364 BOTANICAL GAZETTE [November 



edge of the cotyledon on one side, and the other branch from the 

 same protoxylem group into the corresponding edge of the other 

 cotyledon. This may be stated in another way. In tracing down- 

 ward, the four strands of each of the petioles of the cotyledons may 

 be said to join two by two. Just before reaching the central cylinder 

 the inner strands of each fuse, and the outer strands of the one fuse 

 with the outer strands of the other, the four strands thus formed 

 giving rise to the four protoxylem groups. This is shown semi- 

 diagrammatically in figs. 4-6. Tracing these strands farther upward, 

 they are found to branch once more, so that in the upper part of each 

 cotyledon there may be as many as eight strands; but before reaching 

 the tips of the cotyledons they reunite into one concentric bundle 

 (fig. 75), which abuts immediately against the epidermis, thus com- 

 ing into very close contact with the gametophyte. At this place the 

 tissue of the gametophyte is so closely attached to that of the cotyledons 

 that it is difficult to separate them. 



The vascular strands of the leaf primordia.— For each leaf or leaf 

 primordium four strands leave the vascular cylinder or vascular plate, 

 at points not definitely located, but quite well distributed, and 

 generally in such a way that approximately one strand for each leaf 

 or primordium leaves on each side of the squarish central vascular 

 cylinder; also those strands belonging to the first leaves have their 

 origin either in the neighborhood of or in the protoxylem groups of 

 the plate. Two strands leave the cylinder approximately on the 

 same side as that on which the leaf for which they are destined is 

 located, and run more or less directly through the cortex into the ven- 

 tral part of the petiole without further branching; while the other 

 two strands leave the central cylinder approximately on the opposite 

 side and describe a curve around it (the one in one direction and the 

 other in the opposite direction) through the cortex, through the sheath- 

 ing leaf base, and finally into the dorsal or adaxial part of the petiole, 

 where they branch and rebranch (figs. 4-6). It should be empha- 

 sized that the point of origin is not at all definite, and that any particu- 

 lar girdle does not describe an arc of any definite extent, but that the 

 length of the arc depends upon the place of origin of the girdle and 

 the position of the leaf to which it belongs. 



It has been said that that edge of the leaf base toward which the 



