Conventional geochemical exploration techniques involving the sampling and analyses of stream sediments, soil, and bedrock have found limited application in the areas of glacial overburden with indefinite and disorganized drainage systems such as exist in large regions of the Canadian Shield. In such areas lake sediment composition could be an indicator of mineralization if either the host rock or the mineralization itself has some diagnostic geochemical characteristic which is transmitted to lake sedments. If these conditions are fulfilled, lake sediment sampling would offer a convenient procedure for reconnaissance-scale exploration in Shield areas. The usefulness of the procedure has been previously demonstrated in permafrost areas of Canada. In the southern Shield, where organic production is much greater, trace element dispersion in lake sediments is controlled by a number of factors contributing to the existence of a complex relationship between lake sediment geochemistry and mineralization. Orientation surveys were carried out over four greenstone belts within the Superior Province to investigate the nature and factors affecting trace element accumulation in lake sediments in order to evaluate the feasibility of utilizing lake sediments in reconnaissance exploration. These investigations indicated that the most representative and homogeneous sample of a single lake basin occurs in the central regions of a lake. It was difficult, if not impossible, to distinguish a lake adjacent to mineralization from one in barren terrain on the basis of the single element compositions of the organic-rich material from the lake centers. This is due to the existence of various dispersion processes, related to coprecipitation and varying pH, that affect the distribution of trace elements in lake sediments. Trace element dispersion in lake sediments related to massive sulfide mineralization as opposed to other causes can be identified by screening the Zn content of the sediments on the basis of Zn/Mn ratios. A reconnaissance survey based on the collection and analysis of organic-rich lake sediments from lake centers showed that the distribution of Ni and Zn when screened by ratioing Ni/Mn and Zn/Mn, respectively, can be used to identify lakes adjacent to the Ni-Cu and massive sulfide mineralization, respectively. The Shebandowan Ni-Cu deposit is clearly indicated by a group of lake sediment samples with anomalous Ni/Mn values extending southwest and down ice from the deposit. The distribution of Ni in the Shebandowan Lake area corresponds markedly to that of Cr indicative of a common source, in this case basic and ultrabasic strata. Lake sediment samples with anomalous Ni/Mn values, however, appear to have an additional component of Ni which is not related to coprecipitation (Mn) or bedrock (Cr) and which is believed to be a reflection of Ni-Cu mineralization. In the Sturgeon Lake area a group of samples with anomalous Zn/Mn values encompasses the Mattabi and Lyon Lake massive sulfide mineralization. There is also a group of anomalous samples, confirmed by followup work, to the east of the Mattabi and Lyon Lake deposits. This anomaly occurs in a favorable geological environment and is associated with geophysical conditions indicative of possible mineralization. Final assessment of the technique must await the results of the current program to evaluate the significance of the anomalies revealed by the reconnaissance. Provisionally, however, it is concluded that sampling and analysis of organic-rich lake sediments followed by screening of the data may be a viable exploration procedure in the search for Ni-Cu and base metal massive sulfide deposits in the southern areas of the Canadian Shield.