The tectonic modes operating at any given time in the Earth are intima
tely related to the thermal evolution, since tectonics is driven by he
at removal from the Earth's interior. Conversely, the viability of a p
roposed tectonic mode depends on its ability to remove heat from the i
nterior as well as on its inferred consistency with geological evidenc
e. On this basis it seems that plate tectonics may have been dominant
only in the later part of Earth history, and that proposed earlier mod
es involving only a subcrustal thermal boundary layer may never have b
een dominant unless the effects of the basalt-eclogite transition or o
f latent heat removal were able to enhance their heat transport effici
ency. More generally, the tectonic mode driven by the cool thermal bou
ndary layer at the top of the mantle may have depended very sensitivel
y on the effects of composition and latent heat on density. Calculatio
ns indicate that plumes could Nave operated through most of Earth hist
ory at about the present level of activity, unless heat conduction fro
m the core into the mantle has been inhibited in later times, in which
case they would have been hotter and more active in earlier times. Pl
umes could not have substituted for plate tectonics because plumes and
plates are driven by different thermal boundary layers that operate l
argely independently.