from the Stems of Vegetables. S33 



supply of water necessary for the development of these vo.u- 

 minous masses of ice. But by what force and through what 

 agency is it elevated and protruded ? 



Impressed with the idea that the phaenomenon is purely 

 physicaly having no connexion with the vitality of the stem, it 

 seemed reasonable that the remarkable exudation of icy cO' 

 lumns from certain kinds of earth, which long attracted my 

 attention, might be referred to a similar cause. Considera- 

 tions of this character induced me to study the latter phaeno- 

 menon more carefully. During the winters of 1 848-49 and 

 1849-50, abundant opportunities occurred of examining the 

 phaenomenon under the most diversified circumstances, the 

 soil in this neighbourhood being peculiarly adapted to its de- 

 velopment. 



The following facts seem to be established by my obser- 

 vations: — 



1. The phaenomenon occurs most strikingly when a warm 

 rainy period terminates in clear freezing weather, with the 

 wind from'the west or north-west. It is more or less distinctly 

 developed at all temperatures below 30° Fahrenheit. When, 

 however, the thermometer was as high as this at sunrise, it 

 was exhibited only in situations most favourable to radiation. 

 It frequently appears during several consecutive nights after 

 a rain ; but usually, when the temperature remains nearly 

 constant, with decreasing conspicuousness. This obviously 

 arises from the diminution of moisture : in situations which 

 are persistently wet, it is always developed in proportion to ^ 

 the depression of temperature. 



2. It takes place in soils that are rather firm, but not very 

 compact. For example, the phaenomenon is beautifully ex- 

 hibited along the sides of the water-worn ravines which fur- 

 row the declivities of the firm red clay hills of this primitive 

 region, as well as along the cuts or ditches by the road-side. 

 This clay seems to be formed by the decomposition, in situ^ 

 of hornblendic gneiss and mica-schist. This soil presents the 

 same phaenomenon when thrown up and lying on the surface, 

 provided it is not trodden down and rendered too compact. 

 For this reason it never appears on the well-beaten high roads, 

 although it is seen abundantly along their margins. The in- 

 fluence of compactness of soil is strikingly illustrated by the 

 fact, that the protrusion of the icy columns will frequently 

 occur around the margins and along the middle cleft of a track 

 of a cloven-footed animal, while none were found on ihe por^ 

 tions where the clay had undergone compression. The clods 

 found at the bottom of the ravines and along the margins of 

 the brooks generally afibrd beautiful manifestations of the 



