arrive in late April, and most are present by early May (Kuchel 1977, Ashley 1994, Reichel and 

 Center unpubl. data). Two-year-old females may arrive later than older females (Ashley 1994, 

 Kuchel 1977:32); this age group may be the unpaired females that Wallen (1987) reported as 

 arriving about 4 weeks later than pairs and then not breeding. Males begin leaving Montana by 

 late-May, and are typically gone by late June (Kuchel 1977, Reichel and Center 1993, Ashley 

 1994). Females begin leaving by early July if breeding is unsuccessful, and otherwise by mid- 

 late July. Juvenile birds leave last, beginning in late July, and both adult females and juveniles 

 are gone by the begirming of September (Ashley 1994, Reichel and Center unpubl. data). 



In Washington, birds arrive on breeding streams in late March or early April (Schirato 

 1993). In Oregon, birds arrive on the breeding streams in late April, although some have been 

 reported as early as late February (Latta 1993). 



Of 249 Harlequins banded in Montana from 1991-1995, a minimum of 24 have been 

 reported from Oregon (2), Washington (1), and southern British Columbia (21), including 

 Vancouver Island and Hornby Island. Sexes and ages at banding show the following numbers 

 and percentages observed: adult females (6, 1 1%), adult males (2, 5%), juvenile females (9, 7%), 

 and juvenile males (7, 5%). Two females radio-marked in Idaho were located in the San Juan 

 and Culf Islands of Washington and British Columbia, while one banded bird was reported from 

 northwestern Washington (Cassirer and Croves 1994). The only known wintering bird marked 

 in Wyoming was observed off of San Juan Island in Washington in August 1989; he returned to 

 Crand Teton National Park as an unpaired male in 1990 (Cassirer and Croves 1991, Wallen 

 1993). 



Some evidence of staging areas on the breeding grounds exists. Some marked harlequins 

 observed in early spring on McDonald Creek, Montana, disappear almost immediately (Kuchel 

 1977, Ashley pers. comm.); these may be going to different drainages in the vicinity. At 

 Kootenai Falls, Montana, in the early 1980s, only 1 pair bred in the immediate vicinity, while up 

 to 6 other adults appeared to loaf there prior to and following the breeding season (Thompson 

 1985, Center unpubl. data). 



There are few records of birds stopping between their breeding areas and wintering areas. 

 A single marked bird has been observed en route from wintering to breeding grounds. She was 

 originally marked in Wyoming and observed on the way back to the breeding stream on Crooked 

 Creek, South Fork Clearwater drainage, in central Idaho and seen about a week later in Grand 

 Teton National Park (Cassirer and Groves 1991, Wallen 1993). 



Migratory behavior. It is believed that nearly all one-year-old birds, and some (perhaps 

 most) two-year-old birds remain in coastal water, not moving to breeding streams until they are 

 2-4 years of age. The proportion of each age class which stays on the coast has yet to be 

 determined, but indications are that perhaps V2 of 2-year-old females and Va of 3-year-old females 

 do not return to the breeding grounds (see DEMOGRAPHY AND POPULATIONS: 

 MEASURES OF BREEDING ACTIVITY - Age at first breeding; intervals between breeding). 

 Wallen (1987) reported that a 1-year-old female (n=l 1) returned to Upper Moose Creek, her 

 natal stream in Grand Teton National Park in 1986. This is the only report of a 1 -year-old female 

 on the breeding grounds. No one- or two-year-old males, out of 176 male observations, have 

 been seen in Montana during 1992-95 surveys (this report, Ashley pers. comm.). 



13 



