462 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1960 



on their backs for several seconds before righting themselves and 

 swimming avray. Capture in this easy fashion does not hurt the 

 birds in any way. 



All possible nonbreeding swan are captured and banded at Red 

 Rock Lakes. In 1956 a total of 103 birds were captured and 66 were 

 taken in 1957. The larger number in 1956 represented 65 percent of 

 the nonbreeding birds in the refuge. Thirty percent of the 66 cap- 

 tured in 1957 were repeats of birds banded in previous years. Normal 

 aluminum bands used on other waterfowl have not been success- 

 ful with swans. Trumpeters wear out bands faster than smaller 

 waterfowl. As a result, stainless-steel locking bands are now in use. 

 Birds in Canada are also extensively banded, and a very few Canadian 

 swan have been recovered in the tristate triangle. 



There are two trumpeter census counts each year. One is in Au- 

 gust on the breeding areas, and the other in January on the wintering 

 areas. The winter census is part of the nationwide annual waterfowl 

 inventoiy. Like all wildlife census figures, the census results are 

 not a complete count, but do represent the trend of bird numbers 

 over a period of years; they are essential in evaluating population 

 increase or decrease. The counts have gradually climbed from 73 in 

 1935 to 735 in 1958, with a general leveling off since 1951. Fluctua- 

 tions are apparent even in this level plateau of swan numbers, but 

 since the birds have such definite territorial requirements, it is most 

 likely that they are due to flaws in the counting method, rather than 

 changes in bird numbers. This does not imply that the census of 

 trumpeter swans is haphazard or that the methods used are not good. 

 On the contrary, swan counts are checked repeatedly. The enumera- 

 tion of any wild creature has inherent inaccuracies simply because 

 the count is made of wild creatures, free to hide, move, or appear as 

 they choose. Such counts are among the most valuable wildlife man- 

 agement tools used to indicate the relative changes in wildlife num- 

 bers. The results of such a census must be considered as an index 

 number, not as the exact population of the species counted. 



Over a long period, it became evident that the refuge was producing 

 about all the trumpeters that would live there, although some spread 

 continues through Island Park as more vacant breeding areas are 

 taken up. It was decided in 1938 to undertake the transplanting pro- 

 gram that culminated last year in cygnet rearing in Oregon and 

 Nevada. Over 200 birds have been transplanted, but last year was 

 the first successful breeding. The way has been pointed, and several 

 other areas are being considered as new homes for trumpeters. Trans- 

 planted swan are no longer pinioned, but the primary feathers are 

 clipped on one wing to hold them flightless at their new home for a 

 year. After that they are free to move. Low water at Malheur 



