17 



the 4!i(h and 50th parallels in the Indian division, and between the 54th and the 55th in the Pacific. The 

 centres entered the Atlantic division on or near the 58th parallel, thence about one-third of them moved 

 north-eastward, and the remainder passed between Danco Land and the South Orkneys. During the 

 Summer months the 53rd was the average parallel along which the centres travelled eastward in the 

 Indian division, and they followed a path between the 56th and 57th in that of the Pacific. After p .."in- 

 the 56th meridian of W. longitude, about half the central areas of depression travelled East-south-east 

 between Palmer Land and Danco Land and the South Orkneys, and the other half ]>a>>ed away to the 

 noi th-east ward. During Autumn and Winter the paths were confined to zones between 48 S. and 49 S. 

 in the South Indian division, and between 55 S. and ~>6 S. in the Pa'-irir. After entering the Atlantic 

 division, rather less than one-third of the centres moved east-south-eastward or eastward across Danco 

 Lund, 01 between Danco Land and the South Orkneys, the remainder passing east-north-eastward or 

 north-eastward into the Atlantic. A few crossed Tierra del Fuego or the southern extremity of 

 Patagonia, 



The paths of cyclonic centres are found to have been more scattered during the Autumn and Winter 

 months than in Spring and Summer in all parts of the Southern Ocean. 



Cyclonic storms, presumably of tropical origin, are represented on each of the charts on which the 

 central positions of Summer and Autumn depressions were plotted, but they are confined to the western 

 portion of the Pacific division of the Southern Ocean. These appear to have moved into the Tasnian Sea 

 from the north-westward, one in the Summer of 1901 and one in each of the Autumns of 1902 and 1903. 



The centre of the 1902 storm moved south-eastward to the north-west coast of New Zealand, thence 



south, passing through Cook Strait, and subsequently zigzagged to the eastward. It ran be identified on 



the daily charts from a position in about 34 S., 169i E., on the 12th December to a position in about 



S., 148i W., on the 10th of that month, representing a travel of 2187 nautical miles, at an average 



rate of translation of 312 nautical miles per day. 



The cyclones of the Autumns of 1901 and 1903 moved east-south-eastward after crossing the 30th 

 parallel, but cannot be traced to the eastward of the 180th meridian. 



The tropical disturbance of December, 1901, above mentioned, is not the only cyclonic depression that 

 can be traced on the daily charts for a number of days consecutively. Similar instances of identification, 

 day after day, of such systems after their initial location can be cited as follows : 



March 7-15, 1902. From 56 S., 140 W. to 55' S., SO' W. = 2243 miles; 280 miles per day. 

 May 2-10, 1902. From 53 S., 102 E. to 17' S., 175 E. = 2840 miles; 355 miles per day. 

 May 29-June 5, 1902. From 47 S., 17 E. to 60 S., 83 E. = 2433 miles; 348 miles per day. 

 September 2-9, 1902. From 57 S., 80 E. to 47 S., 131 E. = 2510 miles; 359 miles per day. 

 May 16-23, 1903. From 54 S., 131 W. to 58' S., 72 W. = 1995 miles; 285 miles per day. 

 September 21-29, 1903. From 47 S., 89 E. to 50 S., 130 E. = 1625 miles; 203 miles per day. 

 December 10-17, 1903. From 40 S., 130 E. to 55 S., 171 E. = 1890 miles; 270 miles per day. 

 February 8-15, 1904. From W S., 154' E. to 52 S., 166 W. = 1890 miles; 270 miles per day. 



If the centres of the respective cyclonic depressions have been correctly located, the average daily rate 

 at which they progressed was nearly 300 miles. 



While the number of cyclone systems that can be identified for four or more days in succession, during 

 their passage over the ocean, is large, the only moving anticyclonic systems that can be traced for more 

 than three consecutive days are : (1) Those that appear to have their origin to the west of Australia, the 

 centres of which travel eastward, join the Australian "high," and subsequently move across or to the south 

 of Australia, over the Tasman SIM and New Zealand or immediately north or south of those islands, and 

 then disappear over the Pacific. (2) Those that move eastwards from the South American Continent, 

 over the sea, and soon after disappear over the Atlantic. 



The former appear to originate as secondary high-pressure systems thrown off the South Indian 

 anticyclone; the latter appear to form over the land. 



As regards the paths of anticyclones, shown on the charts, the instances are rare in which areas of high 

 barometer, other than those that obviously form part of the permanent anticyclones of the great oceans, 



D 



