peaks, edge of landslips, &c., were used as points for triangulating; where this was impracticable then the method of converging angles was had recourse to. Up the Fiords of the Te Anau and Manapori Lakes, where, on account of the inaccessible nature of the mountains, and the shore line being shaded over with foliage, neither a triangulation could be carried on, nor bases measured. Differences of level between the Lake and one or more commanding peaks were used as a base for determining distances. This method, from the rapidity it gave to the execution of the work, was found to be of great value in the circumstances. It was generally no difficulty to find a suitable mountain peak, a mile or so in vertical height above the level of the lake; the angle of elevation to which, after the necessary corrections had been applied, gave an excellent means of determining distances up to seven or eight miles. The bearings were (from the same reasons as rendered a vertical triangulation necessary), magnetic. Care was always taken on returning to the stations of the true meridian to observe if there was any local deviation in the variation of the compass. In every other part of the survey, the work was done on the true meridian. The difference of bearing between the meridians of Mount York and Mount Nicholas was found to be 30'; the difference to be added to the meridian of Mount Nicholas. The difference of bearing between the meridians of Mount Nicholas and Lindis Peak 44', the difference to be added to the meridian of Lindis Peak. These differences are not to be taken as precise, seeing that the instrument had to be set several times to natural objects in taking on the bearings from meridian to meridian; but they may be taken as showing a general agreement throughout the survey as to bearing, for the apparent discrepancies are very nearly such as are accounted for by the convergence of the meridians to the Pole. The difference between the meridians of the Bluff and Mount York, obtained in a similar manner to the other differences, is 29', to be added to the meridian of the Bluff. In plotting the survey, the latitudes of the prime stations were found to close the one with the other, as also with the latitude of Mount Hamilton, as determined by the Reconnaissance Survey of Southland. A discrepancy of rather more than a minute of longitude, or nearly five seconds by Chronometer, exists between the longitudinal positions of Mount Hamilton, as determined from the two surveys; as the discrepancy is one of absolute distance, it does not affect the value of either survey. The desirability of having a check on the chronometrical determination of the longitudes of meridians was kept in view during the survey by carrying on, with as much care as possible in the circumstances, a triangulation based on short lines. After plotting the work to the scale of one half-inch to the mile, it is satisfactory to state, considering the rugged nature of the country, that the difference between the chain and chronometrical measurements of the distance between Lindis Peak and Mount York was not appreciable: the meridian of Mount Nicholas when brought to the same test, shows a difference of 2½ seconds by Chronometer.

To check the altitudes, several peaks were determined both from the data of Mount Pisa and from the data of the Bluff. The nearest agreement of the two determinations was that of Earnslaw, the difference being only two feet. The greatest disparity was in the two determinations of Mount Nicholas, the difference being 107 feet. The angular measurements of the survey were all made, (with the exception of the astronomical observations), by a four-inch Everest theodolite. Throughout the survey, an equal attention was given to the details of each district; so that unnecessary minuteness was not obtained in one part at the expense of vagueness in another.

Physical Geography.—The most marked and striking feature in the configuration of the country now under consideration, is the great and sudden, diversify its surface; the elevations take the form of mountain ridges, and the depressions that of gorges, valleys, and deep rocky basins, the latter filled by lakes. The mountains rise from 4000 to 9000 feet above sea level; and as the line of perpetual congelation is 8000 feet above sea level, (as determined last year from the reconnaissance survey of the Wanaka and Hawea Lake district), it follows that all elevations greater than 8000 feet are within the glacial producing zone. The highest parts of the Forbes and Humboldt Mountains are within this zone, and are covered with ice; they are parts of the great icefields that congregate around Mount Aspiring as a centre. The Earnslaw glacier, although only covering about a square mile in extent, is still, on account of its position, a very imposing object: it lies on the south side of Earnslaw, at an elevation of from 9000 feet down to the melting point: it is fifteen miles N. by E. of the head of the Wakatipu Lake; and, as seen from any part of the most northerly twenty miles of it, is by far the most attractive object in view. The lie of the country is nearly from N. to S.; and while the mountain ridges individually lie in that direction, they may, when taken in the mass, be more correctly described as