Influence of Riparian Woodland
Although the largest single element of the Kansan avifauna that reaches distributional limits in Kansas is made up of birds of the eastern deciduous forest, several species of the eastern woodlands are present in Kansas along the east-west river drainages in riparian woodland; the species are listed in [Table 8]. Twenty-one kinds are involved if we include the Cooper Hawk, Yellow-billed Cuckoo, Orchard Oriole, Summer Tanager, Rufous-sided Towhee, and Chipping Sparrow, all of which breed farther to the west but are present in western Kansas only along river drainages. This leaves 15 species of eastern deciduous woodlands that occur west in Kansas along riparian woodland (versus 30 species that drop out chiefly where eastern woodland drops out). These 15 species are about one-third of all woodland birds in western Kansas. Riparian woodland does not seem to afford first-rate habitat for most of the eastern woodland species that do occur; breeding density seems to be much lower than in well-situated eastern woodland.
The importance of these linear woodlands as avenues for gene-flow between eastern and western populations, especially of species-pairs (grosbeaks, flickers, orioles, and buntings), is obviously great. Likewise significant is the existence of these alleys for dispersal from the west of certain species (for instance, the Black-billed Magpie and the Scrub Jay) into new but potentially suitable areas.
BREEDING SEASONS
Introduction
An examination of breeding seasons or schedules is properly undertaken at several levels. The fundamental description of variation in breeding schedules must itself be detailed in several ways and beyond this there are causal factors needing examination. The material below is a summary of the information on breeding schedules of birds in Kansas, treated descriptively and analytically in ways now thought to be of use.
Almost any event in actual reproductive activity has been used in the following report; nestbuilding, egg-laying, incubation, brooding of young, feeding of young out of the nest are considered to be of equal status. To any such event days are added or subtracted from the date of observation so as to yield the date when the clutch under consideration was completed.
Such corrected dates can be used in making histograms that show the time of primary breeding activity, or the "egg-season." All such schedules are generalizations; data are used for a species from any year of observation, whether 50 years ago or less than one year ago. One advantage of such procedure is that averages and modes are thus more nearly representative of the basic temporal adaptations of the species involved, as is explained below.
When information on the schedule of a species from one year is lumped with information from another year or other years, two (and ordinarily more than two) frequency distributions are used to make one frequency distribution. The great advantage here is that the frequency distribution composed of two or more frequency distributions is more stable than any one of its components. Second, the peak of the season, the mode of egg-laying, is represented more broadly than it would have been for any one year alone. Third, the extremes of breeding activity are fairly shown as of minute frequency and thus of limited importance, which would not be true if just one year were graphed. All these considerations combine to support the idea that general schedules in fact represent the basic temporal adaptations of a species much better than schedules for one year only.
Variation in Breeding Seasons