120 Botanisches Centralblatt. — Beiheft 2. 



heavy thickening of the inner cell-walls in the larger bundles. The 

 outline of the mestome-bundles, when examined in transverse 

 section, is mostly orbicular in Lycurus. 



These allies of Aristida are thus readily distinguished from 

 tliat genus (not including those with plumose awns) by their 

 having only one parenchymasheath, and from all the species of 

 Aristida examined, by the presence of a mestome-sheath. — A 

 colorless parenchyma, mostly located between the ribs is developed 

 in Lycurus^ MuJilenhergia gracilis, gracillima and distichophylla, but 

 is wanting in the other species of Muhlenbergia and was not 

 noticed in any of the species belonging to the genera: Ötipa, 

 Oryzopsis, Eriocoma, Nassella and Pi-ptochaetium, while it exists in 

 all the Aristidae. The mesophyll is in most species developed as 

 palisades, but it occurs also though seldom, as a homogeneous 

 tissue of roundish cells in Stipa avenacea, capillata, ßlifolia and 

 in Kassella. A mesophyll consisting of both palisades and roundish 

 cells was found in Oryzopsis micrantha. 



Thus is Aristida s. s. considered from an anatomieal point of 

 view a well-marked genus, when compared with its nearest allies, 

 at least with the species, which we have had an opportunity to 

 examine, and the most salient structural peculiarity is, of course, 

 the presence of a double parenchyma-sheath. The constant abseuce 

 of a mestome-sheath is of less importancCj although it constitutes 

 an excellent anatomieal character in this particular case. — The 

 subgenus Arthratherum contains, as stated above, some species that 

 have and some that have not a double parenchyma-sheath, but in 

 none of these does there appear any trace of a mestome-sheath. 

 In this way the three species with naked awns in which two 

 parenchyma-sheaths are developed appear to be more closely 

 connected with the genus, than those with plumose awns and a 

 Single parenchyma-sheath. 



The lacking of a mestome-sheath in both may possibly 

 indicate some closer affinity to the other subgenera of Aristida 

 than to Stipa and all those of the same tribe, which have mestome- 

 sheaths. 



These structural divergences are, thus, to be found in a large 

 number of species, which are not only considered as near allies 

 of Aristida, but they are moreover, in many instances associated 

 with the genus. Stipa avenacea and Muhlenbergia cajjillaris inhabit 

 the same kind of soil as several species of Aristida viz. : Aristida 

 oligantha, gracilis, purpurascens etc., while Aristida fascicidata 

 inhabits the piain s of Colorado in Company with species of Stipa^ 

 Oryzopsis and Eriocoma. 



Having thus considered the leaf structure in genera to be 

 reckoned among the nearest allies of Aristida., we shall now present 

 a few notes upon the corresponding structure of other genera, 

 among which are several species that are offen found to be 

 associated with Aristida, especially on the plains and in the dry 

 fields, which are not, however, nearly allied to that genus. 



