In the study of the geographical distribution of terrestrial animals one of the great difficulties met with is the impossibility of defining exactly the limits of the regions into which we divide the surface of the earth. In a general way we recognise that there is an Australian region, an Ethiopian region, &c.; but, when we come to discuss the exact position of the frontier lines that separate these regions from their neighbours, we find all kinds of difficulties to overcome and inconsistencies to meet. For the sake of convenience it is useful to adopt certain arbitrary limits for these regions, notwithstanding these difficulties and inconsistencies, but we must recognise the fact that nature recognises no such limits, that every region overlaps its neighbours to a greater or less extent, and that there are many debateable grounds in the world where the fauna characteristic of one region is mixed with that characteristic of another. But this difficulty in defining the exact limits of the terrestrial faunistic regions is even more pronounced in the case of the regions and zones of the marine fauna. On the dry land we find mountain ranges, forests, deserts, and other barriers, that to a very considerable extent prevent the mixing of one fauna with another, but in the sea there are no barriers of anything like the same importance, but one fauna gradually merges into the neighbouring fauna according to the temperature, the pressure, the amount of light, the salinity of the water or the food supply. This then is one of the difficulties met with in the study of the geographical distribution of the marine fauna. But there is another that leads to almost greater complications. In considering terrestrial life it is customary to refer only to regions of geographical, or perhaps it would be more correct to call it—superficial distribution. It would be quite possible, however, to subdivide the geographical areas into zones of elevation above the sea-level, not very clearly marked off from one another, it is true, but nevertheless each showing a number of characteristic features. This idea is expressed, for example, when we speak of the Alpine fauna, the Himalayan fauna, or the fauna of the great Andes.