Gondolin Cave

Site history and excavations

No records of the date of lime mining activities at Gondolin, maps or photographs of the cave prior to mining are known to exist. Long-term residents near Gondolin report that the system was mined prior to the 1950s. The cave was named after J.R.R. Tolkien's Gondolin by the early 1970s. Exposed fossils in the cave system were noted at least as early as the 1970s and brought to the attention of Elisabeth Vrba and David Panagos at the (then) Transvaal Museum (now the Ditsong National Museum of Natural History), who initiated excavations in early 1979. The first phase of excavation addressed the extremely fossiliferous in situ remnant deposits adhering to the northern wall of the cave system (the GD 2 deposits sensu). The three-week 1979 excavation removed approximately 2 m³ (71 cu ft) of calcified sediments from the northern cave wall. While two distinct sediment phases were recognized during the original excavation, later work with the fossil fauna and geology of the GD 2 deposits did not find any basis for multiple depositional phases within the sampled sequence. The removed sediment blocks were organised into 43 blocks/trays and processed using acetic acid at the Transvaal Museum. A significant gap in research at Gondolin occurred in the 1980s, during which time all of the excavated blocks were processed (but no further in situ or ex situ sampling is known to have occurred). The first description of the result fossil assemblage in 1993 only partially described 4,344 individual specimens that could be assigned to a specific taxonomic levels (out of the 90,663 total specimens recovered from acetic acid processing of the 43 trays/blocks).

In 1997 a survey of the sediments at Gondolin included the first sampling of the extensive ex situ dumpsite deposits at the locality via a test trench (Trench A). Materials removed from Trench A included fossiliferous breccia blocks from most, if not all, of the stratigraphic units present at the site (as well as loose fossil specimens sifted from decalcified sediments). From this sampled material (the GD A faunal assemblage), two isolated hominin teeth (representing two different individuals) were recovered. The first, GA 1, is a worn and fractured left molar that has not been confidently attributed to either genus or species (but resembles Homo in some features. In contrast, GA 2 is a complete left m2 that has been recently analysed and attributed to Paranthropus robustus (if from a large individual).

The most recent excavations at Gondolin were undertaken in 2003 and 2004 to explore the largely decalcified GD 1 deposits along the northwestern rim of the Gondolin locality. A four-week excavation season produced a sample of 4,863 fossil specimens from approximately 50 m³ (1,800 cu ft) of soil overburden and naturally decalcified sediments that were screened with 1mm mesh. Critically, integration of taphonomic and geologic data from the GD 1 excavations indicated that fundamentally different processes of fossil record formation were occurring at Gondolin during the formation of the GD 1 and GD 2 fossil assemblages. This result highlighted the complex, heterogenous geologic processes that can influence fossil deposition and assemblage composition in South African cave systems during the late Pliocene and early Pleistocene.