Our justification for interpreting paleomaps and looking for continuous landmasses that may have held vast underground bodies of freshwater through long periods of geological time, is congruent with the results of our analysis in which the species inhabiting the African Pre-Cambrian shield are primitive forms. Such ideas have been formulated earlier (Ruffo, 1951; Leleup, 1955, Siewing, 1963), but Dahl (1977) remarks that a reduction of pleopods is unusual in species living in larger bodies of subterranean waters and thus he maintained that this feature must have been inherited from interstitial ancestors without fully developed pleopods. Therefore it is unlikely, according to Dahl, that these cave forms could have been ancestral to the smaller interstitial types. However, a detailed description of Leleup (1955) on the ecology of Trogloleleupia leleupi, a large cave-inhabiting ingolfiellidean from Congo, reveals that this species does not employ a free swimming behavior but rather moves across the bottom on its side. This suggests the true bottom dwelling nature of ingolfiellideans. Further speculations on whether small interstitial forms with reduced pleopods transformed back into larger cave forms with, secondarily derived, functionally active pleopods cannot be investigated any further without fossil forms.
A number of alternative evolutionary scenarios concerning ingolfiellidean origins now present themselves. First, a marine ancestor could have invaded the fresh-water underground environment in the Early Triassic, or even Late Paleozoic, with concomitant anatomical reductive adaptations taking place. In some cases, a “rebuilding” of reduced features (pleopods, third uropods) might have occurred especially when the infestation of the interstitial environment (with its confined spaces) was followed by radiation into cave-lake systems and underground rivers. Second, limno-stygobionts, as the large African inland species represent could have evolved into cave forms from surface water limnic ancestors. This phenomenon has been extensively studied in recent times by Culver et al. (1995) in the case of Gammarus minus. In the case of ingolfiellids, however, this route is unlikely because no epigean relatives are known to exist. Third, the peracarids may have been interstitial and/or groundwater forms in origin. This possibility has been little considered. In its favor, however, is the fact that the most primitive gammaroids (crangonyctids, bogidiellids) occupy such habitats. The spelaeogriphaceans also occur in caves, and Spears (pers. comm.) believes molecular evidence supports a common origin of spelaeogriphaceans and amphipods. In addition, the most primitive isopods, the phreatoicids, occupy allied habitats (Wilson, 1998).
In contrast, much has been written as to how preadapted marine benthic crustaceans could passively or actively invade the insular and continental ground water. Recent overviews (with numerous references to authors who have published on this subject) are presented by Holsinger, 2000; Coineau, 2000; Stock, 1993; Humphreys, 1999; Botosaneanu, 2001.
In summary, by plotting the grades of taxa found in our cladogram onto a succession of paleographic maps we can perceive shifting patterns that suggest a possible scenario of ingolfiellidean evolution. From an origin in the tropic cave and ground waters of Triassic Pangaea, the early progenitors of the genus Ingolfiella appear to have dispersed eastward along the sub-tropical and temperate shores of the Western Thetys Ocean in the Early Jurrasic. Dispersal in the westward direction along the northern and western coasts of Gondwana happened in the Late Jurassic. As the Atlantic continued to open through the Cretaceous, species of Ingolfiella found themselves isolated on the proto-Atlantic and Caribbean islands. One can envision that some members of the group rode the spreading seafloor down as the Atlantic deep was created. Afterward further dispersal into the deep Tethys and south Atlantic allowed the ingolfiellids to reach the far reaches of the Indo-Pacific and southern Ocean in the Cenozoic.
Finally, the cladogram we obtained (Fig. 8) suggests, as we pointed out above, that the African and Italian cave species may have given rise to smaller interstitial freshwater species and these have evolved further via brackish forms into marine interstitial and deep-sea species. A way to test this working hypothesis would be to sample for large cave Ingolfiellidea in the eastern parts of South America. The presence of large freshwater cave ingolfiellideans would lend credit to the idea that a freshwater continuum underground in Pangea existed before the break-up of the African and South American landmasses in the Early Cretaceous. In a far inland cave in southwestern Brazil (Gruta do Lago Azul), a large bogidiellid belonging to a newly erected genus, Megagidiella, was found recently (Koenemann & Holsinger, 1999). The size of this species is exceptional in relation to the widely distributed smaller sized bogidiellids. This situation is more or less comparable to what is seen in ingolfiellideans, i.e., large cave species far inland that are rare and found in places with ‘relict’ faunas, as opposed to the small species from the species-rich interstitial habitats closer to the sea. The strictly stygobiont freshwater isopod family Stenasellidae has a distribution that points to a Cretaceous origin. Consequently, the occurrence of this family on the northeastern South American Venezuela-Guiana Shield (Magniez, 1981) strengthens the proposition that ancient bodies of freshwater gave rise to ‘relict’ species that can be found on the fragments of Gondwana.
We predict that future research on early origins of stygobiont crustaceans that concentrates on sampling cave environments deep in the heart lands of old cratons in the tropical and temperate climate zones, will yield new taxa with ancient origins that will further elucidate patterns of evolution reaching back to the Triassic or even earlier.