1887.] On the Fossil Plants of the Coal-measures. d' 



pits, not confined, as in Conifers, to the sides of the vessels in contact 

 with the medullary rays, but covering their entire circu inference. 



On examining tangential and radial sections of the phloem, it is 

 found to contain numerous long narrow tubes, but which cannot be 

 absolutely identified with either sieve tubes or with sclerous fibres. 

 In the outer bark numerous flat plates of coarse sclerous parenchyma 

 pass horizontally outwards. This feature is equally characteristic of 

 Heterangium Grievii. Various forms of lateral outgrowths are also- 

 described. 



Further observations are recorded on the structure of Kaloxylon 

 Hookeri, also originally described by the author from specimens 

 obtained from collieries near Oldham, in ' Phil. Trans.,' vol. 166 

 (Part 1, 1876). These specimens, though revealing the elegant arrange- 

 ments, especially of its vascular structures, to which the plant owes 

 its name, did not fully reveal the true structure of its cortex. But 

 soon after the publication of the memoir referred to, the author 

 obtained from Halifax beautiful examples in which this defect was 

 remedied. These new specimens exhibited within a remarkable epi- 

 dermal layer, a thick corfcex, always abounding in narrow vertically 

 elongated tubes, which appear to have been either gum or resin 

 canals. In some examples these are crowded together in remarkable 

 numbers. But the Halifax specimens exhibited other peculiarities. In 

 some of them the peripheral end of each of the five or six radiating vas- 

 cular wedges, which are so characteristic of the plant, is furnished with 

 a true phloem element. In the larger proportion of the Halifax speci- 

 mens the radiating exogenous extensions of the central vascular axis 

 are wholly wanting, and, whilst in most of the specimens which possess 

 these exogenous growths no cellular parenchyma appears amongst the 

 vessels of that central axis, specimens are described which display 

 a gradual development of such tissue in that position. In some- 

 cases its amount almost equals the area occupied by the medullary 

 vascular bundles. Some of the latter examples give off rootlets from 

 the exterior of the central vasculo -cellular axis, which pass directly 

 outwards through the bark. A further series of specimens is described, 

 each of which contains vascular bundles, of which the transverse section 

 is quadrangular, closely resembling the tetrarch bundles of many 

 roots. Tracing these bundles downwards through a number of 

 examples which diminish in size until very minute ones are reached^ 

 we find evidence that these bundles developed centrifugally instead of 

 centripetally. Nevertheless the author inclines to the belief that these- 

 objects are true rootlets. 



After pointing out the probability that the Lyginodendron Oldhamium 

 previously described by him is, notwithstanding its exogenous mode of 

 growth, a fern, of which Rachiopteris aspera was the foliage, the author 

 states that Mr. Kidston has supplied him with structureless specimens 



