198 NATUEAL HISTOEY BULLETIN 



The following plants were collected on the border strip be- 

 tween stations 1 and 2. They present a mixture of species of 

 both forest and prairie, the latter predominating: 



Amorpha canescens Monarda mollis 



Anemone cylindrica Onosmodium occidentale 



Aster sericeus Poa pratensis 

 Ceanothus ovatus var. pubescens Quercus macrocarpa var. 



Eupatorium altissimum olivaeformis 



Fragaria virginiana Ehus glabra (small) 



Fraxlnus pennsylvanica (small) Sanicula marilandica 



Gerardia aspera Symphoricarpos occidentalis 

 Helianthus scaberrimus 



Station 3 was located at an elevation of 175 feet above the val- 

 ley on the most prominent point on this side of the bluffs, and 

 fully exposed to the south and west. It was about 350 feet almost 

 due south of station 1, and like the latter was surrounded by a 

 distinct prairie flora. Its position is shown at the point marked 

 (3) in plate VIII, figure 1, and plate IX, figure 1. See also plate 

 YII, figure 1. This is the most exposed station of the series. 



Station 4 was located on the east side of the same ridge at a 

 point 375 feet east and 220 feet south of station 3 and at the same 

 altitude as that station. The ridge between stations 3 and 4 rises 

 about twenty-five feet above them and its slope on the east side is 

 sufficiently steep to afford protection from excessive evaporation 

 on that side. The eastern slope was formerly covered with forest, 

 but this had been removed and a young orchard has been set out 

 on a part of the slope. Station 4 was in an open place sheltered 

 only by the ridge but in an area which had been but recently for- 

 ested. This station is shown at the point marked (4) in plate IX, 

 ffgure 2, looking south along the cleared slope. 



The observations at stations 3 and 4 should be compared to 

 "bring out the protective influence of the ridge alone. 



Care was exercised in placing the pieces of apparatus required 

 for these observations in such manner that they did not interfere 

 with each other. 



The observations were continued for a whole day at intervals 

 of about a week from the 13th of August to the 12th of Septem- 

 ber. Each observation day was given wholly to this work and 

 liourly readings (excepting those of the evaporating pans) were 



