312 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1953 



One has to choose from unions the gamut of communities lying be- 

 tween these, and somewliat unfortunately segregated as "grasslands" 

 and "shrublands." Their number and variety should demand a far 

 more complex classification. The type of community should be chosen 

 in respect to its nature as : (1) fulfilling the physical demands of the 

 land in respect to passability for patrol, maintenance, reconstruction, 

 etc.; (2) being cheapest in its "constiniction," i. e., herbicidal root-kill 

 (not just top-kill) of the unwanted existing trees; (3) being cheapest 

 in its maintenance through the years, i. e., with the smallest invading 

 relay of trees; and (4) being highest in conservation and public-rela- 

 tions values, involving landscape, game animals, song birds, and fire 

 hazards. 



The type of herbicide treatment now enters the botanical picture. 

 Summer-foliage blanket spraying tends to remove the broad-leaved 

 plants, trees, shrubs, and forbs, and to leave communities of grasslike 

 plants — some of them grasses, others sedges — devoid of attractive 

 wildflowers, and of legumes so important to wildlife. Such vegeta- 

 tion, extraordinarily varied from region to region and soil to soil, is 

 arbitrarily and unsatisfactorily here lumped under the designation of 

 "grassland." Conversely, selective basal spraying leaves a far more 

 varied mixture of grasses, forbes, and shrubs, each community of 

 which is worthy of separate observation and study. Since shrubs 

 frequently dominate, this entire assemblage of plant communities is 

 unceremoniously referred to as "shrubland," so called by virtue of the 

 herbicidal treatment applied, and not because it is composed con- 

 tinuously and constantly of shrubs. 



The economics of vegetation management of rights-of-way and 

 roadsides has now developed into an evaluation of whether the post- 

 blanket-sprayed "grasslands" or the post-selective-sprayed "shrub- 

 lands" are: (1) cheapest for conversion to them; (2) cheapest for 

 maintenance of tliem in respect to reinvading tree seedlings; and 

 (3) highest in public relations. Brought to play upon this subject 

 are my own investigations at Norfolk, Conn., extending through a 

 quarter of a century. Field studies of rights-of-way have taken me 

 in recent years through a territory stretching from the St. Lawrence 

 River and the entire Atlantic seaboard to Illinois, Colorado, Okla- 

 homa, and the (lulf coast. Witli this teri-itory in view, and other 

 lands in mind, no botanist would ever make dogmatic assertions con- 

 cerning "brush" and "grass," any more than a forester would encom- 

 pass his knowledge of American forest types and forest-management 

 practices in a single paragraph. 



A generalization is showing through the welter of botanical details. 

 Grasslands of forested regions, 90 percent of them, are appearing as 

 relatively "open" communities — open to invading relays of trees, 



