We practically disproved the existence of a reef reported by two whaling captains as having been seen by them on voyages from Cape Town to South Georgia in 35° 40´ S. and 5° 20´ W. (350 miles E. by N. of Tristan da Cunha). We steamed over the position and searched for two and a half days in the vicinity, half the time with a heavy southerly gale, in which a breaking reef would show 6 or 7 miles away. We sounded in 1,940 fathoms 3 miles south-east from the position given, 1,942 fathoms 15 miles east, 1,994 fathoms 15 miles south-east, and 1,989 fathoms 8 miles to the east, besides four soundings of 240 fathoms no bottom and one of 560 fathoms no bottom at varying distances from 15 miles south-west to 5 miles north-west. Although I do not think the reef exists, this instance gives some idea of the time and trouble a survey ship may expend in searching for danger, and then not finding it, through having been given a wrong or doubtful position; but vessels passing this position would be well advised to keep a good look-out for breakers. DEEP SEA SOUNDINGS Thirty-two soundings were taken in the southern ocean, practically all in previously sounded areas, and so of great value in adding to our bathymetrical knowledge of the ocean between the Atlantic Ocean and the Antarctic Continent. They were made with a Lucas machine, driven by a small Brotherhood engine, all kindly lent to Sir Ernest Shackleton by the Telegraph Construction and Maintenance Company, who also provided the _Endurance’s_ Lucas, with which we sounded the Weddell Sea. Our first line of soundings was run from a position 500 miles east of the Sandwich group to our farthest south point in 69° 18´ S. 17° 11´ E., where we unfortunately were barred from further progress by heavy impenetrable pack to the south, south-west and south-east. The soundings here were of great interest, having shoaled from 2,356 fathoms to 1,089 fathoms in a distance of 100 miles. This, with other indications, made it practically certain that land lay a short distance south, possibly not more than 60 to 70 miles. An irregular line of soundings for over 2,000 miles was then carried out from 17° E. to 46° W., mainly within and along the Antarctic Circle. The bottom, as usual, was mostly blue mud, droppings from icebergs, but north of “Ross’s Appearance of Land” we dredged up a large haul of angular rocky fragments, to the joy of the geologist. Very heavy weather unfortunately prevented us sounding the blank area between Elephant Island and South Georgia. Three soundings were taken between South Georgia and Tristan da Cunha, but heavy weather again prevented our doing more. Our last series were taken from 50 miles north of Gough Island to 35° 40´ S. and 5° W., the bottom over this area consisting mainly of white clay (globigerina ooze). [Illustration]