Time and Localities of Recovery 



In both 1954 and 1955 the recovery rate was 

 slightly higher for units released in early summer 

 than for those released in late surnmer and fall 

 (tables 1 and 2). Recoveries were highest on week- 

 ends and holidays. A similar time distribution of 

 recovery of drift bottles has been noted in Saginaw 

 Bay and Lake Huron (Johnson 1958). 



It is of course necessary that a unit reach shore 

 if it is to be recovered. The actual efficiency of re- 

 covery of beached units is strongly influenced also 

 by the conspicuousness of the unit once it has landed 

 and the amount of shore traffic through the area. 

 The latter factor, in turn, depends upon shore top- 

 ography and the density of the human population. 

 In southern Lake Michigan where the population is 

 dense and sand beaches are extensive, most areas 

 are covered equally well. Here the uneven distribu- 

 tion of shore traffic offers a scant source of bias. In 

 northern Lake Michigan, on the contrary, the exten- 

 sive areas of sparse population and the more rugged 

 coast line contribute to lesser and unevenly distrib 

 uted shore traffic. Records of recovery are corres- 

 pondingly biased. 



High visibility of units on shore is essential to 

 efficient recovery and differences in conspicuousness 

 probably inject a bias in comparison of percentage 

 recoveries of different types of units. In early Sep- 

 tember 1954, causes of differential rates of recovery 

 were studied by inspection of 5 miles of shore line 

 in southern Lake Michigan, The following possibil- 

 ities were revealed : plastic envelopes can be blown 

 far back from the beach by strong winds and can be 

 covered by blown sand much more easily than tubes 

 and bottles; the color of the cards in plastic envelopes 

 blends well with the flotsam in windrows along the 

 beach; on open sand the glass bottle with drag is much 

 mere conspicuous than the tube or envelope because 

 both the bottle and the drag attract attention; and 

 glass bottles appear to be sought for rifle practice as 

 was evidenced by the great number of Ixoken bottles 

 along the beach and the everpresent empty cartridge 

 cases. 



SURFACE CURRENTS IN LAKE MICHIGAN. 

 1954 and 1955 



The understanding of discussions in the follow- 

 ing sections should be made easier by a brief pre- 



liminary explanation of the methods of graphical 

 representation of the data (figs. 3-18). The fig- 

 ures on wind direction (figs. 3 and 8) were designed 

 in the same way for each month in each year and 

 the plan of jxesenting records of recoveries (figs. 

 4-7; 9-18) is the same for each transect, regard- 

 less of type of drift unit. 



The information on wind within a month is 

 given in terms of lines that show the percent- 

 age of the total time the wind blew toward each 

 of 8 sectors. The individual lines bisect the 

 sectors 000° -045°, 045°-090°, • • • ; 315° -360°. The 

 percentage of time is indicated by the length of line; 

 the radii of the concentric circles are in multiples 

 of 10 percent. 



The graphs on recoveries give direction and 

 number. The sectors are the same as described for 

 the data on wind. The lengths of the radiating lines 

 indicate numbers recovered in each direction from 

 each station in acccffdance with the scale at the bot- 

 tom. Care should be taken not to interpret these lines 

 as showing actual locality of recovery. Aline leading 

 from an inshore station dkectly onto the shore (on oc- 

 casion passing beyond it) does not indicate that the le- 

 coveiies were made from the immediately adjacent 

 beach. Many could have been, and indeed were, re- 

 covered in remote sections of the lake not shown on 

 the strip map, but which would have been traversed by 

 an extension of the line. To reach these points the 

 units had to drift around intervening headlands or islands. 

 Release stations on some of the transects do not lie in 

 a straight line. This irregularity was because of prob- 

 lems in navigation in aossing the lake. 



Tables 7 -22 in the appendix give the location of 

 release and recovery of drift units, number of days a- 

 diift, milestraveled, and average drift rate in miles per 

 day. A bottle recovered on the day it was released was 

 considered to have been adrift 1/2 day. If it was re- 

 covered the day after release, the drift period was one 

 day. Stations niunbered in the tables are located in 

 figures 4-7 and 9-18 by counting from right to left (east 

 to west) with the exception of the Manitowoc -Sturgeon 

 Bay transect (figs. 9, 10); on this transect stations are 

 located by counting from south to north. 



Central and Southern Lake Michigan, 1954 



South Hayen - Waukegan transect. --The recov- 

 eries from releases along this transect illustrate the 



