364 
Psyche 
[Vol. 94 
occupy a web site for many days, and even weeks, whereas others 
relocate frequently. Since web site relocation requires a greater 
energy expenditure than does remaining at a site, the decision to 
leave may be related to the quality of the web site. This study used 
residence time at a particular web site as an indicator of microhabi- 
tat suitability and examined the relative contributions of feeding 
rates, web destruction, and sun-exposure (heat stress) to the ten- 
dency of a spider to remain at a web site. 
Methods 
This study was conducted in an approximately one-acre area of 
deciduous forest on the University of Georgia campus, Athens, 
Georgia (Clarke County). A thick understory of Rubus sp. (black- 
berry) and Rhus radicans (poison-ivy) and a canopy of Quercus alba 
constituted the major plant species of the study site. 
One 12 July 1984, 25 M. gracilis were collected from the study site 
and from the University of Georgia Botanical Gardens. Only penul- 
timate and mature females were collected. Spiders were marked 
with fast-drying enamel paint, using the positioned dots technique 
(Walker & Wineriter 1980). Ten spines on the abdomen of M. gracilis 
made convenient locations for different combinations of paint 
marks, giving each spider a unique identity. This paint mark identity 
was assigned a number to simplify data collection. Blue, green, 
white and black paint were used on the rationale that these colors 
would have the least effect in making the spider more conspicuous 
to vertebrate predators (though Rypstra (1984) has found that such 
paint marks may have very little effect on predation rates). Spiders 
were retained for 24 hours to assure that they would survive the 
marking procedure (which involved no anesthesia). 
Spiders were released at random locations (using the wandering 
quadrant technique of Catana (1955)) at 1800 hrs on 13 July, and 
allowed 6 days to establish a web site. Observations were initiated 
on 20 July. The position of each spider was marked with a flag 
bearing its number identity, placed in the ground below the web. 
Unmarked spiders were marked in the field as they were discovered. 
Each day the distance (if any) moved by each spider was measured 
with a tape measure and recorded. Each web was checked hourly 
from 1000 hours until 1700-1900 hours. The following information 
was recorded at each hourly check: if the spider was feeding, was in 
