of marine origin, and of material which closely resembles the loess of the south-eastern portion of the South Island. This plain slopes back to the base of the low downs near Amberley, where it reaches a height of about 150 ft. However, on looking up the coast to the north from the mouth of the Waipara, decided remnants of a shore platform can be seen at an estimated height of 250 ft. above the sea, and fronting the plain there are several small remnants at lower levels marking stationary periods during the prolonged elevation. It is therefore certain that a long stretch of coast-line has experienced the effect of this movement. That the rise is of recent date is very clear from the species of shells found at Motunau, and also from the forms of the stream-valleys that have been established on the elevated marine shelf. With few exceptions, the streams which run across it have very short courses, and are little more than extended gullies or washouts. Through the somewhat loose Motunau marls, which form the solid base of the land, they have eroded deep channels, at times over 100 ft. in depth, extremely narrow, and with sides so precipitous that they are absolutely impassable for long distances. The whole plain is dissected by them, and they render communication a matter of difficulty where roads and tracks do not exist. The district furnishes a most remarkable example of the effect of a recent upward land-movement on the gradient and cross-section of the stream-channels. The phenomenon is intensified by the uniform seaward dip at moderate angles of the beds under the plain, and the parallelism of the strike to the coast-line. A similar phenomenon is to be observed near the mouth of the Waipara, but the features are not quite so perfect.

This plain of marine denudation once extended much further seaward, and the small island at Motunau is a remnant of it, its flat top showing a marked alignment of its surface with that of the coast-line opposite. How far this plain extended seawards it is impossible to say at present, but at the mouth of the Waipara the river-terraces appear high above the present level of the water, and are terminated suddenly when they reach the edge of the old marine cliff which marks the edge of the coastal plain. At a former period the river must have extended much further seaward, and flowed on the top of the plain, the terraces with their shorn ends giving positive proof of its higher level and seaward extension at that level. As river erosion was proceeding the sea was eating back the margin of the plain, thus giving the streams a steeper gradient and increasing their erosive power, and the truncated ends of the terraces mark the limit to which the plain was destroyed. When one takes into account their perfect condition he must conclude that either terraces are stable land-forms or that marine erosion on this stretch of coast has been vary rapid and comparatively recent

There is also, evidence of a more recent land-movement still. Along the base of the old marine cliff, referred to previously as bordering the coastal plain, there is a strip of flat land consisting of shingle-beds, sanddunes, and swamp, half a mile wide and but slightly raised above the sea. It has, without doubt, been formed of detrital matter brought down by the rivers in the vicinity, such as the Ashley and Waipara, as well as by the small streams which flow directly into the sea, their load of waste being distributed by waves, tides, and currents along the base of the old cliff. These accumulations are several miles in length, and their size suggests that there has been either a remarkable increase in the supply of detritus or that there has been a small recent upward movement of the coast. There is no apparent reason why the streams should have been suddenly furnished with an increased load of detritus, although

calcareous sands with concretionary bands, and gravel - beds more or less cemented, and are highly fossiliferous. The chief genera to be collected are Mactra, Chione, Ostrea, Pecten, Crepidula, and also Bryozoa. Some beds are almost wholly formed of the remains of Crepidula, but they are usually in a poor state of preservation. A characteristic feature of the beds at this point are the massive bands of coarse cemented gravels, dipping to the west at an angle of 60°. For a long distance one of these beds forms one bank of the river, and it can be traced some distance to the north-east on the north side of Mount Cass. Here it dips at a steeper angle, and on going further it is apparently lost under the covering of soil. In all possibility it will reappear in the creeks which flow from the northern side of the Limestone Range.

The same direction of dip and strike is approximately maintained on following down the river to the immediate neighbourhood of a pronounced meander of the stream. The strike here begins to swing round in a positive direction, so that, while just above the loop it is N. 5° E., at the loop itself it is N. 5° W., with a westerly dip, several hard bands of cemented gravel occurring at this point rendering an accurate determination easy. Just past the loop, on the eastern side of the river, and also on the flank of the high escarpment a little further down-stream, the beds consist of sand and concretionary bands full of shell-remains in excellent state of preservation. This is one of the best localities that I know of for the collection of Tertiary fossils, and when thoroughly exhausted will be found to yield a very rich harvest. A list of species collected by Dr. Marshall and myself is given later in this article, and it will be found to show a marked agreement with those collected at the typical Pareora locality, in South Canterbury.

The structure of the beds becomes at this point somewhat complicated, and its unravelment is an interesting problem. On following down the western side of the river below the loop the strike is observed to swing round in the same direction as higher up the river. Just at the mouth of the gorge proper the strike is N. 15° W., with a westerly dip, and just below the Teviotdale Bridge, half a mile further on, it becomes N. 55° W.

In passing through the gorge the strike has thus swung round through a right angle, and its effect is to be seen in the shape of the ridges of the downs towards Amberley, which are found to circle round with it, the outward slope of the downs being generally towards the dip of the beds. At a point about 600 yards above the bridge, at the mouth of a small gully, there has been a marked dislocation of the beds. They have been apparently folded down in an acute isocline, so that the two limbs are approximately parallel; but the beds immediately on either side of it do not appear to be affected, and they do not change their proper level or alter their dip or strike. The disturbance appears to be quite local, and it is not strongly in evidence on the opposite bank of the river. This is the only marked dislocation to be observed throughout the whole length of the gorge, and I have not been able to find any sign of the fault mentioned by Hector (loc. cit.).

If we now consider the arrangement of the beds on the eastern side of the river, the structure on the western side can be readily understood Below the loop mentioned previously, on the slope of the high escarpment which fronts the river to the west and north-west of the Teviotdale Station, the beds dip to the south-east, but an open anticline is clearly visible at the point which projects into the river half a mile below. The axis of this anticline is not horizontal, but pitches to

the south-west, and thus the gradual swing-round of the strike of the beds to the west of the river, is easily explained. The amount of pitch must be considerable, for the lowest beds actually exposed in, the gorge, at a height of about 100 ft. above sea-level, are the Grey Marls, whereas the limestones which underlie the marls are exposed on Mount Cass, about three miles to the north-east, at an elevation of 1,700 ft. It is possible, also, that the anticline not only pitches, but dies out as well. It is certain, However, that it is distinctly unsymmetrical, and the western limb dips down at a steeper angle than the eastern limb.

The direction of the strike observed in the river near the axis of the anticline changes on following the escarpment towards Mount Cass. It is at first parallel with the river, but, on being followed further, strikes north-east with a dip to the south-east, and the beds forming it lie in just the same relation to the limestones of Mount Cass and the Limestone Range as the Mount Brown beds in the Weka Pass do to the limestones occurring there. The similarity of the arrangement is most marked. If the creeks between the escarpment and Mount Cass be examined, the “Grey Marl” of the Survey is found in its proper position and with characteristic development; but only the upper sandy beds of these marls are visible in the gorge itself in the immediate vicinity of the loop of the river and in the reach below it. The axis of the anticline which runs out in the Waipara continues to the north-east, the limestone of the Mount Cass ridge forming the limb dipping to the south-east, while the north-east limb is represented by isolated blocks to the north of the Limestone Range. At the core of the anticline, immediately to the north of Mount Cass, lies a prominent hill, formed of the underlying greywackes of Mesozoic age. This arrangement is exactly what might have been expected from a consideration of the structure and relationship of the beds in the Weka Pass and the Mid-Waipara

As the anticline is traced to the north-east from the river it appears to change to one of increasing asymmetry, so that some of the hard concretionary bands on reaching to the inland side of Mount Cass become nearly vertical. It is possible that on being followed further north-east still it grades into a fault with a downthrow to the north-west, since the limestones near the southern edge of the Omihi Valley, in the neighbourhood of Limestone Creek, show a marked discordance in level between those forming the crest of the ridge of Mount Cass, although they dip in the same direction. More accurate examination of this part of the country is necessary before a satisfactory conclusion can be arrived at.

The apparent bend in the axis of the anticline near the Waipara River is perhaps due to this asymmetrical character, associated with the pitch of the axis; it may, however, be due to a disturbance caused by folding of the beds to the east of the mouth of the river in the neighbourhood of the Teviotdale Station.

On all the stretch of country forming the triangular area between the Waipara River and the Teviotdale Creek the strike of the beds is uniformly N. 55° W., with a southerly dip. They consist of sands, sandy marls, loose gravels, and hard bands of conglomerate, composed of large pebbles of greywacke and full of fossil-fragments. The beds are so hard, however, that they rarely yield good specimens. One of the hard bands forms the escarpment to the south-west of the Teviotdale Station; another forms a low indistinct parallel ridge to the north of this: but the most

prominent is a massive and solid bed on the north of the station, which runs out to the coast at a prominent rocky point to the east of the river-mouth. The immediate coast-line at this point consists of large blocks derived from this bed, and its continuation seaward is marked by a line of submerged reef. Fossil remains are very common in this bed, specially remains of Mollusca, and notably a flat sea-urchin (Arachnoides placenta), but the rock is so firmly cemented that it is almost impossible to obtain good specimens.

This hard bed is primarily responsible for the shape of the ridge known as Bill's Hill, which lies to the north-east of the Teviotdale Station. Its peculiar position presents a somewhat difficult problem till it is recognized that the hill is an anticline, and that it is flanked on the north by a small syncline now coinciding approximately with the upper valley of the Teviotdale Creek. The northern side of this creek is formed of beds dipping to the south-east and rising to the north-west till they form the prominent escarpment facing Mount Cass on its southern side. The Bill's Hill anticline owes its preservation from denuding agents to the protection of its upper surface by the layers of hard conglomerate which covers it almost continuously, although individual layers are somewhat discontinuous in their extension, one band being frequently replaced by a slightly lower and parallel one on frequent occasions. Nevertheless, the total effect is to cover the hill and protect it from active denudation.

The axes of both the anticline and its accompanying syncline run approximately N.E.–S.W., but they peter out between the Teviotdale Station and the river. It is possible, however, that they have exerted some disturbing effect on the main structural anticline, which runs from Mount Cass towards the river; and perhaps the curvature of its axis is due to the coalescence of the two lines of folding as they are followed to the south-west. A very complete examination of the locality is, however, necessary before the precise effect of each fold on its neighbour can be determined.

It will be observed that all the axes of folding enumerated above are approximately parallel, and they are also parallel to the folds which the Cretaceous and Tertiary series at Amuri Bluff and Kaikoura exhibit so markedly. These folding movements have therefore extended well into North Canterbury. Their presence in that locality, and also in the Trelissick basin, described by McKay and confirmed by examination by the present writer, suggest strongly that earth-movements connected with the folding of the great alpine chain had probably not ceased even late in the Tertiary era, although they were certainly more acute in the Kaikoura district than further south, and were, besides, of a different order of intensity and character from those primarily responsible for the formation of the great mountain-range.

The thickness of the beds exposed in the gorge certainly exceeds 1,800 ft., and all through them, as well as in the underlying Grey Marls and limestones, there is no sign of any discordance or dislocation other than folding, with the exception of the local disturbance referred to on page 228. The special importance of the absence of any evidence for a physical break will be understood when the fossil content of the beds has been considered.

In many parts of the area the solid strata are covered with a veneer, of greater or less thickness, of what are evidently river-gravels, in addition to the covering of recent marine shingle on the coastal plain. These may have been derived from rivers which flowed over the country at higher levels than now, of which there is abundant evidence in the downs

• In addition to the above, the following species are recorded by Haast as occurring at the Lower Waipara Gorge (“Geology of Canterbury and Westland,” p. 321):—

• Cytheria enysi Hutt.

• Venericardia intermedia Hutt.

• Modiola albicosta Lam.

• Modiola sp.

• Lima crassa Hutt.

A careful comparison of this list with the list of present species of Mollusca found fossil given by Suter (Trans. N.Z. Inst., vol. 42, 1910, p. 8) shows that more than 30 per cent. of the species given above are now living in New Zealand seas. Although the list of Waipara fossils will no doubt be greatly amplified by more careful search, the relative proportion of species to those existing now is not likely to be much altered. Judging from this percentage, the beds should be classified as Upper Miocene or Lower Pliocene.

A further comparison with the list of fossils found at the typical Pareora locality, in South Canterbury, shows that of sixty-four named species given in the Waipara list thirty-two are to be found in the lists of species collected at Pareora given in Haast's “Geology of Canterbury and Westland,” in Park's paper “On the Marine Tertiaries of Otago and Canterbury” (Trans. N.Z. Inst., vol. 37, 1905, p. 530), and among the fossils exhibited in the Canterbury Museum. This is at first sight a somewhat small proportion, but the forms common to both include a very large number of characteristic species, and it is possible that further collection may bring about further accordance. In any case, the number of characteristic genera common to both localities renders it a matter of certainty that the beds in the Lower Waipara are contemporaneous with those in the typical locality at Pareora.

A further comparison with the list of the fossils collected by Park on the Mount Donald escarpment (loc. cit., p. 540), and with the lists of Mount Brown fossils given by Haast (“Geology of Canterbury and Westland,” pp. 306–11), and also by Hutton in his paper on the “Railway-cuttings in the Weka Pass” (Trans. N.Z. Inst., vol. 20, 1888, pp. 261–62), shows a certain amount of agreement of the Lower Waipara fossils with those collected in the typical Mount Brown localities. There are, however, some differences, notably the absence of Brachiopods from the Lower Waipara, in marked contrast to their extraordinary numbers at Mount Brown. This may be due either to the fact that the proper horizons for these fossils have not been discovered in the Waipara, or that the conditions for their existence or for their entombment were not favourable in that locality when the beds were laid down. The accordance of the fossil content is, however, sufficiently close to assign both sets of beds to the same age; especially when the associated fossil species from other localities of the same age are taken into consideration. The stratigraphical relations also strongly support this conclusion.

Since by far the greater number of the fossils enumerated in the list can be collected on one horizon in the gorge—i.e., just above the Grey Marls—it is reasonable to consider that the lowest beds intersected by the river are of the same age as the Mount Brown beds, while the upper members are probably of the same age as the Motunau or Greta beds, and the conformity of the sequence in the gorge supports the opinion of Hutton that the Mount Brown beds are the base of the Pareora system, and the absence of any unconformity in the gorge also supports his contention