72 BULLETIN OF THE 



and there thus exist a large number of these green oases, as it were, 

 interspersed over the otherwise botanical desert. 



In consequence of this fact it requires experience in order to 

 improve the facilities which the place affords. A botanist unac- 

 quainted with the proper localities for successful collection might 

 spend a month almost in vain, and depart with the conviction that 

 there was nothing here to be found. It may not be wholly pecu- 

 liar, but these favored localities are here often of very limited ex- 

 tent, and in situations which from a distance afford no attraction 

 to the collector. Civilization is, however, very perceptibly encroach- 

 ing upon many of them, and it is feared that in another half cen- 

 tury little will be left but a few bare rocks or inaccessible marshes. 



In naming localities the principal authorities relied upon are : 



1. A recent Atlas of fifteen miles around Washington, including the 

 County of Montgomery, Md., Compiled, Drawn, and Published from 

 Actual Surveys, by G. M. Hopkins, C. E : Philadelphia, 1879 ; and, 



2, a military map of Northeastern Virginia, published in the work 

 of General J. G. Barnard, on the Defences of Washington, 1821. 



From the former the names of many roads, streams, estates, &c, 

 have been obtained, while from the latter those of forts, batteries, 

 &c, are often employed as more convenient. In this respect, how- 

 ever, much remains to be desired. While the military map is 

 antiquated, the other is frequently defective in omitting what is 

 required and incorrect in erroneously locating streams and other 

 objects well known to the writer. In his extensive rambles he has 

 learned many local names not found on the map, and in a few 

 cases of special botanical interest, where names are wholly wanting, 

 he has long been in the habit of designating the localities by names 

 of his own christening, and for which he offers no apology. 



The following are a few of the principal places of botanical in- 

 terest which will be found to recur most frequently in the notes, 

 and for this reason brief descriptions of them are appended. 



1. The Rock Creek Region. — Rock Creek which forms the bound- 

 ary line between Washington and Georgetown (West Washington), 

 has escaped to a remarkable degree the inroads of agriculture and 

 population. For the greater part of its length within the District 

 of Columbia its banks are still finely wooded for some distance 

 back, and afford a rich and varied field for botanical exploration. 

 The character of the surface along Rock Creek is most beautiful 

 and picturesque, often rocky and hilly with frequent deep ravines 



