This geophysical study of the Indonesian Archipelago deals first with the belts of strong negative gravity anomalies; it studies the Archipelago also in the light of Ewing's results in the West Indies. The hypothesis of crustal downbuckling seems to cover all the facts. The main tectonic arc has been brought about by movement of a great inner crustal block south-southeast relative to the crust on the outside of the arc and, for the second tectonic arc, by movement of a northeastern block in a direction slightly diverging to the east from the first direction, relative to the first inner block.
In most places the relative block movements not only bring about a component of compression at right angles to the arc but also a component of shear in the sense of the arc. The hypothesis is advanced that this shear movement, which is strongest on the sides of the arc—i.e., in Sumatra and in Mindanao—takes place in the inner arc, causing volcanic and seismic activity there and thus explaining the presence in all island-arc areas of a volcanic arc inside the tectonic one.
The writer discusses the crustal deformation in the tectonic arcs, which are supposed to have been of a plastic character, and the many shapes the surface features may assume, from a deep trough to a high mountain ridge.
In the second part, the hypothesis of convection currents in the deeper layers is discussed, first assuming currents over the whole or at least a great part of the thickness of the mantle—exerting a drag on the crust which may account for the great forces causing the relative block movements and the crustal compression mentioned—, secondly assuming currents over a depth of 500–750 km below the crust which may explain the deep and intermediate earthquake foci, the sinking of the deep basins, as, e.g., the Banda basin, the Celebes Sea, etc., and the origin of a third ridge inside the volcanic arc, and thirdly assuming a still shallower type reaching down only a few hundreds of km, which may have brought about the basins of a depth of about 2000–2500 m as, e.g., Strait Makassar, the Gulf of Bone, and the Gulf of Tomini.
It is usually assumed that for convection a homogeneous layer is required. Although from a depth of 900 km to 2900 km the mantle seems to fulfill this condition, it appears that from 200 km to 900 km depth the density increases too quickly to allow this assumption. We must here suppose a gradual change of phase or of chemical constitution. This point is discussed, and an attempt is made to show that a convection current can break through a layer of phase change of this kind and thus account for the origin of the basins. Also, the basins do not disappear when the currents stop.
For this hypothesis, however, one must assume that the mantle has a certain strength which must be overcome before flow or creep can set in. For this purpose a trigger effect must be supposed, brought about by a secondary phenomenon causing a horizontal temperature gradient. Griggs made this same supposition in 1939 to explain the pseudoperiodicity of the tectonic phenomena of the Earth's crust.