Assessment of available geochronological information, as well as new whole-rock Rb-Sr data from several granitoid rocks of Saskatchewan, shows a close relationship between magmatic-metamorphic events in the Hudsonian orogen and uranium mineralization. Most uranium deposits lie to the west of the Needle Falls Shear Zone and occur as either: (i) vein-type deposits or (ii) unconformity-type deposits close to the contact between the Athabasca sediments and their basement. At least two metamorphisms have affected the pre-Athabasca rocks: the Kenoran at about 2500 Ma ago, and the more pervasive ‘main’ Hudsonian event at 1740 Ma. A much younger thermal event (perhaps associated with uplift and cooling) at 1540 Ma is also indicated. The post-Kenoran K-Ar dates suggest prolonged thermal activity from about 1900 Ma through to about 1500 Ma ago. Granitoid events at 1870 Ma and 1740 Ma ago are outlined by both U-Pb zircon and Rb-Sr whole-rock isochron data. Whole-rock Rb-Sr data from the unmetamorphosed Athabasca sediments suggest an approximate depositional age of 1450±50 Ma, a figure that is consistent with the age of the underlying Hudsonian basement and the truncation of the sediments by the Cree Lake diabase dyke swarm at about 1200–1300 Ma ago. Although several episodes of uranium deposition have been documented, the main ones seem to have occurred at 1860 Ma (syngenetic uraninite in pegmatites), 1740 Ma (the Beaverlodge vein-type deposits) and between 1300 and 800 Ma (the epigenetic uranium of the unconformity-type deposits). Whereas the two earlier episodes can be correlated with periods of either magmatic or metamorphic activity, the late Proterozoic episodes cannot. The close agreement between the age of the Cree Lake dyke swarm and the late Proterozoic mineralization suggests that at about 1300 Ma ago possible hydrothermal activity from relatively deep-seated fractures may have been responsible for the solution and transportation of the uranium of the unconformity-type deposits. The period 1300 Ma to about 900 Ma, in other parts of the Canadian Shield, was a time of crustal rifting, basic magmatism, carbonatite activity, and intense deformation. Prior to the deposition of the Athabasca sediments uranium was concentrated by Hudsonian magmatic and metamorphic processes whereas subsequently, transportation and intermittent deposition of the unconformity-type deposits were related to fairly long-lived, low-temperature hydrothermal activity.

In Niger, uranium occurs in upper Palaeozoic and lower Mesozoic continental sedimentary basins west of the Aïr mountains, but the source of the uranium had not been identified. Geochemical studies and fissiontrack observations on alkaline ignimbrites preserved in two Palaeozoic anorogenic centres in Aïr show that uranium is concentrated in the matrix and on secondary iron-oxide coatings surrounding lithic and crystal fragments. Based on variable Th/U ratios and degree of oxidation, it is concluded that the original ignimbrite field was enriched in uranium, but that a considerable proportion was leached during the weathering of the volcanic pile. Tectonic uplift, anorogenic magmatism, followed by weathering and erosion of the volcanic cover, with sedimentation in nearby continental basins, have all contributed to the development of uranium mineralization in Niger. The petrological and geochemical similarities between the Palaeozoic ring complexes in Niger and the Nigerian Mesozoic ring structures suggest that sedimentary uranium deposits may also be found in Nigeria if the tectonic and sedimentological controls were favourable.

Enriched concentrations of uranium have been discovered in the exposed granitic roof zones of the Nigerian subvolcanic centres, with Th/U ratios approaching unity. Thus local vein deposits of uraninite, as well as dispersed uranium in recent sedimentary horizons, could be discovered particularly in the drainage systems entering the Chad Basin.

Field gamma-ray spectrometry is a rapid and effective quantitative method of mapping variations of radioelements within igneous intrusions. A field procedure for radioelement mapping demonstrates the value of the method to studies of late-stage magmatic processes during the emplacement of granite intrusions and the determination of their present-day heat productivities. Methods and problems of instrument calibration using both natural and artificial sources are discussed. Calibration based on neutron activation analysis of samples from natural outcrops achieves results comparable with those obtained using artificial sources and has the advantage that it relates directly to field conditions; furthermore it enables secular disequilibrium in the uranium and thorium decay series to be recognized.

Some genetic models for Lower Proterozoic gold- and uranium-bearing pyritic conglomerates favour a modified placer origin in which low levels of atmospheric oxygen are used to account for the survival of uraninite and pyrite. There are many difficulties with such models—for example magnetite is absent in the ore-bearing horizons although it is stable in anoxic conditions, while it is abundant in over- and under-lying strata. Compact and porous pyrite grains are not in hydraulic equivalence and the deposits lack a normal detrital heavy-mineral assemblage. Moreover uranium and gold show evidence of diagenetic remobilization, the uranium becoming associated with secondary titaniferous phases and uranium and gold being enriched in reduzate facies sediments.

New evidence concerning the genesis of the deposits is derived from a clast of ferric iron clay thought to represent a precursor sediment of the Witwatersrand Basin. Reworking of such clays and transport of a magnetite and ferric clay assemblage with subsequent sulphidation, could account for the porous pyrites, the absence of magnetite and the lack of hydraulic equivalence of the mineral grains in the conglomerates. The presence of oxygen, as indicated by the ferric iron clasts, would account for the evidence of mobility of uranium and of gold and would enhance their extraction from source rocks; particularly the release of gold from a precursor auriferous iron formation source. It is suggested that some aspects of the genesis of uranium deposits of the Witswatersrand and Elliot Lake may be similar to those of the Phanerozoic ‘Roll Front’ ores involving interaction between oxidizing uraniferous groundwaters and previously sulphidized and reduzate facies sediments.

The concentrations of U, Th, and K in a charnockite-alkali granite intrusive complex from SW Sweden and associated, partly charnockitized country rocks are presented. The charnockitic components have constant low U and Th levels throughout the compositional range with a mean Th/U ratio of 1.2 to 1.5. The transition from charnockite to granite is marked by a sharp increase in Th and U concentrations with mean Th/U ratios increasing to 3.4 4.4. However, the K concentration increases relatively smoothly with increasing acidity through the compositional range studied. Late shearing of the alkali granite led to significant Th depletion with little change in the U and K concentrations. Charnockitization of the country rock granite gneisses is marked by depletion in Th and K while the U content is almost unchanged.

It is suggested that the behaviour of U and Th reflects the tendency for Th to associate with magma-dissolved water whilst U shows a greater affinity for coexisting free volatiles. In the absence of a significant free volatile phase during crystallization of this charnockite-granite suite, U was incorporated into primary mineral lattice sites whereas Th entered sites from which it could be liberated more easily.

Hydrothermal uranium veins, associated with the Hercynian leucogranites show important variations in their mineralogical, chemical and structural features in relation to the host rock lithology. These are described with particular reference to the Chardon deposit, Vendée where the veins cut granite, basic lithologies, and shales. The following features are described:

1. 1 Changes in the thickness of veins near to contact zones, particularly those between granites and basic lithologies, lamprophyres, and shales.

2. 2 Changes in the gangue mineral assemblage with the preferential development of carbonate in veins cutting basic lithologies, and of silica in veins which cut granite.

3. 3 Paragenetic zoning in the veins in passing from granites to their metamorphic aureoles.

Comparisons between deposits of Vendée, Limousin, and Erzgebirge allow the following generalizations to be made:

1. 1 Open faults and subsequent mineralization are concentrated at boundaries between competent and more plastic lithologies.

2. 2 Mineralizing fluids cause wall-rock alteration characterized by the removal of Si from granite and of Ca, Mg, Fe from metamorphic and basic rocks.

3. 3 The chemical and structural characteristics of wallrocks are important controls on the mineralization but in acid lithologies the main controls on the pitchblende vein formation are the structural characteristics of the wallrock.

The availability of reconnaissance scale geochemical maps for large areas of Canada enables spatial associations between major crustal structures and surface uranium content to be identified. Maps of the distribution of uranium for an area greater than 2 million km2 compiled from airborne gamma-ray spectrometry data are supplemented by maps for uranium, based on stream and lake sediment and some bore hole sampling. These are examined in relation to gravity, aeromagnetic and geological maps.

The radioelement distribution can be related in detail to exposed bedrock and surface geology, but in addition there is evidence of the control of uranium distribution by major structural features which are marked by granitoids containing elevated levels of radioelements; several of these granitoids are associated with large negative Bouguer gravity anomalies. The distribution of such granitoids appears to be related to ‘megashears’, as in the case of the South Mountain batholith in Nova Scotia, or zones of tension. A belt of uranium enrichment, the Athabasca axis which is characterized by uraniferous granitoids with negative Bouguer gravity anomalies and associated tension faulting extends 2500 km northeast-ward from Edmonton, Alberta to the Melville Peninsula. This structure passes under the Athabasca basin which contains many large uranium deposits. Recent evidence that granitoids enriched in radioelements can provide low-grade heat sources over periods of hundreds of millions of years, capable, in favourable situations of maintaining low-temperature hydrothermal circulation is considered to account for uranium and other mineralization in the basin. It is suggested that the transfer of radioelements into the crust at the end of the Lower Proterozoic, was a factor in the stabilization of North American craton.

The lead isotope systematics, zircon uranium concentrations and whole-rock rubidium concentrations of 11 Scottish Caledonian granites are examined for clues to the origin of their uranium. A positive correlation between U in zircon and initial lead isotope ratios suggests that U and Pb were derived from the same source which, as some of these granites contain their U in inherited zircons, is likely to have been within the crust. It is argued, therefore, that most of the uranium in these granites had a crustal derivation but lead isotope ratios indicate that any Lewisian contribution was minor in comparison with those from postulated Grenville, Morarian or Caledonian metamorphic reservoirs. However, additional data are required before this conclusion can be extended to include uraniferous Caledonian granites such as Cairngorm.

The distribution of uranium in a suite of variably deformed and metamorphosed rocks from the leucocratic member of the Glendessarry syenite has been determined using the fission track method. The uranium content of the magma increased during crystallization and uranium was concentrated in accessory minerals such as monazite, zircon, sphene, allanite, apatite, and microinclusions of a Zr- and Ti-rich phase. Contamination of the magma by pelitic metasediment enhanced the uranium content and monazite and zircon formed instead of sphene, allanite, and apatite.

Evidence of subsolidus uranium mobility in late stage magmatic or metamorphic fluids is presented here and shows: (a) Intracrystalline redistribution of uranium, especially in grains of sphene. (b) Intergranular mobility in a fluid phase, which affected the uraniferous accessory minerals in several ways.

A method of epithermal neutron activation analysis for uranium and thorium is discussed in relation to other rapid analytical methods for these elements. Caledonian granites are divided into four main groups on the basis of U and Th contents: (a) a group of evolved late granites with high U and Th contents and high Th/U ratios, (b) a group of uranium-enriched mineralized granites, (c) granites containing approximately average levels of U and Th, and (d) a group of granites with low thorium content.

Trace element variations in stream sediments from an area of 76 000 km2 in central Colorado are used to identify uraniferous granitoids on the basis of whole rock geochemical criteria developed to distinguish barren from metalliferous granitoids in Britain. These criteria (which include enhanced Ba, Be, Cs, Cs/Ba, K, La/Eu, Li, Lu/Eu, Nb, Rb, Rb/K, low Sr and Mg, and RE patterns with marked negative Eu anomalies) are used to formulate an index based on the Pikes Peak batholith of the Front Range as a type uranium source rock.

Uraniferous granitoids in Colorado, which are associated with sedimentary basins containing major uranium mineralization, are identified using this index which may be applicable to the interpretation of stream sediments from elsewhere. The use of stream sediment geochemistry as an exploration method for similar uranium source rocks, which may indicate potential uranium provinces, is thus possible.

The uraninite-albite veins of the Mistamisk area occur in the argillite member of the Dunphy Formation, which is near the base of the slightly meta-morphosed Lower Proterozoic sequence of the central Labrador Trough. The vein minerals are albite, uraninite, dolomite, and chlorite, and minor quantities of quartz, tellurides, sulphides, gold, and organic material. Pitchblende and calcite are related to late remobilization.

The veins were deposited in fractures by hydrothermal solutions, and metasomatism caused albitization of wall rocks. Fluid inclusions have an unusual composition, described here for the first time in connection with soda-metasomatism; the aqueous solution of the inclusions is oversaturated in NaCl and contains Ca2+ and Mg2+, and the gas phase mostly consists of N2, CO and CO2. The presence of hematite and absence of hydrocarbons indicates that the vein-forming solution was oxidizing.

The temperature and pressure of vein formation, estimated from fluid inclusion data, was 300°–350° and 2.5 kbar respectively consistent with the composition of the phengite which is a common metamorphic mineral of the host rock. Vein emplacement occurred in the waning stages of the Hudsonian Orogeny, the hydrothermal solution possibly originating by metamorphism of sodic schists of the Mistamisk area, which are possibly of evaporitic origin.

Data for U, Th, and K, determined by neutron activation analysis, in 128 rock samples from 14 Irish Caledonian and Tertiary granite plutons are sum-marized. Irish Newer Caledonian granites (ninety-four samples) have values of radioelements (average 3.6 ppm U, 12.1 ppm Th, and 3.4% K) consistent with derivation from a source of broadly uniform composition. U mineralization occurs in the voluminous and forcefully emplaced Main Donegal and Leinster plutons whereas smaller passively emplaced granites (Galway, Barnesmore, Rosses) lack mineralization. Irish Tertiary granites (thirty-four samples), which represent small batches of highly fractionated acid magmas, show moderate or high radioelement contents (11.5 ppm U, 39.9 ppm Th, 4.5% K) but are not mineralized. It is concluded that the use of elevated radioelement contents as a criterion of granite ‘fertility’ may be misleading in the British Isles Caledonian Province. The radioelement chemistry of granites may reflect their source composition more closely than expected for differentiated and/or contaminated acid melts.