzed with the finiterange distorted-wave Born-approximation computer code MERCURY (which includes recoil effects exactly). The general features of the data were reproduced only by assuming that the a particle was transferred in an excited state. Good fits to the data were found by also including 1,-dependent absorption in the optical potentials.
946