FUNCTION.] THEORY OF ENDOSMOSE. 329 



next ; and thus the whole mass of fluid is set in motion, from 

 the extremities of the branches down to the roots. Du Petit 

 Thouars is, therefore, of opinion that the expansion of leaves 

 is not the effect of the motion of the sap, but, on the contrary, 

 is the cause of it ; and that the sap begins to move at the 

 extremities of the branches before it stirs at the roots. That 

 this is really the fact, is well known to foresters and all per- 

 sons accustomed to the felling or examination of timber in the 

 spring ; and to gardeners who are occupied with forcing the 

 branches of plants in winter, while their trunks are exposed 

 to the weather. Some good observations upon this were 

 communicated to Loudon's Gardener's Magazine, by Mr. 

 Thomson, gardener at "Welbeck ; who, however, drew a wrong 

 inference from them. 



The following observation gives additional weight to the 

 opinion of Du Petit Thouars : 



M. Gaudichaud found, when in Brazil, that, upon cutting 

 through one of the creeping Cissi (C. hydrophora), the sec- 

 tions only slightly discharged fluid when the upper part was 

 merely divided from the under ; but that when a truncheon, 

 of whatever length, was separated from the stem, the sap then 

 ran out in great quantity from either end, according to which 

 was held downwards, and that it only dropped out slowly 

 when held in a horizontal position. Upon examining the 

 next day the cut end of the lower part of this stem, it was 

 found dry for 5 or 6 inches below the wound. M. Gaudichaud 

 ascribes the latter circumstance to the pressure of the atmo- 

 sphere upon the orifices of the tubes ; and the absence of any 

 considerable amount of bleeding in the upper half, to the 

 power of suction in the leaves, &c. ; while he attributes the 

 ready discharge of fluid from either end of the separated 

 truncheon to atmospheric pressure, which, he supposes, ope- 

 rates upon the vessels of the Cissus, as it would upon inert 

 tubes. (Ann. Sc., n. s., vi. 142.) 



Dutrochet has formed a theory of all the motions of fluids 

 in plants depending upon the agency of galvanism. He found 

 that small bladders of animal and vegetable membrane, being 

 filled with a fluid of greater density than water, securely 

 fastened, and then thrown into water, acquired weight ; he 



