The fossiliferous beds can be divided into two groups,—

The absence of identical or related species at most of the localities is perhaps due more to difference of station than to difference of age. However the matrix of the Flower-pot fossils resembles that of the Waikaripi ones and both places have the genera Nemocardium, Venericardia, Corbula, and Chlamys so that the environment was probably similar. The lack of specific agreement therefore is a fair indication of difference in age. Further, a number of the Mollusca from Whenuataru and the Flower-pot have strong Wanganuian affinities, e.g., Thyasira flexuosa, Venericardia beata, Cardita north-crofti, Corbula howesi, Merelina avita, Phenatoma decessor. It is therefore possible that a considerable time-interval separates the calcareous tuffs from the bryozoan limestone; but for the present it seems best to consider all as belonging to one period.

Seventy-nine species of Mollusca have been recognised from the marine tuffs and limestones; and of these only two (2.5%) are Recent. Both identifications of the Recent species are founded on poor material, so it is likely that they too can be separated as new when more specimens are collected.

Much has been written about the Lyellian method of reckoning position in the Tertiary sequence by means of the percentage of contained Recent Mollusca; but the work done on this subject by modern American palaeontologists has not received sufficient attention in New Zealand. Using only those localities with 50 species or over Dall (1903, p. 1617) got for the Eastern American Oligocene from 2 to 12% Recent, and for the Miocene 13 to 20%. (The upper part of Dall's Oligocene is now classed in the Miocene).

Martin and Glen's work on the Miocene of Maryland (Clark, Shattuck and Dall, 1904, p. exlix) gave only 10% of Recent Mollusca.

The San Pablo group of Middle California supplied 23.5% Recent forms to Bruce Clark (1915, p. 424) who placed the beds as Upper Miocene or possible Lower Pliocene.

In a recent paper on the comparative value of various organisms in zoning, Vaughan (1923, p. 519) says, “The percentage of Recent Mollusks in the Lower Miocene ranges between 3 and 7.5 per cent.”; he also gives a table showing eleven Miocene localities with percentages ranging from 3 (Lower Miocene) to 27 (Upper Miocene), and two Oligocene localities where there are no Recent species.

The very low percentage of Recent forms in the Chatham Island fauna (2%, perhaps 0%) shows that it belongs to early or middle Tertiary times, and the generic assemblage favours classifications as Upper Oligocene or Lower Miocene.

Of the eighty species only four occur in New Zealand. They are: Thyasira flexuosa, Tawera marshalli, Rissoina chathamensis, and Cochlis notocenica. None of these species is represented by well-preserved specimens, so that the identifications are not above suspicion. Most of the genera have been common in New Zealand since the Waiarekan, and some of the species have close relatives in our middle and later Tertiary strata.

The occurrence of the genus Perotrochus one of the Pleurotomariidae is interesting. A fine new species was found by Mr. Allan in the cliffs at Waikaripi, below the Wireless Station. Though always exceedingly rare in the Tertiary, the family has remarkably wide distribution, and there are four Recent species. P. tertiaria McCoy has been found at only one locality in Australia in Lower Tertiary Beds, and the specimens from the Wharekuri greensand, Waitaki Valley, were classed by Hutton (on rather poor grounds) under the same species. The exact age of these Wharekuri beds remains to be demonstrated, but it is probably Ototaran or Hutchinsonian; that is, somewhere about the Oligocene.

A large Perotrochus, perhaps the same species as that at Wharekuri, is fairly common in the Hutchinsonian greensands at Allday Bay, Kakanui; but the specimens are difficult to collect. The Chatham Island shell, though belonging to the same imperforate genus, Perotrochhus, is quite distinct specifically, so that no correlation can be made by means of it.

Another noteworthy shell is “Gryphaea” tarda Hutton. Specimens from the Balcombian and Janjukian of Southern Australia were identified by Tate (1886, p. 98) as the same species; and Ortmann (1902, p. 113) compared with G. tarda, a shell supposed to be from the Patagonian beds. Recently, Finlay has shown that the Australian shells are specifically distinct from the Chatham Island ones and has named them Notostrea lubra. Ihering (1907, pp. 6, 245) cast a certain amount of doubt on the Tertiary occurrence of Ortmann's specimen and identified it as the Upper Cretaceous G. burckhardti Boehm.

The writer agrees with Finlay (in ms.) that G. tarda is related to the genus Nostostrea Finlay, and thinks it probable that the Australian and South American species have arisen independently from Ostrea. Trueman (1922, p. 264) has already shown such to have happened with Jurassic “Gryphaea.”

b. Faunal Lists of Wharekauri-Waitangi Series.

(Recent species marked *)