86 • THE ROYAL SOCIETY OF CANADA 



They are not, however, as Sanio apparently supposed, merely the 

 sloping edges of the hollow. 



Fig. 1, Plate I, a portion of a tracheid from the petiole of Zamia 

 integrifolia, illustrates the type of broad-bordered scalar i form from 

 which the bordered pits of more specialized woods are, from the 

 writer's standpoint, considered to have developed. It will be seen 

 that while in some parts of the tracheid the scalariforms are horizontal, 

 in other parts they are less regular, passing in a diagonal direction 

 across the tracheid. Filling in the space between the diagonal and 

 the horizontally placed scalariforms are the ends of others. The 

 long, slit-like pores reach from end to end of the pits, and the com- 

 paratively wide borders can be plainly seen. Between adjacent pits 

 dark lines are shown in the photograph. These appear dark blue in 

 the section, being portions of the primary wall, stained with Haiden- 

 hain's haematoxylin. The portion covered by the borders is lighter 

 in colour because the primary wall is thinner in this region. In other 

 words, each scalariform exactly covers a long, narrow primordial 

 pit of the same nature as those described by Sanio in Pinus. The 

 strips of primary wall between the pits, where it has retained its 

 normal thickness, are, according to Bailey's view, bars of Sanio. 



Fig. 2, Plate I, is a tracheid from the petiole of Cycas revoluta. 

 It shows in face view scalariforms which have been cut up to form 

 shorter bordered pits. Near the centre of the figure is one scalariform 

 which persists as a single pit reaching from side to side of the cell, 

 but constrictions can be seen in the border, indicating what might 

 be described as an abortive attempt at division into three. On the 

 rest of the tracheid complete divisions are observed, producing 

 bordered pits which are round, or more or less elongated. It is plainly 

 to be seen, that the dark lines extend vertically as well as horizontally 

 between these pits. In other words, each bordered pit has its own 

 separate primordial pit surrounded on all sides by primary wall of 

 normal thickness. This is true though the pits are so arranged that 

 there can be no doubt of the formation of a row of them on what was 

 ancestrally a single scalariform. Even the single scalariform which 

 is left has the thicker primary wall extending into the constrictions 

 on its sides. Sometimes in Cycads one sees a tracheid where the 

 scalariform primordial pit is retained and has a row of bordered pits 

 on it, and this was quite possibly a typical condition in the ancestral 

 forms where multiseriate pitting first originated. The fact that such 

 a habit has largely been lost in forms as loAy as our living Cycads 

 and that the loss is not dependent on alternation of pitting, is cause 

 for doubt as to Bailey's theory on the origin of the bar of Sanio. 



