Fig. 3: ⁴⁰Ar/³⁹Ar isotopic dates versus longitude.

A Map produced in GeoMapApp showing the VER and microplate morphology. Other details as in Figs. 1, 2. B Age-distance relationship for the SO80 sample dates in Supplementary Data 1 defining two diverging trends corresponding to their ___location on the Pacific (red diamonds) or microplate (blue boxes) crust. The Pacific trend (64.6 mm/yr; solid red line and diamonds) is defined by dates from the southern ends of VER 5 (Crough Seamount; 67DS & 69DS), VER 4 (Thomas Seamount; 66DS), and VER 1 (63DS & 62DS). The date for 64DS at the southern end of VER 2 is aligned with the trend and is not a plateau age (see discussion in the results section). The microplate trend (191.6 mm/yr; diagonal solid blue line and boxes) is defined by dates (blue squares) from the northern end of VER 1 (76DS & 74DS), the inner (53DS) and outer (52DS) pseudofaults of the West Rift, and a small VER (51DS) associated with the poorly defined triple junction region⁹. VER 1 was erupting and connecting both trends at about 3.5 Ma and continued at the southern end until 2.7 Ma. The microplate trend is interpreted here as a proxy for the propagation of the SW Rift (ref. ⁹). The red triangles are for published Ar/Ar dates < 10 Ma for samples from the SYGR¹³. Note that sample DR72-2 from the submarine slope of Salas y Gomez is a highly alkalic sample (basaltic trachyandesite). The small yellow boxes represent published ⁴⁰Ar/³⁹Ar dates for submarine volcanism at Umu, Moai, and Easter Island¹². Additionally, a compilation of published K-Ar ages indicates that the three shield volcanoes on Easter Island—Poike, Rano Kau, and Terevaka—underwent two phases of activity between 0.78 and 0.3 Ma and 0.24 and 0.11 Ma¹⁰⁴. These phases are marked by vertical black lines. Three K-Ar ages ranging from 3 to 1.89 Ma are considered unreliable, as discussed in Vezzoli & Acocella¹⁰⁴. The small black symbols are for K-Ar dates (hawaiite & mugearite) for Salas y Gomez Island¹⁰⁵. Error bars are 2σ.