at the opening of the Devonian time. It seems to have come westward from a dispersion area in Africa and it evidently disseminated itself without interruption of continuity from the strands which now, as the Bokkeveld beds of Cape Colony, constitute the only evidence of marine life in the South African Paleozoic, to those of the Falkland Islands, two far distant regions which have much more of organic content in common than do the Falklands and the nearer regions of Parana, Argentina and Bolivia.
This fauna with its special and peculiar features is, however, spread through Bolivia, western Argentina, southern Brazil, including Parana and as far north as Matto Grosso, thence eastward by way of the Falklands to South Africa. From the boreal strands of the period it was separated by a barrier, often narrow and constituted only of deeper water, so that of the boreal Devonian we find no evidence much south of the equator in Brazil nor of the austral Devonian north of that line. This barrier I believe to have been overpassed at times during the early part of the Devonian by species which are of wider distribution south and north but these passages seem to have become rarer as time passed and as more complete geographic isolation was effected.
There are many evidences in this southern fauna that the land bridge was accompanied by insular strands which are evidenced by varying percentages in community of species and by bathymetrical variations. Apart from these possible island masses, there was clearly a Devonian land bridge extending from South Africa to the Falklands, westward into Argentina and northward into Bolivia, embracing also as continental or island lands parts of the states of Paranà, Matto Grosso and even of Pará.
By virtue of the evident derivation of the fauna of this time from the east along newly forming strands which were, throughout the period of the Devonian, kept asunder from the Atlantic-European lands at the north, and by its further development under conditions of isolation, the fauna presents fundamental contrasts to any development of the Devonian elsewhere in the world. It is in itself a unit and a unit also in relation to the sediments in which it is involved. There is no earlier Devonian in this southern region nor is there any later Devonian, for wherever the succession has been determined this austral fauna, bearing no evidence in itself of a later time stamp than early Devonian, is overlain by Carboniferous deposits without demonstrated unconformities between. Deposits and faunas which at the north we are accustomed to regard as of later Devonian age, are absent at the south, either because this austral land was broadly above the sea during these stages and its strands now lie buried or, as seems much more probable, this sedimentation represents the total Devonian sedimentation and this fauna the total Devonian fauna at the south.
