The geology of islands

Islands are attractive places to visit, not just for geologists. Nonetheless, for us, they provide three advantages that favour collecting and research in the Earth Sciences. One of the attractions of an island is its small size in comparison with continents. The corollary of this small size is its relatively long coastline. Assuming that our island is not the mound of sand with a palm tree so loved by cartoonists, a long coastline indicates abundant exposures of rock, commonly well-exposed and accessible.

Second, because of their relatively small size, islands offer a limited possible area of outcrop. The island may be volcanic in origin, so you may have one (or a few) volcanoes and its deposits to map, log and sample, producing a self-contained study. A particular sedimentary deposit may be (probably will be) limited to a single island. If you want to determine the palaeontology or palaeoenvironments of this deposit, the only place it can be studied is on one island. To give one example (among many), the Middle Miocene Grand Bay Formation, exposed on the east coast of Carriacou in the Grenadines, Lesser Antilles, includes the only crinoid-rich deposits in the Caribbean Islands. I had been studying the few Antillean fossil crinoids for ten years until I went to Carriacou and the sum total of specimens I had collected until then could have rested, comfortably, in the palm of one hand. From Carriacou, I collected bags of crinoid-rich bulk sediment samples (Donovan and Veltkamp, 2001; Donovan et al., 2003).

Figure 1
Fig. 1. Outline map of the eastern Caribbean region, showing part of the Greater Antilles (Puerto Rico (PR) and the Virgin Islands (VI)), the Lesser Antilles (LA, including Antigua) and north-eastern South America (Venezuela (VE), Trinidad (Tr) and Tobago) (modified after Donovan & Harper, 1999, fig. 1A).

Third, the geology of islands is commonly unique. The geology of the Isle of Wight is similar to that of the adjacent mainland of the south of England. However, the island status of the Isle of Wight is ephemeral – during times of low sea level, such as during glaciations, the Solent is just a river valley or lagoon, and it may be possible to walk and wade from Portsmouth to Ryde. But not so for an oceanic island, which has no major landmass to which it might be linked, whatever the state of sea level. Jamaica’s origins were as part of a volcanic arc in the mid-Cretaceous, close to what is now the position of Central America. Plate tectonic processes have ‘smeared’ this volcanic arc eastwards. Jamaica became subaerially exposed as it ground past the Yucatan Peninsula in the Paleogene, sinking below the waves for about 40myrs in the mid-Cenozoic to form a Bahamas-like carbonate bank and eventually being uplifted for the past 10myrs or so. This geological history has been deciphered from the Jamaican stratigraphic succession (Draper, 1998). There are aspects that are similar to other Antillean islands, but no other island has a rock record and a geological history quite like Jamaica.

Figure 2
Fig. 2. Geological map of Tobago (after Frost & Snoke, 1989, fig. 2; copied from Jackson & Donovan, 1994, fig. 11.2A).

To illustrate these ideas, I have written thumb-nail sketches about the geology and geological history of two of my favourite islands and research sites. I have worked on the geology of seven Antillean islands – Trinidad, Tobago, Barbados, Carriacou, Antigua, Hispaniola and Jamaica – and each has ample diversity to satisfy any geologist. From these, I choose Tobago and Antigua, islands with strongly contrasting geological histories and much to recommend them to the geological ‘tourist’.

Figure 3
Fig. 3. Aspects of the geology of Tobago (images after Donovan & Jackson, 2010, figs 8, 10, 4, 5 and 7, respectively). (A) Rockly Bay Formation: a roadside locality between the type section and the lateral unconformity (Fig. 4, centre), showing the weakly lithified sandstone beds rich in Megabalanus shells and valves. (B) Rockly Bay Formation: sandstone with balanids and bivalves. The rock is derived from weathered basalt. This site is close to the unconformity (Fig. 4, left). (C) NCSG: Parlatuvier Formation at Englishman’s Bay, showing polyphase deformation features. Lens cap (centre) for scale. (D) TVG: tonalite (light coloured) intruded by dark mafic dykes, near Plymouth. (E) Rockly Bay Formation: cobbles eroded from the type section preserving clusters of Megabalanus shells (Fig. 4, right). Each side of the tape measure is 50mm.


Tobago is one of the geologically most diverse of the Caribbean islands (Figs. 1-4). It is small, long and thin, oriented northeast-southwest, and is about 42km long by 12km at its widest. Snoke et al. (2001) interpreted the island as part of an allochthonous terrane (that is, originating in a place other than where it is now found) that forms the easternmost fragment of the Caribbean Mountain System of northern South America. The island can be divided into essentially three geologic provinces: one sedimentary, one igneous and one metamorphic (Jackson et al., 1988; Jackson and Donovan, 1994; Donovan and Jackson, 2010; and Fig. 2). Snoke et al. (2001) divided these provinces into five major units. These are:

  1. Metamorphosed –North Coast Schist Group (NCSG).
  2. Igneous – Tobago Volcanic Group (TVG).
  3. Ultramafic-felsic plutonic suite.
  4. Mafic (that is, containing dark coloured silicate materials) dyke swarm.
  5. Sedimentary – Tertiary and Quaternary sedimentary rocks.

The oldest of these units, the NCSG in the metamorphic province, is comprised of a succession of Early Cretaceous(?) or older, low-to-medium grade metamorphic rocks that crop out in the north-eastern third of the island. The Mesozoic metamorphic and igneous rocks that form the greater part of Tobago represent two stages of oceanic arc growth with an intervening phase of regional metamorphism. Before deformation and metamorphism of the NCSG, basaltic, andesitic and dacitic volcaniclastic rocks and minor lava flows were extruded onto a basement of oceanic crust. This first stage of arc development occurred in either the Late Jurassic or Early Cretaceous. The second stage of arc growth occurred during the mid-Cretaceous, and was preceded by penetrative deformation and lower greenschist facies metamorphism of the NCSG (Fig. 3C).

Figure 4
Fig. 4. Schematic reconstruction of the palaeoenvironment of the mid-Pliocene Balanus Bed, Rockly Bay Formation, Tobago (slightly modified after Donovan, 1989, fig. 7). The full vertical development of the bed is not shown. Relationships between localities not to scale. Vertical scale approximates both to water depth and bed thickness.

The younger arc (= TVG) was built directly on, or adjacent to, the older arc terrane (= NCSG). The NCSG is juxtaposed to rocks of the igneous province, which consists of a complex pluton (that is, a body of igneous rock formed beneath the surface from magma) of batholithic proportions that intruded the TVG, a thick sequence of volcaniclastic rocks and related lava flows of mafic to intermediate composition. The emplacement of the plutons led to the alteration of the TVG and the NCSG, producing a strongly deformed contact metamorphic aureole of amphibolite facies, that is, composed of a suite of minerals formed under conditions of medium to high temperature and pressure. These rocks crop out in the centre of the island. Mafic dykes intrude the pluton, TVG and NCSG (Fig. 3D). The preferred orientation of the dykes and the presence of small-scale extensional faulting suggest that the arc may have evolved in an extensional, rather than contractional, setting, which would have expedited the emplacement of the pluton to a higher level within the crust.

Figure 5
Fig. 5. Outline map of Antigua (modified after Donovan et al., 2015, fig. 1, redrawn and modified after Weiss, 1994, fig. 3), showing the principal geological subdivisions and the city of Saint John’s.

The sedimentary province is located in the southwest end of the island, and consists of Pliocene(?) and Quaternary terrigenous and carbonate rocks, markedly unconformable on the Mesozoic rocks of the igneous province. Possible down-faulting in the south was coupled with a mid-Pliocene(?) rise in sea level, which led to the deposition of the siliciclastic, shell-rich Rockly Bay Formation. During the Pleistocene, there was normal, followed by reverse, movement along the Southern Tobago Fault System, which accounted for the deposition, and subsequent uplift and tilting, of the Pleistocene coral limestone, which is a raised reef.

My own interest in the geology of Tobago centres on the palaeontology and palaeoecology of the Rockly Bay Formation (Donovan, 1989). A section can be walked from an open marine setting (= type section; Fig. 3E) to the lateral unconformity with an old sea cliff (Fig. 3B). The type section is exposed towards the southwest end of Rockly Bay, southwest of the capital, Scarborough (Fig. 4). About 11m of section is exposed in a low cliff, mainly poorly cemented, dull coloured mudrocks with some better indurated horizons cemented by calcite. Macrofossils are concentrated in the Balanus Clay Bed, which has a well-cemented base overlain with an orange-brown to grey claystone and a well-cemented limestone cap. The dominant macrofossil is the large balanid barnacle, Megabalanus tintinnabulum (Linné) sensu lato. Other fauna includes bivalves, gastropods, echinoids, bryozoans, fragments of crabs and rare shark teeth (Lewis and Donovan, 1991; Taylor and Foster, 1994; Donovan et al., 2001). Megabalanus tintinnabulum occurs as fist-size aggregations growing from originally small attachment points, such as pebbles (Fig. 3E).

Inland, at the Government Farm Road, is an unconformable lateral contact with the underlying Cretaceous basalts, which are well exposed on the north-western side of the highway and the south-western end of the exposure. The latter occur about five metres beneath an exposure of orange-brown, highly weathered basalt, in which quartz veins are still apparent. This, in turn, is about 150m southwest of a superficially similar lithology. On close examination, this is found to be sandstone dominated by individual balanids and barnacle debris, with less common oysters and rare scallops. Bedding is difficult to recognise, but this section appears to include six to seven metres of balanid-rich sandstone. Therefore, close to the lateral unconformable contact with the Mesozoic volcanic rocks, in situ weathered basalt is superficially identical with fossiliferous sandstones. The inferred palaeoenvironment of the Rockly Bay Formation and its Megabalanus-rich beds is shown in Fig. 4.

Figure 6
Fig. 6. Aspects of the geology of Antigua (images after Donovan et al., 2014, figs 5, 7, 10C, 12, 10A and 13, respectively). (A) BVS: long section at Nelson’s Harbour on the south coast; SKD for scale (see also Jackson and Donovan, 2013). (B) CPG: exposure on the north side of Corbison Point, yielding freshwater gastropods preserved in chert; the late Prof Trevor Jackson for scale. (C) Antigua Formation: part of a section at Hughes Point to show the rubbly nature of these shallow water limestones. (D) Antigua Formation: large, thin-walled sponge preserved at Half Moon Bay in deeper water limestones. (E) Antigua Formation: echinoid Echinolampas sp. at Hughes Point, length about 60mm. (F) Antigua Formation: Devil’s Bridge, a natural bridge in well-bedded, deeper water limestones; the right side is green with algae


Antigua is a small island in the northern part of the Lesser Antilles volcanic arc (Fig. 1). It is limited in extent, about 27km east-west and 20km north-south. In this region, the Lesser Antilles are formed of a double volcanic arc, which diverges northwards. To the west are the northern Volcanic Caribees, a line of islands formed by volcanoes during the Neogene, some of which are still active. To the east are the Limestone Caribbees, islands that were volcanic in origin, including Antigua, but which are older (mid-Cenozoic) and are now quiescent (that is, inactive or dormant).

Antigua is unusual in having a rock record that records the transition from island arc volcanism to quiescence and limestone deposition from a limited period (Fig. 5). This change happened during a relatively brief geological interval, the Late Oligocene (Chattian), and probably took just a few million years. The changes in lithologies from southwest to northeast demonstrate a rare complete example of such an environmental transition in the Paleogene of the Antilles.

The geology of Antigua is in strong contrast to that of Tobago. Tobago has a strongly sorted rock succession, with major intervals of time unrepresented by rocks, extending from the Late Jurassic or Early Cretaceous to Pleistocene. Antigua has a tilted, essentially conformable succession of rocks limited to Late Oligocene in age (Donovan et al., 2014). It has a mixed rock record of igneous (mainly volcanic) overlain by sedimentary rocks. The regional dip of Antigua is towards the northeast (Fig. 5), the oldest (volcanic) rocks outcropping in the west and south. The rock record of the island is divided into the Basal Volcanic Suite (BVS; Jackson and Donovan, 2013); the Central Plain Group (CPG); and the Antigua Formation.

The BVS outcrops in the mountainous southwest part of Antigua (Fig. 6A) and consists mainly of lava flows and pyroclastic rocks erupted from five centres. These are mainly andesitic with lesser amounts of basalt and dacite. Minor limestones in the BVS have yielded marine fossils of species known also from the Antigua Formation.

The CPG outcrops in a low-lying belt extending from the northwest to southeast of the island (Fig. 6B), and is comprised of mixed siliciclastic and limestone sedimentary rocks of both marine and non-marine origin. The siliciclastic rocks were derived by weathering, erosion and re-deposition of the BVS. Silicification is a common phenomenon in rocks of the CPG and Antigua Formation; the former includes petrified wood (Donovan, 2014) and cherts preserving freshwater snails.

The Antigua Formation is rich in limestones exposed in the north and east of the island (Fig. 6C-F), interbedded with some minor tuffaceous and sandy horizons. Lithologies and macrofossils are varied, indicative of different palaeoenvironments. Common taxa include larger benthic foraminifers, scleractinian corals, benthic molluscs, echinoids (Fig. 6E), crabs and bryozoans. Deeper water environments high in the succession – that is, towards the northeast – are indicated by the presence of crinoid columnals, sponges (Fig. 6D) and rarer brachiopods (Donovan et al., 2015).


I gratefully acknowledge the support for fieldwork in Antigua provided by National Geographic Society grant #GEFNE55-12 and I dedicate this article to the memory of the late Professor Trevor A Jackson, Caribbean geologist and contributor to Deposits.


Donovan, S.K. 1989. Palaeoecology and significance of barnacles in the Pliocene Balanus Bed in Tobago, West Indies. Geological Journal, 24: 239‑250.

Donovan, S.K. 2014. The petrified wood of the Oligocene of Antigua, West Indies. Deposits, 40: 12-14.

Donovan, S.K. and Harper, D.A.T. 1999. A new paleobathymetric interpretation of the Middle Miocene Grand Bay Formation of Carriacou (Grenadines, Lesser Antilles). Ichnos, 6: 283-288.

Donovan, S.K., Harper, D.A.T. and Portell, R.W. 2015. In deep water: a crinoid-brachiopod association in the Late Oligocene of Antigua, West Indies. Lethaia, 48: 291-298.

Donovan, S.K. and Jackson, T.A. 2010. Classic localities explained 6: Tobago. Geology Today, 26: 233-239.

Donovan, S.K., Jackson, T.A., Harper, D.A.T., Portell, R.W. and Renema, W. 2014. Classic localities explained 16: The Upper Oligocene of Antigua: the volcanic to limestone transition in a limestone Caribbee. Geology Today, 30: 151-158.

Donovan, S.K., Nagassar, V. and Sankar, K. 2001. A fossil shark from the Plio-Pleistocene of Tobago. Caribbean Journal of Science, 37: 119-122.

Donovan, S.K., Pickerill, R.K., Portell, R.W., Jackson, T.A. and Harper, D.A.T. 2003. The Miocene palaeobathymetry and palaeoenvironments of Carriacou, the Grenadines, Lesser Antilles. Lethaia, 36: 255-272.

Donovan, S.K. and Veltkamp, C.J. 2001. The Antillean Tertiary crinoid fauna. Journal of Paleontology, 75: 721-731.

Draper, G. 1998. Geological and tectonic evolution of Jamaica. Contributions to Geology, UWI, Mona, 3: 3-9.

Frost, C.D. & Snoke, A.W. 1989. Tobago, West Indies, a fragment of a Mesozoic oceanic island arc: petrochemical evidence. Journal of the Geological Society, London, 146: 953-964.

Jackson, T.A. and Donovan, S.K. 1994. Tobago. In Donovan, S.K. & Jackson, T.A. (eds), Caribbean Geology: An Introduction: 193‑207. University of the West Indies Publishers’ Association, Kingston.

Jackson, T.A. and Donovan, S.K. 2013. Going green: chloritized tuffs from the Oligocene of Antigua. Deposits, 36: 42-44.

Jackson, T.A., Duke, M.J.M., Smith, T.E. and Huang, C.H. 1988. The geochemistry of the metavolcanics in the Parlatuvier Formation, Tobago: evidence of an island arc origin. In Barker, L. (ed.), Transactions of the 11th Caribbean Geological Conference, Dover Beach, Barbados, July 20th – 26th, 1986: 21.1-21.8.

Lewis, D.N. and Donovan, S.K. 1991. The Pliocene Echinoidea of Tobago, West Indies. Tertiary Research, 12: 139‑146.

Snoke, A.W., Rowe, D.W., Yule, J.D. and Wadge, G. 2001. Petrologic and structural history of Tobago, West Indies: A fragment of accreted Mesozoic oceanic arc in the southern Caribbean. Geological Society of America Special Paper, 354: iv+54 pp.

Taylor, P.D. and Foster, T.S. 1994. Bryozoa from the Plio-Pleistocene of Tobago, West Indies. Tertiary Research, 15: 1-16.

Weiss, M.P. 1994. Oligocene limestones of Antigua, West Indies: Neptune succeeds Vulcan. Caribbean Journal of Science, 30: 1-29.

 Stephen K Donovan (The Netherlands)

Buy Fossils, Crystals, Tools
Subscribe to Deposits
Join Fossil Hunts
UK Fossil Locations