The ninth satellite of Jupiter was discovered and its orbit first computed by Nicholson in 1914 at the Lick Observatory. His most recent orbit,1 epoch about 1940, which we have been carrying on for a number of years, latterly with automatic computing machinery, has proved to be one of the most accurate of all those determined for Jupiter's satellites. The representation of his most recent observation in 1951, inclusive of the usual solar perturbations only, gave residuals of the order of 15", but when the perturbations of Saturn were included, preparatory to a differential correction of the orbit, the residuals were reduced to practically zero, thus removing the indication that an improvement might be desirable. An intensive search by Nicholson in 1938 resulted in the discovery of JX and JXI. After 1938 he photographed Jupiter's faint satellites many times but J XII, if recorded, was not recognized until September 28, 1951, when it was detected on a photograph taken to obtain the position of J X. As soon as the observations of this fall and winter have enabled us to determine a sufficiently accurate orbit, we shall endeavor to carry it back to 1938, to see if the new satellite can be found on the Mount Wilson search plates. Our first orbit of Jupiter XII,2 November 14, 1951, was a "disturbed solution" taking into account the perturbations of the sun as first proposed by Leuschner3 but patterned on the method of Gibbs,4 which proved to be well adapted to such work. The excellent results obtained are a tribute in part to the adaptability of the Gibbsian procedure and in part to the accurate and painstaking calculations of Charles Geoffrey Hilton and Joseph L. Brady, Jr. The elements obtained, together with those of a confirming orbit published by Cunningham5 a week later, those of our latest corrected orbit, and averaged arc residuals for the basic and later dates, are shown in the following tabulation :