130 Trows. Acad. Sci. of St. Louis. 



taxonomy by Fournier (334), 1865, who has given details of 

 structure of different members of the order Cruciferae, 

 especially Sisymbrium^ in which good characters were found. 

 Fournier has published several additional papers on the use 

 of anatomical characters in classification. A. Prunet has 

 reviewed the more important works on anatomy with refer- 

 erence to classification, in Bonnier' s Kevue Generale de Bot- 

 anique. Kobinson (434a, 137),who has recently studied our 

 North American species of the order, says in regard to 

 Sisymbrium: " The pubescence, which, if all species of both 

 continents are considered, passes from simple or occasionally 

 forked hairs to dense stellation, fails to give a really satisfac- 

 tory generic distinction." Dennert's (384) paper indicates 

 one general type of stem structure and then gives an artificial 

 key based on the presence or absence of hard bast, showing 

 no parallelism between related genera. This order Cruciferae 

 is one in which anatomical characters are much used, but not 

 always in a satisfactory way. 



The minute structure of the leaves of Coniferae has long 

 been used as an aid in the diagnosis of species, as in the works 

 of Engelmann (330), McNab (403-405), Coulter and Kose 

 (318), Penhallow (421) and others. The diagnostic value 

 of anatomical characters has been verified by many students. 

 Duval-Jouve (328), 1870, discussed the anatomical char- 

 acters of Gh-afnineae with special reference to Agropyron. 

 His researches show some most striking differences. Hackel, 

 in his monograph on European species of the genus Festuca^ 

 states that the form of the cross-section of the lamina affords 

 a valuable means for distinguishing some species, though one 

 must cautiously consider variation with reference to soil and 

 climate. Many of our manual species afford valuable ana- 

 tomical characters, as the work of Mrs. Hansen (418) shows 

 for Sporobolus. Holm (381), in a series of valuable papers, 

 describes numerous North American grasses and finds most 

 distinctive characters. Ball (299) has established the same 

 fact for Eragrostis, and Miss Sirrine (446), for Panicum. 

 It is not necessary to cite further facts in this order. 



J. Vesque (458-461) has published several very important 

 papers on the application of anatomy to classification. His 



