Tholeiitic diabase dikes that trend northwest-southeast, parallel to the coastline, are common in northwestern Liberia. K-Ar whole-rock and mineral ages determined from dikes that intrude Precambrian crystalline rocks are discordant and range from 186 to 1,213 m.y. Incremental heating experiments on three neutron-irradiated samples of these rocks give “saddle-shaped” 40Ar/39Ar release diagrams that reach minima of less than 300 m.y. at intermediate temperatures and that do not fit a 40Ar/36Ar versus 39Ar/36Ar isochron. K-Ar ages determined from diabase dikes and sills that intrude Paleozoic sedimentary rocks near the coast are all within the range 173 to 192 m.y. 40Ar/39Ar incremental heating data for one of these samples gives a plateau age and a 40Ar/36Ar versus 39Ar/36Ar isochron age that are concordant with the conventional K-Ar age. The conventional and 40Ar/39Ar K-Ar data show that the dikes intruding Precambrian basement rocks contain large and variable amounts of excess 40Ar, whereas the diabase intruding Paleozoic sandstone does not. All of the intrusions are probably earliest Jurassic in age.
Mean paleomagnetic directions in six dikes and sills that intrude sedimentary rocks are nearly parallel to mean paleomagnetic directions in 19 dikes that intrude Precambrian rock, further evidence for contemporaneity. The paleomagnetic pole derived from all 25 diabase units is at lat 68° N., long 242° E., with α95 = 5°, in close agreement with other Mesozoic paleomagnetic poles from the African continent. A mean paleomagnetic pole for northwest Africa has been calculated using these data and published paleomagnetic directions from 19 other intrusive rock units that have similar radiometric ages in Morocco and Sierra Leone. This pole is compared with another paleomagnetic pole calculated from published data from 16 localities in igneous rocks of latest Triassic to earliest Jurassic age distributed from Nova Scotia to Pennsylvania. The comparison shows that, with the African and North American continents in their present positions, the two poles differ by 44° of arc, but when the continents are restored to the predrift configuration proposed by Bullard and others (1965), the angular difference diminishes to 3°. This coincidence of paleomagnetic poles provides an earliest limit of 180 ± 10 m.y. for the separation of Africa from North America.