ABSTRACT. There is substantial evidence that most of the material in the envelopes of some globular-cluster giants is mixed so deeply into the interior that the abundances of many of the light elements, including magnesium and aluminum, are reshuffled by proton capture before the material is returned to the envelope above. In order to understand these abundance changes we must know all of the appropriate proton-capture rates. We have recalculated the thermonuclear reaction rates, N A (av) T , for the 24 Mg(p,y) 25 Al reaction using the canonical 57 temperatures from Caughlan and Fowler (At. Data Nucl. Data Tables, 40, 283, 1988) in the range from T9 =0.004 to T9 = 10.0, where T9 is the temperature in units of 10 9 K. We have done these calculations using a range of values for the unmeasured total width, T l , of the lowest resonance, (l/2) + , in the p+ 24 Mg system. This resonance is located at 2486 keV excitation in 25 A1 and there is a large uncertainty in the value of . The rates currently used are based upon a value of T 1 =75 meV, whereas the most recently accepted experimental value is an upper limit for of 32 eV. For values of T9 <0.07 the rates are quite sensitive to this uncertain width. We discuss the consequences of this uncertainty and the need for experimental data to determine Tj. We have also found that for T9 >5.0, the recalculated rates are somewhat higher than those in Caughlan and Fowler (1988). We present our rates in tabular form.