which has not cut down deeply in its upper portion and now lies in a somewhat open trench till it approaches the sea, where it has cut down deeply into the basement beds. The entrenchment of the Rakaia in the plain gets less and less as the sea is approached. Also the contours show that the axis of highest level of the plain as it is followed seaward does not lie along the course of the stream, but is oriented in a N.W.–S.E. direction from the gorge, so that it crosses the Main South Railway between Dromore and Chertsey, the former ten miles and the latter seven miles south-west of the point where the railway crosses the river. The two places just mentioned are practically on the 400 ft. contour line. The bulge which marks the contours as the are followed seaward continues right to the coast, which it reaches north of the mouth of the Ashburton River and near the Wakanui Creek. Allowing that the cliffs here are seventy feet high, the inclination of the surface of the plain from Chertsey to their summit, a distance of twelve miles, is twenty-six feet to the mile. The general position is that the Rakaia has run along the crest of its fan for some distance, is deflected therefrom, and the effect becomes more pronounced as the coast is approached. The deflection from the crest of the Rakaia fan—or as he called it, the Ashburton fan—has been noted by Hilgendorf (1907, p. 210), when he suggested that the rotation of the earth might account for anomalies in the directions of the Canterbury rivers in their course from the base of the mountains to the sea. He relies on the average steepness of the banks of the rivers to prove his case, the steeper bank indicating the direction of deflection. In the case of the Rakaia he notes that points on the southern bank are higher than those immediately opposite on the northern side—according to his criterion a proof that the river is being deflected towards the south; but this is contrary to the direction the river should take if the deflection were due to the earth's rotation. The average steeper northern bank of the Rangitata he considered proof that it was moving north—the correct direction if deflection were due to the earth's rotation. It is only fair to say that Hilgendorf was conscious of the discrepancy. While admitting that rotation of the earth is a causa vera, which must be considered when accounting for the deflection of rivers, it frequently happens that some other accidental influence has caused a stream to depart from the crest of its fan and occupy an apparently anomalous position. Numerous instances can be observed in the mountain regions where the Canterbury rivers rise. It seems to me that a dominating external influence in the case of the Rakaia has been the sinking of the land to the north, and that the area of depression extended on to the northern flank of the fan. There is definite evidence of a recent lowering of the land both north and south of Banks Peninsula, and this is likely to be a more potent influence than the earth's rotation. The question of the level of the land at the mouth of the river will be referred to later. The greater height of the middle area of the plains between the Rakaia and Ashburton prompted Haast (1879, p. 299) to suggest that this higher portion was formed by the coalescence of the fans of the three main rivers. There are no facts that I am aware of that support