IQI2] 



STOLAND— SPIKE OF BOTRYCHIUM 



529 



pinnae. As shown in figs. 16-18, the strands 

 e may have well developed primary xylem in 



xylem 



formed. Here the strands are distinguisl 

 gated cells. In other cases no traces c 

 throughout strands of the abortive spike. 



xylem 



em 



Figs. 16-18. — -Transverse sections through the strands of the abortive spike, 

 showing development of xylem; fig. 16, 12 mm. below abortive spike; fig. 17, 4 mm.; 

 % 18, 3 mm.; X210. 



If figs. 3-15 are compared with Chrysler's figures showing the 

 origin of the strands to the fertile spike, it becomes evident that 

 the changes are essentially the same. A few exceptions may be 

 noted. The first strand (a and a') which breaks off on the adaxial 

 side does not unite with the inner face of the large strand as it does 





in the case of the fertile spike. The gaps left by the strands to the 

 abortive spike are much more evident. These are shown in the 

 photographs of models in figs. 19-21. These models show only 

 one-half of the vascular supply. In fig. 19, which shows the inner 



