Because they are more influenced by winds, the nonballasted bot- 

 tles reach the shore more quickly than the ballasted bottles. The rate of drift 

 is determined from the bottles with the shortest period of drift. Onshore 

 winds tend to accelerate the rate, while offshore winds, which are more com- 

 mon in the winter, slow the shoreward movement. The rate of drift bottle 

 return decreases during "poor weather when few people frequent the beaches 

 and markedly accelerates during the tourist season. The unusually large in- 

 flux of Sargassum weed on the north Gulf coast in the spring of 1962 covered 

 bottles and discouraged beach use. These factors, together with other more 

 subtle influences, make the calculation of current velocities from drift bottles 

 alone somewhat tentative. The instantaneous measurement of current velocity 

 with the Savonius rotor current meter should strengthen the drift bottle data. 



30- 



• aRtltos* points 



-330-doy directional orrows 

 with drift 8p««d in knots. 



TEX. 



28 



.... \vr / ^ 



GULF OF MEXICO 



30 



-28 



96° 



94° 



92° 



90° 



Surface drift on the continental shelf of the Gulf of Mexico in 

 February 1962 as deduced from drift bottle releases. 



25 



