RESULTS 



From January 1961 to June 1963, a total of 

 16,285 bottles and cards were released, and by 

 January 1964 a total of 538 returns had been re- 

 ported. This represents a recovery rate of 3.3 

 percent, which is remarkably high considering 

 the ratio of the area covered by releases to the 

 coastline available in the mid-Pacific. It can be 

 compared, for example, with the return rate of 

 4.6 percent reported by Schwartzlose (1963) for 

 releases off southern California. 



On the island of Oahu, where some 80 percent 

 of the population of the State of Hawaii lives, the 

 timing of returns suggests that almost no bottles 

 or cards remain undetected on landfor more than 

 about 1 day, which considerably enhances the 

 reliability of the minimum velocities which can 

 be computed from the time the drift bottle or 

 cards spent in the water and the minimum dis- 

 tance covered. These minimum velocities range 

 from less than 1 mile per day to about 15 miles 

 per day, but within this range there are fairly 

 distinct groups of speeds which appear to be 

 characteristic of seasonal current patterns and 

 of the distance traveled by the drift bottle or 

 card. For returns to the islands of Oahu and 

 Kauai, which represent the largest samples, the 

 frequency distribution of computed minimum 

 velocities shows modes at speeds of about 10 

 miles per day, 6 miles per day, 3 miles per day, 

 and less than 1 mile per day. Table 1 shows the 

 relationship between these characteristic modes 

 in the velocity spectrum and the time of year 

 and the distance traversed. 



49 returns from distances greater than 100 

 miles were not enough to permit breakdown by 

 time of year. 



The choice of time periods in table lis based 

 on a distinct seasonal pattern in the direction of 

 travel of the drift bottles and cards. Figures 4 

 through 9 show the locations of releases and re- 

 coveries, together with the inferred trajectories. 

 The data are grouped by season, except for the 

 period from April to June, where it was possible 

 to chart returns by month. Figure 10 shows the 

 returns from releases outside of the immediate 

 vicinity of the Hawaiian Islands. All releases 

 are shown in figures 4 through 10 as dots; re- 

 leases with returns have trajectories associated 

 with them. In cases where more than one return 

 resulted from a release, only one trajectory is 

 shown, unless the recovery points were relatively 

 widely separated. Trajectories were drawn 

 using the convention that trajectories from near- 

 simultaneous releases cannot cross; Inflgure 10, 

 the trajectories were drawn with the aid of the 

 Atlas of Surface Currents (H. O. Publication 

 no. 570, 1947). 



During Charles H. Gilbert cruises 58 and 63, 

 in July 1962 and January to March 1963, respec- 

 tively, bothdrift bottles and cards were released 

 simultaneously. This was part of a series of 

 tests to determine the relative effectiveness of 

 the two packages. Essentially the same propor- 

 tions of both bottles and cards were reported 

 found, and there was no significant difference in 

 the directions of movement, so that the two pack- 

 ages were considered equivalent in performance. 



Table 1.— Most frequently occurring drift bottle 

 and card speeds (miles per day) as a function 

 of distance and time of year 



Note: The underlined values represent the 

 dominant modal speeds in each case. 



The sample sizes on which table 1 is based 

 are about 50 returns in each case, exceptfor the 

 two sets of values for distances of 100 miles or 

 less from July to December, each of which has 

 more than 100 returns represented in it. The 



DISCUSSION 



The velocity figures shown in table 1 suggest 

 a seasonal change in the velocity of the North 

 Pacific Equatorial Current in the latitude of the 

 Hawaiian Islands. In the early part of the year 

 there was little evidence of speeds of drift of 

 more than 6 nautical miles per day; later in the 

 year, and particularly in returns from more 

 than 10 nautical miles away, there was a consid- 

 erable proportion of returns with minimum ve- 

 locities of from 9 to 11 nautical miles per day. 



This change in speed coincides with a change 

 in the direction of drift. Figures 4 through 6 

 show that most recoveries during the first half 

 of the year came from releases made to the south 

 of the recovery points, so that the bottles and 

 cards must have been carried north through the 

 inter-island channels before coming ashore on 

 the windward (northeast) coasts 'of the Islands. 

 Figures 7 and 8, on the other hand, show a much 



