21 



froze over, but see* it rise up from the bottom, so as to 

 dart up in pieces edgeways, half 3 foot, sometimes a 

 foot above the surface. In this posture it continues a 

 little time, and then turning fiat upon the water, swims 

 along the stream, till it meets with other pieces, which 

 if the frost continues, all harden into one, till the river is 

 froze over. 



" In a part of the Thames, where there was very 

 little stream, I found the water, (says Dr. Hale) in a cold 

 morning, fro;:e one fifth of an inch thick, under which I 

 saw abed of ice at the bottom. Breaking away some of 

 the upper ice, I took up some of the lower ice, which 

 was about half an inch thick. It adhered close to the 

 bottom, where the stones and sand were incorporated 

 with it. When it freezes to a considerable thickness, it 

 will raise up with it, from the bottom, the fishermen's 

 osier wheels, although they are sunk down with stones or 

 bricks tied to them. 



" Standing waters indeed freeze first at top, because 

 they are coldest there : whereas in a stream the upper 

 and lower waters being continually blended together, are 

 equally cold; and the upper water mean time having 

 more motion, cannot freeze so soon. But here, where 

 the motion of the water was so small, its surface was 

 froze as well as the bottom, though not so thick : where- 

 as the main river, where its motion was greater, was not 

 froze over, though cakes of ice were continually rising 

 from the bottom." 



It has been commonly supposed, that fluids not only 

 dilate, but evaporate by cokl. And this has long passed 

 for an incontestable truth ; yet it is -altogether a mis- 

 take. From latter experiments it undeniably appears, 

 1. That cold does not increase, but lessen the-evsipots- 

 tion of water, if it be not exposed to the afri'ati*-;.. ed he 

 air: 2. That the evaporation of witter ( LT rinds 0*1 an 

 intestine motion, which it preserves- as ' jid, 



,aud that the air only contributes th.rt t'ly 



transporting the particles detached fromth" 

 thereby giving other particles rcora to disengage them*- 



