Anomalous Forms of Dicotyledonous Stems. 129 



tating condition at its apex. Each scar presented on its roundish 

 disc the almost perfect circle of vascular bundles of the fallen 

 leaf; above the scar was a dried bud, and below it a strong, 

 blunt ridge ran downward on the stem. On each side of the 

 cicatrix was a little semicircular scar indicating the articulation 

 of the fallen stipules, and from each of these lateral scars an 

 acute ridge originated which became united with a similar one 

 coming down from the leaf next above. On the examination of 

 the living leaves it was perceived that the vascular bundles of the 

 petiole formed the central woody mass, and those of the wing of 

 the petiole and of the stipules, the lateral bodies. These were 

 quite isolated just below the node ; but in another twig which 

 was examined, either two or the whole of them were always 

 united to the central mass, and this was particularly the case in a 

 twig which had a roundish instead of the usual triangular form. 

 So that this anomalous structure of the wood of Sapindacea has 

 its origin at the earliest stage, and is connected with the forma- 

 tion of leaves rather than of branches, and depends upon a pecu- 

 liar tendency of the vascular bundles to develope independently 

 of each other, round several centres, which tendency however 

 they occasionally lose and subsequently blend with the central 

 mass. 



The structure of the wood of the Malpighiaceous Lianes agrees 

 to a certain extent in appearance with that of the Sapindacea, 

 and here it is evident that the lateral bodies do not belong to the 

 liber. But according to A. de Jussieu the lateral bodies show no 

 disposition to arrangement of their fibrous tubes and vessels 

 around a pith, as occurs in the Sapindacea. He has also shown 

 that the wood lying immediately around the central pith is very 

 regularly formed, and has narrower and straighter medullary 

 rays than the layers subsequently produced ; while in the Sapin- 

 dacea the separation of the lateral from the central masses is evi- 

 dent in the very earliest stages of the formation of the wood. 



In the same memoir Jussieu has mentioned several climbing 

 dicotyledons of very different families, where the masses of wood 

 have a tendency to become separated from each other ; to these 

 may be added the climbing species of Begonia. Those species of 

 Begoniawiih. an upright stem have the wood symmetrically formed, 

 but in, for example, B. hirtella, the wood on the side of the stem 

 next the wall on which the plant grows is scarcely half so thick as 

 upon the other side which has been exposed to no pressure ; on 

 this side the wedges of wood are much expanded and quite un- 

 symmetrical, being separated from each by medullary rays which 

 equal them in breadth. 



Lastly may be mentioned some peculiarities in the wood of 

 certain climbing Bignoniacea, figures of which are given by Lind- 



Ann.^ Mag. N. Hist. Ser.2. Vol.i. 9 



