METEOROLOGY 151 



tions of any ships in the neighbourhood are also included. But considering the vastness of the area 

 affected, these few points must only allow a very approximate picture to be drawn. 



The anti-cyclone rarely appears as a single entity, but is usually composed of several distinct high 

 pressure centres slowly travelling across the Atlantic in an easterly direction towards South Africa. 

 When one of these has passed, conditions become more diffuse until the next high pressure centre 

 takes control. The cycle appears to average about four days, but it should be noted that the wind is 

 dependent on the pressure gradient between the anti-cyclone and the lower pressure regions to north 

 and east, and consequently does not vary directly with the position of the anti-cyclone. 



Table 2. Ships' observations (positions approx.) of the wind at sea during the 



' William Scoresby ' surveys* 



* Compiled from information collected by voluntary observing ships for the Meteorological Office. 



In February and March 1950, the pressure gradient was not so pronounced as in September- 

 October, and the centres of high pressure lay rather farther south in February and March, in about 

 30 S. Moreover the anti-cyclone in the September-October period was more intense, and frequently 

 associated with quite marked low-pressure areas over the coastal region. The trade winds were, there- 

 fore, stronger and more constant before and during survey II. 



Regular observations of wind were of course maintained aboard the 'William Scoresby' throughout 

 both surveys, and these have been rather useful in illustrating some of the short-term effects on the 

 water movement. 



During survey I, after the ship left Lobito, the wind was fairly light from the south-westerly sector, 



