808 Transactions of the Society. 



chymatous cell-forms, in the vfood of Ailanthus,* the details of 

 which correspond in every particular with my own observations, 

 and the special elements under consideration also agree in form, 

 dimensions, and number of cells in combination, with the cell- 

 groups of the cambium. It is therefore naturally suggested that 

 wood-parenchyma results simply from the non-absorption of the 

 transverse dividing septa, and that while spiral or other thickening 

 has taken place, yet their outward form is the same as it was when 

 the group occupied its place in the cambium. 



In the development of the vessels also there are one or two 

 points of interest. In the mature vessel, the remains of what has 

 hitherto been regarded as the original transverse septa, are always 

 inclined more or less to the course of the vessel. That this should 

 be so is apparent, if we take into consideration the fact that the 

 cells from which they are developed are combined in the form now 

 described ; the septa, the remains of which we see upon the walls 

 of the vessels, being the terminal ones of the cell-group, and 

 therefore originally oblique, the intervening ones, which are at 

 nearly right angles to their length, being wholly absorbed. In 

 Fig. 15 is shown a portion of a section of Samhvcus nigra, in 

 which is seen a rudimentary vessel, the dotted lines indicating the 

 position of the septa, the outer portion of which will remain upon 

 the wall of the fully developed vessel. The transverse dividing 

 septa are still present. 



Another fact worthy of notice, is, that these oblique septa 

 become before absorption in all respects like the septa of sieve tubes, 

 being pierced with minute pores arranged in groups as shown in 

 Fig. 1(5, drawn from a young vessel in Ailanthus glandulosus, the 

 septa being very oblique and cut through. These pores gradually 

 enlarge and coalesce, till either all the central portion is removed, 

 or the bands which separate the sieve-like plates remain, forming 

 the scalariform septum. 



It would seem that absorption in cell-walls generally, or at 

 least frequently, commences by this thinning down and piercing of 

 the wall at minute points ; for in the Coniferae I find in the form- 

 ing wood immediately adjoining the thickened tissue, similar sieve- 

 like plates corresponding to the positions eventually to be occupied 

 by the weU-known coniferous bordered pits, whether it be on the 

 radial walls of the tracheides, or on the medullary rays. This is 

 well shown in Salisburia adiantifolia, and quite as certainly 

 in all others that I have examined, though not always so readily 

 to be made out. 



The results of my observations may now be summed up, and 

 from them it appears that the cambium -layer is not a portion of 

 the procambium remaining over after the differentiation of the 

 * See London Science Class-Book, Botany, Morphology, &c., p. 40. 



