40 DOUGLAS HOUGHTON CAMPBELL ON THE 



Ou account of the foliar gaps, as well as the attachment of the Ijundles from the 

 leaves, roots and stolons, the separate parts of the stem-bundles are usually very irreg- 

 ular and run, for short distances only, in a vertical direction. 



Tiie first permanent tissue to appear in the bundle is composed of small traeheids 

 near the foci of the bundle which is elliptical in section. Those at one focus, and these 

 appear rather earlier than at the other, are spiral or reticulate, although they are often 

 scarcely to be distinguished from the later scalariform ones. Those at the other focus 

 are not to be distinguished in appearance from the other traeheids, except that they are 

 smaller than most of them. In the completed bundle they are indistinguishable. 



The large scalariform traeheids, making up the bnlk of the xylem, are larger than are 

 found elsewhere in the plant. The cells from which they are formed are broader than 

 those of the i-est of the procambium, but are not veiy long. Their ends are at first simply 

 oblirpie, but as growth continues, they become slightly curved, and each cell somewhat 

 spindle-shaped, and strongly prismatic from the pressure of the neighboring cells. In 

 the earlier stages, before any traces of the markings on the walls are distrngiiishable, 

 the cells contain abundant protoplasm, in which are embedded numerous coarse granules' 

 and a modei'ately large oval nucleus. Pi-evious to the formation of the scalarifonn mark- 

 ings, the contents become somewhat clearer, owing to a partial disappearance of the 

 granules. The markings first appear as faint transverse ridges on the walls that are in 

 contact with another tracheid, corresponding ridges appearing each side of the wall. 

 These soon become thick enough to show that they end at the angles where the traeheids 

 meet, and enclose long, narrow, transverse pits, or spaces where the wall is not thickened. 

 This process begins with the cells lying next the primaiy traeheids, but after the first 

 begin to form, the remainder of the xylem develops with great rapidity, and almost si- 

 multaneously. The ridges separating the pits soon begin to show an inequality in the 

 thickening. At points where they touch the division wall they increase but little in 

 breadth after they are first formed, but as they increase vertically in thickness, they also 

 increase in breadth, until at apointabout half-way up they reach a breadth of about twice 

 that at the base. From this point they narrow again, so that a transverse section of one 

 of these bars is nearly rhomboidal. In this way a pit is formed broader at the bottom 

 than at the top, or funnel-form, the broad end of the funnel being in contact with the 

 wall separating the tracheid from its neighbor. 



The development of the sieve-tube begins shortly after that of the large scalariform 

 traeheids, and in some respects resembles it. The cells of the procambium, from which 

 the sieve-tubes originate, are uniform in diameter and have squarer ends than the young 

 traeheids, so that the separate elements of the sieve-tubes are more nearly cylindrical 

 than the traeheids. They occupy an irregular area at each end of the l)undle, and can 

 be best seen in a section passing through the major axis of the elliptical bundle. Their 

 contents are colorless and more finely granular than those from which the traeheids 

 arise, and the nucleus is not so readily seen. The sieve-discs are more abundant in the 

 lateral walls, and are developed almost simultaneously in all. 



Faint, transverse ridges appear upon the lateral walls, much as in the young scalari- 

 form traeheids, but they are broader and not so close together, so that the spaces or pits 

 L'd l)y them arc larger. The bars are frequently somewliat oblique, so that they 



('ni-l(>se( 



