Mar. 1S99.] BOTANICAL SOCIETY OF EDINBURGH 195 



of spring wood with large vasa ; the number as well as the 

 size of the vasa decreases as the passage is made into the 

 autumn wood. Here the trachea? are very numerous, but 

 of small size, and instead of occurring singly or in radial 

 or oblique pairs or threes, they occur in little oblique tails 

 or rosettes. While a spring trachea may measure "1927 

 mm. by '289 mm., those of the autumn wood average 

 •0214 mm. in diameter. Again, the fibre elements, forming 

 the general wood matrix, also vary in size in the spring 

 and autumn wood, the average diameter of one of these 

 being, over all, about "01 53 mm. 



In addition to what has been already given, the junction 

 of the annual rings is marked by the lateral broadening 

 out of the rays as already described. 



The interior of many of the large vasa presents a peculiar 

 appearance. Each appears to be filled with a network of 

 irregular meshes, the threads of the mesh being of a 

 different colour from that of the general mass. This 

 might lie due to some peculiarity of silicification, or to 

 fungal hyphiie, or to thyloses. A careful and repeated 

 examination of all the vasa leads me to think that the 

 latter is the correct interpretation, and that the delicate 

 threads are the preserved cell membranes of the thyloses 

 parenchyma. 



The resemblance in some respects to certain Cupuliferffi 

 is fairly obvious ; but in the number and breadth of rays, 

 as well as the presence of thyloses, it resembles some 

 Leguminosffi, such as Rohinia, but in these respects also it 

 has closer affinities with the woods of the order Meliacese, 

 which are distinguished, however, by their somewhat more 

 rapid growth. 



Explanation of Plates T., II., and III. 



Photomicrographs of opaque polished surfaces of the transverse 

 sections of the stems, taken by the methods described in 

 " Trans. Bot. Soc," 1897. 



