The rock structure is at once seen to be dominated by the development of white sediments that appear to be limestones, though it is found on a closer examination that many of these rocks are highly siliceous and that they are associated with marls, mudstones, and sandstones. There are also volcanic tuffs and lavas to a more restricted extent. The distinct has been greatly disturbed by earth-movements, and the rock outcrops show much folding, and probably faulting has taken place. There is also much and rapid change in the direction of the strike of the rocks, which necessarily greatly increases the difficulty of the stratigraphical problems.

The only parts of the district that were closely examined were the following: (1) Otamatea Arm from Komıti to one mile and a half above Batley; (2) Arapaoa Arm to one mile above Pahi; (3) Pahi Arm; (4) Oruawharo Arm from Port Albert to Oneriri. (See fig. 1.) The information gained from observations within this area cannot be regarded as in any way complete, though some general conclusions were arrived at. It appears that the lowest rock in the district examined is the limestone of the Gibraltar Rocks, in the Pahi Arm. This was also Park's conclusion. This limestone has often been correlated with the Whangarei limestone and with the Waiomio limestone. In the last locality, near the Kawakawa coal-mine, in the Bay of Islands, a bore showed that this type of limestone occurred very nearly at the base of this younger series of rocks and certainly below the hydraulic limestone.^* Above this Whangarei limestone there is a series of marls and mudstones which is perhaps 500 ft. thick, at any rate in the Pahi Arm. Interstratified with the marls there are some bands of limestone not very different from that of the Gibraltar Rocks, and, like it, containing some glauconite. In places the glauconite forms thick beds of greensand, especially in the Pahi neighbourhood. Locally this greensand appears to give place to some very soft marly mudstones, which are of special importance in the Otamatea Arm half a mile to the north-east of Batley, and on the Arapaoa Arm two miles to the north-west of Batley. In the Pahi Arm it appears that some Globigerina limestone, called generally “hydraulic limestone,” occurs beneath the greensand, for it is in this material that the bands of Whangarei limestone already mentioned are found. This Globigerina limestone is highly arenaceous and contains a good deal of glauconite, but in places it is siliceous. Occasionally this siliceous character is shown in the presence of flinty bands, though typical rounded flints are seldom found.

The main mass of the “hydraulic” or Globigerina limestone is found above the greensands. This relationship is particularly clear near Pahi, on both the Arapaoa and the Pahi Arms. The greater part of this limestone is composed of broken tests of Globigerina (Plate XXXII, fig 3), but in places it contains a great number of sponge spicules and marine diatoms and radiolaria (Plate XXXII, figs. 1, 2). This is markedly the case at the main bluff at Batley and at Kaiwaka. In the upper part of this formation extremely fine sediment makes its appearance and the organic contents dwindle. The fine-grained sediment consists almost entirely of very minute grains of quartz, well rounded, and seeming therefore to owe their transport to aeolian influences rather than to those of water. This material forms the top of the Batley Cliff, Paukıhi, and also the whole of the cliff on the opposite side of the Arapaoa Arm. On the foreshore of this cliff and farther to the

south this white material becomes somewhat harder, and then is succeeded rather abruptly, but along an uneroded stratification-plane, by a dark-grey marly bed. Forty yards farther along the foreshore these beds are succeeded along another parallel stratification-plane by a more arenaceous bed with some tufaceous material. These beds form the prominent Pakaurangi Point at the north-west end of the Funnel. In previous reports this point has been generally called Komiti Point, and the beds of which it is formed have been called the Komiti Point beds. This, however, is a misnomer, and it is likely to lead to much confusion, as Komiti is at the other end of the Funnel and is now the location of a considerable settlement of fruitgrowers.

Fig. 1.—Map of Central Kaipara.

These Pakaurangi Point beds, under the name of the Komiti Point beds, have been correlated with the Waitemata series of Auckland, which is referred to the Miocene period. The stratification of these beds is much disturbed on both sides of the Funnel. The upper members, especially as seen on the south side of the Funnel, have a highly variable strike, and they are composed of coarsely tufaceous material, amongst which are numerous fragments of hydraulic limestone. The presence of these fragments does not, in the opinion of the author, imply erosion of the limestone before the Pakaurangi beds were deposited. Their occurrence in this tufaceous material is rather due to volcanic action, which was extremely

violent whilst the land was still submerged and sedimentation was still in progress. Evidence of this is found at Mohınui, which is a volcanic neck. It is here clearly seen that water penetrated into the crevices of the volcanic rock and caused its sudden solidification in the vitreous state. It is also noticeable that the scorıaceous volcanic matter round this neck is quite unoxidized. This inclusion of fragments of sedimentary beds in the volcanic material of submarine eruptions is well seen at Cape Horn, on the western shore of the Manukau Harbour. Here large masses of the Waitemata beds are included with the andesitic volcanic matter deposited by contemporaneous volcanic eruption.

There appears to have been a centre of volcanic activity near the western entrance of the Funnel, for a large mass of andesitic lava occurs here. The rock is a typical hypersthene-andesite. On the shore opposite to Pakaurangi Point the rocks have been violently disturbed, hydraulic limestone, “Waitemata beds,” and volcanic lava being closely associated. The only visit paid to this locality took place at high tide, and the details could not be observed. It is probable that the formation would be uncovered at low tide to a sufficient extent to allow of an interpretation of the structure being found.

At this point there is a coarse conglomerate resting apparently with unconformity on the Pakaurangi beds. This conglomerate has an interest beyond the ordinary in that it contains some pebbles of a diorite which is not known to occur in place anywhere in this district. The nearest locality where plutonic rocks are definitely known to occur in place is Ahipara on the west side and Mangonui on the east side of the North Island respectively, but both of these places are some seventy-five miles distant.

In the series of rocks thus arranged no distinct stratigraphical break was observed, though this point is most difficult to decide because of the extent to which the rocks have been disturbed. Even at Pakaurangi Point, where the stratification is clear and the exposures continuous, there are some difficulties, and the strata are here clearly seen to be so disturbed as to change completely in strike and dip within very short distances.

Some previous observers have described several stratigraphical breaks in this rock series. S. H. Cox^* in 1879 represented the hydraulic limestone as Cretaceo-Tertiary in age, while the limestone of the Gibraltar Rocks is classed as Eocene, and is represented as folded in a synclinal manner while resting on a highly eroded surface of the hydraulic limestone. However Cox correlates the limestone of the Gibraltar Rocks with the Waiomio limestone at Kawakawa. In that place, as mentioned earlier, a bore has clearly shown that the Waiomio limestone is lower in the series than the hydraulic limestone. Cox also places the fossiliferous beds of Pakaurangi Point (called by him Komiti Point) in the Miocene period, but he does not indicate precisely the stratigraphical relation between the Pakaurangi beds and the limestone of the Gıbra tar Rocks which he calls Eocene.

Park^† in 1885 classed the beds of Pakaurangi Point in the Eocene, but all the other strata in the district are classed in the Cretaceo-Tertiary. In 1887 Park reported further on the same district. The Pakaurangi (Komiti) Point beds are in this second paper placed in the Miocene. Most of the other strata are placed in the Cretaceo-Tertiary. However, the white clays

on which the Pakaurangi beds rest and the concretionary beds at Batley in which Inoceramus was recorded were classed in the Jurassic. A strong unconformity is indicated in a section showing the stratification of the Pakaurangi Point. This break is represented as occurring between the “chalky marls” (called in the present paper “white mudstones”) and the “Waitemata beds”^* or tufaceous beds with fossils. The former of these beds are said to strike north-west and south-east and to dip 60° to the northeast, and the latter strike east and west and dip 20° south. Careful observations failed to support these statements. The white mudstones are traversed by numerous prominent joints which have nearly the bearings stated, and it appears that these have been mistaken for stratification-planes. The true stratification-planes are hard to distinguish, but when found it is seen that they strike 157° and dip 24° north-east. This result was obtained twice in different parts of the exposure at an interval of three years. The first of the observations was made in company with Dr. C. A. Cotton, of Wellington.

Some 150 yards farther south-east as seen on the foreshore close to Pakaurangi Point the mudstones acquire a bluish tint rather abruptly along a well-defined plane. The strike is here 120°, and 40 yards farther on, where the beds become more sandy and tufaceous (Waitemata beds of Park and Cox), the strike becomes 107°, and again the plane of contact shows no sign of erosion. Though this change of strike within the short distance mentioned may appear considerable and important, other variations as great are found in the Pakaurangi, or tufaceous, beds themselves. The strike here changes from 65° to 110° within a distance of 30 yards, and afterwards to 170°, with accompanying great changes in the dip, as shown in the map of the district. In order to see the stratigraphical facts mentioned the district must be visited at low tide.

At the extreme end of Pakaurangi Point there is considerable irregularity in the stratification. This is represented by Park as a fault. It might almost as well be represented as a disconformity. The beds here consist of tufaceous material and fine brecciated matter, and at the extreme point contain a large number of fossils of Miogypsina, which is also found below the disconformity. There are also some mollusca, such as Pecten aldingensis Tate, which are also common in the rest of the sandy beds between the sandy white mudstones and the point. The structure at the end of the point is therefore of very little importance.

McKay^† subsequently visited the district and reported a complete conformity from the Inoceramus beds to the top of the hydraulic limestone. He did not, however, visit the region of Pakaurangi Point.

The sequence of the younger rocks of New Zealand, so far as it is developed in the Central Kaipara district, appears to give no indication of such a break as is required to mark the dividing-line between two periods of sedimentation—that is, between two geological systems. Since the hydraulic limestone has always been correlated with the Amuri limestone of North Canterbury, it is natural in this place to review the additional observations that have been made in regard to the stratigraphical relation between this rock and the Weka Pass limestone which rests on it. It must again be stated that this is the dividing plane between Hutton's Cretaceous and

Miocene, and within late years between Park's Cretaceous and Miocene also. Morgan, however, suggests that the rocks are of Eocene and Miocene age respectively. Additional evidence consists very largely in the rediscovery by Morgan^* of a number of phosphatic nodules between the two rocks. Wherever the two rocks occur the one rests on the other without any change in dip or strike, and both are mainly foraminiferal. The phosphatic nodules are apparently regarded as rolled pebbles derived from some phosphatic stratum, which, however, has never been located. It would be much more reasonable to regard them as nodules formed on the sea-floor in situ. It is well known that such nodules are of relatively common occurrence in dredgings from water of moderate depth. From the Agulhas Bank several phosphatic nodules were dredged by the “Challenger”^† from depths of 98 and 150 fathoms. The form is capricious—generally rounded, but also angular. The concretions from the shallower depths were the larger (6 cm in diameter) and contained much more glauconite, and therefore possessed a green external appearance. The concretions are said to be more abundant along coasts where there are great and rapid changes of current, which cause frequent deaths of organisms. Merrill^‡ states that it seems probable that Cretaceous and Tertiary deposits have been formed under similar conditions in all parts of the world Stutzer also accepts the “Challenger” results as accounting for many phosphate deposits.^§ F. W. Clarke, in the Data of Geochemistry, p. 104, describes the methods of formation of such nodules.^∥

It thus appears that the phosphatic nodules are probably original marine deposits. Their occurrence implies rapid current changes, which are also implied by the replacement of the Globigerina ooze of the Amuri limestone by the arenaceous glauconitic limestone called the “Weka Pass stone” that rests on it. The phosphatic nodules occur just where they would be expected in a conformable rock succession deposited on a rising sea-floor.

The gritty limestone of the Gibraltar Rocks is similar in most respects to the hard bands in the hydraulic limestone in the Pahi Arm. As stated by Marshall in the paper previously referred to, the limestone of the Gibraltar Rocks consists mainly of Polyzoa, echinoderm fragments, and Foraminifera belonging to the following genera: Carpenteria, Globigerina, Rotalia, and Amphistegina. At Colbeck's Landing the rock is mainly mudstone, but on the west side there is a band of gritty limestone which contains many joints of stems of Pentacrinus. Some distance to the south-east of the landing, at the point B (fig 1), there is a thick mass of white limestone with a highly crystalline appearance. This appearance is due to the abundance of echinoderm fragments, and with these there are a few Foraminifera, mainly Globigerina. Some distance farther south, at C, there is some more limestone much silicified and apparently brecciated. Close to it there is a