728 
THE WILSON JOURNAL OF ORNITHOLOGY • Vol 123. No. 4. December 2011 
in the campos rupestres (rocky fields) habitat in 
the Alto do Palacio, Region of Serra do Cip6 
National Park in southeastern Brazil. 
METHODS 
Study Area. —The study was conducted in the 
Alto do Paldcio (AP) (19° 15’ S, 43 31' W) in the 
northern part of Serra do Cipb National Park 
(SCNP) and comprises the southern portion of the 
Cadeia do Espinha?o (Rodrigues et aJ. 2005). The 
study area is within the municipality of Moito do 
Pilar, State of Minas Gerais, in southeastern 
Brazil. 
The AP region is in the eastern portion of Serra 
do Cipd and is characterized by wet campos 
rupestres habitat and a strong influence of certain 
vegetation that is typical of the Atlantic Forest 
biome. The landscape is a mosaic of: ( 1 ) typical 
campos rupestres , which are areas of rocky 
outcrops with herbaceous vegetalion and shrubs 
including Bromeliaceae, Cactaceae, Velloziaceae, 
Urchidaceae. Asteraceac. and scattered trees up to 
3 m in height, especially Eremanthus ervthropap- 
pus; (2) open fields, composed predominantly of 
herbaceous species of Cyperaceae, Poaeeac. and 
Enocaulaceae, and small shrubs with an average 
eight of 1 m; and (3) capdes de mala , which are 
small areas of dense forest-like vegetation asso- 
““ areas ’ such as Sprin s s and »nall 
streams with trees up to 8 m in height and many 
epiphytic plants. Representatives of the Melasto'- 
mataceae and Asteraceae. principally Miconiu 
chartacea and Eremanthus crotonoides. respec¬ 
tively. typically dominate the edges of these 
capoes de mata. 
raiS,^ 8 ™ '•■ Xpene,,Ces variation in 
Jan) airi 5Ummeni (mainl - v ,rom to 
Jan) and dry winters (mainly from Jun to Sep) 
(Madeira and Fernandes 1999). There is, usually 
a soil water deficit from May to August which 
coincides with the coldest months of the year 
Ma^ch The “ S ° n W3ter from N °veraber to 
MI rT" m ° nths ° f lhe ^ (Rodri- 
gues 2 UIJ. Rodrigues ct al. 2011 ). 
Data Collection. —Data were collected from 
December2008 t’* Wi,h the eXCeptio " of 
ecembcr 2008. Ten mist nets (12 X 2.6 m ?5- 
;m me «h) were used in an area of 2 ha of tTpical 
ZiaZTnTu ^ Vidni ‘ y ° f ,W ° * 
1 week in month for a period of 
week in each habitat. Th,- . 
mostly during transit between the typical campos 
rupestres and capdes de mata but. at times, visits 
blooming species in this area (L. C. Rodnguev 
pers. obs.). We chose not to conduct mist-nei 
sampling in the open fields. Mist nets were 
opened at dawn (0600 hrs) and kept open for 6 
consecutive hrs with checks for any captured birds 
every 30 min. Each captured Hyacinth Visor- 
bearer was banded and identified to gender and 
age, and taking standardized morphometric mea¬ 
surements by the same person with the aid of a 
manual caliper of 0.05 mm precision. Measure¬ 
ments included body mass, total length, bill 
length, wing chord, tail length, tarsus length, and 
occipital width. Individuals were classified as 
juveniles or adults based on plumage patterns 
\ oung males have some blue feathers on the sides 
ol the neck and the throat (Fig. IB), while these 
feathers in females are more yellowish-green 
(Fig. 1C). Monthly searches for active nests of 
the Hyacinth Visorbearer were conducted in areas 
where mist nests were set, and occasionally in 
other areas of the AP. 
Statistical Analysis. —A D’Agostino test was 
pertormed to examine the normality of data. At¬ 
test was used to assess whether or not the capture 
late of adults and juveniles differed between the 
dry and rainy seasons. A /-test was performed to 
examine whether or not adult males and females, 
as well as juvenile males and females, differ in 
body mass, wing length, tail length, tarsus length, 
occipital width, total length, and bill length. The 
highest and lowest values of each morphometric 
measurement were excluded from analysis (Go- 
tclli and Ellison 2011). Systat 10 software (2000) 
was used to conduct /-tests. We used a principal 
component analysis (PCA) to reduce the number 
of variables under consideration. The PCA was 
also used to examine which variables had the 
greatest loadings in the linear combination of the 
first principal component (PC I and II). A 
discriminant analysis was performed to evaluate 
our ability to correctly classify adults by gender. 
The PCA and discriminant analyses were con¬ 
ducted using Program PAST.'Version 2.00 
(Hammer et al. 2001). 
I week i.. .. k . . 7 * ,,,WUUI lor a period of 
ThC t>Pen ,1e,ds « areas 
T he Hyacinth VisorTea^! by birdv 
oroearer frequents this habitat 
RESULTS 
Morphometries. —Juvenile and adult females 
did not differ in any morphometric measurements 
that were analyzed, while juvenile males had 
shorter bill length (P = 0.004. / = 2.6, n = 66). 
narrower occipital width (P = 0.04, / = 2.1, n = 
