TABLE 5. Mean ± SE of slope (degrees). DBM (cm), and number of stems (counts) discriminating the six study sites for nesting Mariana Crows. 
Ha el al. 
MARIANA CROW NEST SITE SELECTION 
241 
TfooTj- vc x O' — ac r-4 h~ 
O'. fO O' -O ^ q O) <r> S3 
- oo „iriN o '-(N« 0 - 
tititiqq? +i +i © +i +i +i +i +i 
or-''T OCC ’r'30 O 0'r~-fi-'0-i 
in m « q q in in ri oo a 
oiri— — ci no in od o ib 
't 'f C r-i ' 5 t — o- n-, ro 
co © _ fri C4 
+1 +1 +i 25 5 +1 +1 S q tiq +i 4i 41 
O O © © o 
Mng 'Cm, w 't w t'0'''. 
rtO'O — -rf — oo in 
't tn « o cri © e ni oi 
<n o 
ro tcmMnin'tO' 
^ ^ ri O'’ ~ ^ - r ! 
" - O 5 C C m “ rl r ' 51 01 M © 
+1 +1 <= 25 £5 5 +| +| +| +| +| +| +| +| 
v, n-> 0000 r*'. r-~ © © O' 
i « r- « !n 2 
in O' »* ci O' >C r~ o’ 
■n — — rv 
“ - - F o 8 
+i +i +i 8 8 8 8 8 8 +i 8 8 +i +i 
Tj-r-cocodootNodr-vc 
a in q q q r- 
'ton — — ; '6 
t^; O) Cl oc 00 ^ q rf — 
ddo'ts'd d ind -r d 
+1 +1 +1 +1 +1 3 8 +1 +1 +1 H 8 +1 8 
O O' o 
vO O' 
q q t-; -} d d 'c t q q 
o — ro m «o n - om« O 
+1 +1 8 +1 8 8 +1 +1 +1 +1 +1 +1 +1 +1 
tind-doV, rflBOin-O'N 
O) rt oo r~; — q q — q q 
—• d ro ro d oo d d O', ro d 
~ 5 g 
,111 
w ^ ^ so 
S'* 
= I -3 H = •§ 
C Cl 
| 3 a g 
p £• fc .§■ Q 
'’.3 c r 
•2 _|" 5 £ L g 2 2"?-? 
t ? Cl C u o 3 G 5 ^ 
x x x x x x x xx 
ca ea m eo cn m as cn aa 
^ciaaaaan £ o c s 
i = c = c = c = 2§ = 5 
ioSSSifiowSowi 
Woolfenden 2002; J. C. Ha, unpubl. data) or 
mobbing by Black Drongos (Dicrurus macrocer- 
cus) or Micronesian Starlings (Aplonis opaca). 
The understory may be too sparse to provide the 
vegetative complexity for successful foraging if 
the forest is too mature. Similarly, the Florida 
Scrub-Jay has been shown to favor scrub oak 
(Quercus dumosa) in 80% of its nest substrate 
(Bowman and Woolfenden 2002). 
A limitation of our work is that the six study 
plots were not originally chosen to be represen¬ 
tative of the forest on Rota. They represented 
contiguous crow habitat and were considered for 
special habitat protection by the U.S. Fish and 
Wildlife Service (Morton el al. 1999). Thus, they 
represent a range of Mariana Crow densities 
ranging from 3 to 9 pairs per site, or an average of 
one pair/22 ha (Morton et al. 1999). We believe 
the sites are representative of Mariana Crow 
nesting habitat. There were significant differences 
in vegetation between study sites (Table 5), but 
only one species associated with successful nests 
varied between sites (Lipstick tree. Ochrosia 
mariamensis). This tree species is known to have 
been used for nesting by Mariana Crows on Rota, 
but is rarely used (Morton et al. 1999). Island¬ 
wide comparison of nest site selection and habitat 
use by Mariana Crows, as well as a nesting tree 
species and territory tenacity analysis, and on¬ 
going efforts to film nest activity may reveal 
micro-differences in habitat re.sources that affect 
reproductive success in these endangered birds. 
ACKNOWLEDGMENTS 
We are grateful for the critical reviews of earlier versions 
of this manuscript by anonymous reviewers. Jennifer Marsh 
aided in data entry and quality control. Melanie Colon 
helped copy data sheeis. JMM thanks, in particular. Thomas 
Sharp of the USFWS. and Stan Taisacan and Dave 
Worthington of CNMJ-DFW. This study was only possible 
because of the excellent professionalism shown by 
numerous biologists who volunteered their services: Glenn 
Desy, Beeca Dymzarov. Jim Herrigcs, Shame Holman, Jane 
Jackson. Nathan Johnson. Shona Lawson. Tina Lee, Kim 
Livcngood, Brigid O'Neil, Meaghan Parker, Mark Philip- 
part. Shane Pruett. Robert Reed, Tom Reid ; Michelle 
Rogne, Joey Shousky, Nicole Shutt, Lance Tanino. Mike 
Vamstad. Elaine Wells. Chris Willet. and Sarah Wyshynski. 
Laura Williams and Shelly Kremer provided logistical and 
funding support for our work in the CNMI. RRH and JCH 
thank the Whiteley Center at Friday Harbor Laboratories, 
University of Washington, which provided support for 
writing this manuscript. Funding was provided by the 
Commonwealth of the Northern .Mariana Islands. Division 
of Fish and Wildlife and the USDI, Fish and Wildlife 
