448 Proceedings of the Royal Irish Academy. 



spherical masses of silica, with a rough nodular surface. These masses 

 axe found in small cells lying in the groove between two sclerenehy- 

 matous fibres, and forming a chain along the outside of the sheath. 

 There are several such chains along a bundle. In a mature internode 

 the lumen of the cell in which these bodies are formed is little larger 

 than the mass of silica. Strasburger (Leitungsbahnen in der Pflanzen) 

 describes similar bodies in ChamcBdorea elatior, Cocos flexuosa, Phoenix 

 sihestris. Pfitzer ("Flora," 1877, nr. 16), Link (" Bot. Zeit," 1849), 

 and Kohl ("Anaton. Physiol. Umtersuch. derkalksalze und Kieselsaure 

 in der Pflanze," 1889), also refer to similar structures in various plants. 

 In the stem of V. teres, however, they are only to be found probably 

 in connection with the large leaf -traces immediately after they enter 

 the stem. On the other hand, similar chains of cells, containing these 

 silicious bodies seem to accompany all the bundles in the leaves, and 

 I have found them also in stem and leaves of several succulent species 

 species of Dendrobium. Martin Mobius ("Ueber den Anat. Ban der 

 Orchideenblatter und dessen Bedeutung fiir das System dieser Pamilie." 

 Fringsh. Jalirl. f. Wiss. Botanik. Bd. xvin. Heft. 4) found these 

 silicious bodies in every orchid leaf he investigated. The develop- 

 ment can be easily traced in V. teres. In the young leaves, which 

 are still partly enclosed by the petioles of others, the bundles are 

 represented by one or more series of spirally thickened cells, imbedded 

 in a strand of elongated cells. The bundle sheath in this stage is as 

 yet thin, walled, and its cells are indistinguishable from the elongated 

 cells representing the sieve-tubes, companion-cells, and wood paren- 

 chyma. On the outer surface of this strand is to be seen in longitu- 

 dinal section (PL xni., fig. 17) a chain of small cubical cells, with large 

 nuclei, wholly different in character to the elongated cells of the sheath. 

 The nuclei of these cells are large, and almost fill the cell. As the 

 bundles develop, and the cells of the sheath begin to acquire a 

 sclerenchymatous character, a clear body may be seen imbedded in 

 the protoplasm of these cells (fig. 18), beside the nucleus. This is the 

 origin of the silicious body. It gradually increases in size till it 

 finally almost completely fills the cell (PL xttt., figs. 13 and 14), and 

 displaces the nucleus, so that it renders the latter very hard to 

 see. 



The development of the vascular bundles of V. teres shows that 

 the sclerenchymatous sheath arises from the same meristematic 

 strand as the bundle itself, and consequently is to be classed along 

 with those sheaths to which Strasburger gives the name of ''Stelo- 

 lemma," i.e. a sheath developed from the central cylinder, in contra- 



