1869.] C ARRUTHEBS SIGILL AEI A . 249 



mere analogical resemblances give no assistance in determining 

 systematic position, but rather tend to lead astray, as they have done 

 in the genus in question. There is one point in which the fossil 

 differs from the hving plants, which must at once, if it can be es- 

 tablished, separate them widely in systematic position. I refer to 

 the absence of medullary rays in the vascular cylinder of Stigmaria. 

 This cylinder is composed entirely of scalariform tissue pierced only 

 by meshes for the passage from the inner surface of the cyhnder of 

 the vascular bundles that supply the rootlets. 



It is important to remember that medullary rays are composed 

 entirely of muriform cellular tissue, and keep up the connexion be- 

 tween the internal and external cellular structures — the pith and 

 bark. Medullary rays are confined to and characteristic of exogenous 

 stems. They break up the wood into innumerable fine meshes. But, 

 besides these, there are in exogenous stems other and larger meshes 

 for the passage outwards of the vascular bundles which supply the 

 axial appendages — the leaves and branches. In some stems, as in 

 Phytocrene, Tupa, Cyccis, Euphorbia, Cactus, &c., from the size of 

 the vascular bundles these meshes are very large ; but in ordinary 

 trees those formed by the leaves early disappear, because, the leaves 

 being deciduous and produced only on the younger portions of the 

 tree, the vascular bundles supplying them are not continued through 

 the whole woody cylinder. It is of the utmost importance to notice 

 these two sets of horizontal radiating structures in the dicotyledonous 

 stem — the one, entirely cellular, being the medullary ray, the other, 

 vascular, being the vessels of the axial organs. 



Among recent cryptogams the only plants which have anything 

 like a continuous woody cylinder in their stems are the Ferns. In 

 the Lycopodiacece and Equisetacece, as represented among living 

 plants, the vascular bundles of the stem are symmetrically arranged 

 in a large mass of parenchyma, and there is consequently no true 

 separation between the cellular tissue of the medulla and that of the 

 cortical layer. It is however different in many tree ferns, which are 

 composed of a mass of parenchyma traversed by vascular bundles of 

 scalariform tissue, which form a closed circle separating the me- 

 dulla in the interior from the cortex of the exterior. The tissue 

 of this vascular cylinder is entirely destitute of medullary rays ; but 

 it is penetrated by large meshes, through which pass the vascular 

 bundles that supply the fronds and invariably rise from the inner 

 surface of the cylinder. 



The trunk of the cryptogam differs from that of the dicotyledon, 

 as regards the points in question, in having its vascular cylinder 

 penetrated by only one kind of horizontal tissue, namely, the vascular 

 bundles belonging to the fronds, while the exogenous stem has, be- 

 sides, another horizontal tissue of a very different structure and per- 

 forming a totally different function. 



The woody cylinder of Stigmaria possesses only the vascular 

 horizontal layer ; but from not distinguishing aright these two struc- 

 tures, this has been described as a medullary system. Where a 

 medullary ray has been described in addition to the vascular system. 



