Since the mid 1950s, limonite deposits of a new type have been found in the northwest of Western Australia. The deposits consist of extensive fossil valley fillings up to several hundred feet in depth, of medium to high grade limonite, frequently containing limonitized fossil wood, and of Mesozoic or Tertiary age. The distribution of the deposits is closely linked with hilly outcrops of Archaean and Proterozoic jaspilite in the headwaters of the old valleys, from which the iron has been derived by leaching, and presumably re-precipitated by a process not unlike that which operates in forming bog iron deposits. Due to the relative resistance of tlie limonite to erosion, the rejuvenation of streams in the area in recent times has left limonitic remnants as mesas up to several hundred feet above the levels of the present creeks, the courses of which generally are adjacent to or within a few miles of those of the old limonite filled valleys.The Robe River limonite deposits are the largest of this type known and can be traced for approximately 130 miles along the old stream, over 30 sq. miles of limonite remnants remaining at the present time.The present quantity of limonite in this system alone is probably in excess of 5,000,000,000 tons, and the original quantity must have exceeded four times this tonnage when allowance is made for erosion.Some thousands of millions of tons of this material contain in excess of 54 per cent Fe, with approximately 10 per cent combined water. The chief impurities are SiO2 and Al2O3, and minor elements are not present in objectionable amounts. A portion of these deposits was extensively tested by The Broken Hill Proprietary Company Limited (the B.H.P. Co. Ltd.) during 1962-1964. Most of the data presented in this paper are based on the work done on these deposits which are typical of those discussed. The overall problem of the origin of the limonite is discussed. It is apparent that the limonite originates from the jaspilitesof Archaean and Proterozoic age which form the highlands at the upstream ends of the deposits. The iron oxide has been chemically and/or mechanically derived from these jaspilites and deposited by one or more of the following mechanisms.A.-Chemical-sedimentary mainly by biological agenciesB.-Replacement of fluviatile sediments by iron-charged watersC.-Clastic sedimentary processesD.-Lateritization of fluviatile sediments.These various mechanisms are discussed; the authors favour machanismA.Manuscript received' at The Institute, 14th July, 1964.