21.2°C at the surface), (2) high SST east of the front (23.2°C), and 
(3) temperature inversion (minimum temperature = 10.2°C at 85 meters). 
A poorly defined cold filament (SST = 12.6°C) is located inshore of the 
northern edge. Complexity of this particular cross section is due, in 
part, to recurving of the gradient between the warm water and adjacent 
cold water. 
A second warm area was observed in the northeastern quadrant of 
the survey area from January through completion of the study in May. 
Because this feature was of secondary importance to the study, detailed 
investigation was not possible. However, several differences are evi- 
dent from the limited data available; namely, (1) larger size, (2) life- 
time measured in months, (3) SST generally 5° or 6°C below Gulf Stream 
temperatures, and (4) separation from the northern edge by a cold fila- 
ment. The frontal zone enclosing the warm area appears to be well- 
developed with temperature change and gradient across the front of 2° 
to 5°C and about 1°C/km, respectively. Thickness (generally less than 
100 meters) gives the feature a lenticular appearance. Persistence of 
the system suggests that resupply occurs periodically, due perhaps to 
(1) downstream advection of smaller eddies or (2) convolutions in the 
northern edge as reported by Bratnick (1970). 
Cold-Water Intrusion 
A cold intrusion adjacent to the Gulf Stream was reported by Ford 
and Miller (1952), who tracked a narrow band of cold, low salinity 
water along the northern edge from Cape Hatteras approximately 150 
kilometers downstream. Thickness of the cold water was approximately 
40 meters. Observed T-S characteristics were such that the authors be- 
lieved that the cold water could have originated only on the shelf 
north of Cape Hatteras rather than from upwelling along the northern 
edge or advection from the shelf south of Hatteras. Ford, et al. (1952) 
noted that the cold water was within the Gulf Stream system. In dis- 
cussing the above papers, Stommel (1966) observed that: 
",..Were this filament of fresh water a permanent feature, 
the supply of fresh water required to maintain it would be 
of the order of magnitude 10"m3/sec., which could be sup- 
plied bv river discharge along the coast. The very fact 
that such a slender filament can preserve its integrity 
alone at least 1200 miles of the Gulf Stream is an indi- 
cation that small-scale turbulent processes tending to 
transfer properties across the Stream in the upper layer 
are inconsiderable." 
The cold intrusion observed during the flights suggests entrain- 
ment of Shelf Water bv the Gulf Stream northeast of Cape Hatteras. 
Additional details of the cold intrusion are provided by data collected 
aboard the GILLISS in conjunction with the flights of 12 and 16 May. 
Minimum SST (hull-mounted thermistor) and salinitv (bucket sample) ob- 
