THE PLANT WORIvD 65 



picked up by the rolling ball gradually widened as the bottom of the slope 

 was reached. This natural ball rolling was so abundant on the side of one 

 cut that much of the snow was removed in this way from the south side 

 of the cut, and was accumulated as " snail-shells " at the bottom of the 

 slope. Subsequent freezing weather preserved these odd formations for 

 nearly or quite a week. The melting snow maintained the coils, melting 

 fastest where the layers were in contact. The same thing occurred on 

 some cuts along the B. & O. R. R. near Relay. 



The necessary condition for the formation of these snow rolls is a 

 steep-topped slope, and shrubs or other plant growth from which snow 

 can fall upon the steepest part of the slope, and the snow in a sufficiently 

 moist condition to pack easily. Under these conditions the snow drop- 

 ping from the branches of the plants will form the ' ' snail-shells " as it 

 rolls down the side, sometimes tipping over when part way down the 

 slope, or forming a slanting track on the snow. 



On the night of January 30 there was a decided fog covering the 

 higher portions of College Park, while the lower portions were compara- 

 tively clear. The next morning the objects which had been covered by 

 the fog were thick with hoar frost. Plants, trees, wires, fences, and even 

 the snow itself had the frosting of the night. The effect was as striking 

 perhaps in the case of the grasses as with any other one group of objects, 

 unless one excepts the twigs of Norway Spruce. The whole surface of 

 the exposed objects was covered by the frost crystals, sometimes to a 

 depth of a half inch. The frost was especially abundant on the north- 

 eastern side of the trees or other objects. When these were first seen in 

 the morning, the wind was blowing from the south, that is, against the 

 clean side of the frosted wires or shrubs, but at midnight there was very 

 little wind from any direction. It may be that the action of the southerly 

 breeze through the night or early morning checked the formation of frost 

 crystals on the more exposed side. 



In connection with the frost plants it may be noted that ice crystals 

 are sometimes formed of a similar character but with no connection with 

 any plant or other living thing. On January 3d curved crystals of ice 

 were seen in the cracks of a sunken barrel forming the coping to a spring, 

 which resembled in all respects those formed on frost- weeds. They were 

 formed at places where there was a very slow flow of a small amount of 

 water, which froze as fast as it reached the air, the temperature falling 

 to 6° during the night. The curved forms that most of the crystals 

 assumed was probably due to the water coming through the crevices a 

 little faster on one side than on the other even of the same spot of leak- 

 age, freezing as it came. That side of the crystal was made longer than 

 the other on which the supply was greatest, assuming a convex curva- 

 ture, while the slower-formed side was forced or formed into a concave. 



