1542 U.S. NATIONAL MUSEUM BULLETIN 2 37 part 3 



dry logs on the sea beach, along roadsides and in partly cleared 

 places, in tule swamps, about beaver ponds, and even in the dense 

 Douglas fir forest. Everywhere it keeps fairly close to the ground, 

 and in general it remains in wet or marshy places." Frank S. Tompa 

 (1962), in an important study of the population on Mandarte Island, 

 B. C, found that the birds defended territory in shrubby growth, 

 included grassland within their home ranges, and foraged in the tidal 

 zone as a common feeding ground. 



Territory. — Tompa's work on the sedentary population of Mandarte 

 Island provides the following data: Territorialism revived in late 

 January, when singing and chasing began in males. At tins time 

 adults with previous breeding experience were occupying their terri- 

 tories and home ranges of the year before, substantially without 

 change in the former boundaries. Land left vacant by mortality 

 was occupied by birds hatched the preceding year, and a surplus of 

 such young birds was distributed in loose groups of five to ten in- 

 dividuals. With the revival of territorial behavior, most young 

 birds that were still unsettled on a site emigrated; one was found 

 1,300 meters from its hatching site and another 6 kilometers. Pair 

 formation occurred in February and March. 



Territory size of mated males averaged 288 square meters for 47 

 cases, and home range averaged 473 square meters. This average 

 territory area is one-tenth the minimum size Nice reported for terri- 

 tories of euphonia in Ohio. Unmated males on Mandarte Island had 

 territories averaging only 82 square meters. Tompa suggests that the 

 amount of shrubbery defended determined whether a male could 

 attract a female. 



Territorial behavior gradually declined during the breeding season 

 until it reached its minimum in late July and August when molt began 

 in adults. Song was very rare by late June and early July, except in 

 unmated males. Aggressive behavior revived in October. Some 

 young of the year became territorial in late summer, and from late 

 August until October there was an emigration of certain of these birds, 

 apparently as a result of mutual aggressiveness and the shortage of 

 vacant habitat on which to settle. Aggressiveness declined again after 

 October and reached the minimum in November and December. 



Tompa's investigation establishes that the upper limit of territory 

 size in this local population is fixed primarily by territorial behavior. 



Details of early spring behavior of birds at Vancouver, B.C., are 

 reported by William M. Hughes (1951). Males sing on bright winter 

 mornings, and other signs of territoriality appear early in the season. 

 Hughes remarked aggressiveness in birds in January and color- 

 banded three, which he sexed as males. He found that these marked 

 birds fed amicably with other song sparrows (he caught all three 



