The lack of systematic variation in the major and trace elements along the chain, discussed above and by Clague and Frey (this volume), implies that the conditions of magma genesis, the composition of the parental material, and the extent of melting at the Hawaiian hot spot have not changed significantly since Suiko was formed. There have clearly been fluctuations in at least one of these, but mostly they appear to be relatively short-lived. Kilauea and Mauna Loa lavas, for instance, are significantly different, but both volcanoes are presently active. The CaO and TiO (sub 2) anomalies for Nihoa to Daikakuji may represent a longer term fluctuation in source composition. The major problem, however, is how to account for the overall stability. If the Hawaiian hot spot is to be relatively fixed (and the magnetic evidence from Suiko [Kono, this volume] indicates that at least the change in latitude has not been great), the heat source for the igneous activity must be at least as deep as the asthenosphere and possibly deeper. It is very unlikely that it is in the lithosphere, since the lithosphere in the Pacific appears to be in motion away from the east Pacific rise. The evidence for the ultimate location of the magma's source material is considerably weaker. The lines of evidence we have are that (1) the depth of melting to produce the parental tholeiitic magma must be relatively constant and probably fairly shallow, or there would not be tholeiitic basalts (Green and Ringwood, 1967); (2) the source material must be fairly uniform through time and must be continually replenished, or the magma composition would change and igneous activity would eventually stop; and (3) on the basis of seismic activity progressing to volcanism, the magma apparently enters conduits at a depth of at least 60 km (Wright, 1971). Because this depth is in the lithosphere, the source region could be in the lower lithosphere or asthenosphere. We do not have criteria to unambiguously distinguish these possibilities.