59 



in which, from its imperfect permeability to the feeble rays of obscure 

 heat, water acts as a kind of trap for the heat it acquires from sunshine. 

 The same property allows the beds of still water reservoirs to retain 

 their temperature, while the bottoms of running streams are cooled by 

 the constant mingling of the upper and lower waters according to the 

 forced convective action already mentioned. 



If we reflect on the physical structure of the bed of the Dodder 

 while looking over the graphical representation of temperature during 

 the periods of frost, we cannot entertain a reasonable doubt as to the 

 sufficiency of the physical conditions which produced the ground ice 

 observed in the first week of January. On the 31st of December the 

 mean temperature was below the freezing point; and from the 1st of 

 January, when it was 28° • 9 it fell to 1 1° * 1 on the third. The fluctuations 

 of minimum temperature are yet more remarkable. On the second of 

 Janaary the minimum temperature was nearly 20° below freezing, 

 while on the third it descended to more than 29° below the same 

 point. On the fifth a thaw commenced, but it proceeded slowly until 

 midnight ; and it was not fully developed until Sunday morning, when 

 I witnessed the uprising through the water of fragments of ground ice 

 which had been detached from the bottom. It remains to account for 

 the action of the thaw in raising these pieces of ice. A reference to 

 the remarks on the margin of the Temperature Table opposite January 

 5 and 6, shows that the aggregate result of the snow melted and of the 

 rain which fell on these days was *940 inch at the Phoenix Park. It is 

 reasonable to suppose that the fall of snow and rain in the basin of the 

 Dodder, owing to its greater elevation, was somewhat larger than this 

 number would indicate. On the night of Saturday the minimum 

 temperature was 43° • 8 and the mean temperature of Sunday was 48° ; 

 and thus one of the first effects of the resulting thaw was a sudden and 

 very considerable accession of water to the Dodder from its feeding 

 streamlets. The coating of surface ice was thus burst from below up- 

 wards, and the slabs were rapidly swept along by the current, which, in 

 accordance with a law of hydraulics, was gaining in velocity while in- 

 creasing in volume. While the down scour of the river in its channel 

 was thus considerably strengthened, the density of the water increased 

 as its temperature rose with the progress of the thaw towards 40°. 

 The combined operation of these causes would necessarily facilitate the 

 detachment and floatation of those fragments of ground ice which were 

 observed emerging at the surface of the river, and bringing with them 

 manifest traces of their origin. 



During the second period of frost I observed ice attached to stones 

 over which water was flowing on the weirs, and also a few specimens 

 of crystalline ice at the edges of the river, and under the current, such 

 as I had noticed during the first period. I had no opportunity for 



alluded to in these words : — " The heat which it [water] has acquired during the day 

 shall have penetrated so deeply as to be incapable of being radiated backwards into space 

 during the night." (See " Philosophical Magazine" for February and March, 1867.) 



