M'Keevor's Voyage to Hudson's Day. 15 



tation to take place : thus he tells us, that it is always formed 

 in greatest abundance in proportion to the magnitude and 

 number of the stones composing the bed of the river, com- 

 bined with the velocity of the current; as also that it abounds 

 most in rough and rapid places, and not at all where mud or 

 clay is deposited. Now it has occurred to me, that, perhaps, 

 the formation of ice in these situations may be owing to the 

 same causes that gave rise to the deposition of dew and hoar- 

 frost on grass, twigs, and other fibrous substances ; namely, by 

 their possessing a greater radiating power. The rough end 

 surfaces of the stones I conceive to operate in the same way as 

 the vegetable fibres do in a clear, unclouded atmosphere, by 

 allowing the " affluent" wave to come in closer proximity with 

 the surface, and thus facilitate the discharge of caloric from 

 the bed of the river. That none appears where mud and earth 

 are deposited, I should suppose to be owing to their present- 

 ing a smooth surface, in consequence of the water constantly 

 rippling over it ; thus the stratum of incumbent fluid is pre- 

 vented coming into as close contact as if it presented a ruggid 

 surface. Just in the same way as if we were to take a highly- 

 polished vessel of silver and fill it with hot water ; it will take, 

 suppose twenty minutes, to cool a certain number of degrees ; 

 but if its surface be scratched with sand-paper, it will cool the 

 same number of 'degrees in nearly half the time. That a great 

 part of the effect is owing, in this case, to the number of pro- 

 jecting points is proved by the circumstance of simply scratch- 

 ing it in an opposite direction, when the effect is considerably 

 lessened : the number of projecting points being thus dimi- 

 nished, it will now take a much longer time to cool down the 

 same number of degrees. It is on the same principle that a 

 thin covering of muslin, instead of preventing the escape of 

 heat, as d priori we should suppose, does actually favour its 

 discharge. The N. W. wind may act in two ways : first, by 

 its greater degree of cold ; secondly, perhaps, being less im- 

 pregnated with the particles of foreign bodies, in consequence 

 of passing over the frozen regions of the north, it may be thus 

 more favourably circumstanced for the escape and transmis- 

 sion of those calorific radiations. 



The rays act on the same principle as a clear unclouded 

 sky does in producing the deposition of the aqueous meteors 

 already alluded to. The only way, however, of ascertaining 

 this would be by trying what effect screens of different kinds 

 would produce, when interposed between the surface of the 

 water and the strong current of the N. W. wind. This expla- 

 nation appears to me to be supported by a fact long since ob- 

 served, that water will congeal, though the ambient air should 



