TABLE 6. Location of Species The median day of occurrence is not necessarily the peak density loca- tion; however, for the most part it agrees closely with day of max. lin. density. This table shows a reasonable transition from temperate to sub- tropical to tropical elements of the avifauna. TABLE 7. Range of Species An attempt is made to arrange the species according to restrictedness or broadness of range. The number of days of spread was determined by the number of days from first to last sighting. It is not a reliable indicator of broadness of range because it cannot account for ranges extending past either extreme. It is however useful in estimating some cases of "rareness," and does provide a measure of discontinuity. If the days spread figure is larger than the days seen figure, the difference is the relative amount of discontinuity. Sabines Gull, Dark-rumped Storm Petrel, Jaegers, and Red- billed Tropicbirds have discontinuity differences of 5 days or more in this instance. This should be interpreted as meaning these birds probably occur over at least all the indicated spread but are variable or so low in num- bers that they are often not seen. Starting with the assumption that one or more ecological zones were crossed, I attempted to determine where it or they were located as re- lected by changes in the bird populations. If hypothetical ecological zones did exist they could be most easily delimited by locating their boundaries -- where they come together. At these boundaries I further hypothesized the following: 1) new (southern) species would appear; 2) linear density would increase; 3) daily species total would increase; and 4) northern species would disappear. TABLE 8. Chronological Occurrence of Species This table lists the day each species was seen. Appearances and disap- pearances were arbitrarily weighted according to importance as measured by the overall linear density. TABLE 9. List of Species by Relative Importance Primary species were weighted at 5 units, secondary at 3 units, and tertiary at 1 unit. TABLE 11. Ecological Boundaries of Species The weighted totals of appearances and disappearances for each day are shown in Table 11. The linear density for each day was assigned as positive, if higher than the median and as negative, if lower than the median. A com- parison of the four columns revealed the four hypothesized boundary condi- tions were met on three occasions as indicated by brackets in Table 11. In despite of the highly artificial nature of the approach, comparison of these three areas with current charts showed that ocean features at the same loca- tions were in transition states, and a positive correlation was obtained.