One hypothesis is that the Columbia River Basalts formed from melting of SCLM (Smith, 1992). Prior to melting, the SCLM was enriched by pyroxenite veins from the asthenosphere. The melt ponded in the lower crust and acquired some of the crustal lead signature before erupting.
Many arguments exist against the SCLM as the main source for the Columbia River Flood Basalts. The main argument is that lithospheric mantle is refractory and does not easily produce large volumes of melts (Brandon and Goles, 1995). In order to melt the mantle lithosphere, there must either be an increase in temperature, a decrease in pressure or an addition of volatiles such as water. A few authors argue against the addition of metasomatic fluids to the SCLM. High ratios of Ba/Zr found in Clarkston basalts are not consistent with a metasomatic component in the source (Brandon and Goles, 1995). The Re-Os data is also incompatible with the metasomatic fluid hypothesis (Chesley and Ruiz, 1998).