results has been indirectly noted by other New Zealand writers. Powell, for example, has pointed out (1931, p. 90) that the Waitotaran faunas of Waihi and Waipipi, easily correlated by key species, are from not dissimilar lithologies and differ but slightly in depth, yet have only one of their most abundant species in common; the same author (1934, pp. 261, 262) has convincingly correlated the highest Pliocene beds of Landguard Bluff and Cape Runaway entirely by consideration of rather uncommon key species, noting that the faunal communities and common species differed; the same author again (1937) has shown how faunal communities and ecologic controls can be rightly used and interpreted when applied to a single basin of known age uniformity.

Not that it is here intended to discount the importance of facies, faunal communities, or ecology. The principal underlying H. G. Schenck's definition of homeotopic and heterotopic faunas (1928, p. 164), namely, the control of the environment (as expressed by facies) on the generic and specific constitution of a faunule, is one of the first lessons a collector learns. Most species imply, by their very presence, a certain range of environmental conditions, all the more important because they represent positive evidence. Until the many faunal communities have been clearly designated, however, the absence of given species must be used with extreme caution. In these faunal community studies the importance to the palaeontologist of the dead shells as well as the live ones should not be forgotten, for shells are often transported from their own environment and buried in another. This would have a bearing on the “synecological studies” suggested by Thalmann (1936, p. 364), which are interesting in conception, but largely impossible of fulfilment.

As has often been remarked, the different environmental stations are very unevenly represented in the fossil record. Littoral species of both Foraminifera and Mollusca are rare. Infra-tidal, shallow water Mollusca are much better known, but not the corresponding Foraminifera. The off-shore faunas, from, say, 5–100 fathoms are perhaps the best known, also the yet deeper water Foraminifera. The deep water molluscs, being more dispersed and generally of fragile build, are difficult to collect, and tend to be poorly preserved through distortion by compaction of the fine sediments enclosing them. Indeed, a considerable difference in the preservation of these two phyla is often apparent, and it is not rare to find (e.g., at Hampden, in the Maheno and Burnside marls, etc.) that squeezed, broken, and unrecognisable Mollusca are accompanied by the unharmed, fragile, and hollow tests of Foraminifera, perfectly preserved in every detail. Due allowance must be made for this caprice and its bearing on apparently different accounts of the same geological story.

Studies of faunal communities may be useful in the case of such a clear-cut section as Castlecliff, but they are not usually so feasible elsewhere in the Dominion Tertiary (where the lateral gradation of fossiliferous sediments is seldom seen); no one to our knowledge has yet demonstrated in practice their superior usefulness in exact correlation. The acid test of any criticism of old methods is a demonstration of greater efficiency by the new.

Micro-Faunas.

Since Allan's paper appeared much light has been thrown on correlation problems by intensive study of the Foraminifera. The well-known advantage of this group is that excellent faunas can frequently be obtained from important localities poor in or lacking macro-fossils. The result is that where, before, we had relatively few abundant mollusc or branchiopod faunas, unequally distributed, we now have hundreds of rich rhizopod faunas covering almost the whole Tertiary, and can observe and separate faunal changes due to the time factor in a way previously impossible. Although the stratigraphic implications of the Foraminifera have received special attention for several years, the systematic descriptions are only now being published. In the first of these papers to appear, a revised classification of New Zealand stages has been put forward, based on both macro and micro-faunas (Finlay, 1939A, p. 531). This is the table adopted here, with the exception that the Upper Cretaceous is added, and the two divisions of the Taranakian have been reinstated for reasons given later.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]