so ROYAL SOCIETY OF CANADA 



feel certain satisfied every requirement. For the most j^art all traces 

 of organic matter have disappeared, and it is only here and there thai^ 

 regional areas still retain enough carbonaceous matter to make the 

 parts stand out with prominence. Elsewhere, the structure is repre- 

 sented entirely by a transparent, siliceous cast which makes it exceed- 

 ingly difficult to bring out the details. In addition to this, most of 

 the details of the cell wall have been obliterated by decay, while the 

 ■structure of the vessels has been largly replaced by crysta,ls of silica. 

 By first carefully studying the details of structure in Populus, Salix 

 .and Ehamnus, it was possible to determine what elements should be 

 looked for, and in this way it has been possible to obtain a diagnosis 

 which, while it is deficient in one or two respects, is nevertheless so 

 'Complete as to enable us to answer the questions at issue witli a fair 

 'degree of accuracy. This diagnosis is as follows: — 



Tî-ansiose. — Growth rings rather broad, the thin and inconspicuous summer 

 wood of 2 — 4 flattened cells. Medullary rays distant 4 — 8 cells, 1—4 

 cells wide. The wood cells rather thick walled, small, in radial rows. 

 Vessels numerous throughout the growth ring, becoming abruptly 

 smaller in the region of the summer wood; radially oval or oblong, 

 in radial series of 1 — 7, sometimes in tangential series and then form- 

 ing irregular groups; in one row, sometimes two rows, between each 

 pair of rays. 



J?a(ZtaZ.— Medullary ray high, the cells of two kinds; the low, central cells 

 about 3 — 4 times longer than high, numerous and about one-half the 

 height of the high, less numerous and chiefly terminal, short cells, 

 which are about as long as high or sometimes longer; the lateral 

 walls multiporose when opposite a vessel. Vessels with multiseriate, 

 hexagonal pits with transverse, slit-like pores. Scalariform vessels 

 not determinable. Wood parenchyma cells thick-walled, three times 

 longer than broad. 



Tangential.— Medullary rays of two kinds; the uniseriate rays upwards of 

 twelve cells high, the cells large, oblong, all of one kind; the multi- 

 seriate rays from 2—4 cells wide, lenticular and composed of two 

 kinds of cells, the smaller numerous and composing the principal 

 structure, the larger and less numerous terminal and single, often 

 extended into a single series or interposed between areas of the 

 smaller cells. Vessels as in the radial section. 



It is now in order to secure an answer to the first question, as 

 to whether this wood belongs to the Rhamnacese or the Salicacese, 

 and the answer will, to some extent, involve also, the second question 

 as to whether it is a poplar or a willow. These two families present 

 many features in common. In the Rhamnacese the wood cells lie in 

 radial rows, they are chiefly small, rather thin walled and the structure 

 as a whole, is somewhat open and soft. In the Salicaceae the wood 

 cells also lie in radial rows, they are rather large and the walls are 



