108 EXPERIMENTS AND OBSERVATIONS ON THE 



also, which was generated by these inverted cuttings, accumulated above 

 the bases of the annual shoots, as in the preceding instances. 



These facts appear to prove, that the vessels of plants are not equally 

 well calculated to carry their contents in opposite directions ; and, I 

 think, afford some grounds to suspect that the vessels of the bark, like 

 those which constitute the venous system of animals, (to which they are 

 in many respects analogous,) may be provided with valves, whose extreme 

 minuteness has concealed them from observation. 



The experiments, and still more the plates, of Hales, have induced 

 naturalists to draw conclusions in direct opposition to the preceding. 

 But the plates of that great naturalist are not always taken correctly 

 from nature * ; and plates, under such circumstances, however fair and 

 candid the intentions of an author may be, will too often be found some- 

 what better calculated to support his own hypothesis than to elucidate 

 the facts he intends to state. 



The preceding peculiarities in the growth of inverted cuttings, appear 

 to have escaped the observation of Duhamel ; and, as very few instances 

 of error, or want of accurate observation, will ever be found in the works 

 of that excellent naturalist, I must request permission to send you some 

 of the subjects of my experiments, as vouchers for my own accuracy. 



Of the inverted cuttings employed by Duhamel, a small portion only 

 appears to have remained above the ground ; and, under such circum- 

 stances, the different forms of those growing in their natural, or inverted, 

 position would be scarcely observable. It appears also, from his experi- 

 ments, that such inverted cuttings, in subsequent years, grow with as 

 much vigour as others that are not inverted ; whence we must conclude 

 that the organisation of the internal bark becomes again inverted, and 

 adapted to the position of the branch. The growth of some inverted 

 plants of the gooseberry-tree, which I obtained, many years ago, from 

 layers, gave me reason to draw a different conclusion ; for these always 

 continued weak and dwarfish. I do not, however, entertain the slightest 

 degree of doubt but that the assertion of Duhamel is perfectly correct. 



I intended to have added some observations on the reproduction of buds 

 and roots of trees ; but these would necessarily extend the present paper 

 to an immoderate length ; I shall therefore reserve them for a future 

 communication, and conclude with an account of an experiment which 

 more properly belongs to the paper I had the honour to address to you 

 last year, but which had not then succeeded. 



I have stated in that paper, that the leaf-stalk, the fruit-stalk, and the 



* The eleventh plate (Vegetable statics) is that to which, in this place, I particularly allude. 



