140 



pi'iidiciihii' lo il. 'I'licy orii^iiuilcd on the wood, hclow tlio 

 corh'X, and horcd llicii- \\;i>' llii'onuii ci'acks in the latter. 

 Tlic ice lihcrs were ix'rpendiculai- lo tlie stem. The at- 

 tac'limeiit of the riband to the stem ^vas very light and did 

 not correspond to any crack in the stem. 



Dnnal (1848) described on the square stems of some 

 hd)iates, fonr ril)bons of ice comin,<>' out from the sap wood, 

 through the torn cortex. 



Simihir ice blades on plant stems were observed, two 

 years later, in Georgia, by Le Conte wlio depicts the same 

 details as the previous observers. 



Sachs (18(50) attempted to induce ice formations of the 

 kind described, by exposing sections of beets to the frost, 

 under a cover to avoid an excessive evaporation. He ob- 

 tained plates of ice, of velvety appearance, made of ice 

 columns perpendicular to the cut surface of the tissue. 

 The author observed these phenomena on several types of 

 vegetables. He noticed, furthermore, that the process 

 could be more easily induced in succulent plants and when 

 freezing was not too rapid. 



Several attempts have been made to explain this exuda- 

 tion of ice in plant stems. Herschel notes that the water 

 comes from within the plant and that it must finally come 

 from the ground. The plant would work as a "chimney," 

 moisture being exuded from the earth by "every open 

 spiracle, ' ' 



According to de Mohl ("Vermischte Schriften," Tiib- 

 ingen, 1845), the exudation of fluid from the tissues is due 

 to the contraction of the latter under the action of cold. 



Le Conte holds that water will begin to freeze at the 

 external end of each capillary pore, where the contact 

 with tlie air maintains a low^er temperature. When the 

 water freezes into ice columns, a lateral pressure would 

 be exerted between tlie lattei', due to the ex])ansion which 

 accompanies freezing. The ice plate constituted by the 

 columns would then separate from the substratum in a 

 direction perpendicular to its surface. AVhen the ice 

 columns have moved away from the broader open end of 



