In 1949 Eskola described structures that he called mantled gneiss domes from the Caledonides of Finland and pointed to the common occurrence of similar structures in other orogenic regions. As described by Eskola, these structures comprise a core of granitic migmatites or gneisses overlain by a layered metasedimentary and metavolcanic cover or "mantle." The core rocks in the central portions of some domes have an igneous microstructure, but near to their contact with the mantle they are foliated and this foliation, the contact, and the layering in the mantle are all parallel to one another and generally dip outward, away from the gneiss core, in such a way that they define a dome. Thus in outcrop the mantle rocks form a ring of generally outward dipping strata (where the domes have steep walls the simple pattern is sometimes complicated by locally overturned dips) surrounding an area of granitic gneiss. Eskola points out that this interpretation is supported by the way in which the contact faithfully follows the same stratigraphic horizon in the mantle rocks within a given area. He also points out that such a relationship can be proven for some domes where the basal mantle rock is a conglomerate containing pebbles of the granitic rocks of the core. Notwithstanding this, the core rocks are locally seen to intrude the mantle as dikes so that the situation is complex.

Similar structures are known from other orogenic regions. Excellent examples

are found in the Appalachians in the Maryland-Pennsylvania region and in New England where more than 25 domes occur in two linear belts that parallel the local Appalachian trend. The easternmost of the latter two belts stretches from northeastern New Hampshire to Long Island Sound (ca. 420 km) and includes approximately 20 domes. The core rocks of these particular domes are again predominantly granitic in composition and mainly of igneous, plutonic origin but there are also probable metasediments and metavolcanics present. Along most of the length of the belt the core rocks are in contact with the same stratigraphic layer of the mantle, the Ammonoosuc Volcanics.

Another example of gneiss domes is found in the Pine Creek Geosyncline of northern Australia. For example the Rum Jungle granite is a complicated body including schists, gneisses, and intrusive granites. It is essentially dome shaped and is mantled by a sequence of low-grade metasediments, the lowest member of which contains clastic material derived from the core. The core differs from those of the Finnish domes in that it is not foliated parallel to its contact with the mantle.

Various origins have been postulated for the various examples of mantled gneiss domes. Eskola (1949) envisaged the development of the Finnish domes and possibly the development of gneiss domes elsewhere as a process requiring two "orogenic revolutions." During the first orogeny, granite plutons were emplaced in metasediments and metavolcanics that were then lowered by erosion to expose the plutons. The plutons and country rock were then covered by a new sequence of sediments. Finally, during the second orogeny, the old plutons were reactivated by injection of new granitic magma causing them to expand upwards and thereby deform the overlying strata into a dome. The new magma and deformation converted the old granite into migmatites and gneisses and gave rise to the small igneous bodies seen to intrude the mantle in the Finnish gneiss domes. As explanation of the orientation of the foliation in the core rocks, parallel to the core/mantle contact, he claimed that the granitic materials, when the dome was swelling up, would naturally be transported along the paths of differential shear. This explanation of the foliation seems inadequate and we suggest that in general, where there is a foliation parallel to the contact in the rocks of the core and mantle, it is more likely that the foliation predates doming. In some cases the core/ mantle contact may be a thrust plane or thrusted unconformity and the foliation and thrust may be contemporaneous. In the Chester Dome of New England we have seen two generations of tight to isoclinal folds predating development of the dome and there, the folded foliation and possibly the contact are related to the earlier deformations.

Fletcher (1972) has modeled this mathematically and is able to simulate the Bronson Hill domes very successfully. The process has also been simulated experimentally by Ramberg (1967).

As in the case of the New England domes the Rum Jungle granite was initially interpreted as an intrusive body. However subsequent work showed that the core was older

than the mantle and unpublished work by one of the writers indicates that both the Rum Jungle and neighboring Waterhouse granites represent culminations, occurring where anticlines, of two separate generations and different trends, combine to produce a structural high

This mechanism has also been suggested by Ramsay (1967, p. 384) as a possible mechanism for the domes of Uganda and Ruanda. In addition he points out that the complex form need not be due to multiple deformation but could result from a single deformation in which "compressive strain was acting in all directions within the surface layers" (our italics). It should be noted that this mechanism is not necessarily incompatible with the gravity instability mechamism described above. Both processes could operate synchronously.