In Region 1, where the current is most restricted in width, the 
greater percentages of observations occur at higher speeds as expected. 
For example, the constancy of the current throughout the year is 
shown by 40,000 surface drift observations in the region 25°-30°N, 
79°-80°W, which is almost entirely occupied by parts of the Florida 
Current and the Gulf Stream. About 90 percent of these observations 
are of a set directly north at a mean speed of about 3 knots and a 
maximum speed of 6.5 knots; in each of the other seven directions the 
number of observations averaged about 1 percent or less. 
In Regions 2 through 5 the stream widens, and frequencies in the 
lower speeds are higher. In the northeastern part of the current in 
38°-39°N, 64°-67°W, where the axis of the stream lies, 80 percent of 
1,650 observations for all months are of an east-northeast set ata 
mean speed of 1.2 knots and a maximum speed of 3.5 knots. 
In Figure 2]),A the location of the Gulf Stream as determined by a 
survey of surface currents in the easternmost part of the stream in 
the spring of 1960 is compared with the location as determined from 
historical drift data for the same region. The results of the survey 
have indicated that the current may extend to the bottom, that a major 
meandering pattern occurs east of 63°W, that the shapes of these 
Jarge meanders may be influenced by seamounts, and that the meandering 
path of the Gulf Stream changed very little over a period of 10 weeks. 
Anomalies in the different types of surface current data are 
surprising, particularly since comparisons between results of the two 
methods in other areas show reasonable agreement. It is obvious that 
varied interpretations of the Gulf Stream pattern in this region will 
be made until additional and more definitive evidence is available. 
60 
