Mat 1892.] THE BOTANICAL SOCIETY OF EDINBURGH. 433 



Note on the occurrence of Tannin in Dacrydium 

 CUPRESSINUM, Solan d., and Dacrydium Franklinii, Hook. fil. 

 By James Adam Terras, B.Sc. 



In both species the phloem-elements of the stem are, as 

 in most of the Taxinere, arranged in a comparatively regular 

 manner, and the general structure will be most easily under- 

 stood if we conceive of the whole phloem as composed of 

 concentric hollow cylinders fitting closely outside one 

 another. These are alternately thin and thick ; the former 

 are, as a rule, one cell thick, but in places two cells in 

 thickness, and are seldom complete. The latter, on the 

 other hand, contain from six to eight cells in each radial 

 row. Both series of cylinders are traversed at right angles 

 by numerous medullary rays which vary in height from two 

 to six cells and in length according to their age. 



The thinner cylinders are composed of tannin-sacs, and 

 these in Dacrydium FranJclinii lie above one another in small 

 spindle-shaped groups of from seven to nine cells. The 

 central cells of each group are roughly cylindrical, while 

 the terminal ones are somewhat wedge-shaped with the 

 edge directed radially ; the whole having the appearance 

 of being derived from a single cambial cell by transverse 

 partition. 



The tangential walls of these cells are equally thickened 

 all over, and contain minute crystals of oxalate of lime in 

 considerable quantities ; while the radial and transverse 

 walls show irregular cribriform markings with calcium- 

 oxalate crystals in the thickened portions. Each cell is 

 lined with a thin layer of protoplasm, imbedded in which is 

 a nucleus lying usually on one of the lateral walls, but the 

 remainder of the cavity is occupied by tannin. 



In the thicker hollow cylinders alternating with those 

 just described, the elements are arranged in distinct radial 

 rows composed of fibres, sieve-tubes, and cells. Each row 

 usually contains from one to four fibres, alternating with 

 which, and in most cases terminating the row, are the sieve- 

 tubes. At intervals where, if this alternating arrangement 

 were strictly carried out, one would expect to find a fibre, 



