19 



and below tbe bandies and at the margin are very thick- walled, with 

 the cavity almost obliterated by the inner thickening layer. 



In all the species that we have described it will be observed that the 

 general arrangement of the tissues is the same — that is, the chlorophyll- 

 bearing tissues are arranged in two rings concentric with each hbro- 

 vascular bundle, and each concentric system is separated from the 

 next by a band of colorless parenchyma. All but J. straminea are 

 deeply furrowed on the upper face, and all, either on one or on both 

 faces, are well furnished with exodermal exi^ansions. The furrowed 

 leaves are either permanently closed as in J. pilosa and S J. straminea, 

 or have the power of rolling or unrolling in accordance with varying 

 degrees of moisture. The 5 J. straminea alone lacks this power, and 

 thus shows itself less fitted for extremes of drought and heat than do 

 the others. Bands of large thin-walled, colorless parenchyma cells are 

 very common in grasses of dry regions. It is thought that such cells 

 supplement those of the epidermis as storage places for water. It has 

 also been suggested that when situated under bulliform cells they aid 

 these to i^erform their function. In some cases, however, this is clearly 

 not the actual effect; for, as in the exami)le before us, J. straminea, 

 such cells may be present in unusual size and abundance and still the 

 leaf may prove singularly immovable. It is possible that such an 

 arrangement aids in admitting light to the interior of the leaf, and this 

 view is supported by the fact that the lobes of the chlorophyll cells are 

 perpendicular to the band, just as palisade cells are perpendicular to 

 the surface of many leaves. The advantage of additional means for 

 utilizing light is evident when we consider that the extent of leaf sur- 

 face exposed to the air is frequently reduced to a minimum in order to 

 prevent excessive evaporation. 



A strong development of fibrous tissue and strong bundle sheaths 

 are common to grasses of dry regions, but it is rare that isolated strands 

 of such tissue ap])ear below the bands of colorless cells as in J. pilosa; 

 they are usually found opposite the bundles only or form a continuous 

 layer along the whole under surface of the leaf. It is also of rare 

 occurrence to find so marked a difference between basal and median 

 sections of the same leaf. The hypodermal fiber of J. pilosa, as we 

 have seen, presents much variety both in quantity and in character. 

 Its cells range from those with scarcely thickened walls to those whose 

 cavity is almost entirely obliterated, and range from a group of a fex? 

 cells only to very large, strong masses. A comparison of leaves of very 

 rigid plants of D. spicata with the least rigid specimens of the same 

 species shows that in this grass also the same thing is true of fibrous 

 tissue. Its presence in certain positions is practically constant, but the 

 size of the strands and the thickness of the walls of the individual cells 

 vary much. 



Although the leaves of the grasses under consideration diifer from 

 each other sufiicieutly to render it unlikely that necessity should arise 



