and some diatoms; the diatoms blocked the meshes of the silk and made the tow-net slimy. In the evening we sighted another berg to leeward, and at night two other icebergs on either bow of the ship. The sun set only a little to west of south, and a light band of brilliant sky stretched along the southern horizon much the same as is seen in Scotland in June. During that night we passed several bergs in the fog, which came down and enveloped us again; we also met some nasty irregular ragged bits of hard clear ice, each about the size of a cottage, called "growlers" by Arctic seamen on account of the sound they made when rolling in the waves. These growlers are literally floating rocks which would rip the sides out of an ordinary iron steamer. We were truly in the Antarctic Regions, although more than 300 miles north of the Antarctic Circle. For this and other reasons I prefer to define the Antarctic Regions as being bounded by the average limits of floating ice. This line is almost entirely north of 60° S., except to the south of the Indian Ocean and to the south of New Zealand and Tasmania, where it dips to the southward. It trends farthest north in the South Atlantic Ocean, reaching about 50° S. to the south of Cape Colony, and 55° S. to the south-east of the Falklands. Within this limit we find the con-
