Avian Distribution Patterns in Forest Islands of Different Sizes in Central New Jersey

THE number of species present in a particular habitat is strongly influenced by the size of that habitat. Early ecological study of the relationship between species number and habitat area dates to Arrenhius (1921) and Gleason (1922). A large volume of literature on the theory and practical applications of the species-area concept has been developed since and is reviewed by Cain and Castro (1959) and Greig-Smith (1969). Hopkins (1955), Preston (1960), and MacArthur and Wilson (1967) give examples of species-area curves for both plant and animal communities, and the shapes of the curves are characteristic of the communities under study (Cain and Castro 1959). Usually there is a direct linear relationship between species number and the logarithm of area. The use of islands of habitat for establishing species-area relationships differs from the use of parcels within the habitat, as enrichment of a parcel by species from adjoining terrain is minimized when islands are used. Also, islands of habitat provide an edge effect, discussed below. In central New Jersey patches of forest habitat surrounded by open fields exist in relative abundance. Because of their relatively discrete nature, these habitat patches may be considered terrestrial islands. They vary in size from less than a hectare to over 30 ha. The purpose of this study was to determine (1) the relationship between size of a habitat patch and bird species richness in the mixed oak forests on the New Jersey piedmont, (2) the forest size requirements of different bird species, and (3) the importance of forest edge and forest interior as distinct zones for birds. iETHODS Forest islands ranin from 0. O1 to 24. 0 ha were available in the study reion and ten size classes were chosen (0. O1, 0. 2, 0. 8, 1. 2, 2. 0, 3. 0, 4. 0,? . , 10. O, 24. 0 ha). For statistical analysis the study region was divided into three eoraphic sections: northern, central, and southern. Each section contained 10 study islands, one of each size class. Forest islands in the northern and southern sections were censused three times and those of the central section were censused twice. The first census of 30 islands was done -29 June, the second census 30 June-23 July, and the third census of the northern and southern sections 24 July-S Auust 1972. Woods within a section were censused in random order. The northern section was censused first followed by the southern and central sections. All censuses were made from 0630 to 0900, and none was conducted on days of heavy rain orTo assure comparable samplin of each forest, the followin procedure was employed.