fication, since it is more distinctive of the genus. 



Wiesner's (1900) findings are of interest at this 

 point, He says that the bast cells of the four cereal 

 straws (here under consideration) are so similar as to 

 afford no characters for certain differentiation, while 

 the size and form of the epidermal cells are quite dis- 

 tinctive. Those in rye, wheat and oat he continues, have 

 a pretty regular rectangular general outline. Those of 

 rye have strongly sinuous limiting outline, while those 

 of wheat are straight. Oat on the other hand has epi- 

 dermal cells with a gently wavy outline. Those of barley 

 are rhomboidal or trapezoidal in contour. The writer 

 in view of the facts below presented feels that Wiesner's 

 description is quite inadequate. ... 



Fragments of the epidermis, or at least individual 

 cells are easily found in paper pulp, and since they, of 

 all the elements show the least amount of distortion, 

 their characters -can be seized upon with corresponding 

 readiness and certainty. Of the individual elements 

 (cells) it is difficult or impossible to find those of stom- 

 ata and of dwarf cells in bleached papers, since in 

 these the treatment completely dissociates the cells in 

 question, while their small size allows them to be wash- 

 ed away. In unbleached papers, on .the other hand, 

 fragments of epidermis sufficiently big to carry all the 

 kinds of cells composing it can be found, thus permitting 

 identification. 



The epidermis of the outer surface of the leaf-sheath 

 differs from that of the stem in certain details, and, in- 

 asmuch as for constant length of straw there will be 

 more leaf epidermis present in pulp, it is quite as im- 

 portant an element to be considered as the stem epi- 

 dermis, if not more so. The leaf blades are in minor 

 quantity and are almost negligible in this respect. 



As has already been said above, the epidermis is not 

 uniform in structure, but consists of alternate longitu- 

 dinal zones ; those roofing over the chlorenchyma spaces, 

 and recognizable by the stomata (stomatal zones), and 

 those underlaid by fibrous tissues, devoid of stomata 

 (non-porous zones). The latter, except in wheat, do 

 not afford readily differentiable characters. The com- 

 ponent cells are straight-walled (except in the leaf of 

 wheat) and narrow, roughly about one-half the trans- 

 verse diameter of the cells of the stomatal zones. These 

 alternate, in a longitudinal sense, with the minute 

 dwarf cells, which are either single, or in pairs. In the 

 former case the dwarf cell is from the point of view of 

 morphology a trichome of small size. When two are 

 present, the upper (distal) one is a trichome. In the 

 wheat in addition to the dwarf cells, there is a peculiar 

 cell, roughly circular (ca 30 microns in diameter) in 

 outline en face, and round or quadrangular in the stem, 

 which, on account of the large amount of unmodified 

 cellulose present, take up stains (magenta, Bismarck 

 Brown, safranin, etc.) more readily than the remain- 

 ing epidermal cells, excepting only the accessory cells 

 of the stomata. By their consequent darker staining 

 reactions, they are more readily seen in suitably stain- 

 ■ ed preparations, as seen in figures 9 and 13, and es- 

 pecially in figure 27. In the leaf they assume the form 

 of trichomes. 



On account of their shape I call these cells 'crown' 

 cells. Since they are absent from the other grains which 

 we are considering, the presence of crown cells definitely 

 identifies wheat. 



In barley wheat and rye, trichomes are present. They 

 are especially developed on the terminal internode of 

 rye, and, in fragments of epidermis, their bases are seen 

 as large oval cells in. the positions of dwarf cells. Smaller 

 ones sometimes occur on the leaf sheath (outer surface) 

 in wheat. 



They are absent from the oat excepting only that 

 they are present on the upper surface of the leaf-blade 

 and its margins. Their form is such that the sharp 

 claw-like end lies for the most part parallel with the 

 leaf -surface and so the. roughness produced is not mark- 

 ed. This permits their recognition with sonic certainty. 

 In the epidermis of the stem and outer (lower or dorsal) 

 surface of the leaf sheath, we note the following : 



The cells (exclusive of stomata and dwarf cells) have 

 much more strongly (1) wavy or sinuous longitudinal 

 walls in the leaf-sheath. In the stem they are only slight- 

 ly undulating, or, in oat, straight, or nearly so. Ar- 

 ranged in series based on waviness both in leaf-sheath 

 and stem the four straws are as follows : Wheat=rye ; 

 barley; oat. 



It is difficult to judge on this point as between wheat 

 and rye, since the walls are thinner in the latter and so 

 the waviness is masked. The waviness is most irregular 

 in wheat and least so in oat, the four straws falling, as 

 regards this feature, in the same order as the above. 



The waviness and irregularity in the wheat are very 

 characteristic, and enable one to identify it with con- 

 siderable ease. The oat is equally distinctive by virtue 

 of the relative or entire absence of undulations from the 

 cell walls of the middle of the stomatal zones of the 

 upper region of the leaf-sheath (3) the sinuosities in- 

 creasing in amplitude toward the non-porous zones and 

 toward the base of the sheath. When isolated, the 

 straight Availed cells are indistinguishable from the epi- 

 dermal cells from the inner (upper) surface of the leaf- 

 sheath and from the outer surface of the papery leaf 

 margin, so that, unless leaf-fragments of sufficient size 

 are available, positive evidence can be gained only from 

 the sinuous cell-walls. The character of these sinuosities 

 is well shown in figure 11, which however is from a part 

 of the leaf sheath which does not show the straight cell 

 walls, that is, approximately, the middle region. 



-Rye and barley can be distinguished readily enough 

 from wheat or oat, but not easily as between themselves, 

 if the character of the undulations is depended upon 

 (figures 10-12). They are however easily separable 

 by the absence or presence of trichomes at least as far 

 as the lower part of the leaf -sheath is concerned and by 

 the character of their walls, which are much thicker gen- 

 erally in the wheat. ' 



The character of the undulations is more difficult to 

 judge when the cells are isolated, since the lateral pro- 

 cesses of the cell, corresponding to the undulations but 

 lying below the surface, are much more pronounced. It 

 is therefore necessary to obtain a microscopic image of 

 the outline of the upper face of the outer cell wall, (see 

 figure 14) that namely to which the cuticle is applied. 



(1) That is as to depth, but not as to the number of 

 undulations per unit -of length. 



(3) AViesner's description of the epidermal cells of 

 cat has already been cited. It is evident that he hao 

 regard only to the epidermis of the stem. The waviness 

 of the walls of the leaf-sheath epidermis is so evident, 

 and it is so certain that the Zeaf -sheath contributes large- 

 ly to the fiber in paper that it is quite necessary to 

 have regard to the character of its epidermal cells. 



An effort has been made to base distinctions upon the 

 thickness of the epidermis^cell walls, but this is diffi- 

 cult. The thickness varies with the level, the extreme 

 thickness being found at the base (6 microns for oat) 

 where the amount of fiber is also greatest, and the least 

 thickness (2 microns for oat) just below the ligule. This 

 character varies also with the maturity of the plant. In 

 seneral however, the series seems to be the following 

 in order of thickness of cell wall. 



