The ionospheric phenomenon of "Spread-F" makes itself manifest on ionosonde records by additional traces, which are generally similar in shape to the main h 'f trace, but with critical frequencies (f0F2) which may be greater than, or less than, that of the main trace. It is seen at Brisbane associated with the one-hop trace obtained at night, the frequency of occurrence depending on season, and state of the sunspot cycle (McNicol, Webster, and Bowman, 1956 - subsequently referred to as A; Singleton, 1957). The "Spread-F" records at Brisbane display a variety of forms. Observations of the satellite traces at the low frequency end of ionograms, where retardation can be neglected, indicate, at certain times, ranges more than 10 km greater than the true range of the main echo. At other periods the true range of the satellite echo is little greater than that of the main trace, but the critical frequencies are different. The first type of spreading has been called "range-spreading'', and the other "frequency-spreading". Other ionograms showing "Spread-F" merely exhibit diffuse traces. Evidence presented in this paper indicates that nearly always these result from satellite traces so close in true range or critical frequency that the resolution of the equipment is insufficient to separate them. It is convenient to classify spreading into the groups "resolved" and "unresolved", as illustrated in Plate 3, on the basis of the ionograms. It must, however, be remembered that such distinction is a function of the instruments used for observation. Many borderline cases occur. A study of resolved range-spreading, using mainly normal-incidence reflection, has already been published (A). Further information on this phenomenon, together with unresolved range-spreading and unresolved frequency-spreading, has now been obtained employing a rotating spaced-loop direction-finding system, based on that of Ross, (1947). This was located at Brisbane (lat. 27 5°s., long. 152.9°E., geomag. lat. 35.1°s., magnetic dip 57°) and observations made on reflected pulses originating from (a) a transmitter at Armidale, distant 355 km and bearing 202° (from true north) from Brisbane, and (b) a transmitter located close to the loops, and operating on the same frequency (3.84 Mc/s). The spaced-loop direction-finding system while it does not provide information on the elevation angle, as does that of Thomas and McNicol, (1955), has the advantage of allowing the azimuths-of-arrival to be determined, even when satellite echoes are closely spaced in range. In addition, further analysis of h'f and other records described in A, has been undertaken.