THE LEAVES OF VASCULAR PLANTS. 255 



of pinnate nerves ; the latter are divided into leaves 

 with curved nerves, diverging or converging ; lastly, 

 pedalinerved leaves are found, the principal nerves of 

 which form angles and other curves, and notwithstand- 

 ing the resemblance they bear to each other, the former 

 belong to Dicotyledons, the latter to Monocotyledons. 



Let us now examine how we may deduce from these 

 primitive dispositions the theory of the leaves. 



Section IV. 



Of Lobed Leaves. 



It is particularly in all that relates to the divisions of 

 leaves, that the method of considering the limb as an 

 entire surface, which, in certain cases, wovild present 

 divisions, has been pushed to the greatest degree ; but 

 when these pretended causes came to be examined, it 

 was impossible to elucidate them. It is, in fact, setting 

 out with a wrong idea : leaves are not entire surfaces 

 which are cut; they are portions of limbs, which in 

 uniting, or remaining united in different degrees, consti- 

 tute sometimes projecting or retreating angles, some- 

 times entire surfaces. All the terms intended to 

 indicate the different degrees of division of leaves have 

 been believed, and universally admitted, under the sway 

 of the first hypothesis ; I am about to give the details of 

 the second, and, not to innovate too much, I will adopt 

 the old terms. If some do not appear either sufficiently 

 applicable or exact, it is in order to spare a superabun- 

 dant multiplication of words that I employ them, and I 

 must not be rendered responsible if this collection of 



