312 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1934 



of Para, Maranhao, Piauhy, Ceara, and Rio Grande do Norte, a 

 part of Brazil in which few botanical collections, especially of 

 grasses, have been made. In much of this area the chief means of 

 transportation is by rivers which flow from south to north. To 

 reach the grassy savannas of the interior Mr. Swallen was obliged 

 to travel hundreds of miles on muleback. The result was a fine col- 

 lection of grasses now being studied at the National Herbarium. 

 The notes on habit, distribution, and habitat are important aids in 

 determining the systematic relation of the species. 



The collections made by trained agrostologists are especially val- 

 uable additions to the section of grasses of the National Herbarium, 

 which now contains the largest number of specimens of grasses of 

 any herbarium in the world, the number of sheets being more than 

 210,000. 



The classification of plants is an attempt to show systematic or 

 genetic relationships. In many cases the relationship is obvious, but 

 there are groups which cannot be placed with assurance in a system 

 of classification. Not inf reqviently the systematic position of a genus 

 (for example) is clarified by a study of developing organs, the inter- 

 nal anatomy of tissues, or the germination of the seeds. Recently, 

 there has been published a work ^ which bears strongly on the classi- 

 fication of grasses. By a microscopical study of the early stages in 

 the development of the inflorescence of certain grass genera {Setaria, 

 Pe7inisetimi, Cenchrus), Dr. Arber has shown the morphological 

 similarities of the bristles, spines, and involucral bracts which at 

 maturity appear so diverse. She has investigated the rhizomes of 

 bamboos, the corms of Arrhenatherum elatius var. hulhosum^ and 

 the inflorescence of Hordeum. She reviews the work done upon 

 Indian corn or maize by Collins and others in this country and adds 

 many observations of her own upon this remarkable grass. Spartina 

 townsendii (see p. 303) has received her attention, and she lends the 

 weight of her opinion to the theory that the species is of hybrid 

 origin. Many other morphological puzzles have been studied in 

 this important book. 



2 Arber, Agues, Gramineae, a study of cereal, bamboo, and grass. Cambridge Uni- 

 versity Press, 1934. 



