Journal of the Royal Society of Western Australia, 87(3), September 2004 
• Tree Martin (H. nigricans ): Rare; only one individual 
seen. 
• Australian Pipit (Anthus australis): Rare; not seen on 
quadrats; seen on two occasions in low limestone 
ridges along coast. 
• Spinifexbird (Eremiornis carteri ): Common; the most 
abundant species on island; in all habitats with 
spinifex. Many pairs were breeding. 
• White-winged Fairy-wren ( Malurus leucopterus): 
Common; abundant in most habitats, especially those 
with complex vegetation structure. Many pairs were 
breeding. 
• Singing Honeyeater ( Lichenostomus virescens): 
Common; abundant in all habitats with dense 
vegetation or with emergent shrubs and trees. Many 
pairs were breeding. 
• Zebra Finch (Taeniopygia guttata): Common on clay 
pans; uncommon in other habitats. 
• White-breasted Woodswallow ( Artamus 
leucorynchus ): Common; abundant in all habitats with 
dense vegetation or with emergent shrubs and trees. 
In Table 1 we list, for each of the six land forms/ 
vegetation units, the number of species and individuals 
recorded, as well as the estimated densities of the six 
common species. Across all vegetation types together, the 
density of birds averaged 2.20 birds/ha. Across the 31 
vegetation types, there was a significant correlation 
between the area of habitat that we surveyed and the 
total number of species (F x 29 = 11.585, R 2 = 0.261, P = 
0.002) and total number of individuals (F, w = 79.588, R 2 = 
0.724, P < 0.001) recorded. Similarly, there was a 
significant relationship between the total number of 
species and number of individuals recorded on the 
transects (F 1#29 = 36.289, R 2 = 0.541, P < 0.001). 
Total population estimates for all 16 species recorded 
on quadrats are presented in Table 2, along with the 
population estimates for these same species published in 
Sedgwick (1978). 
Habitat Relations 
Land birds were distributed across the entire island, 
but both species diversity and abundance varied with 
habitat (Table 1). Each of the common species observed 
in this study was found in a variety of habitats, but 
densities varied considerably (Table 1). 
For the 31 vegetation types separately, we examined 
whether any of the habitat variables correlated with 
number of species and individuals recorded, and with 
the densities of the six common bird species (Table 1). A 
complete summary of habitat measurements and species 
abundances on each of the 31 vegetation subtypes is 
available on request from the authors. There was no 
significant correlation between any of the habitat 
variables and species richness, or total number of 
individuals. Nevertheless, several significant 
relationships between habitat measurements and 
densities of individual species emerged: the density of 
Spinifexbirds was significantly correlated with the mean 
emergent score (F, 29 = 11.046, P = 0.002) and with 
vegetation cover (Fj ^ = 5.393, P = 0.027; Fig. 1). The 
density of Singing Floneyeaters was significantly related 
a. 
c 
o 
6 
5 
o 
73 
C 
a 
cr 
£ 
£ 
3 
C/D 
£ 
X 
73 
S S 
C O 
o — 
> £ 
5 c/i 
^ £ 
a .3 
N u« 
73 
<y > 
O 
u £ 
at 73 
.tS o 
r- O 
UD % 
tC 
.5 g 
C/D O 
X 
•? 5 
73 
- 
1H 
Dh 
C/D 
-O 
E 
3 
2 
Dh 
C/D 
rj 
G j- 
O w 
rs 73 
2J c 
< £ 
V) 
Dh 
H 
o 
s 
73 
c 
(N vO N h 
CN i—< O O 
o o o o 
n m (N in 
<N O O nO 
O O O rH 
CN 
rH 
o 
IN CN O 
dodo 
3 
o o o o o 
o o o o o 
CO 
O 
ON 
© ^ 
NO 
O' in tx • m 
oo no no in on 
o oo o oo no 
cd 73 
I o 
w> 
73 
e 
O 
73 
C 
u cue 
U -a Q lL 
q, 
— <T3 
J U ^ 
J oS > 
^ .5 
D* 
o 
£ j 
£-< 
35 
£ H 
104 
