If the specifically nuclear part of nuclear interactions is charge independent then the energies of the various members of an isobaric multiplet will differ only because of the neutron-proton mass difference and the electromagnetic interactions among the protons. These electromagnetic contributions are calculable 1 and could be removed; the remaining masses of the multiplets thus could be examined with regard to charge independence in nuclear forces. These calculations, however, are somewhat model dependent, but to first order in the Coulomb energy it has been shown, 2 quite generally, that the masses within an isobaric multiplet are characterized by
M(A,T,T ) = a(A,T) + b(A,T)T +c(A,T)Twhere M is the mass of a member of the multiplet, A is the number of nucleons, T is the isobaric spin, and T z = |(AT-Z); a(A,T), b(A,T), and c{A,T) are taken as constants within a given multiplet. The adequacy of this formula has never been tested empirically because at most three members of a multiplet are known so that no verifiable predictions can be made using Eq. (1).Recent (p,t) experiments, 3 however, have been able to find a T = 2 level in certain T z = 0 nuclei. The T = 2 level that is located is the isobaric analog of the ground state of the T = 2, T z = 2 isobar. For example, a T = 2 level is found in Mg 24 that is the analog to the ground state of Ne 24 . In the mass-24 system the three lowest lying T = 1 levels with T z = 0, ±1 are also known. In a recent paper 4 Wilkinson suggested that one assume the coefficients b{A,T) and c(A,T) within the same A be taken to be T independent. With this assumption, using the levels mentioned above he was able to show that the resulting prediction for the mass of the ground state of Al 24 is in agreement with the observed mass, 5 though the experimental uncertainties are large.We have just completed a study 6 of (p,t) and (£,He 3 ) reactions on O 18 and Ne 22 . These reac-tions on O 18 allowed us to locate the analog to the ground state of C 16 in N 16 (9.91 ±0.