ICEBEKGS IN NAVIGATION BARNES. 



733 



fall as the ship passed abeam of the island, about 300 yards. The 

 elTect is here more marked than in the case of the Fastnet, but it 

 must be remembered that the ship passed much closer in the former 

 case than m the latter. The proximity of the English coast caused a 

 rapid fall of temperature, as shown in figure 8, when the ship was 

 only 2 miles off the Somersetshire coast, steaming up the Bristol 

 Channel. It is probable that the rise of temperature in both cases is 

 due to the effect of fresh water from the land floating out over the 

 surface of the sea and dilutmg it to a considerable distance. 



Early in June, since giving this address, I was enabled, through 

 the kindness of the Allan Steamship Line, to obtain measurements of 

 the sea temperature on the northly route from Ireland across the 



Fig. 8.— Miceo-thermogram showing effect of land. 



Atlantic to Capo Kace and through the Gulf of St. Lawrence to 

 Montreal. These charts, which I intend to publish later, completely 

 confirmed the previous results. During fog in the ice track the ice- 

 berg effect was clearly obtained, and later a large iceberg was dis- 

 covered ahead. 



The icebergs all produced an elTect in the mstrument, even those 

 passed at distances ranging from 8«to 12 miles. The temperature of 

 the water through the ice track was between 40° and 41° F. Even 

 when passing within a quarter of a mile of a berg it was only 39°, yet 

 by the iceberg effect, i. e., the sharp rise of temperature to a maxi- 

 mum above the sea tem))orature, the influence of the berg could be 

 clearly seen. This shows very well that the actual temperature of 



