128 Mr. A. IJc'iii'rey un the Prugress of Physiolntjicul Buiimy : 



placed so as to converge more tow avd each other, since they take 

 away a portion of pith with them, and the centripetal arrange- 

 ment is finally perfected. In a directly opposite manner occurs 

 their reunion with the central mass by the loss of the concentric 

 arrangement of their woody bundles, and their reception into a 

 cavity which is produced at a corresponding point in the central 

 mass. 



From the natui-e of the composition, therefore, the lateral bo- 

 dies have a pith, like the central, also medullary rays and fibrous 

 tubes, but the author has not observed annual rings in either. 

 The existence of this pith in the lateral bodies has been denied*, 

 but on insufficient grounds; the round or oval central cellular 

 mass into which the medullary rays enter, for example in Pmil- 

 linia pinnata, cannot be called anything but pith. A. de Jussieu 

 also describes a pith in the lateral woody masses of several Sapin- 

 dacete, especially Serjania cuspidata, which pith was inclosed in a 

 medullary sheath containing spiral vessels, and was of a cylin- 

 drical or flattened form. Gaudichaud figures the latter form of 

 pith, which is centrally situated in the central woody mass, but 

 more or less excentrical, toward the external surface, in the 

 lateral bodies. 



It is important to observe that the central and lateral bodies 

 are all inclosed in a common bark which contains a common 

 layer of liber ; since this proves that the lateral woody masses are 

 not liber-bundles of the bark, as Martins t appears to have as- 

 sumed. But the author observes that, so far as his limited ma- 

 terials allowed him to see, the circle of liber above-mentioned 

 does not increase in diameter proportionately with the woody 

 masses. 



An attempt has also been made by Martius to explain the 

 presence of these anomalous lateral masses by considering them 

 as undeveloped branches running under the bark in the manner 

 that the roots do in some Lycopodia, as vv^as pointed out by 

 Ad. Brongniart J, who thereby explained some phaenomena ob- 

 served by him in the fossil genus Sigillaria. Lindley § has 

 observed a similar condition in a Barbacenia from Rio Janeiro. 

 Before these phaenomena can be applied to an elucidation of the 

 structure of the Sapindaceo', it is necessary to investigate these 

 lateral woody bodies in their earliest conditions. With this view 

 the author examined Paullinia pinnata, taking a yet herbaceous 

 twig about eighteen inches long on which three leaf-scars ex- 

 isted on the three angles, while two leaves were still in a vege- 



* Scbleiden, Grundz. 2nd ed, ii. 162. 



t l-iebei- die Veg. d. uniicht. u. aclit. Parasit. Miincb. gel. Auz. 1812, 

 N. 44-49. 390. 



X Avchiv du Mus. i. § Introd. to Bot. 3rd ed. 316, figs. 191-3. 



