119 



about as numerous as the thin-walled diy-dock type or the rounded 

 water-worn type. The undulating top surface of the tabular bergs 

 often appears to be but a very slightly modified form of the surface 

 of the original glacier from which the ice was set free. From a short 

 distance it appears to be composed of rough grains about the size of 

 marbles and it is often muddied and soiled by the abundant bird life. 

 Many of the tabular bergs have walls that are kept vertical through- 

 out almost all of their life history in the Grand Banks area by calving 

 off of overhanging pieces as the waves eat into the berg about the 

 water line 



On May 31 a tabular berg approximately 115 feet high and 400 

 feet square was seen near the Tail, and on July 15 another large berg, 

 seen around noon 90 miles south-southeast of the Tail, was of a form 

 bordering upon this type. At about 2 p. m. attention was called to 

 the latter berg by a cry from the bridge. It was calving heavily and 

 pitching as it did so. Whenever its sheer end walls became over- 

 hanging ones they gave way, and then the suddenly lightened end of 

 the berg would lurch upwards. This caused the process to be re- 

 peated from the other end. The berg calved three or four times in 

 this manner duiing the space of about a minute. It was not over 2 

 miles away and could be clearly seen at the time by those on the 

 bridge and about the decks. Soon it became quiescent very close to 

 its former position of trim. 



6. LOCAL CONVECTIONAL CIRCULATION ABOUT ICEBERGS 



Barnes ^ slates that melting bergs draw in the surface waters 

 toward them and that they have warmer surface water immediately 

 about them than is the case farther away. It is quite possible that 

 bergs do draw in, chill, and sink certain amounts of surface water 

 under some conditions of melting. This might easily be concluded 

 in view of the rather small surface temperature effects described in 

 the section of this chapter on ice disintegration. 



Nevertheless it is difficult to see from theoretical consideration how 

 bergs can do much along the line of sinking chilled water in the 

 "melting area" south of the forty-eighth parallel. During the ice- 

 patrol season the surface water of the region is in general much warmer 

 and somewhat fresher than the layers 25 and 50 meters down. Such 

 conditions indicate a marked stability of the water column, and are 

 directly opposed to the production of vertical convection currents. 



In the early season many of the bergs keep in water that is below 

 32° F. at all levels about them until they are well south of the TaU. 

 Of course, in the case of such conditions they can not cause much 

 local circulation through chilling and sinking water that they draw in 



' Annual Report of the Smithsonian Institution for 1912, p .737. 



