Sf. Pearson et Da. Manuwal, Influence of niche overlap and territoriality on hybridization between Hermit Warblers and Townsend's Warblers, AUK, 117(1), 2000, pp. 175-183
Hybrid zones between Townsend's Warblers (Dendroica townsendi) and Hermit W
arblers (D. occidentalis) in the Pacific Northwest are narrow relative to e
stimated dispersal distances and appear to be moving, with Townsend's repla
cing Hermits. We examined whether the habitat-transition and parental-fitne
ss asymmetry models can explain why these zones are narrow and moving by co
mparing habitat variables associated with warbler territories in the Washin
gton Cascades hybrid zone. Habitat variables did not differ among phenotype
s, suggesting that the habitat-transition model cannot explain the narrow a
nd dynamic nature of this hybrid zone. Habitat characteristics of Hermit Wa
rbler territories did not differ inside versus outside the hybrid zone, als
o suggesting that this zone is not associated with a region of habitat tran
sition. The lack of difference in habitat use could be the result of compar
ing variables that are not important to pairing success. However, warblers
tended to select territories on west-southwest aspects. South aspects in th
e southern Washington Cascades are dominated by Douglas fir (Pseudotsuga me
nziesii) and true fir, which is a habitat selected by female warblers when
choosing among territories and males. The parental-fitness asymmetry model
does not necessarily make predictions about habitat use within the hybrid z
one but predicts the superiority of one parental species over the other. Ho
wever, if significant overlap occurs in habitat use or niche las in these w
arblers), then competition between parental species is likely to occur. To
determine whether these species compete, we mapped 12 warbler territories a
nd monitored an additional 94 territories throughout the breeding season an
d found that all males with neighbors compete for and hold exclusive territ
ories. Thus, the pattern of habitat use and territoriality is consistent wi
th the parental-fitness asymmetry model.