1987 SCHNOES AND HUMPHREY: TERRESTRIAL COMMUNITIES IN FLORIDA 95

haps the most significant difference was with the woodpeckers. Four spe-
cies (red-bellied and pileated woodpeckers, yellow-bellied sapsucker, and
common flicker) were found in the older category, whereas only the com-
mon flicker was found in the 15-30 year class. This reflects the increased
availability of larger trees and dead snags for both nesting and foraging.
 Another interesting difference was that wood ducks were found in all
three of the older sites, but in none of the 15-30 year old sites. Again,
this could be due to the presence of trees large enough to contain cavi-
ties. The hypothesis that nesting sites limit wood ducks to use of the
forested sites could be tested by erecting artificial nesting boxes in younger
areas and observing subsequent use of the sites.
 Palm warblers, typically occurring in open ground and brushy habi-
tats, declined in importance with increasing age of the sites after reach-
ing their peak numbers in the 5-15 or 15-30 year old category. The fairly
large number of palm warblers on the older Bartow South site may have
been in response to the surrounding areas that were dominated by old-
fields.
 Unlike the diversity measures, the abundance figures for the 15-30
year class and the >30 year class with lakes were quite different (Fig.
19). This pattern is misleading because the younger treatment was not
sampled during the fall quarter and, among the older sites, only Bartow
South was sampled. A great many small birds (2071) were recorded; over
half of these were wintering yellow-rumped warblers in large flocks. This
extraordinarily high count affected the four-quarter mean for this treat-
ment. It may have been representative of the site (the winter count was
even higher), but it was not balanced by the other two sites, which gen-
erally were lower. In fact, using the three quarters for which complete
data are available, the mean number of individuals is higher for the 15-
30 year class than for the >30 year class (623 vs. 502). Domination of the
younger site by wax myrtle created a very concentrated, high energy
food source for wintering birds. Because the large flocks of wintering
warblers have such a major effect on the total autumn and winter quarter
counts, their habitat use is directly related to the total year counts. This
is supported by the fact that Bartow South had far more yellow-rumped
and palm warblers than Saddle Creek Park and Sanlan Ranch, where the
importance of wax myrtle was much less.
 The larger differences between old sites with lakes and those without
(Fig. 19) resulted from the absence of water and the small wintertime
berry crop in the latter category. Obviously all wading and water birds
were absent from the bird community, which greatly reduced the diver-
sity and biomass of birds found on these sites. Though the passerine
species composition was very similar, abundance values were much lower