T. Holm — Studies in the Cyperacew. 4'29 



cms short and obtuse papillae around the stomata. The ventral 

 epidermis is rather thin-walled and shows only a few papillae ; 

 above the midrib is a band of bulliform cells in a single stra- 

 tum. In regard to the stomata, these occur on both faces of 

 the blade, but are most numerous, however, on the dorsal ; they 

 are sunk below the surrounding epidermis, and the air-chamber 

 is deep and quite wide. 



The stereome is well developed and distributed as hypoder- 

 mal groups accompanying the veins, on both the leptome- and 

 hadrome-side of the larger ones, but only on the leptome-side 

 of the smaller ; a small isolated group of stereome is to be 

 observed in the leaf-margins. The chlorenchyma represents 

 a homogenous tissue of roundish cells (in cross-section), and no 

 palisade-cells were observed ; it is broken down into wide 

 lacunes, one between each two mestome-strands. All the mes- 

 tome-bundles are collateral and possess a thin-walled, green 

 parenchyma-sheath and a mestome-sheath with the inner cell- 

 walls thickened. 



The perigynium. 



The perigynium is generally described as being glabrous, 

 but the surface is, nevertheless, distinctly granular from numer- 

 ous papillae like those observed in the green leaves. The 

 cuticle is thick and smooth, and the outer epidermis is quite 

 thick-walled (fig. 10) in contrast to the inner one (Ep.* in tig. 

 10). A thick-walled stereome occurs as hypodermal strands 

 on both faces of the two mestome-bundles, besides as three 

 isolated strands on the flat, narrow side of the perigynium and 

 as nine on the much broader, convex one. The chlorenchyma 

 is poorly developed as a single layer between the veins, and as 

 two to three layers around these. The mestome-strands are 

 very thin and consist mostly of leptome ; no parenchyma 

 or mestome-sheath was observed, at least not as continuous 

 sheaths. 



We have, as stated above, placed C. holostoma among the 

 "format centrales"* as an ally of C. stylosa C. A. Mey. and 

 C. Raynoldsii Dew., and we might now add a few remarks 

 upon the structure of these species. The difference in structure 

 is relatively slight, inasmuch as they all show the principal 

 anatomical features exhibited by the genus Car ex in general. 



Carex stylosa C. A. Mey. 

 The roots. 



Our figure 11 shows a transverse section of one of the strong, 

 secondary roots, and we notice here nine strata of very thick- 



* " Greges C'aricum." This Journal, vol. xvi, p. 457, Dec. 1903. 



